6 - 1 tf CONTRIBUTIONS THE TEXTUAL CRITICISM OF THE DIVINA COMMEDIA. ILonUon : C. J. CLAY AND SONS, CAMBRIDGE UNIVERSITY PRESS WAREHOUSE, AVE MARIA LANE. CambrtUge: DEIGHTON, BELL, AND CO. ILeip>ifl: F. A. BROCKHAUS. CONTRIBUTIONS TO THE TEXTUAL CRITICISM OF THE DIVINA COMMEDIA INCLUDING THE COMPLETE COLLATION THROUGHOUT THE INFERNO OF ALL THE MSS. AT OXFORD AND CAMBRIDGE. BY THE REV. EDWARD MOORE, D.D. PRINCIPAL OF S. EDMUND HALL, OXFORD, AND BARLOW LECTURER ON DANTE IN THE UNIVERSITY OF LONDON ; AUTHOR OF 'TIME-REFERENCES IN THE DIVINA COMMEDIA.' O degli altri poeti onore e lume, Vagliami il lungo studio e il grande amore Che m' ha fatto cercar lo tuo volume. Inf. i. 824 AT IE UNIVERSITY PRESS {All Rights reserved.] GTambrtoge : PRINTED BY C. J. CLAY, M.A. AND SONS, AT THE UNIVERSITY PRESS. THIS WORK UNDERTAKEN AT THE SUGGESTION AND IN ITS EARLIER STAGES ASSISTED BY THE ADVICE OF THE LATE PRE-EMINENT DANTE-SCHOLAR DR CARL WITTE, THE DEDICATION OF WHICH HE HAD KINDLY CONSENTED TO ACCEPT, IS NOW INSCRIBED WITH PROFOUND FEELINGS OF RESPECT AND ADMIRATION TO HIS MEMORY. PREFACE. THIS work, which has been the labour of several years, and has been carried on amidst the pressure and interruption of many other duties and engagements, has at last reached not indeed completeness, but a convenient halting-place. Completeness must neces- sarily be out of the question when the MSS. known to exist are between 500 and 600 in number : but nearly half of that total having in the case of the more im- portant passages been examined, it seemed likely that the evidence thus collected would be a sufficient sample of the whole, so that the results and conclusions obtained from such materials would not be likely to be much altered by the examination of an increased number of MSS. As Dante himself says molte volte taglia Piu e meglio una che le cinque spade. Par. xvi. 72. The manner in which this part of the work, I mean the collation of selected passages, has entirely outgrown my original purpose, is accountable for the, I fear, rather inconvenient system of nomenclature by which the MSS. have been referred to. Had I foreseen the dimensions to which that list would grow, I should PREFACE. have adopted from the first a uniform system of numerical designation. There remains however this advantage of the system of various alphabets that certain important collections of MSS. are thus kept uniform in their nomenclature. Thus the magnificent Ashburnham Collection is included in the small Greek alphabet. The British Museum, Holkham and Chel- tenham MSS. are designated by the small Roman letters. But an unfortunate result has been the increased liability to errors and misprints from the similarity (espe- cially in manuscript) of certain letters such as, v and v, a) and w, o and o, i and i, &c. Some few cases of such errors I have already detected, and it is only too probable that others may remain. Apart from this, I fear that in a work in which the references are to be counted by thousands, and even tens of thousands, the utmost care will not have availed to avert the occur- rence of several mistakes. Should such be found, I can only ask the indulgence of fellow-students, in considera- tion of the difficulty of the work, for errors quas aut incuria fudit Aut humana parum cavit natura. It only remains that I should acknowledge with gratitude the kind assistance which I have received in many quarters. Chiefly are my thanks due to the Syndics of the Press of the University of Cambridge, who have with very great kindness and liberality undertaken the whole expense of the publication of a work which I fear must prove both costly and unremunerative. For assistance in the consultation of the splendid PREFACE. Libraries in their possession or under their charge, I must especially acknowledge the kindness and courtesy of the Earls of Ashburnham and Leicester, the late Lord Vernon, the Marchese Podesta of the Magliabecchian Library at Florence ; the late Count Porro of the Trivulzian Library at Milan; the late Rev. Alexander Napier, the Librarian at Holkham Hall ; and above all, of Mr Maunde Thompson, keeper of the MSS., and now I am glad to say Principal Librarian, at the British Museum. His unrivalled knowledge of MSS. has always been at my service if I desired advice, and he has never grudged any pains or trouble which my inquiries may have involved. From the Librarians of foreign Libraries in Italy and at Paris I have never failed to receive courtesy and attention. To enumerate them specifically would be almost equivalent to giving a simple list of the Libraries which I have visited. To two Oxford friends I owe a very special debt of gratitude. The Rev. C. H. Daniel, Fellow of Worcester College, whose skill and taste as an amateur printer is well known, has most kindly gone over the proof-sheets of a large portion of the work. The Rev. H. F. Tozer, Fellow and Tutor of Exeter College, has not only done the same, but he has also assisted me with criticism and advice to which his minute acquaintance with the Divina Commedia lends a special value. He has also very kindly contributed an Essay on the Metre of the Poem, for which my readers as well as myself will have occasion to feel grateful to him. S. EDMUND HALL, OXFORD, Dec. 31, 1888. D. I. b FULL TITLES AND EDITIONS OF THE PRINCIPAL WORKS REFERRED TO. Angeletti, N. Cronologia delle Opere Minori di Dante. Citta di Cas- tello, 1886. Antonelli, G. Di alcuni Studi speciali risguardanti la Meteorologia, la Geometria, la Geodesia, e la Divina Commedia. Florence, 1871. Balbo, C. Vita di Dante. Florence, 1853. Barlow, H. C. Contributions to the Study of the Divina Commedia. London, 1864. Francesca da Rimini, her lament and vindication. London, 1859. Baroszt, N. \_Fulin and Gregoretti\. I Codici di Dante Alighieri in Venezia. Venice, 1865. Batines, Colomb de. Bibliografia Dantesca. Prato, 1845. Blanc, L. G. Grammatik der Italianischen Sprache. Halle, 1844. Versuch einer bloss philologischen Erklarung mehrerer dunklen und streitigen Stellen der Gottlichen Komodie (Inferno and Purgatorio i. xxvii.). Halle, 1860, 1865. Vocabolario Dantesco. Florence, 1859. Borghini, V. Studi sulla Divina Commedia di Galileo, Borghini ed altri. Florence, 1855. Butler, A. J. The Purgatory with Translation and Notes. 1880. The Paradise with Translation and Notes. 1885. Carlyle^J. A. Dante's Divine Comedy : The Inferno. A Literal Prose Translation, &c. 2nd ed. London, 1867. Castelvetro, L. Opere Varie Critiche. Berne, 1727. [See also under Commentaries.] Clerici, G. P. Studi Vari nella Divina Commedia. Citta di Castello, 1888. Commentaries on the Divina Commedia. See separate list. Contarini. Dizionario Tascabile delle voci e frasi particolari nel Dialetto Veneziano. Venice, 1852. Diez, F. Etymologisches Worterbuch der Romanischen Sprachen. Bonn, 1869. Grammatik der Romanischen Sprachen. Bonn, 1876. b2 FULL TITLES AND EDITIONS OF Dionisi. Dei Blandimenti Funebri. Padua, 1794. Preparazione istorica e critica alia nuova Edizione di Dante. Verona, 1806. Serie d' Aneddoti. Verona, 1785. Esposizione Dantesca in Firenze. Maggio 1865, Cataloghi. Fanfani, P. Studi ed Osservazioni sopra il Testo delle Opere di Dante. Florence, 1874. Febrer, IV' Andreu. La Comedia di Dant Allighier traslatada de Rims Vulgars Toscans en Rims Vulgars Cathalans (Siglo xv). Barcelona, 1868. Foscolo, U. Discorso sul Testo del Poema di Dante, being the first Vol. of 'La Commedia di Dante Allighieri illustrata'. 4 vols. London, 1842. Fransoni, D. Studi Vari. Florence, 1887. Fraticelli, P. Le Opere di Dante Alighieri con Illustrazioni e Note. 4 vols. Florence, 1873. Storia della Vita di Dante Alighieri. Florence, 1861. Gaiter, L. II Dialetto Veneto nel Secolo di Dante. Bologna, 1877. Gentile, L. II Codice Poggiali della Divina Commedia. Florence, 1888. Gherardini, G. Supplemento ai Vocabolari Italiani. Milan, 1855. Giordano, G. Studi sulla Divina Commedia di Dante. Naples, 18846. Giorgi, Ignazio. Aneddoto di un Codice Dantesco. Leghorn, 1880. Giuliani, G. Dante spiegato con Dante. Florence, 1868 1882 (in- cluding all the works of Dante with Commentaries, except the Divina Commedia). Gregoretti, see under Barozzi. Guerrini e Ricci. Studi e Polemiche Dantesche. Bologna, 1880. Hegel, C. Ueber den historischen Werth der alteren Dante-Commentare. Leipzig, 1878. Hettinger, F. Dante's Divina Commedia, its scope and value, from the German of Franz Hettinger, edited by H. S. Bowden of the Oratory. London, 1887. Hueffer, F. The Troubadours. London, 1878. Lorini, A, Le Varianti della Divina Commedia tolte dal Codice mem- branaceo Cortonese. Cortona, 1858. Lubin, A. Studi Preparatori Illustrativi all' Edizione della Divina Corn- media. Padua, 1 88 1. Mastrqfini, M. Dizionario Critico de' Verbi Italiani Conjugati. Rome, 1844. Midoux, E. et Matton, A. Etudes sur les Filigranes des Papiers em- ploye's en France aux xiv e et xv e siecles. Paris, 1868. Moore, E. The Time-References in the Divina Commedia, and their bearing on the assumed date and duration of the Vision. London, 1887. Mortara, A. Catalogo dei Manoscritti Italiani che sotto la denomina- xii THE PRINCIPAL WORKS REFERRED TO. zione di Codici Canonic! Italian! si conservano nella Biblioteca Bodleiana a Oxford. Oxford, 1854. Muratori. Rerum Italicarum Scriptores. Milan, 1723. Mussafia, A. Sul Testo della Divina Commedia ; I Codici di Vienna e di Stoccarda. Vienna, 1865. Nannucri, V. Analisi Critica dei Verbi Italian!. Florence, 1843. Teorica dei Nomi della lingua Italiana. Florence, 1847. Intorno alle Voci usate da Dante in grazia della Rima. Corfu, 1840. Intorno al Pronome Lei usato dagli Antichi nel Caso Retto 1 . Corfu, 1841. Negroni, C. L' Avverbio "Parte" e i Commentated di Dante. Novara, 1880. Norton, C. E. The New Life of Dante, An Essay, with Translations. Cambridge, U. S. A., 1859. Pallastrelli, B. II Codice Landiano della Divina Commedia. Pia- cenza, 1865. Papanti, G. Dante secondo la Tradizione e i Novellatori. Leghorn, 1873- Pelli, G. Memorie per servire alia Vita di Dante Alighieri. 2nd Ed. Florence, 1823. Perez, P. I sette Cerchi del Purgatorio di Dante. Verona, 1867. Philalethes. Dante Alighieri's Gottliche Comodie. Dritter Abdruck, be- sorgt von J. Petzholdt. Leipzig, 1877. Plumptre, E. H. The Commedia and Canzoniere of Dante Alighieri. 2 vols. 1886 7. Poletto, G. Dizionario Dantesco. 7 vols. Siena, 1885 7. Rossetti, G. La Divina Commedia con Comento analitico in sei Volumi (only 2 vols. published). London, 1826. Scarabelli, L. Esemplare della Divina Commedia donato da Papa (Benedetto XIV.) Lambertini...allo Studio di Bologna... illustrate dai confronti di altri xix Codici Danteschi. 3 vols. Bologna, 18703. Scartazzini. La Divina Commedia di Dante Alighieri. Riveduta nel Testo e commentata. 3 vols. Leipzig, 1874 1882. Vita di Dante, and Opere di Dante. Nos. xlii. and xliii. of the Manueli Hoepli, Milan, 1883. Schneider, C. F. Ueber den Reim in Dante's Divina Commedia. Bonn, 1869. Sicca, A. Rivista delle varie Lezioni della Divina Commedia sinora avvisate. Padua, 1832. Slimming. Bertran de Born, sein Leben und seine Werke. Halle, 1879. 1 This should have been referred to in the discussion ot Purg. xxi. 25 supra p. 400. xiii WORKS REFERRED TO. Todeschini, G. Mende e Contradizioni nella Divina Commedia, in Vol. i. of Scritti su Dante. Vicenza, 1872. Vaccheri e Bertacchi. Cosmografia della Divina Commedia. Turin, 1881. Villani, G. Cronica. Ed. Magheri. Florence, 1823. Vocabolario Universale Italiano. Tramater e C ia . 7 vols. fol. Naples, 1829 1840. Witte,C. Dante-Forschungen. 2 vols. Heilbronn, 1868 and 1879. Prolegomeni Critici all' Edizione della Divina Commedia ricorretta sopra quattro dei piu autorevoli Testi a penna. Berlin, 1862. Zani dd Ferranti, M. Di varie Lezioni da sostituirsi nell' Inferno di Dante Alighieri. Bologna, 1855. Zingarelli, N. Parole e forme della Divina Commedia aliene dal Dialetto Fiorentino forming Fascicolo I of the Studi di Filologia Romanza. Ed. Monaci. XIV LIST OF THE EARLY COMMENTARIES ON THE DIVINA COMMEDIA AS REFERRED TO IN THIS WORK. Referred to thus : Ch. Anon, 6. Probable date : Chiose Anonime alia prima Cantica della Divina Commedia. Ed. Selmi. Turin, 1865. 1320 ? Commento alia Cantica dell' Inferno di Autore Anonimo 1 . Printed by Lord Vernon at Florence 1848, in 100 copies only. 1324 Chiose alia Cantica dell' Inferno attri- buite a Jacopo di Dante. Printed by Lord Vernon at Florence 1848, in i oo copies only. before 1333 Commento di Jacopo della Lana. Ed. Scarabelli. 3 vols. Bologna, 1866 7. c. 1328 ? L' Ottimo Commento. Ed. Torri. 3 vols. Pisa, 1827 9. 1333 Petri Allegherii super Dantis ipsius gen- itoris comcediam Commentarium sumtibus Bar. Vernon, curante V. Nannucci. Florence, 1845. 1340 Commento di Giovanni Boccaccio. Inf. i. to xvii. 17. 1373 Chiose sopra Dante (The so-called "False Boccaccio") published by Lord Ver- non. Florence, 1846. 1375 Benvenuti de Rambaldis de Imola Com- mentum sumtibus Guilielmi Warren Vernon, curante Jacobo Philippo La- caita. 5 vols. Florence, 1887. 1379 1 I have endeavoured to shew in an Article in the Academy of Oct. 8, 1881 that this is nothing but an Italian Translation of the Latin Commentary of Ser Graziuolo de' Bambagioli, which is twice quoted by the author of the Ottimo Commento. Dr Witte, who shortly before his death was engaged on a special study of the Seville MS. of that very ancient Commentary (long supposed to be lost), was kind enough to write to me to say that he thought that I had quite established this point. XV Com. Anon. Jac. Lan. Ott. Pet. Bocc. Fals. Bocc. 8. Benv. EARLY COMMENTARIES REFERRED TO. Referred to thus : Probable date : Buti. 10. Commento di Francesco da Buti. Ed. Giannini. 3 vols. Pisa, 185862. 1393 Anon. Fior. n. Commento d' Anonimo Fiorentino. Ed. Early 15 th Fanfani. 3 vols. Bologna, 1866 74. cent. Barg. 12. Lo Inferno... col Commento di Guiniforto delli Bargigi. Florence, 1838. c. 1440 (?) Ric. 13. La Commedia di Dante Alighieri col commento di Stefano Talice di Ri- caldone per cura di V. Promis e C. Negroni. 3 vols. Milan, 1888. 1474 Land. 14. Commento di Cristoforo Landino. Ed. princeps Firenze per Nicholo di Lo- renzo della Magna. 1481. 1480 Veil. 15. La Comedia...con la nova espositione di Alessandro Vellutello. Ed. prin- ceps Vinegia per Francesco Marco- lini. 1544. 1544 Varch. 16. Lezioni sul Dante per Benedetto Varchi per cura di G. Ajazzi e L. Arbib. Florence, 1841. 1 543-5 Dan. 17. Dante con 1' Esposizione di M. Ber- nardino Daniello da Lucca. Venice 1568. 1568 Cast. 1 8. Sposizione di Ludovico Castelvetro a xxix Canti dell' Inferno Dantesco 1 . Ed. Franciosi. Modena, 1886, in 300 copies only. c. 1570 UNPUBLISHED COMMENTARIES. Guid. 19. Commento Latino sopra 1' Inferno di Dante di Guido Pisano. before 1333 Only one copy of this ancient com- mentary was known to Colomb de Batines (see Bibl. Dant., n. pp. 137, 299). It was at Milan in the Library of the Marchese Archinto, which has since been dispersed. I have not 1 See a fuller description of this interesting Commentary in an article by myself in the Academy of July 10, 1886. xvi UNPUBLISHED COMMENTARIES. i Referred to thus : Probable date : been able to trace this MS. 1 The Sun- derland Library however contained another copy, which was purchased by the British Museum at the sale of that Library in i882 2 . The Commentary is evidently a very old one, since in the Commentary on Inf. xiii. 146 7, the statue of Mars, destroyed by an inundation 1333, is spoken of as still standing, and hon- oured by an annual festival. The Commentary on each Canto has three divisions (i) l Deductio textus de Vulgari in Latinum? This is sometimes a pretty close translation, at other times a diffuse paraphrase, and at others a very much abbreviated compendium. (2) ' Expositio littere? (3) An enumeration of the chief his- torical allusions, prophecies, compari- sons and 'notabilia' in the Canto. Serr. 20. The Latin Commentary on the Divina 1417 Commedia of Giovanni da Serravalle, Bishop of Fermo. This was composed by him at the instance of (i) Cardinal Amidei of Saluzzo ; (2) Nicolas Bub- with, Bp of Bath and Wells ; (3) Robert Hallam, Bp of Salisbury. It is further 1 I have discovered quite lately that this MS. (Bat. 256) is now in the magni- ficent Library of the Due d'Aumale, where through the kindness and courtesy of H.R.H. I have been permitted to inspect it. It is a splendidly executed folio MS. on vellum, and certainly, I should say, of the first half of the I4th century (not towards the end of the century as stated by Batines, II. p. 299). The writing throughout both of the text and commentary is beautifully executed Italian Gothic, and there are also some very interesting drawings, illustrative of scenes in the poem, which seem to be of the School of Giotto. Unfortunately I had not with me the full list of test-passages for the Vatican family, but I looked out a good many of these from memory and found the Vatican reading in all cases (e.g. ii. 12; xi. 90; xii. 16, 125; xvii. 50; xxxiii. 43, 74; xxxiv. 93; &c.). This, it will be remembered, is the case with the copy of this Com- mentary in the British Museum (see under 'q' supra p. 603, and Table I. in Appendix II.). I observed also the same blank in the Commentary at Alino (iv. 141) which is mentioned supra p. 283. 2 I have also seen in Lord Vernon's Library an Italian Translation of a portion of this Commentary, Cantos i. xxiv. (see Bat. II. p. 298). ii UNPUBLISHED COMMENTARIES. stated in the colophon that it was written at Constance 'vacante sede Apostolica et tempore Generalis Con- cilii Constantiae,' the actual date of its completion being Jan. 16, 1417. The Commentary commences with eight 'Preambula' occupying about 15 or 1 6 pages. Then follows a Latin Com- mentary on the whole poem. The MS. concludes with a Latin verse trans- lation (wrongly described as 'prose' by Colomb de Batines, I. p. 247), stated in the colophon to have been commenced in Jan. 1416 and finished in May of the same year. The chief interest of this Commentary is due to the fact that it is the sole authority for the tradition that Dante visited England and Oxford. This statement occurs in the fifth Preambulum. This Commentary, like that of Guido Pisano, is almost unique. Only one copy was known to Colomb de Batines which was, and still is, in the Vatican Library. See Bibl. Datitesca, II. pp. 333 5, for a very full account of this MS. Most of this description applies to another copy which formerly be- longed to the Marquis of Donegal, then to Mr Wodhull of Thenford House, near Banbury. When his fine Library was dispersed in 1886 this almost unique MS., specially interesting to Englishmen for the reason above mentioned, was happily secured for the British Museum 1 . 1 I have given some description of this MS. and its contents in the Academy of Feb. ao, 1886. CONTENTS. PAGE Prolegomena v xlvi Prefatory note on Interpretation of MSS xlvii Ivi Text of the Inferno, with complete Collation of 17 MSS. . I 254 Collation and Discussion of Selected Passages . . . 255 507 Account of the MSS. collated or examined .... 509 679 List of MSS. collated or examined 680 685 List of Lines omitted, transposed, or repeated in error . 686, 7 List of peculiar Readings 688 691 APPENDIX. I. Dante's references to Classical Authors as bearing upon textual criticism . . 695 II. The ' Vatican ' family of MSS., and other groups 701 III. The interpolated lines in Inf. Canto xxxiii. . 706 IV. On the text of Witte's Berlin Edition . . 711 V. On the Metre of the Divina Commedia (by the Rev. H. F. Tozer) .... 713 ERRATA AND ADDENDA. Page 21, line 8 from bottom, for la read la. ,, 22, 19 from bottom, for disc read disc. ,, 46, ,, 12 and 13 from bottom, for dl raw? di' (bis). 263, 8 from top, for tremesse read tremasse. ,, ,, 9 from top, for E read E. 2 73> 8 from bottom, for non raz 16 from top, dele j. ,, ,, ,, 22 from top, after rivolge insert I (rinvolge) j. 33 2 > *5 fr m top, dele TAA; and in 1. 17, after "L insert T*AA, and after bn insert 56.57.90.119. 336, ,, 12 from top, for I and 86 read 1.86 and 104. D 363, 15 from top, dele 9, and after 1. 23 insert due d' elli Q. 364, ,,17 from bottom, after osso mr# Tr. 365, >, ii from bottom, after ^f insert f. 369, 20 from top, for a read a. 381, after 1. 3 from bottom, insert I has e nel minor. ,, 383, line 5 from top, after 10 add 94.117. 3^5) 7 from top, for 62 raz. 457 '4 from top, for v ra&/ v. PROLEGOMENA. THE difficulty of determining the text of the Dimna Corn-media is far greater than might have been expected from its comparatively recent date. It would be nothing surprising if even the autograph had been preserved of a work not yet six centuries old 1 . As a matter of fact, we 1 It is a remarkable fact that not a single stroke of Dante's handwriting is known to exist, though his father's signature has been found, and a fac- simile of this is given, with a reprint of the document in which it appears, in a monograph entitled La Casa di Dante (1865), pp. 13 and 25. It has, I think, been asserted that Dante's own signature as a witness to a legal document exists. This erroneous statement has probably arisen from the two facts (i) that his father's signature exists exactly in this manner (as explained above), and (2) that Dante himself is mentioned as present at, and as witness to, a contract made at Padua on Aug. 27, 1306, but it does not appear that his own signature is attached. (See Pelli, Memorie, &c., p. 115 n., Balbo, Vita, p. 246.) The Florentine archives have been dili- gently searched for his signature but in vain, though it would certainly be expected that some trace would have been found there of the hand- writing of one who held so prominent an official position in the city. It is singular also that this fatality has befallen the handwriting of other celebrated authors in much later times, e.g. Shakspeare and Moliere, of whom at most there are five signatures in each case supposed to exist, but absolutely none of the autographs of their writings. Is the interest in the autographs of great men perhaps a comparatively modern feeling ? The handwriting of Petrarch, it may be mentioned, exists in abundance. Boccaccio says of Dante's numerous Epistles, " delle quali ancora appa- riscono assai." It is also interesting to note that Leonardo Aretino (d. 1444), in his Vita di Dante, minutely describes Dante's handwriting, and quotes 8 or 10 lines from an Epistle written by Dante himself which he had seen : so that for at least a century after his death some of these precious documents were still in existence, which makes their subsequent dis- appearance still more strange. Leonardo's description is worth quoting : "Fu scrittore perfetto, ed era la lettera sua magra, e lunga, e molto corretta, secondo io ho veduto in alcune pistole di sua propria mano scritte." This description at any rate seems intrinsically most probable. VI PROLEGOMENA. have no dated MSS. (excluding cases of palpable blunders of date) professing to be older than 1335 or 1336, even sup- posing those dates were beyond doubt, which is far from being the case 2 . In other words, we have no MS. which pro- fesses to be within about 15 years of the author's death. Unfortunately too, even those documents, which, whether dated or undated, seem to be the earliest or the most trust- worthy, all exhibit signs of a text that has been tampered with 3 . So among others does the celebrated Codice di Santa Croce at Florence, dated 1343 (which however seems almost certainly an error for 1443)*, which Witte agrees with Dionisi in considering as, on the whole, probably the best MS. in existence. More than that, the earliest commentators, some of whom, as Jacopo della Lana, go back at least to 1328, or, as the Ottimo, to 1333, not only display undoubtedly false readings, but they also recognize and discuss various readings as already then existing. Witte, D. F. I. p. 392, gives examples from both these early Commentaries. He cites, as an instance of a false reading, arme for arnie in Inf. xvi. 3 (Ott.). I would add also fama for infamia in Inf. iii. 36 (Lan.), and nuovo for nona in Purg. xxvii. 4, certainly in Otty and apparently also in Lan. Other instances will be found recorded in these collations, though one must always be careful to observe whether a given reading is distinctly recognized or implied in the Commentary itself. Its exist- 2 The three earliest dated MSS. are those which I have designated a (1335, but this date is in a later hand and is almost certainly spurious), A (1336) and r (1337). See the description of these MSS. later on. There are indeed others bearing the dates 1328 and 1329, but these figures are obviously and admittedly erroneous. (See Witte, Prol. p. Ixvii.) 3 Besides instances that will be found in these collations, see further, Witte, Prol. p. Ixviii., and in reference to the Codice di Santa Croce, ib. p. Ixx. n. 2. 4 This date must surely be erroneous. A glance at the writing of the MS. seems to disprove the genuineness of such a date as 1343, and to make 1443 more probable. A similar probable error of a century is noted by Batines (n. p. 81) in another MS. The difficulty of this simple solution in the present case lies in the accompanying reference to the expulsion of the Duke of Athens, which would rightly correspond to the earlier date. See further Witte, Prol. p. Ixix., Dionisi, Anedd. v. pp. 43 66, and the description of this MS. in its place later on, PROLEGOMENA. vii ence in the accompanying printed text is of no authority. The following are some of the instances given by Witte of variants discussed or implied in these early Commentaries. Inf. viii. 78, fusser fosse for ferro fosse (Law.), (this is only implied however); Purg. vii. 15, ove '1 minor s' appiglia, and dove '1 notrir s' appiglia (discussed by Lan.}. We may add affetti and effetti (recognized by Ott.} in Purg. xi. 3. Of the first of these there is, as far as I know, absolutely no evidence in existing MSS. ; and the second is extremely rare 5 . I have only found it in 6 MSS. out of about 200 examined, and in two of these cases the text appeared to have been tampered with. At any rate about the middle of the I4th century, or even earlier, we find what Witte styles a ' lezione volgata ' already constituted, and no existing MS. is independent of it. The great problem then for textual criticism is to get behind this 'lezione volgata; to reconstruct from a large number of ' disjecta membra ' the extinct original of whose existence they afford the evidence. The Divina Commedia made its appearance at a time and under circumstances in some respects peculiarly unfavourable for the preservation of its correct text. The work immediately 5 There are certainly some cases in which a reading once known to exist has either entirely, or almost entirely, disappeared from existing MSS. This of course only means that the change was made very near the foun- tain head, possibly in some cases in the earliest copies of the autographs, and that such copies do not happen to have been often recopied (see p. xxvi.) : e.g. in Purg. xxii. 5 the certainly true reading sitiunt (where see my notes) : and again v&Purg. ix. 17 the reading Men... pi ft commented on by Pietro (c. 1340) p. 356, instead of Piu...nien, does not occur in more than four of about 200 MSS. that I have examined on this passage. In Par. xxviii. 64 the reading enno for sono found in the Ed. Nidob. and others (and which as lectio dijficilior is probably correct), does not occur in any MSS. that I have seen, though it is true I have not examined so large a number here. In Purg. ii. 13 the reading sttl presso, which is (I am inclined to think) not improbably the correct one, has been found by me at any rate only twice in about 200 MSS. examined. It is recognized however by Benv., Dan., Buti, Land, and Veil. We might also compare Purg. xxi. 61, where the true reading sol volerfa, though that of the old Commentators without exception, and without even the recognition of any other, is found only in about one in five of existing MSS. This would probably imply that the lect.fals. has since been most frequently copied, and that at any rate the proportion of MSS. supporting the two readings was formerly different. Vlll PROLEGOMENA. attained to a vast reputation and attracted universal atten- tion. Copies of it were multiplied with great rapidity, and unfortunately by copyists who were just able to take an unintelligent interest in what they were doing. Witte, for example (Prol. p. xlv.), mentions the case of one person who is said to have made 100 copies to provide dowries for his daughters. The language in which it was written, the sub- jects of which it treated, the nearly contemporary interests on which it touched, and the well-remembered incidents with which it abounded, combined with its frequent obscurities and difficulties, all contributed to give the copyist an interest in his work which affected most injuriously its accuracy and faithfulness. We are reminded of the treatment of the Homeric poems by the Btopdcoral of the Alexandrian School. 'Tlavros yap Trpoadeivac TO eXXewroz/ seems to have been un- fortunately a recognized principle with these copyists. This experience illustrates the exaggerated statement which is sometimes made that the correction of proof sheets is often better done by those who are unacquainted with the lan- guage in which they are printed. If Dante had carried out his original intention of writing in Latin (which we may be very thankful that he soon abandoned, to judge from the specimen which is preserved), I have little doubt that we should have had handed down to us a much more accurate transcript of his original. The copyists were in fact at the same time amateur, un- critical, and anonymous editors, unchecked by the editor's sense of responsibility; and their experiments in 'restoration ' have been as reckless and as fatal as any that we have had to deplore in other departments of art. It is indeed only in comparatively modern times that the true function of an editor or textual critic has been recognized to be the dis- covery of what the author actually did write, not the sug- gestion of what he ought to have written. The process of emendation is thus to be applied to a text only when there is good reason to suppose that corruption of the original has occurred : in other words, when we believe that we have before us something that the author did not write. It cannot be too clearly and emphatically asserted that the sole object of the textual, as contrasted with the artistic PROLEGOMENA. IX or literary critic, is to determine, on any grounds of evidence or principles of probability that may be available, what Dante is most likely to have written. He has nothing to do with the superior correctness or clearness or elegance or taste of a reading, except so far as these considerations may per- haps sometimes bear upon the question of its originality. He is not bound to maintain that the reading selected on critical grounds as probably original is the best possible on artistic and literary grounds. No writer is in every line perfect and infallible. There are times when even "bonus dormitat Homerus." Yet it is comparatively speaking re- cently that this, as we would venture to call it, axiomatic principle of textual criticism has been generally admitted. Many of the emendations of even so great a scholar as Bentley in the last century (notably those on Horace and Milton) went on quite a different principle. Assuming a species of infallibility in his author (and apparently still more in himself), he argued This or that expression is open to such and such objections, and therefore it should be altered, even though not a shadow of suspicion rested on the text from MS. authority. This audacious treatment of an author by a really great critic and in comparatively modern times so well illustrates the process to which the text of Dante has been freely subjected by unknown and far less competent hands, that I cannot refrain from giving an example of it. I borrow from Prof. Jebb's Life of Bentley the following startling, but typical, instance. Bentley finds a number of objections, mostly of a minute and hypercritical kind, to the two well- known concluding lines of Paradise Lost: They hand in hand with wandering steps and slow Through Eden took their solitary way. He proposes consequently to emend them thus: Then hand in hand with social steps their way Through Eden took, with heavenly comfort cheered! This is scarcely an exaggeration, indeed, in the case of some few MSS. which I could name, it is no exaggeration at all, of the way in which copyists have treated Dante, un- happily not limiting themselves like Bentley to retaining the text and placing the suggested emendations in a note or in D. I. c X PROLEGOMENA. the margin 6 . The language of Dr Mill respecting the well- known Codex BezcB of the New Testament may well be applied to some of these cases. " Vix dici potest quam supra omnem modum licenter se gesserit ac plane lasciverit interpolator." It will be well now to illustrate in some detail, and to attempt to classify, the chief causes which have been in operation in the deterioration of the text, since this will 6 The following are some typical instances : In Inf. iv. 59, the scribe of D, thinking apparently that Isaac deserved a more explicit mention, reads Israel con Isaac instead of Israel con lo padre. A singularly bold and amusing case occurs in Par. xxvi. 136, where ' j ' makes Adam teach the doctrine of the Trinity by reading " Et Tre si chiamb poi n \ In Inf. i. 28 one MS. ('62') reads Poiprese lena unpoco; and in xxviii. 135 '83 ' has Che diede al re Riccardo(^. See ' 83 ' also in Inf. xxxiv. 13 ; ' 14 ' in Par. xix. 141. The scribes of F, j, 6 (and to some extent ;) seem to have been especially possessed with this prurigo corrigeudi, as I think Bentley calls it. In many other MSS. besides these, the substitution, by way of improvement, of synonymous single words or expressions is ex- tremely common, e.g. parti iorfuggl Inf. xxv. 16 (D), modo for guisa Inf. xxx. 49 (a), niuna for alcima Inf. xxx. 58 and 1 17 (L), and similarly veruna in Inf. iv. 12 (CK), andar for ir Inf. xxxi. 141 (EFL), acerbi for lazzi Inf. xv. 65 (Q), trasmuta for converte Inf. xxv. 99 (F), tirando for strac- ciando Inf. xxii. 72 (O), trafitto for feruto Inf. xxv. 105 (O), saliva for montava Inf. i. 38 (F), bnttare for gittare Inf. xxii. 108 (E), and xxxiii. 68 (K), aspettar for sperar Inf. xxiv. 93 (L). ' I ' twice substitutes for scheggion the more familiar scoglio, Inf. xxi. 60, 89. The interchange of -vendetta and giustizia in Inf. xi. 90 and similar changes elsewhere (see note on Inf. xi. 90 in the discussion of selected passages) may be added. Also in Inf. viii. 63 -volgea has been deliberately changed to rodea in 'C', and to mordeva in 'L'. In Inf. xvi. 119, 120 'M' first alters -veggon to miran, and then miran to vede (sic). Some scribes habitually change Signor into Maestro or mio Duca, or vice -versa. Tenses again are not unfrequently changed to others supposed to be more suitable, e.g. passa and passb in Inf. iii. 127, Entrb and entra in Purg. xxvi. 41, con-verrb and convien in Inf. xix. 84 (L). Or again, we have the dialectic form sute substituted for state in the Tuscan MS. ' G ' in Inf. xiii. 39. There is a curious case at Inf. xxiii. 89 91, where in C and M the plural ' son ' is altered to , and 56 and 57, but I know of no others. (See further on this point Witte's ProL p. xiv.) xxiv PROLEGOMENA. A more extended acquaintance with MSS. would probably shew that they are not all 'peculiar', but they are all, I think, at least uncommon, and in that case they have a singular value and interest of another kind, viz. as establishing ^prima facie relationship between any two or more MSS. in which they are found: and thereby advancing us a step towards that most important, but at present rather distant, result, the classification of MSS. in 'families'. The common possession by two MSS. of only one or two striking and unusual variants has often proved in my experience very fruitful in suggesting further comparison by which sometimes an unmistakable relationship has been established. They are like what Bacon calls 'instances of the sign-post'. In this way it becomes im- portant to register even mere blunders ('errori madornali as Witte calls them), since their precise reproduction often affords, in the very proportion of their unreasonableness, convincing proof of connexion between the MSS. in which they are found. Thus then, to sum up, we have, as Francesco Palermo has tersely put it, "ignoranza, arbitrio, forza di dialetti," all in active operation upon the text for some two centuries. After that, it was subjected, with the invention of printing, to a fresh visitation of editors and grammatical purists. So that Vellutello, writing in 1544, complains that the text, both in MSS. and printed Editions, was 'incorrettissimo' ; and declares that Dante, could he return to life, would be the first to re- cognize the necessity for its thorough reconstruction. This being so, by what means can we hope to recover the original text, or at least to make some approach to it ? Surely by retracing, as far as possible, the various paths through which the errors have been introduced. But let us first clear the ground by shewing the futility of other methods that seem at first sight to offer a shorter, or more obvious means to this end, and by explaining what has already been attempted in these several directions. (i) It follows from what has been said that mere an- tiquity of a MS. is no guarantee of the purity of its text 27 . Of 9 It will be observed that the list of the best texts as selected by Dr Witte and as judged by my own tests, includes many of the isth century MSS., and some even beyond the middle of that century. PROLEGOMENA. XXV course there is, cceteris paribus, more chance of such purity as we draw nearer the fountain head, since every tributary stream has brought in some fresh elements of corruption. But, as we have already stated, a very short examination of the best and oldest MSS. in existence reveals readings which on critical principles are undoubtedly secondary and derivative. Therefore we cannot pin our faith to any one MS. now existing, however ancient ; they are all echoes, some more, some less, distinct ; not original voices. Our only hope therefore lies in a comparative study of MSS. At the same time, many editors have adopted the plan of reprinting the text of some one MS. which on various grounds they have exalted to preeminent authority, e.g. the Codice Bartoliniano of Udine (/"), by Viviani 28 ; the Codice Estense of Modena (M}, by Montfaucon, Parenti and others 29 ; the Codice di Santa Croce (A), to some extent by Dionisi 30 ; the Codice Vaticano, No. 3199 (B}, by Fantoni 31 , and so forth. I have said 'to some extent' when speaking of Dionisi, since while practically almost always falling back on the Codice di Santa Croce, he fully admits, as in the following passage (Anedd. v. p. 3), that no one MS. has preserved the true original, but that " they have all gone out of the way ", " lo cercava un Testo di buona mano e fedele che mi fosse d' esemplare alia ristampa della Divina Commedia da tutti i Letterati rico- nosciuta omai necessaria. In tutta Fiorenza non m' e riuscito di ritrovarlo, sicche degno d' alcuna scusa mi sembra ... il famoso Bastian de' Rossi, il quale volando (m' immagino) sovra le private e pubbliche Biblioteche, come la colomba di Noe sovra 1' acque, ne trovando ove fermar il piede, si ridusse a posarsi su d' una Stampa," and of this he adds, " Qual ella sia, parole non ci appulcro." (2) It would seem superfluous (but for the warning of 28 See Witte, Prol. pp. xxxvii xlii, and Foscolo, Discorso, pp. 16 &c. 115 &c. 29 See Witte, Prol. p. liii, and Foscolo, Discorso, p. 135. Parenti says of this MS., " che solo -vale per cento" 30 See Dion. Anedd. II., Witte, Prol. p. xxxiv (init.'). I venture to think that, in spite of its acknowledged excellence, Witte himself some- times adopts an inferior reading in deference to it. 31 In his Ed. of 1820. See Batines, I. p. 148, n. p. 167. D. I. d XXVI PROLEGOMENA. actual experience) to insist that the comparative study of MSS. does not aim at ascertaining the mere number of MSS. by which one reading or another is supported. Mere numerical support is as fallacious a test of merit in this as in any other subject, and happily in this case no one has any interest in believing, or affecting to believe, that it is otherwise. It is obvious that among MSS. (and not in this case alone) some are mere echoes of echoes, two or three or more "removes from the truth" (as Plato says), or from original and inde- pendent judgment, and so their support adds nothing in the way of evidence. This is very well illustrated by an instructive (but unhappily purely imaginary) diagram of the genealogical relation of MSS. constructed by Scartazzini in No. II. of his excellent little Manuali Danteschi published in Hoepli's Popular Manuals. Again, (and here the same diagram may be referred to), it is obvious that it is a mere accident, whether an early or original MS. has been copied frequently or seldom, or not at all. Some have many, some few, some perhaps no descendants. An interesting indication of this may sometimes be found in the rareness of a reading for which very early authority can be adduced. It is evident in that case that the MS. in which it is found has seldom or never been copied. The pro- cess of counting heads is therefore quite futile. This is clearly and preeminently a case ' non numerare sed appendere' If a posteriori proof were desired, we could point to many readings undoubtedly erroneous which have a large numerical majority of MSS. on their side. The numerical support, so to speak, of a false reading depends in fact on a mere com- bination of accidents: viz. (i) whether the error happened to be introduced by an earlier or a later copyist ; and (2) whether the MS. containing it happened to have been copied often or seldom. Many instances of this will be found among the test-passages here collated. Of course I fully admit that in many of such passages my judgment may be in error, still I hope I may say without presumption 'OVK ev\oyov [e/*e] Sia- fiapravetv TOIV KI>K\O> davlffrcu\ A similar inconsistency, or intentional exception, is quoted from Virgil in the case of Dido. For Dido, who should have been among the Suicides (jEn. vi. 434, &c.) is placed in the 'myrtea silva' with those 'quos durus amor crudeli tabe peredit ' (lines 442 4, 450). It is curious to note that Dante followed Virgil in this exceptional treatment of Dido (see Inj. v. 85), though he definitely mentions the fact of her suicide in the same Canto at 1. 61. Compare also Inf. v. 69 with jn. vi. 442. (7) Frederic II. of Sicily and James of Arragon are censured in Purg. vii. 118 &c., and Frederic especially again in Conv. iv. 6, and in the curious passage about ' lettere mozze* in Par. xix. 134, as well as in De Vulg. Eloq. i. 12. With this the expression onor di Cicilia e d> Aragona in Purg. iii. 116 applied to Frederic and James is said to be inconsistent. It should however be remembered that in these words Manfred, their grandfather, is describing his own kin. It may perhaps be added that, as in the case of Clement V., Dante may have held different opinions at different times, for Frederic had made a 'gran rifiuto ' (see Plumptre, Introd. p. 116) after Henry VII.'s death. It seems that there was a time when Dante had thought of dedicating the Paradiso to Frederic, if we may trust the letter attri- buted to Fra Ilario. This at any rate, considering Dante's almost uniform denunciation of Frederic in his existing writings, and in the Paradiso in particular, is an extremely unlikely statement for a forger to have invented. (I observe that Lubin (Vita e Opcrc, p. 79) unhesi- PROLEGOMENA. XXXIX obvious or important we may mention the Bible (of course in the Vulgate Version), Aristotle (also in a Latin Translation) 41 , Virgil, Ovid, Lucan, and Boethius. This however is so large, and as I venture to think, so highly interesting a subject of investigation, that I have followed it out in an Appendix, not to interrupt the present discussion by too long a digression. (5) Akin to this is the consideration of the relation which a reading bears in respect of truth or falsehood, accuracy or inaccuracy, to questions of historical fact, which may be in- volved in a passage. Here it must be remembered we are on highly dangerous ground, and much caution and judgment must be exercised in applying our knowledge of what the fact alluded to should be in truth, in order to maintain the reading which embodies that fact correctly. For of course if the author, being fallible, and also having access to imperfect sources of information, should have made a slip, the tempta- tion to a copyist to set him right would be very obvious, and tatingly maintains the genuineness of this interesting letter, confidently dating it as written in 1308.) As to Clement V. see Epist. v. (Ed. Frati- celli) ad fat, We must carefully distinguish between these so-called 'discrepancies', and a few cases of apparently deliberate changes of opinion on the part of Dante. I know of the following such " Retractations" : 1 i ) as between Div. Comm. and Convito : (a) Par. ii. 61 and Conv. II. 14. (The theory of Lunar Spots.) (ff) Par. xxviii. 135 and Conv. II. 6. (The Orders of Angels.) (2) as between Div. Comm. and De Vulg. Eloq. : (a) Par. xxvi. 124 and De Vulg. Eloq. I. 6 fin. (The language spoken by Adam. Note also the strange theory broached ib. c. 7 fin.} (b) Par. xxvi. 134 and De Vulg. Eloq. i. 4. (The primitive name of God ; unless indeed (which is improbable) we were to read El and not 7 or J in Par. xxvi. 134.) (3) as between Convito and De Man.: Conv. iv. 7 and De Man. ii. 3. (On Ancestry as a Condition of ' Nobilta', the former admitting one ground only, and the latter two, ^ propria scilicet et majorum.') (4) as between Convito and Vita Nuova : The descriptions of the 'Donna Gentile' in V.N. ( 39 and 40) and in Conv. ii. 2 and 13 seem quite irreconcilable. (See this discussed by Eliot Norton 'The New Life' App. C.) 41 The question of the precise Latin translation in which Aristotle was read by Dante is also discussed in an Appendix. xl PROLEGOMENA. so the reading which corresponded with the truth of fact would be the secondary and false one. We must judge of such cases as they arise, and at any rate two considerations should always be borne in mind : (i) whether the facts were too notorious and well ascertained for ignorance or forget- fulness on the author's part to be conceivable, or whether they were such that error on his part may not have been improbable. One instance of the application of this principle will be found in the discussion of the celebrated variants re giovane and re Giovanni in Inf. xxviii. 135. (2) We must carefully consider what were likely to be the actual sources of the author's own information, and we must not expect him to have avoided errors of history, any more than those of physical science, the correction of which, however obvious and glaring they may now seem, has been due to the re- searches of later times. These considerations will be found to bear on the comments sometimes made on, and cor- rections suggested (as for instance by Dr Barlow) in Inf. xi. 9, A nastasio Papa guardo. We may now briefly indicate how, speaking generally, the nine classes or causes of corruption already enumerated may be counteracted by the application of these methods. The first two (Dialectic varieties and ' Personal Equation ') are comparatively unimportant, since, however audacious and unjustifiable such changes may be, they are frequently merely orthographical, and are at any rate easily detected and elimi- nated. Classes (3) and (4) (Changes for euphony or to avoid unusual words, inflexions, &c.) are to be remedied, as will have been already perceived, by the application of the prin- ciple Difficilior lectio potior. Classes (5) and (6) (Changes from involved construction, or misunderstood sense) by a modified application of the same principle, i.e. by preferring a reading which, though prima facie 'more difficult,' is yet found on reflection (which the copyists seldom bestowed) really to suit the deeper sense, or the whole context, of the passage more easily. The reading to be preferred in such cases is in fact that which is only superficially, but not really, ' dirficilior.' The miscellaneous classes (7) (8) and (9) (Correc- tions in the interests of orthodoxy, accidental or clerical errors, the incorporation of glosses, &c.),must be dealt with in different PROLEGOMENA. xli ways suggested by the different circumstances of the passages in question. Some of the minor critical aids above described occasionally apply, and the a posteriori evidence of MSS. is frequently available in these cases. It occurs to me to suggest, in connexion with the genealo- gical relationships of MSS., another interesting question as to variants, the materials for investigating which are as yet too imperfect. I mean the relative date of the introduction of particular variants. It should be possible, at any rate in reference to some of those which occur rarely, to fix a limit before which they are not found. Thus we should obtain some test- passages in a chronological sense, which when sufficiently established might serve as presumptive or con- firmatory evidence of the date of a MS. I am inclined to suspect, for example, that Rende for Vede in Inf. iii. 1 14, and probably also oppressura for pressura in Purg. vi. 109, are not found in any very early MSS. But at present at any rate it would be most hazardous to dogmatize from want of sufficient evidence, since the assertion would imply the proverbially difficult task of ' proving a negative.' I merely throw it out as a theoretically possible, and if possible, certainly an in- teresting, subject of investigation. Of course however in the positive application of the argument we are on safe ground. I mean, if a reading, whether common or rare, is found in a MS. of early date, the fact speaks for itself, as to the age of the variant in question, whatever may be its value on other grounds. Numerous instances will be found (as has been already indicated, p. vi) in the test-passages which I have collected, of certainly false readings so ancient as to be nearly contemporary with Dante himself. It is further important to observe that a very large number of variants are practically incapable of determination by the evidence of MSS. Differences of orthography or dialect often obscure the very distinctions between which we have to decide. Hence the reading of a considerable number of passages, which would otherwise serve as useful and interesting test- passages must be set aside as 'indeterminate.' I will only mention one or two obvious illustrations; such as (i) where the difference in the reading depends on the doubling or not of a consonant, e.g. saria and sarria in Purg. vii. 5 1 or corregger D. I. C xlii PROLEGOMENA. and coreggier in Par. xi. 138: this detail of spelling being habitually neglected in the numerous MSS. of Lombardo- Venetian origin. (2) Owing to the differing practice in respect of elision, we can never be sure whether such a word as Che (for example) in a MS. stands for Che simply or Che i or even Che ei. So e, de, a &c. may or may not have been meant to carry an elided i. (3) The frequent neglect of the divi- sions between words, combined with the other considerations just mentioned, obscures the evidence of MSS. in many cases: e.g. Senzarostarsi may stand either for Senza rostarsi or Sens' arrostarsi in Inf. xv. 39. So in Par. v. 9 the MSS. help us but little between the readings, Che vista sola, and Che vi sta sola. Another good instance will be found in the collations, at Inf. xxxii. 34. Many other instances could easily be added, but as this subject will be found more fully discussed in the Prefatory Note which follows, I need not pursue the subject further here. I conclude by a brief recapitulation and summary of the results, which seem to me to be established or confirmed by the collations here published. These results, I must admit, (as will have been already anticipated) partake more of a negative character than I at one time hoped 42 . Still even negative results are by no means equivalent to no results at all. As Prof. Earle lately wrote in reference to his researches on Beowulf: "I have often gone a long way on the wrong 42 I certainly had hoped to conclude, if not with a genealogical classi- fication of many of the MSS. examined, at least with some tabular or statistical view of their comparative value and trustworthiness. After very much labour and analysis of the results of the readings here regis- tered, I had to abandon the task as hopeless. The most I feel able to do, and that with much diffidence, is to offer a tentative list of those MSS. which appear to me on the whole, and as judged by these tests, to have the purest and best texts, and so to constitute, as Dr Witte has done, a kind of provisional ' first class ' of MSS. Dr Witte (Proleg. p. Ixxiv) confesses himself to have been similarly disappointed and baffled of the positive results in the way of genealogical classification for which he had hoped. And Professor Mussafia remarks in reference to his own work in the same field (already referred to, p. xxix) Per ora rac- cogliamo i fatti ; in grande numero, con grande esattezza; non ci affret- tiamo a trarne conclusioni, che forse ben tosto saremmo costretti a ricognoscere fallaci : ufficio umile ma piu sicuro, e 1' unico che per ora sia possibile (Introd. p. 5). PROLEGOMENA. xliii track. But such failures are not merely waste. If nothing else, they tend to the accumulation of a store of negative evidence and they reduce the area of future effort." The results then may be summarized as follows : (1) Corruption of the text began at a point anterior to any existing MS. or commentary, and probably it took place in the first copies of the autograph itself, which seems to have very quickly disappeared. Dr Witte has arrived at a similar conclusion. (2) So far as my collations go (and in reference to some of the selected passages they extend to fully one-third, in some cases to not very far short of half, of all the known MSS.), no existing MS., or group of MSS., stands out as possess- ing preeminent or indisputable authority 43 . (3) It is scarcely worth while to repeat that the mere numerical support of MSS. (even when there is no evidence of any special relationship between them) is of very little value. This is at any rate confirmed by these collations, since they shew that some undoubtedly false readings are found in a very large majority of MSS., while some almost certainly true readings have very slender MS. support. (4) A more positive, and to some extent a more original result, is the demonstration (which I have already referred to) that the character and relationships of a MS. differ in different parts of the poem. (See p. xxxiii.) This may have been noticed before, though I am not aware of it. I feel sure however that it has not before been so clearly demonstrated, and that consequently sufficient weight has not been assigned to it. The way in which it vitiates some of the conclusions of so careful and diligent a scholar as Dr Witte himself has been already pointed out. (See p. xxxiii, xxxiv, and note 38.) This fact goes far to discourage the hope of anything like an ' albero genealogico ' of MSS. ever being constructed. (5) At the same time, it seems to be proved possible to 43 It is perhaps worth while observing that the evidence of so large a proportion of MSS., both on this and other points, may fairly be taken as probably representative of their general verdict. There is no reason to suppose for instance that the proportionate amount of support given to two disputed readings by say 200 MSS. taken at random, would be materially altered by an examination of 300 more. 2 xliv PROLEGOMENA. construct certain well-marked groups or families of MSS., such as that described by Witte (in his Proleg. and Dante Forsch- ungeri) as 'the Sienese,' and derived by him from his examination of the statistics collected in reference to Inf. Canto III. only. As a result of these collations, I have also traced distinctly another such group, which I have described as the ' Vatican ' family, since its prototype, or at least best known example, is the celebrated MS. (No. 3199) in the Vatican Library (Witte's B), which is sometimes, though on insufficient grounds, supposed to be in the handwriting of Boccaccio. (See Witte, Proleg. p. Ixxvii.) The evidence on which this ' family ' rests, and its characteristic features, will be found described later, in an Appendix 44 . 6. In the mass of facts and references here brought together there will I trust in any case be found the 'raw materials ' by means of which many questions about the text, and the relationships and families of MSS., may be readily brought to the test of facts, and either followed up or abandoned. I have myself here found the means of thus testing and rejecting many tentative hypotheses, and also of 44 The phenomena of these ' families ' or groups of MSS. may be variously explained. It may mean little more than that some one MS. (or some of its descendants) has chanced to be more often copied than others, though the fact itself would be some presumptive evidence of a high value attached to such MSS. It is also just possible that these distinct types of text, or some of them, may represent a revision or re- cension of the text by the author himself, such as some students are on other grounds inclined to believe in, and such as is intrinsically likely enough to have occurred. (See above, p. xxii.) Again, it is possible that some early copyist may have introduced these changes into the lezione volgata, either from some good text comparatively untampered with, or by the exercise of his own judgment. The character of the variations themselves would probably throw some light on the probability of these alternatives, and it becomes therefore a matter of great importance to determine whether the typical readings which distinguish such families seem to bear generally a primary or secondary character, as compared with the lezione volgata in the same passages. My strong impression is that whenever a MS. is found either with the 'Sienese' or ' Vatican ' pecu- liarities, but more particularly in the latter case, it has nearly always a good text, though I am not prepared to maintain that the special readings characteristic of these families are in themselves uniformly or predomi- nantly good. On the possibility of the coexistence of a good foundation text with even numerous errors, see above, p. xxii. PROLEGOMENA. xlv establishing the relationship of many pairs, and small groups of MSS., and of at least one distinct 'family.' If 'families' are ever to be traced, such must be the first steps in the process. The MSS. here examined are, I believe, sufficiently representative in quantity, quality, date, place of origin, &c., and the passages selected (quite apart from any judgment that may be formed as to my own critical conclusions re- specting many of them) are sufficiently various and testing in character to enable any student to work out many textual problems without the labour of visiting numerous Libraries in various countries. The plan too is one which could easily be extended, if necessary. For instance at Florence, where there are more than 200 MSS., of which I have only examined about one-third, there is an unexplored mine out of which the evidence on any special passage or passages could be largely added to with comparatively little trouble 43 . At any rate (and this remark applies especially to the complete col- lation of seventeen MSS. throughout the Inferno) if MSS. are worth preserving in our Libraries, it cannot be useless by pub- lishing collations to place a series of them at once for practical purposes in the hands of students, and so to make their con- tents readily and generally accessible. The results here gathered up may seem to some both slight and negative ; but if a building is to be secure and lasting, much work must be spent upon its foundations, and though these are not seen afterwards, yet the function they perform is none the less important and indispensable. And if I may be allowed in taking leave of this part of the subject to add a few words as to my own personal work in the matter, I will venture to say thus much. I am deeply convinced that, be the tangible and positive results great or 4 " Colomb do Batincs registers as many as 190, but this number includes several private libraries since dispersed, which would certainly reduce the number by 10 or 15. On the other hand, the Biblioteca Lau- renziana has lately been enriched by the addition of the magnificent col- lection of Lord Ashburnham, 27 in number of the Commcdia alone. Through the very great kindness and courtesy of Lord Ashburnham, which I cannot too gratefully acknowledge, I was able to examine and consult these MSS. on various occasions before they unfortunately left this country for one more able to appreciate their value. xlvi PROLEGOMENA. small, no labour bestowed on the study or elucidation of this, perhaps, the greatest work of human genius in any language, can be felt to be in vain by one who has expended it. All will remember, Volto...in sulla favola di Esopo Lo-.-pensier 46 , how the farmer who turned up the soil of his vineyard in search for hidden treasure found that, although disappointed of the expected treasure, he was more than compensated for his time and labour by the increased productiveness of his land. I know of no author in whose study we can feel more confidently assured of a similar result. I know of none with whom it so often happens, that according to his own beautiful saying, in the Convito (ii. 13), "Esser suole che 1' uomo va cercando argento, e fuori della intenzione trova orb." Apart from this personal profit and pleasure however, I cannot but again express a hope, that beyond the comparatively meagre positive results I have been able to establish myself, these materials may prove to be of value and service in abbre- viating the labours of future scholars. So that I may at least aspire to the usefulness of one Che porta il lume dietro, e se non giova, Ma dopo se fa le persone dotte 47 . (Purg. xxii. 68, 9.) 46 Inf. xxiii. 4. 47 Scartazzini, I think, observes somewhere that these beautiful, and in their context, pathetic lines were possibly suggested by S. Aug. Conf. iv. 16, " Dorsum enim habebam ad lumen, et ad ea quse illuminantur faciem ; unde ipsa facies mea, qua illuminata cernebam, non illuminabatur." In his note on this passage he gives a more recent model and a passage of much closer resemblance, which Dante may have had in his mind, from Messer Polo da Reggio : Si come quel che porta la lumiera La notte quando passa per la via, Alluma assai piu gente della spera, Che se medesmo, che 1' ha in balia. PREFATORY NOTE ON DIFFICULTIES IN THE INTERPRETATION OF MSS. AND ' INDETERMINATE ' READINGS. IT is desirable to explain the principle on which these collations have been conducted, so that any one employing them may know within what limits they are to be relied on. My aim having been to place as it were in the hand of students, for all practical purposes, the actual record of the several MSS. collated, I have taken as a standard text Witte's 8vo. edition 1 , Berlin, 1862, and recorded, with the exceptions now to be explained, any variations from that text. So far as this is done, any person, by a comparison of the printed text with the variants recorded below, can, so to speak, re- construct the MS. and write out for himself its text on any given line. It is necessary therefore to explain precisely what are the exceptions which I have referred to, what are the variations which have generally speaking not been re- corded. They fall under two main heads, (I) those which seemed to be not worth recording ; and (II) those which could not be recorded with any certainty, because the evidence of the MSS. was inconclusive. I. As to the former class, the question as to what is, or is not, worth recording has often been extremely difficult to determine. It might seem at first sight that obvious 1 See Witte, Proleg. p. Ixxx, note, for an explanation of the relation between the 4to. and 8vo. Editions. The following are the differences which I have noted : 4to. 8vo. Inf. x. 88 mosso scosso 117 lui stava lui si stava xiv. 122 del dal xviii. 79 del dal xxvi. 137 della dalla xxvii. 21 istra ista xxviii. 26 curata corata. xlviii PREFATORY NOTE. blunders (' errori madornali' as Witte calls them) might safely be neglected. But, as I have already indicated in the Prolegomena, the servile repetition of such blunders by copy- ists often affords conclusive evidence of the relationship of MSS. The same may sometimes be said of strange varia- tions in orthography, though, as a general rule, variations in orthography are the most obvious class of differences not worth recording. Indeed it would be impossible to record them at all fully, for it would often amount almost to a transcription of the MSS., the orthography being as variable as the handwriting itself. Dr Witte mentions (Proleg. p. Ixxxi) that he soon abandoned this attempt as hopeless. Sometimes however the difference, though one of ortho- graphy, may involve a principle, as for instance the adop- tion of a more archaic, and especially a Latinized form of spelling. There is no doubt that in the language of Dante, not only were actual Latin words and phrases very readily admitted (of which the Divina Commedia affords numerous examples), but the language itself had been much less modi- fied than was afterwards the case, by assimilation of letters, or changes of inflexion, from the actual Latin forms. I have already pointed out in the Prolegomena how the frequency of Latin and unassimilated forms, especially in proper names, affords evidence of a text unmodernized in form, and there- fore, presumably, untampered with in other respects. (See Proleg. p. xii, and especially Dionisi Anedd. II. p. 102, there quoted.) I have generally, therefore, but not invariably, recorded the different orthography of proper names, though it often exhibits the strangest distortions and the most hopeless ignorance on the part of the copyists. In any case I have done so, if I thought it likely in any way to afford evidence either of the archaic character of the text, or of possible rela- tionships between MSS. In the case of copying by the eye, the peculiarities of the original in a word or form in itself unintel- ligible would be likely to be faithfully preserved, the scribe throwing the responsibility for the result on the original. In that case we should have evidence of relationship. On the other hand, in the case of oral copying, strange results might be expected which would at any rate serve as evidence of such a method of transmission, and possibly of relationship PREFATORY NOTE. xlix also. Sometimes, moreover, variations of orthography may suggest the missing link between two readings which at first sight seem unconnected and unaccountable. Some instances of this will be found given later on, under the head of 'inde- terminate' readings. Another doubtful case of ' negligible variants,' not exactly a matter of orthography, is that of the slight variations of the form of pronouns such as el, ei, egli ; quei, quello, qnei ; il, el, lo, &c. These I have not generally recorded, though sometimes it is not safe to neglect even these, since some of them, e.g. the change of quei into quei, has often been a deli- berate alteration due to the misunderstanding of the archaic and obsolete form quei used for the singular number. (So also altri for altro, very commonly.) This ignorance has some- times led to other conjectural tamperings with the text, as in Inf. iv. 95. Occasionally a variation slight or trivial in itself has been recorded when it seemed rare, and only appeared in one or two of the MSS. examined, and I may say generally that I have thought it best, with Witte and Mussafia (Introd. p. 6), when in doubt, to err rather on the side of excess than of defect, especially in cases where small changes, which I should ordinarily have passed over, have been noted by Dr Witte in reference to some of his standard MSS. ABCD, since I have aimed at embodying his marginal variants of these MSS. in my collations. There are, however, some of the changes which he has occasionally recorded (I do not know whether even he has done it invariably) so extremely slight and trivial, and apparently devoid of any interest or significance, being purely orthographical, and not in any true sense ' variants,' that I have ventured to neglect them. Lest, however, I should be thought guilty of carelessness or inac- curacy in not recording a variant in some of Witte's MSS. which he has registered, it may be well to mention the chief cases in which I have adopted a rule differing from his. The following, then, are some of the trifling (as I think) and rather common variations recorded by Witte, which I have as a rule intentionally disregarded 2 : ove and dove ; It may happen that occasionally these are noted. This is due 1 PREFATORY NOTE. volvere and volgere (vii. 29, 96, &c.) ; sopra and sovra ; sapere and savere (vii. 73, &c.) ; furon and furono ; avrebber and avrebbon (and similar forms in other verbs) ; ainendue, ain- bidue, ambedne (or -dui), &c. ; tata? quanti, tutti quanti, tutti e quanti ; dietro, diretro, drieto, direto, &c. ; ristare and restare, &c. (x. 24, 74); dfotf, dua, duo, &c. (xii. 84, &c.) ; aire, aria, aura, &c. ; el and ', &c. II. We turn now to the more important class of variants that have had to be neglected, though often of much interest in themselves, on account of the indeterminateness of the reading in the MSS., sometimes even when no doubt could exist as to the actual letters presented by them. I proceed therefore to enumerate the principal causes of such ambiguity which I have myself encountered. Many of these again are mere questions of orthography, but unfortunately they often involve, and so serve to disguise, just the very differences of reading between which we have to decide. (a) I will first mention the varying practice of copyists in respect of the fusion or elision of final letters. This occurs most commonly in reference to the letter i when representing the nominative plural of the article. MSS. of course do not use the apostrophe. Consequently a word like tra may represent tra\ and so it may stand either for tra or trai. So Che may stand for Che' , and that either for Che i or Ch' ei. (See notes in the collations on ix. 133, xvi. 32, xix. 34, and espe- cially xxiii. 89, and many other places.) So in i. 90, e polsi may stand for e i polsi ; in ii. 20, Che fu may represent CK e' fit, and in iv. 6^,perche dicessi may mean perch ei dicesse*. So de may stand for de , i.e. dei, or as an archaic form ofdi merely (e.g. Inf. i. 17, vi. 2). Thus too a may be for a' or at, and so a confusion may arise between the second or third person present of the verb avere, and the preposition a with or with- out the article. A remarkable instance of this confusion occurs in Inf. xxxii. 1 36, where most MSS. have the blundering sometimes to some special reason seeming to exist for doing so ; and sometimes probably to my neglecting to erase them, since in the actual process of collation a large number of such small variations were noted which have since been deliberately omitted. 3 In this way the copula V has probably been lost in many MSS. at Inf. xix. 3, xxx. 115, and Purg. viii. 94. PREFATORY NOTE. li reading at (or hai) ragion for a ragion*. This ambiguity is especially common where the same letter would occur twice in the full and unelided form, e.g. tutti may stand for tutti i ; Gente avara for Gente e avara (Inf. xv. 68) ; Quali fioretti for Quali i fioretti (Inf. ii. 127); contra for contra (or -fro) a, &c. (b) The article is often written in the form El (as in ' I ' habitually) so that El may stand either for // or E il ; e.g. Inf. xxv. 66, 74, &c. (See note pn the latter passage.) (c) The constant tendency to fusion or irregular division of words is a source of endless difficulty. We meet for instance with such a form as perokivegio for perb eft io veggio at Inf. xv. 116 in ' C ', and though in this particular instance the meaning is clear enough, this is often not the case, e.g. Senzarostarsi may stand for Sens' arrostarsi, or Senza rostarsi in Inf. xv. 39 ; colle or cholle may be for colle or for ctt ho le in xxx. 81. The following readings in Par. xxxii. 26 must be considered altogether indeterminate on MS. evidence : devoti, de j voti, di voti. The interchange of dio, idio and iddio is perhaps in any case scarcely worth noticing, but certainly not when the previous word ends in i as in Inf. vii. 19, 87, &c. In Purg. iv. 138, some Edd. give ed alia riva, but though e dalla riva is doubtless right, the MSS. would not help us much to decide. Nor again would they do so in reference to the true reading convolto, as contrasted with con volto, in Inf. xxi. 46 ; or between a dir and ad ir in Purg. xx. 119. (d) No stress can be laid on MS. evidence where the difference of reading depends on the doubling or not doubling of a consonant. Indifference to this point of orthography is a common feature of the Lombardo-Venetian dialects, and as 4 The same difficulty is met with in MSS. of the New Testament, and has often been overlooked by ' worshippers of the letter,' who sometimes would have us adopt a reading giving little or no sense on the strength of an almost infinitesimal difference of spelling in a MS. three or four cen- turies later than the autograph. Readings, for example, depending on the mere difference of o and 00, t and et, &c. are often practically inde- terminate. See inter al. Field's Otium Norvicense Pars Tertia, pp. 148-9, "It should be borne in mind that IAE and EIAE are rather different spellings than different readings," and he proceeds to give instances where one is written when the other is certainly intended. If this be so, the distinction cannot be insisted on cofite que coiite elsewhere. Hi PREFATORY NOTE. far as my observation extends, a very large number of MSS. exhibit this and other dialectic peculiarities of those pro- vinces. If in conjunction with this we remember that some scribes think that i must be inserted regularly after c and g to make them soft (just as others similarly insert h to make them hard}*, we shall see that the interesting variants in Par. xi. 138, coreggier and corregger become absolutely inde- terminate. Even the distinction between the different readings re giovane and re Giovanni in Inf. xxviii. 135 would according to the practice of some scribes become indistinguishable, when we further take into consideration the frequent carelessness as to terminal e and i (presently to be mentioned) and the close resemblance of these two letters in some MSS. (e) The neglect of the superposed mark of abbrevia- tion ( ~ ), indicating the omission of some letter or letters (most commonly of ), and indeed sometimes its meaningless insertion, is another very common source of confusion 6 . Hence inter alia the distinction between the third person singular and plural of verbs is often obliterated, e.g. avea and avcan, which latter would be commonly written avea, but the super- posed mark has sometimes been forgotten or has faded. An important case occurs at Ptirg. xxii. 5. Another common result is the confusion of modo and mondo; some MSS. writing modo when we feel sure they meant modo, i. e. mondo, and others modo where the sense requires modo. Without going so far as to describe this as a regular case of ' indeterminate ' reading, we cannot at any rate feel quite sure that a given MS. always intends the actual word which it presents in such cases, for these words are interchanged so often that we cannot but suppose accident to have originally played some part in the matter. A well-known and very interesting case 6 Other curious irregularities of orthography in respect of these letters are noted at Inf. xxx. 148. 6 In a MS. in the Taylor Institution Library at Oxford of Cecco d' Ascoli's Acerba, the chapter ominously headed ' Qui in questo capitolo deride Dante dicendo che non si deve scriver fabule ' commences thus : Qui non si canta al modo delle rane Qui non si canta al modo del poeta Che finge imaginando cose vane, &c. where it is quite clear that modo is in no sense a variant, but a sheer blunder for modo. This confusion is especially common in the Bodleian MS. 'I '. PREFATORY NOTE. . Hii is that of Inf. v. 102. The following are other instances that I have noticed of this interchange, /;//. iv. 75, x. 64, xi. 61, xviii. 30, xxviii. 21, Purg. xv. 1 8, Par. ix. 108. The slightly different word moto is interchanged with modo or mondo, in Inf. ii. 60 and Par. xxvii. 106. The interesting variants Dante and Zte fe in Par. xxvi. 104, are another case in point. I next add a few cases of letters which are often confused from their close resemblance in some handwritings, and first (/) final i and e (i's not being dotted). I am not speak- ing of cases where the interchange merely represents archaic usage, as in the singular person-terminations of the conj. mood of verbs 7 , but of cases like gente, parte, &c., which very often stand tor genti, parti, &c., e.g. Inf. xxiv. 144 and xxvi. 90, (where see notes), or when the second person singular and plural of verbs are thus confused. This for instance tends to obscure the evidence in an interesting passage suggested to me in a letter by Dr Witte for examination as a test- passage, viz. Inf. v. 1 20, where he was strongly of opinion that conoscesti and not conosceste was the true reading, the latter being as he thought an obvious lectio facilior* . I found, how- ever, after examining a large number of MSS. that the differ- ence was too slight and uncertain to lay much stress upon it. A combination of the sources of uncertainty mentioned under (a) (e) and (/) will shew how very slight and unreliable is the difference in MSS. between the various readings in Inf. xxii. 101 : SI che (or ch' ei) non teman (i.e. tema). Si che (or ch' ei) non tema. Si ch' i' (or io) non tema. Compare come, com' e\ com' el t com' io, parlava in Purg. viii. 94. (g) Another common interchange is that of final m and n. Very often it is clear that the substitution of one for the other is accidental, and there is no doubt from the context as to the word intended. But as this tends to disguise the dis- 1 See on this Scartazzini's note on Purg. xx. 100, and Nannucci, Verbi Ital. p. 284. 8 We may compare the interchange of sing, and plur. due to a similar consideration in Inf. xxiii. 89 (see note in collations). liv PREFATORY NOTE. tinction between the first and third persons plural of verbs, and as in some cases either person may be possible, no stress can be laid on the evidence of some MSS., at any rate when this interchange is frequent (as e.g. in GDI), to support either reading. A case occurs at Purg. xxii. 5 (already re- ferred to) where we have three possible readings, all of which occur in different editions, avea, aveam and avean. Another important case occurs in Purg. xxi. 19. (See also Inf. xxii. 70, and note on that passage in the collations.) (Ji) Again c and t are in many MSS. nearly undistinguish- able, and in some cases, we may say the difference between c, e, t, and even i (when it is remembered that i's are not gene- rally dotted in MSS.) becomes almost evanescent. Hence the false reading centri for cencri in Inf. xxiv. 87, which is found in many MSS. and repeated in the Aldine and other Editions. This ambiguity may perhaps have facilitated the interchange of zenit (i.e. cinit or cienif) with tiene in Par. xxix. 4. (i) One of the commonest and most puzzling sources of ' indeterminateness ' in MSS. is the similarity of up and down strokes, hence n and u (or v) are sometimes extremely difficult to distinguish, and this combined with the absence of clear divisions between letters (and even words) and also of the dots to i's, involves other letters and combinations of letters in the same ambiguity, such as m, in, ni, iu, ui, iv and vi (see notes on Inf. xxiii. 78, 143), all of which might be disguised in such a common form as in. When we have (as often) four of such strokes together instead of three thus mi (or even five sometimes), the number of possible combinations that they may represent is much increased. To take a few out of almost numberless instances of this difficulty. We are frequently unable to decide whether a MS. means to read nostro or vostro, e.g. Inf. xxvi. 115, xxviii. 5, xxxiv. 124, Purg. xiv. 126; or again nosco or vosco in 1. 105 of the last- named Canto. In this way the words vinto and unito, and even giunto in some MSS. (since some scribes regularly omit the initial g in such words, and would write iunto 9 }, may be undistinguishable. Hence I cannot but regard the inter- 9 See note in the collations at Inf. xxv. 67. This probably accounts for the frequent occurrence of ire andgtre as variants : e.g. Inf. xxvi. 141, xxxi. 124, 141, &c. PREFATORY NOTE. Iv esting variants in Par. xix. 141 as indeterminate, since atusto may stand for a visto, or aiustb, i.e. aggiustb. So again tttse may have been taken for vise (visse), iuse, or iuse, i.e. giunse, and by this means readings apparently so different as Visse and Gitinse in Purg. xxi. 101 may possibly have arisen from an accidental confusion, since the mark over iuse might stand either for an omitted n, or for another s. This might also be the case with vinse and giunse in Inf. v. 72, and with avvinse and aggiunse in Inf. xvii. 96, since by a combination of the different sources of confusion which we have mentioned either might arise from an original atitnse. As I remarked about modo and mondo, it is scarcely likely that these interchanges would be so common unless they had been aided by, or had arisen from, the accidental resemblance. The curious and widely spread false reading lieve in Inf. xxxiii. 26 may thus easily have arisen accidentally from lune. (/) Lastly, I will mention a few pairs of often-recurring words which, from their similarity no doubt, are very often confused or interchanged. Such are the following : alto and altro (very commonly in ' B '), e.g. Inf. i. 9, &c. (see note on Inf. xxiv. 135), and similarly mente and mentre (especially in ' C ') ; also effetti and affetti in several places. Note especially the interesting passage in Purg. xi. 3, where the disputed reading is recognized by the author of the Ott. Comm. as early as 1333. So also effettnoso and affettuoso in Inf. v. 87. (See note on this passage, where numerous instances of this con- fusion are collected.) Again, cagion and ragion are almost habitually confused, this interchange being facilitated by the similarity of sense as well as of form (e.g. Purg. xiii. 20, xviii. 65, xxii. 30, xxxi. 90, Par. vii. 101, xi. 21, &c.). Another pair of words constantly confused is caro (or chard) and chiaro and their derivatives, as in the disputed line Par. xxv. 33. Add also Inf. xxxi. 28, Purg. i. 75, and Par. xxii. 146 (though of course here chiaro must be right, and caro is a mere blunder). The various ambiguities and uncertainties here enumer- ated have obliged me to abandon several interesting test- passages as ' indeterminate,' on which I had collected a large number of references to MSS V and concerning which in many cases the evidence of MSS. would have been peculiarly wel- come, if it could have been given with no " uncertain sound." Ivi PREFATORY NOTE. I will conclude with an amusing instance of ambiguity ingeniously constructed by Dionisi Anedd. V. cap. iv. nidi iiiiione titida, dione me dea cJieintoce canta ttiente crudele; which may be read Indi vivo v e Mida : Dione vi e dea, cJu ' voce canta niente crudele ; or thus : Vidi Junone nuda : Di, ov e Medea, che nuoce cautamente crudele*. 1 Dionisi also gives some curious instances in which Commentators and others have actually been misled. In Par. xvii. 32 s'invescava has been evidently read si muscava by Jacopo della Latia, since he makes these words an occasion for a disquisition on Beelzebub as ' deus muscarum'! Again he states that in Par. xxii. 151 Vaiuola has been read la mola by an anonymous Commentator, who explains it of the circle of the Zodiac, and then proceeds to justify the application of the term mola to the Zodiac. I N FERNO CANTO PRIMO Nel mezzo del cammin di nostra vita Mi ritrovai per una selva oscura, Che la diritta via era smarrita. 3 Eh quanto a dir qual era e cosa dura Questa selva selvaggia aspra e forte, Che nel pensier rinnuova la paura ! 6 Tanto e amara, che poco e piu morte : Ma per trattar del ben ch' i' vi trovai, Dir6 dell' altre cose, ch' io v' ho scorte. 9 I' non so ben ridir com' io v' entrai ; Tant' era pien di sonno in su quel punto, 1 de chamino L. 2 in una I. scura EKP. 3 me era A. avea sm. P. 4 Ha CQ. Ay AE. Ah Z (marg.}. E or Et BDGHIKLMOPZ^^Ca. O quanto F. al dire I. quell' BDFL. ell' I. e w;z. BiCFIKLM. 5 Esta CFGHIOPQZtfa. e(^ret) aspra B J (?)C 1 (?)DHMOQZ J ffCa. 6 del pensier a. 7 Tanto era I. 2nd e om. E (? erased after piu). piu che morte L. 8 per retrar K. ch' io atrovai C. 9 delle cose with 'alte' supplied in margin F. degli altre P. alte QA. 10 vi so C. che mi io v' ent. L. come v' entrai P. 11 del sonno O. a quel ACEGHKLMOP. in om. B*FCa. 11 di has been altered to dl in B. D.I. I 2 INFERNO Che la verace via abbandonai. 12 Ma poi che fui al pie d' un colle giunto, La dove terminava quella valle, Che m' avea di paura il cor compunto, 15 Guardai in alto, e vidi le sue spalle Vestite gia de' raggi del pianeta, Che mena dritto altrui per ogni calle. 18 Allor fu la paura un poco queta, Che nel lago del cor m' era durata La notte, ch' i' passai con tanta pieta. 21 E come quei, che con lena affannata Uscito fuor del pelago alia riva, Si volge all' acqua perigliosa, e guata ; 24 Cos! 1' animo mio, che ancor fuggiva, Si volse indietro a rimirar lo passo, Che non lascio giammai persona viva. 27 Poi ch' ei posato un poco il corpo lasso, Ripresi via per la piaggia diserta, SI che il pie fermo sempre era il piu basso ; 30 12 Ch' io la Q. vera F (perhaps an error, ce supplied above by later hand). la diritta K. 13 Ma puo K. ch' i' (or io) fui (or fu) ABCEFGHILMOPQZ^a. a pie FIPQZC. 16 alti DGHO. 17 del raggio KC de raggio L. di rago Q. de la pianeta L. di pianeta Q. 18 drieto P. ricto Q. a ogni F. 20 eran durata EH. indurata KP. 22 E om. L. quegli E. quello I. quel Q 2 . Quale e colui F. Et qual quei M. 23 Usciti e fuor CF. Usito e fuor K. di pel. A. 24 Si volta Q. 25 mio om. C. che om. K. 26 a drieto BOZ. a dietro Pa. a rietro DELM. a retro IGHO.5. per rimirar L. 28 Poi ch' ebbi ripossato el corpo A. Com' io posato E(?)G(?)H. Et riposato F. Poi posato ebbi IZC E poi che riposato il K. Poi ch' ebbi posato Pa. Poi che un poco il corpo (sic) L. 29 Presi '1 camin Q. della piaggia L. spiagia A. 30 ist il om. E. 2nd il om. ACK. sempre om. L. er 5 al B. 13 15 This terzina is omitted in D. 20 Prob. eran durata may stand for era indurata. 28 Q x had apparently the reading in the text. It has been altered thus : Poi ch' ebbi riposato un poco (sic}. E has passato and G pasato, both are probably clerical errors for posato, since E is habitually careless about doubled consonants. Poi che posato (as in CDM) is no doubt the same as the reading in the text. CANTO I. 3 Ed ecco, quasi al cominciar dell' erta, Una lonza leggiera e presta molto, Che di pel maculate era coperta. 33 E non mi si partia dinanzi al volto ; Anzi impediva tanto il mio cammino, Ch' io fui per ritornar piu volte volto. 3 6 Tempo era dal principio del mattino ; E il sol montava su con quelle stelle Ch' eran con lui, quando 1' amor divino 39 Mosse da prima quelle cose belle ; Si che a bene sperar m' era cagione Di quella fera alia gaietta pelle, 42 L' ora del tempo, e la dolce stagione : Ma non si, che paura non mi desse La vista, che mi apparve, d' un leone. 45 Questi parea, che contra me venesse Con la test' alta e con rabbiosa fame, SI che parea che 1' aer ne temesse : 4 8 Ed una lupa, che di tutte brame 31 E cosi quasi G. 32 Legiera una leonza P. leonza A. e orn. L. 33 Che del mac. B. maculata L. 34 non mi si movia K. davanti K. 35 Ma impediva M. 37 del p. CDFGI(?)KLM/4Ca. Temp' era ia ( = gia as in 1. 51 &c.) dell' ora del Q. 38 che '1 sol CI. saliva F. in su HO^. strelle A. 39 Ch' era M. eran lassu D. 40 de p. AK. di p. FGHPa. di pria L. 41 Si che ben isperar C. SI che bene sperar L. al ben KZ. 42 questa C. lag. BEFGHIK OPQZa. 43 el tempo A. a la dolce L. 46 Questo I. Costui Q. che 'ncontra me D. che 'ncontro a me I. c. a me DFGLQa. 47 bramosa DC 48 tremisse BQ. 32 The following are orthographical varr. h. 1. lionca legiera A. lic,iera C. leonza E. lionza K. Over leonga in E a Chiosatore has written vana gloria, and in 1. 49 over lupa avaritia. There is no comment on leone in 1. 45. 35 B (as often elsewhere) has mie for mio. 38 The form strclla recurs in ii. 55, iii. 23, vii. 98. I believe it is a dialectic form of the district about Parma and Modena with a pronunciation of the st something like German. 49 &c. From 49 to 98 L is so much injured by damp and worms, especially in the last parts of the lines, that its readings are often impossible to determine. I 2 4 INFERNO Sembiava carca nella sua magrezza, E molte genti fe' gia viver grame. 51 Questa mi porse tanto di gravezza Con la paura, che uscia di sua vista, Ch' io perdei la speranza dell' altezza. 54 E quale e quei, che volontieri acquista, E giugne il tempo, che perder lo face, Che in tutt' i suoi pensier piange e s' attrista : 57 Tal mi fece la bestia senza pace, Che venendomi incontro, a poco a poco Mi ripingeva la, dove il Sol tace. 60 Mentre ch' io rovinava in basso loco, Dinanzi agli occhi mi si fu offerto Chi per lungo silenzio parea fioco. 63 Quand' io vidi costui nel gran diserto, Miserere di me, gridai a lui, Qual che tu sii, od ombra, od uomo certo. 66 Risposemi : Non uomo, uomo gia fui, E li parenti miei furon Lombardi, Mantovani per patria ambo e dui. 69 50 con la B. 51 Che molte K. facea D. gia fe' F. 53 suo IQ. 64 che perdi C. 55 E come quei C. Et qual quei E. Quale colui che F. Qual e colui K. quello I. quei MPQ 2 . acquisti E. 56 Poi g. CF. E om. L. al tempo KL. 1' il fazze K. 57 Con tutti suoi D. Con tutto el suo pensier K. 58 la fiera F. 59 venendomi contra M. contro D. 60 M' impingeva B. sol si tazze K. 61 Mentr 1 io A. io om. EL. ritornava D. remirava K. 62 un mi B. 63 Che CKLO. Un che E. Cui M. lungo tempo parea stato f. Q. 64 Quando v. BCDEFIKLOQ. 65 Misere EF. 66 chi A. ist od om. F. sondonbra (sic) L. o domobra (sic) o domo (!) O. 67 Rispose F. non huom ma huom E. non homo gia fui H. 69 ambidui ABCM. ambeduni D. ambedui EGHKOPQ. amendui FLa. ambo dui IZD. 61 The reading (apparently rare) rimirava is strongly advocated here, and also rimtrar, similarly for ruinar, in Par. xxxii. 138, by Zani de' Ferranti, p. 4. 69 These and similar variations occur passim and will not generally be noted further. CANTO I. 5 Nacqui sub Julio, ancorch& fosse tardi, E vissi a Roma, sotto il buono Augusto, Al tempo degli Dei falsi e bugiardi. 72 Poeta fui, e cantai di quel giusto Figliuol d' Anchise, che venne da Troia, Poiche il superbo Ilion fu combusto. 75 Ma tu perche ritorni a tanta noia ? Perchk non sali il dilettoso monte, Ch' & principle e cagion di tutta gioia ? 78 Or se' tu quel Virgilio, e quella fonte, Che spande di parlar si largo fiume ? Risposi lui con vergognosa fronte. 81 O degli altri poeti onore e lume, Vagliami il lungo studio e il grande amore, Che m' ha fatto cercar lo tuo volume. 84 Tu se' lo mio maestro e il mio autore: Tu se' solo colui, da cui io tolsi Lo bello stile, che m' ha fatto onore. 87 Vedi la bestia, per cui io mi volsi : Aiutami da lei, famoso saggio, Ch' ella mi fa tremar le vene e i polsi. 90 A te convien tenere altro viaggio, Rispose, poi che lagrimar mi vide, Se vuoi campar d' esto loco selvaggio : 93 Che questa bestia, per la qual tu gride, 70 su Giulio BL. sub Giulio EH. su Julio Da. sot (?) Giulio G. ben che I L. 72 Nel tempo ABGa. iddei FIL. 74 Figlolu Q(and in iii. 121). d'AnchisesM. di Tr. AEFGMOP^4a. 77 al dil. FIKM. 79 O se' D [? Or erased] F. 80 spandi ABCDEF 1 (?)GHIKMOQ(?)Z^A dil parlar M. 81 Risposi a lui B*CIKOZ. Rispos' io a ADG^A. Rispos' io lui EF*HOQa. 83 et grande M. 86 Tu solo E. solo om. L. 88 per la qual mi F. per chi io I. 89 e saggio AB*CDFIG*KMQP^f. 90 e (or et) polsi ADEFGHIKL(?)MPQZa 93 Se vol M. scampar FG. 94 quella DFGMZZ?. 90 As e may stand for e 1 or et, this is not conclusive, and consequently will not generally be noticed hereafter. 5 INFERNO Non lascia altrui passar per la sua via, Ma tanto lo impedisce, che 1' uccide : 96 Ed ha natura si malvagia e ria, Che mai non empie la bramosa voglia, E dopo il pasto ha piu fame che pria. 99 Molti son gli animali, a cui s' ammoglia, E piu saranno ancora, infin che il veltro Verra, che la fara morir con doglia. 102 Questi non cibera terra ne peltro, Ma sapienza e amore e virtute, E sua nazion sara tra Feltro e Feltro. 105 Di quell' umile Italia fia salute, Per cui morl la vergine Cammilla, Eurialo, e Turno, e Niso di ferute : 108 Questi la caccera per ogni villa, Fin che 1' avra rimessa nello inferno, La onde invidia prima dipartilla. m Ond' io per lo tuo me' penso e discerno, Che tu mi segui, ed io saro tua guida, E trarrotti di qui per loco eterno, 114 Ove udirai le disperate strida, Vedrai gli antichi spiriti dolenti, 95 Non lascierk L. 97 E sua natura e si C. 99 piu fame ha C. 101 fin che CEFPD. insino che I. 102 di doglia D. 103 Costui ACEFHMOQ(Costuoi). 104 ist 'e'0w. AB*CEFGHIKLMOPQa 2 (?). 105 tra '1 Feltro e '1 F. L. 107 morio G. 108 Eur. Turno B*GKOZ J ff. Niso Eur. e Turno P. feruto D. 109 Costui FM. 110 havera messa E. ill La dove K. 112 io om. E. meglio B (altered to mei) CDEIKa. meio A. lo tu mei H. 113 tuo EIZ. 114 qua AL. per lo luogo D. lungo et. C. 115 Li ud. C. La ove K. spietate grida A. dispietate KILD. le dispietati P. 116 Et vedrai K. Di (or De) quegli ant. AGIZa. 104 In B 'e' has apparently been inserted as an omission and after- wards erased. 116 In favour of the insertion of vedrai is perhaps the reference, almost amounting to a citation, in iii. 17, though this may have also suggested it. Per contra (a) the repetition of vedrai in 1. 118 seems questionable; 0) it would seem more natural to have the source of the dispietate strida given, as in the reading di quegli; and CANTO I. 7 Che la seconda morte ciascun grida : 117 E poi vedrai color, che son contenti Nel fuoco, perch& speran di venire, Quando che sia, alle beate genti : 120 Alle qua' poi se tu vorrai salire, Anima fia a ci6 di me piu degna ; Con lei ti Iascer6 nel mio partire : 123 Che quello imperador, che lassu regna, Perch' io fui ribellante alia sua legge, Non vuol che in sua citta per me si vegna. 126 In tutte parti impera, e quivi regge, Quivi e la sua citta e 1' alto seggio : O felice colui, cu' ivi elegge ! iig Ed io a lui : Poeta, io ti richieggio Per quello Dio, che tu non conoscesti, Acciocch' io fugga questo male e peggio, 132 Che tu mi meni la dov' or dicesti, 117 Ch' alia (or Cha la) ABDHKMO^Z?. 118 Puo vedraiA. E ve- derai BCEGMOQZ^. E vedrai HKLPa. 120 bead DG. 121 qual EM. qualiLa. 122 di cio a. piu di me ABCDEFGHILMOPQZ^a. a cio di me sara piu K. 125 a sua A. 126 vuole in sua F. ch' a suo citta sen vegna M. 127 imperio D. 128 om. a*. citate M. e '1 lato Q. 129 E felice D. colui A. che ivi B*DIGKPQ 2 . chui gli L. 130 Ond' io F. Et io, Poeta ancora ti K. 131 Iddio ACDEFGHIKLMPZZ?a. idio i cui Q. 132 io om. Q. c' e peggio (?) B. 133 dove dicesti AF. la ove BjDGOQa. (y) the antithesis of the two classes of spirits in the terzine 115 117 and 118 120 would be more complete with the latter reading. 117 Che la seems certainly right. Against Ctt alia is : (a) the construction of grida in Purg. viii. 125. (/3) Grida here, as in Purg. I.e., does not mean 'cry out to', but proclaim, declare, set forth, (y) If, as is possible, the idea in gridare alia was intended to correspond with such passages as Job iii. 21, or Rev. vi. 16 &c., this is inappropriate, as Scripture never uses such expressions of "the second death". (8) Finally Zani de' Ferranti appropriately illustrates grida by Virgil's use of testatur in ALn. vi. 619. 129 gli (in L) = vi as in Purg. xiii. 7, 152 &c. (see Scart. note on Purg. xiii. 7). g INFERNO SI ch' io vegga la porta di san Pietro, E color, cui tu fai cotanto mesti. i 35 Allor si mosse, ed io li tenni retro. 134 le porte M. 135 E gli altri F. Pria choloro che fai L. che fai A. cui fai M. che tu fai BCEFIOPZZ?a. 136 dietro BCEFGKPZa. detro M. 134 Z has Piero in spite of rhyme. This is I believe a specially Florentine form, e.g. Porta di San Piero &c. CANTO SECONDO Lo giorno se n' andava, e 1' aer bruno Toglieva gli animai, che sono in terra, Dalle fatiche loro ; ed io sol uno 3 M' apparecchiava a sostener la guerra SI del cammino e si della pietate, Che ritrarra la mente, che non erra. 6 O Muse, o alto ingegno, or m' aiutate : O mente, che scrivesti ci6 ch' io vidi, Qui si parra la tua nobilitate. 9 Io cominciai : Poeta che mi guidi, Guarda la mia virtu, s' ella e possente, Prima che all' alto passo tu mi fidi. 12 Tu dici, che di Silvio Io parente, Corruttibile ancora, ad immortale Secolo ando, e fu sensibilmente. 15 Pero se 1' avversario d' ogni male Cortese i fu, pensando 1' alto effetto, 2 animali ACDEFHIKQa. animal GMOPZ. 3 sono 1' uno L. son 1' uno a. 4 Ma perch' io avea L. affaticava B. 5 Si della mente F. e om. M. 7 Musa A 2 BC(?)DOa. 8 scriveste O. 11 s' egli e EP. 12 Anzi AZB. che 1' alto EHM. 14 all' im. A. et in mortale M. 16 Perche B. Percio a. 17 Cortese fu BCGIMP Cort. li fu FHa. gi (?gli) fu E. 3 L (or perhaps the MS. from which it was copied) frequently omits final 1, n, r &c. and hence probably so F uno was mistaken for son /' uno. a (as often) is found in agreement with L, but does not go with it in the further emendation intended to restore the sense in 1. 4. 10 INFERNO Che uscir dovea di lui, e il chi, e il quale, 18 Non pare indegno ad uomo d' intelletto : Ch' ei fu dell' alma Roma e di suo impero Nell' empireo ciel per padre eletto: 21 La quale, e il quale (a voler dir lo vero) Fur stabiliti per lo loco santo, U' siede il successor del maggior Piero. 24 Per questa andata, onde gli dai tu vanto, Intese cose, che furon cagione Di sua vittoria e del papale ammanto. 27 Andovvi poi lo Vas d' elezione, Per recarne conforto a quella fede, Ch' e principio alia via di salvazione. 30 Ma io perche venirvi ? o chi '1 concede ? lo non Enea, io non Paolo sono : Me degno a ci6 ne io ne altri '1 crede. 33 Perche se del venire io m' abbandono, Temo che la venuta non sia folle : Se' savio, intendi me' ch' io non ragiono. 36 18 e il cui AK. e chi B. e '1 qui E. lui il che a. il che F. ist e om. P. 19 E non A. parra I. 20 Ch' el A. Che fu BCDEFGHIKLMOPQZa. alta ID. del suo F*IK. 21 Nello imperio C. Nel' omperio A. 22 Lo quale e '1 quale CEFHLMOQa. La quale il quale GP. El chi el qual K. Lo quale al quale Q. 23 Fu stabilito AEFGHILMOPQ^4CZ?a. Fu stabilito gik per loco s. C. Fue stabilita D. 24 O sede M. La u' siede a^. del m. Pero M. 25 Per questa onde onde E. onde egli F. 27 manto ACEO. 28 vaso di lectione F. vaso d' elez. \A C. 30 Che principio e alia M. e via A* (a via having been written and then corrected by same hand) D. 31 vi venni D. 32 Io non sono Enea ne Pagolo non sono L. Paghol F. e io non Paulo a. 33 Ne degno ADFGILM (digno) PQa. altri crede ACDGLQ^?)^. altro I. 34 Pero K. 35 Teme H. che mia v. F. venuta mia P. 36 Se se' savio L. entende s. ACF. entendi BIQ. mei che non C. meglio EI. non om. a. 28 B has de lectione with gap between and marks of erasure. 36 As e and * are often interchanged this may stand for e intende or merely intende. CANTO II. II E quale e quci, che disvuol cio che voile, E per nuovi pensier cangia proposta, Si che dal cominciar tutto si tolle ; 39 Tal mi fee' io in quella oscura costa : Perche, pensando, consumai la impresa, Che fu nel cominciar cotanto tosta. 42 Se io ho ben la tua parola intesa, Rispose del magnanimo quell' ombra, L' anima tua e da viltate offesa : 45 La qual molte fiate 1' uomo ingombra, Si che d' onrata impresa Io rivolve, Come falso veder bestia, quand' ombra. 4 8 Da questa tema acciocche tu ti solve, Dirotti, perch' io venni, e qucl che intesi Nel primo punto che di tc mi dolve. ?[ Io era tra color che son sospesi, E donna mi chiamo beata e bella, Tal che di comandare io la richiesi. 54 Lucevan gli occhi suoi piu che la Stella : E cominciommi a dir soave e piana, Con angelica voce, in sua favella : 57 O anima cortese Mantovana Di cui la fama ancor nel mondo dura, 37 E qual quegli che di su vuol E. Quale e colui F. Qual e quei O. e quello I. cio ch' 1 voile A. cio ch' ei voile H. quel che DMQ. de suol cio (sic) M. 38 nuovo pensier CIKM. canbia G. perposta C (cf. preposto in 138). preposta M. 39 del com. G. 40 tosta G. 41 Si che FGI. 43 la parola tua HLOQ/^a. la tua par. bene int. K. 44 dal L. 45 vilta BCEFIKLa. 47 Che d' onorata Q. durata E. e da rata F. dorata G. d' ornata I. diorata (sic) M. la rivolve EHIKL/A rinvolve D. 48 Com a f. B. Con falso E. 50 perche venni BGP. che io int. DEFIKO/?. 52 intra DOO^ 53 cortese e bella />'. 54 del A. 55 una s. BKL. di una M. 56 comincio a d. D. 57 in la CHOP. 58 e Alant. E. 59 ancor la fama C. al in. C. la voce ancor F. 57 The word \>ciorQ favclla has been erased in 'a' and is illegible. 12 INFERNO E durera quanto il moto lontana : 60 L' amico mio, e non della ventura, Nella diserta piaggia & impedito SI nel cammin, che volto & per paura: 63 E temo che non sia gia si smarrito, Ch' io mi sia tardi al soccorso levata, Per quel ch' io ho di lui nel Cielo udito. 66 Or muovi, e con la tua parola ornata, E con ci6 ch' & mestieri al suo campare, L' aiuta si, che io ne sia consolata. 69 Io son Beatrice, che ti faccio andare: Vegno di loco, ove tornar disio : Amor mi mosse, che mi fa parlare. 72 Quando sar6 dinanzi al Signor mio, Di te mi Ioder6 sovente a lui. Tacette allora, e poi comincia' io : 75 O donna di virtu, sola per cui L' umana spezie eccede ogni contento Da quel ciel, che ha minor li cerchi sui : 78 Tanto m' aggrada il tuo comandamento, Che 1' ubbidir, se gia fosse, m' e tardi ; 60 mondo ADFGILP. quanto mondo G. moto allontana B* (evidently 'moto' has been substituted). C^has 'moto' clumsily altered to 'modo'. Z* has 'moto', but evidently 7. nd(l manu. K reads -jaefide- moto luntana (erased thus i ma manu). 61 ma non F. 62 e erased in a. 63 che e v. A. che volto per GPQ. che volta E. 64 si om. C. cosi ismarito E. sia gia sia L. 65 tarda Q. 66 io om. A. di lui. o a fugg. FI. a far lor ponta (?) a fugir (sic) E. ' ill dapoi ABM. dopo tali I. cotal M. persone Z (an obvious clerical error, see 1. 109). 112 del ABCDEGMZ5C. 113 nel AEHOPQ5. 115 Poiche F. 118 Et a te venni si M. cornel v. E. 120 mi tolse B. 121 che hai BC. che e om. L. che e perche ne stai E. Dunche perche perche perche restai K. 122 elette F. tanta om. L. 112 The terzina 112114 is omitted in K. 118 In M after line 118, 1. 117 is repeated (an error evidently arising from the o/noiore'Aevra in 116 and 118), and then 119 &c. follow on rightly. CANTO II. 15 Poscia che tai tre donne benedette Curan di te nella corte del cielo, E il mio parlar tanto ben t' impromette? 126 Quali i fioretti dal notturno gelo Chinati e chiusi, poi che il Sol gl' imbianca, Si drizzan tutti aperti in loro stelo ; 129 Tal mi fee' io, di mia virtute stanca : E tanto buono ardire al cor mi corse, Ch' io cominciai come persona franca : 132 O pietosa colei che mi soccorse, E tu cortese, che ubbidisti tosto Alle vere parole che ti porse ! 135 Tu m' hai con desiderio il cor disposto Si al venir, con le parole tue, Ch' io son tomato nel primo proposto. 138 Or va, che un sol volere e d' ambo e due Tu duca, tu signore e tu maestro : Cosi gli dissi ; e poiche mosso fue, 141 Entrai per Io cammino alto e silvestro. 124 Poiche cotai P. tal EM. tra tre K. 125 Che cur. A. per te I. Cur' han A. 126 ti promette FGKOPQZ^Z). 127 Quali f. ABHLM (quagli) OPJSDa* del CDKLM (di]) a. 128 gli om. A. lonbiancha K. 129 Si drizzano aperti F. lor ostelo A. loro ostello C. 130 da EHP. 131 porse C. 132 Che com. C*. Io com. L. 133 che ti L. se corse K. si corse M. 134 te AEHOP. tu obedisti E. 137 Si advenir Z. che le par. L. 138 preposto K. 139 Or movi che un volere D. con un sol E. sol om. F. sol vedere L. 140 e om. LMQ. 141 Cosi diss' io A. e om. CK. poi comosso D. 142 Io entrai D. aspro e s. DO. 129 The reading of A, a common mis-division of words, has probably led to the more unintelligent blunder of C. 137 Z regularly has ad for a. 142 aspro is perhaps an alteration from reminiscence of i. 5. CANTO TERZO " Per me si va nella citta dolente, Per me si va nell' eterno dolore, Per me si va tra la perduta gente. 3 Giustizia mosse il mio alto fattore : Fecemi la divina potestate, La somma sapienza e il primo amore. Dinanzi a me non fur cose create, Se non eterne, ed io eterna duro : Lasciate ogni speranza, voi, ch' entrate!" 9 Queste parole di colore oscuro Vid' io scritte al sommo d' una porta : Perch' io : Maestro, il senso lor m' e duro. 12 Ed egli a me, come persona accorta : Qui si convien lasciare ogni sospetto ; Ogni vilta convien che qui sia morta. 15 Noi siam venuti al loco ov' io t' ho detto, Che tu vedrai le genti dolorose, Ch' hanno perduto il ben dello intelletto. 18 E poiche la sua mano alia mia pose, 1 sen va M (also in lines 2 and 3). 3 alia perduta Q. 5 Fecimi A. 6 il p. honore D. e om. G. 7 a noi EH. 8 eterno duro AB^FIMQ^ZtejC?). eternal duro Z. eterno (bis) DL. Se non e et. FLZ. 12 Perche Maes. K. seno C. 13 E quei C. quegli DQ. quelli Za. 16 sem Q#. siano I. Tu sei venuto O. io om. M. 17 U' vederai G. Che vederai B. le gente BEL 19 suo IK. porse Q. 17 gente is often written for genii, and this interchange therefore will not be generally registered. CANTO III. 17 Con lieto volto, ond' io mi confortai, Mi mise dentro alle segrete cose. 21 Quivi sospiri, pianti ed aid guai Risonavan per 1' aer senza stelle, Perch' io al cominciar ne lagrimai. 24 Diverse lingue, orribili favelle, Parole di dolore, accenti d' ira, Voci alte e fioche, e suon di man con elle, 27 Facevano un tumulto, il qual s' aggira Sempre in quell' aria senza tempo tinta, Come la rena quando a turbo spira. 3 o Ed io, ch' avea d' orror la testa cinta, Dissi : Maestro, che e quel ch' i' odo ? E che gent' e, che par nel duol si vinta ? 33 Ed egli a me: Questo misero modo Tengon 1' anime triste di coloro, Che visser senza infamia e senza lodo. 36 Mischiate sono a quel cattivo coro Degli angeli, che non furon ribelli, Ne fur fedeli a Dio, ma per se foro. 39 Caccianli i Ciel per non esser men belli : 20 viso Z. 21 Mise mi CGK. 22 con pianti I. e pianti a. altri gguai (sic) L. 23 Risonava C. Risonava in quell' B. 25 Div. voci G. 26 dolori C. 27 ficoche alte K. a suon LM. 28 Fanno un tumolto intorno il qual K. 29 Sempre per 1' aere M. 30 al turbo ACK. la nebbia quando turbo sp. FZ?a*. quando a tempo L. quando turbo OZ. i turbo Q : (altered to il). 31 error ABCDEFGKLMOPa*. la testa d' orror Q. tinta A. 33 qual gente BDQ. che gente che AEOP [but gente prob. =gente e]. qual genti che C. Et e gente K. siavinta L. 34 E quelli a me C. Ed elgle (sic) Q. 36 fama AB^CDEHIKLMOPQa^ 37 in quel K. 39 Ne fue M. a Dio fedeli B. per lor f. CK. per se furon (sic) E. 40 per n' esser K. ben belli IL. 36 B* has infamia. a l evidently had infamia. 39 Variations such as for, fuor, fuoro, furon, furo are common in this line and elsewhere, but are differences of orthography rather than ' variants '. The same applies to the endless variations found in writing the words Caccianli i Ciel in 1. 40. D. I. 2 r g INFERNO Ne lo profondo inferno gli riceve, Che alcuna gloria i rei avrebber d' elli. ^ Ed io : Maestro, che e tanto greve A lor, che lamentar gli fa s\ forte? Rispose: Dicerolti molto breve. 45 Questi non hanno speranza di morte, E la lor cieca vita e tanto bassa, Che invidiosi son d' ogni altra sorte. 48 Fama di loro il mondo esser non lassa, Misericordia e giustizia gli sdegna : Non ragioniam di lor, ma guarda e passa. 51 Ed io, che riguardai, vidi una insegna, Che girando correva tanto ratta, Che d' ogni posa mi pareva indegna: 54 E dietro le venia si lunga tratta Di gente, ch' i' non avrei mai creduto, Che morte tanta n' avesse disfatta. 57 Poscia ch' io v' ebbi alcun riconosciuto, Vidi e conobbi 1' ombra di colui Che fece per viltate il gran rifiuto. 60 41 Nello profondo DGI. Nel profondo BCEKQ 2 . Nel profondo d' inferno B. E lo (altered from Ne lo) B^. 42 gloria rei ne avrebbe M. n' avrebber Q. 43 Ed io a lui che e cotanto g. G (see 1. 53). 46 dicerotil AB*P. dirotel E. 46 Quelli K. 49 al mondo F. 50 e om. F. lu (?) degna K. 51 ragiona (? m or r to supply) AB. ragionar CDKMOD. ragionamo E. ragionare FI. ragionia L. rasgionian Q*. 52 ch' a riguardar D, 53 cotanto G. 54 ogni passo G [perhaps an error for possa as found in O &c.]. 55 dietro li (or gli) AP. rietroliC. drietoliBK. drietoleL. retrod, lei C, retro lei D. retro li OQ. rietro gli E. dietro le F. diretro li G. drieto gli IM. 56 che non averei E. mai om. ACDEFGHLMOPQ ZAzBCDo,. ch' io noil' aurei a. 57 avessi I. n' avisse tanta Q. 68 E poi che n' ebbi M. n' ebbi ACGKLOa. B* has v' ebbi. 60 Che per viltk fece ilgran refutoA. per viltk il (^rlo)BCDEFGHIOPQCa. per viltk si fecie L. 41 The readings Nel prof . and Nello prof . probably stand for Nl V and Ne lo respectively. 42 grolia D (cf. xv. 56, xxxi. 116). groria L (a common Tuscan form). Comp. Vorsersi iv. 98, Giuria iv. 128, &c. 56 The interchanges of retro, rietro, drieto, dietro, diretro &c. will not be noticed hereafter. CANTO III. 19 Incontanente intesi, e certo fui, Che quest' era la setta dei cattivi, A Dio spiacenti ed ai nemici sui. 63 Questi sciaurati, che mai non fur vivi, Erano ignudi e stimolati molto Da mosconi e da vespe ch' erano ivi. 66 Elle rigavan lor di sangue il volto, Che, mischiato di lagrime, ai lor piedi, Da fastidiosi vermi era ricolto. 69 E poi che a riguardare oltre mi diedi, Vidi gente alia riva d' un gran fiume : Perch' io dissi : Maestro, or mi concedi, 72 Ch' io sappia quali sono, e quel costume Le fa di trapassar parer si pronte, Com' io discerno per Io fioco lume. 75 Ed egli a me : Le cose ti fieri conte, Quando noi fermerem li nostri passi 61 Allora int. F. Immantanente A 2 . 62 quell' era FA. di ACEKMOP. 63 spiacentev4Ca. a nem. MPa. 64 Quei MO. Sti sciag. K. sciagurati ABDEGHIKMOPZ^CDa. sagurati C. sciurirati (?) F. sciagrati Q r isciagruati L. non furon L. 65 e om. ABEHMC. 66 erano vivi K. 67 Egli A. Elli CI. Et ele L. Et le C. Ello rigava FL. 68 E mischiato FM. meschiate CK. mischiati P. a lor ACDEGKLMOPQZa. 69 Di v. fast. CD. vermi fast. B. 70 E io ch' a AK. E poi che rig. CL. Poscia ch' a I. oltre a rig. F. oltra ABEMP. 71 alia ripa M. gran om. F. flume M. 72 Maestro disse FIQ. dissi al Maestro Z. 73 Ch' io sapio chi son quegli M. quai son quelli B. quali e' sono GZ. chi ei sono A$\ chi son color K. 74 Li fa AC*DFKZyi 1 (?). Mi fa L. Gli fan M. Lor fa a. del tr. EMA. parer di trap. FI. esser si pronti K. si 'npronte (?) a. 75 in questo poco F. fioco I. lu puco (?) Qi. 76 Et eglia a me F. quelli C. fier BC*DEHKLO5. fiar (?) F. fieno I. fie M. 77 fermeren ACDEFLZ. ferme- reno I. 68 B* and G* have meschiato, but both prob. originally meschiate. 72 a (though nearly obliterated) seems to have read Perch" 1 io Maestro mio or mi. 77 But final in and n are constantly interchanged (esp. in C) e.g. Carom 11. 94, 109, aveam 99, debatteam 101, com for con passim, simele- memte 1 1 5 (all in C). This will not be noted generally hereafter. 2 2 20 INFERNO Sulla trista riviera d' Acheronte. 78 Allor con gli occhi vergognosi e bassi, Temendo no '1 mio dir gli fusse grave, Infino al fiume di parlar mi trassi. 81 Ed ecco verso noi venir per nave Un vecchio bianco per antico pelo, Gridando : Guai a voi anime prave : 8 4 Non isperate mai veder lo cielo ! I' vegno per menarvi all' altra riva, Nelle tenebre eterne, in caldo e in gelo. 87 E tu che se' costl, anima viva, Partiti da cotesti che son morti. Ma poi ch' ei vide, ch' io non mi partiva, 90 Disse : Per altra via, per altri porti Verrai a piaggia, non qui, per passare : Piu lieve legno convien che ti porti. 93 E il duca a lui : Caron non ti crucciare : Vuolsi cos\ cola, dove si puote Ci6 che si vuole, e piu non dimandare. 96 Quinci fur quete le lanose gote Al nocchier della livida palude, 78 In sulla L. 79 Et io con F. occhi lacrimosi K. e om. B. 80 ne '1 mio dir AD (mie ?) ILQZ5. nol mie B*. che '1 mio CKMP. ne '1 mi dir non fosse Z. nol me dire a. 81 dal p. ADEFOC del IKLMPQ/2. 82 inverse a r 83 veglio D. 84 guai om. E. a voi om. (? cler. error) Q (a voi 2^ manu in marg.}. 85 None sperate a. gia mai D. 86 Ch' io vegno K. Vegno puoi Q. ad altra F. 87 E 'n tenebre K. il caldo L. e 'n caldo M. e gielo ACEFOP. 88 chi se' O. 89 coteste...morte M. costoro P. 90 E poi K. che vide ABCDEFGHIKLMOPQZa. che non P. 91 altre vie L. e per altri D. o per altre porte L. altre porte M. o per a. 93 Piu leggier F. 94 duca lui B*CGHLQ. E '1 duca mio Charon DK. 95 colk Ik dove B. cusi Ik E. 97 Quinde B. 98 Del nocchier K. livide BCF. 80 The construction of temere seems to have caused the copyists difficulty elsewhere also, e.g. Inf. ii. 35, xvii. 76, xxi. 93. Comp. also ' a' in ii. 36. 90 This may of course stand for ch' e' (or ch' ei) vide. CANTO III. 21 Che intorno agli occhi avea di fiamme rote. 99 Ma quell' anime ch' eran lasse e nude, Cangiar colore e dibattero i denti, Ratto che inteser le parole crude. 102 Bestemmiavano Iddio e lor parent!, L' umana specie, 11 luogo, il tempo e il seme Di lor semenza e di lor nascimenti. 105 Poi si ritrasser tutte quante insieme, Forte piangendo, alia riva malvagia, Che attende ciascun uom che Dio non teme. 108 Caron dimonio, con occhi di bragia, Loro accennando, tutte le raccoglie ; Batte col remo qualunque s' adagia. in Come d' autunno si levan le foglie L' una appresso dell' altra, infin che il ramo Vede alia terra tutte le sue spoglie, 114 Similemente il mal seme d' Adamo: Gittansi di quel lito ad una ad una, 99 aveam C. avean M. avea om. E. avien L. 100 anime triste ismorte e nude F. ch' era M. quelle genti O. 101 debatteam li C4(?)a. dibatter li EDKA 2 . 102 Tantoche P. Tosto chenteser FIMZZ>a. Poi ch' inteser K. 103 dio BEA. e i lor BKZ?. 104 el tempo BDEK. el luogo el tempo CIMQa. al tempo L. il luogo om. F. el tempo el luogo el seme G. 105 sementa LM. 106 raccolser A X E (racolse) FHPZ>. ristrinser B. siritrasstuti (sic) C. riccolser L. ricolse M. tutti quanti ABCEHKLOPa. 108 ziaschadun che K. ciaschedun ke a. ciascun che idio F. Iddio CD. 109 con i occhi AC. con gli DMO. con cenni K. con occhio P. 110 tutti li AB*EGHQ. tutti gli DILMOPZa. HI con remo CKLMOCa. 112 daun tuono (sic] G. da vento Q. 113 degli altra P. fin che FGZZ). che ramo GLa. 114 A terra vede F (vede omitted at first and inserted above in this order), per terra P. 116 Gittasi ADILZjBD. Gittarsi B. Si gittan F. Gitavasi la dentro C*. per quel I. da quel G. quel limo M. linto (?) a. 101 In B* and I* dibattero has apparently been corrected from dibat- terno. 114 Vede is supported perhaps by Georg. ii. 82 (Scartazzini h. /.). The reading Rcnde has very slender MS. support. Barlow found it in 2 only out of 72 MSS. examined, one however being the Codice Landi having the date (if genuine) of 1336. I have found it in 6 only out of about 240 MSS. which I have examined in this passage. 22 INFERNO Per cenni, come augel per suo richiamo. 117 Cosl sen vanno su per 1' onda bruna, Ed avanti che sian di la discese, Anche di qua nuova schiera s' aduna. 120 Figliuol mio, disse il Maestro cortese, Quelli che muoion nell' ira di Dio Tutti convegnon qui d' ogni paese : 123 E pronti sono a trapassar lo rio, Che la divina giustizia gli sprona SI che la tema si volge in disio. 126 Quinci non passa mai anima buona ; E per6 se Caron di te si lagna, Ben puoi saper omai che il suo dir suona. 129 Finite questo, la buia campagna Trem6 si forte, che dello spavento La mente di sudore ancor mi bagna. 132 La terra lagrimosa diede vento, Che baleno una luce vermiglia, 117 Perche nui E. come uccello fa per F. augieli L. 118 Con esse seneva F. 119 di Ik sian K. 120 Ancor CG. Anco FMZ. nova gente M^4 1 (?). 121 ffiol mio K. Fiol dis el maistro mio cort. C. Figliuolo disse F. 122 Color che BDK. Questi I. 123 pervegnon F. 124 pronte a. 125 Perch& la K. 126 se volgion M. 127 pass6 ACFG1KLMPZ. B* has passa. 128 Inpero L. Et pero lo Caron di M. 129 che suo DA^. Intender puoi che '1 F. ormai M. omai suo parlar suona I* (with 'che '!' erased before ' suo'). 132 m' imbagna D. 133 un vento G. 134 E balen6 F. la luce L. 117 Can Perche nui in E possibly be an oral blunder for Per cenni by an ignorant scribe who thought ch was soft, as I believe the initial ch at any rate is so sounded in the Venetian provinces, where this MS. was undoubtedly written ? At any rate che is written ze in Inf. v. 132. In 1. 1 20 schiera is written sciera and in xviii. 71 scheggia is written scieza. So also chiusa (which would be pronounced in Venetian something like our word "chooser? just as chiesa would be sounded like our "church"} is written ciuser in viii. 115, and we find ciuso in ix. 55, alcimia in xxix. 119, disciomi in xxxii. 100, &c. In this case the word Perche, even if pronounced soft, would probably be too familiar to the eye to admit of its spelling being varied. CANTO III. 23 La qual mi vinse ciascun sentimento: 135 E caddi, come 1' uom cui sonno piglia. 135 E che mi K. 136 E om. A. come huom D. che AFKP. che '1 DEHLMQa. cu' il O. 136 See Zani de' Ferranti's note h. I. in favour of che '/ sonno, which he says gives the more forcible idea, " come 1' uomo ch' e preso e quasi stretto dal sonno," or as Cic. says ' quern somnus complexus est,' as contrasted with cut sonno which would be "uom che s' addormenta." CANTO QUARTO Ruppemi 1' alto sonno nella testa Un greve tuono, si ch' io mi riscossi, Come persona che per forza e desta : 3 E 1' occhio riposato intorno mossi, Dritto levato, e fiso riguardai Per conoscer lo loco dov' io fossi. 6 Vero e che in su la proda mi trovai Delia valle d' abisso dolorosa, Che tuono accoglie d' infiniti guai. 9 Oscura, profond' era e nebulosa, Tanto che, per ficcar lo viso al fondo, Io non vi discerneva alcuna cosa. 12 Or discendiam quaggiu nel cieco mondo, Cominci6 il poeta tutto smorto : Io sar6 primo, e tu sarai secondo. 15 Ed io, che del color mi fui accorto, Dissi : Come verro, se tu paventi, Che suoli al mio dubbiare esser conforto ? 18 Ed egli a me : L' angoscia delle genti, 2 trono ACEHLMa, suono K. 3 ch' per f. desta CEGHMOG>4. 4 E li occhi riposati B. 5 e om. F. 6 Ik V i' fossi B. dove fussi Q. 7 Vero che L. 9 truono Da. trono BHKPQ. Che 'ntorno AEGM. intorno colgie C. Chentrono L. Che 'ntorno a colli (sic) Z. definiti (sic) guai L. 10 era prof. DFLM (with 'rea' for 'era' in LM). Osc. e prof. IZ. e tenebrosa AL. e nuvolosa M. 11 a fondo CKLQ. 12 veruna CK. niuna Q. neuna (sic) a. 13 qui giu AK. 14 Incomincio F. 15 il primo FIQ. il secondo BIK. 16 Ond' io EGHLP. 18 dubbiar mio C. esser al mio dubbiar E. 19 de la genti C. CANTO IV. 25 Che son quaggiu, nel viso mi dipigne Quella pieta, che tu per tema senti. 21 Andiam, che la via lunga ne sospigne : Cosl si mise, e cosi mi fe' entrare Nel primo cerchio che 1' abisso cigne. 24 Quivi, secondo che per ascoltare, Non avea pianto, ma' che di sospiri, Che 1' aura eterna facevan tremare: 27 Cio avvenia di duol senza martiri, Ch' avean le turbe, ch' eran molte e grandi, D' infanti e di femmine e di viri. 3 o Lo buon Maestro a me : Tu non dimandi Che spiriti son questi che tu vedi ? Or vo' che sappi, innanzi che piu andi, 33 Ch' ei non peccaro : e s' elli hanno mercedi, Non basta, perche non ebber battesmo, Ch' e parte della fede che tu credi : 36 20 Che qua giu sono a. 21 tu om. E. 22 ci sospigne BP. mi (or me) sospigne KQa. 23 mi disse BF. mosse CK. e si mi fece Y. 24 primo cinghio FL. 25 chie per D. ch' io ACEFHLOQa. 1' ascolt. Q. 26 pianti BFGI. 27 Chella ria ( = Che 1' aria) e la terra D. aire eterno B. aire eterna C. acre eterno Z. 28 E cio ABEGHLZ?. ' Cio adevenia F. Io avvenia C (an obvious blunder). Cio eran di dolor K. 29 molto grandi AEFGHKLMPQa. molte grandi D. 30 E d' infanti D. Di fanti CC. D' inf. de fern. EM. 31 E '1 buon B. 33 Or voi B. che '1 sappi CQ. anzi che piu E. prima che piu G. avanti che M. 34 ebber mercedi DGC. ebber mercede FP. ebbon mercedi IK. mercede CQ. 35 ebbero el bapt. Q*. 24 L has a similar change in 1. 107. 25 chio prob.=<:# f ho (as commonly elsewhere). 28 B has Et...cio with marks of erasure of 3 or 4 letters between. 29 In B there are traces of both readings, but molte e seems to have been the original. 33 Is voi an erroneous com- pleting by the amanuensis of the form ?/0\ which JtJ.=*vogIio1 The same blunder has been again made in 1. 62 and afterwards corrected by erasing the 'z'. 34 C has eseiano mercede. 36 The reading porta adopted in Ed. Crusca is due (as admitted by the Edd.) to considerations of supposed theological propriety. It has no MS. authority. Dr Barlow did not find it in a single one of 138 MSS. examined, though one MS. had porto. Compare the language of ii. 30. 26 INFERNO E se furon dinanzi al Cristianesmo, Non adorar debitamente Dio : E di questi cotai son io medesmo. 39 Per tai difetti, non per altro rio, Semo perduti, e sol di tanto ofifesi, Che senza speme vivemo in disio. 42 Gran duol mi prese al cor quando lo intesi, Perocche genti di molto valore Conobbi, che in quel limbo eran sospesi. 45 Dimmi, Maestro mio, dimmi, Signore, Comincia' io per voler esser certo Di quella fede che vince ogni errore: 48 Uscicci mai alcuno, o per suo merto, O per altrui, che poi fosse beato ? E quei, che intese il mio parlar coperto, 51 Rispose : Io era nuovo in questo stato, Quando ci vidi venire un possente Con segno di vittoria coronato. 54 Trasseci 1' ombra del primo parente, D' Abel suo figlio, e quella di Noe, Di Moise legista e ubbidiente; 57 37 E s' ei A. E se pur fur IZ. in and ( = inanzi) il Cr. M. 38 ado- raron DL. a dio BEGHLACa.. iddio ACDFIKMPQ/?). 39 cotali GLa. cotal P. 40 Per tal difetto FIK. tal defecti EM (diffecti). enon CEFHIKLMQZ^a. 41 Noi sem dannati F. son di tanto L. per tanto K. 42 viviamo B^?). 43 quand' io 1' int. DQ. quand' io intesi K. allor quando 1' int. G. 44 gente ABDEFGHIKMPQ5. grande val. D. 45 quel nimbo B. quel luogo P. 47 io per esserne certo Q v 49 Usine K. o om. BCDKMZ. mai nessum C. veruno Q. 50 1' altrui I. fosse poi B. 51 E quel IMQ.J. 52 Mi disse B. Disse K. 53 Che io ci F. Quand' ioAEHKPQC. 54 incoronato ACDGIKLMQ. 55 Tras- sene Q. 56 Abello F. Da bel. G. suo fio K. 57 Et di Moyse Q. Moyses P. e om. DMQyi 1 (?). 47 In Q voler has been inserted above and ne lightly erased. 57 I have met with an ingenious suggestion (by Zani de' Ferranti) that we should punctuate thus : CANTO IV. 27 Abraam patriarca, e David re, Israel con lo padre, e co' suoi nati, E con Rachele, per cui tanto fe', 60 Ed altri molti ; e fecegli beati : E vo' che sappi che, dinanzi ad essi, Spirit! umani non eran salvati. 63 Non lasciavam 1' andar, perch' ei dicessi, Ma passavam la selva tuttavia, La selva dico di spiriti spessi. 66 Non era lunga ancor la nostra via Di qua dal sonno ; quando vidi un foco, Ch' emisperio di tenebre vincia. 69 Di lungi v' eravamo ancora un poco, Ma non si, ch' io non discernessi in parte, Che onrevol gente possedea quel loco. 72 58 U' Abraam I. 59 Ancho Israel col padre P. Is. con Isaac D. con suo padre G. col suo padre IZ. con suo nati AEGKQ. con soi MP. 61 E molti altri K. 62 Or vo' che K. 64 perch' io L. 67 ancor lunga C. 68 suono A. sommo ILP. sono KACD. sommo (altered to somno by erasure) C; somno O. quand' io ADEGHIKMOPQ^CZ). 69 Che V emisp. ACQ. venia F. 70 n' eravamo ABCEGHKPZ). 71 scorgessi F. 72 genti C. Di Moise legista ; e ubbidiente Abraam patriarca, &c. on the ground that obedience is perhaps the most striking feature in the character of Abraham, whereas Moses was in fact severely punished for disobedience on an important occasion. Certainly we might quote several passages of Scripture in which this characteristic of Abraham is emphasized, e.g. Gen. xxvi. 5 and ch. xxii. generally. Also again Heb. xi. 8 and 17. The MSS. of course do not help us on such a point as this. With the ordinary punctuation, I suppose an antithesis would be intended like that suggested by the centurion who was both 'a man under authority' and also had 'soldiers under him.' So Moses gave laws and yet submitted himself to God's commands. This does not seem to me very forcible. 64 B and Z have lasciavan, I has lasciavono, and E lasavemo, but these and similar variations are not generally worth recording, as the same reading is intended. Also most of the MSS. examined have perche dicessi, but this is inconclusive, since it may be meant for perch' ei. 28 INFERNO O tu, che onori e scienza ed arte, Questi chi son, ch' hanno cotanta onranza, Che dal modo degli altri li diparte ? 75 E quegli a me : L' onrata nominanza, Che di lor suona su nella tua vita, Grazia acquista nel ciel che si gli avanza. 78 Intanto voce fu per me udita : Onorate 1' altissimo poeta ; L' ombra sua torna, ch' era dipartita. 81 Poiche la voce fu restata e queta, Vidi quattro grand' ombre a noi venire ; Sembianza avevan ne trista ne lieta. 84 Lo buon Maestro comincio a dire : Mira colui con quella spada in mano, Che vien dinanzi a' tre si come sire. 87 Quegli e Omero poeta sovrano, L' altro e Orazio satire, che viene, Ovidio e il terzo, e 1' ultimo Lucano. 90 Perocche ciascun meco si conviene Nel nome, che son6 la voce sola, Fannomi onore, e di ci6 fanno bene. 93 73 cui honora scienza AFLM. che onori sc. EH. che omori (sic) di siencia K. ogni scienza I. Q* has Onde io con ogni scientia e con ogni arte (the last three words shewing marks of erasure). 74 Chi son costoro BF. Chi son coloro K. Questi cui M. che son O. che han (or hanno) tanta AEFGKMQa. 75 Che lino do dai (sic) K. di altri A. dagli altri Ba. mondo QiCCj. Che '1 mondo si dagli L. 76 Et egli (or elli) ABDFIKMOQC>a. 1' anticha nom. G. 77 Che dir lor L. ancor nella F. 78 acquistar C. in ciel AEGHLPa 2 . 79 noi K 2 (altered from 'me'). 81 era partita L. 82 la om. P. 83 grandi om. LM. quattro ombre (om. grande) a. ombre grande BK. ombre grandi CECZ?. a noi om. F. verso noi Ma*. 84 avea MZ?. 85 Lo mio Maes. K. a me comincio C. incomincio B. 86 spata AM. con la spada P. 87 a te Da. 88 homera D. soprano ACKM. 90 Ovidio il terzo AEM. terzo, 1' ultimo Luc. L. terzo, 1' ultimo e Luc. Z. e Luc. ACDFIPQa. 91 meco ciascun E. 75 modo in B seems to have been substituted for mondo. CANTO IV. 29 Cosi vidi adunar la bella scuola Di quei signer dell' altissimo canto, Che sopra gli altri com' aquila vola. 96 Da ch' ebber ragionato insieme alquanto, Volsersi a me con salutevol cenno : Perche '1 Maestro sorrise di tanto: 99 E piu d' onore ancora assai mi fenno, Ch' esser mi fecer della loro schiera, SI ch' io fui sesto tra cotanto senno. 102 Cosi n' andammo infino alia lumiera, Parlando cose, che il tacere e bello, SI com' era il parlar cola dov' era. 105 Venimmo al pie d' un nobile castello, Sette volte cerchiato d' alte mura, Difeso intorno d' un bel fiumicello. 108 Questo passammo, come terra dura: Per sette porte intrai con questi savi ; Giugnemmo in prato di fresca verdura. m 94 vid' io ACDEHKPQ. 95 Di quel B*DEFGH5>a. quei signori ILQ. queli signo K. 96 agli altri IZa. egli altri L. ogn' altri K. com' aghulglia (sic] a. 97 Poi ch' ebber BFQ 2 Z>. Quand' ebber a. ebbe M. 98 Volsero I. Vorsersi G. Volsesi L. 99 E '1 mio Maestro ABCDEFGHILMOPQZ^a. 100 E ancor piu d' onor D. 101 Che dei mi B. Ch' essi mi fenno D. Che essi mi fecer AEFI*Z>. Che si mi fecer CHLOa. Che si mi fece M. Che mi fecer K. Che si mi fero P. Ch' ei si GA 2 SC. 102 il sesto I. fra cotanto BG. 103 Cosi and. AiBCDEHLMOPZ^a. Cosi andando Q. insino EGHLa. fin' alia CKP. 105 Si come il parlar cola F. 106 Giugnemmo AEHLPZ*. ad pie Z. a pie CL. d' un molto bel cast. a. 107 ginchiato L. cerchiate P. 109 cosa dura K. 110 porti BCD. Ill Venimmo AEFGHILP. bella verd. C. 95 In B quei has been altered to quel, as it has been passim (though not h. /.) in Q. 97 Q, had Da che. 101 The reading in B is probably meant for Ched ei i.e. Che ei. Z has Chesimi (sic} which prob. represents CK 1 essi mi. 103 ne has been inserted 2 da manu in A. ill C has Gongemo as also in viii. 76. Comp. the interchange of these two words in 1. 106 sup., also in vii. 44, viii. 76, ix. 89. 3O INFERNO Genti v' eran con occhi tardi e gravi, Di grande autorita ne' lor sembianti : Parlavan rado, con voci soavi. n 4 Traemmoci cosl dall' un de' canti In loco aperto luminoso ed alto, SI che veder poteansi tutti e quanti. 117 Cola diritto, sopra il verde smalto, Mi fur mostrati gli spiriti magni, Che del vederli in me stesso n' esalto. 120 lo vidi Elettra con molti compagni, Tra' quai conobbi Ettore ed Enea, Cesare armato con gli occhi grifagni. 123 Vidi Cammilla e la Pentesilea Dall' altra parte, e vidi il re Latino, Che con Lavinia sua figlia sedea. 126 Vidi quel Bruto che caccio Tarquino, Lucrezia, Julia, Marzia e Corniglia, E solo in parte vidi il Saladino. 129 Poi che innalzai un poco piu le ciglia, Vidi il Maestro di color che sanno, Seder tra filosofica famiglia. 132 112 Quivi eran genti F. Gente v' era EM. Genti v' avea a. con i occhi CK. con gli occhi M. 114 Parl. raro AK. Parl. tardo P. 115 da un QLA^BD. 117 si potean ABCDEFGHKLMOPQZ.ff^a. 118 E Ik D. Quivi M. diretro G. drieto K. Con la dricto P. vado(?) smalto C. 120 Che di CD. del vedere in me ABCDEGHLOQZ^a. di vedergli me stesso F. de vederli K. vedere me stesso M. del vedere me P. m' esalto DIK. 121 Vidi El. K. 122 Fra quai F. Fra qual P. con. ed Ett. Btf. 123 Cesaro AEMP. 124 Cam. vidi I. 125 Dagli altra P. parte vidi BCEFIKLMPQZ/4 2 CZ>a*. altre parte Z. 126 colla vina (sic) a. 128 Martia Giulia BDI. Giuria L. 130 Poi ch' io DEFIMZZ?. Poscia che P. alzai EIMPZZ?. la ciglia CE. 115 The terzina 115 to 117 is omitted in I. 117 Most MSS. read tutti quanti but this slight variation is scarcely worth noting. 122 This line was omitted at first in Q, but added in margin apparently i ""'- manu. CANTO IV. 31 Tutti lo miran, tutti onor gli fanno. Quivi vid' io Socrate e Platone, Che innanzi agli altri piu presso gli stanno. 135 Democrito, che il mondo a caso pone, Diogenes, Anassagora e Tale, Empedocles, Eraclito e Zenone: 138 E vidi il buono accoglitor del quale, Dioscoride dico : e vidi Orfeo, Tullio e Lino e Seneca morale: 141 Euclide geometra e Tolommeo, Ippocrate, Avicenna e Galieno, Averrois, che il gran comento feo. i 44 Io non posso ritrar di tutti appieno ; Perocche si mi caccia il lungo tema, Che molte volte al fatto il dir vien meno. 147 133 tutto onor (gli fanno omitted) G. lui mira M. 134 e Socr. BDZa. e '1 Plat. LQ^?). 136 Democrite D. Democlito AGHQ. ch' a casu (sic) el mondo A. 138 Eracrito G. Era collito L. et Eracl. P. Impedocres iraclito e zanone (sic} F. Gienone A. 140 Dioscorides DKMO. 141 T. e Merlino D. e Alino ABCFH^. e Plinio I. Tullio Lino Z. Tullio Alano E. T. e Alano a. Tullio Alino GKP. E Tullio D. e Livio MQ*. e '1 buon Senacha L. 142 Euclides DK. E vidi geom. FL. 143 Avicenna Ypocrate EHP. Ypocras D. Ypocrate ABFM. Ipocras CG. Avicenna Epocrates L. 144 E Averr. A. gran om. F. 145 ritrargli tutti I. tutto K. 146 Poiche D. Percioche B. si mini (?) caccia in 1. E. il om. K. mi stringe il B. lunga CD. in lunga CGHP. a lungo a. 147 Perche molte F. il fatto al dire B*. al fatto al dir E. 141 See Blanc's Erklarungen on this passage. He thinks Greek and Latin authors are purposely alternated as Greeks and Arabs just below and hence defends 'Lino! Dante's habit of giving examples alternately from sacred and profane history (esp. in the Purg.) is well known. See the Author's Time-references in the D.C., p. 132. 143 See Dionisi Anedd. ii. p. 102 on the occasional importance of the ortho- graphy of proper names as indicative of antiquity in MSS. 147 B has obviously been altered from the ordinary reading. The result is a sentiment which though true in another sense is quite unsuitable to the context. Curiously enough a similar alteration occurs in the same MS. in xxxii. 12, though there is no mark of correction there. 32 INFERNO La sesta compagnia in due si scema : Per altra via mi mena il savio duca, Fuor della queta, nelF aura che trema; 150 E vengo in parte, ove non e che luca. 149 men6 '1 G. Mi mena il savio duca per altra via L. 150 all' aura P. 151 venni FG. dove A. ove ne piu che P. chi luca DEHKLOQZ^a. Cosl discesi del cerchio primaio Giu nel secondo, che men loco cinghia, E tanto piu dolor, che pugne a guaio. 3 Stavvi Minos orribilmente e ringhia: Esamina le colpe nell' entrata, Giudica e manda, secondo che avvinghia. 6 Dico, che quando 1' anima mal nata Li vien dinanzi, tutta si confessa ; E quel conoscitor delle peccata 9 Vede qual loco d' inferno e da essa : Cignesi colla coda tante volte, Quantunque gradi vuol che giu sia messa. 12 Sempre dinanzi a lui ne stanno molte : Vanno a vicenda ciascuna al giudizio ; Dicono e odono, e poi son giu volte. 15 O tu, che vieni al doloroso ospizio, Disse Minos a me, quando mi vide, Lasciando 1' atto di cotanto ufizio, 18 1 nel c. Y)B. 2 Gia G. 3 tanto ha piu B^?). piangne (sic) M. dolor pugna e guaio F. e guaio GKL. 4 Stava P. Minoso L. Minossi F. e om. L. 5 Dis"amina F. 6 che vinghia M. 7 lo dico FK. 8 Li va K. Gli ven dananci M. tutto M. 9 della I. 11 E cingnesi la coda F. 12 Quanti gradi F. 13 dinanzi lui A. molto E. 15 rivolte B. 17 Grido Minos. I. Disse mi nosso ame (sic) G. 18 tanto (corrected to cotanto] la. 4 F and L similarly in 1. 17, F in xxix. 120, and L in xiii. 96. Cf. Semiramisse in 1. 58 and Parissi 1. 67, Minosso in xx. 36, Antroposo xxxiii. 126, Caosso xii. 43. D. I. 3 34 INFERNO Guarda com' entri, e di cui tu ti fide: Non t' inganni 1' ampiezza dell' entrare! E il duca mio a lui : Perche pur gride ? ai Non impedir lo suo fatale andare : Vuolsi cosl cola, dove si puote Ci6 che si vuole, e piu non dimandare. 24 Ora incomincian le dolenti note A farmisi sentire : or son venuto La dove molto pianto mi percote. 27 lo venni in loco d' ogni luce muto, Che mugghia, come fa mar per tempesta, Se da contrari venti e combattuto. 30 La bufera infernal, che mai non resta, Mena gli spirti con la sua rapina, Voltando e percotendo li molesta, 33 Quando giungon davanti alia ruina, Quivi le strida, il compianto e il lamento, Bestemmian quivi la virtu divina. 36 Intesi, che a cosl fatto tormento Enno dannati i peccator carnali, Che la ragion sommettono al talento. 39 E come gli stornei ne portan 1' ali, Nel freddo tempo, a schiera larga e piena, 19 di chi CO. de cui tu te infide M. 20 Non ti vincha L. 21 E '1 buon Maestro a lui B. mio a lor L. 22 None F. il tuo F. al suo a. andata(in spite of rhymes in 11. 20 and 24) G. 25 Ora comincia M. Or s' incom. PQ. 28 in parte A-tf)D. 29 il mar IB 2 LMQZZ>. 32 per la F. 34 Quando vTgon Q (=vegnon). dinanzi AEFGIa. denanti Q. di venti (sic) O. 35 e la strida C. con pianto in lamento D. con p. e lam. CFILMa. col pianto e lamento GD. e col pianto e lamento B. el (? e il) comp. AQ X . il com- pianto il lamento HP. pianto ilamento E. e lamento K. 37 a si fatto CF. 38 Eran dann. CFGl(eron)MA l (?)D. Erano A. Sono P. 39 sotto meteiio E. 40 E om. K. storneli EL. stornei FZ. li stornelli portan G. 41 Nel tempo freddo G. lunga e piena H. a piena D. 29 a has an erasure before mar; prob. '//' erased. 35 In A el often stands for il e.g. 11. 65, 91 &c. CANTO V. 35 Cos! quel fiato gli spirit! mali. 4 a Di qua, di la, di giu, di su gli mena : Nulla speranza gli conforta mai, Non che di posa, ma di minor pena. 45 E come i gru van cantando lor lai, Facendo in aer di se lunga riga ; Cosl vid' io venir, traendo guai, 4 8 Ombre portate dalla detta briga : Perch' io dissi : Maestro, chi son quelle Genti, che 1' aura nera si gastiga ? 51 La prima di color, di cui novelle Tu vuoi saper, mi disse quegli allotta, Fu imperatrice di molte favelle. 54 A vizio di lussuria fu si rotta, Che libito fe' licito in sua legge, Per torre il biasmo, in che era condotta. 57 Ell' e Semiramis, di cui si legge, Che succedette a Nino, e fu sua sposa : Tenne la terra, che il Soldan corregge. 60 L' altra e colei, che s' ancise amorosa, E ruppe fede al cener di Sicheo ; Poi e Cleopatras lussuriosa. 63 42 flato A. 43 di su di giu ABCEKMA 45 minor mena (from 6fj.otoT\fVTa in 1. 43) L. 46 E come gru (or grue) A 2 CEMQ* (gruoy). va M. 47 aere si lunga L. 48 vidi venir HOQa x . 60 Maestro dissi I. che son E. 61 Gente BDGKMP. aer ID. aere nera FG. aera negra K. 52 de chi Q. 53 vol M. mi om. M. quest! CEHLMOPQa. 55 Alvizio D. 56 Che libido A. fe' libero G. 57 torse A. tuor C. torsi F. dov' era a^. corrotta DIQ. 58 Egli e Sem. KMP. Semiramisse FI. 69 suger dette EZ> (marg.\ 60 le terre M. 61 s' uccise KPa. 63 L' altra GlPa^. Cleopatra IKD. L' altra Cleopatra Z (-patras secunda manu}. e om. CM. 53 Q has quisti (as usual). 66 In A, Benv. da Imola (as translated) comments on libido, "cioe la sua libidinosa e lussuriosa voluntate parer ed esser 'licito in sua legge.'" 69 The Chiosatore in E adds 'donavit lactem filio Beli.' 36 INFERNO Elena vidi, per cui tanto reo Tempo si volse, e vidi il grande Achille, Che con amore al fine combatteo. 66 Vidi Paris, Tristano; e piu di mille Ombre mostrommi e nominolle a dito, Che amor di nostra vita dipartille. 69 Poscia ch' io ebbi il mio dottore udito Nomar le donne antiche e i cavalieri, Pieta mi giunse, e fui quasi smarrito. 72 Io cominciai : Poeta, volentieri Parlerei a que' due, che insieme vanno, E paion si al vento esser leggieri. 75 Ed egli a me : Vedrai, quando saranno Piu presso a noi ; e tu allor li prega Per quell' amor che i mena ; e quei verranno. 78 SI tosto come il vento a noi li piega, 64 vedi EG^ 2 C. 65 vedi GA 2 C. e om. Q. il forte Achille P. 66 per amore ADCZ>. si comb. Q 2 . 67 Vedi AC. Tristano, Paris A. Parissi FI^ 68 Ombre nominommi e dimostrolle F. e nominommi a dito ABCDEGHKLMOPQZ^a*. 70 Poich' io D. Poscia ch' ebbi a. 71 1' antiche donne FGIZ. e caval. BEFHIKOPQa. 72 mi vinse DFGILMZZ>. 73 Et com. G. Maestro EHLPa*. 75 paion a vento esser si P. Ke paion a 2 (k evidently inserted). 76 a me quando presso saranno F. come saranno K. 77 Piu qui a noi F. 78 Per 1' amor I. ch' elli mena qui verr. D. ed ei verranno BFGMQZZ?. 79 E tosto C. 64, 5, 7 I am strongly inclined to prefer vedi in these 3 cases to vidi. It not only carries on Virgil's description in the preceding lines, but it seems to be almost required by the language of lines 67 71. Zani de' Ferranti very aptly compares Inf. xx. 118 121, as strongly supporting vedi. Add also xx. 37, 40. Buti reads vedi and comments (as I think correctly) Parla ancora Virgilio, e dice, Tu, Dante, vedi Elena &c. and similarly in lines 65 and 67. So also Bargigi ; O tu, Dante, vedi Elena &c. No other of the older Comm. recognize this reading. Ac- cording to Scarabelli, Batines 405 has vedi in all 3 places, and the Cod. Filip. now at Naples, has vedi in 1. 64 and vidi in 65 and 67, while the Cod. Cassin. has vedi in 64 and 67, and vidi in 65. The terzina 64 66 is omitted in G. 73 a t seems to have had Poi com. A similar sub- stitution of Maestro for poeta occurs in some MSS. in 1. in. CANTO V. 37 Mossi la voce: O anime affannate, Venite a noi parlar, s' altri nol niega. 81 Quali colombe dal disio chiamate, Con 1' ali alzate e ferme, al dolce nido Volan per 1' aer dal voler portate : 8 4 Cotali uscir della schiera ov' e Dido, A noi venendo per 1' aer maligno, SI forte fu F affettuoso grido. 87 O animal grazioso e benigno, Che visitando vai per 1' aer perso Noi che tignemmo il mondo di sanguigno : 90 Se fosse amico il re dell' universe, Noi pregheremmo lui per la tua pace, Poiche hai pieta del nostro mal perverse. 93 Di quel che udire e che parlar ti piace Noi udiremo e parleremo a vui, Mentreche il vento, come fa, si tace. 96 80 Muovi (or-ve) ABGHIKLOPQZ5. 82 Quali le I. Quale M. Qual le EZ. 83 ale EKP. alie a. 1' ali aperte G. alzati L. 84 Vegnon ABC(vegniam)DEFGHIKMOPQZ^ 2 CZ>a. dal disio CF. del voler L. 85 Cotal usi (sic) C. Cotale E. usciron L. 86 Vendo (=venendo) ver noi F. Venendo a noi I. aria maligna (violating rhyme) L. 87 1' effectuoso KZ. 91 Se fussi amico a re (altered later to al re) I. 92 della tua ABCDEFGHKLM (di la) OPQA^Da.. 93 Da ch' ai GMQa. 94 o di parlar F. e di parlare I. ed ascoltar ti K. vi piace AEHOP. 95 a voi ABDELPQZa. 96 che '1 velto (sic) L. come fu G. ci tace ACDEFHKLMOQZ^a. qui tace A v 87 effetto and affetto are often confused by copyists e.g. 1. 125, Purg. xi. 3 (var. lect. noticed by Oft.}, Parad. viii. 123, ix. 107, xx. 41, xxvi. 127, xxxii. I, 66, &c. 90 Z has the abbreviation tenemo (?). 96 The reading Vz tace' is I think certainly preferable. The 'bufera infernal che mai non resta' (1. 31) was temporarily lulled for the two who 'uscir della schiera ov' Dido' (1. 85), but probably not for all, as was the case in Orpheus's visit in Georg. iv. 481 4. There is thus no ground for the charge of inconsistency with 1. 31 (though in any case it would surely be hypercritical to insist on it), which is urged with his usual bitterness by Castelvetro in his Comm. h. I. : nor any need to adopt the grotesquely pedantic suggestion of a modern critic that the bufera infernal was cyclonic, and that here we may have reached the central calm (!). 38 INFERNO Siede la terra, dove nata fui, Sulla marina dove il Po discende Per aver pace co' seguaci sui. 99 Amor, che al cor gentil ratto s' apprende, Prese costui della bella persona Che mi fu tolta, e il modo ancor m' offende. . 102 Amor, che a nullo amato amar perdona, Mi prese del costui piacer si forte, Che, come vedi, ancor non mi abbandona. 105 Amor condusse noi ad una morte : Caina attende chi vita ci spense. 97 foi BP. 98 In sulla L. 99 con seg. MP. 100 che '1 cor M. 101 Mi prese costui mi de la persona E. Mi prese costui HL. 102 tolto L. mondo ACDGKMC^?). al mondo LP. 103 che nullo CL. che nulla G. amor perdona CF. 104 di costui CDGIKLMZZ>a. del piacer costui AF. il piacer M. 107 Chay A. ChaynZ*. Chain Q*. Chaino (or Caino) B*CDGHIKMPa. cui vita K. la vita A. a vita CHEP^a. in vita IZA(?). che a vita L. 97 &c. In A these lines are misplaced thus 98, 99, 97, 100, and cor- rected by b, c, a, d, written in the margin. 97 nata clumsily altered to nati in a! In F at line 100 &c. we read Amor che a nullo amato amor perdona Prese costui della bella persona Che mi fu tolta. At this point the copyist seems to have discovered his mistake and to have begun again at 'Amor che al cor' &c. Compare a similar mistake at xiv. 56 [a good instance of blunders of eye due to o/xoioreXfura]. 102 If mondo be read, cf. a similar sentiment more at length in xiii. 77 8 ; modo and mondo are again interchanged in iv. 75, x. 64, xi. 61 &c. 107 I think it is probable that Dante wrote Cain (or possibly Caino\ thus describing the first murderer himself as awaiting the arrival of this modern fratricide with malicious eagerness. This seems much finer, and more in Dante's fierce and bitter manner, than a reference merely to the region afterwards described as Caina. Besides (as Zani de' Ferranti urges) ought not that to be La Caina f The same critic aptly quotes Foscolo as follows : " Dante nel nominar Caino ebbe intenzione ad accennare ch' ei per consolarsi con la compagnia e la sciagura de' suoi complici gli aspettasse impazientamente : quindi scoppia di subito 1' indole invidiosissima di quel tristo, e la malignita comune alle anime della sua stampa." It is also exactly the spirit so graphically described CANTO V. 39 Queste parole da lor ci fur porte. 108 Da che io intesi quelle anime offense, Chinai '1 viso, e tanto il tenni basso, Finche il poeta mi disse : Che pense? m Quando risposi, cominciai : O lasso, Quanti dolci pensier, quanto disio Men6 oostoro al doloroso passo ! u 4 Poi mi rivolsi a loro, e parla' io, E cominciai : Francesca, i tuoi martiri Al lagrimar mi fanno tristo e pio. 117 Ma dimmi : al tempo de' dolci sospiri, A che e come concedette amore, Che conoscesti i dubbiosi desiri ? 120 Ed ella a me : Nessun maggior dolore, Che ricordarsi del tempo felice Nella miseria ; e ci6 sa il tuo dottore. 123 Ma se a conoscer la prima radice Del nostro amor tu hai cotanto affetto, 108 Cotai F. da loro ci furon L. ne fur IZ. 109 Poi che ebbi udite F. Poi ch' io int. IZD. Quand' io EGOA. Poscia che 'ntesi A. Poi ch' ebbe intese M. ill Che '1 mio maestro mi disse F. Finchfe '1 maestro Ca. chi P. 112 Quand' io ABHKLOPa. incominciai D. basso (cler. error) H. ai lasso M. hei lasso A-^T). 113 dolci sospiri I. e quanto F. 115 riv. alora L. 118 il tempo a. 119 O che D. 120 conosceste BDFGKQ.& 121 E quellaACEGHLMPQ^a. niun C. 122 E che a^ 123 il mio dottore K. 124 se conoscer D. 125 effetto CEFGIKL. in Maestro Adamo in Inf. xxx. 76 87. Compare too the splendid passage in Isaiah xiv. 9 &c. The form Cain occurs in Par. ii. 51. Cain and Caino are both found in Inf. xx. 126. The present is of course a case in which MS. evidence goes for next to nothing, except as in favour of Caino when it is written thus in full. The usual form in MSS. would be Cainattende, which might stand for Cain, Caino, or Caina, with a decided chance in favour of the last being chosen on account of the following a in attende. Cod. Filip. has Caym (auct. Scarabelli). Also of the 4 first edd. two have Caino and one has Caym, one only having Caina. 117 Several MSS. have A lagr. or A lagrimar, but this is inconclusive. So some have allagrimar in vi. 59, which may stand for a lagr. or al lagr. So F has chelle = che le in vii. 92, and I has allei for a lei in vii. 85. 40 INFERNO Far6 come colui che piange e dice. 126 Noi leggevamo un giorno per diletto Di Lancelotto, come amor lo strinse: Soli eravamo e senza alcun sospetto. 129 Per piu frate gli occhi ci sospinse Quella lettura, e scolorocci il viso : Ma solo un punto fu quel che ci vinse. 132 Quando leggemmo il disiato riso Esser baciato da cotanto amante, Questi, che mai da me non fia diviso, 135 La bocca mi baci6 tutto tremante : Galeotto fu il libro e chi lo scrisse : Quel giorno piu non vi leggemmo avante. 138 Mentre che 1' uno spirto questo disse, L' altro piangeva si, che di pietade lo venni meno' si com' io morisse ; 141 E caddi, come corpo rnorto cade. 126 Dir6 ACDEKLMOPQv4 2 a. colei CIZ. 128 costrinse E l (co- erased later), e come Da.. 129 e om. ADM. 131 scoloricci ACDKQ*Z. 132 che om. A. fu sol quel L. 133 viso B* (clumsily altered from riso). 134 D' esser K. basciata D. 135 Costui O. fu CLZ. B has fia apparently altered from fu. 136 tutta LP. 137 fu libro D. e cui AK. e che L. 138 non leggevamo C. davanti L. 139 Ma mentre L. mi disse A. 140 da pietade AM. 141 men cosl ABDFHIKOPQZ^. meno come s' io D. a meno AM. si come s' io L. 142 morto corpo A. el corpo M. CANTO SESTO Al tornar della mente, che si chiuse Dinanzi alia pieta de' due cognati, Che di tristizia tutto mi confuse, 3 Nuovi tormenti e nuovi tormentati Mi veggio intorno, come ch' io mi mova, E ch' io mi volga, e come ch' io mi guati. 6 Io sono al terzo cerchio della piova ^ Eterna, maledetta, fredda e greve : Regola e qualita mai non 1' e nuova. 9 Grandine grossa, e acqua tinta, e neve Per 1' aer tenebroso si riversa : Pute la terra che questo riceve. 12 Cerbero, fiera crudele e diversa, Con tre gole caninamente latra Sopra la gente che quivi e sommersa. 15 Gli occhi ha vermigli, la barba unta ed atra, E il ventre largo, e unghiate le mani ; Graffia gli spiriti, scuoia, ed isquatra. 18 2 Davanti F. Dinanzi la C. 6 O come ch' io F. E come ch' io I. Com' io mi G. E che mi a. o ch' io mi guati F. ech'iomil. ch'io guati ABCDEGHKLOPQ^a. 8 e fredda Q. 9 giam- mai li nuova F. gli e IMO. non li nova A. non linnova F. 10 in neve Q. ist 'e' om. EHIMOZ?. 11 tenebrosa ADFGIL. rinversa D. roversa A. 12 questa DI*K. queste Q. 13 fiero L. 14 comunamente E. 15 Sopra alia L. gente om. E. che e quivi EL. qui e D. 16 ha vermei C. e la barba BIZB. 17 II ventre OPCZ>. Largo il ventre F. a largo C. 18 inghoia AC*E FGHILMOPQv4. e ingolla B. ed inghoia DKZ^a. inghia (sic) C. e squatra F. e disquatra BHLOPa. e disquarta AM. 42 INFERNO Urlar gli fa la pioggia come cani: Dell' un de' lati fanno all' altro schermo ; Volgonsi spesso i miseri profani. 11 Quando ci scorse Cerbero, il gran vermo, Le bocche aperse, e mostrocci le sanne : Non avea membro che tenesse fermo. 24 E il duca mio distese le sue spanne; Prese la terra, e con piene le pugna La gitt6 dentro alle bramose canne. 27 Qual e quel cane che abbaiando agugna, E si racqueta poi che il pasto morde, Che solo a divorarlo intende e pugna; 30 Cotai si fecer quelle facce lorde Dello demonic Cerbero che introna L' anime si, ch' esser vorrebber sorde. 33 Noi passavam su per 1' ombre che adona La greve pioggia, e ponevam le piante Sopra lor vanita che par persona. 36 Elle giacean per terra tutte e quante, Fuor ch' una che a seder si Iev6, ratto Ch' ella ci vide passarsi davante. 39 20 e 1' altro A. all' altro fanno Fa. 21 spessi K. 22 ci sporse E. 23 La bocca CDGKZ?. Aperse le bocche F. mostronci M. 24 avean C. venbro (sic) L. 25 Lo duca AEHLCZ?a. 26 e om. DFaj. piena la pugna C. 27 Gittolla CI. 28 Qual quel E. abagiandoMQ. che baiando O. 29 Che si racqueta come '1 F. araqueta Q. il om. a. 30 E solo Z. a vorarlo C, al divorarlo a. 31 Cotal si EM. Cotali DGI. 32 Del demon Cerbero che trona C (this is the whole line). Di quel dim. Z. 33 li, si ch' esser vorrien A Z C. voriam sorde C. vorrebbero esser Ea. vorrebbe esser OP. 34 sopra 1' ombre G. 1' ombra ch' aduna C. 35 tenevam a. 36 Sopra allora I. Sopra a lor L. 37 Ed elle L. 38 Fuor d' una BDELMOQZ5. Fuor una P. si levo a seder E. In K, 'cha seder ratto' was written i ma manu, and this is clumsily altered to 'cha se levo ratto/ 39 passar si EGIL. passarci Q. Perche ci vide passeggiar D. 22 In F, 1. 23 was at first written after 21. It is left standing under- lined, though afterwards appearing in its right place. CANTO VI. 43 O tu, che se' per questo inferno tratto, Mi disse, riconoscimi, se sai : Tu fosti, prima ch' io disfatto, fatto. 42 Ed io a lei : L' angoscia che tu hai Forse ti tira fuor della mia mente, SI che non par, ch' io ti vedessi mai. 45 Ma dimmi chi tu se', che in si dolente Loco se' messa, ed a si fatta pena, Che s' altra e maggio, nulla e si spiacente. 4 8 Ed egli a me : La tua citta, ch' e piena D' invidia si, che gia trabocca il sacco, Seco mi tenne in la vita serena. 51 Voi, cittadini, mi chiamaste Ciacco: Per la dannosa colpa della gola, Come tu vedi, alia pioggia mi fiacco; 54 Ed io anima trista non son sola, Che tutte queste a simil pena stanno Per simil colpa: e piu non fe' parbla. 57 Io git risposi : Ciacco, il tuo affanno Mi pesa si, che a lagrimar m' invita : Ma dimmi, se tu sai, a che verranno 60 Li cittadin della citta partita ? 40 chi se' P. per questo mondo O. 41 Diss' egli a me C. conoscime C. raconossimi M. 43 a lui CFIKQ*Z. B* has lei. 44 tiran L. 47 messo FGI. e a cosi f. EL. messa a cosi P. hai si f. BDIQZZ?. in si fatta C. 48 maggior D*EIMOP Q (magiur) a. E se alcuna e magg. o si spiac. F. fe piu spiac. IK. molto e dispiacente C. 49 ella O. 51 Teco Q. alia vita L. 52 che mi chiamasti DL(with 'che mi' bis). chiamasti EIMQZ. 54 alia spiaggia C (cf. piaggia 1. 101). alia piova a. 58 Io comin- ciai I. 59 che lagrimar L. 61 E citt. I. 47 We find hai and a similarly interchanged in xxxii. 136. If hai be read here, we may compare 1. 43 just above. 48 51 In Q these 3 lines were omitted, and afterwards supplied in the margin by the same hand. 49 Ella seems a hypercritical emendation to suit lei in 43 ; a change converse to that of several MSS. which have altered lei to lui there. 61 The MS. ' I ' has e for the plur. masc. of the article passim, e.g. Quali e fioretti ii. 127, Sotto e mia piedi xii. 30 &c. 44 INFERNO S' alcun v' e giusto : e dimmi la cagione, Perche 1' ha tanta discordia assalita. 63 Ed egli a me: Dopo lunga tenzone Verranno al sangue, e la parte selvaggia Caccera 1' altra con molta offensione. 66 Poi appresso convien, che questa caggia Infra tre soli, e che 1' altra sormonti Con la forza di tal che teste piaggia. 69 Alte terra lungo tempo le fronti, Tenendo 1' altra sotto gravi pesi, Come che di ci6 pianga, e che ne adonti. 72 Giusti son due, ma non vi sono intesi : Superbia, invidia ed avarizia sono Le tre faville che hanno i cori accesi. 75 Qui pose fine al lagrimabil suono. Ed io a lui: Ancor vo' che m' insegni, E che di piu parlar mi facci dono. 78 Farinata e il Tegghiaio, che fur si degni, Jacopo Rusticucci, Arrigo e il Mosca, E gli altri che a ben far poser gl' ingegni, 8r Dimmi ove sono, e fa ch' io li conosca; Che gran desio mi stringe di sapere, 62 E s' alcun F. n' e ADGMP. 63 Perche tanta discordia 1' ha Fla. Perche ha C. Perch' ell' e 'n t. d. salita GMQ. Perche la tanta disc, e salita OP. 64 E quegli (or quelli) ABCDKOy^a. da poi M. di po' B. 65 ala parte L. 67 Appresso poi I. 70 Alto DF. terram F. li fronti C. 71 1' altrui I. 72 E convien che ne pianga G. E convien che di piangi (sic) P. Come chi CO. e ch' io n' ad. F. v' adonti A. o che n' agonti (?) I. e chi n' adonti K. o che EHLOZa. 73 e non ABCEILMOPQa. 75 L' altre fav. L. 76 Quivi fin puose Q. Poi pose I. lagrimal D. 77 che 'n mi segni M. 79 e Teggh. CDFGOCD. Far. Teggh. I. si om. CO. di segni H. 80 e Rust. C. Herrigo P. 81 al ben DFIKMQZ^Z?. far preser E. 82 esta ch' i' li G (prob. clerical error for efa). 64 Q has deppo longa entenoe (sic). 72 Q (with both insertion and erasure) now reads pianghe e che ' ad., but this is evidently not its original reading. 79 If, as would appear from Inf. xvi. 41, Tegghiaio is not a surname, it would be more correct to omit the article here. CANTO VI. 45 Se il ciel gli addolcia o lo inferno gli attosca. 84 E quegli : Ei son tra le anime piu nere ; Diversa colpa giu li grava al fondo : Se tanto scendi, li potrai vedere. 87 Ma quando tu sarai nel dolce mondo, Pregoti che alia mente altrui mi rechi : Piu non ti dico e piu non ti rispondo. 90 Gli diritti occhi torse allora in biechi : Guardommi un poco, e poi chino la testa : Cadde con essa a par degli altri ciechi. 93 E il duca disse a me: Piu non si desta Di qua dal suon dell' angelica tromba ; Quando verra la nimica podesta, 96 Ciascun ritrovera la trista tomba, Ripigliera sua carne e sua figura, Udira quel che in eterno rimbomba. 99 Si trapassammo per sozza mistura Dell' ombre e della pioggia, a passi lenti, Toccando un poco la vita futura: 102 84 o inf. E. li toscha A. olninferuno (sic} L. 85 Ed egli (or elli) FGI^4a. Et egli a me : Tra 1' anime O. E quegli a me : Tra &c. D. E quilgli (sic\ Suono tra Q, 86 Diverse colpe ABCDEHIKLMOPQZa. Diverse pene G. piu gli grava I. gia gli gr. L. li tira EC li aggrava Z. 87 la i potrai ABEHOIM 2 ^a*. 88 sarai tomato al mondo FI. 89 Priego ch' alia C. che la mente L. m' arrechi G. 91 Gli occhi diritti I. allora torse FZ. 92 Guatommi a. 93 E cadde F. al par D 2 Z>. a om. Q. di altri A. 94 Lo duca disse FC. Lo buon Maestro a me OD. disse a lui B. 95 Da qua P. 96 vedra AFHLMOPQ^. vedrai Z. le nim. P. nimical F. 97 Ciascadun A. Che ciascuno rivedera ABDEGKLMQZZX rivedra CFHIOPa. 98 so carne et so figura E. suo (bis) K. 99 E vedrk quel M. 100 per trista mist. G. per forza mist. H. 101 Dell' ombra Aa. piaggia COM. 102 della v. ADKa. 86 It is difficult to account for the ungrammatical Diverse colpe found in many MSS., and many Edd. also. I suppose it was intended to be more severe. The diversa colpa however is the hateful sin referred to in Cantos xv and xvi. 96 B* has verra. 97 Zani de' Ferranti gives rivedera as reading of Vatican 'B'. Not so Witte. 46 INFERNO Perch' io dissi : Maestro, esti tormenti Cresceranno ei dopo la gran sentenza, O fien minori, o saran si cocenti ? 105 Ed egli a me: Ritorna a tua scienza, Che vuol, quanto la cosa e piu perfetta, Piu senta il bene, e cosl la doglienza. 108 Tuttoche questa gente maledetta In vera perfezion giammai non vada, Di la, piu che di qua, essere aspetta. m Noi aggirammo a tondo quella strada, Parlando piu assai ch' io non ridico : Venimmo al punto dove si digrada: u 4 Quivi trovammo Pluto il gran nemico. 103 Perchfe dissi C. io gli dissi I. dissi, di Maes. B*. I' cominciai G. Ch' io diss? di Maes. KQ. Et io a lui Maestro M. 104 Creseranci C. Cresseranno dapoi M. gran om. I. 105 fier CDHO. fir B. fie (sic) M. piu cocenti P. 108 e simile F. e om. E. e anco la G. e piu la P. 109 Tanto che KP. 110 Mudera (sic) perficion E. A vera perf. D. 112 intorno a quella I. a quella M. 113 assai cose ch' io F. assai piu E. che non a. dico DF. 114 Ven. in parte G. la disgrada C. si disgrada K. 116 Pruto L. 105 B apparently had originally chontenti, which is clumsily cor- rected to chocenti. From some MS. notes of Sir A. Panizzi, I find that this error (contend] occurs in a codice at Holkham, No. 515. 114 G's reading seems to be a reminiscence of Inf. iv. 151. Compare also a similar variant in v. 28. CANTO SETTIMO Pape Satan, pape Satan aleppe, Cominci6 Pluto colla voce chioccia. E quel Savio gentil, che tutto seppe, 3 Disse per confortarmi : Non ti noccia La tua paura, che, poter ch' egli abbia, Non ti torra lo scender questa roccia. 6 Poi si rivolse a quell' enfiata labbia, E disse : Taci, maledetto lupo : Consuma dentro te con la tua rabbia. 9 Non e senza cagion 1' andare al cupo : Vuolsi nell' alto la dove Michele Fe' la vendetta del superbo strupo. 12 Quali dal vento le gonfiate vele Caggiono avvolte, poiche 1' alber fiacca ; Tal cadde a terra la fiera crudele. 15 Cosi scendemmo nella quarta lacca, Prendendo piu della dolente ripa, Che il mal dell' universe tutto insacca. 18 I uleppe A*. 2 Plruto L. 4 Per confortarmi disse FI. 5 La sua Z. B* has tua, and so presumably sua i ma manu. per podere D. ch podere EQ. per poter P. 6 Non ci CQCZ>. ti terra B 1 (?)DK(?)Z^ 2 5Ca. ci terra AB 2 EHLMOPa. discender F*. della rocca C. d' esta G. 7 innate B. enfiate EZ. 8 mala- detta D. 11 Vuolsi cosi nell' alto ove F. Vuolsi cosi cola dove Q*. 12 obstrupo F. 13 del vento F. 14 avolto L. quando 1' alber F. 1' arbor BM. 1' arber GK(?)O. 1' albor PQ. 15 in terra F. 16 discendiamo O. 17 Pigliando ABCDEFGHIKLMOPQZCZ>a. 18 tutto sciacca a lt II The reading in Q is an evident adaptation to iii. 95 and v. 23. 17 Prendere and Pigliare are interchanged again in Purg. i. 108. 48 INFERNO Ahi giustizia di Dio, tante chi stipa Nuove travaglie e pene, quante io viddi ? E perche nostra colpa si ne scipa ? 11 Come fa 1' onda la sovra Cariddi, Che si frange con quella in cui s' intoppa, Cosl convien che qui la gente riddi. 24 Qui vid' io gente piu che altrove troppa, E d' una parte e d' altra, con grand' urli, Voltando pesi per forza di poppa : 27 Percotevansi incontro, e poscia pur li Si rivolgea ciascun, voltando a retro, Gridando : Perche tieni e perche burli ? 30 Cosl tornavan per Io cerchio tetro, Da ogni mano all' opposite punto, Gridandosi anche loro ontoso metro : 33 Poi si volgea ciascun, quando era giunto 19 Ad giust. D. A giust. BIL. tanta e M. quante chi F. che stipa DEIM. 20 Nuovi travagli BFIL. quanto viddi FL. 21 Perchfe la nostra G. vita se ne D. sene scipa ABCEFGHIKMPQa. sene istipa L. 22 1' umbra de sovra Caribdi E. 23 infrange BDK. in quella A. con om. M. che s' intoppa EF. frange quella che si L. 24 quella gente EL. 25 Li DFZA Z CD. Poi I. Quivi ELa^?). vidi g. ADEGHKLMOPQ.ffa. 26 Ball' una...e dalP altra A. dell' altra E. dalP altra K. e in altra C. 28 insieme B*F. incontra MQi(?)- poi pur li P. 29 rivolgean(o) Lja. tornando retro F. voltando retro I. a dietro GP. 30 Dicando A. Dicendo M. perch& vieni a. p. torni L. e om. G. o perche MZ>. 31 correano I. ritornavan E. cerchio retro E. 32 E da P. da ogni parte F. 33 Gridando a loro si anche E. Gridando ancora FIO. Gridavan si G. si lor anche C. anche il loro a. anco DM. anco il lor Z. 34 si om. M. rivolgea ACEHL. com' era I. 26 In C, E d' una is over an erasure, so perhaps C : had Ed in una to correspond with its own reading e in altra later in the line. 32 It is curious to note that in L, at xvii. 113 and again at xxii. 56, parte is changed to lato. K is very indistinct and hard to decipher here- abouts. 33 The order in E has been afterwards corrected by letters written above, as sometimes elsewhere. 34 Q has iunto, as else- where it has iu and id, for & iu and gia, and even irando in iii. 53. This omission of the initial g (owing no doubt to a local peculiarity of its pro- nunciation) is a fruitful source of ambiguous readings. (See Prolegomena^) CANTO VII. 49 Per lo suo mezzo cerchio all' altra giostra. Ed io che avea lo cor quasi compunto, 36 Dissi : Maestro mio, or mi dimostra Che gente e questa, e se tutti fur cherci Questi chercuti alia sinistra nostra. 39 Ed egli a me : Tutti e quanti fur guerci SI della mente, in la vita primaia, Che con misura nullo spendio ferci. 42 Assai la voce lor chiaro 1' abbaia, Quando vengono a' due punti del cerchio, Ove colpa contraria li dispaia. 45 Questi fur cherci, che non han coperchio Piloso al capo, e Papi e Cardinali, In cui usa avarizia il suo soperchio. 48 Ed io: Maestro, tra questi cotali Dovre' io ben riconoscere alcuni, Che furo immondi di cotesti mali. 51 Ed egli a me : Vano pensiero aduni : La sconoscente vita, che i fe' sozzi, Ad ogni conoscenza or li fa bruni ; 54 In eterno verranno alii due cozzi ; Questi risurgeranno del sepulcro 35 alta g. LP. 36 quasi lo cor L. 38 questa se tutti C. se fur tutti a. 40 tutti questi I. tutti furon g. E. 41 alia vita L. 44 giungono FO. al punto F. 45 contraria colpa F. colpe con- trarie L. la disp. C. 46 non a cop. O. 47 il capo DFGIZ*. capo, Papi ADFKLMOZ^CZ'a. Papa B. 48 us6 GMOZA^CD. 49 di questi F. 50 Dov. bene CI. Dovre' io conoscer bene ale. K. ben om. G. conoscer DP. ricognoscer ben B. 52 vani pensieri BGa. nuovo pensier K. vano i pensieri L. 53 conoscente Zj. e vizi sozzi F. 64 Et d' ogni K. choscienza L. or om. D. 56 E questi surgeranno BDKMOQa ('-"' added i nda manu in a), di sep. P. 38 L here and elsewhere regularly writes furon (orfurono, as in 1. 40) in spite of metre. So in lines 46, 5 1 &c. 43 abbaia is here used ' 81 vftpiv ' to express an idea similar to that assigned to grida in the note on i. 117. 52 The reading vani pensieri is probably a somewhat pedantic adaptation to the notion of plurality suggested by aduni. D. I. 4 50 INFERNO Col pugno chiuso, e questi co' crin mozzi. 57 Mai dare e mal tener lo mondo pulcro Ha tolto loro, e posti a questa zuffa: Qual ella sia, parole non ci appulcro. 60 Or puoi, figliuol, veder la corta buffa De' ben, che son commessi alia Fortuna, Perche 1' umana gente si rabbuffa. 63 Che tutto 1' oro, ch' e sotto la luna, O che gi& fu, di queste anime stanche Non poterebbe fame posar una. 66 Maestro, diss' io lui, or mi di' anche : Questa Fortuna, di che tu mi tocche, Che e, che i ben del mondo ha si tra branche ? 69 E quegli a me : O creature sciocche, Quanta ignoranza e quella che vi offende ! Or vo' che tu mia sentenza ne imbocche:- 72 Colui, lo cui saper tutto trascende, Fece li cieli, e die lor chi conduce, SI che ogni parte ad ogni parte splende, 75 Distribuendo ugualmente la luce: 67 This line is omitted in error by L. Con pugni chiusi DK. Coi pugni chiusi A Z CD. pugnio chiusi G. chiuso om. M. con crin M. 59 tolti ELP. tolto a loro O. posto L. in questa F. 60 parlare AG. non e pulcro A. nona li pulchro(?) a. non li pulcro C. ci pulcro DEFGHIMPQZ5. ci 'mpulcro O. noo(?) pulcro L. 61 Qui poi M. veder figliuol ACDEFGIK (fiol) LMPQZ^a. 62 Di ben F. commossi H. 63 Per cui Q. Onde 1' um. B. 65 E che ABCDEGHKLMOPQ^C'- chemaiF. 66 Nonnepotr. AEFHILPZ?. Non vi M. far posar pure una F. Non ne potrebbe far posar sol una G. sol far poser D. 67 Edio M. mioF. M. mio diss'io AEHILMPa. Et io a lui, Maestro or O. Et io M. dissi or D, dissi a lui C. dissi lui GZff. 69 che beni DI. che ben OZ. che '1 ben CK. L reads, Che e' ben del mondo asai trabache. 70 Ed elli D. Et egli AEFGHILMOP^CZ?. 72 Or voi HLM. che mia sent, tu F. che tua sentenza E. mia scienza M. non ne imb. L. ne om. P. 73 tanto transc. C. 74 a lor O. che cond. M. 76 in quel mente (sic) K. 61 The lines in M are misplaced thus, 61, 64, 65, 66, 62, 63, 67. 69 Che a. 105 Intra giu P. 106 In la pal. va EHOPQ. Nella pal. va IZAtf). Una pal. va Ga. ha om. M. 107 disteso L. 108 A pie FGIKLCZ?. malvage coste F. mal- vage piagge AIZ. e grige O. rigi (sic) P. glige CD. 109 mirare (i)stava EFGHPZ?. a mirare G. del mirar C. atteso PGD. 110 gente fangosa L. Ill Nude e graffiati con G. e om. BCLOQZA sembianti D. offese L. 112 Queste FIL. 113 colle teste D. e col petto BLQZ?CZ?a. con petto e con piedi C. testa col corpo G. con piedi M. 114 Traendosi G. con denti CEM. 115 figluol vedi BC. 117 Ed om. L. anco vo AGPZ. 118 acqua gente e che C. e gente che D. 106 F has istigC) this prefixed i being almost invariable in this MS. In the reading Una pal. va, va probably = v' ha as Witte suggests. C reads as in the text, but has manifestly been erased. ill The read- ing in G is evidently a reminiscence of xiii. 116. CANTO VII. 53 E fanno pullular quest' acqua al summo, Come 1' occhio ti dice, u' che s' aggira. 120 Fitti nel Hmo dicon : Tristi fummo Nell' aer dolce che dal sol s' allegra, Portando dentro accidioso fummo : 123 Or ci attristiam nella belletta negra. Quest' inno si gorgoglian nella strozza, Che dir nol posson con parola Integra. 126 Cosl girammo della lorda pozza Grand' arco tra la ripa secca e il mezzo, Con gli occhi volti a chi del fango ingozza: 129 Venimmo appie d' una torre al dassezzo. 119 pupolar L. in summo F. al fondo Z. 120 1' occhio ridice F. 1' occhio om. L. ci dice M. o che s' agg. AC. ove s' aggira G. unque L. ove segira M. 121 decontristi F. de contristi E. dicontristi H. di contristi M. 122 Nel mondo dolce F. del sol GKQ. 124 arrestiamo D. restiam G. ci tristian Q. nera L. 125 Questi si ghorgoglian D. si om, EH. E questo si gorgolgia C. Questi non si ghorghoglia L. se gorgola M. grorgolan (sic] Q. 126 non p. DP. intera F. 127 girando B*G. giognemo K. dalla lorda K. nella lorda Q. 128 dalla ripa K. della ripa L. al mezzo L. sega il mezzo Z. 129 Volgendo i volti F. involti L. di fango K. 130 al pi ABCEGHKLMOPQZ^PCZ'a. 120 B* has if che for something longer \ ma manu. CANTO OTTAVO Io dico seguitando, ch' assai prima Che noi fussimo al pie dell' alta torre, Gli occhi nostri n' andar suso alia cima, 3 Per due fiammette che i' vedemmo porre, E un' altra da lungi render cenno Tanto, ch' a pena il potea 1' occhio torre. 6 Ed io mi volsi al mar di tutto il senno; Dissi : Questo che dice ? e che risponde Quell' altro foco ? e chi son quei che il fenno ? 9 Ed egli a me: Su per le sucide onde Gia puoi scorger quello che s' aspetta, Se il fummo del pantan nol ti nasconde. 12 Corda non pinse mai da se saetta, Che si corresse via per 1' acre snella, Com' io vidi una nave piccioletta 15 Venir per 1' acqua verso noi in quella, Sotto il governo d' un sol galeoto, Che gridava: Or se' giunta, anima fella? 18 1 Io dico comincando (sic) G. Indico or P ridico (?) L. 2 a pie D (altered to at) FILA 2 C>, d' un alta C. 3 suso la A. su alia BM. in sulla C. n' andarono alia G. 4 che vedemmo BCFGIKLMPQZZ?. portte (sic) K. 5 dalla lungi F. di lungi Ka. 6 ch' appena la poteva F. puote C. pote G. poeta (sic) L. 7 tutto senno C. 8 E dissi F. 9 Quell' alto C. quei son quei K. e chi fur quei MQ. quel che L. 11 scorg. puoi EFHIOPQZ^a. che vi si F. 12 Al fumo L. tel nasc. FIMOZ. molti (cler. error) E. 13 pinse via D. 14 si om. P. IB Come vidi C. Com' io vid' io E. 17 Sotto gov. P. 18 giunto C. CANTO VIII. 55 Flegias, Flegias, tu gridi a voto, Disse lo mio signore, a questa volta : Piu non ci avrai, che sol passando il loto. i\ Quale colui, che grande inganno ascolta Che gli sia fatto, e poi se ne rammarca, Fecesi Flegias nell' ira accolta. 24 Lo duca mio discese nella barca, E poi mi fece entrare appresso lui, E sol, quand' io fui dentro, parve carca. 27 Tosto che il duca ed io nel legno fui, Secando se ne va 1' antica prora Dell' acqua piu che non suol con altrui. 30 Mentre noi corravam la morta gora, Dinanzi mi si fece un pien di fango, E disse : Chi se' tu che vieni anzi ora ? 33 Ed io a lui : S' io vegno, non rimango ; Ma tu chi se', che sei si fatto brutto ? Rispose : Vedi che son un che piango. 3 6 Ed io a lui : Con piangere e con lutto, Spirito maledetto, ti rimani : Ch' io ti conosco, ancor sia lordo tutto. 39 Allora stese al legno ambo le mani : 19 Fregias TB. Plegias I. Flegias (three times) G. Flegas K. tu si gridi CF. 20 Rispose el mio I. mio maestro G. 21 cha sol E. che '1 sol P. 'che' erased in G. 22 Quale e AZ. Qual e MPAD. gran dannaggio F. 24 Tal si fe FI. Si fece G. Fecesi tal Z. Cosi fece Fleg. D, 26 E poi fece entrarmi F. entrar mi fece G. appresso a lui BI. presso a lui G. 27 E solo ch' io fui F. quando fu F. quando fui MQZ. 29 A seconda LM. 30 Per 1' acqua I. 31 Mentre che corravam F. passavam A. 1' antica gora P. 32 si fe C\D. 33 Dicendo D. vieni aghorra (sic) L. chi vien P. 34 si vengo io non ACEFQ. si vengo non rim. BGHLMa. s' io vengo io non ZA. 35 sei fatto si br. F. si fatto se' I. si se' CEHKMPQZ5. che sei fatto O. 36 ch' io sono IKMOP. 37 piu lutto K. 38 or ti F. ten rimani M. 39 Ben ti cognosce O. Che ti CE. anche O. 40 Allor distese GA. 33 L has probably misread azhora. 56 INFERNO Perche il Maestro accorto lo sospinse, Dicendo : Via costa con gli altri cani. 42 Lo collo poi con le braccia mi cinse, Baciommi il volto, e disse : Alma sdegnosa, Benedetta colei che in te s' incinse. 45 Quei fu al mondo persona orgogliosa ; Bonta non e che sua memoria fregi : Cos! s' e F ombra sua qui furiosa. 48 Quanti si tengon or lassu gran regi, Che qui staranno come porci in brago, Di se lasciando orribili dispregi ! 51 Ed io: Maestro, molto sarei vago Di vederlo attuffare in questa broda, Prima che noi uscissimo del lago. 54 Ed egli a me : Avanti che la proda Ti si lasci veder, tu sarai sazio : Di tal disio converra che tu goda. 57 Dopo ci6 poco vidi quello strazio Far di costui alle fangose genti, Che Dio ancor ne lodo e ne ringrazio. 60 Tutti gridavano : A Filippo Argenti : E '1 Fiorentino spirito bizzarre 42 E disse F. va costk GZ. costk via C. 43 m' avvinse FI. 44 Basciandomi Q. 45 si cinse ACDILOPZCZ)a. 46 Quel fu DKPQ 2 #. Quegli fu M. 48 Cosi e C*EFIK. 49 di lassu F. colassu D. 60 Che quivi A. 51 Di loro F. 52 troppo sarei A. 53 vederla CL. tuffare B. quella D. 54 Anzi che B. Avanti che F. usciemo E. usiamo K. 56 tu om. A. 57 convien AEHLOP. 58 Dapuo cio puocho C. Da poi cio poco M. Di poco poi io D. io vidi ADEGHKOPQa. vidi io BIC. vidi in M. 59 colui F. dalle fang. K. 60 Chidio A. Che idio EFH (? = che i' Dio, or che iddio, prob. the latter). Che io ancor ne '1 lodo e '1 ne ringr. D. Che ancor ne '1 lodo L. Che Dio ne lodo ancor Z. e ne '1 ringr. B. et regratio E. ne li ingratio M. 61 cridavano Felipo K. 62 Al Fior. F. II Fior. MPa. bizardo K. 45 Lines 45 47 are omitted in G. 49 The terzina 49 51 is omitted in K. 51 Di se has had another s inserted later in a. See note on xvi. 67. CANTO VIII. 57 In se medesmo si volgea co' denti. 63 Quivi il lasciammo, ch6 piu non ne narro : Ma negli orecchi mi percosse un duolo, Perch' io avanti 1' occhio intento sbarro : 66 Lo buon Maestro disse : Omai, figliuolo, S' appressa la citta che ha nome Dite, Co' gravi cittadin, col grande stuolo. 69 Ed io : Maestro, gia le sue meschite La entro certo nella valle cerno Vermiglie, come se di foco uscite 72 Fossero. Ed ei mi disse : II foco eterno, Ch' entro 1' affoca, le dimostra rosse, Come tu vedi in questo basso inferno. 75 Noi pur giugnemmo dentro all' alte fosse, Che vallan quella terra sconsolata : Le mura mi parean che ferro fosse. 78 Non senza prima far grande aggirata, Venimmo in parte, dove il nocchier, forte, Uscite, ci grido, qui e 1' entrata. 81 Io vidi piu di mille in sulle porte Da' ciel piovuti, che stizzosamente 63 Che in se FM. se rodea C. si mordeva L. se mordean M. con- denti K. 64 Quivi lasc. ABCKL. e piu F. non vi B*C. ne narcho(?) K. 65 nelF orecchie ABCDEFHLMP (urechie) Qa. 66 attento 1' occhio I. intento 1' occhio CGHPQBZ5Z). 1' occhio avanti E. intento e stancho K. intento gli occhi L. 67 ormai I. 68 alia citta M. 69 Coi grandi C. Con gravi M. e grande F. col grave stuolo I. con grave stuolo KZ. e col gr. C. 71 La dentro EP. certeCGHLMP^4a. 73 Fossero e egli il fuoco eterno F (whole line}. 74 Entro la faccia le dimostre rosse F. entra E. dimo- stran M. 76 pur om. Q. venimmo E. altre CLP. 78 mure M. parea BF. 79 far prima CE. prima senza M. girata ACM. 81 Usciteci ACDFGHKLMOPQZa. quivi P. Grido a noi, qui e la vostra D. 82 Vidi piu C. 83 Dal ciel ACGIMOPQ. Dei ciel C. Del ciel a. piauri (?) L. 63 Several MSS. read z/a. incontro CEHOQ5. contra 7.A. a me A. 100 cosi diss' io K. 101 1' andar MZ. piu avanti F. 102 Ritorniam P. le vie nostre K. 104 disse non sgomentar F. 105 de tal C. c' e dato F. 106 che '1 nostro passo (obvious blunder) L. 104 The word sgomentar occurs again in Purg. xiv. 60, and is not infrequent in Boccaccio. [Derived, according to Diez, quasi ex com- mentare, f aus der Besinnung bringen.'] CANTO VIII. 59 Conforta e ciba di speranza buona, Ch' io non ti lascerc- nel mondo basso. 108 Cosl sen va, e quivi m' abbandona Lo dolce padre, ed io rimango in forse ; Che '1 si e '1 no nel capo mi tenzona. m Udir non pote' quel ch' a lor si porse: Ma ei non stette la con essi guari, Che ciascun dentro a prova si ricorse. 114 Chiuser le porte que' nostri awersari . Nel petto al mio signor, che fuor rimase, E rivolsesi a me con passi rari. 117 Gli occhi alia terra, e le ciglia avea rase D' ogni baldanza, e dicea ne' sospiri : Chi m' ha negate le dolenti case ? 120 Ed a me disse : Tu, perch' io m' adiri, Non sbigottir, ch' io vincer6 la prova, Qual ch' alia difension dentro s' aggiri. 123 Questa lor tracotanza non e nuova, Che gia 1' usaro a men segreta porta, La qual senza serrame ancor si trova. 126 Sopr' essa vedestu la scritta morta : E gia di qua da lei discende 1' erta, Passando per li cerchi senza scorta, 129 Tal che per lui ne fia la terra aperta. 107 da sper. L. 108 Che non CG. 109 mal mi dona K. 110 Lo dolce mastro P. ill Che si e no ADEFGHLOP^Z?a. Che '1 no e '1 si IZC. Che no e si BCKQ. Et si et no M. il capo LP. mi om. O. 112 ch' a lor porse ABEHKLMOQ#a. che lor porse P. 113 Ma non stette I. non fece L. li con M. 114 appruovo D. si ricolse KLA. 115 porti BCDLQZ. 116 al duca mio a. 117 E ritorno ver me F. Ed ei si volse G. 118 avean D. 119 con sospiri K. 120 negato IL. 121 Et ello a meTu E. io om. M. 123 Qualunqu' a C. Qual in la K. Qual che le L. Qual che la Q. 124 Questa tra costanza E. tragontanza D. stracuitanza M. non m' nuova I. 125 a me in EF 1 HO5a. a me segreta L. 127 Sopra vedetu E. vedesti tu BDKQ. 130 ne sie K. per lei C. la porta averta A. la porta ap. GCD, CANTO NONO Quel color che vilta di fuor mi pinse, Veggendo il duca mio tornare in volta, Piu tosto dentro il suo nuovo ristrinse. 3 Attento si fermo com' uom che ascolta ; Che 1' occhio nol potea menare a lunga Per P aer nero e per la nebbia folta. 6 Pure a noi converra vincer la punga, Cominci6 ei : se non...tal ne s' offerse. Oh quanto tarda a me ch' altri qui giunga ! 9 lo vidi ben, si com' ei ricoperse Lo cominciar con 1' altro che poi venne, Che fur parole alle prime diverse. 12 1 che vilta nel core I. 2 Vedendo MAD. 3 al suo I. Dentro il suo nonvi ristrinse is the whole line in L. 4 Intento DK. e como Q. 7 Pur a me Q. 8 sene s' off. D. se n' offerse O. 9 Ahi quanto tardi m' e F. tardi D. e tardi D. a me ananzi che qui A. altre B. 10 si come ricop. AEGHLMOP. si como ric. C. 11 L' un cominciar che 1' altro LM. El ricominciar D. L' in- cominciar D. altra C. pria venne B. 12 fu C. alia prima a x (?). In M there are here interposed four pages of a very closely written prose treatise upon several articles of the Christian faith, with the various errors and heresies related to them, apparently with a view to the sin of heresy dealt with in Cantos 9 and 10. 2 a is almost illegible here (as frequently), but the line does not seem to have ended with in volta. 8 Ei, el, etliand ello are found here, but these interchanges will not generally be noticed hereafter, nor similarly those of quei and quel. The latter change (sometimes obviously 2*a* " 86 Volsimi AEHLOP. Volsimi al mio Maes. Aj. e quei ACFGHLOPQZ^a. e quel EIM. 87 Che stessi E. stessi fermo I. 88 e' mi parea a. pien di sdegno LQi. 89 Giunse IZ. e om. FLQ. 90 V aperse Z. v' ebbe BCEHLOPQZZ). 91 A cacciate B. dal ciel BDIKLPZ^. da ciel a* (comp. 1. 85). 92 Incomincio A. orribili solgla Q. 93 Un.de stra (sic) A. Onde e sta E. Onde istra L. tralgotanza D. tracorranza F. tracotanza ABGIKMy^Z). tracontanza PQ. 89 Giunse for Visse occurs in some MSS. in Purg. xxi. 101, aggiunse for awinse in Inf. xvii. 96, and giugnere is interchanged frequently with venire as h. /. (see note on iv. 1 1 1). 93 Tracorranza in F is -perhaps intended for iracuranza, a reading registered in one MS. by Scarabelli. For the forms stra, istra, c. comp. Inf. xxvii. 21. CANTO IX. 65 A cui non puote il fin mai esser mozzo, E che piu volte v' ha cresciuta doglia ? 96 Che giova nelle fata dar di cozzo ? Cerbero vostro, se ben vi ricorda, Ne porta ancor pelato il mento e il gozzo. 99 Poi si rivolse per la strada lorda, E non fe' motto a noi : ma fe' sembiante D' uomo, cui altra cura stringa e morda, 102 Che quella di colui che gli e davante. E noi movemmo i piedi in ver la terra, Sicuri appresso le parole sante. 105 Dentro v' entrammo senza alcuna guerra : Ed io, ch' avea di riguardar disio La condizion che tal fortezza serra, 108 Com' io fui dentro, 1' occhio intorno invio ; E veggio ad ogni man grande campagna Piena di duolo e di tormento rio. m Si come ad Arli, ove Rodano stagna, Si com' a Pola presso del Quarnaro, Che Italia chiude e suoi termini bagna, n 4 Fanno i sepolcri tutto il loco varo : 95 A cui il fine mai puo esser F. A cui '1 fine non puo mai a. puo BH. Cui non puo al fine essere L. A cui il fine mai non puo O. Alia qual non puo mai '1 fin A v . 96 E chi L. cresciuto BEFIKMZ*. 97 nella fata D. nelP effette F. nelle facta E. nella fate G. nelle fate LPZ. a dar C. 98 vo ricorda a. 99 el mento el gozzo DIQ. e gozzo Z. 101 Ma non K. 102 D' uomo '1 quale D. 104 il pie Z?!. 106 n' entrammo C. li (or gli) entr. AEHOPQ. senza niuna L. 107 che di mirare avea F. di guardar disio E. di riguarda (sic) a. 109 Come fui DKM. gli occhi I. intorno vio C. 110 E vidi ACDFMQ^ZCa. d' ogni mano gran YA^. compagnia EG. compagna DD. Ill pio (?) L. 112 Argli L. il Rod. BDIMZC irodano a. 113 E come FGI. come Pola C. appresso IZ. Carnaro BCDFGHMQZ. dal Carn. L. dal Quarn. D. 114 Che om. G. e i suoi LP. 115 Fanno sep. MQ. tutti LMOQ. in loco DFGMj5. i loco (? il or in} L. il lito CD. 106 For gli= vi, see Purg. xiii. 7, &c. 115 Q has T loco, which in this MS. may stand for il or in. D. I. 5 66 INFERNO Cosi facevan quivi d' ogni parte, Salvo che il modo v' era piu amaro ; 117 Che tra gli avelli fiamme erano sparte, Per le quali eran si del tutto accesi, Che ferro piu non chiede verun' arte. 120 Tutti gli lor coperchi eran sospesi, E fuor n' uscivan si duri lamenti, Che ben parean di miseri e d' offesi. 123 Ed io : Maestro, quai son quelle genti, Che seppellite dentro da quell' arche Si fan sentir con gli sospir dolenti ? 126 Ed egli a me : Qui son gli eresiarche Co' lor seguaci d' ogni setta, e molto Piu che non credi, son le tombe carche. 129 Simile qui con simile e sepolto, E i monimenti son piu, e men caldi. E poi ch' alia man destra si fu volto, 132 Passammo tra i martiri e gli alti spaldi. 116 faceva FL. quinci A. 117 modo era K. avaro (sic] K. 118 Che 'ntra D. 119 Per le qua (sic) D. incesi M. 120 a verun' arte F. in veruna GI. 121 i coperchi loro F. 122 Che fuor C. n' usciro C. usciva L. grandi lam. D. si gravi K. 123 parea CL. miseri offesi DL. e offesi F. da miseri e da offesi C. 124 quali D. qual E. chi son YD. queste g. BGKCD. 125 Che son sepolte G. Chen seppelite K. son dentro a quelle M. queste arche BKZC. 126 Si fa Q. E sentir fansi con sospir G. sospiri a. spiriti dolenti L. 127 E quegli D. Quei son A v le eres. A. 128 Con lor CHMOPZ?. d' ogni parte F. e d' ogni GIKZ. sette G. 130 Sim. con sim. qui F. e om. M. 131 E mon. ACEFGHIMOPQZ.SCa. movimenti LC. E molimenti K. piii e morti caldi (?) L. e piu e men Q. 132 Poscia ch' alia B. E om. Q. mi fui a*. 133 i om. ACEHILOPZ. altri QDGIKLMOP Q t ZCa. (In C and Q altri has been altered to alti.} e alti A. 117 a has an erasure and blank between salvo and z/' era. 125 In K, chen=cK en=cK enno (see v. 38). Thus the reading would practically be the same as in G. 126 L seems to be thinking of i. 116. 133 I shall not generally record such doubtful variants as tra for trai, or e for ei or ^ (see 1. 131), since many MSS. frequently omit a final i in cases where they must have understood it to exist. Thus it is often a question of mere orthography. CANTO DECIMO Ora sen va per un secreto calle Tra il muro della terra e H martiri Lo mio Maestro, ed io dopo le spalle. 3 O virtu somma, che per gli empi giri Mi volvi, cominciai, com' a te piace Parlami, e satisfammi a' miei desiri. 6 La gente, che per li sepolcri giace, Potrebbesi veder? gia son levati Tutti i coperchi, e nessun guardia face. 9 Ed egli a me : Tutti saran serrati, Quando di Josaffat qui torneranno Coi corpi che lassu hanno lasciati. 12 Suo cimitero da questa parte hanno Con Epicuro tutti i suoi seguaci, 1 si va G. uno stretto K. 3 direto le spalle I. per le suo K. dopo le sue (sue underlined) a. 4 ampi AD. impi CEI. imperii ziri K. 5 come ti A. piaque K. 7 Le gente K. sepolti L. ziaze (sic} K. 8 Potrebbensi (-onsi, or -ersi) CDG^LZte,. sollevati D. 9 i sepolcri B. iface M. 10 E quegli BCEHMP. E quei a me A. E queli anime L. quili Q. 11 Josafa (or -pha) AKLOP5. Josaffa BH. Giusaffat D. Giosafa GQa^ Josaphat M. 12 Con corpi M. Li corpi a. 13 Suoi cimiteri Z. 14 Come Ep. F. Cone picuro (sic) G. e tutti DF. tutti suo AEM. tutti suoi HKLOP. sua I. 1 K's reading is perhaps suggested by xviii. 100. 7 Q has morta in margin with a mark of omission before giace. 9 In M iface prob. = f face, i.e. ibi. Cf. f vedemmo in viii. 4. 14 The form mia, sua, &c. for nom. plur. (masc. and fern.) is usual in I, e.g. in 1. 39, Le parole tua. Cf. 1. 47 in G. 52 68 - INFERNO Che F anima col corpo morta fanno. 15 Pero alia dimanda che mi faci Quinc' entro satisfatto sarai tosto, Ed al disio ancor che tu mi taci. 18 Ed io: Buon Duca, non tegno riposto A te mio cor, se non per dicer poco; E tu m' hai non pur mo a ci6 disposto. i\ O Tosco, che per la citta del foco Vivo ten vai, cosl parlando onesto, Piacciati di restare in questo loco. 24 La tua loquela ti fa manifesto Di quella nobil patria natio, Alia qual forse io fui troppo molesto. 27 Subitamente questo suono uscio D' una dell' arche: pero m' accostai, Temendo, un poco piu al duca mio. 30 Ed ei mi disse : Volgiti : che fai ? Vedi la Farinata che s' e dritto : Dalla cintura in su tutto il vedrai. 33 I' avea gia il mio viso nel suo fitto ; Ed ei s' ergea col petto e colla fronte, Come avesse Io inferno in gran dispitto : 36 16 Poi la dimanda chenne L. che tu faci GK. 17 Qua entro F. Qui dentro A^D. sara ADLOQ. saro P. 19 Ed io Maestro F. io non tegno KQ. risposto E^Ca. nascosto ABFZ>. 20 Ate michuro (sic) L il mio cor E. mio dir ZJ5. parlar poco F. 21 pur non m' ai mo a cio Ej (but order corrected later by letters written over), non m'hai pur A^D. 24 restart! K. da restare L. 27 forse fur (or fu) troppo ACDEGH I KMOPZ^Z?a. forse fu' io FQ. fui forse B* (but apparently prima manu fu' io forse). forse fu non troppo modesto L. 29 perch' io m' ace. BF. pero io A. e pero GP. 30 piu un poco C. Un poco piu temendo M. el duca EK. 31 volviti ABK8CD. Et egli a me, Rivolgiti G. 32 Io Far. ABCO*. la e Far. E. che e E. 33 cintolaABDEFGHKLMOPQZ.frDa. 35 surgea ADFI. E s' ergea L. 36 il ninferno a. a gran AEGKLMQ^. 19 Cf. the converse change made by F in 1. 115. The Chiosatore in E has a note ' al. nascosto? 33 K writes zinltola. CANTO X. 69 E 1' animose man del duca e pronte Mi pinser tra le sepolture a lui, Dicendo : Le parole tue sien conte. - 39 Com' io al pie della sua tomba fui, Guardommi un poco, e poi quasi sdegnoso Mi dimand& : Chi fur li maggior tui ? 42 Io, ch' era d' ubbidir desideroso, Non gliel celai, ma tutti gliel' apersi : Ond' ei levo le ciglia un poco in soso; 45 Poi disse : Fieramente furo avversi A me ed a' miei primi ed a mia parte, SI che per due fiate gli dispersi. 4 8 S' ei fur cacciati, ei tornar d' ogni parte, Rispos' io lui, 1' una e 1' altra fiata ; Ma i vostri non appreser ben quell' arte. 51 Allor surse alia vista scoperchiata Un' ombra lungo questa infino al mento : Credo che s' era in ginocchie levata. 54 D' intorno mi guard6, come talento Avesse di veder s' altri era meco ; Ma poi che il suspicar fu tutto spento, 57 37 duca pronte I. 38 pinse ADM. prese L. tra sepult. E. et lui KM. 40 E com' io E. a pie ILZA 2 CD. 41 e om. Q. 42 chi son D. 43 Io era d' un bel dir desideroso L. 44 Non li (or gli) celai EDO. Non il C. tutto BCFGIKLMZ5C7?a. gli ap. AM. li ap. BK. i ap. C. glielo ap. G. 45 Onde levo DLPa. Onde elev6 QI. eli levo K. el levo M. in om. CK. suso ABCDEFLQ. 47 e mia primi G. a mia parente L. 48 io gli F. 49 Se fuor CDMQa. Se fur AEFGHILP. cacciati tornar C. torna M. 50 Risposi a lui CDIKO. Risposi lui FGHP. Risposi 1' una L. Risposi io a lui MZ. 51 quell' altre L. 53 questo C. insino Da. fin al E. fino al FM. infino el K. 54 in ginocchi DA V in ginocchia BFGLZ?. inginocchioni I. inginocchio Z. inginochion EMQ 2 . 57 E poi ACDEFGILOQZX sospic(c)iar CD^LOa. sospecciar FQZ^C. solspiccar G. che sosp. Z. 49 I has etoriiar=ei tornar. Hence elei'b in 1. 45 is not conclusive, nor sc fur in this same line. So perhaps Ma iiostri in 1. 51 in M, and many similar cases. 70 INFERNO Piangendo disse : Se per questo cieco Carcere vai per altezza d' ingegno, Mio figlio ov' e, e perche non e teco ? 60 Ed io a lui : Da me stesso non vegno : Colui, che attende la, per qui mi mena, Forse cui Guido vostro ebbe a disdegno. 63 Le sue parole e il modo della pena M' avevan di costui gia letto il nome : Per6 fu la risposta cosl piena. 66 Di subito drizzato grido: Come Dicesti : egli ebbe ? non viv' egli ancora ? Non fiere gli occhi suoi lo dolce lome ? 69 Quando s' accorse d' alcuna dimora Ch' io faceva dinanzi alia risposta, Supin ricadde, e piu non parve fuora. 72 Ma quell' altro magnanimo, a cui posta Restate m' era, non mut6 aspetto, Ne mosse collo, ne piego sua costa. 75 E se, continuando al primo detto, 59 1' altezza CL. 60 figliuolo E. fio K. ov' e perche non teco F. ov' e perche LMP. e con teco M. non e ei (el) teco HOPa. 61 Et io risposi a lui, Da me non vegno F. 63 Forse che C. nostro L. a sdegno E. 64 mondo (corrected to modo) G. mondo L. 65M'aveaAF. gia di costui DK. eletto ADFHKLO. detto IZ. 67 Di om. G. Di sub. levato K. e (? = ei) grido F. disse Come Z. 68 e non F. 69 fieron BF. negli occhi Q. le dolze K. 71 Ch' io gli facea P. intorno alia F. 72 Supino cadde G. poi non K. non fe' parola G. 74 1' aspetto C. 75 ne muto sua L. 69 Nearly all the MSS. here have the more familiar lume in spite of rhyme. Cf. suso in 1. 45, &c. So also sotto for sutto in xi. 26 in most MSS. On this however consult Nannucci's Saggio intorno alle voci, &c. (P- 37)- He thinks that o and u were formerly so nearly alike in pro- nunciation that such changes as nut, vut, lome, soso are needless orthographic changes of copyists or editors. Possibly this may explain u son for o son in 1. 103 in 'B.' Comp. nodrl and nudrl in xii. 71. 72 G's curious alteration in violation of rhyme is a reminiscence of vi. 57. See another instance in this MS. at vii. 1 1 j. CANTO X. /I S' egli han quell' arte, disse, male appresa, Ci6 mi tormenta piu che questo letto. 78 Ma non cinquanta volte fia raccesa La faccia della donna che qui regge, Che tu saprai quanto quell' arte pesa. 81 E se tu mai nel dolce mondo regge, Dimmi, perche quel popolo e si empio Incontro a' miei in ciascuna sua legge ? 84 Ond' io a lui : Lo strazio e il grande scempio, Che fece 1' Arbia colorata in rosso, Tale orazion fa far nel nostro tempio. 87 Poi ch' ebbe sospirando il capo scosso, A cio non fui io sol, disse, ne certo Senza cagion con gli altri sarei mosso : 90 Ma fu' io sol cola, dove sofferto Fu per ciascun di toglier via Fiorenza, Colui che la difesi a viso aperto. 93 Deh, se riposi mai vostra semenza, Prega' io lui, solvetemi quel nodo, Che qui ha inviluppata mia sentenza. 96 E' par che voi veggiate, se ben odo, 77 Egli han B. S' elli in quella disse male ap. C. S' elli (or egli) in quell' EH. Se li aquel...mal apresa L. impresa a. 79 racceso D. 81 quanto chelarte C. presa L. 83 qual L. 84 Incontro a' miei et a ciasc. E. Incontra a ciascuna sua legge L. in tutte sue legge A. 85 Et io ACD. crudo scempio Z. esempio EK. Io gran strazio et Io scempio M. 86 colorare DFC. 87 fan far G. vostro AMOQa^?). 88 sospirate el (? = e '!) CELOPQa. mosso MOPQa. 89 Disse, a cio non IZ. disse om. LPQ t . solo ancho ne P. io solo ne certo E. 90 sarei cogli altri F. 91 Ma fui solo AQa. solo Ik CEGLMP. 92 di ciascun C. torre via ACDHKLMOPQZZ)a. tor FIC. toller B. de doller (sic) G. di torvi a Fir. (sic) E. 93 difese P. 94 rispose GP. omai OP. 95 Priegha lui L. Pregai lui Q. 96 Che piu a E. c' inviluppa F. a aviluppata P. viluppata A. inviluppate C. inviluppato ILMZZ>. 97 s' io ben KM. 88 Q has sospirato. 89 disse omitted in Q 1? but inserted later above. 72 INFERNO Dinanzi quel che il tempo seco adduce, E nel presente tenete altro modo. 99 Noi veggiam, come quei ch' ha mala luce, Le cose, disse, che ne son lontano ; Cotanto ancor ne splende il sommo Duce: 102 Quando s' appressano, o son, tutto e vano Nostro intelletto ; e s' altri nol ci apporta, Nulla sapem di vostro stato umano. 105 Per6 comprender puoi, che tutta morta Fia nostra conoscenza da quel punto Che del futuro fia chiusa la porta. 108 Allor, come di mia colpa compunto, Dissi: Or direte dunque a quel caduto Che il suo nato e co' vivi ancor congiunto. m E s' io fui innanzr alia risposta muto, Fat' ei saper che il fei, perche pensava Gia nell' error che m' avete soluto. 114 E gia il Maestro mio mi richiamava : Perch' io pregai lo spirto piu avaccio 98 D. il tempo quel che D. D. quel che tempo CZ. D. a quel MD. D. a quel che tempo L. 99 E om. L. Che nel P. 100 comequel(lo) ADHIOP. quei ch' an GKQ (?). ha la mala E. che mala M. 101 disse, ch' erano e son F. non son Z^a,.,. 102 Tanto F. Tanto ne splende ancora M. al sommo D. 103 s' appr. sono tutte vano A. s' appr. n' e del Z. s' appressa divien tutto vano M. u son B (see note on 1. 69). e son P. tutte in vano F. o vano Ca. tutto vano ZS. 104 e om. K. altro L. non ci ACFGHILMOPQZ^a. 107 Sia K. vostra E. 108 dal f. D. sia CD. ne fia E. 110 Diss' io ACDGHLMPQ. Diss' io direte G. Or dicerete dunque a quel B. Dissi or ci dicerete adunque I. dunque om. ACEHLMOPQ. Ill Che '1 nato suo M. Che '1 suo figlio O. con vivi AMPQ. tra' vivi B. ancor giunto B. e om. L. 112 E io L. dinanzi AEHLa. dianzi GIOZ. denanti Q. dinanzi la M67?. 113 Fategli a saper A. Fagli a saper DI. Fateli saper HL. Fatil saper O. Fatei a saper KQ (altered ~2. da manu tofatti in K). Fotti ad saper Z. Fategli saper che gia pensava F. che '1 feci che COQ t (?) (Q* has ch' io) a. che fu BG. che '1 fu K. ch' io il fici ch' io pensava M. perch' io pensava ABGIK. 114 Entro 1' error F. solvuto AD. assoluto K. 115 E gih il duca mio F. 116 io om. M. piu om. I. CANTO X. 73 Che mi dicesse chi con lui si stava. 117 Dissemi : Qui con piu di mille giaccio : Qua dentro e lo secondo Federico, E il Cardinale, e degli altri mi taccio. 120 Indi s' ascose : ed io in ver 1' antico Poeta volsi i passi, ripensando A quel parlar che mi parea nimico. 123 Egli si mosse ; e poi cosl andando, Mi disse : Perche sei tu si smarrito ? Ed io li satisfeci al suo dimando. 126 La mente tua conservi quel ch' udito Hai contra te, mi comando quel Saggio, Ed ora attendi qui : e drizz6 il dito. 129 Quando sarai dinanzi al dolce raggio Di quella, il cui bell' occhio tutto vede, Da lei saprai di tua vita il viaggio. 132 Appresso volse a man sinistra il piede : Lasciammo il muro, e gimmo in ver lo mezzo Per un sentier ch' ad una valle fiede, 135 Che infin lassu facea spiacer suo lezzo. 117 si om. MQa. 118 Risposemi, con piu G. 119 Qui dentro K. dentro v' e A. 120 di altri A. 122 mossi i piedi F. 123 chomio (?) L. 124 Et e' (or el) si mosse GIL. mosse poiZ>. 125 Ne disse B. Mi om. Z. se' cosi sm. DGIM. se' tu cosi AEZ. 128 contro a CFG I. ne comanda ACH. ne comando LQ X . 129 Ed om. K. a te da cui C. attendi a cui M. 130 davanti M. dinanti (as in 1. 112) Q. 131 Delia donna il cui F. beli ochi (sic) L. 132 Di lei L. 133 Appresso mosse BG. 134 Lasciando K. in om. FM. inverso il L. 135 che una valle IM. cha una K. siede Z. 136 Che fin EGLM. fata (?) L. spicciar CDFLP. spicar C. 1. 136 is omitted altogether in K. 122 F's reading is bad because mosse occurs two lines below. Com- pare the var. lect. in B and G at 1. 133. 129 Is C's an oral blunder? In su 1' estremita d' un' alta ripa, Che facevan gran pietre rotte in cerchio, Venimmo sopra piu crudele stipa: 3 E quivi, per 1' orribile soperchio Del puzzo, che il profondo abisso gitta, Ci raccostammo dietro ad un coperchio 6 D' un grande avello, ov' io vidi una scritta Che diceva : Anastasio papa guardo, Lo qual trasse Fotin della via dritta. 9 Lo nostro scender conviene esser tardo, SI che s' ausi un poco prima il senso Al tristo fiato, e poi non fia riguardo. 12 Cosl il Maestro ; ed io : Alcun compenso, Dissi lui, trova, che il tempo non passi Perduto ; ed egli : Vedi che a ci6 penso. 15 Figliuol mio, dentro da cotesti sassi, Cominci6 poi a dir, son tre cerchietti 1 altra BE t Q. 2 faceva AFHILOPZ^a. facie Q. rotto G. 3 sopra a L. 5 che profondo L. 6 Ci ritraemo F. ricostamo C. indietro ad un ABDGHPQ^. a drieto E*. 7 Ad un grande G. ove vid' io M. dov' io A Z C. e vidivi una a. 9 misse Fot. K. Fortin(o) CDLP. Futino F. dalla EK. ritta D. 11 Finche D. prima un poco BGIKMZ>. inprima un poco AEHLOPZ. in- prima il Oa. un poco il schussio (sic} K (?). 12 tristo puzzo F. non fe M. fa Q t . 14 Dissi a lui IO. Diss' io ABEHKPQCa. Diss' io a lui L. Dissi trova M. trovava E. 15 egli a me, Vedi E. E elli a me Ve' Q*. ch' io penso C. 16 A figliuol G. 17 Comincib egli a F. 15 Q t seems to have read simply ' El a me Ye 1 ? CANTO XI. 75 Di grado in grade, come quei che lassi. 18 Tutti son pien di spirti maledetti : Ma perche poi ti basti pur la vista, Intendi come, e perche son costretti. -zi D' ogni malizia, ch' odio in cielo acquista, Ingiuria e il fine, ed ogni fin cotale O con forza o con frode altrui contrista. 24 Ma perche frode e dell' uom proprio male, Piu spiace a Dio ; e pero stan di sutto Gli frodolenti, e piii dolor gli assale. 27 De' violenti il primo cerchio e tutto : Ma perche si fa forza a tre persone, In tre gironi e distinto e costrutto. 30 A Dio, a se, al prossimo si puone Far forza, dico in loro ed in lor cose, Come udirai con aperta ragione. 33 Morte per forza e ferute dogliose Nel prossimo si danno, e nel suo avere Ruine, incendi e toilette dannose : 36 Onde omicide e ciascun che mal fiere, Guastatori e predon, tutti tormenta 23 Ingiuria al fine C 1 (?)F. Ingiuria il fine M. 24 E con forza e L. O con froda o con forza OD. 26 e om. L. sta M. percio C. 27 Lo frodolente L. gli sale AM. li salle E. 29 in tre A v 30 distinto confruto (or consruto (?)) L. 31 e a se e al pr. F. e al pross. P. 33 in aperta K. 34 Morta P. fedite L. ferite M. toilette dannose F (an obvious blunder, see 1. 36). 35 Al proximo O. e om. L. 37 Odio e micidio F. Odii micidii C (omicidii 2 nda manti). Odii omicidii AEGHKMPQZa. Odii chominciadii (?) L. Onde homicida O. 38 Guastori P. prodoni E. pedoni L. 23 C seems to have had alfine altered into e } I fine. 27 This is perhaps for assale in A, as in this MS. the final vowel of a crasis is gener- ally the one elided. In the commentary here it appears as assale, though this is not conclusive, as it is not uncommon for the readings of a commentary and the accompanying text to differ [comp. on this Witte Proleg. p. Iviii]. See also notes inj. on 11. 55, 65, 105. 37 40 Here as elsewhere L abounds in mere careless clerical errors, which it would be useless to reproduce. 76 INFERNO Lo giron primo per diverse schiere. 39 Puote uomo avere in se man violenta E ne' suoi beni : e per6 nel secondo Giron convien che senza pro si penta 42 Qualunque priva se del vostro mondo, Biscazza e fonde la sua facultade, E piange la dove esser dee giocondo. 45 Puossi far forza nella Deitade, Col cor negando e bestemmiando quella, E spregiando natura e sua bontade : 48 E per6 lo minor giron suggella Del segno suo e Sodoma e Caorsa, E chi, spregiando Dio, col cor favella. 51 Lo frode, ond' ogni coscienza e morsa, Pu6 1' uomo usare in colui che 'n lui fida, Ed in quei che fidanza non imborsa. 54 Questo modo di retro par che uccida Pur lo vinco d' amor che fa natura ; Onde nel cerchio secondo s' annida 57 40 in se avere Q. in sua I. violente C. 41 e om. L. 42 si senta H. 43 nostro BC(flrimd mamu)MPZA. di questo mondo E. 45 dee esser D. 47 dispregiando (corrected 2 rfa manu to bestemmiando) F. 48 I spregiando F. spixiando K (spressiando 3 lines below). e om. a. 50 ist e om. FMO. 51 E che L. Idio ADFHIKLMQZ5CZ?. 52 Le fraude a. frode ontte (sic) cose. K. 53 Puote uomo F. che lui fida B. che si fida DFIM. che lui si fida G. se 'nfida K. in chi fida C. cui lui A. chi in lui E (se erased), in cui fida LQ*. in cui se fida P. in quei che C. che lui fida a. 54 in quei ABCDEFGHIKOPQZa. in om. L. non ne CL. 55 E questo L. di sotto A. di dietro BFZa*. incida A. ancida^. 56 Pur lo nimico d' amor DLa. Pur lo vincol EFGIKPZ. vinclo B. C* has 'vinco,' but apparently \ ma manu 'nimico.' 46 In H after line 45 there follow in error lines 88 101 inclusive, afterwards lightly erased. Probably the copyist turned over two pages by mistake. 53 K t seems to have had colui in cuy. Then che has been added above, and cuy altered to luy. 55 Benvenuto has ' di retro ' in the Comm. in A. CANTO XI. 77 Ipocrisia, lusinghe e chi affattura, Falsita, ladroneccio e simonia, Ruffian, baratti e simile lordura. 60 Per 1' altro modo quell' amor s' obblia Che fa natura, e quel ch' e poi aggiunto, Di che la fede spezial si cria : 63 Onde nel cerchio minore, ov' e il punto Dell' universe, in su che Dite siede, Qualunque trade in eterno e censunto. 66 Ed io : Maestro, assai chiaro precede La tua ragione, ed assai ben distingue Questo baratro e il popol che il possiede. 69 Ma dimmi : Quei della palude pingue, Che mena il vento, e che batte la pioggia, E che s' incontran con si aspre lingue, 72 Perche non dentro dalla citta roggia Son ei puniti, se Dio gli ha in ira ? E se non gli ha, perche sono a tal foggia ? 75 Ed egli a me: Perche tanto delira, Disse, lo ingegno tuo da quel che suole ? Ovver la mente dove altrove mira ? 78 Non ti rimembra di quelle parole, 58 lusingha B. chi a factura (sic) H. 69 ladrocinio G. 60 barattieri B(with -eri erased later)LP. barattera G. 61 Per altro E.IPHCD. mondo K. 62 e om. L. che poi b giunto A. e quei che poi v' & giunto F. che poi ha g. A v 63 Onde la D. 64 Onde quel Q. minor cerchio C. 65 in su om. A (? clerical error), sul qual dico sede M. 66 e om. L. 67 chiara ABCEGHIOPQZa. 69 baratto FC (apparently \ ma manu} a. bar. al popol M. che pos- siede AGKLMPZa. 70 Ma di' de quei che la M. che la F. 71 Chei mena e chei B. abbatte E. e combatte D. 72 incontra ACEKLMOPQ^a. cosi aspre DF. 73 della ELQ. dentro alia cittade G. 74 Sono pun. CEGQ. have in ira DP. gli anno M. 75 sono ei K. foglia M. 77 ing. mio L. di quel G. che '1 sole M. 78 o dove D. mente tua altrove G. altro M. 62 Z has adiuto (for aggiunto). This evanescence of g (which is, I believe, Venetian) occurs however most frequently in Q. 65 Benvenuto has ' in su che' 1 when quoting in the notes. 78 INFERNO Colle quai la tua Etica pertratta Le tre disposizion che il ciel non vuole : 81 Incontinenza, malizia e la matta Bestialitade ? e come incontinenza Men Dio offende e men biasimo accatta? 8 4 Se tu riguardi ben questa sentenza, E rechiti alia mente, chi son quelli, Che su di fuor sostengon penitenza, 87 Tu vedrai ben, perche da questi felli Sien dipartiti, e perche men crucciata La divina vendetta gli martelli. 90 O Sol che sani ogni vista turbata, Tu mi contend si, quando tu solvi, Che, non men che saper, dubbiar m' aggrata. 93 Ancora un poco indietro ti rivolvi, Diss' io, la dove di' che usura offende La divina bontade, e il groppo solvi. 96 Filosofia, mi disse, a chi la intende, Nota non pure in una sola parte, Come natura lo suo corso prende 99 Dal divino intelletto e da sua arte ; E se tu ben la tua Fisica note, Tu troverai non dopo molte carte, 102 80 Colle qual K. Delle qual M. 82 e malizia C. 83 Bestialita ABCFGI. 84 Pero men Dio H. piu bias. Z. 86 E rechi alia L. E recati M. 88 di La. 89 Son EFM. e om. L. e percW men KZ. crucciati M. 90 div. giustizia DFGIKZ^Z?. 91 faccia turb. F. ogni cosa A. 92 sollevi E. 93 cha sapere Q. me grata M. 94 indrieto un poco L.A. a dietro D. 95 did 'che s' offende DK. usura dichon fonde (sic} L. di caorsia off. a*. (Cf. 1. 50.) 96 bonta ACDEFIK. isvolvi D. svolvi H. aasolvi K. dissolvi C. 97 1' attende DGI. lattende B. 98 No ti no pur (sic) M. 99 corpo EP. apprende F. verso pende I. 100 Al div. C. et a sua C. et e suo K. 102 da poi M. 95 C has had some other (apparently longer) reading erased under the words che usura. G has dicu usura = (I suppose) df ctt usura. 96 See a similar variant in K in x. 114. CANTO XI. Che T arte vostra quella, quanto puote, Segue, come il maestro fa il discente, Si che vostr' arte a Dio quasi e nipote. Da queste due, se tu ti rechi a mente Lo Genesi dal principio, conviene Prender sua vita ed avanzar la gente. Ma perche 1' usuriere altra via tiene, Per se natura, e per la sua seguace Dispregia, poich& in altro pon la spene. Ma seguimi oramai, che il gir mi piace: Che i Pesci guizzan su per 1' orizzonta, E il Carro tutto sopra il Coro giace, E il balzo via la oltra si dismonta. 79 105 1 08 114 103 Che la nostra arte M. vostra tanto quanto P. 104 maestro sol discende (sic) G. fa om. L. fa discente C. 105 in Dio A. quasi a Dio Ej (order corrected later), qua sene pote C. 106 questi BCFIZ?. Da queste cose AELMPQ. Di queste cose Oa*. se tu '1 ti D. tu te '1 rechi K. se te rechi M. se tu rechi Q. se tu cerchi L. 107 Le gienesi D. Lo genesis EIK. del prin. a. si conviene L. 108 sua via F. la sua figlia E. 109 E perche ABCDEFHIKLOQ5. 1' usurieri AGIM. 1' altra K. 110 nat. per ADEHKLP^ 2 C/?a. nat. colla sua I. Ill poi chi in B. ten la spene K. altri tien P. 112 omai D. ch' omai gir K. che gir Z. giro L. 113 i om. ACEGHKLMOPQa (scarcely con- clusive). 114 E om. G. sovra il toro DF^C. sovra tauro L. B* has Choro. 115 basso B. banzo D. balco E. vie EHI. e via C. oltre BGIKQZCa. si dimostra L. 105 A has a Dio in the Commentary. CANTO DECIMOSECONDO Era lo loco, ove a scender la riva Venimmo, alpestro, e per quel ch' ivi er' anco, Tal, ch' ogni vista ne sarebbe schiva. 3 Qual e quella ruina, che nel fianco Di qua da Trento 1' Adice percosse O per tremuoto o per sostegno manco ; 6 Che da cima del monte, onde si mosse, Al piano e si la roccia discoscesa, Ch' alcuna via darebbe a chi su fosse ; 9 Cotal di quel burrato era la scesa: E in su la punta della rotta lacca L' infamia di Creti era distesa, 12 Che fu concetta nella falsa vacca: E quando vide noi, se stesso morse SI come quei, cui 1' ira dentro fiacca. 15 Lo savio mio inver lui grido: Forse 1 ove slender L. 2 Veniva CF. quel che gli era F. e om. M. che v' era AB*GIMZ. 5 Di Ik FIZ. dal Trento Ga. F Atige AM. Athyce A. ladissce Q. la disce K (Poral blunder). F Adige a. 6 terremoti A. sostegni manco AHPQ. 7 di cima Ca. ove si mosse FL. dal monte K. 8 Dal piano K. rocca C, 11 E om. KMPZ. 12 Crete AM. Cretei O. Cretite D. discesa ABC*GLPZ. dessessa K. 14 se om. L. 15 come colui D. cui dentro F ira CGOZ/?. che F ira K. 16 El savio duca ver lui F. mio Virgilio gr. KZ. forte GM. 2 B had apparently i' manu chiveraco. L has chioveranco. 16 Among other reasons against 'Virgilio' is the consideration that Virgil's name is very rarely mentioned in Inf., only five or six times CANTO XII. 8 1 Tu credi che qui sia il duca d' Atene, Che su nel mondo la morte ti porse? 18 Partiti, bestia, che questi non viene Ammaestrato dalla tua sorella, Ma vassi per veder le vostre pene. i\ Qual e quel toro che si slaccia in quella Che ha ricevuto gia '1 colpo mortale, Che gir non sa, ma qua e la saltella, 74 Vid' io lo Minotauro far cotale. E quegli accorto grido : Corri al varco ; Mentre ch' e in furia, e buon che tu ti cale. 27 Cosl prendemmo via giu per lo scarce Di quelle pietre, che spesso moviensi Sotto i miei piedi per lo nuovo carco. 30 Io gia pensando ; e quei disse : Tu pensi Forse a questa rovina, ch' e guardata Da quell' ira bestial ch' io ora spensi. 33 Or vuo' che sappi, che 1' altra fiata Ch' io discesi quaggiu nel basso inferno, 17 que (=quei ?) sia D. quei A 2 . 19 non ene G. 20 dalla sua G. 22 laccia Q*. lancia GILD. slancia e Z* (ap- parently altered from slaccia]. si iscoglie P. 23 ric. lo colpo DMZ?. il om. Z. gia om. AM. 24 e qua AM. ma 'n qua G. martella L. 25 Vidi lo G. 27 Mentre che furia EILa. e ben L. e '1 bon Q. buon e che te cale M. 28 su per ZB. 29 Da quelle F. 30 sotto miei AEHKMOSCD. sotto e mia I. sotto mia L. 31 et ei disse, Che pensi I. e om. K. 32 Forse questa ruina G. Forsa in questo Q. in questa ruina ACEHMOPa. quella ruina F. guardando G. 33 questa rea bestia C. che ora ACE. che ora aspensi D. Da quella bestia la quale ora M. 34 che tu sappi D. ch' all' altra P. 35 Ch' io scesi F. qui giu K. I believe, whereas it occurs more than 20 times in Purg. See Blane's Dizionario s. v. Virgilio. It has most likely been introduced here from a marginal, or still more probably a superposed note, which commonly occurs both as a gloss and as a correction, and is very liable to be mistaken for a correction when intended for a gloss. 20 The order of lines 20 and 21 has been accidentally interchanged in G. 34 B has here and often elsewhere voi for vuoi. P. I. 6" 82 INFERNO Questa roccia non era ancor cascata. 36 Ma certo poco pria, s' io ben discerno, Che venisse Colui, che la gran preda Lev6 a Dite del cerchio superno, 39 Da tutte parti 1' alta valle feda Tremo si, ch' io pensai che 1' universe Sentisse amor, per Io quale e chi creda 42 Piu volte il mondo in Caos converse: Ed in quel punto questa vecchia roccia Qui ed altrove tal fece riverso. 45 Ma ficca gli occhi a valle ; che s' approccia La riviera del sangue, in la qual bolle Qual che per violenza in altrui noccia. 48 O cieca cupidigia, e ria e folle, Che si ci sproni nella vita corta, E nell' eterna poi si mal c' immolle ! 51 Io vidi un' ampia fossa in arco torta, Come quella che tutto il piano abbraccia, Secondo ch' avea detto la mia scorta: 54 E tra il pi& della ripa ed essa, in traccia 36 non iera C. anc. tagliata B. 37 certo om. F. se ben ABCDEGHIMOQZ#Z?a. 40 1' altra L. 41 Tremo si forte AEH. che pensai EL. 42 chi e creda C (partially erased). 43 volto H. Cios E. Caoso G. in un chaos P. e converse Z. 45 ed om. L. fece tal E. 46 Or ficca I. V occhio G. ch' el s' app. M. 47 et in E. in Io qual DA Z CD. il quale L. nel qual M. 48 Quel che EF. Qualunque O. in altri G. 49 ist e om. FP. et ira et folle E. e ira folle B*GMO. o ira folle IKZ. cup. gira (?) folle L. 50 ci om. E. ci si M. 51 E om. L. ci voile C. ci molle MP. 52 in arco volta C. 53 Come colei LM. il om. L. 55 E om. O. tra pi a. ed om. L. 43 In B there is an erasure and blank space between mondo and in. C seems to have had 'Caosso' i ma manu. 47 Perhaps Io qual is a change to conform to the gender of the nearer noun. Possibly the reading gli or li in xiii. 78 may be due to a similar instinct, though there it seems grammatically inadmissible. 49 C has e ria e folle, but it seems to have been altered from e ira e folle. CANTO XII. 83 Correan Ccntauri armati di saette, Come solean ncl mondo andare a caccia. 57 Vedendoci calar ciascun ristette, E della schiera trc si dipartiro Con archi ed asticciuole prima elctte : 60 E 1' un grido da lungi : A qual martiro Venite voi, che scendete la costa ? Ditel costinci, se non, 1' arco tiro. 63 Lo mio Maestro disse : La risposta Farem noi a Chiron costa di presso: Mai fu la voglia tua sempre si tosta. 66 Poi mi tento, e disse : Quegli e Nesso, Che mori per la bella Deianira, E fe' di se la vendetta egli stesso : 69 E quel di mezzo, che al petto si mira, E il gran Chironc, il qual nudri Achille : Quell' altro e Folo, che fu si pien d' ira. 72 D' intorno al fosso vanno a mille i a mille, Saettando quale anima si svelle Del sangue piu, che sua colpa sortillc. 75 Noi ci appressammo a quellc fiere snelle : 56 e armati L. 57 solea M. si suol A(?)D. andando Z. 59 dalla K. 60 cd om. La t . in prima F. 61 Et lui (cler. error?) M. di lungi FIKP. a quel L. 62 dcsccndetc E. 63 Dite Z. 64 Lo buon Maes F. 65 da presso BDFIKPOZ/2C. 66 Ma fu CL. 67 Colui e Xcsso C. 69 egli in stesso E. 70 E quel d' innanzi F. che '1 pecto M. 71 nodri AEHLP. nudrio DO. nodrio B. che nodri F. che nodrio K. nutrio O. che nudri A. 2 . che nutrico /). 72 Folio D. Polo I. il qual si L. che par 1). 74 anime O. sasvcle (sic) E. 76 Noi pervenimmo F. non ne appr. G. Nui sc app. K. 64 F makes the same alteration in 1. 83. 66 The lines 66, 67 and 68 were at first accidentally omitted in O, but arc inserted i"" 1 manu in the margin. 74 Is sa sc in E ?. See again 1. 128. 76 sc in K probably arose from the soft Venetian pronunciation of ci and may be an oral blunder. Comp. xiii. 2 in M. There seems to be a converse change in '95' at Par. xxvi. 136 El ci chiamo e poi el ci convenne. 62 84 INFERNO Chiron prese uno strale, e con la cocca Fece la barba indietro alle mascelle. 78 Quando s' ebbe scoperta la gran bocca, Disse ai compagni: Siete voi accorti, Che quel di retro move ci6 ch' ei tocca ? 81 Cosl non soglion fare i pie de' morti. E il mio buon Duca, che gia gli era al petto Dove le duo nature son consorti, 8 4 Rispose : Ben e vivo, e si soletto Mostrarli mi convien la valle buia : Necessita '1 c' induce, e non diletto. 87 Tal si parti da cantare alleluia, Che mi commise quest' uficio nuovo ; Non e ladron, ne io anima fuia. 90 Ma per quella virtu, per cui io movo Li passi miei per si selvaggia strada, Danne un de' tuoi, a cui noi siamo a pruovo, 93 Che ne dimostri la dove si guada, 78 Si fe F. dietro A. drieto M. 79 Da poi ch' ebbe F. 1' ebbe M. scoperto ED. 80 sietevi voi DFC. non siete A. 81 di dietro (detro, &c.) ABCFIMPZa. di rietro EKL. cio che t. BDFGA^C. 82 a pie L. 83 Lo buon Maes F. II mio buon G. eran M. 84 dua I. doie Q. costretti L. 85 vero Z* (ap- parently altered from vivo). 86 Mostrar migli C (melli) FM. 87 Nee. il conduce C*DIKMPQ*CZ>. 1' induce AEF. Nee. c' induce BLOZ. Nee. conduce G. 88 dal cantar K. 89 Che commise a me A. Che ne B. 91 per om. L. per la qual movo F. mi muovo L. per cui movo M. 92 Li piedi G. per am. L. 93 a chi A. per cui P. nesciamo a prova C. siamo puo (with a inserted later) D. a puo or a pruo (?) B* (B 1 seems to have had siamo puo}. siamo apso (sic) E. cui nesian K. a cui sia a L. 94 Che ne mostri CF. E che ne mostri ABEGHILMOPZa. 83 See 1. 64 in F. 87 C had apparently 'c' induce' \ ma manit, and so perhaps Q. 93 Q has Damun = Dammi un? (Cf. xiii. 89.) In 1. 120 however Q reads emsu Tamts, so that m may stand for n h. 1., as it does frequently in Q, e.g. in next Canto, 1. 78, emvidia for invidia and xvi. 75 tempiagni. 94 Guada has had di inserted above in Q, appa- rently in reminiscence of vi. 114. CANTO XII. 85 E che porti costui in su la groppa; Che non e spirto che per 1' aer vada. 96 Chiron si volse in sulla destra poppa, E disse a Nesso : Torna, e si li guida, E fa cansar, s' altra schiera v' intoppa. 99 Noi ci novemmo colla scorta fida Lungo la proda del bollor vermiglio, Ove i bolliti facean alte strida. 102 lo vidi gente sotto infino al ciglio; E il gran Centauro disse: Ei son tiranni, Che dier nel sangue e nell' aver di piglio. 105 Ouivi si piangon li spietati danni : Quivi e Alessandro, e Dionisio fero, Che fe' Sicilia aver dolorosi anni : 108 E quella fronte ch' ha il pel cosl nero E Azzolino ; e quell' altro ch' e biondo E Opizzo da Esti, il qual per vero in Fu spento dal figliastro su nel mondo. Allor mi volsi al Poeta, e quei disse: 95 in om. E. costui sopra a*. 96 per acre vada E. 98 guidi Z. 99 cessar CF. scansar M. fiera rintoppa L. schiera int. F. s' int. EGHIOPCZA 100 Or ci mov. ABCEHKLMOPQa. 101 color verm. G. 102 i om. C*D. i bollenti L. fecer C. altre B. alte le strida DK. grande strida Q. 103 Quivi vidi AE (vid' io) HL. E vidi C. Qui vidi O. Qui vid' io P. genti C. sotto fino ADEKMOQ. insino FI. 104 E om. G. disse, Son tir. C* (i son \ ma manu}. 105 Che dieder E. 106 dispiatati Q. 107 Qui e AG. e om. PQ. Dioniso CDFHQa. Quivi Aless. Dionisi fiero L. 108 Cicilia ABCDEFLMZ^a. Cecilia HKO. Scicilia Q. Cicillia C. fer FI. 109 E om. CQ. ch' a il capel si FL. il pel si nero G. fronte col capel si M. 110 Encerino F. Asolino I. Ecelino M. Ezzerino A. altro si biondo LM. Ill Obizo ACDOQZCa. Opiso B. Obiso I. in '1 qual (?) G. 112 dal figl. spento FI. suo B. dal fiolo su K, 113 E volsimi Q. io mi volsi FC. mi disse DFKQ. 108 fer (in plural) is of course historically false, whether 'Alessandro' be Alexander the Great, or Alexander of Pheras, as to which the Comm. are divided. It is curious to find Buti describing the latter as "Alessandro Fereo di Sicilia? 86 INFERNO Quest! ti sia or primo, ed io secondo. 114 Poco piu oltre il Centauro s' affisse Sopra una gente che infino alia gola Parea che di quel bulicame uscisse. 117 Mostrocci un' ombra dall' un canto sola, Dicendo: Colui fesse in grembo a Dio Lo cor che in sul Tamigi ancor si cola. 120 Poi vidi gente, che di fuor del rio Tenea la testa ed ancor tutto il casso: E di costoro assai riconobb' io. 123 Cosl a piu a piu si facea basso Quel sangue si, che cocea pur li piedi : E quivi fu del fosso il nostro passo. 126 SI come tu da questa parte vedi Lo bulicame che sempre si scema, Disse il Centauro, voglio che tu credi, 129 Che da quest' altra piu a piu giu prema Lo fondo suo, infin ch' ei si raggiunge Ove la tirannia convien che gema. 132 114 Questosifia primoet ioorsec. E. sifiaLP. tefiaM. 115 Poi che piu I. 116 SurunaD. una.om. L. chefinoMOQ. 117 che fuor del bul. F. che dello bul. Atf). 118 Mostrami L. da un canto FI. 119 Colei B*CGHLOPZ5. fosse a. il grembo P. 120 in su Tamise (-si or -sa) ADM. in su Tamissy K. in su Tamisci CEGHPa. in su Tamigi BIZA 2 BCD. emsu Tamis Q. in sul Tamigio O. in sul Tamisi F. chiuso (? che in su) Tamisci L. 121 dario CKLC. dalrioMOZZ?. derrio a. 122 Tenean ACFMOZ. ed om. KQ t . 124 Cos! a piu si facea E. facean Ma. 125 Lo sangue che cociva Q. che '1 cocea A. copria K (? copra) WLBD. che om. P. 126 E qui fu F. E quindi EGHLPZ^. 128 a piu a piu si scema F. sa sema E. 129 il om. L. 130 a piu a piu ABDE 2 FGHKOPQZa. piu e piu I. \st piu om. LA 2 CD. 131 fin che si F. che si KMQa ch' el se guigne E. 119 With a's reading comp. the well-known interchange of fosse and fesse in xx. 69. 128 F is evidently wrong, see lines 124 and 130. The phrase could not certainly have occurred three times in seven lines. 130 Comparing 1. 124 it is probable that the first a has been lost in the terminal a of altra. CANTO XII. 87 La divina giustizia di qua punge Quell' Attila che fu flagello in terra, E Pirro, e Sesto ; ed in eterno munge 135 Le lagrime, che col bollor disserra A Rinier da Corneto, a Rinier Pazzo, Che fecero alle strade tanta guerra : 138 Poi si rivolse, e ripassossi il guazzo. 134 Quella tila L. fragello FLQa. 135 Sesto in et. AEFGHLMOa. E Pirro esto (sic) L. or in eterno P. 136 Le lagrimine D. Le om. P. che con quel P. 137 A Riniari da C. I. A Corn. Rin. K. A Ranierda Corn. Q. A Rainier di C. MP. Rainieri da C. E. E Rin. P. CC. A Ranieri P. E. E a Rin. P. D. 139 A poi (?) G. e ripasso '1 A. riposesi K. ripassoe M. 139 G apparently means E poi. CANTO DECIMOTERZO Non era ancor di la Nesso arrivato, Quando noi ci mettemmo per un bosco, Che da nessun sentiero era segnato. 3 Non frondi verdi, ma di color fosco, Non rami schietti, ma nodosi e involti, Non pomi v' eran, ma stecchi con tosco. Non han si aspri sterpi ne si folti Quelle fiere selvagge, che in odio hanno Tra Cecina e Corneto i luoghi colti. 9 Quivi le brutte Arpie lor nidi fanno, Che cacciar delle Strofade i Troiani Con tristo annunzio di future danno. \i Ale hanno late, e colli e visi umani, Pie con artigli, e pennuto il gran ventre : Fanno lamenti in su gli alberi strani. 15 2 se met. M. 3 niun(o) sent. ADOa. 4 fronda verde ACDEGHKLMOPZ^o^C fronde verde B. 5 e void F. 6 v' era F. 7 Non a M. asp. stecchi C. 8 che nido hanno F. che odio K. 9 in luoghi BC. in vecchi colti F. 10 Qui le D. il lor nido D. nido BFGKLMQ. nidio Z. lor nidio le br. Arp. B. 11 dalle D. Scrofade AHLP. Trogliani A. 12 augurio F. 13 Ali CDHKLOPQZ^.Da. Alie F. ist e om. CFIMA^C. e om. bis L. 2nd e om. a^ 15 lamento A. lamenti su alberi E. in om. CM. albori CDFKL. arbori BMPQ^6'^. arberi O. 8 F has clearly an inferior reading, it being at first sight simpler, and also a jejune repetition of the phrase in v. 10. Compare a very similar instance in ' I ' at 1. 118. CANTO XIII. 89 Lo buon Maestro : Prima che piu entre, Sappi che se' nel secondo girone, Mi comincio a dire, e sarai, mentre 18 Che tu verrai nell' orribil sabbione. Pero riguarda bene, e si vedrai Cose, che torrien fede al mio sermone. 71 lo sentia da ogni parte traer guai, E non vedea persona che il facesse ; Perch' io tutto smarrito m' arrestai. 24 lo credo ch' ei credette ch' io credesse, Che tante voci uscisser tra que' bronchi Da gente che per noi si nascondesse. 27 Pero, disse il Maestro, se tu tronchi Qualche fraschetta d' una d' este piante, Li pensier ch' hai si faran tutti monchi. 3 o Allor porsi la mano un poco avante, 16 E '1 buon BCDEFGHIKLMOPQZ^Z?a. che tu entre DFGILOZ?. B* has^z'0 entre, pio being crowded in with smaller letters over an erasure doubtless representing tu in B^ 18 Incomincio KCD. 19 terribil C. 20 E pero guarda ~DKA 2 CD. ben se vederai A. ben si vederai (or bene si vedrai) BEHLOPZZ?. bene si (with e inserted later) a. e tu vedrai I. 21 torria F. torrian CMS. toran Q. che pien fede L. mi B* (with o apparently erased). al mio signore L. 22 trarre da ogni parte \A Z C. d' ogni L. tragger guai BM. trarre j5"a. trarrer D. trar gran guai Q. 23 vidi C. persone K. 24 Perche tutto DFK. mi ristai C. mai rest. C. 25 Cred' io BLOZ^Ca. Credo CK. E credo D. che credette ADEFGHIKLOa* ch' el credette CMPC 26 di quei CMPZ>. 27 Di gente B^D. da noi C. 30 Li pensier tuoi KP. ch'ai tutti fier m. F. farno L. 31 pors' io BCDFKQ^. 2 CZ>a. la mane A. 16 This is not conclusive as El may stand for either //, or Lo, or E '/. 21 A 2 has torien, but A x appears to have had terien. 25 In B cred' io has been clumsily altered to E cred' io. Che may of course represent cK e 1 . Lines 26 and 27 have been transposed in E, and the order corrected in margin. 30 With the form/aruo comp. convertirno in xxvii. 15, and furno xxviii. 130, and aiutorno xxxii. n (all in I). Also riforndarno \. 148. 9O INFERNO E colsi un ramicel da un gran pruno : E il tronco suo grid6 : Perche mi schiante ? 33 Da che fatto fu poi di sangue bruno, Ricomincio a gridar : Perche mi scerpi ? Non hai tu spirto di pietate alcuno ? 36 Uomini fummo, ed or sem fatti sterpi : Ben dovrebb' esser la tua man piu pia, Se state fossim' anime di serpi. 39 Come d' un stizzo verde, che arso sia Ball' un de' capi, che dall' altro geme, E cigola per vento che va via ; 42 Si della scheggia rotta usciva insieme Parole e sangue : ond' io lasciai la cima Cadere, e stetti come 1' uom che teme. 45 S' egli avesse potuto creder prima, Rispose il Savio mio, anima lesa, Ci6 ch' ha veduto pur con la mia rima, 48 Non averebbe in te la man distesa ; Ma la cosa incredibile mi fece Indurlo ad opra, che a me stesso pesa. 51 Ma dilli chi tu fosti, si che, in vece D' alcuna ammenda, tua fama rinfreschi Nel mondo su, dove tornar gli lece. 54 E il tronco : SI con dolce dir m' adeschi, 32 E chosi (? cler. error) L. ramigel B. ramuscello GDI. ra- misselo K. ramoscel Q. ramucelA. ramuscel D, de gran pruno K. 33 suo troncon CFAD. 34 Da poi che fatto fu di F. E poi che fatto fu GQ. fui fatto C. fu fatto B. poi om. L. 35 Comincio M. Ric. a dir G. Ric. gridar P. Comincio a dir Q. sterpi CFGI 1 (?)MOZ. strepi E. 36 pieta BCQ. piata FL. piatate G. 37 scherpi A. fummo or sian Q. 38 doveria I. 39 Se sute G. stati AEHPQZ>. 40 stizon Q. 41 de' lati B. e che I. 43 Cosi della F. Cosi di quella scheggia use. KZQ5. schenc.a rotta A. usciro F. 45 Chavere (sic) L. 47 mio om. K. 48 Cio che ved. E. hai ved. DKL. poi colla G. 49 E' non arebbe F. la mano stesa CDK. 51 Indurlo a cosa BCI. opere K. 52 Ma dimmi G. che tu C. 53 D' alcuna menda ABCEjFILOP. Alcuna amm. K. mi freschi M. 54 dove su Q. 55 E '1 troncon DI. col dolce CGPA. 2 C. m' aeschi BCDFG. CANTO XIII. 91 Ch' io non posso tacere ; e voi non gravi Perch' io un poco a ragionar m' inveschi. 57 Io son colui, che tenni ambo le chiavi Del cor di Federico, e che le volsi Serrando e disserrando si soavi, 60 Che dal secreto suo quasi ogni uom tolsi: Fede portai al glorioso offizio, Tanto ch' io ne perdei le vene e i polsi. 63 La meretrice, che mai dall' ospizio Di Cesare non torse gli occhi putti, Morte comune, e delle corti vizio, 66 Infiamm6 contra me gli animi tutti, E gl' infiammati infiammar si Augusto, Che i lieti onor tornaro in tristi lutti. 69 L' animo mio per disdegnoso gusto, Credendo col morir fuggir disdegno, Ingiusto fece me contra me giusto. 72 56 none grave L. 58 ambur B. 59 e si le volsi I. la volsi O. 61 del seer. DF. ognun tolsi EFIKLQ. ognuno tolsi A. 62 hospitio R. 63 Tanta EH. io om. AP. perdeva B. i seni (=senni) C. i sensi BF. li sonni AHO. sonni e polsi LMPa. Io sonno e polsi K. li sonni di polsi E. [e prob. = / in KLMPa as often.] 66 M. e commune DKQaj (?). e comm. delle D. com. delle GEEKLC^. e vizio GQ. 67 contro a FGI. contra a LM. 69 Che lieti ABCDIKMPQZ.&Ca. Che glieti G. 70 disd. giusto G. 72 feci DEIM.^. contro Q. contro a FGI. contra a LM 2 . 58 The form ambur which is found in B here, and appears again as ambor in xiv. 82 and xix. 120, occurs three or four times in the Chanson de Roland (e.g. 1. 1546, Ambur ocit, ki que 1'blasmt ne le lot). Le"on Gautier's note on it is as follows: "De'rive" sans doute...du gdn. pi. amborum sub- stantivise", comme francur, paienur, &c." (i.e. Francorum, Paganorum. Also Saraconor occurs). He adds "Dans la Chronique des Dues de Normandie nous avons de nombreux exemples d'amfor, ambore, ambur, avec le sens 'd'ensemble.'" 63 le vene e i p. seems an inappropriate reminiscence of i. 90. If sensi be read, comp. perhaps its use in xxvi. 115 for conscious and intelligent life. Scartazzini defends il sonno e i polsi, i.e. 'rest by night and strength by day.' 69 As often, Che probably stands for Che i in these MSS. 71 and 72 In A these lines are trans- posed but corrected by the letters b, a, in margin. Cf. sup. v. 97 100. The terzina 73 75 is repeated in error in H. 92 INFERNO Per le nuove radici d' esto legno Vi giuro che giammai non ruppi fede Al mio signer, che fu d' onor si degno. 75 E se di voi alcun nel mondo riede, Conforti la memoria mia, che giace Ancor del colpo che invidia le diede. 78 Un poco attese, e poi : Da ch' ei si tace, Disse il Poeta a me, non perder 1' ora ; Ma parla, e chiedi a lui se piu ti piace. 81 Ond' io a lui : Domandal tu ancora Di quel che credi che a me satisfaccia; Ch' io non potrei : tanta pieta m' accora. 84 Percio ricomincio : Se 1' uom ti faccia Liberamente ci6 che il tuo dir prega, Spirito incarcerato, ancor ti piaccia 87 Di dime come 1' anima si lega In questi nocchi ; e dinne, se tu puoi, S' alcuna mai da tai membra si spiega. 90 Allor soffio Io tronco forte, e poi Si convert! quel vento in cotal voce : Brevemente sara risposto a voi. 93 Quando si parte 1' anima feroce 73 di questo E. 74 Ti giuro D. 76 E se alcun di voi nel mondo alcun di voi A. al mondo K. 78 dal colpo M. gli (or li) diede ABCDEGIKMOPZ. mi diede F. 79 Non poco K. attese da che '1 D. attesi GI. et dopo ch' el si tace E. e poi che '1 si tace Q. si cade (sic) K. e poi di quel si tace L. 80 Poeta mio M. il Maestro a me a. 81 ci piace (?) a. 82 Ed io K. domanda tu ABCEHKMOPQ^a. domandai a tu L. 83 che mi M. 85 Per6 AFGMOQ. ricominciai H. se uno D. 88 dirmi KMQ. 89 In questi luochi Q*. dime ( = dimmi) EK. 90 di tai DBD. da tal BC* (altered from tai} EMZ. di tai IK. membre DM. mem- bri O. slega FG. 91 tronco e disse poi Z. 92 Si convertio D. Si comincio K. cotai voce C. 93 Brieve mentre (cler, error?) C. fiaFM. 78 See note on xii. 47. 89 See note on xii. 93 for the reading of EK. Qj apparently had the ordinary reading nocchi. CANTO XIII. 93 Dal corpo, ond' ella stessa s' 6 divelta, Minos la manda alia settima foce. 96 Cade in la selva, e non 1' & parte scelta; Ma la dove fortuna la balestra, Quivi germoglia come gran di spelta ; 99 Surge in vermena, ed in pianta silvestra : L' Arpie, pascendo poi delle sue foglie, Fanno dolore, ed al dolor finestra. 102 Come 1' altre, verrem per nostre spoglie, Ma non per6 ch' alcuna sen rivesta : Ch& non e giusto aver cio ch' uom si toglie. 105 Qui le strascineremo, e per la mesta Selva saranno i nostri corpi appesi, Ciascuno al prun dell' ombra sua molesta. 108 Noi eravamo ancora al tronco attesi, Credendo ch' altro ne volesse dire, Quando noi fummo d' un romor sorpresi, m Similemente a colui, che venire Sente il porco e la caccia alia sua posta, 95 Del corpo CFK. disvelta ABCDEFGHlKLOPQZtf. si dis- velta ACGOa. 96 Minoso L. manda la s. L. 97 non gli e DEGIKMOQ. e ne li parte L. 98 Ma dove A. 100 in gier- miria A. e pianta F. con pianta L. 101 poi pascendo le sue G. 103 verien AEHLP. verren BCKZ. veran D. verreno I. verrieno a. 104 percio B. se rivesta AGK. se rinvesta L. 105 cio horn D. che 1' uom L. 106 Poi le strasc. I. Qui la CL. 108 G reads (in obvious error) Ciascuno all' ombra al prun dell' ombra &c. 109 al tronco ancora ABIT). alteso G. 110 altri P. non volesse a. ill noi om. I 2 . soppresi FI. 113 Sentiva P. alia caccia A 2 BCDF KLMQ. a sua p. AFGHLOP. a suo p. BC. 96 I seems to have Minos with -si erased (see note on v. 4). 100 giermina is paraphrased ' in erba tenerella ' in Benvenuto's Commentary in this MS. 103 The variations given are all probably merely ortho- graphical, final m and n, esp. in some MSS, being so frequently inter- changed that such variations are not generally worth notice. ' I ' also exhibits its usual tendency to full-length forms in ' verreno', as it has furono in 1. 120, portaro, 1. 129, &c, 94 INFERNO Ch' ode le bestie e le frasche stormire. 114 Ed ecco duo dalla sinistra costa, Nudi e graffiati, fuggendo si forte, Che della selva rompieno ogni rosta. 117 Quel dinanzi : Ora accorri, accord, morte. E 1' altro, a cui pareva tardar troppo, Gridava : Lano, si non furo accorte 120 Le gambe tue alle giostre del Toppo. E poiche forse gli fallfa la lena, Di se e d' un cespuglio fece un groppo. 113 Diretro a loro era la selva plena Di nere cagne, bramose e correnti, Come veltri che uscisser di catena. n6 In quel, che s' appiatto, miser li denti, E quel dilaceraro a brano a brano; Poi sen portar quelle membra dolenti. 129 Presemi allor la mia scorta per mano, E menommi al cespuglio che piangea, Per le rotture sanguinenti, invano. 132 O Jacomo, dicea, da sant' Andrea, Che t' e giovato di me fare schermo ? Che colpa ho io della tua vita rea ? 135 114 C* has (evidently 2 dlt manu} Colde le bestie. 115 alia sin. DKQ5. due venire della L. 116 correndo si Q. 117 ogni costa K. 118 E quel FM. Ora om. FM. Quel dinanzi gridava, Accorri morte I. corra, corra, K. 119 L' altro a cui FL. E P altro cui ABMOQ^C. tardi L. 120 Grid, a ladro D. Grid. O Alano F. Lana L. 121 dal Toppo BDFGHOP^CZ'a. 122 fallio G. falli M. le fallia P. 123 Di se ad un DG. Di se d' un L. de cesp. K. fece groppo G. 124 Di dietro BZ. O dirietro G. v' era selva P. 125 cagne nere M. e bram. CD. 127 s' appigliato A (corrected in margin 2 da manu). 128 E lui D. dilacerato AH P. a lacerate E. 129 membri M. 130 lo mio duca B. 131 al bis a. 132 Per la rottura sanguinente E. 133 Jacopo AFIM. dicea om. L. 134 Che t' a M. 135 colpa io (? = i' ho). colpai io P. 118 1 has an inferior reading, since ' gridava' 1 recurs two lines below. See note on 1. 8. CANTO XIII. 95 Quando il Maestro fu sopr' esso fermo, Disse : Chi fusti, che per tante punte Soffi con sangue doloroso sermo ? 138 E quegli a noi : O anime che giunte Siete a veder lo strazio disonesto, Ch' ha le mie fronde si da me disgiunte, 141 Raccoglietele al pie del tristo cesto : lo fui della citta che nel Batista Muto '1 primo patrono : ond' ei per questo 144 Sempre con T arte sua la fara trista : E se non fosse che in sul passo d' Arno Rimane ancor di lui alcuna vista ; 147 Quei cittadin, che poi la rifondarno Sopra il cener che d' Attila rimase, Avrebber fatto lavorare indarno. 150 lo fei giubbetto a me delle mie case. 137 chi fuse tu E. 138 Sofio L. col sangue DKZ?. con pianto F. 139 Ed egli (or elli) ABCEFGHILMOPQZ^a. a me A. 141 Cham ( = che ban) D. Che le L. delle mia fronde I. Che le mie frondi sien da me F. Con le mie fronde K. mie foglie Q. desunte G. digiunte Ma. 142 Rico- glietela K. Raccoglietimi P. a pi AFILCZ?. 141 patrone AEMQ. padrone BCDFGHIKLOPZ,4 2 #C>a. onde per CM. ond' io L. 145 la facta ( = !' ha fatta) Z. faccia K. 146 E se '1 non AE. che '1 sul D. che sul KMQ. 147 Rim. di lui ancora F. Di lui rim. anc. M. 148 Li cittadin Q. refundaro E. la fondarno L. 149 la cener A. d' Actile P. 150 Avrebbe M. 151 ghibecto E. jubet F. jubbetto I. gibbeto Ma. gibet Q. CANTO DECIMOQUARTO Poiche la carita del natio loco Mi strinse, raunai le fronde sparte, E rende' le a colui ch' era gia fioco. 3 Indi venimmo al fine, ove si parte Lo secondo giron dal terzo, e dove Si vede di giustizia orribil arte. 6 A ben manifestar le cose nuove, Dico che arrivammo ad una landa, Che dal suo letto ogni pianta rimove. 9 La dolorosa selva 1' e ghirlanda Intorno, come il fosso tristo ad essa: Quivi fermammo i passi a randa a randa. 12 Lo spazzo era un' arena arida e spessa, Non d' altra foggia fatta che colei, 2 Ristrinse L. li frondi sparte Q. sparse K. 3 E rende' le al cespuglio F. rendeteli E. a colei M. che parea fioco G. 4 veniano fin dove L. al flume F. onde si parte CGMZ. 5 giro L. 7 Al bene IL. 8 lo dico FGD. banda L. 9 ogni piata (prob. cler. error for pianta) G. 10 gli e DFQ. li e B*. lei M. la ghirl. ACEa. Ian ghirl. (sic) G i.e. la 'nghirlanda (so D). gli ghirlanda K. 11 D' intorno IM. tristo fosso EFZ>. tristo fosso tristo (sic] G. cornel fose (sic) L. 12 i piei B. 13 L' ospizio L. 14 foglia fatta A. di colei F. cha colei E. 3 The scribe of G seems to have had too good a memory, and here (as often noted elsewhere) has reproduced the words of another familiar pas- sage. See i. 63. 4 I had apparently fatme, altered to fine 2" da manu. The relationship between F and I is very close (see Prolegomena). 14 foglia in A is an obvious clerical error and is corrected (though 2> l(f(l jnatni] in margin. The obvious Venetian origin of the MS. may have contributed to an error of this kind, since foglia would be pronounced like folgta, CANTO XIV. 97 Che fu da' pie di Caton gia soppressa. 15 O vendetta di Dio, quanto tu dei Esser temuta da ciascun che legge Ci6 che fu manifesto agli occhi miei ! 18 D' anime nude vidi molte gregge, Che piangean tutte assai miseramente, E parea posta lor diversa legge. i\ Supin giaceva in terra alcuna gente ; Alcuna si sedea tutta raccolta, Ed altra andava continuamente. 24 Quella che giva intorno era piu molta, E quella men, che giaceva al tormento, Ma piu al duolo avea la lingua sciolta. 27 Sopra tutto il sabbion d' un cader lento Piovean di foco dilatate falde, Come di neve in alpe senza vento. 30 Quali Alessandro in quelle parti calde D' India vide sopra lo suo stuolo Fiamme cadere infino a terra salde ; 33 Perch' ei provvide a scalpitar lo suolo Con le sue schiere, acciocche il vapore Me' si stingeva mentre ch' era solo : 36 Tale scendeva 1' eternale ardore ; 15 da piedi DGI. a pie L. gia om. G. Che da pie di Caton fu F. sospesa L. 17 tenuta E. 19 vid' io L. 21 imposta a lor A. lor imposta D. imposta lor H. par posto I. posta a lor EKZ. posto lor L. posto a lor MA 2 CD. 22 giacean E. 23 E alcuna a. 24 E P altra A 2 CD. altri E. andavan M. 26 E quella che giaceva men ha '1 torm. A (with marks of erasure at ha '/). 27 Ma al duolo avea piu F. piu avean al duol A. a duolo a : . 28 Sopral tutto '1 Z. 29 Piovea A. Pioveva fuoco L. 31 Quando Al. I. 32 vede HP. sopra del suo MI). 33 fino a K. 34 Perche ADEGHLMOPQa. scapolar K. 35 perciocche I. lu vapore Q. 36 Meio si AK. Me' si spegnesse Z. Meno strugiea L. Men lo stringea M. mentre ell' era A v 37 Tal discendeva M. which bears some resemblance to foggia. In E cha for che is not uncommon. Cf. K in 1. in. 34 C* has stuolo, an obvious error from 1. 32. D. I. 7 98 INFERNO Onde 1' arena s" accendea, com' esca Sotto focile, a doppiar lo dolore. 39 Senza riposo mai era la tresca Delle misere mani, or quindi or quinci Iscotendo da se 1' arsura fresca. 42 lo cominciai : Maestro, tu che vinci Tutte le cose, fuor che i Demon duri, Che all' entrar della porta incontra uscinci, 45 Chi e quel grande, che non par che curi L' incendio, e giace dispettoso e torto SI che la pioggia non par che il maturi ? 48 E quel medesmo, che si fue accorto Ch' io domandava il mio duca di lui, Grido : Qual io fui vivo, tal son morto. 51 Se Giove stanchi il suo fabbro, da cui Crucciato prese la folgore acuta, Onde 1' ultimo di percosso fui ; 54 O s' egli stanchi gli altri a muta a muta In Mongibello alia fucina negra, 38 Onde la terra DQ. 39 Sotto '1 fucile ADYGA Z CD. a doppiar dolore D. a crescer lor dolore F. a radoppiar dol. D. 41 or quive or quindi K. 42 Existendo A. Escotendo BDEKPv4 2 . Discotendo CQZP. Schotendo L. Et scotendo OJ3C. Da se scotendo la lordura M. 43 Incominciai C. 44 in fuor che I. furonche L. che dimon duri ACDEFHKLMPQZa. 45 incon- tro ABCDEFGHIKLMOPZ^ 2 Ca. uscirci BCDF. 47 L' in- cendi C. e om. L. dispietoxo K. 48 Si che '1 par che la pioggia nol mat. A. non par che la pioggia FM. marturi M.PD. 49 si fa D. 50 al mio EM. 51 tal qual fu' vivo F. cotal io son m. C*. 52 i suoi fabbri Z. 54 Ond' io A. 55 E s' egli CGI Ma. 56 nera DF. 44 furon in L is a curious blunder of the copyist, who habitually writes furon for fur or furo, and here mistook fuor for the verb. So in xvi. 22 he has apparently mistaken far tor fur, and so has again written furono, violating both sense and metre. Compare xv. 18 (sole for suol). His blunders are so frequent and so unintelligent that one can scarcely sup- pose him to have understood Italian at all. 56 In F, lines 58 and 59 follow line 56. They are underlined but not erased, and afterwards re- peated in their proper place. See a similar blunder in this MS. at v. 100. CANTO XIV. 99 Chiamando : Buon Vulcano, aiuta aiuta, 57 SI com' ei fece alia pugna di Flegra, E me saetti di tutta sua forza, Non ne potrebbe aver vendetta allegra. 60 Allora il Duca mio parlo di forza Tanto, ch' io non 1' avea si forte udito : O Capaneo, in ci6 che non s' ammorza 63 La tua superbia, se' tu piu punito: Nullo martirio, fuor che la tua rabbia, Sarebbe al tuo furor dolor compito. 66 Poi si rivolse a me con miglior labbia, Dicendo : Quel fu 1' un de' sette regi Ch' assiser Tebe; ed ebbe, e par ch 1 egli abbia 69 Dio in disdegno, e poco par che il pregi : Ma, come io dissi a lui, li suoi dispetti Sono al suo petto assai debiti fregi. 72 Or mi vien dietro, e guarda che non metti Ancor li piedi nell' arena arsiccia : Ma sempre al bosco li ritieni stretti. 75 Tacendo divenimmo la ove spiccia Fuor della selva un picciol fiumicello, Lo cui rossore ancor mi raccapriccia. 78 57 Gridando M. el buon vul can (sic) K. 58 come fece CFGM. da Plegra I. 59 saetto E. con tutta ABCDEFGHIKLMOPZZ>a. 60 Non potrebbe L. 61 Lo duca mio allor F. grido Q. 62 Tanto che non DGIK. 63 Ha Cap. D. Canpaneo AFKLQ. Campaneo B^CDEGH. 64 ben punito DOA (marg.\ 65 martiro D. martire F. Null' altra pena I. in fuor I. 66 al tuo dolor tutto comp. C. compiuto EGKLQ. 67 mior A. 68 E disse F. qual fu L. quegli e 1' un F. quel fu uno I. quei fo un M. 70 Dio isdegno 13. Dio a disd. M. in dispetto F. in dispregio PJ3. a disd. C. che pregi A. 71 disse lui GZB. 73 e om. E. viemmi dietro F. 74 entro 1' arena F. 75 al bosco tien (or tieni) i (or li) piedi ABEHILMOPQ^a. al b. fa gli tenghi Z. li mantieni C. si li tiene G. 76 divenimmo dove GMZ. 77 Fuor dell' arena AEGHLMOP. 64 This change is probably from a reminiscence of xix. 97. 73 Keno meti (sic) C. F is probably thinking of xix. 93. 72 100 INFERNO Quale del Bulicame esce un ruscello, Che parton poi tra lor le peccatrici, Tal per 1' arena giu sen giva quello. 81 Lo fondo suo ed ambo le pendici Fatt' eran pietra, e i margini da lato: Perch' io m' accorsi che il passo era lici. 84 Tra tutto 1' altro ch' io t' ho dimostrato, Posciache noi entrammo per la porta, Lo cui sogliare a nessuno e negato, 87 Cosa non fu dagli tuoi occhi scorta Notabil, come Io presente rio, Che sopra se tutte fiammelle ammorta : 90 Queste parole fur del Duca mio : Perche il pregai, che mi largisse il pasto Di cui largito m' aveva il disio. 93 In mezzo mar siede un paese guasto, Diss' egli allora, che s' appella Creta, Sotto il cui rege fu gia il mondo casto. 96 Una montagna v' e, che gia fu lieta 79EqualZ>. esce ruse. ABCDEFGHKLOP,4 2 #C7}. esce il ruse. IQZ. esce erusciello (sic] a. 81 Tal giu per 1' arena F. se giva L. 82 suo et anco K. ambor B. 83 Forte eran B (corrected -2" da manu \.ofatte). Fatto pietra F. pietre AD. e gli argini F. pietre i margini Z. dal lato O. a* has pietra I marg. 84 linci A. 85 Fra tutto GI. 86 Da poi che noi F. per questa porta E. 87 sogliaro E. negun EK. 88 occhi tuoi F. fui L. 89 quant' el F. com' il ZA Z BCD. 90 sopra a se FLa. tutta L. 92 Perch' io pr. CA 2 SCa>. Perch' io il pr. ADEFHKLMOZ. che '1 mi AEKM. il passo CL. 94 mezzo il mar ADEIKQ. 95 Dissi alor L. allotta M. che si chiama Creta I. 96 Sotto cui G. mondo guasto G. 97 n' e M. fu g&G. 81 L however constantly omits final letters and especially n [see Pro- leg^. 82 On the form ambor see note on xiii. 58. 83 Several MSS. have e margini, but this is not conclusive. 89 Several MSS. have cornel, which may stand for come V or con? 2 Y. 97 ve kegia fit lieta is a characteristic specimen of C's orthography. So in 1. 90, C reads fiame lamorta. CANTO XIV. IOI D' acqua e di fronde, che si chiamo Ida; Ora e diserta come cosa vieta. 99 Rea la scelse gia per cuna fida D' un suo figliuolo, e, per celarlo meglio, Quando piangea, vi facea far le grida. 102 Dentro dal monte sta dritto un gran ,veglio, Che tien volte le spalle inver Damiata, E Roma guata si come suo speglio. 105 La sua testa e di fin' oro formata, E puro argento son le braccia e il petto. Poi e di rame infino alia forcata : 108 Da indi in giuso e tutto ferro eletto, Salvo che il destro piede e terra cotta, E sta in su quel, piu che in sull' altro, eretto. m Ciascuna parte, fuor che 1' oro, e rotta D' una fessura che lagrime goccia, Le quali accolte foran quella grotta. 114 Lor corso in questa valle si diroccia : Fanno Acheronte, Stige e Flegetonta ; 98 D' acque DICa*. D' oro K. fronda AYZ. si om. M. chiama ABCEFLa. et e chiamata K. chiamava F. 100 Rei P. gia una fida (sic) L. culla fida I. perche una fida (sic) M. lOlDesuoA. DelsuoBCEFGHIKLOPQZ^a. fiolo K. filglolu Q. cercharlo E x . 102 Ouand' e' piang. C. le strida GMCD. faciea far grida K. 103 Denlro quel F. Dritto dal monte K. sta drieto E. diritto un veglio F. sta ritto C. dritto om. L. 104 ritte le spalle C. spalle ver Dam. AEP. le spalle volte ver M. 105 guarda si come DEIMPQ/?/?. guata come G. guarda come ABCFHLOa. E guarda Roma come Z. come nel suo D. 106 La testa sua BIMQZCZ?. di fin oro e M. 107 le braci A. et petto K. 108 rame fin' alia AOP. rame fino alia CGHKQZZ>a. ramo fino alia E. insino FI. infine B. alia 'nforcata DIPQCD. 109 in giu CFGKM. 110 pie A. pie ch' e terra G. pie e de terra K. Ill E sta su 'n quel BDKQ. E sta su quello piu che in AEH. E sta su quello piu che su FQa. E sta su quel C* (in su i >na mami) MO PA E sta in quel G. piu cha su 1' altro K. retto GIM. 112 in fuor I (as usual). 114 Accolte quivi foran F. questa BFG I MZZ?. 115 Lo corso AE. in quella AP. valle oni. L. 116 Fregetonta F. 102 INFERNO Poi sen va giu per questa stretta doccia 117 Infin la dove piu non si dismonta : Fanno Cocito ; e qual sia quello stagno, Tu il vederai : pero qui non si conta. 120 Ed io a lui : Se il presente rigagno Si deriva cosl dal nostro mondo, Perche ci appar pur da questo vivagno ? 123 Ed egli a me : Tu sai che il luogo e tondo, E tutto che tu sii venuto molto Piu a sinistra giu calando al fondo, 126 Non se' ancor per tutto il cerchio volto ; Perche, se cosa n' apparisce nuova, Non dee addur maraviglia al tuo volto. 129 Ed io ancor: Maestro, ove si trova Flegetonta e Lete, che dell' un taci, E 1' altro di' che si fa d' esta piova ? 132 In tutte tue question certo mi piaci, ' Plegiatonta I. stigne il fregientoia (sic) L. 117 giu vanno A. sen vanno F. sen van EGHIP. van su in per sta stretta A. stretta om. F. 118 Insin Ik ove YBD. Infino dove la I. Infin Ik ove BCEGHKLOZa. 120 Tu '1 ti vedrai DA 2 C. Tu Io vedrai ABCEFHILMOQ>a. Tuttel ved. GK. 122 nel nostro D. del vostro KJ?. 123 Perche appare I. ci par pur questo L. pure a questo ABCEFGH I KMOQZtfa. 124 ch 1 e lungo e tondo Z. 125 Con tutto LP. 126 Pur a EGMOQZCZ?. A man sinistra F. calcando P. 128 Pero M. t' apparisce B. n' apparesse EMP. 129 dur E. indur F. a tuo L. 130 Ed io a lui I. Maestro ancor KO. 131 Flegetonte ABCDHKMOPQZ^a. Lethe e Fre- getonta F. Plegiatonta e Leteo I. Fregente Lethe (sic} L. Lethe e Fregetonta F. Leteo D. che 1' un G. di che 1' un M. che tu taci C. 132 E om. F. E 1' altro che si fa G. dici E. si fa questa Q. di questa EG. dema (?) piova I. pruova L. prova D. 133 tutte tuo BCQ. tutte tua G. molto mi G. tutte tue quistion tutte L. 116 Similarly I has Plegra in 1. 58, and Plegiatonta again in 1. 131. 122 There is a difference h. I. between Witte's 4to and 8vo editions, probably owing to a misprint. I have followed the 8vo, which reads dal, while the 4to has del. If del is intentional in the 4to edition, then ABCD CANTO XIV. 103 Rispose ; ma il bollor dell' acqua rossa Dovea ben solver 1' una che tu faci. 135 Lete vedrai, ma fuor di questa fossa, La dove vanno 1' anime a lavarsi, Quando la colpa pentuta e rimossa. 138 Poi disse: Omai e tempo da scostarsi Dal bosco : fa che diretro a me vegne : Li margini fan via, che non son arsi, 141 E sopra loro ogni vapor si spegne. 136 vedra D. mo fuor B. ma non in qu. I. quella P. 138 da colpa D. e pentuta e rimossa F. e ben tutta rimossa Z. rimorsa L. 139 E disse K. a tempo L. da costarsi C. di scostarsi GIKZ. 140 Del bosco L. or fa che tu dietro mi F. dietro a me tu B. direto a me tu I. che retro C. di dietro Z. di rietro mi E. 141 E margini F. 142 sopra a FI. sopra locho L. dispegni I. should be registered above as exhibiting it, all the other MSS. here exa- mined having dal, as in the 8vo text. Similar differences occur at xviii. 79 and xxvi. 137. 123 F has 'pure' bis. CANTO DECIMOQUINTO Ora cen porta 1' un de' duri margin!, E il fummo del ruscel di sopra aduggia SI, che dal foco salva 1' acqua e gli argini. 3 Quale i Fiamminghi tra Guizzante e Bruggia, Temendo il fiotto che ver lor s' avventa, Fanno lo schermo, perche il mar si fuggia; 6 E quale i Padovan lungo la Brenta, Per difender lor ville e lor castelli, Anzi che Chiarentana il caldo senta ; 9 A tale imagine eran fatti quelli, Tutto che ne si alti ne si grossi, Qual che si fosse, lo maestro felli. 12 Gia eravam dalla selva rimossi 1 del om. (cler. error) P. cimporta C. cemporta EH. porta un D. 2 Che '1 fumo M. del fuocho K. de ruscel a. L reads thus : E stranno del...di sopra aduggia. ' 3 e om. ACEFHOa. gli om. K. 4 Q ua l Fiamengi E. 5 il fiocho A. fiato M. fiocco P. che'nver lor ACEFHILOZ/A che tra lor M. 6 lo scholmo L. lor schermi M. pur che BCDEGHKLOPZ^ 2 ^CZ;a. li fuggia L. sen fuggia M. 7 Padoani EQ. Paduan N. 9 Chiarenta D. Charantana A. Charenthana A. 10 imagini ABCGP. era F. eran gia C*. v' eran L. 11 Tanto che AG. Tutto che non era ne L. ist ne om. F. ne che si ampi M. non si alti I. 12 Chi che F. Quai ch' ei si O. 2 It seems as if the scribe of L, possibly writing from dictation, has missed part of this line. I have noticed in other MSS. (e.g. v, <, i, w, 4 and 8) several instances of gaps similar to this. Another explanation would be a defect in the original MS., however caused. A similar case occurs again in M at xxv. 3. CANTO XV. 105 Tanto, ch' io non avrei visto dov' era, Perch' io indietro rivolto mi fossi, 15 Quando incontrammo d' anime una schiera, Che venia lungo 1' argine, e ciascuna Ci riguardava, come suol da sera 18 Guardar 1' un 1' altro sotto nuova luna; E si ver noi aguzzavan le ciglia, Come '1 vecchio sartor fa nella cruna. 21 Cos! adocchiato da cotal famiglia, Fui conosciuto da un, che mi prese Per Io lembo, e grido : Qual maraviglia ? 24 Ed io, quando il suo braccio a me distese, Ficcai gli occhi per Io cotto aspetto Si, che il viso abbruciato non difese 27 La conoscenza sua al mio intelletto; E chinando la mano alia sua faccia, 14 che non DC. dov' io era K. 15 Perche ind. EQ. 16 Quando incomincio K. 17 venien CDLPQ. venian FGHM. argini Q. 18 remirava F. risguardava P. come il sole L. la sera Q. 19 Guardar un altro CHOPZ^a 2 . un 1' altro E. 20 in ver noi I. si in noi C. 21 Come vecchio PZ. sarto L. 23 Fu' io AFLO. Conosciuto fu' io P. 25 E quando GH. E quando il br. suo P. io & il oin. L. suo om. F. 26 Gli occhi ficcai ZZ?. Ficcai ben gli M. 1' occhio Q. corto C*K. 27 brusiato K. 29 E om. L. chin, la mia IMZ>. alia mia faccia B. 28 almion tellette may serve as a sample of G's unintelligent ortho- graphy. 29 In spite of the ill-mannered savagery of Scartazzini's note, 1 think there is very much to be said for the reading chinando la mia alia sua faccia. (i) Is not chinare la mano a very strange expression, whereas we have quite naturally chinare with faccia, just as we have it again in Purg. xi. 73, and as we also have it with fronte in Ptirg. iii. 44 ; with visa in Inf. v. i io ; and with testa in Inf. vi. 92 ? We might add also with occhio and ciglia in Purg. ii. 40 and vii. 13. (2) What would be the meaning here of such an action as chinare la mano alia sua faccia? The same action as applied to faccia is exactly what is described in 1. 44 just below, and is consistent with lines 17 &c. and 26, immediately before. Among the old Commentators only Buti and Barg. adopt the reading here advocated. Lan. Bocc. Veil, and Dan .have la mano, the first two without explanation. Lan. says ' fece schermo all' anima per meglio vederla 106 INFERNO Risposi : Siete voi qui, ser Brunette ? 30 E quegli : O figliuol mio, non ti dispiaccia, Se Brunette Latini un poco teco Ritorna indietro, e lascia andar la traccia. 33 Io dissi a lui : Quanto posso ven preco ; E se volete che con voi m' asseggia, Farol, se piace a costui, che vo seco. 36 O figliuol, disse, qual di questa greggia S' arresta punto, giace poi cent' anni Senza arrostarsi quando il fuoco il feggia. 39 Pero va oltre : io ti verro a' panni, E poi rigiugner6 la mia masnada, Che va piangendo i suoi eterni danni. 42 Io non osava scender della strada 30 se' voi qui G. qui voi P. sier Br. M. Burnetto Ha (and both again in 1. 32). Bronecto Q. 31 Ed egli a me, Figluol non FL (with quelli for egli) M. Ed egli I. Fiol AK. Filglo Q. mio om. AEHO^4a. O om. G. 32 Ser B. Lat. ADEFGHIKL M (Sier) PQZZ>a. Brunuto C. Latino ABCMOZ. 33 Ritorni L. adrieto M. diretro Q. 34 Io dissi lui GHPZS. Ond' io a lui M. vi priego K. 35 mi seggia MQ. 36 Farolo L. se '1 piace colui M. ch' io vo KMQ. 37 O ffiol K. O om. L. que da L. 39 rostarsi ABCDGHLOC. restarsi (or rist.) EKMPQ*/?. fregia AE. foco leggia C (?). foco freggia G. 40 Pero vavlentre (?) L. oltra Q. et io GKMa. ti veroe (?) Z. da panni P. 41 rigiugneremo D. teneri (sic) danni L. 42 van F. i loro F. 43 usava C. dalla P. colla mano,' which can hardly be the meaning of chinando la mano alia SUA faccia. It would rather have implied mia faccia, and would be an action due to excessive light, of which there is no evidence. Dan. says it was an act of extreme intimacy and familiarity. Ott. Anon. Fior. and Land., pass by the passage in silence. Castelvetro, following the corrupt Aldine text, has alia mia faccia, but corrects it in his Commentary to sua, and explains that Dante 'si china ver lui con le mani [but the text has la mano]. ..per careggiarlo e quasi abbracciarlo,' &c. 30 The spelling Burnetto is found on the monumental pillar of B. Latini in Santa Maria Maggiore at Florence [see the figure in the Vernon Dante, vol. in]. 37 L constantly writes quel for qual, and vice versa ; and also frequently omits a final letter (especially /, n, and r), hence que probably = quel=qual. 39 Is leggia in C a clerical error for ^ I feggia ? 39 But sanzarostarsi may stand for sanza rostarsi, or sanz 1 arrostarsi. CANTO XV. lO/ Per andar par di lui : ma il capo chino Tenea, come uom che reverente vada. 45 Ei comincio : Qual fortuna o destine Anzi 1' ultimo di quaggiu ti mena ? E chi e questi che mostra il cammino ? 4 8 La su di sopra in la vita serena, Rispos' io lui, mi smarri' in una valle, Avanti che 1' eta mia fosse piena. 51 Pure ier mattina le volsi le spalle : Questi m' apparve, tornand' io in quella, E riducemi a ca per questo calle. 54 Ed egli a me : Se tu segui tua stella, Non puoi fallire al glorioso porto, Se ben m' accorsi nella vita bella: 57 E s' io non fossi si per tempo morto, Veggendo il cielo a te cosi benigno, Dato t' avrei all' opera conforto. 60 Ma quell' ingrato popolo maligno, Che discese di Fiesole ab antico, E tiene ancor del monte e del macigno, 63 Ti si fara, per tuo ben far, nimico : Ed e ragion ; che tra li lazzi sorbi 44 capo inchino K. ma capo L. 45 Tenevan huomini L. 46 o qual destjno ADKQ. e destino G. o ver dest. M. 47 Nanzi GP. Innanzi H. Si quaggiu anzi 1' ultimo ti mena F. qui giu K. 48 E chi e quei GMA Z CD. E qual P. E chi e quello I. che ti m. GMPAzCDa.^?). mostro M. 50 Risposi (or -e) lui DGKLZa. Risposi a lui IMOC 51 Innanzi che BZ. Anzi che a. 52 gli volsi GM. 53 m' apparse EHPQ. tornando quella L. ritor- nando in Z.Z?. tornando in QC. ritornando io D. in ella A t . 54 E om. L. riducemi qua D. rid. qui G. queste M. 55 segni M. tuo stella DI. sua G. 56 il glorioso F. a glor. AHPZj?. 57 vita fella L. 59 Vedendo DIKMCZ>. Venendo in cielo L. 61 e maligno FGIZ>. 62 da Fies. MQZSD. 63 del mal cigno A. 65 rason E. li acerbi sorbi Q 2 . 47 Qui giu for quaggiu seems to be a regular change in K. 65 doubtless had lazzi here and al in the next line. 108 INFERNO Si disconvien fruttare al dolce fico. 66 Vecchia fama nel mondo li chiama orbi, Gent' e avara, invidiosa e superba : Da' lor costumi fa che tu ti forbi. 69 La tua fortuna tanto onor ti serba, Che 1' una parte e 1' altra avranno fame Di te: ma lungi fia dal becco 1' erba. 72 Faccian le bestie Fiesolane strame Di lor medesme, e non tocchin la pianta, S' alcuna surge ancora in lor letame, 75 In cui riviva la semente santa Di quei Roman, che vi rimaser, quando Fu fatto il nido di malizia tanta. 78 Se fosse tutto pieno il mio dimando, Risposi lui, voi non sareste ancora Dell' umana natura posto in bando : 81 Che in la mente m' e fitta, ed or mi accora La cara e buona imagine paterna Di voi, quando nel mondo ad ora ad ora 84 M' insegnavate come 1' uom s' eterna : 66 il dolce EFGILMOQ 2 ZZ>. a dolce C. 67 gli fa orbi G. 70 tanto ben ti KZ. 71 avera fame MQ. 72 all' erba M. 73 Fac- cino I. Faceam C. 74 medesmi DL. medesmo BZa. Di lor semenza e non guastin I. Di lor semenza D. tocchir BCHOP. tocher A. tochi EKM. tochiron L. 75 S' alcuna vergha surge F. S' alcuna ne risurge G. nel lor ABCDEHlLOPQZ/>Y>>a. de lor K. 76 In cui ruina AG. ruvina L. sementa BCDGH*LOQZ'ZX semenza K. 78 nidio DZ.Z?. mondo L. 79 Se '1 fosse M. pieno tutto El. 80 Rispos'io ABEFHQ6Z>. Risposi a lui 1 MO. saresti AEGHKMOP. 81 Dall' um. AB^CEFGHZA posti LM. 82 Che la mente ACEFGHLMOPQa. Chon la mente me ficha K. aficta E. ma fligge (sic) M. ed om. L. 83 E la cara E. La cara buona DKP^4 2 Z'C'a. La buona ecara M. Lachiara I. imag. e pat. AB 1 DEHLOlM 2 /)Ca. 85 Mi mostravate Z. s' interna MP. 74 An injudicious emendation in I, since semenza occurs two lines below. 82 Line 82 has been omitted in F and added (by the same hand) at the bottom of the page. 83 Cara and chiara are not un- frequently interchanged in MSS., see note on Par. xxv. 33. So also etenui and interna, as in 1. 86. See notes on Par. xvii. 9, and xxiii. 1 15. CANTO XV. 109 E quant' io I' abbia in grado, mentre io vivo Convien che nella mia lingua si scerna. 87 Cio che narrate di mio corso scrivo, E.serbolo a chiosar con altro testo A donna che sapra, se a lei arrivo. 90 Tanto vogl' io che vi sia manifesto, Pur che mia coscienza non mi garra, Che alia fortuna, come vuol, son presto. 93 Non e nuova agli orecchi miei tale arra : Pero giri fortuna la sua rota, Come le piace, e il villan la sua marra. 96 Lo mio Maestro allora in sulla gota Destra si volse indietro, e riguardommi ; Poi disse : Bene ascolta chi la nota. 99 Ne per tanto di men parlando vommi Con ser Brunetto, e domando chi sono Li suoi compagni piu noti e piu sommi. 102 Ed egli a me : Saper d' alcuno e buono : Degli altri fia laudabile tacerci. 86 E quanto 1' abbia DGMPQCZ>. E quanto 1' abbi I. 1' abbo Z. ingrato (prob. = in grata) Qa*. a grato ADEGKA^CD. a grado F. in grato CHOPZ. abbia caro B*. abbia grato M. mentre vivo DKQZ. 87 mia vita ID. si discerna F. si cerna D. che nella mente mia Z. 88 di me io D. del mio M. 89 E om. L. 90 che '1 sapra ADFGIKMa. 91 Tanto voglio che GHOC Ma tanto voglio ve sia M. E tanto vo' che D. chi vi O. 92 mingara ( = me ne garra ?) K. 93 Colla fortuna L. 94 nuovo GOAC-D. all' orechie mie DMPQ. agli occhi miei C. questa arra I. 95 girin A. gira L. 96 Come gli CHOP. li piace Qa. 97 in sulla gote F. 99 ben 1' ascolta AFIZA a chi L. cui la M. 100 Non per tanto DKM. Non per tanto io parlando F. parlando huomini L. 101 Burnetto ACa. Bornetto K. e om. LM. 102 Gli sua G. E suo I. Li suo C. e piu noti I. e i piu noti K. 103 D' algun saper O. 104 tacere A. a tacerci Q. 86 B had a longer reading v a manu, and apparently in grado. 87 F seems to read ilia ( = in la) for nella. In Z scerna has obviously been altered by a later hand to cerna. 100 Oy. (as often) an oral blunder in L ? 102 e piu noti in I probably stands for / piu, as e represents i at the beginning of this line, wcA passim in this MS. HO INFERNO Che il tempo saria corto a tanto suono. 105 In somma sappi, che tutti fur cherci, E letterati grandi, e di gran fama, D' un peccato medesmo al mondo lerci. 108 Priscian sen va con quell a turba grama, E Francesco d' Accorso; anco vedervi, S' avessi avuto di tal tigna brama, m Colui potei che dal servo de' servi Fu trasmutato d' Arno in Bacchiglione, Dove Iasci6 li mal protesi nervi. 114 Di piu direi; ma il venir e il sermone Piu lungo esser non puo, pero ch' io veggio La surger nuovo fummo del sabbione. 117 Gente vien con la quale esser non deggio; Siati raccomandato il mio Tesoro Nel quale io vivo ancora; e piu non cheggio. 120 Poi si rivolse, e parve di coloro Che corrono a Verona il drappo verde Per la campagna; e parve di costoro 123 Quegli che vince e non colui che perde. 105 sara G. corto, tanti sono DLM. 106 furon L. 107 Eli terrati tutti (sic) Q. 108 D' un peccati D. med. pecc. IB. 110 E om. AEHLO. da Corso EM. ancor DFGI. anche ved. CQZa. anco (or -che) e ved. AEHKLOP.5. ill Potei s'avessi di F. di cotal L. 112 E colui ancor F. potevi GDI. 114 mal ser- tisi (?) L. 116 non puo esser DK. non potria esser E. perch' io DE. 117 fummo nuovo A. fummo om, L. dal sabb. CDKZ>. 118 Genti Q. con le qual KQ. 120 ne piu non deggio L. 121 si parti B. 122 corsono D. il palio verde I. pano verde E. al drappo O. 123 e om, L. 124 Colui che v. e non di quei F. Colui che v. 1. e om. ABCEGHLMOPQa. non color M. quegli che perde G. 110 But anche ved. in C &c., may stand for anco e i>cd. Ill, 112 The changes in F in these lines are evidently introduced to avoid the long interval between vedervi and potei. CANTO DECIMOSESTO Gia era in loco ove s' udia il rimbombo Dell' acqua che cadea nell' altro giro, Simile a quel che 1' arnie fanno rombo; 3 Quando tre ombre insieme si partiro, Correndo, d' una torma che passava Sotto la pioggia dell' aspro martiro. 6 Venian ver noi, e ciascuna gridava: Sostati tu, che all' abito ne sembri Essere alcun di nostra terra prava. 9 Aime, che piaghe vidi ne' lor membri Recenti e vecchie dalle fiamme incese ! Ancor men duol, pur ch' io me ne rimembri. 12 Alle lor grida il mio Dottor s' attese, Volse il viso ver me, ed : Ora aspetta, Disse ; a costor si vuole esser cortese : 15 1 illocho (sic) a. onde AEHLMOPa. il om. Ba. ond' io s' udia rimbombo G. 2 alto BGP. nel quarto M. 3 a quei AHL. 1' arme DFHIKLC. 1' arne CGOP^?). 1' arno AE. 1' ape MP 2 . rimbombo Q*. 5 turba LZ. turma ACEH^ 2 Ca. terma K. 7 Veneno A. 8 che ; 1 abito LZ. con 1' abito K. n' assembri B*CDI. 10 O me C. vid' io G. vitti (sic) M. 11 Vecchie e recenti G. e om. L. delle DI. accese I. 12 mi duolADGK. per ch' io C. che me ne K. mi ni rim. AE. mi ne rim. HO. 14 E volsesi ver me F. in ver me C. e disse asp. AC*EFGHLOP^!. a has, e disse ora aspetta, Disse, &c. with ora and Disse slightly erased. 15 Qui a costor C*. Che a costor AFG. Costoro se vol esser E. A costoro si vuoP esser HLOP. Ora a cost. A$). 8 B has evidently been altered 2 lia inanu from nc sembri. 112 INFERNO E se non fosse il foco che saetta La natura del loco, io dicerei, Che meglio stesse a te, che a lor, la fretta. 18 Ricominciar, come noi ristemmo, ei L' antico verso ; e quando a noi fur giunti, Fenno una rota di se tutti e trei. 11 Qual soleano i campion far nudi ed unti, Avvisando lor presa e lor vantaggio, Prima che sien tra lor battuti e punti : 24 Cosl, rotando, ciascuno il visaggio Drizzava a me, si che in contrario il collo Faceva a' pie continuo viaggio. 27 Eh, se miseria d' esto loco sollo 17 i' ti direi C. io direi F. 18 a lor ch' a te D. 19 Rico- minciaron AEHP. E cominciaron I. Rincominciaron come noi restemo D (corrected by later hand to come restamo lei]. restammo AEH^4. ristamo C. come restammo FIL. 20 come a noi FK. e om. L. 21 Fer D. Fero F. Fermo (?) L. Femo M. de lor Ma. tutti trei AELPO. 22 i om. M. nuti (corrected to nudi} D. e nudi G. furono nudi unti L. 25 Et si rot. ABEHMOPQa. Essi C. ciascun lor D. ciascuna-ff. viaggio L. 26 Drizzavam si (corrected to a me si) D. Drizzavan L. Drizzaro BQ. si che tra loro AMA C 2 /?a. che 'n tra loro GHO. si tra lor L. si che contra lor P. che a contrario D. in om. BFIQZ. 27 Faceano (or -vano) BGM. Facendo C. Facea col pie O. Fac. e i piedi C. a piedi M. contrar viaggio G. continuoi B. 28 E (or et) se BDEGHIKLMOPZa. Deh se Q. di questo E. 18 D is corrected by a later hand to a te elf a lor. 22 The reading in L is a curious blunder, and evidently arose thus. Among L's personal caprices are the frequent omission of e, and the substitution of furono for fur or furo even in spite of rhythm. See note on xiv. 44. This has been here substituted for far, which was first no doubt misread fur. M has nuti for nudi two or three times, though not h. /., just as it has -vitti for tw&ft'jost above in 1. io. The interchange is not uncommon in other words, e.g. Sghermidor and Sghermitor in xxii. 142, corritor and corridor in xxii. 4, falsator and falsador in xxix. 57, &c. 26 B has che con- trarol chollo. 27 B has irisaggio (an obvious error) (see 1. 25). Cf. E in 1. 73. 28 Z has ' al. deli ' in marg. i da manu. CANTO XVI. 113 Rende in dispetto noi e nostri preghi, Cominci6 1' uno, e il tinto aspetto e brollo; 30 La fama nostra il tuo animo pieghi A dime chi tu se', che i vivi piedi Cosl sicuro per lo inferno freghi. 33 Questi, 1' orme di cui pestar mi vedi, Tutto che nudo e dipelato vada, Fu di grado maggior che tu non credi. 36 Nepote fu della buona Gualdrada : Guido Guerra ebbe nome, ed in sua vita Fece col senno assai e con la spada. 39 L' altro che appresso me 1' arena trita, E Tegghiaio Aldobrandi, la cui voce Nel mondo su dovria esser gradita. 4? Ed io, che posto son con loro in croce, Jacopo Rusticucci fui : e certo La fiera moglie piu ch' altro mi nuoce. 45 S' io fussi stato dal foco coperto, Gittato mi sarei tra lor disotto, E credo che il Dottor 1' avria sofferto. 48 29 in om. G. in dispregio A. a noi D. e i nostri B. 30 con tristo asp. D. col tristo I. e '1 tristo BKPQZA 2 C>. uno il tristo F. aspetto brollo A. asp. il brollo L. 31 fama neora (?) G. il om. F. prieghi CD. 32 che vivi DFHIKMOPQ 2 CZ?a. che vivi i piedi C. che vivi e piedi L. a vivi E. 33 Cosi sicuri C. Cosi ardito F. 34 Queste A. Quest' orme CB. Queste orme Z l (It inserted later above). Queste 1' orme AH Pa. Queste e 1' orme E. Questo di cui pestar 1' orme G. Costui 1' orme D. che 1' orme sue F. 1' ormi I. cui tu pest. ZB. ne vedi E* (corrected from me) HLO. Q has Quisti, and /' orme altered to horme. 36 Fur ~DB. 38 ed om. KLaj. in om. A. 39 Assai fece col P. Col senno fece B. co senno Z. 40 L' altra CE 2 . che presso ADZ>. ap- presso a me FIL. la terra trita I. 41 E '1 L. 42 dovrebbe DFA si dovrebbe Z. 43 Ed om. a. 44 fu' io L. 45 La mala moglie C. altri FP. mi chuoce D. mi quoce I. 46 Se fosse G. da foco C. 47 saria O. con lor I. 31 In C the r of prieghi has been erased. 32 che vivi is really not conclusive. D. I. 8 114 INFERNO Ma perch' io mi sarei bruciato e cotto, Vinse paura la mia buona voglia, Che di loro abbracciar mi facea ghiotto. 51 Poi cominciai : Non dispetto, ma doglia La vostra condizion dentro mi fisse Tanto, che tardi tutta si dispoglia, 54 Tosto che questo mio Signer mi disse Parole, per le quali io mi pensai, Che qual voi siete, tal gente venisse. 57 Di vostra terra sono ; e sempre mai L' opre di voi e gli onorati nomi Con affezion ritrassi ed ascoltai. 60 Lascio Io fele, e vo per dolci pomi Promessi a me per Io verace Duca ; Ma fino al centro pria convien ch' io tomi. 63 Se lungamente 1' anima conduca Le membra tue, rispose quegli, ancora, E se la fama tua dopo te luca, 66 Cortesia e valor, di', se dimora 49 perchk mi ACE. miaserea (?) L. 51 mi fecie moto L. 52 non dispregio D. 53 m' affisse DAa.. 54 Tanta Ha. tutta tardi E. sene spoglia F. 55 mio Maestro disse L. 57 sieti P. 58 nostra EL. 59 Sopra di voi D. L' opra AEMPQZ. L' ovra BCFHKO^a. L' ovro G. L' oprar I. di voi onorati FL. 60 ed om. L. intesi ed asc. Aj. 61 per li D. pe ( = pei) dolci I. 63 Ama fino (sic) F. Ma infine al DKL. Ma 'nfino GZa, fina al cerchio M. prima C. pur convien L. che tomi ADEGILQ. 65 ripose elli allora D. 66 diponte lungha (with g erased) L. da poi te M. 67 disse, dim. BCDFGHMOPQ^a. dinne, dimora Z. disse ei I. disc ( = di' se, or disse ?) K. 49 Lines 50 and 51 are accidentally transposed in L. 52 Comp. a similar change of dispetto to dispregio in 1. 29 in A. 59 L apparently has F aura, a clerical error for F ovra. 61 I, it will be remembered, constantly omits final i when following a, e, or o. 67 Even disse n\ay = d? se, since this fusion of words and doubling of consonants occur passim in some MSS. A case occurs in xix. 36, where C reads disse for di st, and similarly in xix. 73 misson for mi son. In fact in many MSS. disse, dp se, and di se, would be practically undistinguishable. Z (marg.) has al. disse. CANTO XVI. 115 Nella nostra citta, si come suole, O se del tutto se n' e gita fuora ? 69 Che Guglielmo Borsiere, il qual si duole Con noi per poco, e va la coi compagni, Assai ne cruccia con le sue parole. 72 La gente nuova, e i subiti guadagni, Orgoglio e dismisura ban generata, Fiorenza, in te, si che tu gia ten piagni. 75 Cosl gridai colla faccia levata : E i tre, che cio inteser per risposta, Guatar 1' un 1' altro, come al ver si guata. 78 Se 1' altre volte si poco ti costa, Risposer tutti, il satisfare altrui, Felice te, se si parli a tua posta. 81 Pero se campi d' esti lochi bui, E torni a riveder le belle stelle, Quando ti giovera dicere : lo fui, 84 Fa che di noi alia gente favelle. 68 vostra EL. 69 nel tutto K. getta F. gietta I. gito B. 70 Guiglelmin D. Porsie G. sen duole DA 2 C. 71 Con noi e va in Ik co' comp. DI. poco vale cui compagni L. con comp. M. 72 Molto ne A. Ch' assai D. colle suoi par. Q. 73 La nova gente M. e subiti ACDFGHIKMa. e subito i com- pagni (obvious error) E. e i om. L. 74 e om. D. a dism. a. ha gen. DM. ha ingen. EGQ 1 (?). 75 In te Fir. I. E tu Fior. che gia tene a. che tu tene G. tu om. AI. gia ne p. D. gia conpiangni F. 77 E tre D. E in tre L. che co tesoro D x . 78 Guardo DM. Guardar (or -aro) ABEHIKOQZ^(?)^Z>a. come '1 ver IL. 79 Se altre I. volte om. (obvious error) K. 80 Ris- puose a tutti sosdisfare a. a sodisfare DI. al C* (altered from il). 81 setu parli D. che si parli I KZ>. 82 scampi D. di questi E. 83 e ritorni ED. 84 quanto D. di dire D. 85 alle genti DQa. 77 A later hand in D has inserted i above co (do) and 1 before tesoro, erasing the last o. cb=cib occurs again in xxx. 148, and i after c is very frequently omitted in this MS. under similar circumstances (see description of the MS.). 78 guatar in C has been altered z da manu to guardar. 82 1 1 6 INFERNO Indi rupper la rota, ed a fuggirsi Ale sembiar le gambe loro snelle. 87 Un ammen non saria potuto dirsi Tosto cosl, com' ei furo spariti : Perche al Maestro parve di partirsi. 90 lo lo seguiva, e poco eravam iti, Che il suon dell' acqua n' era si vicino, Che, per parlar, saremmo appena uditi. 93 Come quel fiume, ch' ha proprio cammino Prima da monte Veso in ver levante Dalla sinistra costa d' Apennino, 96 Che si chiama Acquaqueta suso, avante Che si divalli giu nel basso letto, Ed a Forll di quel nome e vacante, 99 Rimbomba la sopra san Benedetto Dell' alpe, per cadere ad una scesa, Ove dovea per mille esser ricetto ; 102 86ed0;. A. al fuggirsi IA t . 87 Ali ABCDEGHKLMOP^Ca. Alie FZ. sembrar F. sembiar P. sembiaron C*M^4 1 (?). senbiavano le loro ganbe I. le dame (?) G. le loro gambe K. 88 amme BDH.5. non s' avrebbe E. non sarei G. possuto OPQ (pussuto). 89 Si tosto DF. como fuor A. come f. EGHKLMOQa. come furon dipartiti D. come furo dispariti BF. come furo spartiti C. come funno spariti I. 90 al mio Duca F. il Maestro L. 91 I o lo 'nsegniai e poi ch' erano L. eravano D. 92 v' era EI. n' eran M. 93 appena sar. D. 94 E come D. che per se il cam. L. 95 Pria L. dal monte Q. monte Verso BCDFM. Vesso A. Niso EQ. Nesso K. Viso L. monte insu in ver P. monte giuso ver lev. C. in o m. F. 96 Delia L. dasinino L. 97 si chiama Queta A. 98 Poi si divalla F. giuso nel D. qui nel O. 99 monte vacante D. e om. M. 101 De lapi (sic) M. 102 Lo qual dovria O. DoveM^4 2 C. dovria DFZB. doveva E. dove' L. dovia QCZ>. mile EKL. 87 A writes ' Ally.' D has been altered by a later hand to sembiaron le lor. 95 insu in P is perhaps, as Barlow suggests (Contributions, p. 68), an error for the almost indistinguishable visu (which would pro- bably be the form of visa in this MS.). 102 The form mile proves nothing for the absurd interpretation miles (sc. Christi) i.e. "nel quale 1' autore dovea essere ricevuto per frate," suggested by Scarabelli h. I. in his CANTO XVI. Cosl, giu d' una ripa discoscesa, Trovammo risonar quell' acqua tinta, SI che in poc' ora avria 1' orecchie offesa. 105 lo aveva una corda intorno cinta, E con essa pensai alcuna volta Prender la lonza alia pelle dipinta. 108 Poscia che 1' ebbi tutta da me sciolta, Si come il Duca m' avea comandato, Porsila a lui aggroppata e ravvolta. m Ond' ei si volse inver lo destro lato, Ed alquanto di lungi dalla sponda La gitto giuso in quell' alto burrato. 114 E' pur convien che novita risponda, Dicea fra me medesmo, al nuovo cenno 103 Cosi da una E. per una ripa Q. ripo G. roccia disc. K. discesa L. 104 Sentimmo ~DYA(marg.)CD. di sonare un' ac- qua L. 105 Si che poc' ora IL. avrei G. 1' orecchia ABGIPQZ?. la lingua B. 106 lo ch' aveva D. 108 ha la pelle A. 109 Et poi ch' io F. Poscia ch' io K. da me tutta ILQA 2 CDa,. HO come '1 mio Duca D. ill raggroppata L. raccolta DLP. avvolta BGI. rivolta AMOQ. et involta K. 112 Et e' si volse F. Onde si ACDEIa. volse ver lo DMZ. in- verso il a. 113 da lungi Q. della sp. D. 114 giu in DFHM. altro RCLOQZA(marg.)CD (a* has alto). barato E. 115 Et pur IZ?a. El pur FQ. E per L. novitk convien che F. 116 Dicea io fra me stesso F. in me C. tra me ID. col novo E. a novo G. al om. L. edition of the Commentary of della Lana. This is of course in allusion to the tradition that Dante once thought of becoming a Franciscan monk, and even went so far as to enter some lower grade in the order ; and indeed, according to one legend, he died clothed in the Franciscan dress. ' L ' for instance has mile again for mille in xxvi. 66. Mile is a form equally unknown and impossible as representing the Latin miles, or rather militem, and Scarabelli's suggestion is justly called by Blanc " eine wun- derliche Auslegung." The ' corda'' in 1. 106 has sometimes been sup- posed (independently of any such interpretation here) to refer to the Fran- ciscan dress. O has a curious marginal variant here " al. de (?) per Emilia." 103 The reading in K is evidently a reminiscence of xii. 8. 115 E pur (as in Z) may represent either E* or Et. The same remark applies to E disse in 1. 121, which is really indeterminate. Il8 INFERNO Che il Maestro con 1' occhio s\ seconda. 117 Ahi quanto cauti gli uomini esser denno Presso a color, che non veggon pur 1' opra, Ma per entro i pensier miran col senno ! 120 Ei disse a me : Tosto verra di sopra Ci6 ch' io attendo, e che il tuo pensier sogna Tosto convien ch' al tuo viso si scopra. 123 Sempre a quel ver ch' ha faccia di menzogna De' 1' uom chiuder le labbra finch' ei puote, Per6 che senza colpa fa vergogna; 126 Ma qui tacer nol posso : e per le note Di questa commedia, letter, ti giuro, S' elle non sien di lunga grazia vote, 129 Ch' io vidi per quell' aer grosso e scuro Venir notando una figura in suso, Maravigliosa ad ogni cor sicuro, 132 SI come torna colui che va giuso Talora a solver ancora, ch' aggrappa O scoglio od altro che nel mare e chiuso, 135 Che in su si stende, e da pie si rattrappa. 117 Cosi il K. Che si Maes. L. E che 1' occhio del Maes. F. cogli occhi BG2?. ci seconda G. si om. K. 118 Alquanto L. quanti cauti G. 119 Presso a colui L. Pure a color F. Presso color EQ. vengion C. non miran M. 1' ora (cler. error) E. 120 Ma dentro al pensiero D. entro il pens. BFKL. entro pensier C. per entr* al pensiero I. entro ai pens. A v vede col M. per senno K. 121 Poi disse D. veran de sovra K. 122 e om. L. il mio pens. B. 123 si copra A. 124 Semple F. Sempre quel M. faccie P. 125 chiuder la bocca I. infino che '1 BDK. quanto puote Z. quant' el puote F. quant' ei B. infin cho petti (?) L. 127 il tacer non posso L. non posso KMQ(?)ZZ>. non posso tacere F. e om. L. 128 lettore ingiro (sic] L. 129 da lunga C. lunga fama F. 131 nodando A. 132 e da ogni E. 134 Talvolta YZA^BCD. a sciogliere IQ. 1' ancora ABCDEFGH IKLMOPQZa. 135 A scoglio DI#. o a altro D. 136 ed a pie A. scende (?) G (but / and c are often scarcely distinguishable). CANTO DECIMOSETTIMO Ecco la fiera con la coda aguzza, Che passa i monti, e rompe muri ed armi ; Ecco colei che tutto il mondo appuzza: 3 Si comincio lo mio Duca a parlarmi, Ed accenolle che venisse a proda, Vicino al fin de' passeggiati marmi: 6 E quella sozza imagine di froda, Sen venne, ed arrive la testa e il busto ; Ma in sulla riva non trasse la coda. 9 La faccia sua era faccia d' uom giusto; Tanto benigna avea di fuor la pelle, E d' un serpente tutto 1' altro fusto. 12 Due branche avea pilose infin 1' ascelle : Lo dosso e il petto ed ambo e due le coste 1 kalla (? = ch ! ha la) C. 2 passa monti ADEHIPQ. monti rompe ZA^ mura DI. i muri BCFHKOZA 2 CD a. e 1' armi ABCEFHKOQZ^ 2 ^CZ)a. e muri e 1' arme L. 4 Cosi com. D. 5 Ed a zenolle (sic) A. accennolli M. 6 Vicin a ABCDEHOOZ/4. Vicinal IP. Infino al fin M. al fiume EHP. a fine L. passegianti EHL. 8 Si venne L. venne a rivo D. 9 Ma sulla AKLO. Ma su di fora M. Ma in su la proda I. 10 La faca soa era de huom giusto (sic) A. sua parea come d' uom D. sua si era K. sua mi pareva d' uom Z> 2 . d' un giusto KLZ. 11 Tutta benigna Q. 12 1' uno e 1' altro I. frusto DEFGHKLZa. 13 Dove (? cler. error for Doie) Q. infine all' D. pilose fin FM. in- sin BD. fra 1' ascelle E. 14 Lo petto e '1 dosso C. 14 There is endless variety of such forms as ambedue, amendue, &c. here and elsewhere in the MSS., but it seems not worth registering. I2O INFERNO Dipinte avea di nodi e di rotelle. 15 Con piu color sommesse e soprapposte Non fer mai drappo Tartar! n& Turchi, Ne fur tai tele per Aragne imposte. 18 Come tal volta stanno a riva i burchi, Che parte sono in acqua e parte in terra, E come la tra li Tedeschi lurchi 11 Lo bevero s' assetta a far sua guerra; Cosl la fiera pessima si stava Sull' orlo che, di pietra, il sabbion serra. 24 Nel vano tutta sua coda guizzava, Torcendo in su la venenosa forca Che, a guisa di scorpion, la punta armava. 27 Lo Duca disse : Or convien che si torca La nostra via un poco infino a quella Bestia malvagia che cola si corca. 30 Pero scendemmo alia destra mammella, E dieci passi femmo in sullo stremo, Per ben cessar la rena e la fiammella : 33 E quando noi a lei venuti semo, 16 Dipinto ACDEFMPa. Dipinti BGL. Dipintavea H. e om. L. 16 son messe ACFKa. sommessi BGQ. commesse BD. son proposte D. 17 fe C. fur IQ(?). drappi ABGIMQZ?a*. 18 fur mai D. tal tele AEFIKM. per Ragnol D. poste K. 19 talora Z. 20 Che stanno parte in acqua F. stanno in BDIP. 21 tran li A. elurchi C. ilurchi a. tedeschi li urchi AH. todeschi burchi E. 22 bivero ABCDFLMOZ^CZ?a. buero GH. bovero K. bivaro Q. Lobrivaro (?) sa senta E. s' aspetta e fa L. la guerra K. 24 Su loro K. che dipinta sabbion L. la petra e '1 sab. M. 25 sguizzava E. 26 velenosa LZa. 27 Ch' a visa di K. scarpion DFGHZa. le punte Q. 29 insino DFI. fino KMPQZ. 32 in su 1' extremo EZ. fenno a*. 34 a lei noi ADG. a lor I. scemo (cler. error) M. 18 The form Aragne occurs again in Purg. xii. 43, 44. Comp. l li tegni di Galieno ' as a transliteration of 77 T^J/T? in Conv. i. c. 8. 21 AEH are so often in agreement that probably liurchi and burchi are in- tended for the same. Burchim&y have been an error from o/^oiore'Aeura with line 19, and then mistaken for or altered into liurchi. 34 M being a CANTO XVII. 121 Poco piu oltre veggio in sulla rena Gente seder propinqua al loco scemo. 36 Quivi il Maestro : Acciocche tutta piena Esperienza d' esto giron porti, Mi disse, or va, e vedi la lor mena. 39 Li tuoi ragionamenti sian la corti : Mentre che torni parlero con questa, . Che ne conceda i suoi omeri forti. 42 Cos! ancor su per la strema testa Di quel settimo cerchio, tutto solo Andai, ove sedea la gente mesta. 45 Per gli occhi fuori scoppiava lor duolo : Di qua, di la soccorrien con le mani, Quando a' vapori, e quando al caldo suolo. 48 Non altrimenti fan di state i cani, Or col ceffo or coi pie, quando son morsi O da pulci o da mosche o da tafani. 51 Poi che nel viso a certi gli occhi porsi, Ne' quali il doloroso foco casca, Non ne conobbi alcun ; ma io m' accorsi 54 35 in om. EM. vidi L. 36 stremo D. propinquo allo stremo F. 38 di questo E. d' esto luogho I. 39 disse va ABCDEFGHIMOPQ^a. pe na M. 40 li corti K. 42 me cone. M. ci cone. P. 43 Cosi alor M. 1' extrema E. 1' es- trema I. la strecta Q. resta D. 45 dove AEFHMPA 2 C. 46 fuora P. scoppia loro un duolo D. accoppiava C. el lor duolo I. lor el duolo Z. 47 di lo (in error for di la) H. 48 al vapore FI. e om. A. Ora al vapore e ora al F. 49 None altrimenti Q. i om. P. 50 Or col pi or col ceffo O. piedi DL. or col pie ACEFGIPQa. con pie M. Or co' piedi, or col ceffo quando morsi ZB. 51 Da pulci AEFGHMOP. Da pulci da mosconi I. Da pulci son, da mosche ZB. o da mosconi L. ovver da mosche P. tavani E. toffani M. 52 in viso E. il viso L. a certo K. accord gli occhi E. adetti (?) gli occhi I. torsi L. 53 Nel quale DI. A quali M. dol. loco D. Venetian MS. the sound of sc and s would be similar, and hence perhaps the error. Compare lascio for lasso in E (1. 130). Somewhat similarly perhaps M writes se for ci in xiii. 2, and K (another Venetian MS.) does so in xviii. 72 and xxiii. 68. 122 INFERNO Che dal collo a ciascun pendea una tasca, Che avea certo colore e certo segno, E quindi par che il loro occhio si pasca. 57 E com' io riguardando tra lor vegno, In una borsa gialla vidi azzurro, Che d' un leone avea faccia e contegno. 60 Poi procedendo di mio sguardo il curro Vidine un' altra come sangue rossa Mostrare un' oca bianca piu che burro. 63 Ed un, che d' una scrofa azzurra e grossa Segnato avea lo suo sacchetto bianco, Mi disse : Che fai tu in questa fossa ? 66 Or te ne va : e perche se' vivo anco, Sappi che il mio vicin Vitaliano Sedera qui dal mio sinistro fianco. 69 Con questi Fiorentin son Padovano ; Spesse fiate m' intronan gli orecchi, Gridando : Vegna il cavalier soprano, 72 Che rechera la tasca con tre becchi : 55 al collo DE. da collo LP. a cadauno A. di ciascuno D. 56 o certo segno A. 57 che 1' occhio lor DM. ch' a lor 1' occhio I. che loro AEGKLOPZC. che lor occhi Q. 58 E come rig. AFKAI. in lor M. 60 di lione I. 61 serciendendo or sernendendo? L. del mio A. sg. oscuro DL. il curtto (sic) K. inchurro (sic) a. 62 Vidi un AD. 63 Mostrava D. Mostrando ACEGHIOPQ (a* has mostrare). b. come burro D. piu om. P (cler. error?). 64 che a una schorza L. 65 Segnando L. 67 pero che se' D. e om. L. 68 Vitiliano D. Italiano K. 69 Sederai C. Si sedrk P. dal tuo sin. G. al mio L. 70 Tra questi D. 71 Spesse volte EL. E spesse IO. introna C. intruonono I. mi tronan AKO. 1' orecchie EK. Che molte volte me tronar M. 73 testa con D. coi tre ABCHIKPj5a. 59 In E gialla has been omitted and inserted above. Also azzurro is written egiuro, perhaps an oral blunder due to the assimilation of g and z in pronunciation. Comp. megio for mezzo in ix. 48, and line 83 inf. 63 Possibly D's exemplar may have omitted piil (cf. P) and the scribe may have assisted the sense and rhythm by substituting come for che. Lines 72 75 are written in the order 74, 73, 72, 75 in L. CANTO XVII. 123 Qui distorse la bocca, e di fuor trasse La lingua, come '1 hue che il naso lecchi. 75 Ed io, temendo nol piu star crucciasse Lui che di poco star m' avea monito, Torna' mi indietro dall' anime lasse. 78 Trovai lo Duca mio ch' era salito Gia in sulla groppa del fiero animale, E disse a me : Or sii forte ed ardito. 81 Omai si scende per si fatte scale : Monta dinanzi, ch' io voglio esser mezzo, SI che la coda non possa far male. 84 Qual e colui, ch' ha si presso il riprezzo Delia quartana, ch' ha gia 1' unghie smorte, E trema tutto, pur guardando il rezzo, 87 Tal divenn' io alle parole porte; Ma vergogna mi fer le sue minacce, Che innanzi a buon signor fa servo forte. 90 Io m' assettai in su quelle spallacce: Si (volli dir, ma la voce non venne 74 Indi storse F. Qui storse Z. dist. la faccia ABCE (ditorse) GHM(discorse)OPQa. 75 come hue ADEFGHIMOPQ 2 Z5a. che naso DGOQ^ 2 C 76 che il piu CDIM. non lo star KL. 77 Quei M. del poco DM. avea ammonite BDFIMQ^a. mu- nito EG. amunito L. 78 Tornamine E. Tornai indrieto K. a rieto E. Tornai dall' anime indrieto L. 79 E trova' il 2,A. 80 Gia sulla AEGHMOPQ5CZ?. 81 sia DFGLZ. sie CEHKOP. sii fiero I. 82 Ormai CKA V scenderk L. cosi fatte E. 83 davanti D. e io D. in mezzo L (a* has an erased blank). 84 non ti fark male D. non ti faccia male M. 85 che s' appressa al IM. ch' e si presso al F. ch' e si presso il BCLOZ. al riprezzo BO. a riprezzo C. 86 che gik AHa. che abbia 1' unghie E. che gik 1' unghie morte Z. ed ha gik K. 1' unghia smorte IO. cheglia 1' unghie L. cha lugni gik (sic) Q. (? che in Qj.) 87 tutto trema sol M. al rezzo D. sguardando E. 88 Tal doveni io B. dovenne E. 89 vergognar I. feco le D. mi fe' AEHMOPZZ?. 90 al buon BDFLP. buon om. C. il servo D. fan servo Bj^FIL (or far?)O^Z?. 92 Si volsi DGIKLA. E volli BF. Et volsi Z. 83 Chi voglio has been altered to che in C. Q has megio, ripregio and re>" 1 manu. CANTO XVII. 125 Che fu la mia, quando vidi ch' i' era Nell' aer d' ogni parte, e vidi spenta Ogni veduta, fuor che della fiera. 114 Ella sen va nuotando lenta lenta ; Rota e discende, ma non me n' accorgo, Se non ch' al viso e disotto mi venta. 117 lo sentia gia dalla man destra il gorgo Far sotto noi un orribile stroscio ; Per che con gli occhi in giu la testa sporgo. 120 Allor fu' io piu timido allo scoscio : Perocch' io vidi fochi, e sentii pianti ; Ond' io tremando tutto mi raccoscio. 123 E vidi poi, che nol vedea davanti, Lo scendere e il girar, per li gran mali Che s' appressavan da diversi canti. 126 Come il falcon ch' e stato assai sull' ali, Che senza veder logoro o uccello, Fa dire al falcon iere : Oime tu cali: 129 Discende lasso, onde si mosse snello, 112quand' ioBCIKQ. ch'eraQ. 113 e d' ogni CEFGHMO. ogni lato L. parte vidi CEGHKMOP. parte io vidi F. 114 vendeta (sic) C. veduto a. dalla C. 115 rotando ADEFGHKP^CZ?. 116 Aora is- cende (sic) D. che disc. L. manomovoacc. (sic) a. 117 al viso disotto GIA V il viso disotto F. m' avventa F. 119 scroscio GIKQ. 120 Pero in giu gli occhi e la testa isporgo D. Pero A^BCD, Perch' io E 2 GHMP. Perche gli occhi KL. Perch' io la testa giu con gli occhi porgo M. occhi giu B. porgo ACFILMZ?. scorgo P. 121 Allora fui piu E. io om. C. fu piu io A. stoscio DFHMZffDa,. scoso E. stroscio K (error from o/LtotoreXtvra). 122 Perch' io EFI. Perocch' io vidio K. sentio K. 124 ch' io nol DEM. che non vedeva avanti A. che non la L. non P udia B. 125 e '1 gridar ACDFHOPZA 2 SCDa.. e girar K. e '1 gitar L. al girar Ej (altered to el or e '/). 126 s' appressava M. di diversi Ka. 127 Come falcon IM. che sta assai Z. in sull' ali DFGI^!. 128 loyro veder P. o 1' uccello I. 129 al falcone ove (sic) L. 130 lascio E. ove L. si muove ABCEFHIKLMOPQZ 120 occhi has been omitted at first in F. 128 Q. 2 has something like loyro. 130 For lascio cf. note sup. 1. 34. 126 INFERNO Per cento rote, e da lungi si pone Dal suo maestro, disdegnoso e fello : 131 Cosl ne pose al fondo Gerione A pik a pie della stagliata rocca, E, discarcate le nostre persone, 135 Si dileguo, come da corda cocca. 131 e om. L. di lunghi BDQ. da lunga I. e da rasso (?) si pone G. 132 Del suo G. Al suo Z. si disd. D. e disd. EC. 133 nel fondo K. 134 A piedi a piei K. A piede B. istangulata boccha D. scagliata ACEZC. stagnata M. scogliata O. della scarata (sic) Q. 135 discaricate E. vostre (?) L. 136 Si de legno (sic) E. Si desleguo K. deslego A. diliquo O. 133 E has the Venetian iormpuse. CANTO DECIMOTTAVO Loco & in inferno, detto Malebolge, Tutto di pietra di color ferrigno, Come la cerchia che d' intorno il volge. 3 Nel dritto mezzo del campo maligno Vaneggia un pozzo assai largo e profondo, Di cui in suo loco dicero 1' ordigno. 6 Quel cinghio che rimane adunquc e tondo, Tra il pozzo e il pie dell' alta ripa dura, Ed ha distinto in dieci valli il fondo. 9 Quale, dove per guardia delle mura, Piu e piu fossi cingon li castelli, La parte dov' ei son rende figura: 12 Tale imagine quivi facean quelli : 1 Loco et inferno K. e d' inferno I. e inferno E. 2 pietre D. e di color ABCDEGHIMOPZCZ>a. di dolor K. 3 la ripa A*F. lo cerchio I. intorno volge Z. 4 mezzo campo si maligno P. 5 si largo P. 6 Di cui suo ABCDEFGHIKMOPZ^CA Di che suo Q. Di qui suo a. suo modo F. dicera AEGHKMQa. contera VLB. dicerai lor digno (sic) C. 7 Quel cerchio BCEIMPQ. Quellangio (?) G. et tondo G. 8 altra BE^Q. 9 Era dist. ADEFGHMOPQa. Et e dist. I. E han K. in d. parti I. in d. parte EI [with valle written above]. al fondo Q. 10 Et dove M. E dove a. 12 E la parte Z>. dove son ABCDEFGHIKMPQZa. dove '1 sol B. rende sicura ACDKPZA Z CD. renden sicura B. rendon sicura FGIM. rendon figura Q. 13 facean quivi FK. qui facevan G. imagini facea quivi C. N.B. Canto xviii. is altogether omitted in L. 1 Perhaps K's reading is an oral blunder for e d' inferno, just as in xix. 1 10 E dalle appears as Et alle in CFK and as E alle in L. 12 O has a marginal note (apparently \ ma manu\ al. sicura. 128 INFERNO E come a tai fortezze dai lor sogli Alia ripa di fuor son ponticelli, 15 Cosl da imo della roccia scogli Movien, che recidean gli argini e fossi Infino al pozzo, che i tronca e raccogli. 18 In questo loco, dalla schiena scossi Di Gerion, trovammoci : e il Poeta Tenne a sinistra, ed io retro mi mossi. 21 Alia man destra vidi nuova pieta ; Nuovi tormenti e nuovi frustatori, Di che la prima bolgia era repleta. 24 Nel fondo erano ignudi i peccatori : Dal mezzo in qua ci venian verso il volto, Di la con noi, ma con passi maggiori : 27 Come i Roman, per 1' esercito molto, L' anno del Giubbileo, su per lo ponte Hanno a passar la gente modo colto : 30 Che dall' un lato tutti hanno la fronte Verso il castello, e vanno a santo Pietro ; Dall' altra sponda vanno verso il monte. 33 14 a tal fortezza D. a tal fortezze EIM. a tai fortezza GQ. dalor BEFGIMP. dei a. 15 Dalla ripa I. di lor C. 16 E cosi uno della F. da un C. da uno GB. delle rocce D. dalla roccia EM. daun di quella P. 17 Muove I. Movea FK. si che D. ricidea F. redian (sic) K. e i fossi BUOZ. effossi C. 18 che tronca ACDEFGHOa. che tronchi P. chen tronque e accogli M. 19 E in Q. della ABCDGIMOQZ^ 2 Ca. 20 ci trovammo C. trovandoci P. e 0m. KMP. 21 dietro ABCDEFHOQZZ?a. de dietro P. drieto GIM. 22 man dritta'E. 23 Nuovo tormento COQ. Nuovi frustati FM. 24 Di chi E. ripieta K. 26 Da mezzo A%C. ne venian M. ver lo E. 27 Di la com' io Z. passy maory (sic) K. 30 al passar A^CD. le gente CDQ. la gente a passar F. il modo D. mondo G. accolto A. tolto I. 31 un canto D. hanno tutti Q. 32 Ver lo c. D. Verso castello F. 33 altra parte GM. ver lo monte D. al monte K. 14 But prob. da=dai,e.g. F has ta fortezze also. 16 'Cossi da ymo de la roc,ia schogly ' may be taken as a sample of A's orthography. As regards the variant uno for imo, in many MSS. these would be simply undistinguishable (fs not being dotted), e.g. in B. CANTO XVIII. 129 Di qua, di la, su per lo sasso tetro Vidi Demon cornuti con gran ferze, Che li battean crudelmente di retro. 36 Ahi come facean lor levar le berze Alle prime percosse ! gia nessuno Le seconde aspettava ne le terze. 39 Mentr' io andava, gli occhi miei in uno Furo scontrati ; ed io si tosto dissi : Di gia veder costui non son digiuno. 42 Percio a figurarlo i piedi affissi : E il dolce Duca meco si ristette, Ed assentl ch' alquanto indietro gissi : 45 E quel frustato celar si credette Bassando il viso, ma poco gli valse: Ch' io dissi : O tu che 1' occhio a terra gette, 4 s Se le fazion che porti non son false, Venedico se' tu Caccianimico ; Ma che ti mena a si pungenti salse ? 51 Ed egli a me : Mai volentier lo dico ; Ma sforzami la tua chiara favella, Che mi fa sovvenir del mondo antico. 54 Io fui colui, che la Ghisola bella Condussi a far la voglia del Marchese, 34 Di Ik di qua BD. fosso retro E. 35 dimoni BEQ. dimon H. correnti I. 36 Che peccator battean forte retro D. crudel mentre C (as frequently in this MS.). dietro BEOZa. 37 Deh come D. comacievan (sic) K. faceallor (sic) G. 38 Che alle M. e gia AI. niuno K. 39 aspettavan DHP. 40 e gli occhi a*. 41 Funno I. ond' io FM. 42 Di ved. gia D>. Gia di veder BGIKZ^a. 43 Perch' io AEGHOPQ^. Pero a Z. E per raffigurarlo D. affigurarlo Ea. affigurando M. gli occhi affissi FZJ5. 44 Duca mio AFIMOPQa. si si rist. FIOPQ. si sen rist. M. 45 assentio ABDGHM. consenti Q. indietro alquanto CK. io gissi ABHOPQ. 47 Abbassando F. 48 Diss' io O tu D. dissi, tu ABCEGHIKM OPQZj^a. in terra 1' occhio D. ch' a terra 1' occhio K. 1' occhi la. 49 la fazion E. 51 chi ti PQZ. 52 tel dico YD. 53 forzami AEK. inforzami P. 54 del tempo ant. CD. 55 Io son colui GZ. D. I. Q ISO INFERNO Come che suoni la sconcia novella. 57 E non pur io qui piango Bolognese : Anzi n' e questo loco tanto pieno, Che tante lingue non son ora apprese 60 A dicer sipa tra Savena e Reno : E se di ci6 vuoi fede o testimonio, Recati a mente il nostro avaro seno. 63 Cosl parlando il percosse un demonic Delia sua scuriada, e disse : Via, Ruffian, qui non son femmine da conio. 66 Io mi raggiunsi con la scorta mia : Poscia con pochi passi divenimmo, La dove un scoglio della ripa uscia, 69 Assai leggieramente quel salimmo, E volti a destra su per la sua scheggia, Da quelle cerchie eterne ci partimmo. 72 Quando noi fummo la, dov' ei vaneggia Di sotto, per dar passo agli sferzati, Lo Duca disse: Attienti, e fa che feggia 75 Lo viso in te di questi altri mal nati, A' quali ancor non vedesti la faccia, Perocche son con noi insieme andati. 78 Dal vecchio ponte guardavam la traccia, 57 favella C. 59 Anzi gli & A. Anco I. tutto pieno A$). 60 accese F. 61 Ravena K. Savona Q. e '1 Reno ADEFGHKM OPQZA 2 ffCZ). 62 e testim. Z. 63 il mio av. P. 65 Con la I Ma. scorigiada GQ 2 . schuriata DI. scoriata P. e om. K. 66 qua G. non ha CDZ. di conio K. 68 Poi con D. ne venimmo D. 69 Dove un (oruno) BCDIMZa. La ove AGHQZ). uscio O. 70 Ed assai D. 71 E om. F. sopra (or sovra) la la. scieza (sic) E. 72 Di quelle D. c' impartimmo C. se partimmo K. 73 dove van. DFIM. la ov' el GD. 74 dar luogo F. ferzati K. 75 attendi I. seggia D. veggia C. 77 Di cui ancor F. non vedesti anc. EGHP. 78 sem colloroZ. insieme con noi I. 79 Del vecchio ABCDEFGHKZ^CX'a. guardava Z. 72 On K's reading cf. note on xvii. 34. 79 Witte's 4to text has /Where. Cf. xiv. 122. CANTO XVIII. 131 Che venia verso noi dall' altra banda, E che la ferza similmente scaccia. 81 II buon Maestro, senza mia domanda, Mi disse : Guarda quel grande che viene, E, per dolor, noh par lagrima spanda: 8 4 Quanto aspetto reale ancor ritiene ! Quelli e Jason, che per core e per senno Li Colchi del monton privati fene. 87 Egli pass6 per 1' isola di Lenno, Poi che le ardite femmine spietate Tutti li maschi loro a morte dienno. 90 Ivi con segni e con parole ornate Isifile inganno, la giovinetta, Che prima avea tutte 1' altre ingannate. 93 Lasciolla quivi gravida e soletta : Tal colpa a tal martiro lui condanna ; Ed anco di Medea si fa vendetta. 96 Con lui sen va chi da tal parte inganna: E questo basti della prima valle 80 venivan ver C. venian K.SC. venia inverse Z. 81 E cui AC. caccia FIM. schiaccia HOQ^. 82 Lo mio AGHPQ. 83 vedi quel D. exguarda (?)Q. 84 Che per M. lagrime AEGHIKOPZX pare lagrime I. 85 regal GA. 86 Quello e Janson F. Gianson DG. Egli e Yason Z. che per forza M. 87 del molton A. 88 Esso passo IZ. Ello passo KOP. 89 e spietate I. 90 lor maschi alia D. 91 Quivi AEGH. Quivi con senno DM. cenni I. senno CZ. G has 'senno' altered to 'segni' i nda manu, 93 Che 'n prima FM. tutte 1' altre avea BF. V altre avea tutte AHPZ. 1' altre tutte avea EQ. avea 1' altre tutte K. 94 e quivi Q. e om. EFHOPQZa(P). 95 colpa om. E (clerical error), a om. M. 96 anche EFKOPQa. 97 Colui sen Z. Collui G. Colui I. di tal DG. 80 venianverso may obviously stand for venian verso, or venia inverso. The reading intended would be indeterminate. 82 Many MSS. (like Witte's A^BCD) have El, but this may represent either // (as regularly in F and I and often in A) or E '/, and the reading is therefore indeter- minate. 94 The writing of * a' is often nearly illegible hereabouts, the ink having scaled off. 92 132 INFERNO Sapere, e di color che in se assanna. 99 Gia eravam la 've lo stretto calle Con 1' argine secondo s' incrocicchia, E fa di quello ad un altro arco spalle. 102 Quindi sentimmo gente che si nicchia Nell' altra bolgia, e che col muso isbufifa, E se medesma con le palme picchia. 105 Le ripe eran grommate d' una mufifa Per 1' alito di giu che vi si appasta, Che con gli occhi e col naso facea zuffa. 108 Lo fondo e cupo si, che non ci basta Loco a veder senza montare al dosso Dell' arco, ove lo scoglio piu soprasta, 1 1 1 Quivi venimmo, e quindi giu nel fosso Vidi gente attuffata in uno stereo, Che dagli uman privati parea mosso: n 4 E mentre ch' io la giu con 1' occhio cerco, Vidi un col capo si di merda lordo, 99 Da sapere F. in om. M. 100 erav. dove BCFGIMPZZ). Ik dove '1 K. secreto calle E. 101 arg. sinislro G. che si circhia K. si crocichia M. 102 E fa quello F. di quella ad un arco spalle D. arco om. a. 103 Quivi DGMO. trovammo M. son- nichia AE. s' innichia DKZC. s' annicchia GD. 104 e om. ADEIQa. col viso EK. sbuffa AI. scuffa (or isc.) BCEFGHKMOQZ^ 2J #CZ>a. sciuffa P. 105 medesmo ABEHMPQa. medesme C. medesmi O. colle man si F. 107 che ivi FP. appiasta Z. 109 e scuro C. che '1 non M. cupo tanto che non D. ti basta Q. 110 L' occhio veder A. L' occhio a veder I. Loco veder Q. montarli K. adosso BDK. Ill Ball' arco M. dove el schoio K. 112 E quindi ne venimmo giu D. Quindi ven. IP. quivi giu Q. 113 La era gente D. 114 privadi ABCDEFGHKOZ^Ca. 115 mentr' io laggiu A. che laggiu IMZ>. che coll' occhio laggiu F. con gli occhi I. cogli occhi laggiu A Y 100 E's reading seems to be a reminiscence of x. i. 102 C* has un altr 3 archo le written over an erasure. 103 Trovammo and sentimmo are again interchanged by MSS. in xvi. 104. 107 After 107 F repeats 1. 105 again in error 'E se med. &c.' 110 On the interchange of L' occhio and Loco see note on Par. x. 4. CANTO XVIII. 133 Che non parea s' era laico o cherco. 117 Quei mi sgrido : Perche se' tu si ingordo Di riguardar piu me, che gli altri brutti ? Ed io a lui : Perche, se ben ricordo, 120 Gia t' ho veduto coi capelli asciutti, E sei Alessio Interminei da Lucca : Pero t' adocchio piu che gli altri tutti. 123 Ed egli allor, battendosi la zucca : Quaggiu m' hanno sommerso le lusinghe, Ond' io non ebbi mai la lingua stucca. 126 Appresso ci6 Io Duca : Fa che pinghe, Mi disse, il viso un poco piu avante, SI che la faccia ben con gli occhi attinghe 129 Di quella sozza e scapigliata fante, Che la si graffia con 1' unghie merdose, Ed or s' accoscia, ed ora e in piede stante. 132 Taide e la puttana, che rispose Al drudo suo, quando disse : Ho io grazie Grandi appo te ? Anzi meravigliose. 135 E quinci sierf le nostre viste sazie. 117 Ch' io non vedea F. pareva ladico n chercho C. parea laico cherco M. ne ladico ne chercho a. 118 grido ADl. si gordo ^a. 119 me piu D. ch' altri D. 120 perch' io C. mi ricordo D. 121 con cap. DEM. 123 Anterminei J3. t' ammiro F. cha i altri K. 124 egli a me M. 125 sommerse GM. 126 ebbi ancor AEGH. sciucca P. 127 Di presso G. E '1 duca disse a me Or fa M. 128 un poco il viso OJ5. El viso tuo un poco M. in avante A v 129 meglio coll' occhio D. coll' occhio BFMZ. Si con gli occhi ben la f. A 2 C. 130 e om. BCIKZa. 131 e colle unghie D. 1' unghia B. 132 Ed ora ace. D. accoscia ora A. e ora in pie CD. e ora in piedi IP. e ora in piede E^KMa. e in piedi D. 133 Taida BC. 134 quando e' disse F. 135 Grande o poche A. apud te I. aput te F. e si marav. D. 136 sia A. 130 K has scapiata which is due to the same Venetian pronunciation as the form schoio in 1. in, and isgaiarda in xxi. 27. Compare the con- verse change of taia into taglia (of which the pronunciation would be similar) in xxi. 60. CANTO DECIMONONO O Simon mago, o miseri .seguaci, Che le cose di Dio, che di bontate Deono essere spose, voi rapaci 3 Per oro e per argento, adulterate; Or convien che per voi suoni la tromba, Perocche nella terza bolgia state. 6 Gia eravamo alia seguente tomba Montati, dello scoglio in quella parte, Che appunto sopra mezzo il fosso piomba. 9 O somma Sapienza, quanta e 1' arte Che mostri in cielo, in terra e nel mal mondo, E quanto giusto tua virtu comparte ! 12 lo vidi per le coste e per lo fondo Piena la pietra livida di fori D' un largo tutti, e ciascuno era tondo. 15 Non mi parean meno ampi ne maggiori, Che quei che son nel mio bel San Giovanni 1 mago miseri B. magro miseri L. 2 Challe cose a. e di bont. L. 3 Den essere D. Debbono CIL. Deggiono K. Denno OM. e voi ABDEFGIMQZ^C/V 4 avolterate BCEFGHKLMOPQa. avulterate Z. avoltrate A. 5 tonba (cler. error) K. 6 bolgia terza D. 7 surgente D. 8 dallo CDZ. 9 sopr' al mezzo ABDFGILOQZ. il om. E. 11 e in terra BGLQZ. terra nel La r 12 Quanta giustizia tua CPZ?. Quanta giusta H. E quanta giusta EQ. E quanto giusta BDIKa. giusto tuo virtu AF. e giusto L. 15 ciascuna era tonda D. 16 parea AL. ampli C*EK. ampia (sic) M. alti Z. piu ampi ne ne (sic) minori F. 17 che sono nel mio san G. DL. Joanni I. 17 bel appears as a late marginal addition in D. CANTO XIX. 135 Fatti per loco de' battezzatori ; 18 L' un delli quali, ancor non e molt' anni, Rupp' io per un che dentro vi annegava : E questo fia suggel ch' ogni uomo sganni. 21 Fuor della bocca a ciascun soperchiava D' un peccator li piedi, e delle gambe Infino al grosso, e 1' altro dentro stava. 24 Le piante erano a tutti accese intrambe ; Per che si forte guizzavan le giunte, Che spezzate averian ritorte e strambe. 27 Qual suole il fiammeggiar delle cose unte Moversi pur su per 1' estrema buccia; Tal era li da' calcagni alle punte. 30 Chi e colui, Maestro, che si cruccia, Guizzando piu che gli altri suoi consorti, Diss' io, e cui piu rozza fiamma succia ? 33 Ed egli a me: Se tu vuoi ch' io ti porti Laggiu per quella ripa che piu giace, Da lui saprai di se e de' suoi torti. 36 19 L' un de' quali FP. di quali O. L' uno de quel M. non e ancor ABDE/^. non sono I. non a molti K. mille anni (sic) M. 21 questa E. sie sugg. DQ. sia ABCEFGHIKLMOZ^C/?a. 22 a om. Z. 23 dei piedi D. e le gambe A. 24 Fin' al AEH. Sino allo G. 25 Le gambe QD. accese a tutti BMPA 2 Zt. ac- cese tutte CD. tutte accese G. accese er. a tutti A v tutte er. ace. D. incese IQ. Le piante a ciascuno era acceso F. 27 Che rotte F. spezzato GLM. averia M. averienli torte A. le torte strambe CG. li torte HOa. le trontte (?) K. 28 al fiam. G. delle consunte (with n partly erased) G. 29 Volgersi pur C. su om. E. su pur G. la istreta L. 30 Tal era delli E. dal calcagno D. dalle calcagne A. dalle calcagna F. dal calcagni P. e le punte D. le punte K. 32 caialttry (sic) K. 33 Diss' io che L. e che K. rossa BCDGIZ5. roggia FK. ruc^a M. E has apparently soga altered to roqa. 34 tu om. E. che ti A. 36 Laggiu saprai di lui D. Da lui di se sap. B. Da lui saprai disse C. 24 A regular change in A. 33 B has rossa, but as ss stand con- stantly for zz in this MS. this is inconclusive, e.g. guissavan above in 1. 26. 34 But che ma.y=chei=che i\ as final i is omitted passim in A. 36 On C's reading comp. note on xvi. 67. 136 INFERNO Ed io : Tanto m' e bel, quanto a te piace : Tu sei signore, e sai ch' io non mi parto Dal tuo volere, e sai quel che si tace. 39 Allor venimmo sull' argine quarto; Volgemmo, e discendemmo a mano stanca Laggiu nel fondo foracchiato ed arto. 42 E il buon Maestro ancor della sua anca Non mi dipose, si mi giunse al rotto Di quei che si piangeva con la zanca. 45 O qual che se', che '1 di su tien di sotto, Anima trista, come pal commessa, Comincia' io a dir, se puoi, fa motto. 48 Io stava come il frate che confessa Lo perfido assassin, che poi ch' e fitto, Richiama lui, per che la morte cessa: 51 Ed ei grido : Sei tu gia cost! ritto, Sei tu gia cost! ritto, Bonifazio ? Di parecchi anni mi mentl Io scritto. 54 Se' tu si tosto di quell' aver sazio, 37 tanto om. E (clerical error corrected 2^ manu). a om. QDM^j. 39 cio che si tace F. 40 Venimmo allora EGHP. in su '1 ABC*DGHILOPQZ#>a. 42 e archo L. 43 Lo buon Maes. ABCDEFGHIMOPQZCZ?a. dalla sua B. 44 dispose BQEFIKLMZ5C7?. sinmi (sic) G. e si mi K. che giugnem- mo F. a rotto G. e si m' aggiunse alorto (sic) L. 45 Di quel ABCDEFGHIKLMOPQZa. si spingava G (see 1. 120). spiangava P. 46 che sia F. chi se' OP. che di su Z. che tieni '1 su di sotto A<. che '1 viso teni disotto Q*. 48 io om. E. 50 assessin (or -ino) CDOPZa. essessino poi F. 51 Lo richiama F. il cessa Z. 54 mentio D. ti menti Z. mente I. mi mena L. 37 C 2 has a ti, 44 In C the s of dispose has been erased. 46 Q has been obviously altered, and apparently from the ordinary reading, except that Qj had e before disotto. 52 3 1. 53 is omitted in F, the error having evidently arisen from the repeated costl ritto, since 1. 52 was written 'ritto o Bonifacio? the words italicised being afterwards erased. Then comes 1. 54, then 53 is inserted and 54 again repeated. CANTO XIX. 137 Per lo qual non temesti torre a inganno La bella Donna, e poi di fame strazio ? 57 Tal mi fee' io, quai son color che stanno, Per non intender cio ch' e lor risposto, Quasi scornati, e risponder non sanno. 60 Allor Virgilio disse : Digli tosto, Non son colui, non son colui che credi : Ed io risposi come a me fu imposto. 63 Per che lo spirto tutto storse i piedi : Poi sospirando, e con voce di pianto, Mi disse : Dunque che a me richiedi ? 66 Se di saper chi io sia ti cal cotanto, Che tu abbi pero la ripa corsa, Sappi ch' io fui vestito del gran manto: 69 E veramente fui figliuol dell' orsa, Cupido si, per avanzar gli orsatti, Che su 1' avere, e qui me misi in borsa. 72 Di sotto al capo mio son gli altri tratti Che precedetter me simoneggiando, Per le fessure della pietra piatti. 75 Laggiu caschero io altresi, quando Verra colui ch' io credea che tu fossi, Allor ch' io feci il subito domando. 78 56 Per la qual F. tuore A. 57 e om. C. di poi fame IB. poi da fame C. 58 qual color E. quali color I. qual son KLM. 59 quel ch' e lor F. 60 e om. C. non (bis) F. 61 disse Virgilio A. 62 "2 nd son om, D. son colui ne colui a. 63 a om. MQ (Q has come fo with me inserted later). 64 tutti storse CIKOPQZ5. 65 Poi mi rispose con F. e om. DIK. e non con E. di canto K. 66 E disse F. adunque El. 67 io fui M. ti cale tanto CK. t' incale E. 70 io fu figl. A. 71 avacciar G. 72 Che su il tesoro F. e quei C. e om. L. 73 mi son CE. 74 procedetter L. 75 la fessura Y17.BD. 76 io om. F. 77 qual io credea che fossi M. cui io Q. tu om. Q. 78 che feci EG. 67 Fui in M is probably a change made to suit 1. ?o. The terzina 7072 is omitted in H. 138 INFERNO Ma piu e il tempo gia che i pie mi cossi, E ch'.io son stato cosl sottosopra, Ch' ei non stara piantato coi pie rossi : 81 Che dopo lui verra, di piu laid' opra, Di ver ponente un pastor senza legge, Tal che convien che lui e me ricopra. 84 Nuovo lason sara, di cui si legge Ne' Maccabei : e come a quel fu molle Suo re, cosl fia a lui chi Francia regge. 87 lo non so s' io mi fui qui troppo folle, Ch' io pur risposi lui a questo metro : Deh or mi di', quanto tesoro voile 90 Nostro Signore in prima da san Pietro, Che gli ponesse le chiavi in balla ? Certo non chiese se non : Viemmi retro. 93 Ne Pier ne gli altri chiesero a Mattia Oro od argento, quando fu sortito Al loco che perde 1' anima ria. 96 Pero ti sta, che tu se' ben punito ; 79 tempo ch' io i pie CF. che pie i mi P. che '1 pie G. gia om. KL. 81 con pie M. 82 Che di po' BC^?)^. di om. M. 84 converra L. che me e lui BCFKP^Ca. 85 Gianson DG^. Nonvi sarra di cui (sic) L. 86 come quel DGjLP. e om. C*. a quel chi fu C*. 87 e cosi fia lui G. Cosi fia lui ancor chi C*. cosi lui chi L. a lui cosi fie Q. fia lui ABDHKOPZa. che Francia EP. chi in Francia K. 89 a lui OQ. per questo D 2 I. 90 Die or A. 91 Signore prima CI. Sign, pria di Santo M. in pria AFGHKOP^Z). da santo CFGHKOPCZ>. 92 Che po- nesse BDFIPZ. Che imponesse C. Che i pon. LOQ^4 X . in sua balia ABDEFGHIKLMOPQZ/^a. 93 Certe G. non disse D. non gli chiese EH. non mi chiese L. non i chiese B. dietro BEFKLZa. detro M. 94 Pietro E. ne altri C. tolsero a M. ABCEFHIKLOPQa. tolseno M. 95 ne argento C*G. 96 Al 1. per quello chi perde C*. Nel loco M. In luogo Z. Nel luogo D. 97 Ora ti sta P. sta ben che sey ponito A. 82 C* reads dopo, but the first o in dopo is 2 n(la manu over an erasure. 86 7 have been much erased and re-written in C. 89 a questo, altered by later hand to per q. in D. 92 But che may be intended for che i as noted elsewhere. 93 C* has retro. CANTO XIX. E guarda ben la mal tolta moneta, Ch' esser ti fece contra Carlo ardito. ' 99 E se non fosse, che ancor lo mi vieta La riverenza delle somm'e chiavi, Che tu tenesti nella vita lieta, 102 I' userei parole ancor piu gravi ; Che la vostra avarizia il mondo attrista, Calcando i buoni e sollevando i pravi. 105 Di voi pastor s' accorse il Vangelista, Quando colei, che siede sopra 1' acque, Puttaneggiar co' regi a lui fu vista : 108 Quella che con le sette teste nacque, E dalle dieci corna ebbe argomento, Fin che virtute al suo marito piacque. m Fatto v' avete Dio d' oro e d' argento : E che altro e da voi all' idolatre, , Se non ch' egli uno, e voi n' orate cento? 114 Ahi, Constantin, di quanto mal fu matre, Non la tua conversion, ma quella dote Che da te prese il primo ricco patre ! 117 E mentre io gli cantava cotai note, 98 Riguarda ben M. 99 contro a DFIZa. contra a L. 100 ancor che CG. la mi L. invieta K. me '1 divieta P. 101 sommi IP. 103 ancor palrole (sic) E. 104 nostra G. 105 Cha- lando D (corrected later to Chalchando). su levando CFGHMA. sul- levando OZ. sublevando E. 107 acqua G. 108 Putanicar E. allor G. fe (sic) vista M. 109 E quella L. teste sette D. 110 Et (or E) alle CFKL. ill virtu CL. 112 n' avete Q(?)Z>. Idio ADFGHKMOCZ). 113 che e altro da F^. e om. L. altr' e che voi M. 114 che ( = ch' ei ?) uno MO. e honorate cento D. e voi onorate (. 120 ma indarno I. 121 che lasciar il lago P. 122 La spola il fuso E. feronsi B. feceronsi E. fecionsi L. indovine BCDELZC fecersi divine GOP. feceser (? fecersi) divine O. 123 E fer malie F. 124 ormai L. tien nel C. tie nel K. i confini E. le confine I. 126 Sobilia ABCEHLZ. leSpie (sic) E. 127 E gib. ABCDEFGHIKLMOPQZ^ 2 a. 128 Ben ti dee ABDGKLMPQZtfZ?. nacque (obvious error) H. 129 fronda C (the word partially erased) M. 130 andamo D. IO 2 CANTO VENTESIMOPRIMO Cosl, di ponte in ponte, altro parlando Che la mia commedia cantar non cura, Venimmo, e tenevamo il colmo, quando 3 Ristemmo per veder 1' altra fessura Di Malebolge, e gli altri pianti vani ; E vidila mirabilmente oscura. 6 Quale nell 1 Arzana de' Viniziani Bolle 1' inverno la tenace pece A rimpalmar li lor legni non sani, 9 Che navicar non ponno, e in quella vece Chi fa suo legno nuovo, e chi ristoppa Le coste a quel che piu viaggi fece ; 1 2 Chi ribatte da proda, e chi da poppa ; Altri fa remi, ed altri volge sarte; Chi terzeruolo ed artimon rintoppa: 15 Tal, non per foco, ma per divina arte Bollia laggiuso una pegola spessa 2 contar G. 3 al colmo C*. il colpo H. 4 Restamo AFHILOPQZa. 5 aid OS. 6 admirabilmente F. 7 nella Terzana BD. Erzana a. Arsena ACEIKM. Arsenal -<4i(?). Argiana O. 8 d' inverno L. L' inverno bolle C. nel verno Q. 9 A ripalmar CMPQ. Per rimp. I. A riparar K. Per ripalmar MA(marg.}D. legni lor ABCDEFGHIKLMPQ^a. mal sani DZ. 10 e om. ACDEFGHLMOPQZa 2 . H Qual fa F. e qual rist. F. e om. L. 12 viaggio DFIZ>. 13 rib. da plura (?) F. 14 fan or fa FGILMQa. rivolge B. e chi rinvolge I. volgon L. 15 E chi terzuolo F. terzeruoli MQ. e chi '1 timo D. e chi timon C. o artimon A. antimon I. autimo (?) L. rimpoppa C. 17 Bolle G. laggiu ACDILM/}. CANTO XXI. 149 Che inviscava la ripa da ogni parte. 18 lo vedea lei, ma non vedeva in essa Ma' che le bolle che il bollor levava, E gonfiar tutta, e riseder compressa. 21 Mentr' io laggiu fisamente mirava, Lo Duca mio, dicendo : Guarda, guarda, Mi trasse a se del loco dov' io stava. 24 Allor mi volsi come 1' uom cui tarda Di veder quel che gli convien fuggire, E cui paura subita sgagliarda, 27 Che, per veder, non indugia il partire : E vidi dietro a noi un diavol nero Correndo su per lo scoglio venire. : 30 Ahi quanto egli era neli' aspetto fiero ! E quanto mi parea nell' atto acerbo, Con 1' ale aperte, e sopra il pie leggiero ! 33 L' omero suo, ch' era acuto e superbo, Carcava un peccator con ambo 1' anche, E quei tenea de' pie ghermito il nerbo. 36 Del nostro ponte, disse, o Malebranche, Ecco un degli anzian di santa Zita: Mettetel sotto, ch' io torno per anche 39 A quella terra ch' i' n' ho ben fornita : 18 rivesciava D. inveschiava C*. d' ogni Z. 19 Io vedea ben lei ma non in essa I. 20 Ma quelle bolgie C. Mai con le bolle K. Ma quele bole che '1 dolor L. 21 Gonfiare E. Ri- gonfiar F. riseder tutta I. 22 Mentre laggiu \BD. 24 dove stava C. 25 Io mi rivolsi AEGHP. Allora D. che tarda DEF. chiC*. 27EconF. subito DGIMZ?. ighagliarda D. isgaiarda K. 28 Che per fuggir I. nonne D. none F. 29 dimon nero A. 32 negli atti D. 33 ali ABEGHKMOPZ^. alie DFQ. e om. ABCHKPa. i pie CFKQ. sopra pie A}LHOA 2 BCDa.. Chonai aperte sopra a pie L. 34 che om. L. 36 Ed egli D. Et ei B. del pie IMQZ?. da pie tenea F. gremito B. 37 Dal nostro I. disse Mai. IKL. 39 Mette '1 di sotto BL. ch 1 io vo per E. 40 che n' e ben BCDEFGHIKLMQCZ?. che v' e ben A. chio (-=ch' i' ho) ben OPa. 28 Qy. nonne non e' 1 ? 150 INFERNO Ognun v' e barattier, fuor che Bonturo: Del no, per li denar, vi si fa ita. 4* Laggiu il butto, e per lo scoglio duro Si volse, e mai non fu mastino sciolto Con tanta fretta a seguitar lo furo. 45 Quei s' attuffo, e torno su convolto ; Ma i demon, che del ponte avean coperchio, Gridar : Qui non ha loco il santo volto ; 48 Qui si nuota altrimenti che nel Serchio ; Pero, se tu non vuoi de' nostri graffi, Non far sopra la pegola soperchio. 51 Poi 1' addentar con piii di cento raffi ; Disser : Coperto convien che qui balli, SI che, se puoi, nascosamente accaffi. 54 Non altrimenti i cuochi ai lor vassalli Fanno atuffare in mezzo la caldaia La carne cogli uncin, perche non galli. 57 Lo buon Maestro : Acciocche non si paia Che tu ci sii, mi disse, giu t' acquatta Dopo uno scheggio, che alcun schermo t' haia ; 60 E per nulla offension che mi sia fatta, 41 Ciascuno DCZX Ogn' uom ABEHOPQZ^a. 42 Del non A. si fan M. 43 e om. L. 44 da sciolto M. 45 suo furo F. lor furo P. 46 Quel s' attuffo AKLQa. E' s' attuffo D. El s' attuffo EGH P. col volto AC DEGIOM Q*Z>a. con volto PQ^?). 47 E' dimon F. dimonii A. che havia del ponte coperchio E. 48 Disson I. Disser O. Gridaron L. 49 si nuocha E. nuta M. 50 E pero E. se tu non vuol EQ. desti nostri A. raffi ADQ. non vuoi sentir de'Z. 51 Non star Q. 53 qui convien che DGK. con- vien che tu ABCEHIMQ. 55 quochi I. a lor ACDEFGHIMOPa. i lor L. 56 tuffar Z. nel mezzo Fl. alia cald. F. 57 con 1' uncino D. col uncin G. ancini AE. ancim C. con le man K. 59 sia DFGIKL,4CZ?a. sie OPQ. aguata E. 60 Dopo un schermo ch' uno scoglio taglia K. uno segio C. uno scoglio I. schermo aia F. 61 E om. F. ch' a me FI. 60 The reading taglia in K is probably for /' aglia : this dialectic, and I believe Venetian, form for haia being adopted in spite of rhyme. See note on xviii. 130. I has again altered scheggion to scoglio in 1. 89. CANTO XXI. 151 Non temer tu, ch' io ho le cose conte, Perche altra volta fui a tal baratta. 63 Poscia passo di la dal co del ponte, E com' ei giunse in su la ripa sesta, Mestier gli fu d' aver sicura fronte. 66 Con quel furor e con quella tempesta Ch' escono i cani addosso al poverello, Che di subito chiede ove s' arresta ; 69 Usciron quei di sotto il ponticello, E volser contra lui tutti i roncigli ; Ma ei grido: Nessun di voi sia fello. 72 Innanzi che 1' uncin vostro mi pigli, Traggasi avanti alcun di voi che m' oda, E poi d' arroncigliarmi si consigli. 75 Tutti gridaron : Vada Malacoda ; Perche un si mosse, e gli altri stetter fermi ; E venne a lui dicendo : Che gli approda ? 78 Credi tu, Malacoda, qui vedermi Esser venuto, disse il mio Maestro, Sicuro gia da tutti vostri schermi, 81 Senza voler divino e fato destro ? 62 chio alle cose D. 63 Che un altra DKL. Per altra C. Che altra O. E altra BFI. fu' io Q. a cotal K. 64 da co CD. dall altro ponte Q. 65 E come giunse CIL. Ma come giunse M. E quando ei Z. in om. K. 68 Che schovan (sic) K. i om. M. in dosso HPQa. il (or el) pov. DE. 69 Di sub. che ch. D. ovvero I. 70 di sotto al ABCDEFHIKLMOPQZ^a. 71 puoser AO. volse C. volson M. volserer B. porser EGHP. contro a FGI. in verlui L. lioncigli K. 72 niun QC. 73 m' appigli K. 74 Tr. innanzi GD. 1' un di voi BCEFGKLOPQZ^ 2 ^Ca, un di voi A. che '1 m' odi K. 75 di roncigliarmi AGKLO. a runcigliare mi si D. arronciarmi F. 76 Gridavan (or -vano) DIKLZ.fr vadavi DF. 77 un om. CD. 78 Vegnendo a D. che ti app. Q\A(mar.). or che Q. che si app. F. 79 tu qui Mai. ved. E. 81 tutti i v. A.DYB. de tutti v. K. 82 fatto ADEFa. facto O. e flitto (?) H. o fato Z. afatto vestro L. 71 Compare xxii. 149 for the reading porser. 77 C has flen'/ie si altered to perche 'I si 2'"*" manu. 82 fatto may perhaps stand for 152 INFERNO Lasciane andar, chb nel cielo e voluto Ch' io mostri altrui questo cammin silvestro. 8 4 Allor gli fu 1' orgoglio si caduto, Che si lascio cascar 1' uncino ai piedi, E disse agli altri : Omai non sia feruto. 87 E il Duca mio a me : O tu, che siedi Tra gli scheggion del ponte quatto quatto, Sicuramente omai a me tu riedi. 90 Perch' io mi mossi, ed a lui venni ratto : E i diavoli si fecer tutti avanti, Si ch' io temetti non tenesser patto. 93 E cosi vid' io gia temer li fanti Ch' uscivan patteggiati di Caprona, Veggendo se tra nimici cotanti. 96 Io m' accostai con tutta la persona Lungo il mio Duca, e non torceva gli occhi Dalla sembianza lor ch' era non buona. 99 Ei chinavan gli raffi, e, Vuoi che '1 tocchi, Diceva 1' un con 1' altro, in sul groppone ? 83 Lasciami ACIMZa. Lasciaci L. Pero lascia and. F. 84 a lui C*. sto camin K. 86 cader ABDFGIPZ. 1' uncin cascar C. a piedi ACDEFGHILMOPQZa. 87 Dicendo A^C. 88 a me tu riedi (o/tcuort'X. with 1. 90) L. 89 scheglion A. scogli I. guato guato A. quanto quanto L. 90 a me omai DEFMQZ?. ti riedi BZ?. ten riede MZ. 91 mossi a lui e venni F. 92 Quei diavoli F. I diavoli E. Et gli demon Z. si fer C. 93 Si ch' io credetti F. che tenesser O. che li tenesser AH. ch' ei tenesser KQA 2 BC. che li non ten. E. che non ten. CFGL. che noi ten. B*. 94 E om. ADEHIKLOMQa. Cosi vid' io ben gia temer 1' infanti O. 95 Ch' usciron BDMZ. da Caprona PZ. 96 Vedendosi tra EIMP. tra gli nimici tanti D. trai nim. EPA. Quando si videro tra nim. tanti F. 98 Verso '1 mio MQ. 99 Delia semb. ALa. che n' era bona M. 100 chin, gli uncini F. li graffi IK. e om. LM. or vuo' D. ch' io '1 tocchi FMP. ch' i' toc- chi Q. 101 Dicean E. E dicea 1' uno all' altro D. 1' uno all' altro BCFM. 1' un gli altro P. fato, especially in Venetian MSS. like A and E. Conversely in xxiii. 74 C has fato for fatto. 86 a piedi is not conclusive however. 87 noi (in error for non?) D. 89 See note on 1. 60. CANTO XXI. 153 E rispondean : SI, fa che gliele accocchi. 102 Ma quel demonic che tenea sermone Col Duca mio, si volse tutto presto E disse : Posa, posa, Scarmiglione. 105 Poi disse a noi : Piu oltre andar per questo Iscoglio non si puo, perocche giace Tutto spezzato al fondo 1' arco sesto : 108 E se T andare avanti pur vi piace, Andatevene su per questa grotta ; Presso e un altro scoglio che via face, m ler, piu oltre cinqu' ore che quest' otta, Mille dugento con sessanta sei Anni compie, che qui la via fu rotta. 114 lo mando verso la di questi miei A riguardar s' alcun se ne sciorina: Gite con lor, ch' ei non saranno rei. 117 Tratti avanti, Alichino e Calcabrina, Comincio egli a dire, e tu, Cagnazzo, E Barbariccia guidi la decina. 120 Libicocco vegna oltre, e Draghignazzo, Ciriatto sannuto, e Graffiacane, E Farfarello, e Rubicante il pazzo. 123 102 Egli risp. E. Ei risp. HPA< 2 C. fa si B. che gli ace. LP. 104 Con duca P. 105 Dicendo MO. 106 a me AEHP. a noi om. L. 107 Scoglio ABCDEFHIKMOPZ^?. Scoglio ben non D. Scheggio Q. non si potrk BCIKLM. puo ben P. perocch '1 giace M. 108 1' arco al fondo sesto OZ. il fondo al M. 109 piu oltre piu D. piu ti C. piu vi I. Ill Presso a CDQ. Presso v' e K. 112 cinque ore piu oltre DF. cinque ore oltre piu KL. 113 dugento un con MZ. e sessanta Ej (altered to con}, e sei Q. 114 compier Z. anni conpli K. che piu E. questa via GM. e rotta A. 117 a lor L. 118 Tirati DI. Traiti KP. Traggasi O. davanti BF. 121 Bilichocho A. Lobicoco E. 122 Cu- riato M. 123 Rub. pazzo ABCDEGHIKLMOPQa. 113 For the curious correction inserted by M and Z, and also in refer- ence to the date generally, see the Author's Time-Reftrences in the Divina Commedia, pp. 46, &c. 154 INFERNO Cercate intorno alle boglienti pane ; Costor sien salvi insino all' altro scheggio Che tutto intero va sopra le tane. 126 O me ! Maestro, che e quel che io veggio ? Diss' io : deh ! senza scorta andiamci soli, Se tu sai ir, ch' io per me non la chieggio. 129 Se tu sei si accorto come suoli, Non vedi tu ch' ei digrignan li denti, E colle ciglia ne minaccian duoli ? 132 Ed egli a me : Non vo' che tu paventi : Lasciali digrignar pure a lor senno, Ch' ei fanno cio per li lessi dolenti. 135 Per P argine sinistro volta dienno; Ma prima avea ciascun la lingua stretta Coi denti, verso lor duca per cenno, 138 Ed egli avea del cul fatto trombetta. 124 int. le ABCDEFGHIKMOQZZ>a. 125 Costui sian salvo M. Questi sia salvo Z. infino BCDGHMOPQZ. salvi fino all' AEKa. 126 intorno C. 127 e che e quel K. 128 Deh diss' io B. Dissi, deh G. Deh sanza scorta, diss' io Z. deh om. Q. 129 sa gire E. sai gir M. che per me O. da me M. non le I. 130 sei accorto si com' esser suoli I. 131 che congrignan E. 132 ciglia min. di doli F. 133 spaventi F. che paventi E. 134 pur digrignar E. 135 Che fam cosi C. lesi AEFGIMPOQ^. li elessi K. lasi L. 136 sinistra DL. 138 Con denti Q. il lor DZ. il mio duca M. 139 avean P. fatta B. 135 Little stress can be laid on the distinction between lesi and lessi, especially in Venetian MSS., as the majority of these are. D has lessi with one s erased. Also C constantly interchanges m and ;/, e.g. comsigli gridarotn, &c. CANTO VENTESIMOSECONDO Io vidi gia cavalier mover campo, E cominciare stormo, e far lor mostra, E talvolta partir per loro scampo : 3 Corridor vidi per la terra vostra, O Aretini, e vidi gir gualdane, Ferir torneamenti, e correr giostra, 6 Quando con trombe, e quando con campane, Con tamburi e con cenni di castella, E con cose nostrali e con istrane ; 9 Ne gia con si diversa cennamella Cavalier vidi mover, ne pedoni, Ne nave a segno di terra o di stella. 12 Noi andavam con li dieci dimoni : l caval. gia I. 2 lor stormo BK. lo stormo CGL. la mostra C. lor meschia(?) E. 3 talotta C. 4 Corritor GMOQCZ>. nostra LM. 6 Fedir ACj^DEFGHMOQZCa. EfarZ?. tornea- mento L. muover giostra BZS. 7 e om. Q. 8 E con tamb. F. tamburini D. e cenni F. 9 nostrane E. 10 Ne mai D. gia cosi KL. cieramella A. ciaramella CQD. zaramella K. cialamella M. 11 vid' io D. 12 di torre Q*. di castella Ea. 3 Talotta = talora, Cf. xxi. 112 for otta = ora. Also allotta for allora in v. 53, xxxi. 112. The origin of otta'vs, most uncertain. Diez sug- gests a similar derivation for the disputed word dotia in Inf. xxxi. no. Tramater suggests oretia, Diez the Gothic ttkt=Katpos whence uohta 'spring' and . e ancor me n' accapr. G. cor me ne capr. KM. 32 Uno ascoltar E. aspettare si come F. come li contra E. 33 Che 1' una F. rana om. E. rimane om. L. e 1' altra ABDFIMOPQZa. 30 QJ seems to have had allet bollori. One / has been erased and a line drawn through e. CANTO XXII. 157 E Graffiacan, die gli era piu d' incontra, Gli arroncigli6 le impegolate chiome, E trassel su, che mi parve una lontra. 36 lo sapea gia di tutti e quanti il nome, SI li notai, quando furono eletti, E poi che si chiamaro, attesi come. 39 O Rubicante, fa che tu gli metti Gli unghioni addosso si che tu lo scuoi, Gridavan tutti insieme i maledetti. 47 Ed io : Maestro mio, fa, se tu puoi, Che tu sappi chi e lo sciagurato Venuto a man degli avversari suoi. 45 Lo Duca mio gli s' accosto allato, Domandollo ond' ei fosse, e quei rispose : Io fui del regno di Navarra nato. 4 8 Mia madre a servo d' un signer mi pose, Che m' avea generate d' un ribaldo Distruggitor di se e di sue cose. 51 Poi fui famiglio del buon re Tebaldo ; Quivi mi misi a far baratteria, Di che io rendo ragione in questo caldo. 54 E Ciriatto, a cui di bocca uscia 34 ch' era Ik piu Q. piu ad incontra F. piu rincontra D. di contra ECKLZS. 35 Gli runciglio O. le pegolate AO. 36 ch' e' parve F. come fusse una D. 37 quante P. 38 E si K. 39 che si contavano altresi come V. cosichiamargli att. L. chiamato B. intesi M. 41 L' unghion(e) DFQCZX adosso e fa che P. lo suoi K. 43 Maestro deh fa Q. 45 di avvers. K. 46 i s' accosto C. si gli costo E. da lato M. 47 Domando dond' ei B. E domandol chi fosti F. dove il fosse K. ed ei BEFGHMQZZ?. et el risp. KP. 48 de (altered to dal in D) regno Da. 50 ingenerato I La. 62 Puo K. famiglia HKPQ^?)^. Trebaldo L. 53 barteria F. 54 Ond' io D. Di che rendo CIMQZ?. 65 Ciriaco K. Curiato M. 37 gid has been erased and a space left blank in C. 41 Suoi in K (an obviously Venetian MS.) may be compared with posta, diserno, disicsi ( = discesi) &c., elsewhere. 158 INFERNO D' ogni parte una sanna come a porco, Gli fe' sentir come 1' una sdrucia. 57 Tra male gatte era venuto il sorco ; Ma Barbariccia il chiuse con le braccia, E disse: State in la, mentr' io lo inforco. 60 Ed al Maestro mio volse la faccia : Domanda, disse, ancor se piu desii Saper da lui, prima ch' altri il disfaccia. 63 Lo Duca: Dunque or di' degli altri rii : Conosci tu alcun che sia Latino Sotto la pece ? E quegli : Io mi partii 66 Poco e da un, che fu di la vicino ; Cos! foss' io ancor con lui coperto, Che io non temerei unghia, ne uncino. 69 E Libicocco : Troppo avem sofferto, Disse, e presegli il braccio col ronciglio, SI che, stracciando, ne porto un lacerto. 72 Draghignazzo anco i voile dar di piglio 56 D' ogni lato LA. come porco L. 67 come la sua I. struscia A. scruscia C*. sa drussia K. 58 male branche DGlKLOPZAfaarg.^CD. era arrivato F. 59 chiude I. ciuse D. trale braccia L. 60 E dice I. Dicendo YA 2 C. state la FMPZ>. tratti in la K. io om. Ea. 61 Et el Maestro El. 62 Dimandal CD. Disse, domanda Z. disse om. A. gli disse E. se tu desii D. 63 di lui PQ. il om. CL. 64 Lo duca disse or B. dunque o degli D. di' di altri A. adunque di' degli K. addunqu' addi' degli F*(?). adunque or di' I. dunque ci di' L. or dunque di' Q. 66 S. la pegola D. ed egli F. 67 e om. L. 68 con lui ancor CQAi. 69 Che non KL. curerei C. 70 abbian D. assai v' avem L. 71 prese il suo D. con rone. DKOZ?. con 1' unciglio M. 72 tirando O. porto lacerto CD. portonne I. 73 D. gli volse dar DGKLZ (voile). E Dragh. voile dar CF. anche ABPQCa. anche voile O. volse B. li volse A. anche i volse B. ancor voile I. 58 Cf. 1. 100. 64 F t appears to have read adunque or di\ agreeing (as often) with I. 70 abbian in D of course = abbiamo (as again in xxiii. 23, &c.), final m and n being constantly interchanged in this MS. as also in C and I. CANTO XXII. 159 Giuso alle gambe ; onde il decurio loro Si volse intorno intorno con mal piglio. 75 Quand' elH un poco rappaciati foro, A lui che ancor mirava sua ferita, Domando il Duca mio senza dimoro : 78 Chi fu colui, da cui mala partita Di' che facesti per venire a proda ? Ed ei rispose : Fu frate Gomita, Si Quel di Gallura, vasel d' ogni froda, Ch' ebbe i nimici di suo donno in mano, E fe' si lor, che ciascun se ne loda : 84 Denar si tolse, e lasciolli di piano, Si com' ei dice : e negli altri offizi anche Barattier fu non picciol, ma soprano. 87 Usa con esso donno Michel Zanche Di Logodoro : ed a dir di Sardigna Le lingue lor non si sentono stanche. 90 O me ! vedete 1' altro che digrigna : lo direi anco ; ma io temo ch' ello Non s' apparecchi a grattarmi la tigna. 93 74 Giunse D. Giu dalle S. e '1 decurio C. 75 intorno tutto D. ciglio CG. 76 E quando un C. 77 Colui che D. che riguardava C. 79 Chi e M. da chi CD. 81 E quei YCD. Egli risp. K. fu om. F. 82 Gullura Z. 83 Ed ebbe nem. L. i om. OD. di sua donna M. 84 E fe' lor si BCDFIMa. E fecie si che L. a lor O. 85 egliasolle (sic} C. 86 Si come dice ABCDEFGHK LOPQZa. Come se dice M. dice cogli altri F. 88 Uso AB^L. Usan M. 89 Di luogo doro GZ. Di loco doro F. lagho doro L. Ligodoro O. Lacodoro M. Lodogoro P. al dar di Sard. F. e aver di (sic} D. 90 vedettono stanche D. si om. M. sentiro F. sentiran (?) L. 91 disgrigna C. 92 et io direi C. anche BLPZ. d' anche M. 93 di grattarmi D. 75 Ciglio is probably an emendation to avoid the repetition of piglio. 88 B 2 has Usa. 89 The line in C begins with Dcllo, followed by a partially erased word scarcely legible, but looking somewhat like honor. Was this an oral blunder, the word being unintelligible to the scribe ? The strange jumble in 1. 85 in this MS. looks also like an oral blunder. 160 INFERNO E il gran proposto, volto a Farfarello Che stralunava gli occhi per ferire, Disse : Fatti in costa, malvagio uccello. 96 Se voi volete vedere o udire, Ricominci6 lo spaurato appresso, Toschi o Lombardi, io ne faro venire. 90 Ma stien le male branche un poco in cesso, Si ch' ei non teman delle lor vendette ; Ed io, sedendo in questo loco stesso, 102 Per un ch' io son, ne faro venir sette, Quand' io sufoler6, com' e nostr' uso Di fare allor che fuori alcun si mette. 105 Cagnazzo a cotal motto levo il muso, Crollando il capo, e disse: Odi malizia Ch' egli ha pensata per gittarsi giuso. 108 Ond' ei ch' avea lacciuoli a gran divizia, Rispose : Malizioso son io troppo, Quand' io procure a' miei maggior tristizia. m 94 preposto KOj. e volto Q. 95 fedire CGHOQa. 96 E disse L. E disse fatti in Ik F. fatti costa AC K. 97 o vedere BIKL^,. e udire C. 98 Incominci6 DKD. Rincomincio C. Comincio L. spauroso A. 99 e Lomb. CDM^4. io om. D. vi faro A. ve faro M. i' vi faro K. 100 le om. K. a cesso DFL^,. poco cesso Z. 101 Si ch' io non tema BDFGKLMZ>a. Si che i non temano E. Si che non Q5. de lor A. 102 seggendo ABCDG HKOPQZ^Ca. seguendo E. segnando L. loco instesso E. 103 io so EHK#. che son M. 104 Quando suf. BCDLMZ^Da. Com' io F. 105 a lor O. ch' alcun di fuor BF. alcun fuori E t (order altered 2" da manu) A v che alcuno se ne mette M. che fuor nessun Q. 107 e om. BDF. 108 buttarsi E. 109 E quei D. Et ei O. havean E. i lancuoli K. Et e' che di lacciuoli avea dovitia F. in gran dovizia L. HO io om. L. ill Quando proc. BDEILP5Z?. Quando precacio (sic] K. a mia DFGILUZA^BCD. 108 K has a similar change in xxxiii. 68. Ill The various interpre- tations of this line are well given by Blanc, Versuch &c. p. 202. He points out that not only the construction is ambiguous, either with a' miei or a mia, but even a mia itself is ambiguous, as it may be only a ' Florentinism ' for a miei. It is singular that (with the exception of M) the CANTO XXIi. l6l Alichin non si tenne, e di rintoppo Agli altri, disse a lui : Se tu ti cali, lo non ti verr6 dietro di galoppo, u 4 Ma battero sopra la pece 1' ali : Lascisi il colle, e sia la ripa scudo A veder se tu sol piu di noi vali. 117 O tu, che leggi, udirai nuovo ludo ! Ciascun dall' altra costa gli occhi volse ; Quei prima, ch' a cio fare era piu crudo. 120 Lo Navarrese ben suo tempo colse, Fermo le piante a terra, ed in un punto Salt6, e dal proposto lor si sciolse. 123 Di che ciascun di colpa fu compunto, Ma quei piu, che cagion fu del difetto ; Pero si mosse, e grido : Tu se' giunto. 126 Ma poco i valse : che Y ale al sospetto Non potero avanzar : quegli ando sotto, E quei drizzo, volando, suso il petto: 129 Non altrimenti 1' anitra di botto, 112 e or/i. DL. si ritenne I. 113 e disse DQ. 115 batteron D. 116 Lascia '1 collo KL. collo DGHKLOP/^a. le reni scudo D. al scudo K. 117 pio B (? cler. error). 119 dell' altra LC^a. altra parte I. Cias. gli occhi all' altra ripa F. 120 Quei che prima a far cio era piu duro FL (cio fare in L). primo DZ. che cio fare EGHa. che '1 cio fare Z. 122 in terra G. 123 preposto KOQ. stolse D. tolse CI. a lor si volse L. 124 di colpa ciascun D. ciasc. di loro F. colpo EHZ^4J?a. era comp. EGH. 125 Ma piu quei B. Ma quei (or quei) che piu ACFGKZ?. 126 Prima si mosse F. e om. L. 127 Poco gli D/). E poco BCJ3CSL. E poco valse CK. Ma poco valse FGILMOPZ. 1' alie DFZC. 1' ali ABCEGHKLMOPQ^a. el sospetto I. 128 Nol potero F 2 (altered from non\ di sotto D. cogliando (?) sotto C. colui ando F. che gli P. 129 Quei dirizzo C. Quei dir. L. insuso D. su volando il F. 130 1' anedra A. 1' anatra BP. 1' anetra E. Ik entra L. list of MSS. containing a mia corresponds with that given in the Proleg. as being of Tuscan origin. 123 Comp. prccacciarc in 1. in, and preposto in 1. 94. D. I. I I 1 62 INFERNO Quando il falcon s' appressa, giu s' attuffa, Ed ei ritorna su crucciato e rotto. 132 Irato Calcabrina della buffa, Volando dietro gli tenne, invaghito Che quei campasse, per aver la zuffa. 135 E come il barattier fu disparito, Cosi volse gli artigli al suo compagno, E fu con lui sopra il fosso ghermito. 138 Ma 1' altro fu bene sparvier grifagno Ad artigliar ben lui, ed ambo e due Cadder nel mezzo del bogliente stagno. 141 Lo caldo sghermitor subito fue : Ma pero di levarsi era niente, SI aveano inviscate 1' ale sue. 144 Barbariccia, con gli altri suoi dolente, Quattro ne fe' volar dall' altra costa Con tutti i raffi, ed assai prestamente \ 4 - Di qua, di la discesero alia posta: Porser gli uncini verso gl' impaniati, Ch' eran gia cotti dentro dalla crosta : 150 E noi lasciammo lor cosi impacciati. 132 E quei YA Z C. E io ritornai L. in su DKL. giu D, 134 vaghito C. 136 E come quei A. dispartito ADLMQv5C7A dipartito EIZ. 137 Volse cosi O. volse om. E. a suo L. 138 fossone I. ingremito B. inghermito Q. gremito EL.5. scher- nito K. 139 E 1' altro D. si fu ben C. parve ben M. sparvier selvastro G. 140 E d' artigliar Q. Ad artigliarle in lui et anche i due E. ben om. F. sicche amenduni D. 141 Cade A. Caddero insieme F. 142 scherm. ACDGHIKMPQZZ?a. scernitor E. 143 Ma poco Z. era ricente K. 144 Si erano F. Si erano vescate C. 1' ali ABDEGHLMOPQC 1' alie FZ. 145 e gli altri L. 146 ne fa B&. 147 graffi ID. 149 E porser F. Porse G. Presor L. Volser K. Appresso li Q. in verso ABIKL. 150 costa BCEGHLMA(marg-.). 151 lasc. costoro imp. a. 142 The interchange of . 21 Quando dissi BC 1 (?)DEILMPC Maes, senza cessi (?) E. che non celi L. 22 mi pavento C. 23 Dei Mai. A 2 C. Da Mai. Q. retro B. non ci venghin dietro F. 24 Io sento A. 25 se fossi BCDM. si fosse a. di piombato ABCEGLMOPA 26 di fuori tutto D. tua di fuor F. 27 Si tosto FI. Si t. come quella F. quel che dentro M. 28 li om. M. vennero I. coi mei P. 30 d' entrambi un bo (? un bon) K. d' intrambo A^D. 31 e cosi la AP. si che la FM. 32 alta G. 33 si 'maginata G. 34 compio M. compier C. non pi& (with com added in margin) F. cotalcons. G. 35 coll' alie DFZC ale IMPD. ali aperte G. 36 volerci Z. 19 Compare a similar construction in which both tutto and tutti are found, in xix. 64. 21 A has Quando altered to Quanto. 22 Several MSS. have io pav., but this may stand for f ho pai>. Cf. 1. 96. 24 A seems to prefer Io to gli: see again xxiv. 121. 30 K is very indistinct hereabouts. CANTO XXIII. 165 Lo Duca mio di subito mi prese, Come la madre ch' al romore e desta, E vede presso a se le fiamme accese, 39 Che prende il figlio e fugge e non s' arresta, Avendo piu di lui che di se cura, Tanto che solo una camicia vesta: 42 E giu dal colle della ripa dura Supin si diede alia pendente roccia, Che 1' un dei lati all' altra bolgia tura. 45 Non corse mai si tosto acqua per doccia A volger rota di molin terragno, Quand' ella piu verso le pale approccia, 48 Come il Maestro mio per quel vivagno, Portandosene me sopra il suo petto, Come suo figlio, non come compagno. 51 Appena fur li pie suoi giunti al letto Del fondo giu, ch' ei furono in sul colle Sopresso noi : ma non gli era sospetto ; 54 Che 1' alta provvidenza, che lor voile Porre ministri della fossa quinta, Poder di partirs' indi a tutti tolle. 57 Laggiu trovammo una gente dipinta, 38 la donna F. ch' a romore COBa.. ch' a remor EQ. 39 Che vede M. 40 ist 'e' om. LP. 42 sola AFQZ. camisa resta E. 43 dal collo BCDEFGHKLMOPZ^Ca. del colle I. riva P. 44 siede Z. alia dolente M. 46 mai cosi acqua F. mai acqua si forte L. acqua si tosto Q. di doccia I. 48 Quinci della piu (?) L. le spale C*M. sopra le pale D. 50 Portandose e me a 2 . sopra al LQ. 51 e non AKQ. 52 furo i suoi pi EPOa. furono giunti i pi& suoi L. 53 Del collo giu O. fuorono sul C. che giunsero I. 54 non v' era FI. 56 Poner min. KDIA. Porer CL. dalla C. 57 Poter di dipartirsi a tutti D. Poter IQZ. Poder di dipartirsi indi lor K. Poder e dipartir se in ver lor tolle L. di om. PQ. Di partirsi potere a tutti M. di partirsi Q (with di erased 2 nda manu}. d' indi partirsi Z. 38 F's emendation is in as good taste as the suggestion made some years ago to substitute 'lady' for 'woman' in the Prayer Book! 53 O has, I think, generally collo for colle (rhyme permitting). 1 66 INFERNO Che giva intorno assai con lenti passi Piangendo, e nel sembiante stanca e vinta. 60 Egli avean cappe con cappucci bassi Dinanzi agli occhi, fatti della taglia Che in Clugnl per li monaci fassi. 63 Di fuor derate son, si ch' egli abbaglia ; Ma dentro tutte piombo, e gravi tanto, Che Federico le mettea di paglia. 66 O in eterno faticoso manto ! Noi ci volgemmo ancor pure a man manca Con loro insieme, intenti al tristo pianto : 69 Ma per lo peso quella gente stanca Venia si pian, che noi eravam nuovi Di compagnia ad ogni mover d' anca. 72 Perch' io al Duca mio: Fa che tu trovi Alcun, ch' al fatto o al nome si conosca, E gli occhi si andando intorno movi. 75 Ed un, che intese la parola Tosca, Diretro a noi grido : Tenete i piedi, 59 Che intorno gia D. 60 E nei sembianti loro istanca F. ist 'e' om. K. 61 havea E. cappi P. 62 E fatti tutti quanti a quella taglia D. fatti a quella I. fatte UBC. 63 Che in Cicilia D. Cologna ABEGIKM. Cologni HLOQ. Cachologni ( = Che in Cologni ?) C*. Crugni C. Che indigiu (?) Z. Che per li mon. in Cologna VBD. Che cocollon (?) F. vassi D. 64 son om. E. barbaglia E. 65 E dentro FZ. Ma dentro piombo e gravi tanto A. son di piombo F. piombo tutte IZ. piombo tutto G. tutti p. P. tutto e piombo CD. tutte e O. tutto p. Q. 67 e fatigoso P. fa di cose (!) L. amanto DF. 68 Noi si volg. K. volg. pur alia man A. volg. pur dalla m. D. 69 attenti BFZ>. 71 Venian BCFGKLMO^a. 72 volger d' anca KL. 73 Perche al G. Et io Maestro mio Z. 74 Qualchun F. fatto il nome BD.Z?. che '1 fatto e '1 nome KLQ. o il MPD. che '1 fatto o '1 nome Z. 75 E io si andando Q t . cosi F. 76 E que' L. 77 Drieto da B. 63 Possibly the reading in Z arose from cligni=clugn\. See a dis- cussion of the reading here in Dionisi, Aneddoti v. p. 56. The terzina 64 66 is omitted in L. 67 Qy. oral blunder in L? atnanto is perhaps a reminiscence of ii. 27. 68 On K's reading see note on xvii. 34. CANTO XXIII. 167 Voi, che correte si per 1' aura fosca : 78 Forse ch' avrai da me quel che tu chiedi. Onde il Duca si volse, e disse : Aspetta, E poi secondo il suo passo procedi. 81 Ristetti, e vidi due mostrar gran fretta Dell' animo, col viso, d' esser meco ; Ma tardavagli il carco e la via stretta. 84 Quando fur giunti, assai con 1' occhio bieco Mi rimiraron senza far parola : Poi si volsero in se, e dicean seco : 87 Costui par vivo all' atto della gola: E s' ei son morti, per qual privilegio Vanno scoperti della grave stola ? 90 Poi disser me : O Tosco, ch' al collegio Degl' ipocriti tristi se' venuto, Dir chi tu sei non avere in dispregio. 93 Ed io a loro : lo fui nato e cresciuto 78 per la via AD. su per FGQCD. che om. L. si om. L. 79 tu da me quel che chiedi Q. quello che chiedi L. 80 Perche '1 Duca F. 81 secondo al B. il om. A. passo suo L. 82 a due DM. 83 Nell' animo B. Dell' anima L. Coll' animo e col viso FM. Innell' atto del viso I. e col viso A. e del viso C. 84 impacciavagli F. 85 Quand' ei fur A^C. 86 rimiravan DI. riguardaro C. 87 si ri volse DM. si rivolser insieme C. si rivolsero ALO. dicea seco AM. 88 Questo IP. Questi M. 89 E se son ABDEFGHIOPQZ^Ca. E s' egli e morto CM. 90 Va iscoperto C. Va discop. M. clalla gr. I. grave schola D. 91 Poi disser O Tosco C. Poi disse O Tosco D. Poi disse a me FIMQ. Poi disser a me GP. Poi mi disser CD. 93 Di (or Di'?) BCEGHLOPQZ^a. Di che D. Di cui M. non 1' avere CDGLM. non n' avere B. a dispr. B. 94 Ed io a lui Q. allora ADEHZ. 78 la via (i.e. laida) in AD is probably a blunder for laura. Comp. the confusion of oura (i.e. ovra or oprd) and cura in Par. ii. 27. 89 But se may stand for se 1 or s' ei, as often in similar cases (e.g. xxiv. 9, &c.). So also que = quei xxiv. 112 and passim, qua=quaixx.v. 36, e = ei xxv. 40, suo = suoi xxvii. 38, a = hai xxix. 7. Perhaps this last may explain the curious lect.fals. 'hai' in xxxii. 136. C's reading is a dull and matter of fact alteration, because only one had been mentioned in 1. 88 as ap- parently living. 90 Stola has been altered to scola in C (cf. D), perhaps as an echo from collegio in next line. 1 68 INFERNO Sopra il bel fiume d' Arno alia gran villa, E son col corpo ch' i' ho sempre avuto. 96 Ma voi chi siete, a cui tanto distilla, Quant' io veggio, dolor giu per le guance, E che pena & in voi die si sfavilla ? 99 E 1' un rispose a me : Le cappe ranee Son di piombo si grosse, che li pesi Fan cosi cigolar le lor bilance. 102 Erati Godenti fummo, e Bolognesi, Io Catalano, e questi Loderingo Nomati, e da tua terra insieme presi, 105 Come suole esser tolto un uom solingo Per conservar sua pace, e fummo tali, Ch' ancor si pare intorno dal Gardingo. 108 Io cominciai : O frati, i vostri mali... Ma piu non dissi : ch' all' occhio mi corse Un, crocifisso in terra con tre pali. m Ouando mi vide, tutto si distorse, Soffiando nella barba coi sospiri : E il frate Catalan, ch' a cio s' accorse, 114 Mi disse: Quel confitto, che tu miri, Consiglio i Farisei, che convenia Porre un uom per Io popolo a' martiri. 117 95 Sovra a L. gran fiume K. 96 sempre tenuto F. 97 a chi G. 98 Quanto veggio ILO. su per CG. 99 pene G. e am. L. favilla E. 100 Et un AB. L' un L. 101 grosso ELA V si gravi F. lor pesi DFM. 102 le sua bil. I. 103 furono D. e om. I. 104 questi e Lod. DI. 105 Nom. da tuo terra L. 106 a uom D. un che uom I. 108 d' intorno MQ>. 109 fratri B. nostri I. i om. L. 110 che 1' occhio D. chi- gliocchi (sic) C. ch' agli occhi GIZB. 112 Quande ( = Quand' ei) D. discorse D. 113 con sosp. ADEGHKMPQ^. cosospiri CZ. 114 Et frate QA.f. Catalan di cio G. 116 Falisei CQ^). che'l convenia K. 117 Porer C (as in 1. 56). Deporre M. L in hopeless confusion reads, Porte vemeni a martiri. 96 In C a superfluous (K]o has been inserted before avuto, it not being observed that chio = cK f ho. See 1. 22. CANTO XXIII. 169 Attraversato e nudo e nella via, Come tu vedi, ed e mestier ch' ei senta Oualunque passa com' ei pesa pria : 120 Ed a tal modo il suocero si stenta In questa fossa, e gli altri del concilio Che fu per li Giudei mala sementa. 123 Allor vid' io maravigliar Virgilio Sopra colui ch' era disteso in croce Tanto vilmente nell' eterno esilio. 126 Poscia drizzo al frate cotal voce : Non vi dispiaccia, se vi lece, dirci Se alia man destra giace alcuna foce, 129 Onde noi ambo e due possiamo uscirci Senza costringer degli angeli neri, Che vegnan d' esto fondo a dipartirci. 132 Rispose adunque : Piu che tu non speri S' appressa un sasso, che dalla gran cerchia Si move, e varca tutti i vallon feri, 135 Salvo ch' a questo e rotto, e nol coperchia : Montar potrete su per la ruina, Che giace in costa, e nel fondo soperchia. 138 Lo Duca stette un poco a testa china, 118 Attraversato giace per M. e om. BLP. nudo nella AFZ. sulla via C. e nudo per la I. entro la via L. e indelavia (sic] B. 119 tu '1 vedi B. ed om. ACDEHKLQj. ed e om. B^. mestier e P. che senta CDFI. 120 come pesa ACDEFGHIOPZ>a. Qual. pasacome pasa L. 121 E tal modo E. 122 e om. L. dal cone. ABEFHOPQZ^a. 123 Che fur O. semenzaa. 124 io om. L. 125 costui M. 127 Poi dirizzo D. a' frati I. 128 se '1 ve lece EKM. 132 di questo fondo E. d' esto loco KLZ. a om. D. a partirci G. 133 Risp. appresso F. Risp. dunque D. 134 un scoglio M. della Ma. 135 tutti altri feri L. 136 Salvo che questo ABiCDEFGHIMOPQZ^P^ZJa. nol soperchia F. 137 potete la. 138 e om. L. costa che '1 D. 118 Z has c inserted above before nella, apparently i nda manu, and per contra a has I erased here. 119 B has et % inserted above 2 ntia manu. So also e in 1. 118. 127 The terzina 127 9 is omitted in L. I/O INFERNO Poi disse : Mai contava la bisogna Colui, che i peccator di la uncina. 141 E il frate : lo udi' gia dire a Bologna Del Diavol yizii assai, tra i quali udi' Ch' egli e bugiardo, e padre di menzogna. 144 Appresso il Duca a gran passi sen gl, Turbato un poco d' ira nel sembiante : Ond' io dagl' incarcati mi parti' 147 Dietro alle poste delle care piante. 140 mal conta D. 141 Ch' a lui ch' era pec. L. di qua ABCDEFGHIKLMOPOZ(quae)^CZ?a. vicina CA (marg.\ 142 dire gia D. io vidi AL. dire om. F. dicer gia Bologna G. dire Bol. B. 143 Che '1 diav. L. fra quali F. 144 menzogne a. 145 passo AB. 146 d' ira un poco I. 147 mi diparti' EH. 143 G has iude or uide (z's not being dotted), which may be f udi or vidi. CANTO VENTESIMOQUARTO In quella parte del giovinetto anno, Che il sole i crin sotto 1' Aquario tempra, E gia le notti al mezzo dl sen vanno : 3 Quando la brina in sulla terra assempra L' imagine di sua sorella bianca, Ma poco dura alia sua penna tempra ; 6 Lo villanello, a cui la roba manca, Si leva e guarda, e vede la campagna Biancheggiar tutta, ond' ei si batte 1' anca : 9 Ritorna in casa, e qua e la si lagna, Come il tapin che non sa che si faccia ; Poi riede, e la speranza ringavagna, 12 Veggendo il mondo aver cangiata faccia In poco d' ora, e prende suo vincastro, E fuor le pecorelle a pascer caccia: 15 Cosl mi fece sbigottir lo Mastro, Quand' io gli vidi si turbar la fronte, 1 In questa L. 2 inchrin D. sole intiene (?) L. i crini 1' Aq. C. templa (and in 1. 6) F. 3 a mezzo FGI. al mezzo sene vanno A. ne vanno F. 4 exempra AA. assempla F. sempra M. 6 poco basta F. e la sua GDI. prima tempra AE. pena CDKLM. 8 leva guarda GO. leva vide e guarda P. 9 onde (? = ond' ei) CDFKLMQa. 10 Kit. in cha A. a casa QZB. 11 Come tap. IO. taupin B. 12 Poi torna M. E has a la altered i nda manu to e la. riguadagna I. il ring. CD. 13 Vedendo MZ?. cangiato F. 14 poca A Z C. e vede '1 suo K. e prender suo viaggio D. il suo Q. 17 Quando gli BCDGILMOZ^?)/?. Vedendogli cosi cangiar la F. 1/2 INFERNO E cosi tosto al mal giunse lo impiastro: 18 Che come noi venimmo al guasto ponte, Lo Duca a me si volse con quel piglio Dolce, ch' io vidi prima a pie del monte. 11 Le braccia aperse, dopo alcun consiglio Eletto seco, riguardando prima Ben la ruina, e diedemi di piglio. 24 E come quei che adopera ed estima, Che sempre par che innanzi si proveggia ; Cosi, levando me su ver la cima 27 D' un ronchion, avvisava un' altra scheggia, Dicendo : Sopra quella poi t' aggrappa ; Ma tenta pria s' e tal ch' ella ti reggia. 30 Non era via da vestito di cappa, Che noi a pena, ei lieve, ed io sospinto, Potevam su montar di chiappa in chiappa. 33 E se non fosse, che da quel procinto, Piu che dall' altro, era la costa corta, Non so di lui, ma io sarei ben vinto. 36 Ma perche Malebolge in ver la porta Del bassissimo pozzo tutta pende, Lo sito di ciascuna valle porta 39 Che 1' una costa surge e 1' altra scende : Noi pur venimmo alfine in sulla punta 18 giunse al mal L. 19 Et come F. Como noi Q. mastro (sic) ponte L. 20 Lo duca mio IQ. 21 ch' il vidi DGIZ. ch' io '1 vidi HP. imprima BC. al pie EK. ponte (obvious error) EH. 22 e dopo FM.#. 25 adopera exstima M. e stima DI. e istima BFQ5. et extima EHKPZ. 26 E sempre innanzi par che si pr. F. 27 portando su me M. per la DG. su in ver FL. mi in ver Q. 28 ronchone D. d' un altra L. 29 a quella DL. quella omai G. 30 tienti imprima si tal che ti r. E. prima BCD. s' ell' e tal che ti Q. che ti BCDFLC se credi che ti Z>. 31 Non v' era a. da cappa a. 33 montar su F. d' in chiappa in a. 34 precinto ACEGHMOPZ^a. pcinto BQ. 35 dell' altro CI. altra G. costa torta IMQ. 36 ma sere' io Q. 38 Del suo bass. A. Del bass, cerchio F. prende C. 39 da cias. C, valle morta F. 40 pende (o/ioiorc'Xeura) M. 41 dal fin Z. infine B. in una punta G. CANTO XXIV. 173 Onde 1' ultima pietra si scoscende. 42 La lena m' era del polmon si munta Quando fui su, ch' io non potea piu oltre, Anzi mi assisi nella prima giunta. 45 Omai convien che tu cosi ti spoltre, Disse il Maestro, che, sedendo in piuma, In fama non si vien, ne sotto coltre, 48 Senza la qual chi sua vita consuma, Cotal vestigio in terra di se lascia, Qual fummo in aer ed in acqua la schiuma : 51 E pero leva su, vinci 1' ambascia Con 1' animo che vince ogni battaglia, Se col suo grave corpo non s' accascia. 54 Piu lunga scala convien che si saglia : Non basta da costoro, esser partito : Se tu m' intend!, or fa si che ti vaglia. 57 Leva' mi allor mostrandomi fornito Meglio di lena ch' io non mi sentia ; E dissi : Va, ch' io son forte ed ardito. 60 Su per Io scoglio prendemmo la via, Ch' era ronchioso, stretto e malagevole, Ed erto piu assai che quel di pria. 63 Parlando andava per non parer fievole, 42 se conscende P. 43 munto (sic) G. 44 Quand' io EFHKMOPQyl Quando fui ch' io L. potei D. pote' I. 45 m' affixi (sic) P. sulla prima Q. 46 Ormai L. cosi che tu D. 47 seggendo ABCDFGHIKLOPQZZ?a. 48 ma sotto EH. 50 vestige D. 51 o in acqua schiuma DF. o in Z. e in acqua ischiuma L. 52 Pero levati su P. lieva vinci L. suso e vinci D. 64 con suo K. si cascia Q. 57 si om. K. 58 Leva' mi su mostr. F. 59 Di lena meglio Z. 60 or va FG. 61 pren- dendo D. proceder la via K. 62 ronchiosa istretta L. e stretto BCZa. ronch. sconcio F. 63 E dentro (corrected to entro) D. E piu alto assai E. erta...che quella L. da pria EM. 64 debole F. parer piu f. E. The terzina 4345 was omitted in Q, and has been added by a later hand. INFERNO Ed una voce uscio dall" altro fosso, A parole formar disconvenevole. 66 Non so che disse, ancor che sopra il dosso Fossi dell' arco gia che varca quivi ; Ma chi parlava ad ira parea mosso. 6y lo era volto in giii ; ma gli occhi vivi Non potean ire al fondo per 1' oscuro : Perch' io : Maestro, fa che tu arrivi 7 * Dall' altro cinghio, e dismontiam lo muro ; Che com' i' odo quinci e non intendo, Cosl giu veggio, e niente affiguro. 75 Altra risposta, disse, non ti rendo, Se non lo far : che la domanda onesta Si dee seguir coll' opera tacendo. 78 Noi discendemmo il ponte dalla testa, Dove si giunge coll' ottava ripa, E poi mi fu la bolgia manifesta: 81 E vidivi entro terribile stipa Di serpenti, e di si diversa mena, Che la memoria il sangue ancor mi scipa. 84 Piu non si vanti Libia con sua rena ; Che, se chelidri, iaculi e faree 65 Ind' una A v Onde una ABCDEGHIKLOPQZa. dell' altro ABCDFGILMOQZa. dell' alto EHKP. 66 Che parole formo M. schonvonevole D. 68 Foss' io AC^ (altered to fossi} M. che guarda D. 69 pareva ad ira E. d' ira M. ad ire O (mirg.']. 70 volto gius' e gli Z*. 71 gire M. 72 tu om. G. fa che mi L. 73 altra cinghia Z. dismonta L. 74 non ne int. DL. 75 giu guardo F. mentre aff. E. niente no M. 77 Se no nel far M. di far F. 78 Si dea fugire L. 79 scendemo D. al ponte L. 80 Dov' el se giunge K. Ove B. s' aggiunge ABCEFGHIMOPQZ^ 2J ffCZ?a. 82 si teribel A. 83 e om. BDGIZA 2 CD. si om. EQ. e di si fatta C. 84 ne scipa FGHOPZa. stipa DK. vi stipa E. ne stipa L. 85 si avanti A. Labia K. che sua D. di sua F. 86 Con suo chrelidi L. quegli idri D. colidre K. colidri M. e achuli D. quelle yacule F (with ydrie in margin). pharetrei G. o faree A x . 84 /'/ sangue was at first omitted in E, and is added above. 86 Che secolidrijacolli epharee is the whole line in C. CANTO XXIV. 1/5 Produce, e cencri con amfisibena; 87 Ne tante pestilenzie ne si ree Mostro giammai con tutta 1' Etiopia, Ne con ci6 che di sopra il mar rosso ee. 90 Tra questa cruda e tristissima copia Correvan genti nude e spaventate, Senza sperar pertugio o elitropia. 93 Con serpi le man dietro avean legate : Quelle ficcavan per le ren la coda E il capo, ed eran dinanzi aggroppate. 96 Ed ecco ad un, ch' era da nostra proda, S' avvent6 un serpente, che il trafisse La dove il collo alle spalle s' annoda. 99 Ne O si tosto mai, ne I si scrisse, Com' ei s' accese ed arse, e cener tutto Convenne che cascando divenisse : 102 E poi che fu a terra si distrutto, La polver si raccolse per se stessa, E in quel medesmo ritorno di butto: 105 Cosi per li gran savi si confessa, Che la Fenice more e poi rinasce, 87 centri A 1 (?)BCDE(?)FGHOP J 5Ca. cintti (?) K. in centri M. L has 'Si riduce incierti chonafi si bena' (!). 89 la tyopia A. la thiopia E. 90 che sopra F. al mar ABCDHILMOPQZa. mare osdei (?) G. 92 Andavan C. Correva L. 93 Senza aspettar L. pertuso (sic) la. egli intropia (sic) D. 94 serpenti D. 95 Quelli (or Ouegli) EGI. facevano E. menavan I. con le L. gli reni G. 97 Ed ecco un E. di nostra B^PJDZa. alia nostra G. 99 La ove FO. il capo M. e le spalle L. 100 N& o ne i si tosto mai F. N io si tosto may n iscrisse E. Ne O tosto L. si presto I. 101 Come quel A. e 'n cener FI. 102 el divenisse F. 103 E poi ch' ei AD. ch' el fu Z. 104 La cener EFHP. s' accolse BQ. ricolse A v per lei stessa K. e per se B. raccolse e poi per L. 105 E quel IL. In quel B. si torno Z. 107 Che la finisce DL. le fenice E. Che lo fen. CHOP, e si rinasce K. 87 A has zencri apparently altered from zentri. It should be noted that c and / are often scarcely distinguishable in MSS. (and especially in E). 100 This line has been erased and rewritten in D. 1/6 INFERNO Quando al cinquecentesimo anno appressa. 108 Erba, ne biado in sua vita non pasce, Ma sol d' incenso lagrime ed amomo ; E nardo e mirra son 1' ultime fasce. m E qual e quei che cade, e non sa como, Per forza di demon ch' a terra il tira, O d' altra oppilazion che lega 1' uomo, n 4 Quando si leva, che intorno si mira Tutto smarrito dalla grande angoscia Ch' egli ha sofferta, e guardando sospira ; 117 Tal era il peccator levato poscia. O potenzia di Dio quanto se' vera ! Che cotai colpi per vendetta croscia. 120 Lo Duca il domando poi chi egli era : Perch' ei rispose : lo piovvi di Toscana, Poco tempo e, in questa gola fera. 123 Vita bestial mi piacque, e non umana, SI come a mul ch' io fui : son Vanni Fucci Bestia, e Pistoia mi fu degna tana. 126 Ed io al Duca : Digli che non mucci, E domanda qual colpa quaggiu il pinse : Ch' io il vidi uomo di sangue e di crucci. 129 E il peccator, che intese, non s' infinse, 108 Quando el FIMQ. s' appressa D. 109 Ch' erba G. biada BCDFIPQ. 110 e lagr. CD. di lagr. L. e lagr. d' amomo FGM. inc. lagr. d' amomo Q. Ill E mirra e nardo D. su P ultime a. 112 e quel LOQa. 114 O altra FIL. 115 che 'ncontro Q. 116 stordito D. della DGLOPa. per la Z. grave ang. D. 117 Che ha sofferta K. e om. a. 118 levato il peccator L. 119 vendetta A*. giustitia Z. quanto severa E. quante ( = quant' e) severa E*IOZA Z S. sey severa A. 121 il om. DI. chi lo era A. ello era E. 123 tempo et in E. 125 Si come malo D. Si come mul FGP. Si come mulo L. 127 non nuoci K. 128 Domandalo che B. E sappi qual peccato F. che colpa EGMPa. qual pena D. qui giii K. 129 Ch' i '1 vidi un uom C. Ch' i '1 vidi uom gik I/?, corrucci BC*ILP/>. cruoci K. cornucci C. 130 si finse AGP. 119 Bj seems to have had quanto: C l had quante Severn, which has been altered to quanto. CANTO XXIV. 177 Ma drizzo verso me 1' animo e il volto, E di trista vergogna si dipinse ; 132 Poi disse : Piu mi duol che tu m' hai colto Nella miseria, dove tu mi vedi, Che quando fui dell' altra vita tolto. 135 lo non posso negar quel che tu chiedi ; In giu son messo tanto, perch' io fui Ladro alia sacrestia de' belli arredi ; 138 E falsamente gia fu apposto altrui. Ma perche di tal vista tu non godi, Se mai sarai di fuor de' lochi bui, 141 Apri gli orecchi al mio annunzio, ed odi : Pistoia in pria di Negri si dimagra, Poi Fiorenza rinnuova genti e modi. 144 Tragge Marte vapor di val di Magra Ch' e di torbidi nuvoli involute, E con tempesta impetuosa ed agra 147 131 inverse KL. 1' anima F. 132 el se dip. E. 133 Poi mi disse L. 134 che tu qui B. Ik dove mi F. 135 quand' io ABFKL. alta Z. 137 ma giu D. 139 E fals. fuo apposto Q Z (fue). fu gia CLA^D. fu posto G. 140 Ma perche tu di tal Z. 141 di Ik da L. dai luoghi ABCEGHKOPQ^. de' cerchi F. 142 F orecchie DEMPQ. le orecchi O. nunzio K. 143 Pist. prima D. in prima KOQa. Pist. pria D. di negra D. dei Neri B. di Neri AFGHIKOPQZ?a. di Nera E (di at first omitted). de Neri LM. dismagra G. dimacra A r 144 E poi Firenze F. Firenze BGIQ^. Fiorenze a. Firenza AP. gente BCDGIP. 145 Macra A r 146 a turb. D. torbi (sic) E. da torb. I. einvol. DEjF. involto AE 2 Z. e involto E t L. 147 acra A v 135 B's reading is probably altra. It is written all a and this seems to be his way of writing a final -tra. Many similar cases occur. So we find sometimes diet , e.g. xxxiii. 135. 143 de in LM may stand for tit or de\ as in many cases elsewhere ; and ought we not to read de 1 Neri instead of di Negri in Witte's text ? The name Neri and not Negri is universally associated with that of the political party in question. And perhaps de 1 rather than di, because, as has been argued, this was what Pistoia did in May 1301, i.e. it was a complete (not partial) expulsion of 'the Neri.' 144 gente is really indeterminate, as in many MSS. the plural is written thus, and the variation is therefore not generally registered. D.I. 12 178 INFERNO Sopra campo Picen fia combattuto : Ond' ei repente spezzer^ la nebbia, SI ch' ogni Bianco ne sara feruto : 150 E detto 1' ho, perche doler ti debbia. 148 Sovra '1 C. Viceno E. e Picem F. 149 Onde rep. BFILOMQZa. 150 ogni biada L. 151 dolor EG. ten debbia B. CANTO VENTESIMOQUINTO Al fine delle sue parole il ladro Le mani alz6 con ambedue le fiche, Gridando : Togli, Iddio, che a te le squadro. 3 Da indi in qua mi fur le serpi amiche, Perch' una gli s' avvolse allora al collo, Come dicesse : lo non vo' che piu diche : 6 Ed un' altra alle braccia, e rilegollo, Ribadendo se stessa si dinanzi, Che non potea con esse dare un crollo. 9 Ahi Pistoia, Pistoia, che non stanzi D' incenerarti, si che piu non duri, Poi che in mal far lo seme tuo avanzi. n Per tutti i cerchi dell' inferno oscuri Non vidi spirto in Dio tanto superbo, 3 Dicendo BFKM. Dio BEIOQ 2 #. or togli...che (with a space left thus) M. 4 D' allora K. Da indi inanzi C. mi fun B. fue M. le serpe AEK. li fuor li P. 5 attorno el M. 6 Quasi dicesse M. io om. CDIKMOZ?a. non voglio L. piu nonvocheZ. che tu diche I. 8 Eribattendo ADGP. Ri- batendo ELMOQ. Ribandendo I. si om. CjDK 2 . 9 essa BDZ>a. dinanzi dare I. di crollo C. 10 che nosstai anzi (sic) L. istanchi E. stanchi K. istanzi PZ. schanzi D. vi stanzi a(?). 11 D' ingenerare D. D' ingenerarti L. 12 in om. D. a mal F. al mal Q. lo tuo seme F. nel male il seme G. il mar far inseme tu L. 13 Per tutto lo inferno per li cerchi iscuri F. tutti cerchi EHOC>. tutti i luoghi Q. inf. duri C. scuro E. 14 Spirto non vidi BC. Non trovai F. in Dio spirto K. 3 For the omission in M see note on xv. 2. 8 si has been inserted in C 2 nda manu, and in K an erased space is left blank. 12 2 180 INFERNO Non quel che cadde a Tebe giu da' muri. 15 Ei si fuggl, che non par!6 piu verbo : Ed io vidi un Centauro pien di rabbia Venir chiamando : Ov' e, ov' e 1' acerbo ? 18 Maremma non cred' io che tante n' abbia, Quante bisce egli avea su per la groppa, Infin dove comincia nostra labbia. i\ Sopra le spalle, dietro dalla coppa, Con 1' ale aperte gli giacea un draco, E quello affoca qualunque s' intoppa. 24 Lo mio Maestro disse : Quegli e Caco, Che sotto il sasso di monte Aventino Di sangue fece spesse volte laco. 27 Non va co' suoi fratei per un cammino, Per Io furar frodolente che fece Del grande armento, ch' egli ebbe a vicino : 30 Onde cessar le sue opere biece Sotto la mazza d' Ercole, che forse Gliene die cento, e non sentl le diece. 33 Mentre che si parlava, ed ei trascorse, E tre spiriti venner sotto noi, 15 Ne quel I. a Tebe cadde F. da Tebe C. gia L. de' (or di) muri FIMPQ. 16 El si parti DP. sen fuggi AM. ch' el non EM. 17 E vidi Q. 18 gridando BFIMPC/?. 19 credo che ACDEKLO. non udi che Q. tanta G. 20 Biscie quante MP. in sulla gr. Z. 21 Insino G. ove ACDEGH KOQ*Z# 22 Sura le D. 23 Aperte coll ali G. ali aperti ABD. ali aperte EHLMOPQZ^a. alie F. 24 El quale I. E questo F. E quella L. gl' intoppa FI. 25 questo e B. questie CEHLOQy^^a. quest' e K. 26 del monte D. 27 Spesse volte fe' di sangue MP. molte volte AI. 28 pe' suoi fratelli L. con suoi (or suo) AEGIMOPQ. fratelli DEI. 29 furto ABCDFGHIKLMOPOa. frutto E. che frod. ABDEFGHIKL MOPQZZ>a. ch' ei fece B. 30 che gli fu vie. F. ebbe vicino EGILMP. 31 Ond' ei a. il cessar L. 33 Ne gli D. che non D. 34 Mentre che '1 si Q. ch' ei su per la via si tracorse L. stracorse M. 21 Lines 21 and 22 are omitted in L. CANTO XXV. l8l De' quai ne io ne il Duca mio s' accorse, 36 Se non quando gridar : Chi siete voi ? Per che nostra novella si ristette, Ed intendemmo pure ad essi poi. 39 Io non gli conoscea ; ma ei seguette, Come suol seguitar per alcun caso, Che 1' un nomare un altro convenette, 42 Dicendo : Cianfa dove fia rimaso ? Perch' io, acciocche il Duca stesse attento, Mi posi il dito su dal mento al naso. 45 Se tu sei or, Lettore, a creder lento Cio ch' io diro, non sara maraviglia, Che io, che il vidi, appena il mi consento. 48 Com' io tenea levate in lor le ciglia, Ed un serpente con sei pie si lancia Dinanzi all' uno, e tutto a lui s' appiglia. 51 Coi pie di mezzo gli avvinse la pancia, E con gli anterior le braccia prese ; Poi gli addento e 1' una e 1' altra guancia. 54 Gli diretani alle cosce distese, E miseli la coda tr' ambe e due, E dietro per le ren su la ritese. 57 36 Di quai G. Di qual K. mio om. LQ^ il Duca mio ne io A-L. 37 quand' ei B. chi siete [voi om.] D. 38 nov. nostra D. 39 attendemmo FEGHIMP. 41 Come '1 M. 42 Che un M. nomare 1' altro CMP. nominar 1' altro FI. uno nomera e 1' altro (sic) L. illaltro P. concedette G. 43 ove P. se' rimaso D. 44 Et io F. intento C. 45 al dito L. detto (sic) E. tra '1 mento e '1 naso PD. 46 O lettore L. 48 m' il cons. CGI KM. i' m' il cons. A. 49 Come tenea E. levato D. a lor LZ. 50 e con sei L. piedi D. 51 a uno D. rapto a lui Q. se piglia M. 52 aggiunse E (see note on xvii. 96). 53 1' enteriore colle braccia K. 54 ist e erased in C, and om. in EGKLPQ. E addento gli F una F. 55 delitani L. discese AEHOP^4 (marg.) a. 56 intr 5 ambedue AEGHOZ). fr' ambedue P. 57 distese DG. su om. L. 55 C* has distese. 1 82 INFERNO Ellera abbarbicata mai non fue Ad arbor si, come 1' orribil fiera Per T altrui membra avviticchio le sue : 60 Poi s' appiccar, come di calda cera Fossero stati, e mischiar lor colore ; Ne 1' un ne 1' altro gia parea quel ch' era : 63 Come precede innanzi dall' ardore Per lo papiro suso un color bruno, Che non e nero ancora, e il bianco more. 66 Gli altri due riguardavano, e ciascuno Gridava : O me, Agnel, come ti muti ! Vedi che gia non sei ne due ne uno. 69 68 Ela era abbarb. E. barbacata C. mai bis P. 59 Ad alberi come D. Ad albero come FI. alberi A. albero ELa. albor CGiQZ. alber GaHKOP^. albore B. 60 Per 1' altre D. Per altrui CE. Coll' altrui Q. avinchiose E. 61 se piccar K. da calda Z. 62 state ACDEFHIKa. fatti L. stricti O. a mis- chiar K. e om. LQ. colori L. lo colore Q. 63 parea gik E. altro non parea KL. 64 dinanzi CD. dell' ard. Qa. 65 Per lor D. pap. solo P. su un C. el color E. 66 Che non e fo (?) L. e om. CLQ. il om. D. e il om. M. e 1' altro more IZ. 67 E gli L. Gli altri lo rig. I. il rivardavano A. due il rig. EFHKOP. lo sguardavano G. e om. CLaj. cattuno A Z CD. 68 Gridavano DIL. Agnello C. Angel DD. Agnol M. perche ti L. 69 non sene vede niuno (?) L. ne duno C. 58 Qy. an oral blunder in E (as may often be suspected in this MS.). bo 66 In Z albo is written 2 tuia manu over altro [thus altro'}. 67 The form rivardavano in A seems to represent the evanescent sound of g in the Venetian dialect. Similarly we find vanni for gianni in this MS. at xxx. 32, vera for guerra in K at xxvii. 86, and vardi for guardi in xxxiii. 51. This is still more striking in the total omission of the letter now and then in Q, e.g. iunto, id, z'u, irando, &c. ior ginnfo, gia, giil, girando, &c. Compare also such forms as meio, maor, tolia, &c., for meglio, maggior, toglia, &c. 69 C's reading is probably equivalent to ned uno, the d being e'^eA/cvo-riKoi' like ched for che, as in xxxi. 138, and apparently sed for se in xxviii. 57 (D). Also ned io occurs in xxv. 36 (a). Castelvetro contends that iii. 52 should be read, Ched ogni posa &c., and gives an unusual interpretation to the passage accordingly. CANTO XXV. 183 Gia eran li due capi un divenuti, Quando n' apparver due figure miste In una faccia, ov' eran due perduti. 71 Fersi le braccia due di quattro liste ; Le cosce con le gambe, il ventre e il casso Divenner membra che non fur mai viste. 75 Ogni primaio aspetto ivi era casso : Due e nessun 1' imagine perversa Parea, e tal sen gi'a con lento passo. 78 Come il ramarro, sotto la gran fersa De' dl canicular cangiando siepe, Folgore par, se la via attra versa : 81 Cosl parea, venendo verso 1' epe Degli altri due, un serpentello acceso, Livido e nero come gran di pepe. 84 E quella parte, donde prima e preso Nostro alimento, all' un di lor trafisse ; Poi cadde giuso innanzi lui disteso. 87 70 Giacean D. due corpi MA. 71 m' apparven B. mi parver C. m' apparve D. n' apparve EMZ>. 72 dove EH. dove eron perduti I. 73 Feronsi E. Fensi I. 74 II ventre colle gambe le coste e '1 casso D. e '1 ventre BCKLMQ. 75 membre M. che mai non fur F. 76 primo D. quivi L. v' era DIK. 77 Due insule margine diversa (sic) L. Dove nessun DQ. D' uno e I. Due nessuno F. 78 cotal E. e om. L. gfo AEHKOPa. 79 Come ramarro I L. 80 Di di C. De idi EMO. Deidi H. Dei can. L. cambiando BCGLQ. 81 traversa M. 82 Si parea ABCEGHKLOPZ^^?a. vedendo E. inverse I. 83 Con gli altri D. Degli altrui A. serpente D. 84 e om. L. granel di C. ner come granello I. 85 E qual E. E in quella FM. onde ABCDEGHLOPa. ove K. dove M. da prima B. ondeglie prima P. donde e pr. preso A 2 C. imprima A r 86 Nostro helemento F. alcun di lor L. 87 all' un di lor steso L. inn. a lui GIZ. giu dinanzi A^C. giu dinanzi a lui F. dinanzi a lui K. avanti a lui M. 74 I has el ventre, but el always stands for il in this MS., and may do so in some of the others cited as reading e 'L In Mline 77 is written in error (owing to the o/zoioreXtura 'casso' in 74 and 76), after line 74, and then repeated in its proper place. 184 INFERNO Lo trafitto il miro, ma nulla disse: Anzi coi pie fermati sbadigliava, Pur come sonno o febbre 1' assalisse. 90 Egli il serpente, e quei lui riguardava : L' un per la piaga, e 1' altro per la bocca Fumavan forte, e il fummo si scontrava. 93 Taccia Lucano omai, la dov' ei tocca Del misero Sabello e di Nassidio, Ed attenda ad udir quel ch' or si scocca. 96 Taccia di Cadmo e d' Aretusa Ovidio : Che se quello in serpente, e quella in fonte Converte poetando, io non 1' invidio : 99 Che due nature mai a fronte a fronte Non trasmuto, si ch' ambo e due le forme A cambiar lor materia fosser pronte. 102 Insieme si risposero a tai norme, Che il serpente la coda in forca fesse, E il feruto ristrinse insieme 1' orme. 105 88 traf. miro CQ. e nulla Q. 89 Anco G. piedi E. con pie M. fermato Q. 90 e febbre L. 91 E lo serpente che egli rig. D. 93 Fiammava D. Fumava BFG. Fumar L. forte al fumo L. s' incontrava IOZ. 94 omai Lucano K. ormai L. dove tocca ACDEFGHKLMOPQ7J?CZ>a. omai dove si toccha I. 95 Sabione L. Sabellio A. 96 Ed om. L. E stia ad udir M. intenda BCD. audir BDGHIQZa. ch' ora scocca C. quel che si tocca Z. che or si tocca FG. si si scocca M. 97 Chano D. Chamo B. Camo E. Canmo G. Cammo H. e om. E. e dare tu se Ov. (sic) D. di Retusa A. e Aretusa I. 98 quello serpente L. 99 Convert! IGQ. Trasmuta F. poet, nollo vidio C. io nol n' invidio F. e non 1' inv. C. 101 trasmu- tar(o) DQ. 102 matere Ba. matera AC*GHOP. materie IKQ-5C 103 si om. K. riposero (or -sono) ILZj. a tal norme EIK. a tal forme F. 104 in forma L. 105 E '1 trafitto O. feritoFIKPZ. 94 On dove tocca see note on xxiii. 89. 96 audir probably = a itdir or it may be an archaic form audir, as in xxvi. 78. 97 nm = mm in G passim. 102 The variation of matera for materia (as in 1. 125, &c.) is very common, and will not be again registered. CANTO XXV. 185 Le gambe con le cosce seco stesse S' appiccar si, che in poco la giuntura Non facea segno alcun che si paresse. 108 Togliea la coda fessa la figura Che si perdeva la, e la sua pelle Si facea molle, e quella di la dura. m lo vidi entrar le braccia per 1' ascelle, E i due pie della fiera, ch' eran corti, Tanto allungar, quanto accorciavan quelle. 114 Poscia li pie diretro, insieme attorti, Diventaron lo membro che 1' uom cela, E il misero del suo n' avea due porti. 117 Mentre che il fummo 1' uno e 1' altro vela Di color nuovo, e genera il pel suso Per 1' una parte, e dall' altra il dipela, 120 L' un si levo, e 1' altro cadde giuso, Non torcendo pero le lucerne empie, Sotto le quai ciascun cambiava muso. 123 106 cose E (s = sc passim in this MS.). coste C. insieme seco E. 107 ch' a poco L. che poco Z. 108 facendo E. che pertenesse K. 109 Toglie C. Tolia A. fissa C. a la figura L. 110 perd. gia DC perd. di la CQ. perd. di lei I. perteva e la la (sic] L. Ill quelle L. quella dell' uom M. 112 intralle D. vidi trar G. per le selle (sic) C. 113 E om. D. i om. KLQZ. della bestia I. forti M. 114 accortavan AG. scorciavan B. accorciaron FM. chorciava L. 115 li piedi D. di dietro BFI. da rietro K. 117 de suo E. parti DKL. 118 e 1' uno L. 119 e om. I. generar pel C. 120 e per 1' altra BKQ. e per altro F. e om. L. dell' altro dip. L. altra si pela K. altra dipela C 2 G. 121 in giuso C. 122 perci6 BCFQ. in pero E. le luc. o pi& (sic) L. 123 cambia il Qj (cambiava Q 2 ). 106 insieme in E is underlined to indicate an error (see previous line). 112 G similarly changes entrati into tratti in xxvi. 132. Also intralle in D = entrarle. Comp. chontallo for contarlo in xxviii. 114: rand / are often interchanged in this (Tuscan) MS., e.g. obriando = obbliando xxviii. 54, archimia xxix. 119, &c., and gro lia gloria, iii. 42, &c. Instances in other MSS. will be found registered at xxiii. 116, xxvii. 22, 47, xxviii. 82, xxix. 83. Note also the curious form delafrito = deW afflitto in xxvii. 10. 120 / has been erased before dipela in C. 186 INFERNO Quel ch' era dritto, il trasse ver le tempie, E di troppa materia che in la venne, Uscir gli orecchi delle gote scempie : 116 Ci6 che non corse in dietro e si ritenne, Di quel soperchio fe' naso alia faccia, E le labbra ingrosso quanto convenne. 129 Quel che giacea, il muso innanzi caccia, E gli orecchi ritira per la testa, Come face le corna la lumaccia : 132 E la lingua, che avea unita e presta Prima a parlar, si fende, e la forcuta Nell' altro si richiude, e il fummo resta. 135 L' anima, ch' era fiera divenuta, Si fuggl sufolando per la valle, E 1' altro dietro a lui parlando sputa. 138 Poscia gli volse le novelle spalle, E disse all' altro : lo vo' che Buoso corra, Com' ho fatt' io, carpon, per questo calle. 141 Cosl vid' io la settima zavorra 124 Quelle D. drieto K. dritto trasse in ver FI. dritto trasse '1 ver L. in ver DIKZ?. trasse per le Ca. 125 che gli avvenne AB*DFKMZ. che Ik venne CILQ. chi Ik Ea. 126 1' orecchie A (le rechie) DEKLMP. le orecchi O. goti P. 127 che ne corse L. in om. MQ. e om. ABFGKL. o se M. 128 il naso A. naso la faccia BCEHIKLOZ. alia om. B. 129 grosse L. 130 innanzi al muso K. innanzi el muso M. 131 le orecchie EKMP. le orecchi O. tiro verso la P. 132 Come fa delle ID. le corna face M. limaccia El. 133 unta L. 134 si fonde HP. se fesse M. a la fore. La. la 'nforcuta G. 135 Nell' altra L. arresta Q. 136 La ruina (sic) L. 137 Sufolando si fuggl ABCDFGHILOQy^a Suf. sen fuggl EKM. fugged. fuful- lando (sic) D. giu per C. per quella valle Q. 138 a om. PQ. 139 si volse KP. rivolse Z. 140 dicea D. agli altri D. Buoso om. M (cler. error]. 141 Carpon com' ho A. Come faccio DEGMQZ. Come fee' io F. Com' io feci K. Com' fo io B. in questo D. questa valle FZ. 125 chila is probably an inadvertence for chlla che in Id,. 141 a has been erased, but apparently Como fattio has been altered to something like Como faccio. CANTO XXV. IS/ Mutare e trasmutare ; e qui mi scusi La novita, se fior la penna abborra. 144 Ed avvegnache gli occhi miei confusi Fossero alquanto, e 1' animo smagato, Non poter quei fuggirsi tanto chiusi, 147 Ch' io non scorgessi ben Puccio Sciancato : Ed era quei che sol, de' tre compagni Che venner prima, non era mutato : 150 L' altro era quei che tu, Gaville, piagni. 143 e om. L. quivi mi L. 144 che fuor D. se for F. la lingua A (marg.} B. 145 Avvegna dio che gli (?) Q. che miei occhi DF. 146 all' animo C. ch' all' animo L. 147 Non poteron fuggire que' tanto D. qui (? = quei) CL. quegli Q. 148 Che non EO. scorgiesen E. 149 quei che '1 CQj(?). da tre M. 150 Che v' eran e prima L. e non DKL. 151 E 1' altro E. 144 C 1 had fior, which has been altered to fuor. CANTO VENTESIMOSESTO Godi, Fiorenza, poi che sei si grande, Che per mare e per terra batti 1' ali, E per 1' inferno il tuo nome si spande. 3 Tra li ladron trovai cinque cotali Tuoi cittadini, onde mi vien vergogna, E tu in grande onranza non ne sali. 6 Ma se presso al mattin il ver si sogna, Tu sentirai di qua da picciol tempo Di quel che Prato, non ch' altri, t' agogna. 9 E se gia fosse, non saria per tempo. Cosl foss' ei, da che pure esser dee ; Che piu mi gravera, com' piu m' attempo. 12 Noi ci partimmo, e su per le scalee, Che n' avean fatte i borni a scender pria, Rimonto il Duca mio, e trasse mee. 15 I Firenze BDEFGIKLCa. Firenza AOPZ>. chi se' P. 2 Che per terra e per mare batti tua ali G. 3 inferno tuo ACEFHLOPQa. 4 Tra gli altri ladron M. 5 ond' ei B. men vien G. 6 non sali G. 7 il mattin M. del ver ABCDEFGHKMOPQZ;4 2 5C>a. 8 Tu vederai M. da qua P. 9 altro DL. altre B. n' agogna M. 11 foss' egli poiche pur D. fosse da che C*L. fosse quando M. quando pur B. poich& P. da ch' egli esser E. da ch' ei H. 12 aggrava D. aggravera GQ. 13 e om. KL. 14 ne venian L. fatti borni ABC 1 (?)DFM^CZ?. fatto bornio (?) I. fatti i borni PQ. i lerni(?) K. 15 il mio Maestro ACDEGHIKLMOPQZa. trasse e mee L. II C has an erasure between fosse and da che. 14 C* has / borni. E has a correction which makes it doubtful whether facte or facti was original. CANTO XXVI. 189 E proseguendo la solinga via Tra le schegge e tra' rocchi dello scoglio, Lo pie senza la man non si spedia. 18 Allor mi dolsi, ed ora mi ridoglio, Quand' io drizzo la mente a ci6 ch' io vidi ; E piu Io ingegno affreno ch' io non soglio, 21 Perche non corra, che virtu nol guidi ; Si che se Stella buona, o miglior cosa M' ha dato il ben, ch' io stesso nol m' invidi. 24 Quante il villan, ch' al poggio si riposa, Nel tempo che colui che il mondo schiara La faccia sua a noi tien meno ascosa, 27 Come la mosca cede alia zenzara, Vede lucciole giu per la vallea, Forse cola dove vendemmia ed ara : 30 Di tante fiamme tutta risplendea L' ottava bolgia, si com' io m' accorsi, Tosto ch' io fui la Ve il fondo parea. 33 E qual colui che si vengi6 con gli orsi, 16 perseguendo K. la suo longa via K. 17 Tra rocchi et tra le schegge M. roncho L. ronchi Z. e li ronchi Q. dallo K. Io scheggio A v 18 le mani L. si partia D. spendia G. s' expedia M. 19 dolvi C. ora piu mi doglio C. et ancor mi doglio M. 20 Quando dr. DILKQ5CZ?. drizzai C. a quel ch' io YKD. 21 affermo C* (after evident erasure) EFH. cha non E. che non GO. 22 Perche e' non I. Perch' el Z. e che virtu K. 23 buona stella F. la Stella buona H. Stella non va La. 24 M' han dato GL. data a. stessi DGO. che stessi E. ben io stesso A$}D. non me '1 vidi D. nol mi vidi C. non m' inv. GIKLB. 25 Quando DGKQ. Quale il BCFM. Quale e il vill. L. al pocho D. 26 di colui O. 27 tien meno a noi nascosa D. men nascosaQ. 28 fede L. zanzara BCEFIKOPQ^a. 30 ove A v o ara D. e ch' ara L. 32 come m' accorsi E. 33 Tosto che fui ABEGHLZ5Z>a. fui dove '1 ADFIKMQZ^ 2 CZ>a. fui ove '1 CG. ove ch' il f. E. 34 vegghio DL. 21 Cha for che is frequent in E. 23 Probably this non va is for nuova. If so, compare nova fata in Juv. vii. 189. 24 In C mividi is probably for mividi, 25 As pozzo is very often written po$o, the vari- ant in D has probably arisen from mistaking po$o io\ pocn. IQO INFERNO Vide il carro d' Elia al dipartire, Quando i cavalli al cielo erti levorsi ; 36 Che nol potea si con gli occhi seguire, Ch' ei vedesse altro che la fiamma sola, SI come nuvoletta, in su salire : 39 Tal si movea ciascuna per la gola Del fosso, che nessuna mostra il furto, Ed ogni fiamma un peccatore invola. 42 lo stava sopra il ponte a veder surto, Si che, s' io non avessi un ronchion preso, Caduto sarei giu senza esser urto. 45 E il Duca, che mi vide tanto atteso, Disse : Dentro da' fochi son gli spirti : Ciascun si fascia di quel ch' egli e inceso. 4 8 Maestro mio, rispos' io, per udirti Son io piu certo ; ma gia m' era avviso Che cosi fusse, e gia voleva dirti : 51 Chi e in quel foco, che vien si diviso Di sopra, che par surger della pira, Ov' Eteocle col fratel fu miso ? 54 Risposemi : La entro si martira Ulisse e Diomede, e cosi insieme 36 Quando cav. C. alti A. erto FG. levarsi G. 37 no potea D. potea con D. 38 Che ved. FlKMffD. Ch' io ved. LC. 39 A modo di nuv. F. in suo DFGa. in sul C. in om. L. 40 muove IOQa. 41 Del fuoco 7LB. giu e niuna F. 42 Ch' ogni L. 43 Io m' era F. sotto al ponte G. 44 io om. Q. ronchio Tpreso Q. 47 Dentro dal foco, disse M. dentro a D. dal fuoco IKL. dentro alle fiamme F. 48 Chatun(o) DLa. Cadun C. s' infascia K. li fascia L. di quel(lo) ch' e DFK. acceso K. 49 M. mio diss' io C. risposi per KL. 50 qui certo E. e gik EG. che gia P. m' era viso ABCDKQ. me aviso L. 51 e om. L. 52 in questo M. 53 dalla pira D. 54 Quive teode D. Eteologo K. Teocle MQa. Teoleso L. e '1 fratel K. 55 Ed egli a me AM. R. a me BCDEFGHOPQZ?a. Ik deniro ACEG HKMOPQ>a. 56 Ulises e Diomedes cosi C. Ulixe et Diomede cosi Z. Ulises e Dyomede cosi A. Ulixe HOP. Ulise Dicomedes L. 56 Similarly C has Circes in 1. 91, and Ercules in 1. 108. See Dion. CANTO XXVI. 191 Alia vendetta vanno come all' ira : 57 E dentro dalla lor fiamma si geme L' aguato del caval, che fe' la porta Ond' usci de' Romani il gentil seme. 60 Piangevisi entro 1' arte, per che morta Deidamia ancor si duol d' Achille, E del Palladio pena vi si porta. 63 S' ei posson dentro da quelle faville Parlar, diss' io, Maestro, assai ten prego E riprego, che il prego vaglia mille, 66 Che non mi facci dell' attender nego, Finche la fiamma cornuta qua vegna : Vedi che del disio ver lei mi piego. 69 Ed egli a me : La tua preghiera e degna Di molta lode, ed io per6 1' accetto ; Ma fa che la tua lingua si sostegna. 71 Lascia parlare a me : ch' io ho concetto Ci6 che tu vuoi : ch' ei sarebbero schivi, Perch' ei fur Greci, forse del tuo detto. 75 Poiche la fiamma fu venuta quivi, Dove parve al mio Duca tempo e loco, 67 con mal ira B. corron B. 68 E dentro a quella F. dalle lor fiamme K. loro om. L. si si geme F. 69 El giunto del cav. I. che fe' alia porta D. 60 Donde F. di Roma K. 61 Piangevasi A. Piangeasi D. Piangesi C. Piangiensi L. dentro CD Fa. 1' arte dentro L. 62 E Deidomia DF. duol Ach. L. 64 Se posson BCDEIQCa. Se possor G. Se posso dentro M. 65 Maestro mio parlare F. ancor ti C. ti priego ABCDEIKQZ?. 66 E priego F. e che '1 E. che 1' un priego F. 69 dal desio A. mi piango L. 70 a me om. L. 71 loda BDEFGHILMOPQa. onde pero I. pero Io certo K. 73 ch' i' 1' ho B. 75 per tuo E. 77 Onde Q. apparve L. al mio Duca parve FM. il tempo L. Anedd. n.p. 102 on the significance of the retention of these pure Latin forms in MSS. 67 B's reading is probably a blunder from comal ira, especially as this MS. always affects unassimilated forms, such as sobmecton, tucto, atttumpno, &c. 75 Most of the MSS. have Perch} fur here and che sarebbero in the previous line, but this is really inconclusive. I9 2 INFERNO In questa forma lui parlare audivi : 78 O voi, che siete due dentro ad un foco, S' io meritai di voi mentre ch' io vissi, S' io meritai di voi assai o poco, HI Quando nel mondo gli alti versi scrissi, Non vi movete ; ma 1' un di voi dica Dove per lui perduto a morir gissi. 84 Lo maggior corno della fiamma antica Comincio a crollarsi mormorando, Pur come quella cui vento affatica. 87 Indi la cima qua e la menando, Come fosse la lingua che parlasse, Gitto voce di fuori, e disse : Quando 90 Mi diparti' da Circe, che sottrasse Me piu d' un anno la presso a Gaeta, Prima che si Enea la nominasse; 93 Ne dolcezza di figlio, ne la pieta Del vecchio padre, ne il debito amore, Lo qual dovea Penelope far lieta, 96 Vincer poter dentro da me 1' ardore Ch' i' ebbi a divenir del mondo esperto, E degli vizii umani e del valore : 99 78 In cotal forma e parlo (?) F. In tal maniera K. udivi DIZ. 79 dua entro I. 82 igli alti F. 84 Ove ZB. Dove e per E. amor gissi L. 87 quello P. che vento D. ch' un vento Q. fatica Z. 88 Ik e qua D. 89 una lingua F. 90 voci PC. voce fuori D. fuora P. 92 li M. la appresso a Q. Ik ap- presso Gaeta E. anno presso da K. 93 Pria che cosi I. Prima che Enea cosi la nominasse EF (nomasse). nomasse ABCFHKQa. 94 Ne donzella D. del figliuolo BE. del figlio M. fio K. di figli F. 95 il om. Q. 96 La (altered later to Lo) qual D. 97 potero ACDHP. poteo M. dentro om. L. a me AEGHPa. a me dentro Q. 98 Ch' i' ebbi di venir D. 82 igli in F does not apparently represent f 1 gli, but is merely a dialec- tic form of c //, since it appears elsewhere in this MS. in many other places. An i prefixed to all kinds of words is most common in this MS., and is I believe a Florentine peculiarity. 90 But final / and e are so often interchanged in P that this is scarcely conclusive. CANTO XXVI. 193 Ma misi me per 1' alto mare aperto Sol con un legno e con quella compagna Picciola, dalla qual non fui deserto. 102 L' un lito e 1' altro vidi infin la Spagna, Fin nel Morrocco, e 1' isola de' Sardi, E. 1' altre che quel mare intorno bagna. 105 lo e i compagni eravam vecchi e tardi, Quando venimmo a quella foce stretta, Ov' Ercole segno li suoi riguardi, 108 Acciocche 1' uom piu oltre non si metta : Dalla man destra mi lasciai Sibilia, Dall' altra gia m' avea lasciata Setta. m O frati, dissi, che per cento milia Perigli siete giunti all' occidente, A questa tanto picciola vigilia 114 De' vostri sensi, ch' e del rimanente, Non vogliate negar 1' esperienza, Diretro al sol, del mondo senza gente, 117 Considerate la vostra semenza : Fatti non foste a viver come bruti, Ma per seguir virtute e conoscenza. 120 Li miei compagni fee' io si acuti, Con questa orazion picciola, al cammino, 100 Ma. am. L. 102 Piccola GIMC>. della quale CDMQ. con la qual K. 103 lito vidi e 1' altro B. fin LM. insin IB. 104 Fin dal I. Fin al D. 105 1' altra B (corrected 2. nda manu to /' altre]. 1' altro M. 106 coi comp. O. 108 Ercules C. segnio CG. con suoi IZ. coi suoi Q. righagni D. 109 Accio- che altri piu F. piu volte K. ill Septa F. 112 frate D. fratei IQ. diss' io BC. 113 giunti siete Af. 114 E questa EH. Di questa F. Da questa L. tanta GL. piccola ACDEG^M^. 115 nostri BCDEFHILOPQZCZ?a(?). et ch' e del rim. K. di rim. ABFILOZ^a. 117 Di Ik dal sol Z. 119 fosti ADEGHIMPZ. 120 Ma asseguir D. per aver Z. . vertu C. 121 feci si FI. arguti KLPCD. abuti (?) Z. 122 queste or. picciole Q. piccola ADM. piccola orazion K. al om. L. 108 Most of the MSS. have Dove here, but this difference is not gener- ally registered, except when it is noticed in the margin of Witte's edition. D.I. 13 194 INFERNO Che appena poscia gli avrei ritenuti. 1 23 E, volta nostra poppa nel mattino, De' remi facemmo ale al folle volo, Sempre acquistando dal lato mancino. 126 Tutte le stelle gia dell' altro polo Vedea la notte, e il nostro tanto basso, Che non surgeva fuor del marin suolo. 129 Cinque volte racceso, e tante casso Lo lume era di sotto dalla luna, Poi ch' entrati eravam nell' alto passo, 132 Quando n' apparve una montagna bruna Per la distanza, e parvemi alta tanto, Quanto veduta non n' aveva alcuna. 135 Noi ci allegrammo, e tosto torno in pianto ; Che dalla nuova terra un turbo nacque, E percosse del legno il primo canto. 138 Tre volte il fe' girar con tutte 1' acque, Alia quarta levar la poppa in suso, E la prora ire in giu, com' altrui piacque, 141 Infin che il mar fu sopra noi richiuso. 123 Che per voler non gli are' poi tenuti F. tenuti AEGHKLOQa*. 124 dal mattino K. 125 Con remi D. alie F. ali ABCDGHKLMOPQZ/X a folle Q. al nostro volo Z. 126 dal alto mane. I. da lato M. 127 alto CF. 128 Veder L. e om. F. 129 Che '1 non Z. Che non fu gia fuor L. 130 era acceso intanto D. e racceso Q. raccese B (corrected 7. nda manu to racceso]. tanto ABFGLQ. 131 Lo sole era Q. Che '1 lume D. della BFGIC7?. sotto la A. 132 Poi che tratti G. all' alto F. altro ACLPQ#. 133 m' apparve BDLO. 134 e parea D. e om. I. 135 veduto DL. noi n' avamo D. non avea ABCEGHIKOPQZ^CZ'a. mai n' avesse M. 136 Poi ci D. torno tosto B. 137 della BDEFHK]VL4Z>'CZ>a. 138 Che percosse Ma 2 (obviously inserted later). di legno D. il primaio Q. 139 feo M. 140 in om.. I. 141 Poi leproran in (sic) L. proda BCPZ. plura andar giu F. ir giu DK. gy giu A. (Qy. gy=g1 ? see Inf. xxiii. 145, xx. 60.) gir giu M. 142 chiuso B. rinchiuso ACGP. inchiuso M. 132 On tratti in G see note on xxv. 1 12. CANTO VENTESIMOSETTIMO Gia era dritta in su la fiamma e queta, Per non dir piu, e gia da noi sen gia Con la licenza del dolce Poeta ; 3 Quando un' altra, che dietro a lei venia, Ne fece volger gli occhi alia sua cima, Per un confuso suon che fuor n' uscia. 6 Come il hue Cicilian' che mugghio prima Col pianto di colui (e cio fu dritto) Che 1' avea temperato con sua lima, 9 Mugghiava con la voce dell' afflitto, Si che, con tutto ch' ei fosse di rame, Pure e' pareva dal dolor trafitto : 12 Cosi per non aver via ne forame, Dal principio nel foco, in suo linguaggio Si convertivan le parole grame. 15 Ma poscia ch' ebber colto lor viaggio 2 piu alor e gia sen gia K. de noi E. 4 altro F. a noi ven. EGHLP. dietro gli M. 5 Ne fe' levare F. 6 fuor uscia A. su n' usciva L. 7 buo KQ. 8 Con mugghio IZ. 10 quella voce C. delafrito L. 12 El pur pareva G. da dolor K. del dolor Qa. 1* del fuocho AEDFlKLOQZA(marg-.) BCDo,. in fuoco E. al suo F. 15 Si convertirno I. 16 ebbe AMAC. che lor colto lor L. suo viaggio E. el lor viaggio K. 10 With delafrito in L comp. fritto in the interpolated lines in E in Canto xxxiii. 11 Most MSS. have che fosse, on which see note on xxvi. 75. 15 See note on xiii. 30. 16 L's blunder has perhaps arisen from chebbor colto, or possibly from some such orthography as that of D, which has che bercholto. 1.3 2 196 INFERNO Su per la punta, dandole quel guizzo Che dato avea la lingua in lor passaggio, 18 Udimmo dire : O tu, a cui io drizzo La voce, e che parlavi mo Lombardo, Dicendo : ista ten va, piu non t' adizzo: 11 Perch' io sia giunto forse alquanto tardo, Non t' incresca restare a parlar meco: Vedi che non incresce a me, ed ardo. 24 Se tu pur mo in questo mondo cieco Caduto sei di quella dolce terra Latina, ond' io mia colpa tutta reco, 27 Dimmi se i Romagnuoli han pace, o gucrra ; Ch' io fui de' monti la intra Urbino E il giogo di che '1 Tever si disserra. 30 Io era ingiuso ancora attento e chino, Quandcj il mio Duca m,i tento di costa, Dicendo : Parla tu, questi e Latino. 33 17 dandoli (or gli) BGMOZ?. 18 data G. avean L. al suo pass. F. a lor pass. K. 19 a cui dirizzo B 1 C 1 FKLa. 20 La lingua F. e om. CFKM. mo bis P. 21 istra ACDEHIKMOPQZ#>. istar BG. in stra A. issa I. sta L. stra ten YC. non piu B. t' adrizzo E. aizzo G. 22 Forse perch' io G. forse otn. E 3 . sia forse zunto arquanto K. 23 ti rincresca astare D, tj ricresca C, ti rincresca L. te rin- cresca star(e) MQZ?. 24 rincresce CDLQ/?. m' incresce a me M. ch' io ardo M. che ardo CDEFGIQZ. ' 26 Cad. su Q* gentil terra F. 27 onde mja CDILMZ. tutta mia colpa A. 2 C. 28 se '1 Romagnuolo ha F. 29 entro a (or ad) Orbino BDIL. intro a Q. Ik qltre a Urb. G. Ik oltra Orb. P. 30 Nel giogo I. da che I. che Tever CHKOPQ^a. che Tevero BG. di che tener E. E giogo donde Tever F. 31 in giu CQ. ancora ingiuso FM. intento EFKL^4 2 C e om. L. 32 da costa GI. 19 C has o inserted before dirizzo, and B has the first / in dirizzo erased. 31 At this point in C (it not being the end of a page) a new scribe seems to have come in. The writing is not so good and the ink much inferior. This continues to the end of the Inferno. At the be- ginning of the Purg. the former scribe seems to have recommenced with the old ink. Many of the corrections over erasures in the preceding Cantos seem to be in the handwriting and the ink which begins here. CANTO XXVII. 197 Ed io ch' avea gia pronta la risposta, Senza indugio a parlare incominciai : O anima, che se' laggiti nascosta, 36 Romagna tua non e, e non fu mai, Senza guerra ne' cor de' suoi tiranni; Ma 'n palese nessuna or vi lasciai. 39 Ravenna sta, come stata e molti anni : L' aquila da Polenta la si cova, Si che Cervia ricopre co' suoi vanni. 42 "La. terra che fe' gia la lunga prova, E de' Franceschi sanguinoso mucchio, Sotto le branche verdi si ritrova. 45 II Mastin vecchio, e il nuovo da Verrucchio, Che fecer di Montagna il mal governo, La dove soglion, fan de' denti succhio. 4 8 Le citta di Lamone e di Santerno Conduce il leoncel dal nido bianco, Che muta parte dalla state al verno; 51 34 pronta om. L. pronta gia C. 35 piu indugio C. comin- ciai CDFGKLQ [O^ had io com.]. 36 laggiu se' B. 37 ne non fu AD. n fu giammai BG. ne e ne fu mai F. 38 nel cuor FM. 39 Ma palese ABDEIKPQZ. Ma il pal. L. Ne pal. neuna M. or nessuna B. nulla or F. or ven IZB. 40 com' e stata ABCFLMQ. come stata gia E. com' stato G. come stancha (sic) a. 41 L' aguglia CEGHOa. di Polenta HI. dopolente D. 42 E anche Cervia cuopre F. con suo GMQ. con soi P. 46 E '1 Mastin BCDEGILQZ^^a. nuovo e '1 vecchio D. e nuovo GIM. dal Verrucchio FLCD. diVerr. O. del Verr. Q. 47 fece DL. fecion I. di Romagna I. del Montagna F. il om. O. mar governo (sic) H. 48 dov' ei soglion K. far(e) ADGILMPQ^a. 49 La citta ClA^-BC^Da,. di Alamone C. da Lamone a. 50 Cond. il choltel (?) L. del nido CDFIP^ 2 CZ>. dal monte b. G. 51 della state Z. aloverno (sic) E. 40 D has come stata molti with . The terzina 43 45 is omitted in H. 46 But as El may either = // or E Y, this is not con- clusive. 47 On mar for mal, cf. note on xxx. 19. 198 INFERNO E quella a cui il Savio bagna il fianco, Cos! com' ella sie' tra il piano e il monte, Tra tirannia si vive e stalo franco. 54 Ora chi sei ti prego che ne conte : Non esser duro piu ch' altri sia stato, Se il nome tuo nel mondo tegna fronte. 57 Peseta che il foco alquanto ebbe rugghiato Al modo suo, 1' acuta punta mosse Di qua, di la, e poi die cotal fiato : 60 S' io credessi che mia risposta fosse A persona che mai tornasse al mondo, Questa fiamma staria senza piu scosse : 6> Ma perciocche giammai di questo fondo Non torno vivo alcun, s' i' odo il vero, Senza tema d' infamia ti rispondo. 66 Io fui uom d' arme, e poi fui cordelliero, Credendomi, si cinto, fare ammenda : E certo il creder mio veniva intero, 69 Se non fosse il gran Prete, a cui mal prenda, Che mi rimise nelle prime colpe ; E come, e quare voglio che m' intenda. 72 Mentre ch' io forma fui d' ossa e di polpe, 52 quella cui ABCEGHKLOQisa=girisa in K at xvii. 27. 200 INFERNO Dentro Siratti a guarir della lebbre, Cosl mi chiese quest! per maestro 96 A guarir della sua superba febbre : Domandommi consiglio, ed io tacetti, Perche le sue parole parver ebbre. 99 E poi mi disse : Tuo cor non sospetti : Finor t' assolvo, e tu m' insegna fare SI come Penestrino in terra getti. 102 Lo ciel poss' io serrare e disserrare, Come tu sai ; pero son due le chiavi, Che il mio antecessor non ebbe care. 105 Allor mi pinser gli argomenti gravi La Ve il tacer mi fu avviso il peggio, E dissi : Padre, da che tu mi lavi 108 Di quel peccato, ov' io mo cader deggio, Lunga promessa con 1' attender corto Ti fara trionfar nell' alto seggio. m Francesco venne poi, com' io fui morto, Per me ; ma un de' neri Cherubini 95 Dentro dasse ( = da se?) D. Dentro a Sir. BIMa. Dentro in Sirato E. Syrapti P. della sua L. delle lebbre B*CQa. 96 questo AEP. 97 superbia (sic} B. 99 le parole sue mi parver F. 100 Poi ridisse CGHKLOP. E poi disse A. Et e' mi disse F. E poi disse E x (with re- inserted later). Poi mi rid. C. 101 Si non D. insegni a fare KMO. 102 Per che modo Pen. F. Si com' io BLa. Pilistrino B. Penestrina I. Pilestrino a. Peles- trino KL. Penestino M. a terra GM. 105 non 1' ebbe B. 106 m' inpinser (?) C. 107 Onde tacer D. Ove '1 tacer CGZ. La ove tacer EH P. Onde el I. Dove il M. Si che '1 tacer al viso mi fu il p. F. li fu L. fu m' avviso G. 108 poi che tu DF. 109 Da quel K. ove mo l~LM.Q2.BD. dove mo FK. 110 Lunghe promesse D. impromessa LQ. Ill sull' alto L. 112 venne a me GIL. poiche fu m. GL. da ch' io fu K. 113 Carubini K. 95 de lelebre (sic} C. Bj apparently had dalla. 112 The read- ing in GIL is evidently bad. The copyist who first made it did not carry his attention even as far as the next line. Shortsighted alterations of this kind are very frequent, and not seldom (as //. /.) alterations are made which take account only of the single line in which they occur. CANTO XXVII. 201 Gli disse : Nol portar; non mi far torto. 114 Venir sen dee la giu tra' miei meschini, Perche diede il consiglio frodolente, Dal quale in qua stato gli sono a' crini ; 117 Ch' assolver non si puo, chi non si pente, Ne pentere e volere insieme puossi, Per la contradizion che nol consente. 120 O me dolente ! come mi riscossi, Quando mi prese, dicendomi : Forse Tu non pensavi ch' io loico fossi ! 123 A Minos mi porto : e quegli attorse Otto volte la coda al dosso duro, E, poi che per gran rabbia la si morse, 126 Disse : Questi e de' rei del foco furo : Perch' io la dove vedi son perduto, E si vestito andando mi rancuro. 129 Quand' egli ebbe il suo dir cost compiuto, La fiamma dolorando si partio, Torcendo e dibattendo il corno acuto. 132 Noi passammo oltre, ed io e il Duca mio, Su per Io scoglio infino in sull' altr' arco Che copre il fosso, in che si paga il fio 135 A quei che scommettendo acquistan carco. 114 non portar ABCDFGHLMOPOJL4C>a. 115 se ne dee giu ABEFHKLOOJL4 2 5CZ?a. quaggiu CI. 116 Perocche F. 117 Da indi in qua li sono stato F. li son stato E. 118 Che om. F. non s' inpente K. 119 E pentere D. pentire EGP. ne volere L. 120 Per la contrarietk che non consente F. 121 Qime ABDM. com' io M. 122 Quando prese C. 123 credevi BDFIZ. che loico E. logico CHOP. chelo dicho (sic) L. 124 ed egli s' attorse D. e que' m' attorse L. 126 E poi per gran DHLOa. E poi per la I. se la morse I. la rimorse EHP. 127 E disse K. 128 Perche la CO. 129 m' inrancuro M. 130 Poscia ch' egli ebbe Z. ebbe suo DZ. cosi om. A. si compiuto Z. 132 Correndo M. al corno D. el torno (?) L. 133 oltre io ADEF 1LPZ. 134 Su om. D. in om. L. fin sopra 1' altro F. fin in M. 135 scuopre I. il focho L. ove si paga B. 136 Da quei CD. sgomentando D. che sotto mettendo L. acquista EF. 117 F's reading seems to be a reminiscence of xxv. 4. CANTO VENTESIMOTTAVO Chi poria mai pur con parole sciolte Dicer del sangue e delle piaghe appieno, Ch' i' ora vidi, per narrar piu volte ? 3 Ogni lingua per certo verria meno Per lo nostro sermone e per la mente, Ch' hanno a tanto comprender poco seno. 6 S' ei s' adunasse ancor tutta la gente, Che gia in sulla fortunata terra Di Puglia fu del suo sangue dolente 9 Per li Troiani, e per la lunga guerra Che dell' anella fe' si alte spoglie, Come Livio scrive, che non erra : r 2 Con quella che senti di colpi doglie, Per contrastare a Roberto Guiscardo, E 1' altra, il cui ossame ancor s' accoglie 15 A Ceperan, la dove fu bugiardo 1 potria D. mai om. C. pur om. AB. 3 Che ora a. innarrare D. piu oltre K. 5 vostro D. 6 Che a tanto comprehender D. Che non a tanto L. tanto a comprender B. 8 Che giace in I. fortuna terra G (clerical error). in su la 'nfortunata Z. 9 Che fe' di Puglia I. il suo sangue KL. 10 Per li Romani B*. O per Troj. D. o per la ACDEIOQ(?)Z^ 2 ^^- e om. K. 11 fer BD. fen E. fue si L. si larghe C. 12 Si come Livio I. SI come scrive Livio M. scrisse A. descrive O. 13 chonsentio D. chonsentin L. che sentio BCEGH^a. di colpe IL. 15 E quella M. E 1' altre O. si coglie A. 16 la om. M. 7 The scribe in G seems liable to miss the last syllable of words occasionally. Cf. fortuna for fortunata again in xxxi. 115, and Arde for Ardere in xxix. 1 17. CANTO XXVIII. 203 Ciascun Pugliese, e la da Tagliacozzo Ove senz' arme vinse il vecchio Alardo : 18 E qual forato suo membro, e qual mozzo Mostrasse, da equar sarebbe nulla Al modo della nona bolgia sozzo. 21 Gia veggia per mezzul perdere o lulla, Com' io vidi un, cosl non si pertugia, Rotto dal mento infin dove si trulla: 24 Tra le gambe pendevan le minugia; La corata pareva, e il tristo sacco Che merda fa di quel che si trangugia. 27 Mentre che tutto in lui veder m' attacco, Guardommi, e con le man s' aperse il petto, Dicendo : Or vedi come io mi dilacco : 30 Vedi come storpiato e Maometto. Dinanzi a me sen va piangendo All Fesso nel volto dal mento al ciuffetto : 33 E tutti gli altri, che tu vedi qui, 17 Ogni Pugl. I. Puiese A. Taialchozzo A. e om. LM. 19 Qual suo membro forato F. 20 Mostrasse adiequar D. ad adequar I. dagualgliar a. 21 II modo ABEGHIKMPZZ?a. Che el modo Q. El mondo L. della nuova Z. nova K. della bolgia nona E. che la nona C. bolgia il puzzo D. 22 veggio L. 23 non si om. K. 24 Rotta L. Fesso FOZ>. dal mezzo Z. fin AM. insin B. turla AF. s' intrulla K. 25 pendeva la DLZ2. gli pendea A. li pendevan B. 26 ap- parea I. e om. TLGA. 2 C3,. 27 Che stereo D. 28 che lui veder tutto D. che in lui veder tutto I. 29 man aperse LD. 30 Dicendo vedi FLOZ. come qui D. come mi FL. 31 strop- piato AF. scoppiato BIC/}. scorpiato P. e om. M. 32 a me piangendo li (sic) L. 33 Rotto O. dal mento nel volto D. 34 vedi cosie D (see 1. 36). 21 nova and nona are interchanged in a passage of some critical in- terest at Purg. xxvii. 4. 24 K seems to have a tendency to introduce compounds with in. Comp. s 1 inpente in xxvii. 118 ; s' inchiude xxx. 27, &c. 26 Witte's 4to. edition reads curata, and the 8vo. edition corata, but the difference seems hardly worth registering. 31 Z has 'al. scoppiato'' written above i cta manu, and similarly B (marg.\ 33 Comp. Rotto and Fesso interchanged in 1. 24. 204 INFERNO Seminator di scandalo e di scisma Fur vivi ; e per6 son fessi cosi. 36 Un diavolo e qua dietro che n' accisma Si crudelmente, al taglio della spada Rimettendo ciascun di questa risma, 39 Ouando avem volta la dolente strada ; Perocche le ferite son richiuse Prima ch' altri dinanzi gli rivada. 42 Ma tu chi se' che in sullo scoglio muse, Forse per indugiar d' ire alia pena, Ch' e giudicata in sulle tue accuse ? 45 Ne morte il giunse ancor, ne colpa il mena, Rispose il mio Maestro, a tormentarlo ; Ma per dar lui esperienza plena, 48 A me, che morto son, convien menarlo Per lo inferno quaggiu di giro in giro : E questo e ver cosi com' io ti parlo. 51 Piu fur di cento che, quando 1' udiro, S' arrestaron nel fosso a riguardarmi, Per maraviglia obbliando il martiro. 54 Or di' a Fra Dolcin dunque che s' armi, Tu che forse vedrai lo sole in breve, S' egli non vuol qui tosto seguitarmi, $ 35 scandali DEFIM. 36 Fin ch' i for vivi, perb son fessi qui A. Fur tutti Z. Fur ivi C. vivi poi son L. e om. FIZ. cosie D. fuor ffessy cosyy (sic) K. 37 Un dimonio D. Un diavol qua K. diavolo che qua L. qui dentro A. qua dentro G. qua entro D. qui dietro C. ci ascisma D. n' ascisma I CD. 38 al taio A. a tagli K. d' una spada D. 39 Che fa tremar ciascun M. 40 avia D. volto I. 41 ferute E. rinchiuse M. 42 davanti BD. innanzi FP. la rivada CKL. 43 qual se' M. 44 indugiar ire IK. 45 giud. gia in sulle Z. 46 Ne 'n morte B. 47 il om. L. Maestro mio E. 48 dar a lui F. 50 Per questo inferno giu F. qui zu K. 51 E cio e D. vero si com' io BE. com' io cosi C. 63 del fosso D. 54 obriando D [see note on xxv. 112]. 55 frate BL. 56 Che tu D. vederay forse E. vedrai il sole forse O. O tu che '1 sol forse vederai in breve Z. un brieve E. 57 Se- deno ( = Sed ei non. See note on xxv. 69) D. asseguitarmi L. CANTO XXVIII. 205 Si di vivanda, che stretta di neve Non rechi la vittoria al Noarese, Ch' altrimenti acquistar non saria lieve. 60 Poi che 1' un pie per girsene sospese, Maometto mi disse esta parola, Indi a partirsi in terra lo distese. 63 Un altro, che forata avea la gola E tronco il naso infin sotto le ciglia, E non avea ma' ch' un' orecchia sola, 66 Restate a riguardar per maraviglia Con gli altri, innanzi agli altri aprl la canna Ch' era di fuor d' ogni parte vermiglia; 69 E disse : Tu, cui colpa non condanna, E cui io vidi su in terra Latina, Se troppa simiglianza non m' inganna, 72 Rimembriti di Pier da Medicina, Se mai torni a veder lo dolce piano, Che da Vercelli a Marcabo dichina. 75 E fa saper ai due miglior di Fano, A messer Guido ed anco ad Angiojello Che, se 1' antiveder qui non e vano, 78 Gittati saran fuor di lor vasello, 58 vivandi D. che e stretta E. da neve I, 59 Navarese AM. Novarese GQ. el Noarese K. alnon aresse (sic) L. 60 Che om. Q. saria breve D. 61 uno. piede per girsi D, pie per girse E, 62 questa E. 63 Poi a F. partire BD. partirlo CI. la dist. D. 64 Et un altro EGH. forato ABCDEHILMOPZ^/?, 65 fin AEMO. insin G. 66 ma' om. 7.D. che om. FI. avea piu d' una M. 67 Ristati D. Ristato ABCFGHKOPQZCZ?a. a guardar A. a veder F. 68 e 'nanzi apri D. ad altri D. 69 da ogni A. 70 O tu DEFHP. 71 E ch' esso vidi D. vidi gia in IB. vidivi su terra E. in su terra BCDFGHKMOPQZ^a. in su la terra A. io om. L. in om. L. in sua terra A. 73 Rimembrati ADIKM. Ricorditi G. di Med. I. 74 ritorni E. da vederli L. 75 Vercello B. Mardabo Z. declina Z. 76 a saper AE. a du' gli migl. A. a duo E. ad due Q. da Fano CEHMPZ?a. 77 anche CDFIKL.O. 79 Saettati D. del lor D. 206 INFERNO E mazzerati presso alia Cattolica, Per tradimento d' un tiranno fello. 81 Tra 1' isola di Cipro e di Maiolica Non vide mai si gran fallo Nettuno, Non da pirati, non da gente Argolica. 84 Quel traditor che vede pur con 1' uno, E tien la terra, che tal e qui meco, Vorrebbe di vedere esser digiuno, 87 Fara venirli a parlamento seco ; Poi fara s\, che al vento di Focara Non fara lor mestier voto ne preco. 90 Ed io a lui : Dimostrami e dichiara, Se vuoi ch' io porti su di te novella, Chi e colui dalla veduta amara. 93 Allor pose la mano alia mascella D' un suo compagno, e la bocca gli aperse Gridando : Questi e desso, e non favella : 96 80 macerati (0r mac.er.) ADEIKMOPQZa. dentro alia I. presso la P. 82 Cipri ABCDEGHIKLMOPQZCZ?a. Maricha F. Maiorica Z. 83 mai om. Q. mai cotal ZB. falli M. mai tafallo F. nessuno DF. niuno L. 84 Non di Ka. Non da partire L. pirate ACHKOPOZ^^a. pirati e non F. ne da G. di gente K. 85 piu P. 86 tal e reggimento (sic) L. 87 vederla DIA^ 88 meco D. seco om. L. 89 E fara D. che '1 vento I. di storchata (?) L. 90 Non gli fara mestiero DG. sara ACMZa. ne voto E. 91 mostrami e dichiarata L. 93 della BCDIL>. 94 porse AK. 95 e oin. L. 96 Digando K. Dicendo P. E disse M. e esso ABCGHIKLOPQa. desso om. E. e om. L. 80 Probably this is only a variation in orthography for mazzerati. 84 The spelling pirate is not conclusive, as final e and i are frequently interchanged in some MSS. The etymology given here for pirate by Anon. Fior. is too good to be lost: "da Pir quod est ignis,. ..da quello fuoco che si facea ne' falsi porti per rubatori" ! 92 'E' combining his dislike to assimilated forms with his Venetian carelessness about doubled letters has written dicte for diite, i.e. di te. Comp. liecti for lieti in xiii. 69, and ghibecto xiii. 151, besides such forms as lucta, alocta, decto, &c. passim. CANTO XXVIII. 207 Questi, scacciato, il dubitar sommerse In Cesare, affermando che il fornito Sempre con danno 1' attender sofferse. 99 O quanto mi pareva sbigottito Con la lingua tagliata nella strozza, Curio, ch' a dire fu cosi ardito ! 102 Ed un ch' avea 1' una e 1' altra man mozza, Levando i moncherin per 1' aura fosca, SI che il sangue facea la faccia sozza, 105 Grido: Ricordera' ti anche del Mosca, Che dissi, lasso ! Capo ha cosa fatta, Che fu il mal seme per la gente tosca. icS Ed io gli aggiunsi : E morte di tua schiatta ; Perch' egli accumulando duol con duolo, Sen gi'o come persona trista e matta. m Ma io rimasi a riguardar lo stuolo, E vidi cosa ch' io avrei paura, Senza piu prova, di contarla solo; u 4 Se non che coscienza mi assicura, La buona compagnia che 1' uom francheggia Sotto 1' asbergo del sentirsi pura. 117 Io vidi certo, ed ancor par ch' io '1 veggia, 97 cacciato A. al dub. DK. il om. L. 98 Cesaro A. affermato che forn. L. che forn. M. 99 Sempre conduno (sic) E. all' attender G. 101 Colla lunga Z (obvious error and corrected above 2a, del monte L. 129 appressarme Z. 130 Che furno I. 131 O tu che vai si riguardando D. sperando L. i om. E. 132 Vedendo M. granda AK. grando G. grave I. 133 novelle CGHIKOPQ^ 2 C 134 Sacci (?) G. Beltram CDEFG HIKMOPQZ/2Z?a. del Borm. CA. di Born. Q. Boxnio (sic) H. Beltramo dal borgho L. quelli om. L, 135 Que' che diedi a re L. are P. re JohT B. giovane A. guane K. i mai conf. CDFGKLZ. i mai M. mai conf. E. e mia conf. I. 136 E feci DK. e figli D. e figli esser rub. M. figliuolo L. figliol E. fio K. rubelli LMZ. 137 Aghito fe' non L. Abso- loneA. Absalone BCEHOQZ/4 2 ^. Assalone GIMa. 130 On the form furno cf. note on xiii. 30. 135 e mia is I's regular way of writing i miei. CANTO XXVIII. 2O9 E di David co' malvagi pungelli. 138 Perch' io partii cosl giunte persone, Partito porto il mio cerebro, lasso ! Dal suo principle ch' e in questo troncone. 141 Cosl s' osserva in me lo contrapasso. 138 N di Dav. ZB. con malv. GILMOQ. con unalgi (sic) E. punzelli AKCGHOPAfia.. poncelli E. puntegli IQ. polzeli K. puntelli LZ. 139 si congiunte DI. 140 Partito posso L. celabro DFLa. celebro EK. basso L. 141 princ. con questo DFGQ. in . 89 qua entro CD. qui a centre E. qui dentro P. qui entro A!. unghie DF. e se 1' unghia Q. 92 dicendo 1' un D. 93 tu om. L. addemandasti Q x (with one d erased later). 94 Lo duca F. io om. Q. 95 in giu Q. 96 E mostrare illonferno F. a om. L. 98 E ciaschedun tremando F. a me ciascuno Q. 99 Cogli altri ABEOZ?. CANTO XXIX. 215 Lo buon Maestro a me tutto s' accolse, Dicendo : Di' a lor cio che tu vuoli. Ed io incominciai, poscia ch' ei volse: 101 Se la vostra memoria non' s' imboli Nel primo mondo dall' umane menti, Ma s' ella viva sotto molti soli, 105 Ditemi chi voi siete e di che genti : La vostra sconcia e fastidiosa pena Di palesarvi a me non vi spaventi. 108 Io fui d' Arezzo, ed Albero da Siena, Rispose 1' un, mi fe' mettere al foco ; Ma quel perch' io mori' qui non mi mena. m Ver e ch' io dissi a lui, parlando a gioco, Io mi saprei levar per 1' acre a volo : E quei che avea vaghezza e senno poco, 114 Voile ch' io gli mostrassi 1' arte ; e solo Perch' io nol feci Dedalo, mi fece Ardere a tal, che 1' avea per figliuolo. 117 Ma nell' ultima bolgia delle diece Me per alchimia che nel mondo usai, Danno Minos, a cui fallar non lece. 120 Ed io dissi al Poeta : Or fu giammai Gente si vana come la sanese ? 102 Ed io com. AC. posa (sic) E. 103 involi DFGIMZ?. raboli (?) K. 104 delle AEHLOP. genti G. 106 Ditene AEGHP. Dimmi L. 107 e om. L. 108 Da pal. G. palesarmi a me Q. si spaventi B. paventi G. 109 da Regio A. Alberto BD. di Siena KQ. 110 Rispose mi fe' A. Rispose lui I. ill perchui E. perque mori' L. 112 che disse E. dissi lui M. 114 E quillo avea Q. sonno LC. 115 Volse BCELP. e om. L. 116 e' mi fece I. 117 Arde (cler. error) G. ch' io 1' avea DM. 118 Me nell' Q. 119 Me om. Q. per 1' archimia DKZ. per 1' alchimia EIMOQ^Ca. per la chimia L. ch' io BD. 120 Minossi F. a cui peccar C. 121 or om. L. 122 piu vana D. come e E. senese 109 With da Regio = (f Areszo comp. batigiatori in some MSS. at xix. 1 8. 117 On arde cf. note on xxviii. 7. 216 INFERNO Certo non la francesca si d' assai. 113 Onde 1' altro lebbroso die m' intese, Rispose al detto mio : Trammene Stricca, Che seppe far le temperate spese ; 1 16 E Niccolo, che la costuma ricca Del garofano prima discoperse Nell' orto, dove tal seme s' appicca ; 1 19 E tranne la brigata, in che disperse Caccia d' Ascian la vigna e la gran fronda, E 1' Abbagliato il suo senno proferse. 131 Ma perche sappi chi si ti seconda Contra i Sanesi, aguzza ver me 1' occhio Si, che la faccia mia ben ti risponda; 135 Si vedrai ch' io son 1' ombra di Capocchio, Che falsai li metalli con alchimia, E ti dei ricordar, se ben t' adocchio, 138 Com' io fui di natura buona scimia. 123 si da assai C*. 124 Allor 1' altro C. leproso EM. 125 al duca mio Z. Tramme Io Stricca D. Trane Io St. I. tramino e stricha L. 127 della costuma F. 128 la schoperse L. 130 E om. Q. 131 fonda EHZ. 132 E 1' Abbl. suo ABCFGHIKL MOPQZ. Alabagliato senno suo E. coperse C. 133 si chi F. ch' io si IK. ti nasconda Z. 134 Contra S. CDEHKMOQZ.5C. Contro a San. FI. Senesi EKP^?. Contro San. G. Chotai sensi (sic) L. in ver me Q. 136 Si om. GK. 137 il metallo Q. archimia DGZ. per alch. L. con 1' alchimia MOP. 138 Ben ti dee D. Et tee de (sic] B. E ten dee CFLP. E te dee ben A. E tee dee (sic) G. Et (El 2* r manu) te de K. s' io ben DO. ben aocchio G. ben adocchio K. 139 Come fui L. 131 Z has ' al. fronda'' written over. CANTO TRENTESIMO Nel tempo che Giunone era crucciata Per Semele contra il sangue tebano, Come mostro una ed altra fiata, 3 Atamante divenne tanto insano, Che veggendo la moglie con due figli Andar carcata da ciascuna mano, 6 Grido : Tendiam le reti, si ch' io pigli La leonessa e i leoncini al varco : E poi distese i dispietati artigli, 9 Prendendo 1' un che avea nome Learco, E rotollo, e percosselo ad un sasso ; E quella s' annego con 1' altro carco. n E quando la fortuna volse in basso L' altezza de' Troian che tutto ardiva, SI che insieme col regno il re fu casso ; 15 Ecuba trista misera e cattiva, Poscia che vide Polissena morta, 1 crucciato G. 2 contra a sangue L. 3 Com' el M. di- mostro I. mostrone P. 1' una e 1' altra FLM. e 1' altra K. 5 vedendo FMP. co' duo R. 6 Venir ~DZA(marg.}B. car- cato A. di cias. D. 7 Tend, le reti grido E. le rete D. 8 et leoncini M. 9 E distendendo M. desperate E. 10 Prese 1' uno M. 11 rotolollo C. 2nd 'e' otn. KL. in un D. 12 quella sane ando (sic} L. . alto E. incarco FILQC. 15 Fin che I. Che 'nsieme col reame F. e regno co re L. 16 misera trista D. misera tapina L e out. Q. 17 Alor che vide M. 2l8 INFERNO E del suo Polidoro in sulla riva 18 Del mar si fu la dolorosa accorta, Forsennata latro si come cane ; Tanto il dolor le fe' la mente torta. 21 Ma ne di Tebe furie ne Troiane Si vider mai in alcun tanto crude, Non punger bestie, non che membra umane, 74 Quant' io vidi in due ombre smorte e nude, Che mordendo correvan di quel modo Che il porco quando del porcil si schiude. 27 L' una giunse a Capocchio, ed in sul nodo Del collo 1' assanno si che, tirando, Grattar gli fece il ventre al fondo sodo. 30 E 1' Aretin, che rimase tremando, Mi dissc : Quel folletto e Gianni Schicchi, E va rabbioso altrui cosl conciando. 33 O, diss' io lui, se 1' altro non ti ficchi Li denti addosso, non ti sia fatica A dir chi e, pria che di qui si spicchi. 36 18 om. L. E bel suo A. E '1 bel suo GHP. E '1 suo bel Pol. DE. 19 Del mal P. 20 E forsennata A. Farse innata D. Fortunata K. si om, DE. 21 Tanto dolor IM. la fe' C. gli fe' GMP. li fe' KLOQ. 22 Mai ne FK. Ma non A\ 23 tante crude G. 24 In punger DMC non con L. membre AQ- 25 Quand' io A. vidi due FlMZff. 26 Che fuggendo cor- reva M. a quel L. 27 Che porco GI. dal M. da porcil I. s' inchiude K. 28 ed om, FL. 29 assanna L. 30 il om, L. fece in fino al M. a fondo K. 31 tirando B. 32 follechio D. Vanni A. Ziani Schiochi K. Gian di Schicchi L. 33 Che va DF. cacciando DGL. 34 O diss' io se F. diss' io a lui B. dissi a lui IO. dissi lui ZBC. se altro M. ffiochi K. 35 Li unghioni B. 36 Dir chi e F. A dir qual e P. A dirmi prima I. A dar A 2 . prima KQ. mi sficchi D. ti spicchi Zj (si 2 nda matnt], 18 Observe the gradual alteration of the text : (i) E del suo, (2) E bel suo, (3) E '1 bel suo, (4) E '1 suo bel. (See notes later.) 19 Com- pare mal and mar again interchanged in xxvii. 47 and xxxiv. 123. In Par. xxii. 95 the variants mal and mar are both well supported by MSS. though the former is obviously a blunder. ' P ' has also gualdar for guardar inf. 1. 48. Comp. arquanto in K at xxvii. 22. 32 On Vanni for Gianni cf. note on xxv. 67. CANTO XXX. 219 Ed egli a me: Quell' e 1' anima antica Di Mirra scellerata, che divenne Al padre, fuor del dritto amore, arnica. 39 Questa a peccar con esso cosl venne, Falsificando se in altrui forma, Come 1' altro, che la sen va, sostenne, 42 Per guadagnar la donna della torma, Falsificare in se Buoso Donati, Testando, e dando al testainento norma. 45 E poi che i due rabbiosi fur passati, Sopra cu' io avea 1' occhio tenuto, Rivolsilo a guardar gli altri mal nati. 48 Io vidi un, fatto a guisa di liuto, Pur ch' egH avesse avuta 1' anguinaia Tronca dal lato, che 1' uomo ha forcuto. 51 La grave idropisi, che si dispaia Le membra con 1' umor che mal converte, Che il viso non risponde alia ventraia, 54 Faceva a lui tener le labbra aperte, Come 1' etico fa, che per la sete 37 quell' anima e anticha D. 38 Mirti K. 39 Del padre M. fuori del diritto arnica D. 40 col padre cosi I. si convenne DF. 41 in altra FK. 42 che 'n la sen va BGI. che va in la M. che sen va Ik Q (but che and Id were omitted in Q t and are added 2*^ manu\ 44 Per falsificar A. Falsificando CD P. Falsified M. 47 Sopra quali A. 2 C. Colui sopra cui F. Sopra a colui cui aveva L. Sop. cui aveva A. 48 Rivolsemi L. Mi volsi a riguardar I. Volsilo a veder F. gualdar (sic] P. malati DP. ammalati C*Q. 49amododiL. leuto BCHLMOP^a. lautto K. del tuto E. 50 avuto BEIMZ. 51 Tronco di Ik K. dall' altro ADEHLPCa. dall' alto A. 2 . (d)ove 1' uomo e IK. che 1' uomo e forcuto FMQZ. 52 La grande I. La grieve P. La quale M. Le grave a. chelli dispaia FL. 53 per 1' umor F. che 1' amor K. che '1 mal DFKMC. 54 Che il volto B. 55 Facie lui E. Fac. lui BCDHLMPZ^a. Fac. all' un O. labbre AK. 56 per la gran sete G. che lapresente(I) L. 52 Dionisi maintains (Anedd. ii. p. 105) that idropesl is the ancient and not a syncopated form from idropesis. So Genesi\vA. xi. 107, Creti (from Curetis) Inf. xii. 12, Tanai xxxii. 27, &c. \Ve might add Tamest in xii. 120. 220 INFERNO L' un verso il mento e 1' altro in su rinvcrtc. 57 O voi, che senza alcuna pena siete (E non so io perche) nel mondo gramo, Diss' egli a noi, guardate ed attendete f>o Alia miseria del maestro Adamo : Io ebbi, vivo, assai di quel ch' io volli, Ed ora, lasso ! un gocciol d' acqua bramo. 63 Li ruscelletti, che dei verdi colli Del Casentin discendon giuso in Arno, Facendo i lor canali freddi e molli, 66 Sempre mi stanno innanzi, e non indarno ; Che 1' imagine lor vie piu m' asciuga, Che il male ond' io nel volto mi discarno. 69 La rigida giustizia, che mi fruga, Tragge cagion del loco ov' io peccai, A metter piu li miei sospiri in fuga. 72 Ivi e Romena, la dov' io falsai La lega sigillata del Batista, Perch' io il corpo su arso lasciai. 75 Ma s' io vedessi qui 1' anima trista Di Guido, o d' Alessandro, o di lor frate, 57 riverte ABCDEFGIKLMOPQZCa. 68 se senza K. pena alcuna D. niuna pena L. 59 son io F. 62 che volli E. 63 goccio D. 64'Gliu (?) russcielleti E. da ( = dai ?) verdi A. de ( = dei ?) DEIQ. di FGKLOZa. 65 Di Cas. Q. scendono DK. che scendon C. giu in E. 66 Fac. lor D. in lor M. i suo can. G. e freddi DEHQA^C. 67 non e ind. DFGL. 68 via piu ABKLMP/?. me sciuga M. 69 dond' io Z. onde nel viso I. del volto M. mal di che nel volto F. 70 che gli fruga F. che m' asciuga L. 71 da luogo K. ond' io CKL. 72 pensieri DI. gli m. sosp. piu E. 73 La sie (sic) D. Ik dove f. C. la ov' io D. 74 Le lega KZ. La leggie L. La lege M. suggellata ABCDEFGHIMOQZ^Ca. sogellata P. suggellate K. sagiela (? sugella) L. 75 col corpo L. suso DIPZ. suo E. 77 e d' Alex, o di E. e d' Aless. e di IQ. o Alexandro Z. e di lor G. 73 sie in D is perhaps for sie'siede as sup. in xxvii. 53. Also comp. Siede la terra, &c. in v. 97. CANTO XXX. 221 Per fontc Branda non darei la vista. 78 Dentro c' e 1' una gia, se 1' arrabbiate Ombre che van dintorno dicon vero : Ma che mi val, ch' ho le membra legate ? 81 S' io fossi pur di tanto ancor leggiero, Ch' io potessi in cent' anni andare un' oncia, Io sarei messo gia per Io sentiero, 84 Cercando lui tra questa gente sconcia, Con tutto ch' ella volge undici miglia, E men d' un mezzo di traverse non ci ha. 87 Io son per lor tra si fatta famiglia : Ei m' indussero a battere i fiorini, Che avean ben tre carati di mondiglia. 90 Ed io a lui : Chi son li due tapini, Che fuman come man bagnate il verno, Giacendo stretti a' tuoi destri confini ? 93 Qui li trovai, e poi volta non dierno, Rispose, quand' io piovvi in questo greppo, E non credo che dieno in sempiterno. 96 78 Brandi ABCa. Blanda A. fronte Branda (cler. error?) Z. non ti darei E. 79 se 1' altra abiate (sic) K. le rabbiate MO. 80 vanno int. ACDEGHIKOPZ^a. a torno BF. 81 ch' i' ho AEGHP. colle membra KMZ. 82 E s' io fossi di tanto D. S' io fossi da tanto pur legg. F. anchor di tanto pur AKMQ. ancora pur di tanto L. S' io fossi tanto pur anchor a. legato (cler. error from 6fjLoioTf\evra) E. 83 Che pot. G. trent' anni L. 84 Io mi sar. gik messo pel A$). Messo mi sarei D. mosso CFIQ. gia mosso F. gia messo L. 85 fra IP. per la valle sconcia F. 86 volga DI. gii-i Q. che la sia L. dodici D. 87 E men di mezzo AGM. E piu d' un mezzo ZA(marg.) B. per traverso AD. e' non ci ha C. 89 Ed ei m' indusser DL. Che m' ind. FI. Perche m' indusse M. 90 ben om. ABCDEFGHIKLMOPQZ^ 2 5Ca. carate GHKOPQZ^a. 91 taupini BCM#. topini O. 92 bagnata ABCDEGHO^a. bagnato L. 93 stricti a tuoi stricti (sic) Q. 94 trovammo D. non denno Q. 95 quando piovvi CLJ?a. pioveti (sic) E. 96 Ne non credo Z. deano YB. diano L. 81 Q has (in obvious cler. error) Mi cole val. 83 D has oca i.e. ftitcia, so schonca and non c'a in 85 and 87. See note below on 1. 148. 86 D's reading dodici is curious. (Qy. how to be accounted for?) 222 INFERNO L' una e la falsa che accuse Joseppo ; L' altro e il falso Sinon greco da Troia : Per febbre acuta gittan tanto leppo. 99 E 1' un di lor, che si reco a noia Forse d' esser nomato si oscuro, Col pugno gli percosse 1' epa croia : 102 Quella sono, come fosse un tamburo: E mastro Adamo gli percosse il volto Col braccio suo che non parve men duro, 105 Dicendo a lui : Ancor che mi sia tolto Lo mover, per le membra che son gravi, Ho io il braccio a tal mestiere sciolto. 108 Ond' ei rispose : Quando tu andavi Al foco, non 1' avei tu cosi presto ; Ma si e piu 1' avei quando coniavi. m E 1' idropico : Tu di' ver di questo ; Ma tu non fosti si ver testimonio, La Ve del ver a Troia fosti richiesto. n 4 S' io dissi '1 falso, tu falsasti il conio, Disse Sinone, e son qui per un fallo, 97 L' uno a. e om. G. Yosepo A. Giuseppe BDGHLOQZ^. 98 L' altra C. E 1' altr' e L. Simon M. credo da Troia Z. di Troglia A. di Troia CDGKMOPQa. 99 getta CD. gita L. 100 Et un AO. che '1 si A. sarecho (? = s' arreco) D. sarico Q. noglia A. 101 Forte A(marg.}C. 102 gli coperse D. croglia A. 104 E '1 maestro KQZ. si percosse A. 105 Col pugno EDFGA^C. suo om, L. 106 Dicendo ancora kemisi tolto (sic) C. 107 delle membra I. 108 Si abbo il D. a tal il braco a tal (sic) E. disciolto EGHMPOZ^a. 109 Onde risp. AKL. Ed ei DF. 110 tu om. IM. tu si presto FK. Ill Ma cosi o piu F. 112 E '1 ritruopico F. E lidropo K. Ello istropiccio e tu L. di' ben ver BIQ. 114 A Troia quando fosti del ver D. Lk dove E. Dove del ver MZ. Ove del ver fosti CGO. Quando del ver fosti B. Quando a Troia del ver F. fosti a Troia AEHKMPQZ^ 2 Ca. a Troia del ver I. 115 Io dissi '1 falso G. dissi falso ACDEMO PQZ^Ca. B* has dissi '1. e tu ABDEFGHIKLMOPQZ^a. 116 e qui son EH. io son qui MQ. 100 So D has ;;/' arrisso for mi risso in 1. 132, and s' aggira for si giro, in 1. 135. CANTO XXX. 223 E tu per piu che alcun altro demonic. 117 Ricorditi, spergiuro, del cavallo, Rispose quel ch' avea enfiata 1' epa ; E siati reo, che tutto il mondo sallo. 120 E te sia rea la sete onde ti crepa, Disse il Greco, la lingua, e 1' acqua marcia Che il ventre innanzi gli occhi si t' assiepa. 123 Allora il monetier : Cos! si squarcia La bocca tua per suo mal come suole; Che s' i' ho sete, ed umor mi rinfarcia, 126 Tu hai 1' arsura, e il capo che ti duole, E per leccar lo specchio di Narcisso, Non vorresti a invitar molte parole. 129 Ad ascoltarli er' io del tutto fisso, Quando il Maestro mi disse : Or pur mira, Che per poco e che teco non mi risso. 132 Ouand' io '1 senti' a me parlar con ira, Volsimi verso lui con tal vergogna, Ch' ancor per la memoria mi si gira. 135 E quale e quei che suo dannaggio sogna, Che sognando desidera sognare, 117 che niuno DL. 118 Ricordati AIMa. lo spergiuro IL. 119 quei BCDFLMOQZ^a. 121 E a te FGLM. reo DQa. che ti A. 123 Che nel D. El ventre F. agli occhi ABCDFG HIKLMOQZCa. ti s' assiepa BCDFGILMPCa. si te siepa E. 125 La lingua A. per mal dir UA(marg.}. tuo mal ABCEHKLPQC per lo mal G. per dir mal I O. a dir mal F. come si suole D. 126 Che se ho sete umor I. e tumor D. che humor M. e ancor mi K. 127 Tu hai la sete G. arsura al capo L. 129 ver- resti G. avresti Q. a vitar queste L. troppe parole Z. 130 io era molto D. al tutto Q. 131 mi om. AL. o pur CFQ. or om. 1. 132 Ch' e per poco che teco non piu B. Che per poco con teco O. Che per poco e teco a. e om. CDZ (first ' che'' = cK % probably). non m' arrisso DQ. non m' adisso F. 133 Quand' io senti' D. Quando '1 senti' FGIKLQ X . senti' parlar cosi con Q. 134 ver lui F. inverse L. 135 s' aggira DEGL. 136 QuaP e quel Q. Quale e colui EFGHP. che '1 suo G. che gran K. dammaggio BK/4. 137 Cognosanclo E. E sognando F. 224 INFERNO Si che quel ch' e, come non fosse, agogna ; 138 Tal mi fee' io, non potendo parlare, Che desiava scusarmi, e scusava Me tuttavia, e nol mi credea fare. u Maggior difetto men vergogna lava, Disse il Maestro, che il tuo non e stato ; Per6 d' ogni tristizia ti disgrava : 144 E fa ragion ch' io ti sia sempre allato, Se piu avvien che fortuna t' accoglia, Ove sia gente in simigliante piato; 147 Che voler cio udire e bassa voglia. 138 Si che non & come se fosse D. E quello che e F. ch' 6 om. L. Si che quel ch' e come e' non (sic) M. 139 Cotal fee' io D. possendo ACEFGHKMOPZ^Ca. 140 Che desiando D. 141 non me'l DI. paria fare D. 146 averrk D. t' aghognia L. ti toglia PM. 147 sien genti ABCDHOPQZ/^tfCa. sian gente EM. sien gente IL. sian genti K. a sim. D. 148 Che cio volere F. cio vedere C. ebbi gran voglia L. 148 D has here the curious orthography cb for do. Besides the instances above at lines 83, &c., there are many others, such as posca xxix. 102 ; marca xxx. 122 ; guanca, &c. xxxi. 2, 4, 6 ; fasco xxxi. 135 ; usco xxxiii. 46, &c. Sometimes g is similarly treated : e.g. git xxxi. 48 ; galla xxxiv. 43; gunto xxxiv. 112, &c. By a curious perversity ch is sometimes made soft, as if =ci, e.g.fascfia xxxiii. 92 ; deficho xxxiv. 7. CANTO TRENTESIMOPRIMO Una medesma lingua pria mi morse, Si che mi tinse 1' una e 1' altra guancia, E poi la medicina mi riporse. . 3 Cosl od' io, che soleva la lancia D' Achille e del suo padre esser cagione Prima di trista e poi di buona mancia. 6 Noi demmo il dosso al misero vallone Su per la ripa che il cinge dintorno, Attraversando senza alcun sermone. 9 Quivi era men che notte e men che giorno, SI che il viso m' andava innanzi poco : Ma io senti' sonare un alto corno, 12 Tanto ch' avrebbe ogni tuon fatto fioco, Che, contra se la sua via seguitando, Dirizzo gli occhi miei tutti ad un loco : 15 Dopo la dolorosa rotta, quando Carlo Magno perde la santa gesta, 3 vi riporse M. 4 ond' io L. solea far L. 5 e om. L. de suo K. a se 'cagione L. 7 Poi demmo D. 8 che cinge EKQ. cinghia I. 9 Attraversato HP. 10 Qui n' era D. Qui v' era AEHIP. 11 un poco BJ^G. 12 Allor senti' sonar un corno D. Ma senti' rison. C. io udi' A (marg!). sentiva L. altro CGLACa.. 13 ogni suono L. 14 E contro a D. Che contro a AGI. Che incontro a F. contro a. via om. L. 10 H has Quivem, which may be Quivi era or Qui if era. 11 ;z afterwards erased in B. 12 alto seems to have been altered from altro in I. D. I. 15 226 INFERNO Non sono si terribilmente Orlando. 18 Poco portai in la volta la testa, Che mi parve veder molte alte torri; Ond' io: Maestro, di', che terra e questa ? i\ Ed egli a me : Per6 che tu trascorri Per le tenebre troppo dalla lungi, Avvien che poi nel 'maginare aborri. 24 Tu vedrai ben, se tu la ti congiungi, Quanto il senso s' inganna di lontano : Pero alquanto piu te stesso pungi. 27 Poi caramente mi prese per mano, E disse: Pria che noi siam piu avanti, Acciocche il fatto men ti paia strano, 30 Sappi che non son torri, ma giganti, E son nel pozzo intorno dalla ripa Dall' umbilico in giuso tutti e quanti. 33 Come, quando la nebbia si dissipa, Lo sguardo a poco a poco raffigura Cio che cela il vapor che 1' acre stipa : 36 Cosl forando 1' aura grossa e scura, Piu e piu appressando in ver la sponda, 19 Pero portai H. 20 Ch' a me a. alti Da. altre GZ. 23 molto dalla E. 24 nel mangiar L. poi imaginando C. 25 se ben la tu ti E. se Ik tu ti AD. la tutti G. 26 senno ACE. t' inganna DL. se inganno E. da lontano KM. 27 Pero te stesso alquanto pungi F. piagni L. 28 chiaramente C. 29 Dicendo F. prima E. 30 non ti paia E. 32 dentro dalla F. della ripa GIKL. 33 Dell' umb. M. Dal bellicho in giu son tutti quanti DQ. Dal bellicho in giu stan tutti quanti F. Dal bolicollo in zusso K. unbellico Ga. umbellicolo O. in giu O 2 . 34 Che come L. distipa (0rdiscipa?) CM. discipa GLOa. 36 il pavore (sic) a. 37 grossa oschura L. 38 Venendo a poco a poco ver F. rappressando B. in om. ABCEGHLMOP. 20 But Chame in a may= Che mi, since at xiv. 80 a has cha for che, as is common in E and other MSS. sometimes. 33 Bellicolo is, I believe, a common vulgarism still (especially about Rome) for umbilico. CANTO XXXI. 227 Fuggi'emi errore, e cresce' mi paura. 39 Perocche come in sulla cerchia tonda Montereggion di torri si corona ; Cosl la proda, che il pozzo circonda, 42 Torreggiavan di mezza la persona Gli orribili giganti, cui minaccia Giove del cielo ancora, quando tuona. 45 Ed io scorgeva gia d' alcun la faccia, Le spalle e il petto, e del ventre gran parte, E per le coste giu ambo le braccia. 48 Natura certo, quando lascio 1' arte Di si fatti animali, assai fe' bene, Per torre tali esecutori a Marte : 51 E s' ella d' elefanti e di balene Non si pente, chi guarda sottilmente, Piu giusta e piu discreta la ne tiene : 54 Che dove 1' argomento della mente S' aggiunge al mal volere ed alia possa, Nessun riparo vi pu6 far la gente. 57 La faccia sua mi parea lunga e grossa, 39 giungemi p. ZB. crebbemi F. 40 Perche F. E pero che B. in om. ABCEGHKOPQa. 41 Molte region (sic) EK. s' incorona BCDIKLMC. 42 prora ACEGHOP. la ripa F. che '1 poco E [prob. cler. error for po$o (as in ~K.)=po20 or poszo\. se- conda M. 43 da mezzo I. di mezzo L. da mezza M. 44 Horribili gig. H. 45 da cielo I. dal cielo Q. 46 Ed io iscopria D. Ed om. F. gik om. F. 47 ist 'e' om. OP. il ventre e del petto I. e degli occhi gran parte Z. 48 giuso M. ambe GMPQ. 51 Per tor via I. toglier D. tor cotali F1VL5. 52 dilionfanti DQ. d' alifanti EFM. d' olifanti P. dirfanti ilon e di (?) L. 53 penteo D. 54 e piu sagreta L. ne la tiene I. ne om. D. la rietene C. 66 Si giunge A^. 58 largha D. 39 Fuggtemi and cresce' mi appear in a variety of forms in the MSS. such &s,fuggiame,f^lggimme) cresciami, crescieami, crescimmi, &c. 40 F first wrote in error Perche come Montereggion di torri si corona. This is underlined and then the line is rewritten correctly. 42 Conversely proda stands for prora in xxvi. 141 (BCPZ). Also F has plura for proda in xxi. 13. Blanc compares the interchange of rado and raro. 51 ' I ' is perhaps thinking of x. 92. 152 228 INFERNO Come la pina di san Pietro a Roma ; Ed a sua proporzione eran 1' altr' ossa : 60 SI che la ripa, ch' era perizoma Dal mezzo in giu, ne mostrava ben tanto Di sopra, che di giungere alia chioma 63 Tre Prison s' averian dato mal vanto : Perocch' io ne vedea trenta gran palmi Dal loco in giu, dov' uomo affibbia il manto. 66 Rafel mai amech zabi almi, Cominci6 a gridar la fiera bocca, Cui non si convenian piu dolci salmi. 69 E il Duca mio ver lui : Anima sciocca, Tienti col corno, e con quel ti disfoga, Quand' ira o altra passion ti tocca. 72 Cercati al collo, e troverai la soga Che il tien legato, o anima confusa, E vedi lui che il gran petto ti doga. 75 Poi disse a me : Egli stesso s' accusa ; Questi e Nembrotto, per lo cui mal coto Pure un linguaggio nel mondo non s' usa. 78 59 Piero CDEFGKLMPQZ^4. Pier I. di Roma I. 60 E 'n sua D. A suo L. tutte 1' altre BD1VL4(?). 62 Da mezzo I. Del mezzo P. 63 aggiungere D. ziugnera la (sic) K. 64 Tre gran Freson L. Fresoni H^4 1 a. Frigion I. Fresion K. si saria DL. sarian (or -en) FQ*. s' averia KM. saren dato I. dati G. 65 Perch' io AE. gran om. K. 66 in su O. 1' uom AEK. s' affibbia CQZ. dove s' affibbia FIM. 67 Rachel L. bai I. ameo B. amer D. amet I. amehe F. amoe K. may mace Q. sabi D. et almi AEH. 69 A cui FI. con- veniaACDEFGHKOP^C". convien IQ. 70 a lui GKQ. 71 al corno K. 73 Cercate M. il collo L. 74 Chi t' ha legato D. Che tien L. 75 Vedi la giu F. Vedi la G. lei DAC. 76 stessi EP. 77 Egli e D. Menbrotto DK. Nembroth EP. Nembrot FHMOZa. Nenbrot I. Nebrotti G. voto M. 78 Piu un IB. nel mondo om. F. 60 K has era la trossa, which looks like an oral blunder. 64 Probably sarian and saren = s > arian and s* aren in F and I, since this Florentine form of avere occurs passim in these MSS. Comp. D in 1. 141. CANTO XXXI. 229 Lasciamlo stare, e non parliamo a voto : Che cosl e a lui ciascun linguaggio, Come il suo ad altrui ch' a nullo e noto. 81 Facemmo adunque piu lungo viaggio Volti a sinistra ; ed al trar d' un balestro Trovammo 1' altro assai piu fiero e maggio. 84 A cinger lui, qual che fosse il maestro Non so io dir, ma ei tenea succinto Dinanzi 1' altro, e dietro il braccio destro 87 D' una catena, che il teneva avvinto Dal collo in giu, si che in sullo scoperto Si ravvolgeva infino al giro quinto. 90 Questo superbo voll' esser esperto Di sua potenza contra il sommo Giove, Disse il mio Duca, ond' egli ha cotal merto. 93 Fialte ha nome ; e' fece le gran prove, Quando i giganti fer paura ai Dei : Le braccia ch' ei meno, giammai non move. 96 Ed io a lui : S' esser puote, io vorrei Che dello ismisurato Briareo Esperienza avesser gli occhi miei. 99 79 Lascialo ACDEFHIKLMOPQZa. parlemo K. parlamo AM. parlammo O. 80 alcun ling. C. ogni ling. D. ogni E x (with ciascuno later). 81 suo altrui F. che nullo E. 82 dunque Q. 83 A man sinistra F. ed om. Q. 85 qual maestro che fosse il maestro E (in obvious error). 87 Dinanzi 1' uno OZ. rietro E. di drieto L. 88 che teneva Z. tenea cinto avvinto (cinto under- lined) E. 89 Del collo M. si che sullo QZ. 90 rav. fino AD. insino FHI. sino al G. giron AC. 92 il om. ACEHOQ. contro al GI. contra! Z. 93 Disse '1 Maestro F. egli om. D. ha tal merto EL. 94 E Fialte I Q. O Fialte L. 95 Quando gig. M. a Dei ACEFGHIOPQa. alii Dei D. 1 96 che meno DFILMQZ. 97 puo CDF. po K. pote se ser (sic] L. Q t had potesse. 99 avesse P. 81 But as cha is often written for che in E, so perhaps the abbrevia- tion cli here may stand for cha or cK a. See note on 1. 20. 94 Perhaps this is for Effialte, and ' O ' in L may be in error for ' E.' (See 1. 108.) 230 INFERNO Ond' ei rispose : Tu vedrai Anteo Presso di qui, che parla, ed e disciolto, Che ne porra nel fondo d' ogni reo. 102 Quel che tu vuoi veder, piu la e molto, Ed e legato e fatto come questo, Salvo che piu feroce par nel volto. 105 Non fu tremoto gia tanto rubesto, Che scotesse una torre cosl forte, Come Fialte a scotersi fu presto. 108 Allor temett' io piu che mai la morte, E non v' era mestier piu che la dotta, S' io non avessi viste le ritorte. 1 1 1 Noi procedemmo piu avanti allotta, E venimmo ad Anteo, che ben cinqu' alle, Senza la testa, uscia fuor della grotta. 114 O tu, che nella fortunata valle, Che fece Scipion di gloria ereda, Quando Annibal co' suoi diede le spalle, 117 Recasti gia mille leon per preda, 100 Onderisp. ACDIKLQa. E quei risp. F. 101 da qui I. e sciolto FQ. 102 no porra K. porta G. del fondo DG. al fondo F. 103 piii om. H. & piu Ik molto M. piu e Ik Q. 104 E deslegato A. fano (sic) L. 106 teremoto A. termoto QA. tormento E. mai tanto MQ. 107 Che sotto se (sic) M. tanto forte G. 108 o fialte L. Effialte Q (see 1. 94). 109 temei io B. temetti piu IM5. di morte CF. 110 non era BCIZ. che piu la K. se non la L. quella dotta D. Ill Se non avessi visto I. 114 uscir (sic) a. 115 fortuna valle G (see xxviii. 7). 116 reda BCDFGHIKLOPQZCa. 117 con soi M. diede co' suoi Q l (corrected later). 107 Is M's an oral blunder, sc being constantly written s or ss in this MS. and double consonants disregarded ? 110 Dotta see note on xxii. 3. Land, and Veil. h. I. declare that dotta is a common Florentine provin- cialism for ' breve spatio di tempo' ; ' minima parte d' un' ora.' The usual explanation however is that it is connected with Prov. doptar, dotar, c. ( = dubitare) and so dotta = dubitanza, and thence fiaura. (So Scart.) The verb dottare occurs in B. Latini, Tesoretto, xiii. 24 : ' Dottai ben della morte'. 116 DGL have the peculiarly Tuscan iormgrolia. CANTO XXXI. 231 E che, se fossi stato all' alta guerra De' tuoi fratelli, ancor par ch' e' si creda, 120 Che avrebber vinto i figli della terra ; Mettine giu (e non ten venga schifo) Dove Cocito la freddura serra. 123 Non ci far ire a Tizio, ne a Tifo: Questi puo dar di quel che qui si brama : Pero ti china, e non torcer lo grifo. 126 Ancor ti puo nel mondo render fama ; Ch' ei vive, e lunga vita ancor aspetta, Se innanzi tempo grazia a se nol chiama. 129 Cos! disse il Maestro : e quegli in fretta Le man distese, e prese il Duca mio, Ond' Ercole sentl gia grande stretta. 132 Virgilio, quando prender si sentio, Disse a me : Fatti in qua, si ch' io ti prenda : Poi fece si, che un fascio er' egli ed io. 135 Oual pare a riguardar la Carisenda Sotto il chinato, quando un nuvol vada Sopr' essa si, che ella incontro penda; 138 Tal parve Anteo a me che stava a bada Di vederlo chinare, e fu tal ora Ch' io avrei volut' ir per altra strada : 141 119 se tu fossi E. fosti O. alia gran guerra DL. altra BCG. 120 fratei ACEHKOP. fradelli G. 121 avristi M. vinti CDEGHKP. figlioli BK. 122 Mettici D. giuso B. ti venga ABCDEIKLMCX4. a schifo IM^. !24gireQ. 125 puon dar a. che piu si T)A(marg.'). 126 Percio Q. 128 anche lunga vita D. 1' aspetta AE. s' aspetta P. 130 e egli in L. 131 Stese le mani F. La man GL. 132 Hercore O. sentia C. sentio B. grandi D. 134 Mi disse M. Dissimi A. vien qua D. fatti qua ACEFHO. fatti qui BKPQ. 135 fascio tra egli A. 137 il om. C. un om. FLQ. 138 che della or ched ella GKPQZ?a. intorno IL. incontra M. 139 a me Anteo DP A. 141 Ch' arei D. volentieri per GK. Ch' i' sarei I. voluto andar EFLyi 1 (?). che gir voluto avrei M. per 1' altra O. 134 B has merely ' q.'. 138 For ched ella see note on xxv. 69, and compare also It's reading in xxxiv. 113. 232 INFERNO Ma lievemente al fondo, che divora Lucifero con Giuda, ci sposo; Ne si chinato li fece dimora, 144 E come albero in nave si levo. 143 ci posbe ADGHIP 2 . ci sposc-e CMOPjQC ci passoe E. ci pos6 FL. si sposo a. 145 Ma come FIZ. E om. P. di nave LM. levoe ACDGHIMOPQC 143 and 145. Most likely the truncated rhythm is intentional and descriptive of the suddenness of the actions. Hence the readings posbe and levoe are probably copyist's corrections. The same remark would perhaps apply to eric and cricchi, &c. in next Canto, 26, 28 and 30. CANTO TRENTESIMOSECONDO S' io avessi le rime aspre e chiocce, Come si converrebbe al tristo buco, Sopra il qual pontan tutte 1' altre rocce, 3 Io premerei di mio concetto il suco Piu pienamente ; ma perch' io non 1' abbo, Non senza tema a dicer mi conduce. 6 Che non e impresa da pigliare a gabbo, Descriver fondo a tutto 1' universe, Ne da lingua che chiami mamma e babbo. 9 Ma quelle Donne aiutino il mio verso, Ch' aiutaro Amfion a chiuder Tebe, SI che dal fatto il dir non sia diverse. 12 O sopra tutte mal creata plebe, Che stai nel loco, onde '1 parlare e duro, 1 ed aspre BG5. 2 converrieno Z. 3 Sovra qual G. pontan om. E. porta L. 4 Io om. L. del mio KM. 6 pero che F. 6 teme H. al dicer Q. 7 Che '1 non e K. presa A. di pigliar AIKLO. 8 Di scriver ABCDEGHILMOPQZC. A scriver F. tutto a fondo D. 9 Ne e lingua E. a lingua FQ. chiama A. o babbo ~BCDEFGlKLMQZA(marg.). 11 Ch' aiutorno I. Chatarono (sic) D. ad Amf. a. 12 il fatto dal dir BI. dal fatto al dir CLa. 13 tutti B. create F. 14 Che sta I. onde parlarne B. onde parlar ACEGHLPZ?. ove '1 parlar KA(marg.). ove parlar Q. m' e duro I. e om. L. 1 a has an erased space, as if ed aspre stood originally. 8 Discriver (sic) may however stand either for Di scriver or Descriver. 12 On the reading in B see note on iv. 147 ; and for the construction involved in the reading of CLa comp. Inf. xix. 113 and Purg. xxiv. 62. 14 Sta is pro- bably for stai, as final i is omitted passim in 'I'. So D has troverra troverai in 1. 59. 234 INFERNO Me' foste state qui pecore o zebe. 15 Come noi fummo giu nel pozzo scuro Sotto i pie del gigante, assai piu bassi, Ed io mirava ancora all' alto muro, 18 Dicere udimmi : Guarda, come passi ; Fa si, che tu non calchi con le piante Le teste de' fratei miseri lassi. 21 Perch' io mi volsi, e vidimi davante E sotto i piedi un lago, che per gelo Avea di vetro e non d' acqua sembiante. 14 Non fece al corso suo si grosso velo D' inverno la Danoia in Osteric, Ne Tanai la sotto il freddo cielo, 2- Com' era quivi : che, se Tambernic Vi fosse su caduto, o Pietrapana, Non avria pur dall' orlo fatto eric. 30 E come a gracidar si sta la rana Col muso fuor dell' acqua, quando sogna Di spigolar sovente la villana : 33 15 Meio A. fosse (or -si) DQ. 16 noi om. E. noi bis (in error) F. 17 Sotto pi CDGHLP. Sotto piedi E. de (or dei) giganti ADCMja. de gigante E. 18 io guardava B. al lato A. altro GlQZA(marg.)B. 19 udimmo I. 20 Va si ABCDEFGHIKLMOP QZCa. tu om. E. non schiacci Q. 21 frategli ADE. di fratei CKQ. dei fratei Z. e lassi EFM. 22 mi mossi E. 23 i pi APQ. sotto piedi EHCX5. 24 e om. M. 25 Non fe' mai al corso BL (il corso). Non fece '1 C. mai al corso grosso M. al suo corso Q x (corrected later). a corso a. 26 Di verno GILOPQ. L' inverno BZZ?C. Danovia A. Danubio C. Osterlichi (or -icchi) AGHMO^ 2 a. Ostericchi FIG Austerricchi ZB. Asterlicchi P. Astarlichi L. Esterlichi C. Isterlicchi BQ. Istralicchi D. ne Ostorichi E. i ni ster gliochi (sic!) K. 27 gielo A. 28 eran MQ. Tambernicchi ABCDEFGHILMOPQZ^^'C. Ciamber- nicchi a. setenbr niochy (sic f) K. 29 suo cad. D. caduta K. piera pana K. 30 pur om. DM. fatto pur dall' orlo C. dell' orlo K. dalveloE. cricchi ABCDEFGHILMOPOZ^tfCa. criochi K. un crichi M. 32 e quando L. 30 See note on xxxi. 143. CANTO XXXII. 235 Livide insin la dove appar vergogna, Eran 1' ombre dolenti nella ghiaccia, Mettendo i denti in nota di cicogna. 36 Ognuna in giu tenea volta la faccia: Da bocca il freddo, e dagli occhi il cor tristo Tra lor testimonianza si procaccia. 39 Quand' io ebbi d' intorno alquanto visto, Volsimi a' piedi, e vidi due si stretti, Che il pel del capo avieno insieme misto. 42 Ditemi voi, che si stringete i petti, Diss' io, chi siete. E quei piegaro i colli ; E poi ch' ebber li visi a me eretti, 45 Gli occhi lor, ch' eran pria pur dentro molli, Gocciar su per le labbra, e il gielo strinse Le lagrime tra essi, e riserrolli : 48 Con legno legno mai spranga non cinse Forte cosi, ond' ei, come due becchi, Cozzaro insieme : tant' ira li vinse. 51 Ed un, ch' avea perduti ambo gli orecchi 34 Lividi ABCDEHKLQ. Liv. si Ik ABCEHKLO^. Li vidi si Ik Ma. Li vid' io si la P. Li vid' io insin Z. Si vid' io li Ik G. Li vidi si Ik dove il viso appar verg. E. infin Ik Q. infin dove F. insin dove si par D. par Qi- 35 Ch' eran M. Eran quivi i dolenti L. 1' umbra E. 36 in nota i denti E. 37 Ciascuna O. tenea giu D. tenea in giu F. 38 e om. F. 39 testimonanza ABCDFGLPQ. 40 Quando ebbi Qa. E quand' io D. m' ebbi A t . intorno F. assai visto D. 42 Che ( = che i?) peli DP. misti G. 43 Ditene F. 44 piegar li D/?. et ei p. F. 45 amene (sic) er. P. a me gli visi MQ. diretti A r 46 erano dentro prima D. eran prima dentro B. pria om. EF. prima a. 47 Docciargiu G. Gocciar giu AEFHP. le om. B. e om. G. giel gli strinse F. giel ristrinse D. 48 La lagrime a. 49 Col legno A. Che legno a legno D. Con legno ispranga ma' non cinse L. spranga mai ABCDFGIKMOPQZ^ 2 C. 50 cosi over come a. 51 Goc- ciaro P. con tant' ira L. 52 Che uno D. perduto ABDEIKL. perdute MQ. orecchie EM (as passim). 36 and 37 are accidentally transposed in A. 52 a hasperduti over an erasure, suggesting perduto in a^ 236 INFERNO Per la freddura, pur col viso in giue Mi disse : Perche tanto in noi ti specchi ? 54 Se vuoi saper chi son cotesti due, La valle, onde Bisenzio si dichina, Del padre loro Alberto e di lor fue. 57 D' un corpo usciro : e tutta la Caina Potrai cercare, e non troverai ombra Degna piu d' esser fitta in gelatina: 60 Non quelli, a cui fu rotto il petto e 1' ombra Con esso un colpo, per la man d' Artu : Non Focaccia: non questi, che m' ingombra 63 Col capo si, ch' io non veggio oltre piu, E fu nomato Sassol Mascheroni : Se Tosco se', ben sa' omai chi fu. 66 E perche non mi metti in piu sermoni, Sappi ch' io fui il Camicion de' Pazzi, Ed aspetto Carlin che mi scagioni. 69 Poscia vid' io mille visi, cagnazzi Fatti per freddo : onde mi vien riprezzo, E verra sempre, de' gelati guazzi. 72 E mentre che-andavamo in ver Io mezzo, Al quale ogni gravezza si raduna, Ed io tremava nell' eterno rezzo : 75 53 Colla freddura Z. 54 Disse perche cotanto ABCDEGHIKLM OPQZZ?a. Disse perche tanto F. Disse perche in noi tanto C. 55 chi for Q. questi altri due D. questi due E. 56 Le valle K. onde el M. Bisenzo ABDEFMOPQ.4Ca. Bisenza G. 58D'un ceppo C. D' un corpo fun K. e om. Q. 59 e om. G. un' ombra G. 60 d' esser piu DQ. fatta in gel. KZ. 61 Non quella FZ. quello Q. rotta D. 62 Con un sol colpo K. un corpo Z. le man G, man d' altrui C. 63 Non fu caeca G. Non fu chaccia P. con questi E. ne questi L. non questo Q. 64 vegno P. oltre om. A. 65 chiamato DG. nomato '1 Z. Sensol I. 66 Se Tosco omai ben sai chi fu D. omai ben sai I. ben de' saper chi e'fu F. chi '1 fu K. ch' io fu CE. 67 mi metta YIA. 68 ch' io sono DFO^4 (marg.)Ca.. i\om. LM. 69 qui Carlin M. 70 vidi mille FM. 71 ond' ei mi Z. vien om. P. 72 E n' era M. 73 mentre nui and. per Io M. ch' io andava F. 74 raguna DGIMCa. rauna ABCFHKLOPQZ5. 75 orezzo I. CANTO XXXII. 237 Se voler fu, o destine, o fortuna, Non so : ma passeggiando tra le teste, Forte percossi il pie nel viso ad una. 78 Piangendo mi sgrido : Perche mi peste ? Se tu non vieni a crescer la vendetta Di Mont' Aperti, perche mi moleste ? 81 Ed io : Maestro mio, or qui m' aspetta, Si ch' io esca d' un dubbio per costui : Poi mi farai, quantunque vorrai, fretta. 84 Lo Duca stette ; ed io dissi a colui Che bestemmiava duramente ancora : Qual se' tu, che cosi rampogni altrui ? 87 Or tu chi se', che vai per 1' Antenora Percotendo, rispose, altrui le gote SI, che se fossi vivo, troppo fora ? 90 Vivo son io, e caro esser ti puote, Fu mia risposta, se domandi fama, Ch' io metta il nome tuo tra 1' altre note. 93 Ed egli a me : Del contrario ho io brama : Levati quinci, e non mi dar piu lagna : Che mal sai lusingar per questa lama. 96 Allor Io presi per la cuticagna, E dissi : E' converra che tu ti nomi, O che capel qui su non ti rimagna. 99 Ond' egli a me: Perche tu mi dischiomi, 76 ist ' o ' om. L. o destine o gravezza (obvious clerical error from o/iotoreXeura) Z. 77 trapassigiando E. fra le DFIP. per le E. 78 nel volto A$). 79 crido K. grido L. 81 Aperto P. mi om. P. 82 Ed om. L. ora m' aspetta F. 83 Tanto ch' io esca C. 84 tu vuoi F. 85 stese G. 86 blasfemava A, fortemente Q. 87 Chi se' tu F. De ( = Deh?) qual se' tu P. che si AEFHMP. rapogni tanto altrui D. 88 O tu A]. 89 Perc. diss' egli B. Rispose perc. FMZ. 90 fosti A. vivo fossi Q^. 92 se adi- mandi A. se demandi infama (sic) a. 93 fra 1' altre FP. 94 contradio D. i' ho G. 99 che '1 capel K. capel costi non A. qua su CEGOP. co su (?) HK. 100 Ed egli CDK. discomi A. 76 ' al. fortuna'' Z marg. 2 ttcia manu. 238 INFERNO Ne ti diro ch' io sia, ne mostrerolti, Se mille fiate in sul capo mi tomi. 102 Io avea gia i capelli in mano avvolti, E tratti glien' avea piu d' una ciocca, Latrando lui con gli occhi in giu raccolti ; 105 Quando un altro grid6 : Che hai tu, Bocca ? Non ti basta sonar con le mascelle, Se tu non latri ? qual diavol ti tocca ? 108 Omai, diss' io, non vo' che tu favelle, Malvagio traditor, che alia tua onta Io porter6 di te vere novelle. m Va via, rispose, e cio che tu vuoi, conta ; Ma non tacer, se tu di qua entr' eschi, Di quei ch' ebbe or cosl la lingua pronta. 114 Ei piange qui 1' argento de' Franceschi : Io vidi, potrai dir, quel da Duera La dove i peccatori stanno frescht. 117 Se fossi domandato, altri chi v' era, Tu hai da lato quel di Beccheria, Di cui seg6 Fiorenza la gorgiera. 120 Gianni de' Soldanier credo che sia Piu la con Ganellone e Tribaldello, Ch' aprl Faenza quando si dormia. 123 101 Non ti ACDFGIMPQZa. Non ti daro (?) chi sia F. chi e' sia E. 102 mille volte A. in om. DFM. mi teni (sic) K. mio tomi Q. 103 gik om. I. 104 tratto A. gli aveva A. n' avia Q. 105 in om. F. 107 delle mascelle Z. 108 latra A. che diavol I. 109 Non vo' diss' io K. che piu EH. 110 che la tua BA. colla tua Za. tua vita (sic} E. 112 Va via disse e F. Va viadiss' egli e Q. Va oramai e cio a*. e om. L. cio che vuolgli si conta D. cio che voli M. 114 Di lui ch' ebbi Q. ebber cos! D. ch' ebb' or la lingua cosi Z. ch' ebbi la lingua cosi MQ. 116 quel bis in K. 118 Se fosti A. 119 dal lato OP. da Bee. DP. 120 A cui K. Fior. sego M. 121 del Sold. ACIMOQ. 123 Ch' apri6 BD. 114 Ebber in D is probably a misunderstanding o{ebbor=ebb > or, or per- haps a change due to the notion that quei was plural. See discussion of Inf. iv. 95. CANTO XXXII. 239 Noi eravam partiti gia da ello, Ch' io vidi due ghiacciati in una buca Si, che 1' un capo all' altro era cappello : 126 E come il pan per fame si manduca, Cosl il sopran li denti all' altro pose La Ve il cervel si giunge colla nuca. 129 Non altrimenti Tideo si rose Le tempie a Menalippo per disdegno, Che quei faceva il teschio e 1' altre cose. 132 O tu che mostri per si bestial segno Odio sopra colui cui tu ti mangi, Dimmi il perche, diss' io, per tal convegno, 135 Che se tu a ragion di lui ti piangi, Sappiendo chi voi siete, e la sua pecca, Nel mondo suso ancor io te ne cangi, 138 Se quella con ch' io parlo non si secca. 124 gia partiti A. 125 Quand' ioA^ schiacciati (sic) Z. 126 e 1' altro FL. 128 Cosi gli denti sopra 1' altro D. Cosi 1'un sopra 1' altro i denti BGKZ.Z?. Cosl el sovra li denti 1' altro C. Cosl sovra gli denti EH. Cosl sopra 1' altro i denti pose F. Cosl sovra li denti all' altro La. Cosi 1' un sovra li denti P. all' altro li denti Q- 129 La dove '1 AEI. Dove '1 DFGLZ. Ove '1 BCIC^. s' aggiunge BCDEFGHIKLOPQ^ 2 Ca. 130 si manuca M. 131 per isdegno E. 132 sfaciea A. rodeva F. Che facea il teschio e 1' altre cose suca M. e altre a. 134 che tu ABCDFGHILMOPQZ^a. tu m' ingagni (sic} C. 135 Dimmi perche CZ. io om. a. Dimmi '1 diss' io perche L. di tal F. Dilmi diss' io perche Q : (perche diss' io Q 2 ). 136 (h)ai ragion ADEHMPa. 137 Sapendo io B. Sappiend' io PZa. la suo pocha K. 138 Che su nel mondo F. su anchor AEHP. ancor ch' io D. 139 sia seccha DP. si schocha K. fia secca A. The terzina 127 129 is omitted in M, and note the curious alterations in the following lines to supply the rhymes. Hence if a MS. could be found in which this terzina is omitted, it would probably be related to M. M's archetype had the terzina omitted, hence the scribe of M altered the rhymes to suit what was before him. 132 The Commentary in M remarks, " Questo vocabulo suca vole dire parlatura" It appears therefore to be a commentary by the writer of the MS. and not any exist- ing commentary copied, but the explanation seems ignotum per ignotius. 138 D's repetition of che is probably due to the length of the construction. CANTO TRENTESIMOTERZO La bocca sollevo dal fiero pasto Quel peccator, forbendola ai capelli Del capo, ch' egli avea diretro guasto. 3 Poi cominci6 : Tu vuoi ch' io rinnovelli Disperato dolor che il cor mi preme, Gia pur pensando, pria ch' io ne favelli. 6 Ma se le mie parole esser den seme, Che frutti infamia al traditor ch' io rodo, Parlare e lagrimar vedrai insieme. 9 I' non so chi tu sei, ne per che modo Venuto se' quaggiu ; ma Fiorentino Mi sembri veramente, quand io t' odo. 12 Tu dei saper ch' io fui Conte Ugolino, E questi e 1' Arcivescovo Ruggieri : Or ti dir6 perch' io son tal vicino. 15 Che per 1' effetto de' suo' ma' pensieri, 1 su levb AEFGHOP. si levo CDIKLM^a. da M. 2 Del peccator M. forbendolo F. forbendogli a I OP. forbendosi M. 4 Puo (sic} E. ricomincio E. 5 Scellerato dolor K. Dispietato A. al cor DFL. che cor M. 6 pur om. F. ch' io favelli I. io om. C. 1 le om. P. dien EHMPOvZC. 8 Che fruttin fama D. fruttin B. fama I. peccator CI. 9 vedra' mi DFA 2 . udirai I. udrai M. mi vedrai a. 10 sie B. 11 Firentino E. 12 M' assembri DEFIQa. sembli P. quando t' odo DF. 13 il conte DI. 14 E questo 1' arch. VLB. 15 tel dir6 E. perch' ei son >A(marg.}. perch e son G. gli son vie. P. vicini G. 16 1' affetto D. per effetto L. mal pens. ABEMPOZC 2 . P is very inaccurate in this Canto. 15 G, reading percht son, probably mistook son for the 3rd plural and wrote vicini to suit ! CANTO XXXIII. 241 Fidandomi di lui, io fossi preso E poscia morto, dir non e mestieri. 18 Per6 quel che non puoi avere inteso, Ci6 e come la morte mia fu cruda, Udirai, e saprai se m' ha offeso. 21 Breve pertugio dentro dalla muda, La qual per me ha il titol della fame, E in che conviene ancor ch' altri si chiuda, 24 M' avea mostrato per lo suo forame Piu lune gia, quand' io feci il mal sonno, Che del future mi squarcio il velame. 27 Questi pareva a me maestro e donno, Cacciando il lupo e i lupicini al monte, Per che i Pisan veder Lucca non ponno. 30 Con cagne magre, studiose e conte, Gualandi con Sismondi e con Lanfranchi S' avea messi dinanzi dalla fronte. 33 In picciol corso mi pareano stanchi Lo padre e i figli, e con 1' acute scane Mi parea lor veder fender li fianchi. 36 Quando fui desto innanzi la dimane, Pianger senti' fra il sonno i miei figliuoli, Ch' eran con meco, e domandar del pane. 39 17 io fu si AGO. 18 E di poi D. E morto poscia C. non m' e CIKMC. 19 Percio F. Pero che quel che E. che ne puoi a. 20 mi fu cruda CF. 21 s' el m' ha AZ. s' ei m' ha B. sem'haiM. 22 della muda I La. 23 ha titol C. 24 E che AB*CDEFGHKLMOPZ#a. In che IQ. ch' altrui ABIOZA ancora altrui L. s' inchiudi U. la schiuda F. 25 mostrate L*. 26 Piu lume ADFGIM (lumi) Q5. Piu lieve BCEHKLPCa. dove feci I. quando f. BDGLMa. 27 mi squarto D. 29 Cacciando lupo K. i lupi BDLMP. 31 cagni grame I. e pronte D. 32 Gual. Sigtsmondi con Lanf. F. con Gismondi L. 33 S' aven L. S' avean M. messo I. davanti E. din. alia f. I. 34 picol tempo KD. 35 et figli KOM. figli con ACDEKLMZ. con acute G. 36 veder om. (obvious blunder) A. 37 Quand' io KLOPQ. innanzi alia KLA. 38 tra '1 AGQ. infra (corrected to 'fra' 2 ntia manu) E. nel sonno I. i om. M. fioli K. 39 con om. A. a dimandar EL, D. I. 16 242 INFERNO Ben se' crudel, se tu gia non ti duoli, Pensando ci6 ch' il mio cor s' annunziava: E se non piangi, di che pianger suoli ? 42 Gia eran desti, e 1' ora s' appressava Che il cibo ne soleva essere addotto, E per suo sogno ciascun dubitava: 45 Ed io sentii chiavar 1' uscio di sotto All' orribile torre ; ond' io guardai Nel viso a' miei figliuoi senza far motto. 48 Io non piangeva ; si dentro impietrai : Piangevan elli ; ed Anselmuccio mio Disse : Tu guardi si, padre : che hai ? 51 Per6 non lagrimai, ne rispos' io Tutto quel giorno, ne la notte appresso, Infin che 1' altro sol nel mondo uscio. 54 Come un poco di raggio si fu messo Nel doloroso carcere, ed io scorsi Per quattro visi il mio aspetto stesso ; 57 Ambo le man per Io dolor mi morsi. Ed ei, pensando ch' io '1 fessi per voglia 40 crudele se gia FI. 41 cio om. FM. ch' al mio BDLA 2 C. che '1 quor I. ch' al cuor D. cuor nunciava K. 43 eram O. era desto Z. trapassava I. 44 ne sole E. ci solea Z. 46 Ouand' io M. sentia P. 47 A quell' orribel P. e io D. 48 i miei L. noli EK. 49 piangevan M. m' inpetrai A. 50 Piangevano eglino D. 51 Mi disse C. Tu vardi K. 52 Percio ABCDGHIKL MOPQZ/2a. Perch' io E. ne lagr. B. e ne risp. L. 53ela notte D. e quella notte F. 54 Fin che AEH. altro giorno sol L. alto M. 55 Ma come Q. si om. Q. 57 il mio peto (sic) L. 58 Ambe le mani DM. Ambur B. per dolor FGKMQ 2 . 59 Ed egli A. Ond' ei M. E quei B. facessi D. fessi per fame K. 44 C has aduto, a Venetian peculiarity, both in the neglect of the double letters and the substitution of u for o. 51 Comp. vera for guerra in xxvii. 86 in K, and see note on xxv. 67. 58 On ambur cf. note on xiii. 58. 59 K of course here violates rhyme, but the two corresponding lines are over the page. CANTO XXXIII. 243 Di manicar, di subito levorsi, 60 E disser : Padre, assai ci fia men doglia, Se tu mangi di noi : tu ne vestisti Queste misere carni, e tu le spoglia. 63 Queta' mi allor per non farli piu tristi : Lo dl e 1' altro stemmo tutti muti : Ahi dura terra, perche non t' apristi ? 66 Posciache fummo al quarto dl venuti, Gaddo mi si gitto disteso a' piedi, Dicendo : Padre mio, che non m' aiuti ? 69 Quivi mori : e come tu mi vedi, Vid' io cascar li tre ad uno ad uno Tra il quinto dl e il sesto : ond' io mi diedi 72 Gia cieco a brancolar sopra ciascuno, E due dl li chiamai poi che fur morti : Poscia, piu che il dolor, pote il digiuno. 75 Quand' ebbe detto ci6, con gli occhi torti Riprese il teschio misero coi denti, 60 Di mucar (?) E. mandicar K. 2nd 'di' om. Q. 61 O disser D. assai ne DQ. 62 Che tu I. tu ci K. ne om. P. 63 e tu ne DFI. e om. Q. 64 non li far F. 65 Quel dl BD1Z. Lii di a*. et altro K. 66 O dura torre come F. 67 alquanto divenuti A. 68 si butto K. 69 E disse Padre EGHPQ. O padre F. 70 mi om. P. 71 Vid' io morire D. Vidi cascar IKL. li tre cascar P. 72 Tra '1 quinto e '1 sesto dl CK. Tra '1 quarto di e '1 sesto BF. 73 arangholar sovra a L. in brancolar Q. 74 E tre di ZB. li om. a*. 75 E poi piu D. Ma poi piu IQ. Poiche '1 dolor pote piu che '1 D. cal (sic] dolor E. il duol Q. 77 con denti M. 60 mucar is probably intended for manucar, a dialectic form of manicar or manducar, which is also found in xxxii. 130 (M). 74 On the readings tre and due see Blanc's Versuch, p. 289. Buti (born 1324) who lived at Pisa, describes minutely the circumstances of Ugo- lino's imprisonment, and says that he saw the chains, &c. He states that the prison was opened after eight days and all were found dead. If so sesto in 1. 72 would require due. Possibly the reading tre may have (inter alia) suggested the alteration in that line of sesto to quinto. 75 cal in E probably = che 7, since cha for che is common in this MS. 1 6 2 244 INFERNO Che furo all' osso, come d' un can, forti. 78 Ahi Pisa, vituperio delle genti Del bel paese la, dove il si suona ; Poi che i vicini a te punir son lenti, 81 Movasi la Caprara e la Gorgona, E faccian siepe ad Arno in sulla foce, SI ch' egli anneghi in te ogni persona. 8 4 Che se il Conte Ugolino aveva voce D' aver tradita te delle castella, Non dovei tu i figliuoi porre a tal croce. 87 Innocenti facea 1' eta novella, Novella Tebe, Uguccione e il Brigata, E gli altri due che il canto suso appella. 90 Noi passamm' oltre, la 've la gelata Ruvidamente un' altra gente fascia, Non volta in giu, ma tutta riversata. 93 Lo pianto stesso li pianger non lascia, E il duol, che trova in sugli occhi rintoppo, 78 Che forar 1' osso DA(marg:)C. Che fora 1' osso LQ,(?). Che foron 1' osso I. Ch' erano M. come di can D. forte A. 79 Ahi Pisa Pisa vitupero FL. Pixia A. A Pisa D. della gientty K. 80 dove si CEGZa. dove se sona M. 1& ove si P. 82 Muovansi F. Cravara A. Cravrara EH. Cavrara CDFLMZa. Chapraia BGl. Crapaia OQ. 84 Si che ann. AD. Si che in te annieghi F. 85 avea pur voce F. la voce D. 86 tradito DIM. tradito a te BQ. colle cast. I. 87 Non dovea porre i figl. D. Non dovei porre i figl. QZ. il figluol M. al tal L. tu cm. FGI. 88 Gli innocenti facean 1' etk M. Innoc. gli facea BFP. face' gli DA 2 C. facien K. lieta (sic) a. 89 Novella eth. ebbe Ug. A. e Brigata G. 90 altri del cui canto Q. suso il canto I. su appella A. 91 oltre ove la g. FQ. oltre dove la g. GILMZA. Ik dove la brigata P. 93 roversata E. rivesciata FI. rinver- sata M. 94 lor pianger FKM. lo pianger L. 95 II duol G. Che '1 pianto truova sugli F. trovava E. intoppo M. 78 Qj seems to have had the reading of the text. The terzina 82 84 was at first omitted in K, but supplied in the margin apparently by the same hand. 90 Here follows in E a very remarkable interpolation of 18 lines, first noticed in 1874 by Dr Greg. Palmieri. A full discussion of them will be found in the Appendix. CANTO XXXIII. 245 Si volve in entro a far crescer 1' ambascia: 96 Che le lagrime prime fanno groppo, E, si come visiere di cristallo, Riempion sotto il ciglio tutto il coppo. 99 Ed avvegna che, si come d' un callo, Per la freddura ciascun sentimento Cessato avesse del mio viso stallo, 102 Gia mi parea sentire alquanto vento ; Perch' io : Maestro mio, questo chi move ? Non e quaggiu ogni vapore spento ? 105 Ond' egli a me : Avaccio sarai, dove Di cio ti fara, 1' occhio la risposta, Veggendo la cagion che il fiato piove. 108 Ed un de' tristi della fredda crosta Grido a noi : O anime crudeli Tanto, che data v' e 1' ultima posta, m Levatemi dal viso i duri veli, Si ch' io sfoghi il dolor che il cor m' impregna, Un poco, pria che il pianto si raggeli. 114 Perch' io a lui: Se vuoi ch' io ti sovvegna, Dimmi chi sei, e s' io non ti disbrigo, Al fondo della ghiaccia ir mi convegna. 117 Rispose adunque : Io son Frate Alberigo, Io son quel delle frutte del mal orto, 96 Volvesi indietro e fa F. in retro C. indietro CG. in dentro IQ 2 Z. volve dentro D. e fa CDFGKLIVL4 2 CZ>a. 97 prima GZa. le prime lagr. F. 98 Et insieme vis. EH. E om. a. 100 Ed om. D. 102 dal mio KMQZ>. 104 Perch' io dissi, Maestro, esto F. che muove O. 105 qui giu AK. qua giuso TLZB. 106 Ed egli ACDEFKOPQ^ 2 CZ>. tosto sarai F. 107 Dove dici6 E. Da cio F. faran gli occhi G. la om. EQ t a. 108 Vedendo M. il om. M. fato (? forfatto) AL. fiato move YD. che '1 pianto K. 109 de' freddi della trista costra G. ill dato AEHILOPa. costa G. 112 Levatevi D. 113 il duol ABCDEFGHIKLMOPQCZ?a. ch' al cor D. mi om. L. mi pregna AEM. 114 Ma poco I. si mi geli K. ralegri (sic) L. 115 Ed io D. 116 chi fusti F.Z?. 118 dunque D. i fui D. 119 E son quei D. dalle frutta HKMOP. delle frutta BCQZ^a. della frutta GC. dalle frutte AEZA 2 . 246 INFERNO Che qui riprendo dattero per figo. 120 O, diss' io lui : Or sei tu ancor morto ? Ed egli a me : Come il mio corpo stea Nel mondo su, nulla scienza porto. 123 Cotal vantaggio ha questa Tolomea, Che spesse volte 1' anima ci cade Innanzi ch' Atropos mossa le dea. 126 E perche tu piu volentier mi rade Le invetriate lagrime dal volto, Sappi che tosto che 1' anima trade, 129 Come fee' io, il corpo suo 1' e tolto Da un demonio, che poscia il governa Mentre che il tempo suo tutto sia volto. 132 Ella ruina in si fatta cisterna ; E forse pare ancor lo corpo suso Dell' ombra che di qua retro mi verna. 135 Tu il dei saper, se tu vien pur mo giuso : Egli e Ser Branca d' Oria, e son piu anni Poscia passati, ch' ei fu si racchiuso. 138 Io credo, dissi lui, che tu m' inganni ; Che Branca d' Oria non morl unquanche, 120 E qui F. qua D. ricevo Q. ripiglo D. dattilo BI. datalo E. 121 Deh diss' io D. O diss' io a lui BF. O dissi lui GLC. O dissi a lui IO. Diss' io a lui Q. Or diss' io lui Z. or om. L. 122 Come che '1 corpo D. 123 su om. D. 124 Talnomea L. 125 ci om. F. 126 Pria che F. Prima che A Antropos ACDGHKOPZ^CZ^a. Antroposo LQ^ li dea AIK. gli dea MP. morte gli dea G. 128 del volto DG. 130 Nel mondo su il corpo A(marg.}C. corpo su EH. gli e tolto AGKM. P e su tolto D. 131 Da mi (sic) M. la governa D. 132 sia tolto K. 133 Et la M. 134 per ancor lo corpo suo E. 135 dietro ABCDEFGHIKLMOPQZA'CPa. m' inverna ADEIKMP. 136 Tu dei sap. CDO^. 137 d' Ori F. molt' anni K. 138 richiuso AK. rinchiuso EFIMQ. 139 diss' io lui ABCDEFGHKPZ^a. dissi a lui IO. diss' io a lui KMQ. 140 d' Ori F. 130 The reading of A (marg.} and C is evidently a repetition of 1. 123. 131 The reading of M is apparently a clerical error for Da un, since utt, nu, mi, im are often almost indistinguishable. (See Prefatory Note.) CANTO XXXIII. 247 E mangia e bee e dorme e veste panni. 141 Nel fosso su, diss' ei, di Malebranche, La dove bolle la tenace pece, Non era giunto ancora- Michel Zanche, 144 Che questi lascio il diavolo in sua vece Nel corpo suo, e d' un suo prossimano Che il tradimento insieme con lui fece. 147 Ma distendi oramai in qua la mano, Aprimi gli occhi : ed io non gliele apersi, E cortesia fu, in lui esser villano. 150 Ahi Genovesi, uomini diversi D' ogni costume, e pien d' ogni magagna, Perche non siete voi del mondo spersi ? 153 Che col peggiore spirto di Romagna Trovai un tal di voi, che per sua opra In anima in Cocito gia si bagna, 156 Ed in corpo par vivo ancor di sopra. 141 magna AKM. mangia beve E. 142 diss' el Malebranche A. disse su di D. diss' el su B. 143 La ove A-^. 144 Non v' era DI. Non n' era G. ancora giunto EFP. ancor Don Mich. M. Micheri B. 145 Che questo EQ*. Che quel I. un diavolo ACEGHILOPQ*. 146 Nel corpo su AP. Nel suo corpo E. corpo suso A. et un AGlKQ*ZACa.. ch' un suo D. dasu presimano (sic) L. 147 Che tradimento Q. 148 oggimai ABCEGHIKOPQZCa. omai quaggiu F. omai L. di qua K. 149 E aprimi F. non gli apersi AE. non gliene I. 150 Che cort. M. fui lui B*(?). fu lui ACEGHKLMPZ>a. a lui FOZ. Q has fo lui altered to_/b //. 152 costum pieni C. 153 voi om. DQ. dispersi D. 155 Trova' io un di voi Z. di voi om. C. di voi un tal ABEGHIKMOPQ^4Z>a. un di voi tal F. di voi un che L. 156 Un' anima IL. gia om. CPZ>a*. L' anima sua in Cocito si bagna FM. 157 E '1 corpo suo FM. 148 B has ogtumai, as also in xxxiv. 26 and 32. 150 The erasure in B seems to imply fui lui in Bj. CANTO TRENTESIMOQUARTO Vexilla Regis prodeunt inferni Verso di noi: pero dinanzi mira, Disse il Maestro mio, se tu il discerni. 3 Come quando una grossa nebbia spira, O quando 1' emisperio nostro annotta, Par da lungi un molin che il vento gira ; 6 Veder mi parve un tal 'dificio allotta: Poi per lo vento mi ristrinsi retro Al Duca mio ; che non li era altra grotta. 9 Gia era (e con paura il metto in metro) La, dove 1' ombre eran tutte coperte, E trasparean come festuca in vetro. 12 Altre sono a giacere, altre stanno erte, Quella col capo, e quella con le piante ; Altra, com' arco, il volto a' piedi inverte. 15 1 regia G. 2 innanzi F. 3 setu discerni CFIMZ. 4 Come una g. n. quando sp. L. nebbia grossa BCPZ. scira (?) C. 5 E quando IM. 6 di lungi ACEFGHIKLMOPa. ch' al vento AEGHLP. che vento Da. ch' a vento aggira Q. 8 mi strinsi E. mi ritrassi G. mi ristrinsi per lo vento Q. a retro K. a detro M. 9 mio om. Ka. gliera ADEHKMOP. che '1 non gli era A. non v' era FIZ. 10 e om. L. il mento in EL. 11 ombra a. tutte eran BEFGHIKOPZ#Z>a. tutte era A. tutte 1' ombre C. tutte 1' ombre era M. 12 in metro E. 13 Altre stanno a giac. BFM. e altre F. sono erte DG. 14 e om. YKA Z . 15 Altre L. il suo volto B. in volto Ca. a piede A. a pi CDKLZ. ai pi BQZ?. il collo a piede I. il capo a piedi F. riverte CDQZ. verte B. rinverte KZ>. 9 li, gli, and vi are often interchanged. Cf. Purg. xiii. 7. &c. Q has gli era die non (but corrected 2 nd ' 1 manu}. 15 But a pic probably stands for ai pit. CANTO XXXIV. 249 Quando noi fummo fatti tanto avante, Ch' al mio Maestro piacque di mostrarmi La creatura ch' ebbe il bel sembiante, 18 Dinanzi mi si tolse, e fe' restarmi, Ecco Dite, dicendo, ed ecco il loco, Ove convien che di fortezza t' armi. 21 Com' io divenni allor gelato e fioco, Nol domandar, Letter, ch' io non lo scrivo, Pero ch' ogni parlar sarebbe poco. 24 Io non morii, e non rimasi vivo : Penso oramai per te, s' hai fior d' ingegno, Qual io divenni, d' uno e d' altro privo. 27 Lo imperador del doloroso regno Da mezzo il petto uscia fuor della ghiaccia ; E piu con un gigante io mi convegno, 30 Che i giganti non fan con le sue braccia: Vedi oramai quant' esser dee quel tutto Ch' a cosi fatte parti si confaccia. 33 S' ei fu si bel com' egli e ora brutto, E contra il suo Fattore alzo le ciglia, 16 fatte D. 17 parve di LQZ>. 20 di te (sic) M. 21 Dove EFMP. Onde K. 22 venni allora DF. 23 Non domandar CD. che non lo C. 26 P. lettore se tu hai D. Or pensa ormai P. omaiAEFHOC. oggimai BGIKQ^a. 27ed'unoK. 29 Dal mezzo '1 BC. Dal mezzo p. A v . fuor om. I. 30 E piu che un gran gigante il mi convegno F. che un M. io om. ACM. 31 Che gig. ACDEGHIKLOPMZAP. E che giganti O_. non fa Aa. sue om. F. 32 Or vedi omai P. omai ACDEHP. oggimai BFGIKLMOQZ.ffCa. che debbe esser F. egli tutto E. 33 fatta parte ABCDEFGHIKLMOPQZtfZJa. 34 come e ora I. 35 Incontro D. E contro al FGIZ^4. E central ACHQ. Contro al L. E contra al a. 32 Cf. P's reading in 1. 26. 33 In A fatta has apparently been altered from fatte probably by a diopda>Trjs, who was not aware that parte (as in many similar words in this and other MSS.) was meant for the plural. 25O INFERNO Ben dee da lui procedere ogni lutto. 36 O quanto parve a me gran maraviglia, Quando vidi tre facce alia sua testa ! L' una dinanzi, e quella era vermiglia ; 39 L' altre eran due, che s' aggiungieno a questa: Sopr' esso il mezzo di ciascuna spalla, E si giungieno al loco della cresta ; \i E la destra parea tra bianca e gialla ; La sinistra a vedere era tal, quali Vengon di la, onde il Nilo s' avvalla. 45 Sotto ciascuna uscivan due grandi ali, Quanto si convenia a tanto uccello ; Vele di mar non vid' io mai cotali. 4 8 Non avean penne, ma di vipistrello Era lor modo; e quelle svolazzava, SI che tre venti si movean da ello. 51 Quindi Cocito tutto s' aggelava : Con sei occhi piangeva, e per tre menti Gocciava il pianto e sanguinosa bava. 54 Da ogni bocca dirompea coi denti 36 prender ogni A. 37 Quanto parve FQ. 38 Quand' io ACFHKOPMZ. 39 Una F. 40 Dell' altre due I. eran om. M. si giongono K. se giungieno a. 41 Di sopra '1 mezzo D. il om. BEHKOPa. e di K. da EHILP. 42 s' aggiugneano IMa. 'E' before s 3 agg. has been erased in I and added i nda manu in 'a.' E aggiugneano F. al sommo della I. al colmo della A(marg.}D, del lor cresta (clumsily altered to testa) K. della testa D. 43 La destra mi parea FIMO. 44 L' altra a vedere F. era a veder E. 45 ove '1 DKLQZ.5. dove il M. 47 li conv. K. conveniano Z. a cosi fatto ucc. E. 48 di navi L. non furon mai tali D. non vidi mai KLQD. mai tali CL. mai om. E. 49 havea E. visper- tello AKQ. vilpistrello DLA^BCD. vespistrello Z. pilistrello B. vispistrello CEOPa. vespertello IM. d' unvisp. K. 50 Eran P. il lor D. solazzava CD. su lazzava D. suo lazava Q. in su alzava I. su alzava 'B>7,D(marg.'). su 1' alzava GK. in su 1' alzava M. 51 da ella G. 52 si giellava E. 53 Per sei MQ. e con tre menti DK. 54 il petto sang. D. al petto sang. Q. il petto e sang. L. con sang. E. 55 E d' ogni D. vi rompea E. con denti M. In G, after 1. 55 lines 2055 are again repeated bodily and in the CANTO XXXIV. 251 Un peccatore, a guisa di maciulla, Si che tre ne facea cosi dolenti. 57 A quel dinanzi il mordere era nulla, Verso il graffiar, che tal volta la schiena Rimanea della pelle tutta brulla. 60 Quell' anima lassu che ha maggior pena, Disse il Maestro, e Giuda Scariotto, Che il capo ha dentro, e fuor le gambe mena. 63 Degli altri due ch' hanno il capo di sotto, Quei che pende dal nero ceffo e Bruto : Vedi come si storce, e non fa motto : 66 E 1' altro e Cassio, che par si membruto. Ma la notte risurge ; ed oramai E da partir, che tutto avem veduto. 69 Com' a lui piacque, il collo gli avvinghiai ; Ed ei prese di tempo e loco poste : E, quando 1' ale furo aperte assai, 72 Appiglio se alle vellute coste : Di vello in vello giu discese poscia Tra il folto pelo e le gelate croste. 75 Quando noi fummo la dove la coscia Si volge appunto in sul grosso dell' anche, 56 manciulla P. 67 facean E. 58 Al qual L. 59 ch' era volta K. tai volte A v 60 Rimane D. 61 laggiu M. 63 dentro ha '1 capo D. 65 Ouello che D. Quel che ABCEFGHIKLM OQZa. prende L. dal terzo I. ceffo nero FL. 66 com' el Z. com' ei A. torce AB* (evidently altered from storce) EGH MPQZ. 67 L' altro AEH. Quell' altro DP. Casso C. 68 La notte E. ris. e torna omai D. 69 abbiam ved. FQ. avea L. 70 Com' a lor B. il capo C. i' m' avvinghai K. 71 prese tempo E. del tempo DGIKM. tempo luogo FL. e poste ABCDFHLMP. e posta E. 72 1' alie A. 1' ali... aperte ABCDEFGHKLMOPQZa. aperte ebbe C. aperte furo A. furno I. 73 Appigliosene P. croste (obvious error) C. lanute coste Z. 75 e le vellute croste C (confusion with 1. 73 q. v.}. 77 in om, FM. same hand, without any indication that the blunder was ever noticed. In L after 1. 58 line 22 is repeated in error, and line 59 is omitted alto- gether. 252 INFERNO Lo Duca con fatica e con angoscia 78 Volse la testa ov' egli avea le zanche, Ed aggrappossi al pel come uom che sale, Si che in inferno io credea tornar anche. 81 Attienti ben, che per si fatte scale, Disse il Maestro, ansando com' uom lasso, Conviensi dipartir da tanto male. 84 Poi usci fuor per lo foro d' un sasso, E pose me in sull' orlo a sedere : Appresso porse a me 1' accorto passo. 87 Io levai gli occhi, e credetti vedere Lucifero com' io 1' avea lasciato, E vidili le gambe in su tenere. 90 E s' io divenni allora travagliato, La gente grossa il pensi, che non vede Qual e quel punto ch' io avea passato. 93 Levati su, disse il Maestro, in piede : La via e lunga, e il cammino e malvagio, E gia il sole a mezza terza riede. 96 Non era camminata di palagio La V eravam, ma natural burella Ch' avea mal suolo, e di lume disagio. 99 Prima ch' io dell' abisso mi divella, Maestro mio, diss' io quando fui dritto, 78 con om. O. 79 piante A. 80 Aggrapandosi F. 81 io om, DELMQ. inferno credetti F. 82 per cotali scale ABCEFGHIK LMPQZZ?a. 83 ansiando MP. 84 Si convien Z. da cotanto E. 85 Poi usci per lo foro d' un gran sasso C. per un foro FL. 86 puosi mivi (?) in L. 87 porse me lo corto passo C. 88 Io chinai A. 89 io om. I. 91 Com' io C. E om. A. 92 il om. KL. nol vede DQ. 93 Qual era il 2.BD. lasciato ACG. pasciato (sic) E. 94 in su A. 95 e om. L. e salvaggio a. 97 caminato L. da M. 98 Ove er. BCD. Dove GLMZZ?. La ove eran I. a ma (?) L. durella G. 99 Ch' avea '1 C. mal sole D. ma il suol E. 100 Prima che dell' AIMOA. che dello abisso io F. che dall' ab. LD. disvella ADFGL. dilieva E. 101 Maestro diss' io lui Q. diss' io om. E. quand' io AKOP. poi ch' io fu ritto F. quanto fu ritto L. CANTO XXXIV. 253 A trarmi d' erro un poco mi favella. 102 Ov' e la ghiaccia ? e questi com' e fitto Si sottosopra ? e come in si poc' ora Da sera a mane ha fatto il sol tragitto ? 105 Ed egli a me : Tu immagini ancora Esser di la dal centro, ov' io m' appresi Al pel del vermo reo che il mondo fora. 108 Di la fosti cotanto, quant' io scesi : Quando mi volsi, tu passasti il punto Al qual si traggon d' ogni parte i pesi : m E se' or sotto 1' emisperio giunto Ch' e contrapposto a quel che la gran secca Coperchia, e sotto il cui colmo consunto u 4 Fu 1' uom che nacque e visse senza pecca : Tu hai li piedi in su picciola spera Che 1' altra faccia fa della Giudecca. ir; Qui e da man, quando di la e sera : E questi che ne fe' scala col pelo, Fitto e ancora, si come prim' era. 120 Da questa parte cadde giu dal cielo : E la terra che pria di qua si sporse, Per paura di lui fe' del mar velo, 123 102 error ADE. un poco d' erro Z. 104 poco d' ora Q. 105 ha om. a*. 107 D'esser ABCDEGHIKLMOPQ^4,(?)5a. da centro O. ond' io mi presi A. ov' io mi presi CDEGHKLMOPQZ^?C/?a. 109 tu fusti tanto F. io om. L. 110 Quand' io ACEHIKOPQZZ>a*. rivolsi M. ill Dal qual KM. A qual L. parte pesi Q. 112 or om. E. 113 Ch' 6 opposite (or opposto) ABEFGHIMZ. Ched e opp. B. Ch' e apposto C. Che n' e opp. P. a posto K. della gran secca a. che la 'n secca B. 114 Converchia CQ. il om. Q. 116 in om. E. 117 altra parte MA. 118 diman DE. doman K. 119 E questi c chende (?) fe' G. ne fecie E. ne fa O. che ci fe' Z. 120 in prima I. Ancora e fitto come F. 121 da cielo L. 122 porse AHILMZCP. si scorse Q. 123 del mal L. 113 See note on xxv. 69, for Z?'s reading here. 119 Possibly G's reading is an oral blunder for che ne or chenne. 123 mar and mal are elsewhere interchanged. See note on xxx. 19. 254 INFERNO E venne all' emisperio vostro ; e forse Per fuggir lui Iasci6 qui il loco voto Quella che appar di qua, e su ricorse. 126 Loco e laggiu da Belzebu remoto Tanto, quanto la tomba si distende, Che non per vista, ma per suono e noto ^9 D' un ruscelletto che quivi discende Per la buca d' un sasso, ch' egli ha roso Col corso ch' egli avvolge, e poco pende. 132 Lo Duca ed io per quel cammino ascoso Entrammo a ritornar nel chiaro mondo : E senza cura aver d' alcun riposo 135 Salimmo suso, ei primo ed io secondo, Tanto ch' io vidi delle cose belle Che porta il ciel, per un pertugio tondo, 138 E quindi uscimmo a riveder le stelle. 124 el misperio E. alle misterio (sic) G. nostro ABCEFGHI KLMOPQ*Za(?). 125 il om. ABEGHKMOPQZa. quel luogo I. 126 che par AEIKLMOQ. 127 quaggiu D. Belzebub AC. Belzabu DLa. Balzebu G. Balzabu K. Belzibub Q. 128 discende LQ. 131 bocca FGM^4. 132 Col corso suo che volge F. a poco K. 133 nascoso F. 135 d' avere alcun D. 136 Salimmo su ACEGHIKMOP^a. in su L. suo (sic) Q. 138 pertuso I. pertusio K. 139 quivi L. COLLATION AND DISCUSSION OF SELECTED PASSAGES. NOTE. Italic Capital Letters are reserved for MSS. which I have not myself examined. The information given is generally derived from printed Editions, or published Collations : sometimes it has been contributed by friends. The authorities relied upon in each case will be found acknowledged under the description of the several MSS. When MSS. are referred to under the Nos. in the Bibliografia Dantesca of Colomb de Batines (e.g. Bat. 444), they have been ex- amined only in a few scattered passages. No special account of them will be found among the MSS. described. For such informa- tion Colomb de Batines may be consulted. When, for brevity in the record of Collations, several MSS. are grouped thus "56 to 62" it must be understood that the latter figure is always included. It will be observed that, in spite of what I have said in the Proleg. (p. xxvii), I have frequently recorded merely the total number of MSS. in which a reading was found. This has been done for economy of space, in cases when either (i) some one reading was found in such an overwhelming number of MSS. that it seemed unnecessary to do more than specify in detail those MSS. which did not contain it : or (2) when the relative merit of the readings could not be so confidently determined as to make the passage serviceable as a test. INFERNO. Vedrai gli antic hi spiriti dolenti Che la seconda morte tiascun grida. Inf. i. 116, 7. I. 28. Poi ch' ei posato un poco il corpo lasso, or Poiche posato (wh. prob. = Poi ch' ei) BCDMOQ(?)STr.VM J &Z)(?)^77V6>2f / Hai^o-bqturt^-r. 4.5.8.9.11.13.16.17.18.19.22.24.26.27.28.31.32.35.36.37.38.39.45.48.49.51.53. 54.59.60.61. 63*.6472. 73.77. 82.97. loo. 103. 104. 106. 1 10. 1 14. Bat. 10. 129. 142. 415.428.431.444 [85] Poi ch' ebbi posato un poco Pyai 7.80.107.109.112. Poi ch' io ebbi posato un poco 1. Poi ch' ebbi posato il c. 1. il. Poi ch' io ebbi posato il c. 1. t\m 91. Poi ch' io posato un poco il c. 1. q. Poi ch' ei pasato (sic] alquanto, &c. *. Poi che posato il C. 1. (with itinpoco in margl) 85. Poi posato ebbi un poco &c. \ZV l Ci\(marg.} vdm^ 29.56.57.79.86.90. 102.105.1 13. Poi ch' ebbi riposato un poco il c. 1. Q 2 . Poi ch' ebbi riposato il c. 1. AUVW/Af^pev^cehknvwzc 20.25.42.71. 74.78. Barg. Poi ch' io ebbi riposato il c. 1. 83. Poi che fu riposato E. Poi che posato fu A. Poi che posato ebbi un poco il c. 1. 99. Poi che riposato un poco il c. 1. Bat. 1 14. Poi fu posato un poco ^23. Poi posato un poco -TV i. Poi riposato un poco Z>(?)Z7 12. Et poi che fu posato il c. 1. s 58. D. I. 17 258 INFERNO. Et riposato alauanto &c. 21. Et poi che riposato il c. 1. K. Et riposato un poco il c. 1. F8\|/d/ 10.33.43.44.46.68.94.95.98. Quando ebbi riposato il c. 1. Rl> 101. Quando ebbi posato il c. 1. g. Quand' io ei posato un poco &c. 0A*. Quando posato un poco il c. 1. 67. Da poi che riposato il c. 1. f. Da ch' ebbi riposato p 66. Com' ei (or Come) posato un poco &c. tr 2*. Com' ebbi riposato. Ott. Com' io posato un poco E(?) G(?) HTV 3 WXGl.r2Kir (passato) xy 3. 14 (passato) 15.34.40.41.65.70.76.84.88.89.92.93.108. Bat. 132.429. Or che posato un poco 8. Como possato (sic) un poco/ Poi che un poco il c. 1. L. (Cler. error ?) Poi prese lena un poco il c. 1. (!) 62. D is differently reported by Witte and Sicca. In 75 a torn page begins ...hei posato, and in p the line is defective, but begins Chom... Out of the extraordinary number of varr. lectt (no less it will be seen than 36) of this line, I believe the one printed first to have been certainly the original text. All the others can I think be easily accounted for as deviations from it, while it can scarcely have arisen from any of them. I suspect that the archaic and obsolete ei (for ebbi) was the chief source of the mischief. It was either recognized, and deliberately changed into the more familiar form ebbi [if so illustrate by enno in v. 38, which has been changed into eran, erano, or sono\: or it was misunderstood [if so illustrate by xx. 69, where fesse (=facesse) has been altered to fosse, or fosse in}, and hence arose the substitution (with or without consequential emendations) of io in several MSS. e.g. Poi c/i 1 io posato, Coin' io posato, &c. In the former case the change would 'pitifully disaster' the rhythm. This however some copyists took no further notice of, while others sought to remedy it by further changes in the text, such as the omission of un poco, combined with the change of posato into riposato (the latter having perhaps also the advantage of being a slightly more familiar term). It may be observed that the omission of such words as un poco is much more probable than their insertion ; and COLLATION OF SELECTED PASSAGES. 259 also that in this passage the words un poco, if. not actually necessary, enhance the picture, while their absence weakens it. After such a struggle, and with danger and fear still pressing, the rest of the poet's weary frame was only brief and partial. There is also, it will be observed, a whole crop of minor variants such as Da che, Come, Or che, Quando, &c. &c. depending merely on the taste and fancy of the copyist. In reference to the form ei, see Nannucci, Analisi Critica del Verbi Italiani, p. 499, where illustrations are given of this form of the perfect (ei, esti, ee or e &c.) from several old poets, including Fra Guittdne, Cino da Pistoia, Brunetto Latini, &c. ; and among prose writers (where there can be no question of ( metri gratia 1 } from Giamboni, the translator (inter alia] of the Tresor of B. Latini from French into Italian. Compare also, both for the scansion, and for the same form ei, B. Latini, Tesoretto, Cap. III. E pd ctt i' /' ei pensato, &c. It is further noticeable that in Inf. iv. 25 Castelvetro reads, " Quivi secondo cti ei per ascoltare," paraphrasing, ' secondo che ebbi e compresi per ascoltare,' and adding, ' di questa lettura appaiono vestigi ne' libri scritti a mano.' Perhaps such a 'vestigio* appears in the readings which I have registered there, chie and cti io. It will be observed that io is one of the changes for ei in the passage now under discussion. I. 42- alia gaietta KCDlM.V*ABCDMOSVv$&mzg 21.29.33.3742.56.57.60. 79.90.94.100.102.105.113. [35] la gaietta BEFGHIKOPQRSTUV^^WXZGl.Tr.T^VQf/rAOAn 2 1 I'Hap-y8'n6iKXv7rpcrn|/f abceghiklnqrsuvwxy abcdefhiklmnpqrtuvx 1.2. 3.4.5.7 to 20.22.23.25.26.28.31.32.34.35.36.38.39.40.41.43.45.46.48.49.51.55. 58.59.61.62.63.66.67.68.70.71.72.74.75.76. 77.78. 80. 82. 83. 84. 85. 86. 88.89.91. 92.93^95.97.98. 101. 103. 104. 106. 107. 108. 109. i io. 1 12. 1 1 5. Bat. \ 14.444. [174] Anon. Fior., Benv., Ott. (?), Suti, Barg., Cast. Bocc. mentions both readings without deciding between them. della gaietta 44. cha la gaietta t. e la gaietta Eu. 172 200 INFERNO. la quieta 24. la lieta 65. da gaetta 1 14. 27 is defective. Of the Venetian Codd. (auct. Gregoretti) none have alia g., while 18 have la g. The reading alia gaietta, though found in comparatively few MSS. (in about i in 6 as compared with la gaietta, according to my collations; in 21 as against 91 according to Dr Barlow) is still I think probably the true one. 1. The general sense of the passage is better suited by it. In lines 34 36 it has been distinctly stated that the ' lonza' hindered Dante and beat him back. How then could the 'gaietta pelle ' of the animal itself be one of the sources of hope and encouragement, as is signified by the reading ' la gaietta'1 In that case, what object remains to cause disquiet, or to call for support and encouragement ? Further, as Scarf, asks, how could the panther's gay and spotted hide be a ground of hope in any case ? On the other hand, every- thing falls into its place when (reading ' alia ') we interpret, that the brightness of the early dawn, and the sweet spring tide (1. 43) [" the gladness of the May " as Wordsworth says] caused him to have a good hope concerning, i.e. of over- coming, (notwithstanding what is said in 1. 36) the beast with gaietta pelle. Comp. the construction in Conv. II. xi., 'fanno della persona bene sperare.' 2. The force of SI che in 1. 41, coupling the recapitulation contained in 1. 43 with the preceding lines 37 40, and im- plying, as it does, a distinct consequence from them, has not, I think, been sufficiently noted. It was early morning (1. 37) and spring time (11. 38 40), consequently (Si che) both the Jiour and the season (1. 43) inspired him with hope 1 . 1 Besides the general association of buoyancy and hope with the spring, there should be added the mediaeval belief, (shared by Dante and referred to here in 1. 40) that the world was created at the spring equinox, an idea found also among the ancients, as is shewn by the well known and beautiful passage in Virg. Georg. ii. 336 &c. Non alios prima crescentis origine mundi Illuxisse dies, aliumve habuisse tenorem COLLATION OF SELECTED PASSAGES. 26 1 3. The reading ' alia' is supported by xvi. 108, la lonza alia pelle dipinta, and illustrated by ix. 36, /' alta torre alia cima rovente. This is of course a two-edged argument, as such passages (or at any rate the former) may have sug- gested the alteration, and in cases where the similarity is too obvious such an argument would probably point in the other direction, but not so I think when the resemblance is (as here) somewhat distant and recondite. 4. There are one or two points of view from which la gaietta appears to be lectio facilior, and more likely to have been substituted for alia gaietta than vice versa, (a) La gaietta is slightly more euphonious in respect of rhythm, (b] The length of construction involved in the reading alia gaietta would not have been unlikely, here as elsewhere, to suggest the supplying of an earlier nominative, (c) We might per- haps add, as another possible reason for such a change, that the construction alia gaietta is a little unusual. In fact it is commonly recognized by Commentators as one of the most distinct instances of a Gallicism (and therefore a non-Floren- tine idiom) in the Divina Commedia. The other two passages already cited (Inf. ix. 36 and xvi. 108) are of course similar cases. To these Par. xi. 1 14 amassero a fede is sometimes added and Inf. ix. 50 Batteansi a palme. See the very exhaustive discussion of the subject of Gallicisms (real or alleged) in the Divina Commedia by Zingarelli in the Studi di Filol. Romanza, Fasc. I. pp. 109 143. (d) It should also be added that from alia gaietta (i.e. ala, or a la, as it would often be written) the reading la gaietta would have been almost certain to have arisen accidentally, since the initial a would have been fused in many MSS. with the final a of Crediderim : ver illud erat, ver magnus agebat Orbis, et hibernis parcebant flatibus Euri, &c. The propitious character of the vernal equinox is again referred to in Parad. i. 37 42 and Conv. ii. c. 4, 1. 50. " Dico ancora, che quanto il Cielo e piu presso al cerchio equatore, tanto e piu nobile per com- parazione alii suoi poli ; perocch ha piu movimento e piu attualitade e piu vita e piu forma," &c. &c. See further on this, my work on Time References in the Divina Commedia, pp. 14 16, 118 120. 262 INFERNO. fera. (See other cases in Inf. ii. 127 ; xiii. 20 ; xv. 39 ; xix. 3, &c. &c., and Prefatory Note, p. li.) On all these grounds then an original 'alia' is more likely to have altered into '/' than vice versa. The early Commentators however seem all to support la gaietta, though they commonly treat la gaietta pelle as the accusative after sperare. Thus Bocc. and Benv., the former of whom mentions both readings without deciding between them, and comments thus " secondo 1' altra lettera (i.e. the reading la gaietta) m' era cagione di bene sperare, di dovere ottenere la pelle di quella fiera la quale esso intendea di prendere." Benv. paraphrases " dabant mihi causam bene sperandi detractionem pellis," &c. So Pietro " duo dicit quae eum fecerunt contra dictum vitium bene sperare, sci- licet... principium diei...et tempus Martiale." The Otlimo comments thus " Sicche, considerate queste due cose, spe- rava e non temeva le cose ree," &c., the 'due cose ' being " 1'ora e la stagione del tempo, cioe la mattina a mezzo Marzo," &c. It is clear from the expression ' due cose ' that the Ott. does not at any rate take la pelle as one of the subjects of era, and so, if that was his reading, he probably considered la pelle to be the object of sperare, as Bocc., Benv. and Buti do (' del detto animale s' intende d' avere la gaietta pelle,' &c. Buti). And so it seems to be understood by all the early Commen- tators, except indeed Castelvetro, who takes it as a nominative. At any rate their explanations clearly exclude what seems to be the usual modern interpretation of the reading la gaietta, to which I have taken objection above, viz. that which makes la gaietta pelle itself one of the grounds of hope. They are not however always sufficiently explicit to enable us to decide whether they read alia gaietta, or la gaietta taken as an accusative. Serravalle seems to have read alia gaietta, since he paraphrases ' Dabat mihi bonam spem vincendi bestiam pellis maculate, sive gaiete pellis.' I may add that in a list of 10 or 12 sample "test- passages" sent to me by the late Dr Witte in Nov. 1879, alia gaietta was one, though he gave no reasons for the selection. COLLATION OF SELECTED PASSAGES. 263 I. 48. temesse. I found this in 187 MSS. and in the following Commentators : Pet., Benv., Lan., Cast. tremesse (or sometimes tremasse) BQV 3 W J pp|/weinu/ 14.34.42.61. 85*88.94.105. [21] Butt. g has no temesse. z has temesse. p and 46 have che '1 mondo, with tremesse in p, and temesse in 46. E has che la terra ne temesse. 7* has che 1' aire ancor tremisse. 91 has il ritemesse. Tremesse seems to me more obvious and commonplace than temesse, and so the former would be somewhat more likely to have been substituted for the latter than vice versa. In fact it might well be an imitation or reminiscence of Inf. iv. 150. The reading temesse is found in such an over- whelming majority of MSS. that it is scarcely serviceable as a test-passage, though this fact, especially if it be the less obvious word, is much in its favour, and would imply the comparatively late introduction of the other and simpler reading. Among the older Commentators, Biiti reads and expounds tremesse, and, while most of the others pass over the passage, Pietro at any rate reads temesse, for he trans- lates 'aer videbatur de eo timere.' Benv. has temesse, giving the strange interpretation of aere, ' idest viri aerei, idest alti vel sapientes, quasi dicat tacite : non solum minores, sed etiam majores, timent superbiam leonis, idest potentis.' Similarly Land, has 'superiori e grandi uomini.' I cannot help thinking that this striking expression is introduced by Dante to account for his own fear at the sight of the lion described in 1. 44. Observe the connexion "Ma non si che paura non rni desse," &c. : i.e. these grounds for cheerful hope that have been mentioned were still not enough to quench the fear which the vision of the lion inspired. Indeed, ' so terrible was the sight ' that the very air around him seemed to fear 1 . Hence I think temesse has a special 1 Castelvetro compares the converse proverbial expression, ' to be afraid 264 INFERNO. point Ugo Foscolo in a note on this passage defends tre- messe "perche 1' acre che parea temere non esprime se non congetture; e il tremare, mostrando un effetto visibile, fa imagine." An argument which, to say nothing of its pedantry, is two-edged, since such a consideration might have suggested the substitution of tremesse for temesse. Further, we might answer Foscolo's objection to the use of temesse by quoting the very same expression in Ovid, Metam. (a work with which Dante was intimately acquainted, and which he often both quotes and imitates) xiii. 406, where Ovid says of Hecuba, ' novo latratu terruit auras 1 .' Or again, Fasti i. 567, ' fragor aethera terruit ipsum.' Once more we find in Shakespeare (Henry V., Prologue), ' the casques That did affright the air at Agincourt.' The expression surely needs no further justification, if it needed even thus much. The other reading however might be illustrated from Guido Cavalcanti, Son. vii. : Chi e questa che vien, ch' ogni uom la mira, E fa di clarita 1' aer tremare. As to the form tremesse being found from tremare, instead of tremasse, it has been suggested by Zani de' Ferranti (Di varie Lezioni, &c., p. 2) that Dante (or, as we should prefer to say, the author of this reading) probably had regard rather to the Latin form tremere than to the Italian tremare. I. 94. questa bestia was found in 131 MSS. quella bestia in 44, viz. DYGMV^GlDE-nQwkm^mrf 1.7.21.29.33. 35.36.37.44.48.56.57.60.65.74.79.82.83.90.95. 102. 105. 108. 1 1 3. [44] This variant is noticed because it seems to be an example of a change from considerations of euphony (as probably in Inf. ii. 94 (M), vii. 108, and many other places), and also because the comparative rareness of quella may perhaps serve to test relationships. even of the air,' and hence argues by a curious non-sequitur that if even the air feared how much more should animate objects do so ! 1 Comp. Inf. xxx. 20. COLLATION OF SELECTED PASSAGES. 265 II. 12. Prima che was found in 136 MSS. Anzi che AZ/hmq? 18.27.29.31.56.57.90.95.102.113. [17] The variant Anzi die for Prima die is noted because it is found in several members of the ' Vatican ' family. It is not in all of them, but generally speaking only in its most pronounced members 1 . In this way I have found it a most useful preliminary test passage ; as if a MS. has this reading it is almost sure to exhibit the ' Vatican ' peculiarities elsewhere, though, as I have already implied, the converse does not necessarily follow. Anzi is again substituted for prim a in Inf. viii. 54 (/>), Purg. vii. 4 (A), and Par. xxiv. 6 (Z>). La quale e 1 quale v mar.cmn/r7/7t72... 1 13. 46.55. 56.57. 58.60.61.63. 75. 79.83.90.91. 95. 97*.98*. 100.102. 104.1 13.1 14. [81] La quale il quale GlY'V 62.109. (This is probably the same as the above, the 'e' being merged in the final letter of quale.) Lo quale e il quale CEFHLMOQRS^UV^^WXTr.^flTiiKiAlirr^x aeklstu.\yz(al T)acdcijknt 1.4^?). 5. 8. 1 1 . 1 2. 1 5. 1 6. 1 7. 19.24.26.3 1 .32.34.37.38.40. 43. 49.54. 59.64.67.68.69.70.74.78.81. 82.84.85. 88.89.92.93. 97 1 (?). 9 8 1 (?).ioi. 103. 105.106.107.1 io*.i 15. Hat. 129. 132.142. [97] Lo quale al quale OGl.rSKvoxJ/wv^ 10.42.65.71.108. [17] Lo qual dal quale 1 1 2. Lo qual del quale N. La quale al quale 77. La qual e qual q. Lo quale e '1 quanto 48. La quale e 1 chi 53. II quale al quale 6. II quale e 1 quale 18.22.27. Hat. 444. Del quale al quale 9i/. Nel quale al quale .r. II quanto e 1 quale pj 66. Per lo quale e '1 quale 5 1 . 1 The ,\iss. A and in, in which this reading is found, arc exceptional in shewing little or no trace of such influence elsewhere. 266 INFERNO. E 1 chi e 1 quale K 3*. Lo (II) chi e 1 quale Irmp 7.39. La chi e 1 quale 73. Lo che e 1 quale 6. Da questi il quale 2. Del quale a voler dir lo vero 80. Lo quale al voler dire il vero. Bat. 440. ' 28 ' omits this terzina. La quale e il quale seems undoubtedly the true read- ing; the other, or rather others, being in fact (as far as I can see) unintelligible. The depravation of the text pro- bably arose from a sort of blind notion that there was no reason for the distinction of gender, the author of the change having failed to see the reference back to Roma and suo impero in line 20, "La qual Roma e il qual impero " as Barg. paraphrases it. Hence probably at first the simple but stupid change of La to Lo, and next the bolder and more radical changes given in the tertiary readings, E '/ chi, 11 quanta, &c., in the vain hope of infusing some sense into the meaningless procession of words to which the first short-sighted alteration had reduced the text. II. 55- piu che la stella. This was found in 196 MSS. piu che una stella BKLMG"aeo-4>lsvxy(?)^7 3.33.452.58.63.70.79.86. 91.101.110.115. Bat. 55.415.128. [32] Ott., Guid. piu che stella X^ 65.72.105.107. Bat. 65.102. quanto la stella U*i/. come la stella Tr.E 51. piu di una stella M. In n and 24 the terzina is omitted. Barg. (text and comm.) has the singular reading piu che a stella. This is perhaps a blunder due to an original ' piu cha (or ka} stella.'' In fact cha occurs here in c, and in Bat. 65 ; and in some MSS. (as in E,) cha is often written for che. The reading una stella is an obvious case of lectio facilior. If it were original, the reading la stella could scarcely have COLLATION OF SELECTED PASSAGES. 267 arisen, while the reverse change is obvious and natural. It is not indeed at all easy to determine the interpretation of la Stella, and this, it need hardly be said, is presumptive evidence both in its favour, and also as to the probable cause of its extrusion from the text. Commentators are still widely at variance, some explaining it to mean the sun ; others, the planet Venus. The majority, I think, of modern commentators incline to explain it as used generically for the ' starry firma- ment,' or in fact ' the stars,' and with this view I unhesitatingly agree. I will first say why I think both of the other explanations named are to be rejected. As against 'la Stella' being 'the sun,' I would urge (i) that (as far as I can find) there is no authority for this use of the word 1 . In the place sometimes cited from the Convito (iii. 5), difficult though the whole passage is, the context makes it perfectly clear (in spite of Fraticelli and other commentators) that 'la stella' denotes the pole-star. The fact that the sun is called ' pianeta? just above (1. 17) is not in point, since the sun was considered to be one of the planets in the Ptolemaic system. So Petrarch, Trionfi (Tempo 1. 86) calls the sun 'il gran pianeta.' (2) Besides, it would surely be rather a tasteless hyperbole to say that the eyes of Beatrice shone brighter than the sun. To describe them as brighter than the stars, "the burning eyes of heaven" (as Shakspeare 2 calls them), seems far more appropriate. Then again as to the planet Venus, no doubt 'la stella' is so applied, but again, as far as I am able to ascertain, not absolutely, but only in such phrases as 'A stella levata 1 ; 'All' ora della stella ; ' Fino alia Stella 1 , &c. in reference to Venus as the morning or evening star [see Vocab. Tram. s. v. stella}*. 1 There seems indeed to be a passage in Guido Guinicelli's Canzone beginning "Al cor gentil" &c., where in 1. 40' la stella'' stands for 'the sun,' which is the main subject of this part of the poem, being mentioned seven or eight times by name. But in this place it is solely in virtue of a special context that it stands for 'the sun,' just as in the passage quoted in the Convito, it is the context which determines it to mean ' the pole-star,' as it does also in Par. xii. 29. 2 Hamlet, Act n. Sc. ii. 3 In the fragmentary Pastille de Torquato Tasso on the first xxiV 268 INFERNO. Next as to the interpretation of 'la stella' as 'the starry firmament,' or 'the stars' collectively, we have I think abun- dant support for this in Dante's own use of the word elsewhere. First, in the Convito, and in the Canzone prefixed to the Tratt. III. near the end, we read : Tu sai che '1 ciel sempr' e lucente e chiaro, E quanto in s non si turba giammai : Ma li nostr' occhi per cagioni assai Chiaman la stella talor tenebrosa. This is not conclusive in itself, but the prose Commentary later on in Tratt. III. c. 9 leaves, I think, little doubt that la stella here means 'the stars' or 'the starry heavens.' For (i) the words " il ciel sempre e lucente e chiaro," &c. are paraphrased by the words " avvegnache la stella sempre sia d' un modo chiara e lucente... per piu cagioni puote parere nn chiara e non lucente." Thus he clearly uses stella here as equiva- lent to both cielo and stella in the Canzone. He proceeds to add that this occasional obscurity is due to the medium through which it is seen, which medium varies with the presence or absence of the sun, who is so full of light ' che e vincente della stella' Hence it is quite clear that la stella certainly does not here stand for il sole, in spite of Fraticelli's short-sighted note on the line in the Canzone, though he comments differently on this passage in the Trattato itself. Towards the end of the Chapter, Dante says that imperfect or diseased eye-sight may sometimes produce the same effect, so that, as he graphically puts it, objects are blurred like writing upon damp paper, and thus also "puote parere la stella turbata." He quotes his own experience in proof of this, when from excessive reading his eyes were so weak that c le stelle' (note this equivalent substitute for the word ' la stella' just before) f le stelle mi pareano tutte d' alcuno albore ombrate.' There is an equally convincing, if not still clearer, proof of his use of the word in this sense in the Canzone to the 4th Trattato. Dante wishes to prove that wherever there is virtue there is 'gentilezza ' or ' nobiltd,' but not vice versa, noting afterwards Cantos of the Inferno (Bologna, 1829) I find the note on this line "La Stella assolutamente di Venere." COLLATION OF SELECTED PASSAGES. 269 in proof that the words 'gentile' and 'nobile can be applied to inanimate objects such as pictures, rocks, &c., and shewing that apart from virtue "ciascuna cosa...in sua natura nobile esser chiamata." Here however he adds this illustration : Siccome e cielo dovunque la Stella, Ma cio non e converse. Again it is the Commentary that determines the meaning: see Tratt. iv. c. 19, where we are told that nobilta is the heaven (cielo) in which ' molte e diverse stelle rilucono' (corresponding therefore to la stella in the Canzone), and these stars are the several virtues, moral and intellectual, many of which he enu- merates, thus reminding us of Wordsworth's well-known line : The primal duties shine aloft like stars. Once more let us turn to the Vita Nuova, and in the Canzone of c. xxiii, beginning "Donna pietosa" we find the line " Turbar lo Sole ed apparir la stella." We should have little doubt as to this being equivalent to 'the stars' even if we did not read in the equivalent prose passage of the same chapter (1. 23) " Pareami vedere il sole oscurare si, che le stelle si mostravano di colore, &c." Blanc in his ' Versuch ' aptly quotes a very similar use by Dante of 'la fiammella as equivalent to ' lefiammelle,' in Inf. xvii. 33. 'La stella' is the reading of a vast majority of the MSS., but it will be seen that the reading ' una stella' is as old as the Ottimo Commento, i.e. 1333. It may be interesting in conclusion to cite the opinions of the older commentators, the ' patristic ' literature, so to speak, of the Divina Commedia. With the exception of the Ottimo already noted, and of Bargigi who has the curious reading ' piu che a stella' both in the text and its citation in the Commentary (where however it is not explained) with these exceptions, I say, they seem all to have had the ordinary reading piu che la stella, but they explain it differently. Boccaccio comments rather ambiguously, " la luce di quella stella che e piu lucente." Buti paraphrases : "qualunque stella... sanza ristringersi ad alcuna, dice, piu che 2/0 INFERNO. stella." Vellutello and Daniello explain it of the sun the latter adding that "la stella" stands for the sun among the stars, just as 'il Poetal in reference to Latin authors, would mean Virgil ; ' il Filosofo,' would always be understood of Aristotle ; /' Apostolo, of S. Paul, and so on. Landings com- ment is certainly very curious. " Questa e ottima compara- zione. Percioche la stella luce, ma non si che a perfezione mostra le cose, come il sole. Adunque Beatrice lucea piu che la stella, e come il Sole, perche la grazia perficiente e con- sumante luce piu che 1' altre grazie." Castelvetro contents himself with saying that he has proved elsewhere that ' la Stella 1 is Venus, adding somewhat contemptuously, " il che non intendono gli spositori"; rather reminding us of Aristotle's dismissal of an opponent with the words 'e/ceivos rovro tfyvoet,' in the Poetics. Benv. comments very briefly, but apparently in the sense which I have contended for : piu che la stella, " quia transcendit omnes caelos." Giiido Pisano has " plus- quam faciat una stella." As in other cases, it was probably the unfamiliar, or perhaps not understood use of 'la stella' that prompted the simplifica- tions found in the other readings, such as ' una stella' &c. Possibly too the readings ' 'quanta la stella' and ' come la stella' were intended to soften the hyperbole or exaggeration that would be implied in the reading ' piu che la stella' when it was understood to mean ' the sun.' II. 60. moto lontana B X*v (marg.) oac 2 egio(?) F*a 1 (?) bdx^ 2 //^r2*.3.4 1 (?) 12.28*. 29.35.41.452.46.48.49.54.55.56. 57. 60.61.62. 68. 70.7 1,5.73.76. 79-82.83.86.9o.93.(aprir il tuo) 95.97.98.100.101. 102,. 109^.113. 1 14*. 1 1 5. j5W. 444. [77] Lan., Bocc., Land., Veil., Guid. R and t have d' aprirmi. t' e uopo ch' aprirmi (or ch' aprire) AB*C*DEGHLQSTUV 2 V 3 V 4 1 W XGLTr.* D 2 GMNQrAAnZ$Vay$tQ K tfoKpa-ty- non t' e piu uo' ch' aprirmi b. Piu non ci voglio ch' empiere 1 tuo w. Piu non voglio (?) ch' aprirmi K. Piu non uo' coprir il to talento 8. coprirmi il tuo 1 1 2. Piu non vorria ch' aprirmi 91. non ci e uopo aprir lo mio 106. ch' aprir lo tuo intellects (!) v. D.I. 18 274 INFERNO. I am inclined to give a very decided preference to the reading f e uopo aprirmi (omitting 'che'}, though it is not that of the majority of MSS. examined. The insertion of ' die ' gives no doubt a very natural and appropriate sentiment in the general, and one which might be compared with that of Purg. xxxiii. 130 2, and also with the well-known lines, &n. i. 76, 77. Tuus, o regina, quid optes Explorare labor, mihi jussa capessere fas est : both of which passages are quoted by Blanc. Whatever be the reading here adopted, there is little doubt that the passage of Virgil was in Dante's mind, especially as he shews his familiarity with it by directly quoting it in the Vita Nuova c. 25. But this tempting general sentiment does not so well suit this context, if we consider (i) that Beatrice's will had been already declared to Virgil (see lines 67, seqq.), and more- over Virgil has just stated that it had been so (see lines 79, 80). (2) That it is not afterwards in any way declared, as the reading ck* aprirmi would imply that it should be, and as the words of the previous line distinctly state that it need not be ; words which quite naturally lead to the decla- ration of 1. 8 1 (omitting the word che) that any further expla- nation of her will would be superfluous. (3) Piii seems almost unmeaning with ch* aprirmi, unless it could be shewn that some declaration of Beatrice's will was still future, which, as we have pointed out, is not the case : whereas with aprirmi it has a natural and obvious reference to the sufficient explanation of it already made in lines 67, scqq. (4) The follow- ing words ' Ma dimmi' come more naturally after the reading I am advocating; i.e. 'There is no need to expound your wishes, but explain to me this other point.' (5) Lastly, it is to be observed (and this alone might perhaps be regarded as almost conclusive) that che introduces a redundant syllable defying scansion, unless the mutilated form no found in some of the MSS. were admissible. COLLATION OF SELECTED PASSAGES. 275 III. 8. eterno was found in 115 MSS. eterna 95 eterne 8 viz. V 2 ^g 19.24.64. eternal 3 viz. Zjp. The MSS. are not very unevenly divided between eterno and eterna. It is perhaps in favour of the former that prima facie both grammar and euphony are against it, and con- sequently eterno would be more likely to be changed to eterna than vice versa. The fact is that eterno is an adverb rather than an adjective, like the Latin neuter in 'sedet aeter- numque sedebit/ ' aeternum latrans,' ' serviet seternum/ &c. We might compare Inf. xix. 12, *E quanto giusto tua virtu comparte,' where, as might be expected, several MSS. have giusta. Cf. also 'tutto (al. tutti) storse i piedi' in xix. 64, and the reading ' tutto arricciar li peli' found in some MSS. in xxiii. 19. Add Inf. xi. 67 ; Par. xvii. 93 (quei che fien presente). In most of these cases the masc. form is not that of the majority of MSS., and I do not mean to imply that it is in all instances correct, but its frequent occurrence even as a variant will serve to illustrate the use here contended for. III. 31. orror A 14. 1 5. 17. 1 8. 19.20.2 1. 22.23.26. 27.28.29.3 1. 32.33.34. 36!. 55. 57. 60.63.67. 70.72. 73. 75. 84.88. 89.91.92. 93.98.99. 100. 102. 107^109. Bat. 30. SS.ii4.i29(?) 132.428.429.431. [93] Anon.Fior., Barg., Veil., Guid. error ABCDEFGKLMOPR\JV 1 V 4 GLTr.FMNOQS[/VT&Q3>V*a,a.p y J niX(xvoirpcTT4>|'w9a*bcgh 2 jk(?)lmnor 2 stuvyz acdefhijmntwx l,2*(?) 3.4*. 5.7. 9.10.11.12.13.16.24.25.30.35.37.38.39.42.43.45.46.48.49.51.54.56.58.59.61.62. 64.65.66.68.69.71.74.76.77.78.79.80.82.83.85.86.87.90.94.95.97.101.103. 105. 106.108.110.112.113.114.115. ^^.10.65.142.415.416.444. [149] Butt, Land, (mentioning both), Dan., Cast., Serr. Q has la testa d' orror. m 74 have la testa d' error. P has d' erro (cf. Inf. xxxiv. 102). e has de dolor (!). 79 has d' error la testa plena cinta (sic]. 1 8 2 2/6 INFERNO. o has d' rror (sic). buv are doubtful. Orror seems proved beyond doubt to be the true reading (as Scarf, has pointed out) by the obvious imitation of Virg. jEn. ii. 559, Me turn primum saevus circumstetit horror. Comp. also ALn. iv. 280, Arrectaeque horrore comae. Possibly too (as Zani de' Ferranti suggests) Dante may have had some recollection of Daniel vii. 15, "//tfrrwzVspiritus meus; ego Daniel territus sum intus, et visiones capitis mei conturba- verunt me." It also suits well the context preceding. Error is so little appropriate (though found in a majority of MSS.) that it probably arose from some very early copyist's blunder, rather than as an intentional emendation of the text. The early Commentators pass over the passage. Anon. Fior. and Veil. adopt orror ; Bull and Land., error.; but Land, cites and explains the other reading also, as being that of ' altri testV. Benv. mentions both without deciding. Cast, has error, and enumerates three special ' errori ' that Dante has fallen into. III. 36. infamia B*FGV*ZG\.Tr 2 .ASCDEOSU 1 T l ^ l a.^ 1 c l ^fi y i (?)^rw6.io 1 .i4.i6.i7 1 .i8.i9.2o 1 (?)2i.23.24.27. 30.31.32.33.35.39.42.46.55. 56*. 58 1 .59 1 .6o.75 1 .87.90.99.ioo 1 .ioi.io2*.io3.io9 1 . [75] Ch. Anon., Ott., Bocc., Dan., Serr. fama A i. 2. 3.5.7.8.9.1 1. 12.i3.i5.i7 2 .2o*.22 1 .26.34. 36.37.40.41.43.44.45. 48. 49. 51.53.54.57.61.62.63.64.65.66.67.68.69.70.71.72.73.74.76.77.78.79.80.82.83.84. 85. 86.88.89.9 1 1 .92.93-94.95.97.98 1 .ioo 2 .io2 1 . 104. 105. 106.107. IC) 8-i 10.112.113. 114.115. Bat. io.30j.55.65. (fame) 114.129.132. 142.4i5-4i6.428.429j.43i. 440.444. [179] Lan., Benv., Butt, Anon. Fior., Land., Veil., Barg., Cast. Some MSS. (e.g. tueg, &c.) have infama, but this no doubt = infamia (comp. matcra and materia, which are often interchanged, as well as other similar words). 38 has senzan fama (sic). n has senza famia. 25 has famia with ' i ' erased. The terzina is omitted in 4, but added below (and later) COLLATION OF SELECTED PASSAGES. 277 The interesting and instructive variation of reading here has evidently had its origin from the superficial difficulty, If these victims lived senza infamia why are they condemned? Hence it was obvious to substitute fama, of which lodo is but a feeble and nearly tautologous repetition, whereas to in- famia it stands in fine contrast. Of course too it entirely misses the point of the passage, which is to reproduce the scornful contempt of the seer of the Apocalypse for the Laodicean Christians : "because thou art neither cold nor hot, I will spue thee out of my mouth." Moreover, as Blanc well points out, the distinction between infamia and lodo is again explained and emphasized by lines 38, 39, 'non furon ribelli' (= infamia), ' ne fur fedeli a Dio ' (= lodo} [ Versuch. &c. p. 33]. The majority of MSS. and early editions however exhibit the false reading fama, and this may consequently be regarded as an important test-passage. Some MSS. write infama here. The same form is found in ' a ' at xxxii. 92 (where of course fama is correct). Compare the reading of ' D ' at xxxiii. 8. The fact of fama being commented on by Lan. (1328) and infamia in the very early Chiose Anonime (ed. Selmi), p. 17 (attributed by C. Hegel to 1320), and also in the Ott. (1333), shews that both readings were in existence very early. Infamia is found in Boccaccio [who in fact paraphrases the v/ords 'senza lodo' ' cioe senza fama '] ; while fama is adopted by Anon. Fior. and Buti [the latter justifying its use in a bad sense curiously, by quoting Virg. ;En. iv. 174, ' Fama malum quo non aliud velocius ullum,' so that he reads fama, but explains it as h. I. equivalent to infamid\: by Land, [who per contra maintains that it has a good sense here], and by Veil. Benv. adopts fama, adding ' aliqui tamen dicunt, senza infamia; sed prior litera melior est, quia non vixerunt sine infamia, ut patebit paulo post.' (A good specimen of the reasoning out of which the false reading fama probably originated.) Cast, adopts senza fama, adding however the curt remark, 'I migliori testi hanno senza infamia' Thus the majority of the early Commentators as well as of MSS. support the inferior reading fama. Dan., adopting infamia, reasons more soundly in its 2/8 INFERNO. favour, " perche se 'fama e lodo ' dicesse, verrebbe anco a dire una cosa stessa," &c. Guid. would evidently be of the same opinion, since (after quoting the passage with 'infamia') he paraphrases, 'qui de mundo sine fama et infamia exive- runt.' It is curious to observe what a large number of the MSS. have been tampered with (see above) in consequence of the uncertainty of reading in this passage. III. 59, 60. Vidi e conobbi is almost universal. After examining more than 240 MSS., I found Guardai e vidi in three only, viz. b, i, and /. Dr Witte found it in nine only out of about 400. Its very rarity however may make it useful as suggesting relation- ship, and this indication in the case of i and / at any rate is fully borne out by many other passages. It is curious to observe that Guardai e vidi appears in an overwhelming majority of Editions, including the two Aldines and the Delia Crusca (all of which however are very inferior texts) and several others of the i6th century; while in modern Edi- tions it is all but universal. Out of 19 such which I have examined I find Vidi e conobbi in three only, viz. Witte, Scartazzini, and Camerini. Per contra, none of the old Com- mentators (except perhaps Castelvetrd] recognize the reading ' Guardai &c.,' while Bocc., Benv., Buti, Land., Barg. have Vidi e conobbi. The rest are indeterminate. III. 114. Similar remarks will apply to the readings Vede and Rende in this passage. Dr Barlow found Rcnde in two MSS. only out of 74 examined, and one of these, the Codice Landi (A), seems doubtful 1 . I found it in six only out of about 240 (viz. i] k 48.85.91.97) and Mussafia reports it as occurring in S. One MS. (d) has Rivede. Witte (Prol. p. xx) states 1 Dr Barlow says that A had Rende altered to Vede. It appeared to me that the word was Vende with the n partially erased, but the V un- altered. ' 48 ' had Rede, which shews how the readings might easily be confused. COLLATION OF SELECTED PASSAGES. 279 that he found it in fourteen only out of more than 400 MSS. Scartazzini argues that Vede is more poetical, less common- place and obvious, and moreover is not improbably a remi- niscence of Virg. Georg. ii. 81, Miraturque novas frondes et non sua poma. Gregoretti on the other hand defends 'Rende,' though stating that it occurs in one only of the Venetian MSS., as against 'Vede' in 16 (App. p. 13). The difference, it should be noted, between Vede and Rede or Rede is very slight, and the change in either direction might conceivably have had an accidental origin. In this case again the almost unsupported reading Rende appears in the great majority of Editions, but it is not recognized by a single early Commentator. Vede is found distinctly in Bocc., Veil., Barg., Dan., Benv., the last- named noting, ' attribuit proprietatem videndi arbori.' IV. 23. si mise. This was found in no less than 156 out of more than 180 MSS. examined on this passage. The few exceptions were as follows : si mosse CKu? 3.61.63.73.108. [9] mi mise RQ(.pjkvfanx 1.27.44.74.75.91. [16] mi disse BFXq^/ 18.82.85. [8] The line in F continues, e si mi fece entrare. g has fui mise (?). 68 has si messe. Si mise is no doubt original, the more commonplace ex- pressions si mosse and mi mise having an unmistakably conjectural air about them, nor, if either of them had been original, would si mise have been likely to have been substituted for them. Curiously enough there is a similar interchange of readings in Purg, ix. 69. For this use of mettersi, meaning to 'enter' or 'proceed,' compare Par. ii. 5 'Non vi mettete in pelago,'and Boccaccio (quoted in Vocab. Tram.s.v.) 'si misero per un giardino.' ' Mi mise' occurs in Inf. iii. 21. IV. 68. sonno. This reading appeared in 141 MSS. sono. This may be, and in some MSS. no doubt is, a mere ortho- graphical variety of sonno, but it may represent a different word, since 280 INFERNO. suono appears sometimes. Sono was found in KACD\cofbcdfgijlmstvwxM^-/;;//w 2. 3.4.1 1.12.21.25.29.30.35.36.37.39. 42.43.44.46.48.49.56.57.59.60.62.63.65.66.68.71.78.86.88.89.90.91.92.97. 100. 102.104.105.113.114. [108] quei signori ILQUAn$^poT9npruyzrtr/rz/ 5.6.7.8.132.16.17.19. 28.32.33.38 t (?) 41.45*58*64.69.72.74.75.76.77.80.95.98.103.107.110.112. [55] . quel Signor K*DEFGHTV 3 XA(marg.)I}DE6<.K l (?)\t.Trxaehq/ijgr(mar.) x i. i3 t . 14. 1 5. 18.20.24.26.27.31. 34.38 2 40.58i(?)73.79.8i.82.84.85. 108.109. [50] 0S have quel signori. K has queli signo (sic}. k and k have quelli signiori. The archaic use of quci for quel (though extremely com- mon in Dante and other old writers) has perhaps here, as it has in many other places, misled copyists. Sometimes it has COLLATION OF SELECTED PASSAGES. 28 1 caused them to modernize the text by the substitution of the more usual form, and sometimes to alter a verb or substantive into a fancied agreement with the misunderstood form, e.g. Inf. xxi. 14. In the present instance, some of those who were misled changed signor into signori, in supposed agree- ment with quei, by this change (be it noted), with the usual short-sightedness of emendating copyists, failing to give sense to the very next line, except by the very awkward device of making Che refer to ' altissimo canto' i.e. the 'grand style' generally. (So apparently Lan., " lo trattato di quelli e sopra tutti gli altri, &c." and similarly Buti, ' lo verso eroico dattilico sopra tutti li altri e eccellente.') Besides this, with signori the scansion of the line becomes impossible. Others changed quei to quel to secure a more obvious agreement with signor. The original reading was probably quei signor. Most of the early Commentators seem to read quel, and those who have quei explain it as plural. Among the latter is Benv., who finds a difficulty in Horace, 'qui scripsit in mediocri stilo, puta satira,' and Ovid (qui scripsit) 'in basso, scilicet comedia,' being included among writers of 'altissimo canto.' Having got over this, he awkwardly explains the next line as though c1ie...uola were plural, ' quia antecellunt alios ' (i.e. ' unusquisque in genere suo'). Another good instance of quei for the singular will be found in Ptirg. iii. 120 (where one modern translator at least has been ludicrously misled), and again in Inf. ii. 37, xxviii. 126, and xxxii. 114; and a parallel form stessi for stesso in Inf. ix. 58, and Par. v. 133 (in both cases however in rhyme, though this is not the whole explanation, since it occurs in other ancient authors, and sometimes in prose). Altri again is found for altro very commonly in Dante and elsewhere, and questi for questo, e.g. v. 135, xv. 48, &c. &c. See further on these forms, Diez Gramm. der Rom. SpracJien ii. p. 83 (Ed. 1876). He suggests their having arisen from an enclitic -c or -ce, so that Elli would come from illic ; esti, cotesti, &c., from is tic ; and other words would be formed similarly by a sort of false analogy, which is of course a very common phenome- non. We might compare with this suggestion the derivation often given of postea, prceterea, &c. in Classical Latin from 282 INFERNO. posteace &c. It is fair to add however that I cannot cite any instance of quci in the sing, used adjectivally with a noun in agreement. If this be thought an insuperable objection, I can only suppose that Dante wrote quel, and that some early copyist supposing that it took more than one to make a 'bella scuola' (see note on Inf. xx. 69), altered it to quei, and that another proceeded further, regardless of metre, to write signori in more obvious agreement with it. IV. 141. Tullio e Alino qrtux 1.3.4. 5.7.8.9.12.13. 15.16. 18.23.24.26.27.31. 32.4345 1 48.54.s8. 59^65.67. 75.78.82.88.92.95.98.101.106.108.109.110. Bat. 30. [83] Tullio Alino GKPWGl 1 .A 1 p8iKjipjms 1 vz' 2.6.10. 14^1 7. 19.22.28. 34.35. 36.38.40.41.46.49.51.62.63.66.72.74.81.89.93.114. Bat. 10.136. [46] Tullio e Lino LOV 4 Tr. ACEFSUVS&\F&gc 42.60.64.68.71.83.86.87. i oo. 1 02. 1 1 2. Bat. 102.136. [32] Tullio Lino V^Zto-vfn 29.56.57.69.90.103.113. [14] E Tullio e Lino D 21.33. Tullio ed Almo Q (? Is not this meant for Alino). Tullio e Alano S^oxj/wact 11.25.30.107!. [12] Tullio Alano Eyrehs 2 x/ 20.61. 73.77. Bat. 440. [13] Tullio e Livio MO*Ro 37. Bat. 431. Also Ed. Nidob. and Cast. Tullio e Merlino D. Tullio e Plinio I. Tullio e Alieno \ 53- Tullio e alicio (sic) 79.91. Tullio Anino p. Tullio cannio (sic} 97. Tulio daluina e (sic} 105. , succia dette k, suco dette V 2 , succio dette 82, s'accedette , savor decte (?)0, succedente Bat. 1 14. In short, out of about 250 MSS., only two (E and f) quite unequivocally present the reading SUgger dette, ' KOI'TOI TroXXol Xo'-yoi n-po? avro Kara- fifpXrjVTcu.' It may be added that succedette is found in II MSS. in the Trivulzian Library which I had not time to examine (so I was informed by the learned and courteous Librarian, Count Porro), and in all the 19 MSS. at Venice examined by Gregoretti. This would raise the number to about 280, or nearly so, as against two or three. This passage is discussed at length by Witte D. F. i. pp. 190 198, and is again referred to in D. F. ii. pp. 377, 378, where he justly expresses his astonishment at Scarabelli having again endeavoured to resuscitate a reading which "Man hatte denken sollen...das ein Menschenalter lang dariiber gefiihrt ist, sei sie endlich zu Grabe getragen." The late Dr Barlow (Contributions, &c. pp. 106, seqg.) also warmly advocates this unfortunate and almost unsupported variant. Most astonish- ing however is his statement that though the early printed Editions have sticcedette, yet the Commentators commonly explain the reading sugger dette. On the contrary, I make bold to say there is no trace in a single one of the early Com- mentators of their acquaintance with it, much less of their having adopted it. The following expressly quote or dis- tinctly imply 'succedette'' ': Ch. Anon., Com. Anon., Lan., Ott., Jac. (?), Pet., Bocc., Benv., Veil, Buti, Dan., Cast., Serr.: while Fals. Bocc., Anon. Fior., and Land, are indeterminate. Castel- vetro would evidently have caught at some such reading as sugger dette, had he known of it, since he complains of the weakness of the lect. vulg. He says there is no need to appeal to testimony ( ( di cui si legge') for so obvious a fact as it implies. 'Laonde e da dire che succedette a Nino suo 286 INFERNO. primo marito, e fu sua sposa, cio& sposa di Nino suo figliuolo (!).' He may well add : 'benche le parole poco s' adattino a questo sentimento.' V. 1 02. ?X(AvoirTuxa*bcefgjlmoqrt vrxyzb^eghltfnnqtw i . 4. 5. 6. 7 2 . 9 (?) I o. 1 2. 1 5. 1 6. 17.18.19.21.23.26.28.29.30.31.32.33.34.35.36.37.39.40.42.43.44.46.51.54.56.57. 58.59.60.63.72.75.76.78.79.81.83.84.85.86.87.88.89.91.92.93.94.95.97.98.99. 102.103.1042.106.108.109.113. Bat. 30.65.102.114.132.136.431.440. [149] Bocc., Benv., An. Fior., Buti, Barg., Land., Veil., Dan., Cast. mondo ACDGKLMPRV 2 ^?) EVAA*Uy aS^eiKpo-^f dh(?)iknpsuv ab^fjkl^prx?). 7 1 .8. 1 1. 1 3. 14.20.22.24.38.41.455.49.53.61. 62. 64.65. 66.67. 68.69. 7- 7 1. 74.77.82.100.101.104!. 107. 1 10. 1 12. 1 1 5. Bat. 10.55.129.142.416.428.429. [88] Lan. al mondo ancor appears in several MSS. above cited, e.g. LPR2tA0ien| kv 11.53.61.77.82.105. Bat. 10.55. U has rfiodo (sic) prob. = mondo. 80 has el mo acor (sic). h has el mon ancor (sic). 1 has del modo. g has tollata ochmodo (?) * has al mondo e anchor mi offende. /has toltalmodo (sic). 6 1 has al mondo che m' offende. 2 has modo, but the mark over 'o' seems later. In A the whole line is written over an erasure. In c the terzina 100 2 is omitted. The question between modo and mondo in this passage is a well-known battle-field of Commentators and Scholars, though the former reading, until comparatively recent times, was most generally, and I think rightly, accepted. The evidence of MSS. preponderates for modo, but the difference between modo and modo is so slight, and the distinguishing mark above the vowel is so often neglected, and even some- times inserted without any apparent meaning, that the read- ing becomes nearly ' indeterminate' so far as MS. evidence goes 1 . But though the difference in form is so slight, the difference of meaning is important and extremely interesting. Speaking quite generally, ' mondo ' would imply that Fran- 1 See further on this " Prefatory Note, &c." p. lii. COLLATION OF SELECTED PASSAGES. 287 cesca and Paolo were wrongly accused of the crime for which they were killed, and for which prima facie they appear to be here punished in Hell : while ' modo ' (though variously explained in detail) would mean that the circumstances of their death still troubled them. Dr Barlow in his Tract on Francesca da Rimini very strongly contends for mondo, which he maintains that Dante must have written. He argues, (i) that the reading mondo saves the reputation of Francesca, for which, if " innocent of the great offence," and guilty only of being surprised into a momentary act of imprudent familiarity (as Dr Barlow maintains was the case), she was naturally jealous ; since the popular form of the story (see the Commentaries passim) undoubtedly was that Francesca and Paolo were killed ' eV avTofywpw! The passage would then, it is said, be similar to that in Inf. xiii. 76 8, where Pier delle Vigne claims to rehabilitate his reputation. To this we would reply, (a) if they are innocent of serious offence, why are they here, ' in si dolente loco...ed a si fatta pena' ? (b) if Dante had meant to set right the general and current report on so serious a matter would he not have done so more clearly and unequivocally ? (c) the case of Pier delle Vigne is not strictly parallel, since he does not attempt to deny the crime (suicide) for which he is condemned, but that of infidelity to his master, to which his punishment in Hell has no reference whatever. (2) Dr Barlow objects that modo is so obscure that those who adopt it differ widely as to its meaning; e.g. Boccaccio, Buti, and Foscolo (see further on this below). But such an ' objec- tion' (?) would be fatal to a good many other passages ; and indeed it is so far from being an objection that it is rather an argument in its favour, since it would shew it to be lectio difficilior. (3) Dr Barlow argues that the reading modo, in its most obvious meaning, seems to confirm, in an inoppor- tune and inartistic manner, the worst interpretation of the vague story of Francesca's death. We reply, (a) that the meaning of modo is not necessarily more definite than that, death having surprised them in a state, and even in an act, of sin, there was no time for so much as the slightest symp- tom of repentance, (ft) This veiled admission of their guilt, 288 INFERNO. merely indicated (as in other similar cases) with Dante's own exquisite delicacy of touch, is neither inopportune nor in- artistic. For while Dante deals, it is true, most tenderly with the fate of one whom, and whose family also, he had known so well 1 , yet having in stern justice placed her here, he is not the man to waver, or to open the door to a false sympathy, by hinting doubts as to the extent of her guilt, or vague sug- gestions of exculpation. For Chi piu scellerato che colui Che al giudizio divin compassion porta ? Inf. xx. 29, 30. It is natural to object to Dr Barlow's explanation : 'If they were not guilty of this crime, why are they placed in Hell, and moreover in such very bad company ? ' Dr Barlow's answer appears inadequate. He says that they were in any case among those Ch' amor di nostra vita dipartille (1. 69) ; 'they had prematurely perished through love, their deaths had been sudden ; there was no time ' (whether they were actually sinning at the moment or not) ' to recommend their souls to God, nor to receive absolution, and so they had not died in that state of grace which was regarded as the neces- sary condition of reception into Heaven.' (Tract, p. 38.) But at any rate the popular story distinctly stated that they were so sinning, and, if not, readers of the Purgatorio will not need to be reminded of the numerous instances in which the occur- rence of a single word, or act, or even aspiration, in cases of sudden death, at any rate when this was not directly associ- ated with an act of sin, and though coming at the close of a sinful life (e.g. Manfred, Buonconte, &c.), made all the differ- ence between Hell, and Purgatory with Heaven in the distant, but certain, future. If therefore Dr Barlow's view be correct, why are not these unhappy lovers found in Purgatory ? For this is clearly not a case of vindictive condemnation, but of stern justice overpowering the tender and scarcely disguised sympathy of friendship. 1 Foscolo (Disc. p. 328) notes circumstances 'che fanno presumere che V episodic fosse o composto o ritoccato nelle loro case'. COLLATION OF SELECTED PASSAGES. 289 I conclude with explaining the views of the old Commen- tators on this passage. As to the reading, Boccaccio, Benve- nuto, Anon. Fior., Buti, Bargigi, Landino, Vellutello, Danielle and Castelvetro all have modo (though with various explana- tions). Jacopo della Lana alone adopts mondo, which he paraphrases by the words " cioe la nominanza e fama." As to the interpretations of modo, Boccaccio says, " il modo nel quale mi fu tolta ancor m' offende, cioe mi tormenta ;" and his explanation which follows of the history in detail is very interesting, since he distinctly declares his own belief that while there was great intimacy and even affection between Francesca and her brother-in-law Paolo, there was no impro- priety in their intercourse. He narrates the well-known story how the handsome Paolo personated his deformed brother Gianciotto on the occasion of the espousals of Francesca, and states that the affection she conceived for him at first sight, supposing him of course to be her real husband, remained after the discovery of the deception of which she had been the victim; as indeed was only natural. Under these circum- stances extreme jealousy and morbid suspicion on the part of Gianciotto were inevitable. Boccaccio then writes as follows : "Vedendosi ingannata sdegnasse, ne percio removesse dell' animo suo 1' amore gia postovi verso Polo. Col quale ella poi si congiugnesse mai non udii dire, se non quello che 1' autore ne scrive ; il che possibile e che cosl fosse. Ma io credo quello esser piuttosto fizion formata sopra quello che era possibile ad essere avvenuto, che io non credo che 1' autore sapesse che cosl fosse." Boccaccio therefore, while evidently believing Dante to have implied the worst, himself distinctly discredits that form of the story. He then proceeds to describe how Francesca was killed in an heroic effort to inter- cept the blow intended for Paolo. Buti's comment is somewhat amusing. ' Modo' he explains to be " ii modo di quest' amore (referring modo to 'amor... prese costui ' and not to ' tolta') che fu disordinato e smodato." And he then proceeds pedantically to object to Dante for- sooth, as though smodato were not Buti's own word, that his language is inappropriate, since ' Io smodamento propriamente non si puo chiamare modo ' ! and he adds that it is like D. I. 19 290 INFERNO. applying the term order to a state of disorder. Surely this is 'a most admired disorder' of his own. Foscolo (Discorso p. 327) offers three suggestions for explaining modo (i) the circumstances under which her life was taken (which had, as Shakespeare says, ' no flavour of salvation in them '), (2) the suddenness of her death, which left no time for repentance, (3) the barbarous character of the assassination itself. V. 126. fgilnorstuvS.g.io. 1 1. 12.13. 14. 1 5.16. 17.19.23.25. 26.27.28.30.31. 34.35.36.37.38. 40.41.42.46.49.53.54.58.59.60.64.65.67.68.69.70.71.72.74.76.77.78.80.81.82.84. 85.88.89.92.93.94.103.104.106.107.110.112. Bat. 30.65.440. [117] Far6 come BFGIlIZA 1 J3CDNy$ipovd(marg.)hjkmpqwdjgr(marg.}t'tv 1.2.3.18.20.21.22.29.32.33.39.43.44.45.48.51.56.57.61.62.63.66.73.75.79.83.86. 87.90.91.97.98. 100*. 102. 105. 108.109.1 13. [70] 24 has Ma fa come colui. 58 has Dir6 si come quel. 6 has lo tel dir6 como die piange e dice. 105 has Faro colui. V. 126 (bis). colui was found in 189 MSS. Only the following have colei CIZOA*^ vk# 3.1 1.39.54.56.57.67.85.86.90.93.106. 1 13. [23] Colei is evidently a pedantic adaptation to the gender of the speaker, but it is too rarely found to offer a discriminat- ing test. (See note on vi. 43 for other instances of a similar change.) The var. lectt. Faro and Dirb are interesting, and both strongly supported by MS. evidence. V. 135- This passage was not taken in hand as early as most others, and only about 150 MSS. were examined for it. fia diviso was found in all except the following, which had fu or fo : Bj(?)CLZo 1 (?)epv<|>j 22.29.48.54.56.57.61. 62.66.74.76.83.91^106.112.113. [26] 72 has nossia diviso. 24 has mai da me ne fie. The grandly defiant tone of fia diviso, as compared with the very tame quasi-historic statement fu diviso, leaves no COLLATION OF SELECTED PASSAGES. 29 1 room for doubt which is the true reading. The fals. lect. 'fu' seems generally characteristic of MSS. of the Vatican family (with a few others) though curiously it does not occur in the archetypal MS. 'B' itself. I can find no trace of ( fu ' in any of the early Commentators. VI. 43- a lei AB*DEGHLMOPSTUV 2 V 3 V*WXGl.Tr.^ J 5C J O J : J P6 < ;7Fr0A* IL%&tyfi8efyK\iLv%oTrT$tywQabdeghimr\opqrstwyzacfg'j7nflr!ivw 2.3.4.5.6.8.10. 11.14.15.16.17.18.19.20.21.22.23.25.26.27.28.29.30.31.32.33.34.35.37.38.39.40. 45.46.47.48.54.58.59.60.62.67.68.72.73.74.75.80.81.82.84.88.89.98. 102.103. 1 04. 1 06. 1 07. 1 08. 1 09. [145] a lui CFIKQ*RV l ZMQAayr\9po-uxFcfkluvxl>deMmytx 1.7.9.13.24.36. 41.42.43.44.49.51.56.57.61.63.65.66.69.70.77.78.79.85.90.91.92.97.1072.112. 113. [67] 92 has Et elli a lui. The change of a lei to a lui is similar in principle to that noted in v. 126, but since as yet the 'ombra' referred to (line 34) has not been identified, the feminine gender is still appro- priate, and is therefore used as it was in lines 37, 38, 39 ; and as it is again in line 47, which the would-be corrector probably did not notice, though a few MSS. (as will be seen in the collations) have consistently made the further change messo in that line also. The masculine egli first appears when Ciacco proceeds to make himself known, in 1. 49. Even then however one MS. (O) reads ella, probably influenced by the previous feminines. There is a precisely similar pair of variants in Inf. xii. 119, 'Colui (Colei) fesse in grembo a Dio,' the former having regard to the historical fact, the latter to the grammati- cal antecedent ' ombra... sola' in 1. 118. So again in P^^rg. xxiii. 133 the variants la and lo depend on the same considera- tions. Other illustrations will be found in Par. xxi. 67, where S. Peter Damian says, ' Ne piu amor mi fece esser piu presta' and again he uses serve of himself and the other spirits (1. 70). This is probably because they are commonly spoken of as vite, fiamme, margheritc, &c. (e.g. in this case, Vita beata, 1. 55); and indeed they do not appear to Dante in bodily form except in the Empyrean, where he is permitted to see them as they will be after the Resurrection and Last Judgment (see on this 19 2 2Q2 INFERNO. Par. xxii. 61 63 ; xxx. 44, 45). Par. xxii. 31 is a still closer parallel to the passage now under consideration, where 'Poi dentro a lei udi" refers to S. Benedict, described under the term marglicrita just before. VII. 1 08. malvage piaggie grige AIV^Z 8uimp// 3.29.43.44.49.53.56.57.60.68. 77.86.90.99*. loo*. 102. 105. 1 06. 1 1 3. [32] maligne piaggie was found in no less than 184 MSS. F and d have malvage coste. R seems to have malvge but the *v' (?) may be a ;/ or ;;. W has marine (!) piaggie grige. Fh 79 have e grige. 6 1 and no have montagne piaggie. 69 has piage strigie (sic). I am inclined to believe malvage to have been the original reading, though it has numerically slender MS. support. This by no means disproves its claim, since the euphonic substitu- tion (as I should consider it) of maligne may have taken place so early, and may have commended itself, when the alternative was known, so generally, as to account for the approximate extinction of its rival. Moreover the minority includes several MSS. having very good texts generally. The obviously euphonic change (found, as far as I know, in F and d only) 'malvage coste' is significant, as indicating a sense of the caco- phonous effect of malvage piaggie grige, but in that case the remedy is applied to a different word in the combination. It is at any rate difficult to see why malvage should have ever appeared in the text if the more obvious and more euphonious word maligne had been original. I do not however feel sufficiently confident to regard the passage as a test, but this is just such a slight blemish as the author might easily have overlooked, or indeed may have himself corrected in a later revision, as has been suggested in respect of other passages. In Inf. ii. 94 we have another case of a somewhat similar unpleasant combination of syllables in gentil ncl del, which one MS. has altered into la su ncl del. Perhaps we may add quella bestia for qucsta bcstia in Inf. i. 94. See this passage above, and other cases discussed in Proleg. COLLATION OF SELECTED PASSAGES. 293 p. xiii. There is nothing in the early Commentators to throw light on the reading here, except that Veil, comments on maligne. The remarks of Benv. are curious, from the (rather characteristic) spiteful illustration in the parenthesis: 'ad plagias grisas quae habent colorem subnigrum (qualis est vestis nigra monacalis), ita quod locus conformatur aquae," &c. See his caustic note on Par. xxi. 134. VIII. 78. ferro fosse in all the MSS. examined (at least 160). fusser fosse in none. This variant is noticed only because it is as old as the Commentary of Jacopo della Lana, which Witte puts at about 1328, see Dante ForscJi. I., p. 384, seqq., and Hegel, Ueber den JiistoriscJien Wertk der dlteren Dante Commentare, p. 1 2. I have never met with it in any MS., but it may well exist somewhere in the large number of MSS. which I have not seen. In any case it is clearly wrong, since (to say nothing of the absurdity of the statement itself) fosse would then be repeated in the same group of rhymes in the same sense. Of this I believe no example occurs in Dante, unless in the well- known and intentional exceptions of the words Cristo (where- ever it occurs in rhyme), and ammenda (JPurg. xx. 65 69), for different and obvious reasons, and also in the Latin quotation of Purg. xxxiii. 10, 12; though he does not hesitate to employ the same word when the sense is different: e.g. in Par. x. 143 and 147, nota occurs as noun and then as participle ; in Inf. xxiv. 2 and 6, tempra as noun and verb ; similar cases to this occur in Inf. xxiv. 37 and 39, and Par. ii. 139 and 141. Add Purg. viii. 64 and 66 (volse from volgere and volere]. The nearest approaches to an identical repeti- tion which I remember occur in Par. ix. 121 and 123, and in Par. xxxii. 125 and 129, where chiavi and palma are substantives in both cases, but different in signification. The case of piglid is somewhat similar in the two phrases dar di piglio and con mal piglio in Inf. xxii. 73 and 75. 294 INFERNO. VIII. 125. a men segreta porta in 75 MSS. a me in segreta porta in 98 MSS. a me segreta (prob. error for me) LWA* pS 9ijko(?> 36(?).tf. 65.66. a me in altra porta x. a me all' altra porta 101. a me la stretta 24. a men distinta xj/. 1' usaro a me serrar la porta 7. 1' usaro ad Re serrar la porta A. a men segretamente Gl. in men(o) segreta FVWz/ 29.82.102.105.113. The reading a men segreta porta is clearly right, the refer- ence being of course to our Lord's descent into Limbo, see Inf. iv. 53, seqq., and xii. 38 ; while the reading a me &c. has no meaning. The distinction however in the MSS. between men and me in or me is too slight a one to lay much stress upon. It is curious to note some of the daring conjectures that have been hazarded here by copyists. IX. 54- Mai non ABDEFHIKLPQRSTUV 1 V 2 V 3 V 4 WXZGl.Tr.^ J ffC/? J ^/^ iux 1.4.5.6.7.10.12 to 19.21.22.23.25 to 33.35. 36. 38.40.41. 42.45*46.47. 48.49. 54.55. S6 1 (?).57.s8. 59.60.61.63.64.66 to 72.74.76.78.79.80.82.83.84.85.88 to 94. 97.98.99.100.102.103.104.107.109.110.113. 114. Bat. 30.55.65.102.114.129. 136. [186] Ma non G/^AIiSinxs^hm^;// 2*.3.8.g. 11.20*. 34. 39.43.44.5 1.53-56*.62. 77.81.87.95.105.106.108. Bat. 132. [37] Ma non vedemo in peseo (sic) d. Ma nol Cobvx/. Ma noi Mp 65.101. Bat. 142. Mai noi O 112. Bat. 10. Mai noi il v. 75. Mai ne n*o 37.73. Mai che veggiamo 86. Mai non a 69^?). Mai non vi chiamo in toscho 1' assalto (!). Bat. 444. Mai non vedemmo ekl i. Ma non veggiamo in che fe' 1' assalto (!) 24. COLLATION OF SELECTED PASSAGES. 29$ NOTE. Several MSS. read veggiamo instead of vengiammo ; these words be- ing elsewhere interchanged, i\ has a mark of aposiopesis, thus : Ma non veng. &c. The differences of reading here are interesting and in- structive. Mai has the idiomatic and somewhat unusual sense of 'unfortunately/ or 'to our sorrow,' which has com- pletely misled the copyists, and it is amusing to see the various devices by which here, and still more remarkably in Purg. iv. 72, they have attempted to impart sense to the passage, since 'mal' taken in its ordinary sense gives a mean- ing just the reverse of appropriate, and, in fact, makes non- sense. Its usage here may be compared with male feriati in Hor. Carm. IV. vi. 14, and perhaps also with male veloci Remo in Ovid Fast. v. 452. There is again a case in Ovid Met. xiii. 58 : Quern male convicti nimium memor iste furoris Prodere rem Danaam finxit. This is specially interesting, since variants 'confict? and l con- tecti' occur here, which seem (as in the case before us) to be due to the supposed difficulty of the expression 'male convicti' Ulysses was actually and rightly convicted of the feigned madness, but to the sorrow (male) of Palamedes who convicted him. Further, the same idiom is found several times in Dante himself, and in most cases with a similar result of inviting blundering emendations of his text. (See Inf. xii. 66, Purg. iv. 72, Par. vi. 69, xvi. 140, xix. I4I 1 .) A similar use of mal is found also in Petrarch Trion. Cast. 1. 119, and Trion. Fam. ii. 100. The reading Mal, as 'difficilior lectio' in the proper sense of the term, may therefore with confidence be pronounced original. The old Commentators who notice the passage all have the correct reading and explanation ; such as, Bocc. (aver mal fatto a non vengiare), Buti, Barg., Dan. (mal fu per noi che non), Cast, (per mal nostro). So also Benv., who further curiously explains this idiom as ' vulgare tuscum et non lombardum,' adding " unde 1 The reading of one MS. ('14') here is amusingly audacious : Che molto ha visto il comune di Vinegia ! 296 INFERNO. vidi multos deceptos hie exponentes : ' mat non vcngiamo' idest bene vindicavimus : imrno debet exponi per oppositum sic... male fecimus quod non vindicavimus," &c. The very rare and as yet unpublished Commentary of Guido Pisano, of which a copy, formerly in the Sunderland Library, is now in the British Museum, explains simply and correctly thus, 'Male fuit factum quod non vindicavimus.' In the unedited Commentary of Serravalle, a similar explanation is introduced, with the curious warning : " Textus est fortis... Auctor intelligit totum oppositum de eo quod sonat volgare lombardicum ; hie quod sonat quod bene vindicavimus," &c. IX. 70. porta fuori (or fori) ACD(fore)EFGHIKLMOPQRSTUV 1 V 2 V 3 V 4 WXZ*Gl.Tr.yi 1 (?) D (marg.} EFGHIJMNQSUWLQAI^VQafrjMlrti. K\|xv 1 (?)|owpoT^<}>x^wsf9aeghijknopqtruvwxyz^bdm 63.86.100.102.105. [10] Dan. abbatte e frondi e fiori r (with ' al. porta ' in marg.} d (marg.}. abbatte frondi e fiori B 83. Bat. 102. (So also Buti and Barg.} abbatte foglie e fiori 28.91 (marg.}. abbatte e foglie e fiori Bat. 415. abbatte e butta fori 80. gitta fuori \x 108. 24 has e batte. Of the 19 Venetian MSS., one only (Bat. 293) has i fiori (auct. Grego- retti), and the other 1 8 have fori or fuori. The change of text here can, I think, be traced almost with certainty. The original reading fuori or fori was mis- taken for, or altered into, fiori. (In Inf. xxv. 144 fuor, fior, and for are again found interchanged.) The text being then in a state of unstable equilibrium,' since fiori evidently would demand an aiticle, a further alteration to' i fiori (as found in a few MSS.) naturally followed. Possibly the reading (also found in a few MSS.) portai fuori (or fori}, i.e. porta i (viz. i rami} fuori, may have helped to suggest the change to COLLATION OF SELECTED PASSAGES. 297 i fiori. The following are some points either against fiori or in favour of fuori. (i) The poet is describing the effect of a hurricane on forest trees. We have (with the reading fuori} a graphic picture of the successive steps of destruction upon the branches: (i) it breaks them off (schianta), (2) it dashes them down (abbatte), (3) it carries them away (porta fuori). With the reading fiori we have an inappropriate subject introduced, and the miserable bathos withal, of the blowing away of the flowers after the crashing of the forest trees. In fact with fiori two further alterations are needed in the text: (i) obviously we should require the article i fiori, as above ex- plained, corresponding with Li rami: hence the reading i fiori in BD and a few other MSS. : and (2) either a connecting particle between the verbs schianta abbatte (since porta has been parted from them), or a still more radical alteration such as is found in three or four MSS., and also in Bargigi and Buti among the Commentators, viz. abbatte frondi e fiori (or abbatte e foglie e fiori as in Bat. 415), by which we are let down more easily from the branches to the flowers through the medium of the leaves. (ii) On the ground of what has been called ' transcrip- tional probability,' the changes from fuori or fori, first to fiori and then to i fiori, are far more likely and intelligible than those changes in a reverse order would be, since either fuori or fori might be thought to be a mistake for fiori rather than vice versa. (iii) Finally the imitation of Virgil Georg. ii. 441 (Quas animosi Euri assidue franguntque feruntque) is unmistakable, and abbatte e porta fuori almost exactly reproduces the ex- pression franguntque feruntque. The large majority of MSS. preserve the reading fuori, though Witte admits fiori into his text presumably in de- ference to the evidence of his selected MSS., three (and per- haps the fourth also) of which have fiori 1 , and two exhibit even the tertiary reading i fiori. I cannot but think it singu- lar that these readings should have the support of Foscolo, 1 The reading of A seems doubtful. 298 INFERNO. Tommaseo, Rossetti (Com. Anal. i. p. 251), Witte, Philalethes, and many other distinguished names; besides the Aldine and Crusca Edd., (though these go for little). In Dr Barlow's Collations 53 MSS. had fuori, and only seven fiori. In the MSS. collated by myself, the preponderance for the former is still larger 1 . The early Commentators are almost all so vague or brief here that it is difficult to be sure of the reading they followed. Serravalle translates, 'scindit transportat' (i.e. porta fuori}. IX. 125. Che seppelite dentro in about 210 MSS. Che son seppelite &c. Ma99 7.65. Che son sepolte G 109. Chen (? = enno) seppelite Ks/8. The introduction of son, either before or after seppelite, is obviously (as often elsewhere) a shortsighted emendation, it not being observed that the construction must be held in suspense, though only till the next line, where the main verb 'fan is found. X. 65. gik letto il nome 2.3.5.7.9.10.1 1.13.14.20.27.30. 32. 109.110.112. Bat. 129. [104] elettO ADFHKLOSTV 3 V 4 XGl.^(w depqt i. 4.6.8. i8.i9.22.24.26.3i.4346.55i.6i. 62.69. 70.72.88.89. 106.108. [65] detto IV^tfpvdfft 29.48.56.57.66.68.86.90.102.103.105.113. [22] X has alletto. r has gik di costui. U has gik di costor si eletto. 63 has si letto. Letto is pretty certainly the true reading here, being a little out of the common, at any rate as compared with detto, which is an obvious lectio facilior, and quite commonplace. 1 I find that in a good many points I have been anticipated by Blanc Versiich, p. 92, &c., but the above note was written when absent from his book, and I am not aware of having previously seen his discussion of this passage. COLLATION OF SELECTED PASSAGES. 299 We have an exact parallel to this use of leggere in Purg. xxvi. 85, even including the combination with nome: In obbrobrio di noi, per noi si legge, Quando partiamci, il nome di colei, &c. We might compare also Par. xxvi. 18. As to the reading eletto, though it is found in a great many MSS., it may have had its origin in an early clerical error, or possibly in a confused recollection of Inf. xxii. 37, 8, where this word and nome occur together in near connection, though the sense of the passage is quite different. XI. 90. vendetta ABCEHLMOPQRSTUWX ACEFGHJMOQSUT&QA n2$p8ei]9iK^voff 54.73.106. E '1 savio Duca ver lui F. E 1 savio Duca a lui d. E 1 Duca mio ver lui 1. Lo savio Duca mio 0. Lo savio mio Duca inver lui h. Lo savio Duca inver lui 20. Lo savio mio Duca grid& 51. Lo Duca mio inver lui 80. E '1 savio Duca mio gridd 24. Lo savio mio grid6 inver lui F . Lo savio mio Maestro gridb w. Bat. 415. Lo mio Maestro inver lui ko 1.37.68. COLLATION OF SELECTED PASSAGES. 301 d (marg.} has Virgilio. f(marg.} has cio& Virgilio. 62 is def. Here we have the case, I think, of a true reading standing out clearly among several false and conjectural ones. That reading is, Lo Savio inio inver lui, which is the one found in the majority of MSS. The expression Lo savio mio is rather an unusual one. Unless I am mistaken, it is found only once again in the whole Commedia, viz. in Inf. xiii. 47, though quel (or z7) Savio is applied to Virgil in Inf. vii. 3, and to Statius in Purg. xxxiii. 15. This would perhaps ac- count for the great variety of alternative titles for Virgil exhibited here by the MSS. The introduction of Virgil's own name however is the most interesting and widely supported of the variants. It seems most probable (as I have already observed in the Proleg. p. xx) that this arose from a marginal gloss (such as indeed actually occurs here in ' d ' and '/"') to elucidate the reference of Lo savio mio. This was mis- taken for a correction (or a var. lect.) for ver lui or inver lui, and so substituted for them in the text. Another instance of the probable embodiment in the text of a marginal gloss will be found in Purg. xxi. 25, as is pointed out by Witte Pro 1. p. xli. Special interest attaches to this variant, as it is one of the most characteristic peculiarities of the ' Vatican ' family (see Appendix). As a further indication of the im- probability of the reading Virgilio here, it is interesting to note how seldom Virgil's name is mentioned in the Inferno ; at most, I believe, five or six times, though it occurs very often (certainly more than twenty times) in the Purgatorio. This can scarcely be accidental, since the occasions for re- ferring to or indicating Virgil are even more numerous in the former 1 . Perhaps this is due to the same kind of feeling which causes Dante never in the Inferno to mention by name, but always by a periphrasis, Christ, the Virgin Mary, and Beatrice. The name Beatrice does indeed occur once, viz. in Inf. ii. 70, but under circumstances that do not con- 1 Inter alia, Virgil disappears in Canto xxx. of the Purgatorio, and becomes a 'KOH/>OI> TTPOO-WTTOC ' after Canto xxvii. 302 INFERNO. stitute any real exception to this statement. See the peri- phrases for Christ in Inf. iv. 53, xii. 38, xxxiv. 115, &c. ; for the Virgin Mary, Inf. ii. 94; for Beatrice, Inf. x. 131. It is also noticeable that the Sun is never mentioned or referred to as affording a datum of time while they are within the limits of the Inferno itself, as I have pointed out in my Time References in the Divina Commedia, p. 61. XII. 66 Mai fu. This was found in 179 MSS. Ma fu only in 10, viz. CLR^X/^z,, 5.9.24.70. This is similar to the passage discussed at ix. 54, but the recognition of the idiomatic use of mat here not being so absolutely essential to the intelligibility of the passage, the palpably secondary reading Ma is found in a very few MSS. Its rarity may perhaps make it worth recording, and it also throws light on the changes occurring in the other similar passages cited under ix. 54. XII. 125. cocea ABCDETGHlLOPQST\JWXTT.ACEFfffOQSV TAOAHS ^^^y^(lj]9\.K^voTTa-T\^F(}abeghikinnrtuvwyza6^e/ 2 ^/iihftnp^ i to 12.14.15. 16.17.19.20.21.22.24.26.28.30.32.33.34.36 to 44.46.47.48.52.53.55.59.60.64.67 to 75.77.78.79.81 to 89.91.92.93.95.101.104.107.108.109.111.114.115. Bat. 10.30.65. 1 02. 1 14. 1 29. 132. 1 36. 142.442.444. [186] copria KMRZ^Z?Z,FfloXpv4osdfjlpqsx/iy^rt/7/ 13.18.27.29.31.35. 45*49.51.54.56.57-58.62.63.65.66.76.90.102.103.106.110.112.113. [57] p has cocea li lor piedi. M^ have pur cocea, &c. r has lo piede. 7.1 1 have cocea lor li piedi. N and Bat. 431 have cuoce pur. / has si che copre. 23 has si che copria li piedi. 6 1 has si coceo (sic] tutti i piedi. 80 has toccava pur. A has che 1 cocea. Gregoretti reports the Venetian Codici thus cocea 1 2, copria 5. Copria has quite the air of a correction, it being a much more obvious word than cocea to apply to a river. The corrector COLLATION OF SELECTED PASSAGES. 303 forgot that it was 'sangue bollente* (see lines 47, 101, 117). Cocea has the support of the more numerous and, I think, generally more authoritative MSS., though most of the 'Vatican' family have copria. It is noticeable that this is again a case of B and D in agreement as against A and C (see note on xi. 90). It is hardly necessary to point out that the use of cuo- cere for ' burning ' in a generic sense, like coquo in Latin, is thoroughly Dantesque, e.g. Inf. vi. 105; xvi. 49; xix. 79: Purg. ix. 32, &c., &c. XIII. 20 and 21. ben, e si vedrai. bene, si ved(e)rai. bene se vederai. These are so similar and so easily confused, that I have not felt sure about the distinction between them in my notes, and have therefore thought it better to group them together. One or other of these readings was found in 152 out of 173 MSS. examined. The following were the exceptions : se (or si) tu vedrai Uptown 7.46.108. e vederai E\h. e tu vedrai I 62.86. che tu vedrai 33. che tu ben vedrai St. ben che vedrai i. Per6 pon ben mente e si vedrai (!) d. In line 21 the reading torrien was found in about 160 MSS. out of about 1 80 examined. The exceptions were as follows : terrian A 1 (?). terran hs 20.54.69.73. tornan (-nen, -nin) Fi?biqt 7. torni ^wz*. cosa che torria a 65. che pien fede (!) L. I suspect that si was mistaken for si or se, and that the rhythm was improved by the insertion of tu. The ' e ' must be regarded as ' indeterminate ', with bene immediately pre- ceding. (Compare note on xix. 3.) " Look, [and] so (si) shall you see," seems in any case better than " Look if (si) you shall see." There can, I think, be little doubt as to the reading torrien in preference to terrien, tornan, &c. in the 304 INFERNO. next line. These, and still more the unsupported (?) con- jecture damn in the Ed. Nidob., have clearly the air of cor- rections, due to the superficial difficulty, that the startling facts proving true ought not to destroy, but confirm the credi- bility of Virgil's statements concerning them. The obvious meaning however is, 'that you will see things which would have been incredible if only heard from me and not actually seen:' "a work which ye shall in no wise believe, though a man declare it unto you." Virgil says in effect that it was cosa ch' io avrei paura, Senza piu prova, di contarla solo. Inf. xxviii. 113, 4. Virgil in fact here puts into practice the very prudent maxim inculcated by Dante through his mouth in Inf. xvi. 1 24 6 : Sempre a quel ver ch' ha faccia di menzogna De' 1' uom chiuder le labbra finch' ei puote, Pero che senza colpa fa vergogna. Compare also Aristotle, Poet, xxivi IO irpoatpelcrdai 8et dSvvara etKora /j,a\\ov rj Svi'ara airidava. XIII. 63. sonno e' polsi KGHIcycjv 7.62.63.78. [13] Bocc., Anon. Fior., Benv., Veil., Cast. sonni e' polsi. AHLMOPSTWXTr.^W06"A0AnSap^vir(?) pjtu^ 9. 36.38.41.66.101.110.111. [18] sensi e' polsi BFR&DSp^kmvifiter 8.24.34.37.43.44.45.48.54.65.68.70. 8 1.88.106. [30] Guid. senno e' polsi z 5.76. senso e' polsi w 53.73 le vene e' polsi DGlQy 1 ZABCZ)JWyi\Q\\>ty has 1' ame (?). R has P aere. Fhas che rive (?). 1 1 has largie (sic) = /' argine ? 53 has che all' acqua fanno. X has P api fan rinbombo. T] has che al arne. o- has calanni. I have no doubt whatever as to the true reading here, but unfortunately for critical purposes the certainly genuine arnie (i.e. swarms or hives of bees) is often almost, and in some MSS. quite indistinguishable from the spurious and conjectural arme 1 . Where however a MS. has distinctly arme it may so far be condemned. Clerical error may account for this change, but ape is an audacious and deliberate substitution of an 1 See Prefatory Note, p. liv. COLLATION OF SELECTED PASSAGES. 313 easier word somewhat similar in meaning. U Arno and /' aere are so far bolder still, that the meaning is changed, especially as it necessitates the further alteration of fanno into fa, which is sometimes carried out. The Ottimo comments on arme thus : " simile al suono del ripercuotere dell' arme insieme." But in spite of this it is clear that the sound of a distant waterfall is more appropriately compared to the hum of a beehive (arnid) than to the clash of arms, being " uno rombare continue o sordo," as Lan. puts it. I cannot find any authority for arme in the old Commen- tators. Barg. reads arvie (a word which I cannot find in Vocab. Tram.}, but explains it as though it were equivalent in sense to arnie, ' le arvie nel rombo 1 .' Zani de' Ferranti strongly defends arme, (i) as being the reading of most MSS., or, as he most inaccurately adds, ' potevan dir quasi tutti' This is absolutely the reverse of the fact, as the above collations show. (2) Because arnie "non e derivabile da nessuna lingua, o morta o viva, ch' io sappia." Diez also admits that no satisfactory derivation can be assigned to the word. But this objection is immaterial, since the word itself exists beyond doubt, and is found in other authors and even in prose. (See Vocab. Tram, s.v.) Cast. (Comm. k. /.) quotes an instance of its use by " Pietro Crescenzo nel libro xii. dell' Agricoltura." XVI. 14, 15. Ora aspetta Disse a costor cdjqswxyz^rz/ 10.13.18.20.21.27.29.35.41.42.452.46.49.56.57.59.60.63.74.90.97. 98.102.105.112.113. [63] disse Aspetta AC*EFGHLOPSTUWXTr.^^rni>nY8 W osf be 44.48.51.53.54.55.61.62.65.68.69.70.78.82.84.88.89.106.107.108.109. [96] disse Ora aspetta Disse a costor AAa*evxaig// 4.7.30.47.52.64.67.72. 76.77.110.111. [25] disse Ora aspetta A costor fyxcniw 2.33.91. disse Ora aspetta Che a costor k 85. disse Ora aspetta Et a costor n. 1 Possibly as r and / are frequently interchanged (esp. in Florentine) arvie may merely represent alvie or alvei. Thus we find arquanto, artro, arberi, &c. 314 INFERNO. disse Aspetta Disse a costor tofybx 3.36.53.54.110.111. disse Aspetta Che a costor AFG&fc 11. poi Aspetta Disse R. disse Ora m aspetta a. disse Ora t' aspetta 24.75. disse Ancor specta 9. disse Ancora aspetta h 73.79. disse Aspetta Perche a costor 25. disse Ora aspetta Perche a costor 38.39. disse Aspetta Par la a costor si &c. 61. disse Aspetta Ecco costoro se vole e (with ora in marg.}. disse Aspetta Qui a costor C*. dicendo Aspetta Disse a costor 101. NOTE. The entries of these minor variations at the beginning of 1. 15, though I trust correct as far as they go, are probably not exhaustive. I believe the collation was not in some cases carried beyond 1. 14. In the great variety of readings here we may, I think, safely be guided by the principle that that reading is to be preferred from whose pre-existence the others can be most naturally derived or accounted for. This is I believe the case with the reading Ora aspetta Disse a costor, c. which is not however that of the majority of MSS. The delay in the occurrence of Disse, which usually precedes the speech it is intended to introduce ; perhaps also the slight ambiguity involved in the collocation of the words, Disse a costor, probably induced some one to make the alteration disse aspetta, the next line being then variously filled up, e.g. Ora a costor Qui a costor Che a costor A costoro Et a costor or (violating rhythm) Ecco costoro ; sometimes too even retaining the vain repetition of Disse a costor without alteration. All these extraordinary variations seem to be only different ways of adapting the text to the introduction of Disse in the previous line ; but with Ora aspetta in 1. 14 there is little or no variation in 1. 15. This ex- plains also the origin of the sort of ' conflate reading,' disse Ora aspetta in 1. 14, combined with various forms of 1. 15 as before, which is found in several MSS. COLLATION OF SELECTED PASSAGES. 315 XVI. 104. Trovammo in nearly 160 MSS. Sentimmo DYA (marg.) CDEtyuydpyr 21. Udimmo In. Ritrovammo 77. Faceva risonare 73. This is not a variant of any special significance, except as perhaps indicating some relationship among the comparatively few MSS. that have changed the common reading Trovammo into Sentimmo or Udimmo. One or other of these words was apparently thought more appropriate in connexion with risonar, and also in regard to the statement of the next line. XVII. 20. Che parte sono. This was found in about 190 MSS. Che parte stannowas found in 14 only, viz. BDIPpo-slq.27.6o.66.8o.85. F and d have Che stanno parte. Tr. and have fanno a riva in the previous line. The whole line in v is E parte in acqua e parte in terra. /has e altra in terra. This passage is one of several exhibiting a tendency of copyists to introduce uniformity of expression into clauses between which there is, or appears to be, an antithesis, or some other relation of correspondence. Compare Inf. xix. 23, xxiii. 83, xxv. 1 20, xxxiv. 13, 53 ; Purg. v. 88. Evidently in the present case the verb stanno in the previous line sug- gested its repetition here, but the circumstances are not quite the same, apart from any considerations as to elegance of diction. XVII. 95. altro forse B^?) DEKMRV^^^avfpdnrw 13.23.29.42.56.57.71.74. 90.98.99.102.105.113. [32] Benv. alto forse LQ/^Ctywcxz 7.9.10.18.27.31.35.60.84.92.104.1 12. [21] Buti. altro forte AK*Z*DJNU'2opg}ovaegk 8.14.21.22.37.51.58.63.66.70.82. 85. [30] Anon. Fior. alto forte CPUV 4 *Gl.Tr. J &FgrSI>Si;iXsariqsuvy bhjmqv 25.30.38.41. 45.47.49.54.62.65.72.73.76.79.91.97.101.106.107.110.111. [52] Barg. 316 INFERNO. alti forte HOSTV^V^WXeAn^KHnro-reflmrt^^ 1.2.4.6.12.15. 16.17.19.20.26.32.34.36.39.40.48.59.75.78.81.103.108.114. [52] altri forte b. altri forti/6i. alti forti k. altra forte 5. altre forte 69. alto forze (? = forse) v. altre sorti P 33. altre sorte 1 1*46. alti sorti 68. altro periglio i/ 77. Ad alto tosto forte I9\\v 3.43.44.52.83.86.95. Ad alti tosto forte G-y 109. Ad altro tosto forse V. Ad alto tosto ch' io F^. Ad alto gridb tosto ch' io 24. Ad alto tosto ch' io forte C. A se mi strinse e tosto ch' io x. Al dicer forte siccom' io montai E. Ad altro forse ch' io tosto MQ. The copious list of variants here is very curious, and I think easily accounted for by the somewhat unusual sub- stantival use of forse in the original reading (as I take it undoubtedly to have been), altro forse. I should consider alto forse to be probably (at least in some cases) a clerical error, since alto and altro are very frequently confused in MSS. (e. g. in B passim. See also numerous cases of their interchange in the marginal variants to Witte's 4to. Ed.) The other variants are conjectures, more or less bold, to get rid of the misunderstood forse. Forte also will have to be taken substantially (though see below), and in the sense of ' diffi- culty.' Forte no doubt often has the sense of ' difficult,' as in Par. vi. 1 02, and Conv. Canz. i. sub fin.; cf. fortezza = diffi- colta in Conv. i. ^ fin. &c. &c., but I am not aware of forte being elsewhere thus used as a substantive. The substantival use of forse is found in two passages quoted in the Vocab. Tram. s.v. ; but it is fair to say that the definite article is added in both cases. Dante however uses in forse ' in doubt,' or ' in danger,' at least three times : Inf. viii. i io, Purg. xxix. 1 8, Par. xii. 41 : and so also does Petrarch, Trionf. Mort. i. 58, 'e poi che 'n forse, Fu stata un poco.' COLLATION OF SELECTED PASSAGES. 317 In illustration of the word altro see a list of such deliver- ances given by Scartazzini in his note on Inf. viii. 97. The curious and unintelligent variations in the order of the words should be noticed, as indicating that the copyists are groping after a lost meaning. Perhaps they meant to take forte adverbially with avvinse, and wished to bring it nearer to its verb, ad alto being joined with montai and interpreted ' on high.' So at least Bargigi and some later Commentators quoted by Blanc ( Versuch, &c.), though I cannot find any authority for his attributing this perversion to Landino. Vel- lutello however takes forte even with sostenne (Ed. 1544). Barg. gives the above explanation as an alternative, though himself preferring ' tosto ch' io, fatto forte ed ardito, montai ad alto,' &c. I do not find any explanation of the passage in the other older Commentators except Anon. Fior., who reads altro forte and explains, ' piu volte et a forte (sic) cose.' Buti (without any comment) adopts the reading alto forse, but notes the existence of the variant forte. Benv. reads altro forse, explaining it briefly and simply as ' ad aliud dubium.' Castelvetro reads alto, but hesitates between uniting it with sovvenne (then strangely explaining it, ' cioe di sopra, ne' cerchi superni') or with m avvinse ('cioe al collo'). Being dissatisfied with the sense thus obtained, he makes the brilliant critical suggestion, ' che questa voce alto sia corrotta per traspozione di lettere, volendo dire lato ' ! Giuliani in his recent Ed. of the Div. Com. (Florence, 1880) adds another guess, 'All 1 alte fosse' which is, I believe, as absolutely devoid of any MS. authority whatever, as it certainly is of poetic taste. The curious reading altro periglio in the related MSS. i, /, and 77 is evidently a reminiscence of Inf. viii. 99. XVIII. 12. rende figura EHOSTUV 1 V 2 V 3 V t WXGnr.A l EGffJ'/MQS[7TQA II < l>atT]K|j.oo-TsF 1 acegkmnortuvxyz^^-/y/^'<7Wjr 1.3.6.7.8.9 (se rende). io. 12.13.14.15.17.19.22.23.24.26.28.30.32.33.34.39.40.42.45.47.48.49.52.59.62.63. 64.65. 67. 69.70.7i.72 1 (?) 74.75. 76j.78.8i. 84.88. 89.92.95. 99. 101.104. 107.108,1 io, 111.112.114.115. Bat. 30.114.132. [132] 3l8 INFERNO. Lan., Off. (?), An.Fior. (?), Benv., Butt, Barg., Cast. rende sicura (or sigura) ACDKO(marg.) PZ bdijqsmttfr 4.5.ii.i6(?) 18.25.27.29.31.35.36.37.38.41.46.54.56.57.60.66.68. 80.82 (Le parti). 83.90.98.100.102.103.105.106.113. Bat. 10. [66] rendon (or -den) sicura (or sigura) EFGIMROS-ySeiXxUX/^/ 21.43. 44-5i-58.79-86.9i.97.io9. [30] rendon figura QD(marg-.) Vw 2.85. render figura 77. dove sol rende figura 3o.45i-65.95. 1142- dove '1 [or dov' e '!] sol rende figura B 45 2 .47-52-7 6 i-78- Cast. dove '1 [or dov' e '!] sol rende sicura 54.106. dove '1 [or dov' e '!] sol rende fotura . La parte dove rende men sicura 61. La parte dove si rende men sicura 53. Serr. (?) La parte dove men rende sicura R\fh 20*. La parte dove si rende sicura v. La parte dove non rende sicura p\ (rendon) j 66. La parte da (?) si rende men sicura f. La parte dove non 6 ben secura e. E la parte D. Da quella parte ch' ei rendon sicura 73. Bat. 55 has Da parte dove son rende figura sicura. This is a passage in which the true reading can without difficulty be discerned, the variants being obviously due to the misunderstanding of a longish construction 1 . The con- struction rendered quite literally is ' Such a form as, when to guard the walls more and more moats surround the forts, the place where they are presents, such a picture here did these make.' Thus the word Quale at the beginning of line 10 does not find its substantive till we come to figura at the end of line 12, which would be ample time for the average copyist to have forgotten it, though it is strange that the immediate occurrence of Tale imagine in the next line should not have recalled Quale figura. Meanwhile the images of walls, ditches and forts would have so occupied his mind that the change of rende figura to rende sigura or sicura would seem imperative. This change is so slight that it may pos- sibly in some cases have arisen as a mere clerical error. The text was now in what we have elsewhere described as a state 1 See other cases of this at xxvi. 25, xxix. 46, xxx. 18 c. See also Proleg. pp. xvi. &c. COLLATION OF SELECTED PASSAGES. 319 of ' unstable equilibrium,' since rende sigura could not stand with the plural verb and its subject (ei son) just before. Hence we have a tertiary reading rendon sigura in many MSS. This last consideration, operating apart from the former, would account for the comparatively rare rendon figura found in some MSS. Another conjecture due to over- looking the suspended construction is that of D, which adds the word 'E' at the beginning of the line from the supposed need of connecting the verbs in lines n and 12. The change in ' e' is amusingly audacious: "dove non e ben sicura" ! I see that Landino's text (Ed. 1564) has " dove il Sol rende figura" though this is not noticed in the Com- mentary. Blanc ( Versuch &c.) appositely quotes (if indeed illustration be necessary) Conv. iv. 7 " tutto copre la neve e rende una figura in ogni parte." Benvenuto explains the passage very well. He describes it as " comparationem satis communem cujus sententia est satis confusa et litera valde intricata." After some explana- tion in detail he paraphrases thus: " Quelle (sc. valli) face- van quivi...tal imagine qual figura la parte dove son li castelli rende (i.e. ostendit, repraesentat), dove piu e piu fossi cingon quelli castelli per guardia delle mura." Castelvetro has a curious note : " M 'e stato detto che in Alamagna e una fortezza cosi fatta, e nell' Atlantico di Platone n' e descritta medesimamente cosi fatta; e, se Dante avesse letto Platone, affermerei che avesse questa comparazione quindi." He further adopts the blundering reading dov e il sol, and ex- plains it as meaning 'in the world above,' in contrast with the Inferno ! Serravalle (in his unedited Commentary) para- phrases, " circumdant castella in parte ubi se reddunt minus secura, i.e. ubi facilius possent expugnari." He seems there- fore to have had the thoroughly bad reading 'dove si rende (or rendon) men sicura.' XVIII. 42. Di gik veder. This was found in about 155 MSS. Gik di veder BGIKZ56^F*a-yXvx<}>acdn 2 oqsz^r 18.27.29.31.35.53. 56.57.61.63.69.79.86.90.108.109.113. [45] 320 INFERNO. Di veder gik DZ>ip/t/ 2 21.77. Di gik costui veder g. Di gik di veder 38. The true reading here is, I doubt not, that of the majority of MSS. viz.: Di gia veder. The alterations Gia di veder or Di veder gia were intended to arrange the words in a more usual and natural order, or at least what seemed so prima facie. But they overlook the fact that gia is meant to qualify veder and not son. Perhaps too the change was to some ex- tent due to a misunderstanding of the rather peculiar (though thoroughly Dantesque) use of esser digiuno in the sense of "not to have had experience of," e.g. Inf. xxviii. 87, 'Vor- rebbe di vedere esser digiuno,' i.e. " would wish never to have seen." The Vocab. Tram, quotes an expression from the Amor. Vis. of Boccaccio, " Forse ne vorrebbe esser digiuno," i.e. "Per- haps they would wish to be without it " ; which looks like an imitation of Inf. xxviii. 87. So in Par. xvi. 135, ' Se di nuovi vicin fosser digiuni ' = " If they had not had experience of new neighbours": and again in Par. ii. 75 digiuno means 'scant' or 'lacking of 1 '. It will be found in a similar sense at Purg. xv. 58, though used metaphorically, just as pasciuto occurs in the converse metaphor in Purg. xxvi. 103. Thus the sense of the present passage is literally " Of having already seen this one I am not inexperienced " : so that the seemingly unusual order Di gia veder is just what is required. The older Commentators generally pass over the passage, but Benv. explains thus, taking gia with digiuno : " non son gia digiuno di veder costui, quasi dicat, imo satur, quia pluries viderat eum vivum ; quasi dicat tacite, quod oculus ejus fasti- diebat et indignabatur videre virum nobilem qui tarn viliter prostituerat carnem suam." The passage is rightly translated by Ford, Carlyle and Longfellow, but most strangely mis- understood by Gary: "Of him... not yet mine eye hath had his fill." The order Gia di veder seems to characterize MSS. of the Vatican family, though it is not confined to them. 1 Dr Carlyle aptly quotes a similar use ofjejunus from Cicero, Orator cxxx, jejunas hiijus,,,orationis (lures. COLLATION OF SELECTED PASSAGES. 321 XVIII. 55. lo fui colui was the reading found in about 190 MSS. Io son colui occurred in 33, viz. GSV^E^Qv^h^tdj^: 3.7.25. 29.5 1.53.54. 56.57.62.68.74.80.90.99*. 106. 109. 1 1 3. The reading Io fid colui is, I think, to be preferred, since Io son colui has the appearance of a euphonic change. The question as to the principle on which Dante uses the past or present tense in passages of this kind will be found discussed under Purg. v. 88 (q. z;.). This however is one of the inter- mediate class of passages in which either tense might be justified. Cf. Io son colui in xiii. 58, where there is no varia- tion in the seventeen MSS. collated for the Inferno. In xix. 67, ' M ' changes sia into fui perhaps with an eye to the expression in 1. 70. XVIII. 91. senno CDG 1 MV 1 ZGl./ 7 Sn\vvxhklu&?;> 7.29.30.41.45.51.52.53.56.57.58. 80.90.95.98*. 101. 102. 105. 1 10. 1 13. [42] segni i to 6.9 to 19.21.22.23.24.25.26.27.28.31.32.34.35.36.37.39.40.42.43.44.47.49. 54.55.59.60.62 to 68. 70.71. 72.73.74.75.77.78.79.81.82.83.84 85.88.89.91.92.93 94 < 97.98 1 (?)99. 100.103. 104. 106. 107. 108. 109. 111.112. [175] cenni Id 86. senni PjJicov 20*. 33. 46.48. segno w 38.69.114. sendo (sic) 61. NOTE. The line begins, Quivi con... in many MSS. * I must confess to a strong preference for senno here, though it has the support of not very many MSS. (i) There is much point in its repetition from 1. 86 (compare its use in Inf. xvi. 39). The senno which enabled Jason to perform the great exploit referred to in 1. 86 (" Ne potea altrimente fare si egregio fatto se non avesse grande animo...e gran prudenza." Land.), that same 'senno' was now prostituted to this base deception. D. i. 21 322 INFERNO. " av yikv ovv 6 OTCOTTO? 77 Ka\oaDXo/3x ilt elm 4.5.36.39.49. 76.80.83.91.106. [27] li piedi e le gambe Ayb/7.i 1.58. ^48 have li piedi con le gambe. Tj 2 has li written above dei. This is an instructive instance of the tendency to assimila- tion or uniformity which has been noticed under xvii. 20 (q. v.}\ since here, as in the other passages there quoted which exhibit the same tendency, there are two ways of effecting the assimilation, and some MSS. adopt one, and some the other method. Doubtless then the original reading was li piedi e delle gambe. The fact too that until we reach the next line there is no apparent government for the genitive delle gambe might possibly have had some influence in cor- rupting the text, though it will be seen that the change to le gambe is that most rarely found. XIX. 64. tutti was found in 87 MSS. tutto was found in 71 MSS. The following minor variations were noticed : spirto storse allora i piedi U& (marg.} \l. spirto storse tutti piedi 7. spirto sparse allora i piedi 77. spirto atorse e piedi k. spirto e storse ei piedi 30. This is a case in which the arguments for tutto and tutti appear to be nearly equally balanced. Blanc ( Versuch, p. 177) contends strongly for tutti, chiefly on the ground that the 326 INFERNO. word tutti was likely to have been altered because of its inapplicability to the two feet 1 . Per contra however it might be argued that the singular tutto may have been altered because of its inapplicability to the plural piedi, by those who were unaware of the quasi-adverbial use of the adjective, which we have had occasion to notice and illustrate under iii. 8. The parallel passages quoted by Blanc are striking (esp. Inf. xxxi. 15, " Drizzo gli occhi miei tutti ad un loco"), and would of course fully justify and illustrate tutti h. 1., but instances of the adverbial use of tutto are also very numerous. The MSS. seem to be not very unequally divided between the two readings. XX. 30. passion porta FGILV^nS* o.^\.w^\v>y.lrw 24.35.39.49.5 1.54.60^73. 75.78.82.923.106. Bat. 444. [35] Lan., Dan., Cast. passion comporta BCEHK 2 MOPQ 1 STV 1 V 2 V 3 WXZGl.Tr..4 1 # 1 J F2 U rAOAS^?) 8K* ( j.vpTvs/' 1 abefgjkmnoqrstuv aegjkptux 1 .2.4.7. 1 5- 1 6. 1 7. 1 8. 19.26.27.28.29.30.31.32.36.37.38.40.41.43.44.45.47.48.52.55.56.57.58.61.62.65. 66.69. 70.72.77. 79.84.85. 88.89.90.91. 92^93.95. 97. 101.102.103. 104. 105. 107. 109. 110.111.112.113.114. [127] compassion porta ADK 1 Q 2 RU^*CZ?//7^^/5F4>p Y *9ace 34(?)6.9. 12. 14. 15. 16.18.19.22.23.24.25. 26.28.31. 1 Some however (as Gregoretti and Fraticelli) take passion porta &c. in the sense of 'feels pain' at God's judgments (with this compare in .hat case Par. iv. 67), and they object to the other interpretation or to the reading compassion, that 'compassion'' would have for its object not the judgments of God, but the sufferings of the lost, resulting from those judgments. 2 Here, as elsewhere, we must be careful not to quote the authority of any Commentary merely on the strength of the text printed with it, unless this is distinctly confirmed by the language of the Commentary itself. 328 INFERNO. 32.34-35-36-374-43-44-47-49-54-57-59-62 r 72.74-75-77-8i.88.89.92.95.io3.io6. 111.112.114. [96] se fosse in quel DGS*UGl.r*8o|//Vx;> 5.8.65.78.99.101.109. [22] se fosse a quel OF 53. se fosser a quel 107. se fesse in quel P. se fosse pel cam. 79. se fesser (or fesson) quel AV^sh^*^- 1 1.20.33.39.46.91*. 102. 105. Benv. se fusser (-en) quel -y 5 1. se stesse quel p. se fosso (?=fosson) k. se fusse qui il cam. 80. se fesse quel paese 13 (a curious o/ioiore'Xfvroi>). se 1 fesse quel ^ 21. chi fesse quel i. chi fosse quel d. se fen 94. This is another case in which the misunderstanding of an archaic form has certainly led to depravation of the text 1 , the steps of which process can be distinctly traced. The form fesse for facesse was obsolete and unfamiliar. Nannucci (Analisi, &c. p. 630) recognizes this form, and it is still more fully illustrated by Mastrofini (Diz. Crit. &c. p. 277) as occurring, though rarely, in Tasso and Ariosto, and in archaic prose 2 . It should be noticed that Dante himself employs it several times, e.g. Par. v. 20, xvi. 146, and xxiii. 45, besides the well-known passage in Inf. xxxiii. 59, "ch' io '1 fessi per voglia Di manicar" ; where ' D ' modernizes it into facessi, which of course will not scan. This form must not be con- fused with fessi si fe' occurring in Ptirg. xxx. 98 and else- where. Fesse then, not being understood, was altered into fosse or fusse, for which it was probably mistaken, since fesser and fosser again occur as variants in Par. xxix. 45. This is the reading of the majority of MSS. It has however the dis- 1 Compare Inf. i. 28, iv. 95, v. 38, &c. 2 There are several other similar archaic forms in this verb, as though derived from fere instead of fare orfacere: e.g. imperf. feva and fea, fut. ferd, and still more commonly a perf. fei, festi, ft?' or fee (see Inf. xxxiv. 19, Purg. xxxii. 12, Par. xxxii. 19, &c., and comp. the collateral form feo, e.g. Inf. iv. 144, Purg. xvi. 106, c., formed in the same way as rompeo, combatteo, poteo, &c.),femmo (Inf. xvii. 32), fcste,fenno and fere (Inf. xvi. 21, xvii. 17, 89, &c.). COLLATION OF SELECTED PASSAGES. 329 advantage of giving no sense whatever, a preposition being now required for quel cammino. Hence the further alteration ' in quell or ' # quel] to complete the construction. Another question is between the singular and plural of the verb. The singular is no doubt correct, the readings fesser,fusser y fussen, &c., being corrections due to the consideration that three bishops had surely a right to a plural verb. It should be noted that the preceding potria is also singular. This kind of ' a-^rjfia TLivSapiKov ' is not uncommon in Dante. I have noted the following instances of plural nominatives with a singular verb: Inf. i. 41, iii. 50, viii. 28, xiii. 43, xxi. 114; Par. ix. 49, xvi. 40. XXI. 14. fa. . . volge. This reading occurred in 1 53 MSS. (Some of these however have chi rivolge instead of altri -volge.} fan (or fa) .. .volge in 15 viz. FGIMQ-yXpajv 24.36.61^109. fan (or fa)...volgon in 20 viz. Lpf>z<# 33.45.46.47.48.52.68.76.83.95. 106.107.111. The following sporadic variants also occurred : Alcun fa...alcun volge 80. Chi fa suo remo e altri volge 73. Altri fan remi e chi rivolge 66.86. Altri fan remi e altre (sic) rivolge 51. Altri fa volger remi ed altri sarte 9. The true reading is no doubt that of the large majority of MSS. Altri fa remi ed altri volge sarte. The archaic altri was mistaken for the plural (comp. quei in iv. 95 q. v.) and one or both of the verbs altered hence the readings fa or fan and volgon. The substitution of chi rivolge for altri volge is perhaps to avoid the same imaginary false concord, with an eye no doubt also to the similar form in the previous and follow- ing lines. It is easy to trace the same motive in the other changes recorded. XXI. 46. convolto 63.65. 67.70.71. 74.81. 82. 85. 88.89.90 99. 102^?) 105. 1 1 1. 1 12.113. [82] Dan., Veil., Cast. 330 INFERNO. con volto PQ 1 (?)SWTr.(?)nefgistf/(?) 2.5.17.19.38.45.51.55.59.61.64. 72.84. 1 14 (cum volto). [30]. col volto ACDEGIMOQ*RUV 4 Z.FFAn^napY8KXvopviflgrz'w 2.12.16.18. 19.20.27.31.32.35.36.42.46.49.60.61.65.70.71.74.75.77.81.84.85.107. [86] Huli, Barg. temetti non tenesser DIMPV 1 (?)Z^ 1 Z>^t/ar](rT4)\|/bfilopx gu*x 4.7. 1 1. 21. 29.37. 39- 5 3- 5 5*. 56. 5 7. 59.68.69.78.86.90. 102.113. [45] [ 54.106. leri cinque ore piu che 97. This is not a passage of much importance. The reading of the large majority of MSS. is probably correct, the various changes in the order of the words being probably intended to improve the rhythm or euphony of the passage, and they are at any rate deliberate and not accidental in origin. XXI. 113. Mille dugento ABCDEFGmKLOPQy l V*V 3 V*ACZ)FQS(7VrA acdefghijklmopq. [45] Mille dugent' uno MZbn (n having e sessantd). [FAA are here reported on the authority of Scarabelli.] This curious variation was unfortunately not observed by me in time to collect fuller information. I have discussed the probable motive of the second reading given above at some length in my " Time References &c" (pp. 46 segq.\ and need not repeat the discussion here, especially as my information respecting the MSS. is so imperfect. I will only now add the evidence of the three of the early Commentators who recognize the latter reading, viz. Lan., Fals. Bocc., and Benv. : since when writing before I did not quote the first two, and could not quote the last, as Mr Vernon's splendid edition of that Commentary was not yet published. Lan. says : " E dice che correano anni MCC. uno e sessanta sei, cioe MCCLXVII. anni erano trascorsi &c." : Fals. Bocc. : " Per queste parole dei intendere che nel MCCC. fece Dante questo libro, e pero dice che dal sessanta sei e uno infino al cento ha XXXIII. " : Benv. alone recognizes both readings thus : " Et hie nota quod aliqui textus magis moderni habent aliam literam sic: Mille dugent itno e sessanta sei, sed ista dis- cordantia accidit propter discordantiam opinionum, quia, ut jam dixi, aliqui volunt quod Christus vixerit triginta tribus COLLATION OF SELECTED PASSAGES. 333 annis, alii quod triginta quattuor, et de hoc audivi magnam disputationem ; sed prima opinio et ita prima litera videtur melior." XXII. 101. SI ch' ei (or che) non teman ACEHIOPQRTUV 1 V 2 V 3 V 4 WXZG1. ABCUQATI2&yliK\)Lvm\ty(:tg/i&mnflrtuvx2.3 4.5. 6.9.10.11.12.16.18.19.20.21.23.25.26.27.30.31.32.38.39.42.43.44.49.55.56.57.59. 60.67.70.71.72.73.74.75.77.78.79.81.82.84.86.90.97.102.104. 105. 107. 108. HI. 112.113. [129] SI ch' io non tema BDFGKLMSZ?\$"FA^ap0opp/ 22.35.61.80.106. [11] The original reading here is clear enough, though a con siderable number of MSS. have gone wrong ; viz. ' Si che (or che i or ch' ei) non teman ' the reference being obviously to the " Toschi o Lombardi" of 1. 99 whom the Navarrese craftily offers to summon. Perhaps this allusion was overlooked, and the speaker was imagined to be still referring to his own fears (with a reminiscence perhaps of 1. 92), though the occurrence of io in the next line might have suggested that there was probably a mutatio subjecti. Perhaps also 'chei* was mistaken for die t, i.e. io (for which of course it often stands) and the verb was altered into fancied agreement with it. XXII. 142. sghermidor (or -tor) in about 60 MSS. schermidor (or -tor) in about 150 MSS. (Several blundering forms occur, which however seem readily to refer themselves to one or other of the above types : e.g. ghermitor, sghrimitor, scremitor, schenitor, schehnitor, sghremitor &c. &c.) The difference in form between these words is so slight, and the liability in a hurried collation to confusion so great, on account of the double interchange (i) of g and c, (2) of d and t the former being essential, and the latter accidental and immaterial that I do not venture to reproduce the re- port of my collations in detail. The general result may be taken to be not far different from that registered above, so 334 INFERNO. that the correct reading sghermidor (or -tor) is probably found only in about one in four of the MSS. Though the difference in meaning of the two rival words is very important, the difference in form is not only (as I have said) slight, but it is also one very liable to arise accidentally from considera- tions of dialect or orthography [e.g. cridare and gridare, alcuno and alguno, torgier torccr (A in /;// xxxi. 126). Cf. also grudele, grede, grude, giascuno, &c. in L. These I believe are Tuscan vulgarisms]. However in the present case the wrong form occurs, I think, too frequently to be accounted for as a mere dialectic variation, and the change as a matter of fact completely alters the meaning of the word, the forms schermidor &c. giving no sense whatever to the passage. Probably both sghermidor and schermidor, being rare words, were little understood by many of the copyists, and the form written was more or less a matter of chance, though schermire and its derivatives, being much more familiar than sghcrmire, &c., would naturally be more often adopted. The real dis- tinction is as follows : Sghermidor is from sghermire which is the reverse of ghermire, i.e. disghitinare, as Vocab. Tram. explains it. So Benv. comments here : " id est subitus sepa- rator istorum, ita quod disghermitivit eos ubi erant primo ghermiti inter se " ! Ghermire or gremire denotes the violent grasping with the claws by birds of prey. So Guido Pisano (unedited Commentary) explains " ScJiemitor (sic} t gremire est ilia invasio quam una avis in aliam facit et sic contrarium istius vocabuli (i.e. of the word in the text) est gremire." The word ghermito occurs just above in 1. 138, and also in xxi. 36. Diez (s. v. gremire} derives it from O. H. G. ' krimman ' and explains it " mit den Klauen packen." Scher- midor is from schermire (cf. schirmeii) to fence, or defend. See Inf. xxi. 8 1. Schermidor is found also in Petrarch, Trion. Cast. 1. 49 seqq., in a passage which well illustrates its meaning : Che giammai schermidor non fu si accorto A schifar colpo, ne nocchier si presto A volger nave dagli scogli in porto, Come uno schermo intrepido ed oncsto Subito ricoperse quel bel viso Da colpo &c. COLLATION OF SELECTED PASSAGES. 335 Comp. the word schermi in the disputed passage, Purg. x. 126, where however there seems no doubt that the word is to be taken in its natural and proper sense of 'screen' or 'pro- tection.' The strange and grotesque variations in the form of the word registered above seem to betray the copyists' ignorance concerning it. In consequence of the difference between these words being so slight, and the accidental interchange of c and ^so frequent, I cannot lay any stress on this as a test-passage, though it is important to insist on sghermidor being certainly the true reading. As further instances of the interchange of letters in question, compare sego (Purg. xvii. 58) prego, lago, drago, Aragne, siguro,figo, &c. &c., in all which cases c is also found. And especially note // tegni di Galieno in Conv. i. 8 as a trans- literation of ?; re^vr). XXIII. 83. Dell' animo col viso. This was found in 170 MSS. Dell' animo del viso a9in. Coll' animo col viso ov. Dell' animo e del viso CFXu// 44.69.77.108.110. Coll' animo e col viso FM^r. Dell' animo e col viso A 38*. Nell' animo col viso Bpjks9b._,cfghijklnstvwxz/7#w 3.8.9.10. 13.2O*.33.394 1.452.46. 5 1.54. 58.62. 63. 66.69.7 1. 73.76.80.82. 85.94.95. 99 1 .io3- 107. 108. 109. 1 10. 1 1 2. [92] I and 86 have ribandendo. 47 has ribiadendo. There can be little doubt that in this line the technical, and therefore little understood, Ribadendo has been altered into the common-place but far less appropriate Ribattendo. Ribadire means ' to rivet ' ; or as the Vocab. Tram, explains it, to knock back the projecting point of a nail so that it cannot be extracted. Diez (s. v. the French 'river'}, while assigning to ribadirc a similar meaning, expresses doubts as to its derivation, though both Diez and Brachet connect the French 'river' with the Danish rive, and the former mentions a Prov. word ribar with a similar meaning. The metaphor of ' riveting' is in this context most graphic, whereas ribattendo, which does not seem to have passed beyond its literal and etymological meaning, would evidently be much less appro- priate, if suitable at all ; but being a far more familiar word it would have been likely to be substituted. It occurs three times in the Commedia (viz. Inf. xx. 43, xxi. 13 ; Purg. xvii. 87) and in all cases I believe with the simple sense ' to beat again,' or something closely akin to it, though in the last named passage other and less probable interpretations are given, and notably by Daniello. The early commentators are not sufficiently explicit here to be worth quoting 1 . It may 1 The naive note of Bargigi's editor (Zacheroni) is a good illustration of the false critical principles from which corruptions like the present arose. " Se 1' origine del verbo ribadire trovasi in ribattere, non puo negarsi che il primo sia un' alterazione di questo secondo fatta in Toscana, e che il solo ribattere sia vocabolo Italiano da tutti inteso, e da Dante adoperato" ! COLLATION OF SELECTED PASSAGES. 337 be noted that a very similar case of corruption arising from ignorance of a technical word occurs in Purg. xiii. 2, where risega has been replaced by rilega, &c. XXV. 120. Per 1' una parte e , , v \ altra. This was found in 153 MSS. ClGil ) una parte e , ,,,Y altra GL-ET^-yiisfksvw&^z/ 1.37.52.68.78. 79.103.107. [24] Per 1' una parte e per 1' altra BF (altro) KQdeiX.vTloquxsb 2 ciksuwxz^or 13.54.58.65.69.85*94.99. 106.112. [32] Quanto 1 villan m 7.23.66.71.101. Quantal villan p 24. Quale al villan w. Qual & il villan L 10. Qualtel villan (sic) q. Come 1 villan Tr.fi 68. 1). I. 22 338 INFERNO. This is an instructive instance of corruption of the text arising from a long construction, of which we have already had an instance in xviii. 12. Others will be found in xxix. 46 and xxx. 18. This is a case very similar to that in xviii. 12, but the construction here is a good deal longer, since quante is not supplied with the substantive Incciole, with which it agrees, till the 4th line below. The construction here (again comp. xviii. 12) is emphasized by Di tante fiamme in 1. 31. The text therefore can be decisively settled here, and the prima facie easier readings Quale, Quando, Come, &c., may be unhesitatingly rejected. None of the commentators except the Ottimo (which has Quale) seem to countenance any of the false readings. XXVII. 21. Issa Ii2\*vhknwri.7.i3.2o*.86.93.98 2 .Q9*.ioo*. Bat. 55. [19] An. Ficr,, Ott., Benv., Land., Veil., Dan. Istra ACDEHIKMOPQTUV^V^XZTr. BDHIJMQS V A*0 AH 2$^p-ySKfMTTS9abcdegil (?) opqrsuxyz ab^cefgijklmpquvwx 2. 3.4. 5.6. 10. 1 2. 1 4. 15.16.17.18.19.21.22.23.25.26.27.28.29.31 to 38.40.41.42.45.47.49.55.56.57.58. 59.60.64.67.70.72.74.77.78.82.84.85.88.89.90.91.9-2.94.98! (?) 102. I03(ista0 I04j(?) 105.109.111.114. [144] Lan., Guid., Serr. Ista RS/rnl/ (?)/" \.hn 1 1.24.39 (?) 53 (?) 54.62.69.71.79.81. 106. 107. 108. [22] Butt, Barg., Cast. Istar BGa. Stra FG1. CETty fv/ 43.44.61.75.83.113. [16] Sta Lfivpo-xwjm 9.48.66.73.76.87.95. 101.115. [18] In stra or Instra VV^o 63. 1 10. In sta 1 1 2. Stare. Estra 30. Extra 52. Estu 97. Isso or ten va 65. Statti o va U. Ora ten va 68. Istate un poco non t' ad. d. Istra non ten va x. 4 had apparently istra. 1 has either istra or ista. " al issa ;; is in the marg. of *dr. In 80 the word is omitted and a space left blank. COLLATION OF SELECTED PASSAGES. 339 This is a case in which I think the vast majority of MSS. must certainly be thrown overboard 1 , the copyists having gone astray (as often elsewhere) over an unfamiliar word, viz. Issa. Yet this word is twice besides used by Dante, viz. in Inf. xxiii. 7, where in fact its meaning may be said to be explained by Dante himself as being indistinguishable from that of ' mol i.e. 'now': and again in Purg. xxiv. 55. On Inf. xxiii. 7 Vellutello comments: "Mo et Issa ; che tanto suona in Lombardia ognuna di que' due dittioni quanto in Toscana Hora e Adesso." So also Landino. In the same place Lan. (unless it be, as Scarabelli conjectures, a gloss) says that mo is vocabolo romagnolo and issa is vocabolo loinbardo. Buti in the same passage just reverses this statement. On Purg. xxiv. 55, however, Buti seems to forget what he had said before, and states that issa is vocabolo hicchese. Benv. at Inf. xxiii. 7 says, ' aliqui Tusci dicunt mo, aliqui Lombardi dicunt issa.' It seems probable from the expression 'parlavi Loinbardo' that the following words, purporting to be a quotation from Virgil's language, embody ' Lombardisms ' in the case of issa and adizzo, and this because Virgil was himself ' Lombardo ' as described in Inf. \. 68. Serravalle comments " loquebaris mine lombardum, i.e. lombardice dicendo istra'' According to Diez issa is an archaic form of esso, which is often found in composition, e.g. adesso, sovresso, &c., and is merely ipsa (sc. hora). Blanc notes that isso is used for esso in Par. vii. 92. Istra, which is found in the large majority of MSS., seems to be a mere blunder, and a word without any meaning at all. Its predominance implies that the error, whether clerical or otherwise, was introduced into a very early copy. The same may be said of fsta, Stra, In stra, &c. &c. Zingarelli however 1 Among the Commentators, however, it will be observed that the majority are for issa; e.g. Ott., Anon. Fior., Benv., Land, ('cioe adesso'), Veil, ('cioe hora'), and Dan. In Buti and Barg. (who generally co- incide) we find ista, and Lan. has istra, but with no explanation. A similar difference between the testimony of the Commentators and the MSS. is to be noticed at Inf. xiv. 126, xxxii. 136; Purg. ii. 13, xiii. 2, xviii. 83, xxi. .61, xxii. 5, xxv. 9, xxvii. in ; Par. ii. 141, and many other places. Thus the early Commentators often supply a very important cor- rective or supplement to the evidence of existing MSS. ; just as patristic quotations do to the surviving MSS. of the New Testament. 22 2 340 INFERNO. (Parole e Fotme, &c. pp. 153, 4) endeavours to justify the form istra as a variety of issa. See the collation of /;// ix. 93, where some of these strange forms are again found in some MSS. Sta and similar emendations are of course in- consistent with ten va, whence arose the more ' heroic remedies' of U, ' Statti o va ' ; or of d, ' I state nn poco ' ! In Purg. xxiv. 55, already referred to, the unfamiliarity of ' issa ' has caused the copyist of ' 85 ' to substitute ' assat.' As a curiosity of perverse ingenuity may be mentioned a suggestion of Canello that this line is to be explained by reading it as a transliteration of Greek, thus : *Ix9b 2 (?) f ijlotf^py^//?'/! (?);/- 3.9.11.13!. 16.30.36.37.47.48.52.53.54.65.66.73.75.76.78.79.80.91.95.101.106.110.111.112. [52] Niuno or Neuno LU^es.^ 46. [8] In niuno 8/ 43.44. TV has si gran fallo uno. d has Non vede mai tal crudelta Nettuno. 24 has Non vidi ma' nessun fallo a Nettuno. The reading Nessuno would probably be at once set down as a mere blunder, except that the occurrence of Niuno also suggests the possibility of its being an intentional, as it cer- tainly is a most tasteless, change. As I find one or other of these variants in about one-fourth of the MSS. examined, and signs of alteration in four or five of those which at present have the true reading, I have thought that this might perhaps be included among the test-passages. 344 INFERNO. XXVIII. 135. giovane or Giovane A(giovane sic)KMQS(G;iovaxie in Q and (giovane)94(Giovene) 1 10. Bat. 429. [22?] Giovanni CDEFGHILMOPQSTV 2 V 3 VnVXZGl.Tr.^CZ?^ZM9 grAOAn^fi o.p - y8t'l.Kp,vjrp(rTv<}>x|'s abdfhikmopqrstuvwxyzrt&/ a$$elifliK\vovfyusf<)abcdfghnpqrsuvxyzafod<'gJMHfigrt'iv 2 . 4.6.7.8.9. i o. 1 1 . 13.16.17.18.19.20.21.24.25.27.29.30.31.32.33.35.36.38.39.41.42.43.44.45.46.47. 102.103.104.105 107.108.110.111.112.113. [151] 1 This is not at all uncommon however in Dante, as may be seen from the numerous examples in Blanc's Dizionario s.v. essere. COLLATION OF SELECTED PASSAGES. 353 E bel suo AV 3 ej 72*101. E suo bel i. El bel suo GHPRTWXTr.AfAji/n-pTJrafe/ 1^.12.14.1^.22.26.28*^4. 40.51.52.59.66.70.75.79.81.84.88.89.92.98*. Bat. 444. [44] E '1 suo bel DE{tykol// 3.23.37.48.49.61.77.80.83. [19] Del suo bel E. E '1 bel Pol. 76.; E '1 suo Pol. w 65. E lo suo Pol. Bat. 55. Del suo Pol. 0-54.106.109. E del suo bel 68. Ebbe 1 suo Pol. St 63. Vide '1 suo Pol. \. Onde '1 suo Pol. ^. Ed el suo 1 14. E '1 del suo x (? cler. error). This is again a very instructive instance of the way in which a lengthy or inverted construction leads to a tampering with the text, the cause here being very obvious, and the con- jectural devices quite transparent, and also unusually varied and numerous. The ' fons et origo mali ' was the inversion by which the genitive del suo Polidoro precedes by nearly two lines the word accorta on which it depends. The first timid change was to substitute bel for del. This occurs in a few MSS. only, and clearly could not stand, as the definite article was a necessity. The next step was to remedy this harshness, either (a) by supplying the article, which gives the reading (found in more than 40 of the MSS. here ex- amined) E 7 bel suo (compare what has occurred in ix. 70 and xxi. 46) ; or else (/3) by introducing the easier order E suo bel (found in 'i'); or (7) more completely by combining both these changes and reading, E 7 suo bel, as found in 19 MSS. These then may be called tertiary readings. We also find the curious (and by reason of metrical difficulties rather un- common) instance of a ' conflate ' reading 1 in E, which has Del suo bel Polidoro. Finally we have bolder conjectures, which are less hampered by the form of the original text, found in one or two MSS. each, such as Ebbe 7 suo, Vide 7 suo, E 7 suo, Onde 7 suo, &c. I find no trace of any but the true reading in any of the old Commentators. 1 See note in Prolcg. p. xxi. D. I. 23 354 INFERNO. XXX. 44. Falsificare. This was found in 169 MSS. Falsificando CDPSUV 2 Tr. U^f^\\i\.\*dklmnrtti 3.1 1.22.23.36.39.43. 44.452-49.52.54.55.62.65.76.77.80.82.91.93.98.106.110. [48] Falsified Mo. Per falsificar A. The reading Falsificando in this line is another instance of corruption arising from a long construction, combined perhaps with the desire for uniformity already more than once noticed. The intervention of 1. 43 has caused it to be overlooked that the verb sostenne is still waiting for its object, and the occurrence of the gerundial forms Falsificando in 1. 41, and Testando &c. just afterwards, suggested a corresponding form here. Some correctors under the influence of the former con- sideration only, i.e. not recognizing the need for an infinitive to complete the construction, have marred it by reading Falsificb or Per falsificar. \ XXX. 115. e tu falsasti. This was found in 193 MSS. tu falsasti (om. 'e') in 18 only, viz. CACDE\rfbjl 47.49.51.62.68.73. 105. [18] e falsasti was the reading of 23 and 86. The reading e tu, which is that of the large majority of MSS., is evidently to be preferred to the lectio facilior which omits 'e.' Its idiomatic use here is the same as that illus- trated under xix. 3, to which reference can be made. The erroneous reading is in this case so rare that I have placed this passage in the second class, though entertaining no doubt as to the correctness of the reading e tu. XXXI. 75. E vedi lui was found in 202 MSS. E vedi lei in DACqdfpefirx- 11.21.30.39.45.52.60.76.100. [20] E vedi la che Gp-yW; 14.46.79.95.104.106.109. [14] E vedi la (or Ik ?) giu Yd. . COLLATION OF SELECTED PASSAGES. 355 The readings lei or la for lui arose from the superficial error that a feminine was required to refer to soga. It is however quite clear that it is the corno that is referred to. The leathern thong (soga) is around Nimrod's neck, and to it the horn is suspended (cJie il tien legato), which itself surrounds his huge breast like the stave of a cask (doga a word oc- curring again in its literal sense in Purg. xii. 105). Instances of a somewhat similar change of gender to effect a concord superficially more suitable will be found noted in the colla- tions at Inf. xii. 47 and xiii. 78. XXXII. 47. Gocciar su KC~DIKLMOQRUV 1 V 2 V*ZACDEF/MQS17&ATI.V apet9i.Xv5op(rv*np8ST ] i\ 1 v^irtrrv^xsfpaeghjmortuvy^?) ahilptwx 2.3.6.8. 10. i ^.16.17.1^.20^2^.26.^2. 35-37- 38i-4-43-44-48.49-54-58.59i-63-64-65-66.67.69.7o 1 .7i.72 1 .73.74.75.76.77. 79.84.89.92.94.95. 97.98.104!. 105.106. 107.108.110.112.114, Bat. 10.30. 142. [117] ' ha ragion 41.45. y appears to have had ai ragion, which has been altered into di ragion. The strange and (as far as I can see) utterly unintelligible reading se tu ai ragion seems to have been a mere blunder arising from the incredibly short-sighted consideration that ha should be hai to agree with tu, though its real verb piangi follows immediately. Still more strange is it that I find this led. fals. in half of the MSS. examined. I should suppose therefore that it was a very early blunder which has been mechanically repeated. I find no trace of it in any of the Commentators, and naturally, since they would have been called upon to explain it. A similar interchange occurs in several MSS. in Inf. vi. 47, but in that case either reading gives good sense. XXXIII. 26. Piu lune gia ORUZADFGfflJMQSIS Tu&jUC&pxzlPcgmnrx 6.14. 16.21. 23.29. 3o*33.42.46*.55*56 1 (?) 57.69.73.77.90.91.94.100.102.104.110!. 115. Bat. 428. [57] Piu lume gia ADFGIMQV^#FA02piX 2 p|/b 2 hijUnsuw//7/ 4*.5.7*-9. 10. 13. 17. 20*. 35.53.58.602.65.66.68.71. 80.82.86 (Iumi)99*.io3. 105. 109. 113. Bat. 10.55.102.142.429.440. [62] Com. Anon., Ott., Benv., Butt, Barg., Land., Veil., Dan. (lumi), Serr. Piu lieve grieve) gia. BCEHKLPSTV 2 V 3 V 4 WXGl.Tr.C(piu lieve 61.62.64.70.75.76.78.81.83.84.85.87.88.89.92.93.95.97.98.101.107.108.112. Bat. 114.129.132.444. [Ill] Guid. Piu le vie (0r levie) Al&z? 18.19.27.31.39.59.67.74*110*. 114. [15] Piu brieve gia v 1 1 (breve). Piu volte gik er\.ug\o...ni;lfar del giorno m' avea motrato piu lume." But there is nothing in the text corre- sponding to he italicized words in either case. - This is 'Guido da Monte Feutro,' as the Chronicler calls him just 360 INFERNO. a di 13 di Marso [This date is repeated a little later, ' Lo quale venne a Pisa a dl 13 del mese di Marso MCCLXXXIX.']... Quando lo dicto Messere lo Conte Guido giunse in Pisa lo Conte Ugolino, &c. [here follow the four names 1 , Gaddo, Uguccione, Brigata and Anselmuccio, each of whom Dante mentions once in lines 50, 68, 89], ch' erano in pregione in della Torre de' Gualandi da sette vie erano in distretta di mangiare e di here per la posta della moneta di libre V mila ch' era loro imposta, che ne aveano pagate tre altre imposte. E fu dicto al Conte Ugolino da Neze a Marti che se non pagasse, u pagasse, era dicto che dovesseno morire. E quando lo Conte Guido giunse in Pisa, gia erano morti lo Conte Gaddo, e Uguccione di fame ; e li autri tre morinno quella medesima septimana anco per distretta di fame perche non pagonno...e dissesi e credeasi che se '1 Conte Guido fusse giunto in Pisa inansi che fusseno cominciati a morire, u che fusseno cosi venuti meno, che non are lassato ne patito che fusseio morti per quello modo, che li are iscampati da morte." below, who is well known by Dante's references in Inf. xxvi, Cotw. iv. 28, &c. 1 The Chronicler distinguishes that the two former were hs sons and the two latter his grandsons (' nipoti ' having evidently this sense here as in Inf. xi. 105), and Giov. Villani also (ap. Muratori, Tom.xiii. p. 320), "due suoi figliuoli e tre nipoti, figliuoli del figliuolo." The l tr here seems due to the error of supposing that Heinrich, another gnndson, was captured. On this see the note of Philalethes on xxxiii. 90, wio also gives a tabular view of these relationships. He there makes both Brigata and Anselmuccio to be sons of Ugolino's eldest son Guelfo, whereas the chronicler h. /. makes Anselmuccio the offspring of anothos. (omp. Arist. Poet. ix. 3, 4. COLLATION OF SELECTED PASSAGES. 361 This interesting extract enables us to calculate almost to the day the duration of Ugolino's imprisonment and the date of his death. The imprisonment lasted from the 3rd or 4th week in July 1288, to about March 15, 1289. Observe too that another chronicler, Ptolemaeus Lucensis (quoted by Scar- tazzini) speaks also of a long period of imprisonment and extortion of money; and Giov. Villani (who died c. 1350) vii. 128 (quoted by Blanc and Barlow h. /.) also states that Ugolino was in prison from August 1288, to March 1289, but his whole account is less precise in its details (see also c. 12 1). Surely Dante was familiar with this fact, or received version, of contemporary history. If so what more natural than that he should have introduced this telling circumstance of the cruel length of the imprisonment and the weary months of uncertainty before the veil of the future was suddenly torn away (1. 27) ? (3) On the principle of what is called by Westcott and Hort " transcriptional probability," lume is I think more likely to have arisen from lune than vice versa, (i) It is a more general and colourless statement, while lune introduces a new and important, and perhaps it might be thought contro- vertible, fact. For instance, the slender argument might have occurred to an early copyist, which is urged seriously by Bar- low (Contributions, p. 167), "that a room of narrow dimensions with one small slit in the wall for the admission of light and air was not adapted to contain five persons in close confine- ment for seven months." This a priori argument from 'adapt- ation ' however will scarcely weigh against direct external evidence; nor have we reason to suppose that the Arch- bishop Ruggieri would have been careful as to the health or comfort of the victims whom he afterwards starved, (ii) The reading lume being consistent with any theory of the length of imprisonment, those who believed that it endured for months would not be tempted to tamper with lume (if original), whereas those who on a priori grounds like Barlow, or a posteriori grounds like Benvenuto (see below), believed its duration to have been only days, would be compelled to remove lune. In short lume if original would not have pro- voked opposition, lune might easily have done so. 362 INFERNO. The old Commentators 1 however, it should be said, are unanimous for lume (or lumi}, though some admit the exist- ence of lune in many Codici, Benv. notices the variant lune, but states "ista litera stare non potest, quia... comes non stetit inclusus in ista turre nisi paucis diebus." (He gives no authority for this assertion.) Anon. Fior. mentions lune as 'come in molti libri si truova scritta,' and also lume without definitely deciding between them, but explaining the latter by the words 'cioe datogli lume il sogno suo allo intelletto.' Lan. has lume with the singular paraphrase ' poich' el fu in- carcerato piu die, vidi altri, e altri lui, per picciola finestra ' ! Butt, Barg., Land, and Veil, explain lume to refer to the light of dawn. Dan. insists that lumi should be read and not lume, justifying its use for giorni by a reference to lumina as em- ployed by Virgil. This emendation of Dan. appears to be unsupported by MS. evidence, though the difference between lume and lumi would be extremely slight, and the forms liable to be confused [see on this, Prefatory Note, p. liii (f)]. The general result is that, while the old Commentators prefer lume, they do not deal with the historical question involved in the difference of reading, except Benvenuto, whose passing and unsupported dictum, writing as he did in 1379, is not to be set against the minute and specific statements of the more nearly contemporary chronicler. There is an interesting passage in Buti (himself a Pisan) in his note on line 78, describing the place of burial of the victims, and their dis- interment which he himself witnessed. XXXIII. 72. Tra 1 quinto dl e '1 sesto. This reading was found in about 225 MSS. quarto. . .quinto Eid 48. Bat. 4 1 5. quarto... sesto BFfl-y^^ 30.45.47.51.52.54.62. 65. 76.106. Oil. Tra 1 quinto e 1 sesto dl CK. Tra '1 quinto e 1 sesto giorno 1. Tra '1 quinto dl e '1 quarto 77. Tra 1 quinto dl al sesto e. 1 I find Gutdo Pisano (in his unedited commentary) alone recognizing ' lieve ' thus : " mostravit mini per suum foramen magis leve jam quum feci malum somnium." COLLATION OF SELECTED PASSAGES. 363 The reading of the overwhelming majority of MSS., Tra 7 qitinto dl e 'I sesto, has been altered in a few cases to quarto... quinto and in a few more to quarto .. .sesto the motive of the alteration in both cases apparently being that the fourth day, when Gaddo died, was the last mentioned (1. 67) and that it would seem more natural to start from that as the inferior limit in describing the successive deaths of the other three victims. The only Commentary in which I find any other than the ordinary reading is the Ottimo, which has in the notes quarto... sesto. Serravalle alone notices the difference of read- ing, thus : " inter quintum, alias quartum, diem et sextum " : and so a little below (on 1. 74) " et tribus diebus, alias duobus, eos vocavi postquam fuerunt mortui." XXXIII. 74. due dl x 1.2.3.4.6 to 17.19.20.21.22.23.24.26.28.30.32.33.34.36 to 49.5 1.52.53. 54.55. 58.59.6o*.6i to 89.91.92.93.99.101.103 to 112.114.115. Bat. 10.30.55.65. 114.129.132.136.142.440.442.444. [210] tre dl V*ZBLvirvc(gr(inarg.') 5.18.25.27.29.31.35.56.57.90.95.97.98.100*. 102.113. Bat. 102. [27] E due gli ch.f. due dili (sic) \.VMwcefghijkmpqrtu'wx 1.2.4.6.8. 1 2. 1 5 to 23.25 to 34.36.38,39.40.43.44.47.49.51.52.54.55.56.57.59.62 to 70.72.73.74. 76.77.78.79.8 1.82.84.87 to 93.98.99. IOO.IOI.IO2. IO3. 105. 1 06. IO7. 108.109^?) 111.113.115. Bat. 55.114.129.132.136.444. [159] Benv., Veil., Guid. Che forar (foran, foron) 1' osso ~D\A(marg.} CM$yt8\trtywa/Hv 3.7*. 9.io.i3.i4(furan) 35.42.46.48.58.60.61.71.75.83.85.86.94.95.104.1092.112. 114. Bat. 65.102.415.426. [45] Buti, Barg. foral' osso LQ 2 n 5.11. forol' osso 24.41.53. erano all' osso Mod 37. furon 1' ossa - faro(?) 1' osso m. foran all' OSSO 80. 1 10 (a sort of 'conflate' reading). Che 'nfino all' osso \. foran la fossa v. foro alo fosso (sic) g. furono all' osso 97. foro insino all' osso. Bat. 440. The reading forar (or foran} /' osso seems to have arisen from a misunderstanding of the archaic form foro 1 =furo or furono. Nannucci in his monograph (Intorno alle Voci, &c. p. 15) says, "da fo, che tutti gli Antichi scrissero per fu, ne venne na.tura.\menteforotio,foro." The form foro occurs seve- ral times in Dante, though generally (if not always, as Blanc 1 For similar cases, cf. i. 28, v. 38, xx. 69. COLLATION OF SELECTED PASSAGES. 365 Vocab. says) l , except h. /., at the end of a line, and conse- quently it is commonly said to be " in grazia della Rima." This is refuted by Nannucci, Intorno alle Voci, &c. p. 15, where he quotes instances from the prose of Giamboni. Furo and fur of course occur in Dante, e.g. Inf. iii. 39 and vii. 51. (In the former passage 6 reads Ne foro, and several MSS. Ne fuor, and per contra at the end of the line both fuoro and furo occur in several MSS.) Another indication that a difficulty was felt here is the substitution of the familiar form erano in a few MSS., just as in v. 38 it was substituted for Enno. The readings, however, forar, or foran, /' osso involve a more radical alteration both of the verb and its construction. Forar is doubtless intended for the archaic and syncopated perf. =foraron which is substituted for it in Bargigts para- phrase. Buti as usual coincides, but I do not find the read- ing in any other Commentary. This syncopated form is not uncommon in old writers, and an instance of it occurs in ornar in Purg. xxii. 108. The reading foran perhaps arose from forar being mistaken for the infin., which would give no sense, and was therefore changed to foran, but it does not suit the preceding perfect Riprese. XXXIV. 13. Altre sono...altre stanno. This reading was found in 145 MSS. sono...sono DGSI/yA&uq&itMaigrv 3.7.42.58.65.69.71.100.101.107. [27] stanno... stanno BFMRj57V0\vT<}>goqw 5i 1 (?). Per... con E, Con soi occhi e per tre 85. Coi soi occhi e per tre 10. The same remarks apply as on the last passage. XXXIV. 99. mal SUOlo. This was found in nearly 190 MSS. mal sole D s 95. mal suono T] 73. ma il suol (sic) E. di sole e di lume e 9.33.46. del sole e di lume j/w 104. mal suolo e di sole g. The lect. fals. 'sole', which is temptingly suggested by di lume, is so rarely found that the passage will not serve as a criterion for the text, though it may be useful for tracing some relationship between the few MSS. in which the obviously false reading occurs. We may compare the way in which il sole is introduced into the line Purg. ii. 1 3, in a few MSS., from a similar association of ideas with mattino. See on this, Proleg. p. xviii, and add to the illustrations there given, Par. xv. 72 'Che fece crescer 1'ali al voler mio'; where some MSS. read volar, influenced no doubt by the proximity of the word ali 1 . So perhaps also in Inf. ix. 70 (already discussed) the 1 Not only readings, but also sometimes, I venture to think, interpre- tations, in the case of obscure words, are traceable to the influence of the association of ideas. For example, it seems to me that the explanations given in Purg. xxix. 147 of the rare and very interesting word brolo such as that it means 'a garden' (so most Commentators), or that it is 'Lom- bardic' for verdura (as in the bolder venture of Buti and Land.), or ghirlanda (as Benv^) are all mere guesses of interpretation suggested by the association of context, just as the present and other passages COLLATION OF SELECTED PASSAGES. 367 fals. lect. 'fiori,' and the still worse 'foglie e fiori', may have been suggested by the association with rami. quoted are guesses of reading due to the same cause. There is, I think, no reasonable doubt (though this is not the place to follow out the discus- sion) that brolo means a thicket, and therefore that Dante is referring to the thick mass and luxuriance of the flowers, and not to their garden-like colours or garland- like use, as the Commentators above-mentioned seem to have assumed as the basis of their explanations of a word unknown to them. PURGATORIO. E poi vcdrai color che son contenti Nel fuoco, perchZ speran di venire, Qtiartifo che sia, alle beate genti. Inf. i. 118 1 20. I. 1 08. Prendere OUACFHIJSUT pcXvs bdln*^?) cr 6.10.282.33.3542.46. 68.7i.8s(?) 9 i 2 . 9 4.io8. [37] Pigliate />'Aa 27.31. Prendete. This reading was almost universally found with the above exceptions. It occurred in fact in fully 160 MSS. out of about 210 ex- amined. 4> has piii greve salita. The reading Prendete is that of a large majority of the MSS. I think however it has probably been an early sub- stitution for Prendere, since (i) it gives a tempting complete- ness to the line in which it occurs (exemplifying a tendency which we have already had occasion to notice) ; and (2) Pren- dere is really almost necessary to complete the construction of ' mostrera,' which has otherwise no object, while the im- perative clause is abrupt and disjointed. " Devono girare il monte secondo che lo gira il sole da levante a ponente" (Bianchi h. /.). Prendere and pigliare will be found again interchanged in the MSS. at /;//. vii. 17. Among the old Commentators Benv. alone distinctly recognises Prendere, while Butt, Land, and Dan. have Prendete. Among twenty of the best known Editions which I have examined I find Prendere in 12; Prendete in 7; and Pigliate in 3. Of the 'first four Editions,' three have Prendete, and one only (Mantova) has Prendere, COLLATION OF SELECTED PASSAGES. 369 II. 13- qual suol (sol, or sole) presso dal (or del) AC (preso) FGHKLOP QSTUV !! V 4 WXZGl.Tr. BFSV(te) rAnZ^ Y tri0Kfxvirp-ytT]0KHTrpcrTvxs a*begijklorstuvwjravg/0'/#;ir i .3.7. 1 2. 1 6. 1 7. 1 8. 19.22.23.24.26.27.28.29. 30l (?) 31.32.35.37.38.39.43.47.48.5 1. 54.55*. 56. 57. 63. 66.68.69.72.74.75.77.78.80.82.84.90.91.92.93.95.97.101.102.105.108.109.113. [123] parea per descritto 7*GHJ. parea per scritto TAh'. parea per rescritto Wm 14.52. parea per soscripto 45*. parea beato in suo descripto A. parea beato pur descritto EVKf+dhnfi) 42.58.71.107 (scritto). pareva beato perscripto f. pareva beato prescritto v. paria beda (sic) per iscritto 106. faria pur descritto RACDJKUV (beato faria) apci 9.1 1.20.21.302.33.40*46. 70. 104. [31] facea pur descritto DFo. faria per scritto 10. seria beato per iscritto 96. COLLATION OF SELECTED PASSAGES. 373 NOTE. 1 has Perche parea. Of the MSS. of the Purg. at Venice, only one is said by Gregoretti to have faria beato pur descritto. The other 16 (?) have parea beato per iscritto. There is a great variety of readings here and a great diversity of opinion (in some cases supported with much warmth) among Commentators and Editors. I cannot assume " tantas componere lites," but I think the variations of the MSS. may be worth recording. The large majority are in favour of parea... per iscritto, which is advocated by Barlow (Contributions, &c. p. 183). Scartazzini contends for faria... pur descritto. It seems to me so far in favour of descritto that iscritto is all but identical with scritto in the same set of rhymes. Without saying that this is quite impossible, I think it is at least a very doubtful combination. Castelvetro in his Op. Varr. reading 'parea beato per iscritto' comments thus: " Modo proverbiale. Parea che 1'Angelo fosse beato in guisa, che non potesse perdere la beatitudine, ne quella gli potesse essere ritolta, poiche glien' era stato fatto stromento di Dio." This means perhaps the same as Buti, " cioe, ben parea con- fermato in grazia com' elli era." Similarly Land. But Veil. and Dan., also reading iscritto, seem to take it in the sense that his blessedness was as it were ' written in his face.' Benv. paraphrases ' nedum visus,' which seems rather to imply pur descritto. None of the Commentators seem to recognise the reading 'faria' II. 93- tant' ora xzegviv* 9.14.20.21.302.33.42.46.70.77.96.107. [56] tanta terra C*EFGHKLPOSTUV 1 V 2 V 4 WXZ5 - 260I^S7 1 eK^virpo- TV<}>XS aeiklmrstuy adikqrtu^w^x 1.11.16.18.19.22.24.26.27.29.31.32.35.36.37. 38.39.40.43.44.47.54.55.56.57.63.66.68.71.72.74.75.76.84.90.93.97.102.105.106. 108.109.110.113. [103] come tant' ora 6 tolta >\HJ\oc. come ora tanta terra e tolta U&d. com' ora 6 tanta terra y. com' e cotanta terra j. 374 PURGATORIO. com' e t' anchora colta (?) p. com' e ta tolta (sic) h. com' e cotanto ora /. com' era cotant' ora 48. com' e"e hora tolta 94. M' a te diss' io com' e tant' ora f. M' a te come ancora tolta (sic) 4.69. M' a te com' e, diss' io, tant' ora 78. tanto terra volta 54. tantu retolta (sic) g. In w, the line ends com' e tan d (marg.) has al. terra. Presumably most of the MSS. reading tanta terra have omitted diss' io. Otherwise the line would not scan. Certainly several have done so, but I omitted to register this point. The alternative readings here ' com e tanta ora ' and ' coin era tanta terra' (with minor variations) are interesting as being recognised from a very early date, either implicitly, as by Lan, and Off., who here, as often, are not independent, (see Witte, D. F., Series I. Art. xx.) or explicitly, as by Anon. Fior. and Dan., while Benv., Buti, Land, and Veil, adopt ora, without referring to the other reading. The MS. evidence for both is considerable, the resulting sense practically much the same, and I record them without being able to give a decided preference to either. Castelvetro (Op. Varr. p. 160) reads terra, and conjectures tanto for tanta, annotating thus : " E torre terra ad alcuno s' e vietargli d' andare in porto, cosi come si dice prender terra per andare a riva, et a porto." A fresh turn in the whole interpretation has recently been attempted by Antonelli (Alcuni Studi Speciali, &c.) who has discussed the passage at great length. He adopts the reading tanta terra, but objects to the interpretations ordinarily given both of this and of tant' ora. He explains tanta terra to be not (as it is usually taken) Purgatory, but 'the land of the living,' so that Dante here merely expresses his surprise at Casella's death, which Antonelli supposes to have occurred since Dante entered Purgatory, and in the flower of his age. To support this he lays stress (i) on ora in 1. 100, and (2) on Se (as = Dato eke, or Posto che) in 1. 96 ; and explains that line on the supposition (a pure invention ?) that Casella had often before prayed that he might die ! COLLATION OF SELECTED PASSAGES. 375 II. 118. Noi andavam. This was found in 176 MSS. I has Noi sedevam (so Butt}. 73 has Noi eravamo immobili et attenti. The reading eravam though that of a very few MSS. seems to suit the context before and after far better than andavam. It will be enough to refer to Scartazzini's exhaustive note h.l. Sedevam, which has the air of a correction, is the reading of Buti, but I have only met with it in one MS. The change from eravam to andavam (if so it was) must have been a very early one, since the latter reading is now nearly universal. It is difficult to trace any motive for the change, as eravam presents no difficulty. It might perhaps be thought to be a reminiscence of Purg. \. 1 1 8, or xv. 139 ; but per contra there are at least four or five instances of lines beginning ' Noi eravam' It was probably a short-sighted change which presented itself as a slight improvement, suitable enough to the line but not equally so to the context. The case, however, is, I think, scarcely clear enough to enable us to speak with absolute confidence. Several of the old Commentators are indeter- minate here. Benv. and Dan. have eravamo, and so apparently Lan., since he paraphrases by 'stavano negligenti.' Land. alone has (apparently) andavam, but his comment seems rather to point to the other reading : " Noi andavam tutti fissi : quasi dicat, eravamo non solamente fermi, ma fermi per restare." Buti (as above noticed) has sedevam. IV. 22. la calla B*EFGHKLOS 1 (?) TUV^^WXZGl.Tr.ASEFMQSVTU ^^yK\^p:vwx- 1.3.16.19. 28.32.35.38.39.46.48.49.52.53.54.55. s8.S9 1 .67.68.72.73.76 1 .78.8o.82. 91. 96. 98. loo. 101. 104*. 106. 108. 109. 1 10. [73] lo calle D I ODHIJ 7*gjknps/*r 7.10.13.20.21.33.41.592.69.71.77.79.852. 93.94.95. [33] 3/6 PURGATORIO. le calle R. la scala K&i\ 42. la callaia v (with calla in marg.} 105. la calcaia e (? in error for callaia). e has non era colla (with blank space and erasure before colla). 45 has Che non era cola onde saline. Here at any rate there can be no doubt that the primitive reading was calla, the unusual form of the word, which is generally written callaia^, being misunderstood, and con- jectural emendations introduced accordingly. The word calla however occurs again in Purg. ix. 123, but being then at the end of a line, it could not well be tampered with. In Purg. xxv. 7 we have the ordinary form callaia. The first impulse was evidently to substitute the familiar word calle, and so the reading la calle is found in a large number of MSS. But as calle is a masc. noun, a further change was almost inevitable, so that we have the tertiary reading lo calle in several MSS. Two or three have taken the bolder step of substituting la scala. IV. 72. Che mal non ABDGHKQRSTUV 2 WXGl.Tr.^CZ^G^/y6TAS ioirpTx/'a 1 (?) bcdefgijlmnprsvw^o7,/;/z^r/'Z/.r 1.3.4.6.7.1 1.12.13.2-14. 15.16.17.19.21.24.26.27.28.32.33.35.36.37.38.40.43.44.45.47.48.49.51.53.54-59- 63.66.67.68.69.70.72.74.76.78.80.81.84.88.89.91.92.96.97.99.100. 101 . 106. 107. 109.110.111.112.114.115. Bat. 132.136. [140] Che mai non OPV 4 FQVMAH&$ti\ty(a. La qual non V l Z(/\* 14.21. 23.24.29.31.43.45.47.55.56.57.66.73.78.84.90.102.105.107.113. [53] qui ritto BIKR(qui recto)^4 1 (retto)6'J /r rnap0iXvo(n|/ws/acfhjknosv w(ripto)xz*/;/m> 1.4.9.10.18.20.33.38.42.46.68.69.71.75.82.96.104.106. [56] qui dritto < 48. In 77 the two lines are almost rewritten thus : Dicendo oramai, Me di' perche assiso Se' tu quinci? Attend! qui escorta? This like the last two passages is a case of correction introduced owing to the occurrence of an unfamiliar word, viz. quiritta, which is nothing more than an emphasized form of qui. It is found again (in the form of either quiritta or quiritto) in Purg. xvii. 86, where it clearly corresponds to qui repeated immediately afterwards. Numerous instances are given of its use in the Vocab. Tram, s.v., where it is stated to be now-a-days merely colloquial in its use. So also 37$ PURGATORIO. Fanfani, ap. Scartazzini h.l. Being misunderstood here, it was first written as two words qui ritta. Then (as in iv. 22) the unsuitableness of the gender led to a further change qui ritto, or the more boldly conjectural qui dritto. In some MSS. however it is difficult to distinguish readings differing only in the division of words, as has been noticed before. (See Prefatory Note, p. li.) The only Commentators noticing the passage are Buti, whose language seems to imply the reading qui ritto, and Benv., who (reading, it should be ob- served, affisso in line 124) paraphrases, "quare es tu firmatus hie ? " He no doubt, therefore, read quiritta. V. 38. prima notte ABCDEFGHKLOPQRSTUV 2 V 4 WXGl.Tr.^CZ^/Z mnqrtww 3.4.6.7.9 to 17.19.20.21.22.24.26.28.30 to 49.52.53.54.55.58.59.64. 66.67.69 to 81.84.85.87.88.89.92.94.97.101.103.104.106 to 112.114.115. Bat. 10.136.444. [180] JBenv., Buti, Veil., Dan. mezza notte IV'Z.SOIlYvv^xsdhkoxjr 1.18.23.27.29.51.56.57.63.68.82. 90.91.93.102.105. [34] j and 96 have pura notte. i) has da prima vera. Bat. 1 29... di prima vera. The difference of reading here is interesting rather as affording a clue to relationship than as a test of excellence of text. It is clear that the interchange of prima and mezza could not have arisen (as is often the case) accidentally, though the precise motive for the substitution of one word for the other (in either direction) is not easy to determine. I am inclined to prefer prima, which is, as will be seen, that of the large majority of MSS. It seems to me that mezza notte may have been introduced to enhance the effect of the meteors by darkness, and is therefore suspicious. There is no doubt some force in Blanc's argument (Versuch &c., p. 13) that an antithesis is intended between lines 38 and 39, and that two distinct phenomena are referred to, viz. falling stars and summer lightning, both of which however were in the physical theories familiar to Dante (from Aristotle, Brunette COLLATION OF SELECTED PASSAGES. 379 Latini, &c.) attributed to the same cause, viz. ' vapori accesi'. We should thus have the double antithesis between the clear sky of midnight athwart which the meteors rush (cf. Par. xv. 13), and the cloudy banks of sunset about which the summer lightnings play. Still this is not conclusive for mezza notte, since it cannot be said that there is no antithesis between prima notte and sol calando. The former indicates a later period, when darkness has commenced and the clouds of sunset have disappeared. It is like the distinction in Purg. xxvii. 67 72 between the setting of the sun (1. 68) and the first uniform darkness of the nocturnal sky (11. 70 2). The greater completeness and obviousness of the antithesis in mezza notte might in itself be a ground for suspicion, just as it is made by Blanc a ground for preference of this reading. I must further confess to some suspicion of a reading whose chief support is found in members of the Vatican family without much independent evidence. Moreover all the early Commentators whose reading can be determined (viz. Benv., Buti, Veil., Dan.} read prima notte. V. 88. son Buonconte apYT]9iKX[Avoirpfv^ 8. 1 3.24.37*38*. 39.63. 76.98. 105.108. [25] Disposando CHIOTUV^vnVX^CZ^S vrAUWfattv^iV-vop ' rvxsf abcdlprux^zY 1.4.6.7.9. 10. 11.12.1 5. 16. 17.1 8. 19.2 1.26.27.28.29.30.3 1.32. 33-34-35443444647-5 I -55-56.57-58.64.66.67.68.7o.7i.74.78.8i.84.8S.89.9o. 92.93.94.97. 102. 103. 104. 107. 1 1 1. 1 13. 1 14. [112] Disponsado E. Dispensando t/ 1 12. E disposata Yd 77. 1 In all cases no distinction has been made between the forms dispo- sata and disponsata, &c. COLLATION OF SELECTED PASSAGES. 381 According to Gregoretti the Venetian Codici are about equally divided between Disposato and Disposando. The three principal readings here, Disposato, Disposata and Disposando have all considerable MS. support and espe- cially the two latter. Though Scartazzini almost anathema- tizes any who presume to doubt that Disposata is the true reading, I think the case is by no means a clear one, and I prefer to record the variants without pronouncing a definite opinion. Of course the point is that Disposata lays stress on the fact of La Pia's previous marriage, and pointedly fixes the guilt of her death (as must of course have been the case) on her second husband, though the mention of this may not have been superfluous if the fact of the double marriage was not generally known. It should be noted that not one of the early Commentators makes any allusion, either in the inter- pretation of or comments upon the passage, to the second marriage. They all therefore presumably, and some explicitly, read Disposando. It was probably the reading of Ott., Lan. and Anon. Fior., and it occurs definitely in Benv., Land., Veil. and Serr. Benvenuto's comment is ' qui primo posuerat mihi annulum sicut fit tempore matrimonii/ There is therefore no trace in the old Commentators of the reading Disposata. The readings may have been more easily confused than per- haps seems likely at first sight. Disposado (with the super- posed mark somewhat irregularly placed, as it often is,) might represent either Disposando or Disponsado, and the latter (see ' E ') is a Venetian dialectic form of Disponsato (see on this Blanc's Grammatik, &c. p. 654). We have a precisely similar interchange of Poggiando and Poggiato in Purg. xxvii. 81. VII. 15. ove il minor was found in more than 200 MSS. ove il nutrir in 6 only, viz. E*DA 2 Cbf. (Also in the marg. of D. See Witte Prol. p. Ixxxiv.) Ik 've 1' umero s' appiglia (!) Bat. 55. This is recorded merely on account of the interesting fact that Jacopo della Lana, writing as early as 1328, notes the 382 PURGATORIO. existence of the two readings (see Witte Prol. p. Ixvi). I observe that Scarabelli in his edition of that Commentary omits the passage in question as being in his opinion an interpolation, but he gives no evidence of this. The reading nutrir however is exceedingly rare, whatever its date. In fact I met with it in the text of only six MSS. (two of them bearing trace of alteration) out of more than 200 examined on this passage. The idea expressed by that reading may be com- pared with that of Inf. xxv. 85. There is no trace of it in any of the other old Commentators, except perhaps in the curiously confused note of Buti, who having first explained il minor as T omo di minore condizione,' gives the following as a possible alternative interpretation : ' ove il minor, cioe lo fanciullo, s' appillia, quando e nel ventre della madre ; cioe al bellico, onde 1' omo. . .riceve lo nutrimento,' &c. This certainly seems to imply some vague knowledge of the other reading, the explanation intended by some one for that reading having apparently been forced into connexion with this, which Buti, as it will be seen, distinctly adopts. VII. 51. o non sarria che non GHOTU'WXSFJVQA 2 'S.lj ] 6K V .-^rbc l e(e non) rs/*/(?)/ i. 6.7^?) 15.16.18.19.22.23.24.26.27.28.30.31.32.35.37.43.44.55.59.67. 70.74.75.79.81.88.89.105.106.114. [63] (24.43 and 44 have chi non. /has satria ?) o non saria (or seria) che non EKLSZGl.rAjIIavpxsac*fgikl movzbcftmnq 3.11.12.13.14.17.29.36.39.41.47.48.49.54.56.57.63.66.82.84.90. 97.101.103.109.110.111.113. [62] Buti. u non saria che non C u. o saria che non Afl-ynwxw 4.8.52 (sere) 99. (w has a space left before saria). ovver saria che non Q*Ry4CZ}//y0djp 21.100. ower saria che '1 non B 68.85. ove saria che non D. o pur saria che (or che 1) non \KUM^\.^xfs\\r -j*. -2.0.^. 963.98*. 102*. [22] Benv. (seria pur), Dan. (?). o che saria che non g. o non saria che 1' uom F. o non saria che uom d. e non sarei che 1' uom 69. e non saria che non V 1 rI>v|/.t- 9.72.76.104.1 12. Bat. 55.' COLLATION OF SELECTED PASSAGES. 383 o non sauria che non 58. io non saria che non P. e non saria ch' io non Tr. ^ (chi non) e (do.). o non saria ch' io non A. o saria pur che 1 non F. o saria dunque perche (or perch' ei) non Sv io. o dunque saria che non 64*. ogn' uom sauria che non t. o non saria che potesse a. o non saria come potesse 93. o non sarebbe perche non 73. o seria che non se potesse 77. oel (sic) seria che non 107. o non saria chi le impedisi(!) 108. 38i had e (?) non sarria che non, but 382 has o vero che non sarria che non. The extraordinary number of variations found in this line are all due to the misunderstanding of the unusual form sarria for salirebbe. Mastrofini gives examples of sarrb for the future and sarrei for the conditional ; the third person of the latter appearing both as sarrebbe and sarria. These occur even in prose, e.g. in Boccaccio and Passavanti, and we have examples quoted from Guido Cavalcanti. See also Nannucci, Verbi &c., p. 794. This word sarria was obvi- ously mistaken by copyists for saria from essere. This misconception evidently underlies most of the numerous varieties of emendation. It does not follow that all those who merely wrote ' o non saria die non ' were in error, as the neglect of double letters both in the way of insertion and omission is so very common that little stress can be laid on it. But as soon as the negative non is felt to be a stumbling- block (as most of the variants imply) the misunderstanding clearly betrays itself. Among the Commentators, Benv. reads the whole line thus : O seria pur perche non potessc, and explains, " an hoc solum proveniret ab impotentia sua." Dan. quotes and comments thus, " o pur saria che non potesse, cioe salirebbe per non potere," which is clearly an expla- nation of sarria, but without ' non' Buti reads, o non saria che non potesse, and expounds curiously : " cioe o sarebbe ch' elli potesse : ma non volesse, s' intende : impero che due negazioni importano affermazione." The other old Com- 384 PURGATORIO. mentators pass over the passage. It does not seem therefore as if any of them had explained the line rightly. VIII. 64. 1' altro ad un . ad)hlnpuvwxyz a(a, 24.28.292. 30. 37. 39.42.44.46.48.52.58.67.68.71. 75. 77. 82. 88. 91. 99. 100^)104. 113. [92] Benv., Butt, Land,, Dan. 1' altro a me BC*EGHKSTUV 2 WXZTr.^2n*^t T ie K virpXsabdejknopstu wybcgknqr'wwx 1 .4. io. 1 1 . 1 3. 1 6. 1 7. 1 8. 19.2 1.22.23.24.27.30. 31.35.36.38.42.47.48.52.54.56.57.58.67.70.71.72.74.77.85.90.93.96.100.102.105. 106.113. [98] Benv., Veil., Dan. Como parlava 9. Come diceva 3. Com' io guardava 62. VIII. 94 (bis). e Sordello was found in about 175 MSS. Sordello \om. 'e'] only in io viz. ACI&foad 13.119. j has il Sordello omitting ' e '. Com' io parlava is obviously false in fact (though found in nearly one half of the MSS. examined, and in all of Witte's four), since Virgil, and not Dante, was speaking. Con- sequently the true reading is undoubtedly, ' Com* ei (or el) parlava.' Benv., Veil, and Dan, observing the above difficulty, explain the reading io parlava by saying that Dante was replying to Virgil, but there is no evidence of this, nor would parlava have naturally been thus used for rispondeva. Buti explains rightly : "Coni z,cioe V\rg\\\o, parlava a me, Dante, le parole ditte di sopra." The omission of ' e ' before ' Sordello ' in a few MSS. is due to its idiomatic use having been overlooked. This is the same as that which has been already explained and illustrated under Inf. xix. 3. IX. 17. Piu... men in about 200 MSS. Men...piu in 4 only, viz. KU\ 20. This is noted merely as being one of those passages where variants are recognized by or traceable in very early writers. In this case, Pietro Dante (c. 1340) reads 'Men dalla carne e pin dal pensier presa' (see Witte, Prol. p. Ixvi.). Not- withstanding, I have found this reading in only four cases out of about 200 MSS. examined. There seems little doubt that Pietro is wrong and the lect. vnlg. right. Scartazzini's ex- D. I. 25 386 PURGATORIO. planation makes all clear and consistent. Further a passage from Cic. de Sen. 80 (with which work, as we know from many passages in the Convito, Dante was familiar), strongly supports ' Piu dalla carne': "Atqui dormientium animi maxime declarant divinitatem suam : multa enim, quum remissi et liberi sunt, futura prospiciunt Ex quo intelligitur quales futuri sint, quum se plane corporis vinculis relaxaverint 1 ." At present it only approximates to this state, when it is " pellegrina piu dalla carne." Compare also the following passage in the de Sen. a few lines earlier : " nee vero turn animum esse insipientem quum ex insipienti corpore evasisset, sed quum omni admixtione corporis liberatus purus et integer esse ccepisset, turn esse sapientem." Lan., Benv., Veil., Buti and Dan. all explain the conditions described in this line by stating that the stomach is empty at that hour and so the brain clearer ! " quando lo stomaco ha fatta la sua digestione, lo cerebro e alleviate dalle fumositadi dello stomacho" (Lan.) ; " il celebro non e occupato da la sua evaporazione " (Buti). X. 14. scemo was found in 174 MSS. lo stremo V^ZZ-'v^xb? 1.3.18.31.56.57.90.105.113. [17] 10 scender Cu. 11 sommo A. lo stretto k. The obviously inferior and more common-place stremo is certainly to be rejected. It is noticed here only because it seems to be almost entirely confined to MSS. exhibiting elsewhere relationship with the Vatican MS. ' B ', and it may therefore be regarded as a good test-passage for that family. X. 30. dritta di salita BOQR.Tr.FNOQa.fr\9(\vov^tafcgxal>^Amn.r 9.24.30. 33.37*45.46.48.58.69.73.79.91.99.102.103.104.108.112.113. [52] 1 This is aptly quoted by Dionisi, Anedd. II. p. 59. COLLATION OF SELECTED PASSAGES. 387 dritto di salita ACDEFGHKLPSTUWXGl.^(?)CZ>.O/S/FrA II2*^npYt K l A ' ir P' r< J ) Xs a beijklmnoprstuvwyz^i(?)^7^r/z/ 1.3.4.6.7.8.10.13.14. 15.16.17.18.19.21.22.23.26.27.28.31.32.35.36.38.39.40.41.42.43.44.47.51.52.54. 55.56.57.59.63.64.66.70.7 1. 72.74.75. 76.77.78.80.81. 82.84.85*.88. 89. 90. 92. 93. 95.97.98. loo*. 101.105. 106.107. 109. 1 1 1. 1 19. [142] Che di dritta salita h 20.1 10.1 14. Che di salita dritta w. Che di salita ritta I. Che diritta salita A. Che dritta a salita R. Che dritta di salire o-. Che dritto di salire u. Che derta (?) di salita d. Che deta(?) e di salita Z 96 (?d' erta). B has drieto, which is probably an error for dricto, from which it is often scarcely to be distinguished in many MSS. f 11.68 have dietro di, which is probably an intentional improvement from the reading in some exemplar, dietro, in error for dricto. This is a passage in which the somewhat unusual con- struction, or order of the words, has given occasion to a variety of alterations, so that a comparatively small number of the MSS. have preserved what I cannot but believe to have been the original text, viz.: Che dritta di salita aveva manco. Among the early Commentators, Anon. Fior. and Dan. read, or explain as if they had read, di dritta salita : Bnti quotes 'di salita dritta' but his explanation is ambiguous: Land, seems to be expounding the reading which I am defending. Benv. and Veil, clearly have dritto, the former paraphrasing, 'quasi dicat rectitudinem ascensus non habebat ;' and the latter, ' mancava di dritto e di ragione di salita.' I believe the explanation of Scart. to be correct, viz., that we should make a pause, or even insert a comma, after dritta, and take it as being equivalent to ' Che, a cagione dell' esser diritta, aveva mancanza di salita,' i.e. 'which, being perpen- dicular, could not be climbed.' This construction of dritta (as Scart. notes) is very similar to that in Ptirg. xxi. 89, ' Che, Tolosano, a se mi trasse Roma.' To this he might have added (if his own reading be adopted), Pnrg. v. 135, ' che, inannellata pria, Disposata m' avea,' &c. ; and Inf. xvii. 25 2 388 PURGATORIO. 24, ' Sull' orlo che, di pietra, il sabbion serra.' The following passages seem also to illustrate the construction contended for: Purg. xvi. 112 ; xix. 132 (if we read dritto, and explain as in the note on that passage later on); and xxiv. 122. This use of manco as a substantive for mancanza is found again at Par. iii. 30, 'per manco di voto'; and several in- stances of this use in prose writers are given in Vocab. Tram. s. v. If dritto be read, it is necessary to give it the unusual sense of ' possibility.' Longfellow strangely says that it is used in the sense of ' right of way.' Moreover the reading dritta, though, as we have said, that of a minority of the Codici, is not so unsupported as is implied by Dr Witte's statement that it is "congettura...sprovvista dell' autorita di antichi codici." It is a most unlikely result of conjecture, since it would tend to introduce, and not to remove, a prima facie difficulty. And as to the number of supporting MSS., it is more than can be found for some other undoubtedly good readings. XIII. 2. risega CDlORGL<4I>^f0(7rAA*Xpn 1 Xe^(?)ta 1 (?) bgjkln,(?) cefghimnrx 1.8.9.20.30.33.42.46.48.52.53.63.67.71.73.77.78.79.82.91.93.98. 104.107.108.110. Bat. 10. [74] Lan., Butt, Land., Veil., Dan. rilega AFGHKlfQST\JV l W 4 VfXZTr.CEFNQSyn&iMkft*l irpo-Tv<|>xK?) Fa 2 dehimn*orstuw^ 1 v|/wcegjlwrt 9.24.28.33.35.46.48.52. 63 (la vostra) 69.102.109. [36] Butt, Land., Veil., Dan., Serr. nostra region Lu. 14 has SI m' a nostra la mente strecta (sic). The correct reading nostra ragion, which is found in the majority of MSS., would probably not have been tampered with but for the somewhat unusual sense of ragion = ' dis- course:' i.e. 'our discourse inclines me rather to weep than to converse.' Comp. ragione again in this sense in Purg. xxii. 130, and indeed in three or four other passages of the Coin- media (see Blanc's Dizionario s.v.). The verb ragionar in the sense of discourse is very common, e.g. Purg. xiii. 132, xiv. 8, xvi. 1 20, xxii. 21, xxiv. 2 ; Par. iii. 35, &c. &c. Among the Commentators, Benv. alone appears to read nostra ragion, which he strangely explains zsjustitia, since it is more just to mourn for one's own sins than to deplore those of the living. Buti, Land., Veil, and Dan. all read vostra ragion, and ex- plain it as ragion umana, which inculcates charity and com- passion ; Buti adding that this is the reverse of the feeling of Envy, which Guido del Duca is now expiating. Lan. and Ott., without specifying the reading adopted, also say that 1 This is a sample of the puzzles and disappointments which generally (so far as I have seen) baffle any attempt to generalize, or to establish definite results as to the value and relationships of MSS., from the materials here collected. COLLATION OF SELECTED PASSAGES. 391 ' pianger' relates to pity for the sad state of the people of Romagna, adding (like Butt) that such a feeling is the direct opposite of Envy. As to nostra region (the reading of Ed. Nidob. and a very few MSS.), 1 cannot suppose that Dante would use such an expression as nostra region ha stretta la mente, meaning by it, ' the condition of our country has con- strained my mind.' It appears here, as sometimes elsewhere, that the majority of Commentators support a reading which is that of a small minority of the MSS. It should be re- membered however in this case that the difference between nostra and vostra is very slight. (See on this Prefatory Note, p. liv, and on the difference between the evidence of Commen- tators and that of the MSS. see note on Inf. xxvii. 21.) XVI. 145. CJosl tornb was found in about 155 MSS. Oosl par!6 ABDV l ZNOA*afaikotwdgr 1.3.18.24.27.29.31.47.552.56. 57.67.76.85.90.105. [36] Cosl parlommi 37. Oosl sparl O. Bat. 428. Oosl sparid 1. Cosln' and6 73.119. Cosl parlando e poi 93. This passage is recorded because the reading parlb for tornb is characteristic of the MSS. of the ' Vatican ' family, which has been several times referred to. The reasons urged by Scartazzini seem conclusive for tornb, which is the reading found in a very large majority of MSS. Sparl and n andb are conjectural emendations occurring very rarely. Parlb pro- bably had a similar origin, but happens to have commanded more support, or to have arisen earlier. Among the Com- mentators, Off. (probably), Buti, Veil, and Dan. have tornb. Benv. gives both readings without comment. XVIII. 83. Pietola ?iu che villa AEFmOQRUZGLTr.AC>EFNOUT'Z4>y 2 2 7*.9.io(?) I7 2 .i8 2 .2o.2i.22*. 29.33. 392 PURGATORIO. 35.382.42.45*46.482.51.52.552.56.57.58.59.69.71.73.77.78.85.87.90.91.94.98.99. 100.102.103.104.109.110.113.119. Bat. 10. [101] Pietola pifo che nulla B^Fixih 3o.3i 1 (?) 322.41.72*74.76.108. [14] Pet., Benv., Anon. Fior., Buti, Land., Veil., Dan., Serr. Pietosa piu che nulla C D A Allflatirsekorz^?) enqvw^x i .4. 1 1 . 1 3 X . 1 6. 1 7j. 82.84.93.95.96.97.101.106.107 (ulla) 112.114. [58] Pietosa pifc che villa PS 3.53. Piatosa fu piu che nulla v. Picciola piu che villa ~Lvakt 44. Picciola piti che nulla 1 17. Cortese piu che nulla GKTWX^^K^pTim/w'/w* 6.12.14.15.23.26.27. 34.43.63.66.68.75.79.81.88.89.92. [36] Cortese pifo che villa h*. Cortesia piu che nulla 6. Pietola piu che villa matutina g. Pietro Ik piu che nulla 105 (? oral blunder). a x seems to have had nulla. We have here a curious case of corruption arising from an unfamiliar word, viz. Pietola. This is a village near Mantua, popularly supposed to represent ' Andes,' the birthplace of Virgil. This word was mistaken for, or altered into, Pieftsa (the long s and / being very similar), with the accompanying change (in most cases) of villa into nulla, and sometimes also of die pur tor per cut in the preceding line. The most currous point to notice however is the considerable amount of sup- port accorded to the further audacious substitution of Ctrtese for Pietosa. There is also a great variety of other con- jectural or blundering alterations, as will be found in the collations above given. It will also be noted that a very large number of the MSS. examined here have been tampered with in the way of erasure or alteration of what was originally written. It is to be observed that the old Commentators without exception (omitting Ott. and Lan. who are too vague to cite) adopt and explain Pietola, without any allusion to any other reading. They only differ as to whether villa Maniovana is to be explained of 'the city of Mantua itself as Be'iv., or of (any Mantuan town' as Pet., Anon. Fior., Buti, Lend., Serr. ' quam aliquod aliud rus Mantuanum '). COLLATION OF SELECTED PASSAGES. 393 XIX. 34 . lo mossi gli occhi e 1 buon Virgilio al men tre ACDdpr 21.35*. al buon Maestro al men tre Tr.i*(?) a r e 1 buon M. e mentre *|/ io2 1 (?). el buon M. mentre peo 46. e 1 buon M. al men tre QRGlTAc (messi) gvxj 9- 3o. 33.52.99i- 102*108. [18] e 1 mio M. al men tre a 20. e '1 mio M. mentre X (with ' al ' inda manu). lo volsi gli occhi e 1 buon Virgilio al men tre V 1 7.B 29.56.57.90!.! 13. al buon Virgilio e mentre AB*P\j kowg i . 1 1 . 1 6. 1 8. 27.31.37.45.59.70.76.85.93.96.105 (el mentre). [24] e 1 buon Maestro al men tre Ffb*fsy^- 14.42.47. 48.552-78. al buon M. al men tre 51.109. al buon M. e mentre CEGHKOSTV^WXF// /6'^7FAn2^'t J Txsabeh*ikGUwz^G7cdghlmp. [18] dritta ABCDGQZtf/f/.Stf'Fabefiknrstuv. [25] dentro j. The evidence here collected is slight, as I noticed the passage rather late. This however is certainly a case in which we can feel, as Scart. says, l persuasissimi ' as to the true reading, viz., dritto. It is quite obvious that the ' facilior lectio' dritta has arisen from the shallow notion that the word must be feminine to agree with coscienza. (Compare a similar case at xxi. 45.) This implies a complete misunder- standing of the passage. Dante had prostrated himself (1. 127) as soon as he discovered that he was in the presence of a Pope (Adrian V), and when the Pope had immediately protested against this (the whole scene being in obvious imitation of Rev. xix. 10, and xxii. 8, 9), Dante replied that his conscience reproached him for standing upright (dritto) in the Pope's presence. If we read dritta, or, reading dritto, take it adverbially ( = rettamente, or gitistamente), there is nothing clearly to shew, as Scart. observes, what his con- COLLATION OF SELECTED PASSAGES. 395 science reproached him for. Among the old Commentators, Scartazzini quotes Benv., Bnti, Land., Veil, and Dan- as reading dritta. This seems to be true only of the two first named ; and it looks as if Scartazzini had merely observed the word in the text that happened to be printed with the Commentaries, which as we have often seen proves nothing whatever. Land, and Veil, say nothing to indicate their reading ; while Dan. commenting, " Pareva al Poeta discor- tesia lo starsi egli dritto in piedi" must clearly have read dritto. In printed editions, according to Scartazzini, the reading dritta is almost universal. He states that he has found it in more than a hundred editions. XX. 90. tra vivi ladroni was found ia more than 200 MSS. nuovi in 12 only, viz. V'ZxI/wghjr 20.29.98.1 13. [12] Some MSS. (e.g. RUTQFf 1.73.108.112 and probably others) have tra i vivi. b has entra vivi (vivi over erasure). 39 has Entra gli due ladroni. 52 has Entra G-iudei ladroni. s* has travi vi with slight erasure, as if nuovi had been original. According to Gregoretti, 2 of the Venetian Codici have nuovi (Bat. 288 and 297), the rest (I suppose about 16) have vivi. Besides the overwhelming MS. evidence for vivi, it may be added that nuovi is (i) suspiciously obvious, being suggested by nuovo in 1. 91, and rinnovellar in 1. 89; (2) somewhat tame, adding no new touch to the picture; (3) only super- ficially appropriate. For in the case of the ' ladroni ' here intended, viz. Nogaret and Sciarra Colonna, the parallel with ' the thieves ' so far fails, that though branded by Dante as 'ladroni' and though assisting at this 'crucifying afresh' of Christ in the person of His Vicar, yet they were not them- selves sufferers, they were not put to death or injured in any way like the thieves to whom they are compared. This distinction is pointedly brought out by l vivi,' and so this epithet gives a fresh character to the scene. Thus while nuovi would have no definite interpretation in reference to the case of Boniface, but would merely imply that Dante 396 PURGATORIO. was working on at the picture by enumerating another detail of the original scene (not at all his habit) ; mm on the other hand would represent to us Boniface as it were crucified between Nogaret and Sciarra Colonna, who were standing on either side of him mocking and insulting him (Comp. Matt, xxvii. 44), so far like the ( ladroni' yet still '-m'vi.' Here then we have an epithet quite precise yet pregnant in mean- ing and caustic withal, having ' sapor di forte agrume ' (Par. xvii. 117), which it need hardly be said is quite in Dante's manner. The fact too that its meaning is not superficially apparent would invite the substitution of a more obvious epithet 1 . The old Commentators pass over the passage, except Ott. and Benv., who read mm', and Buti, who explains both readings without expressing a decided preference for either. Serr. also has 'mm',' but without any comment. The ex- planation of Ott. is certainly curious. He takes anciso with rigid literalness to apply to the actual death of Boniface, about a month after the outrage at Anagni, in Rome sur- rounded by the Cardinals, who are here described as ' mm ladroni? 'perche rubano tutto il mondo !' Though it does not bear on this reading, I cannot help mentioning the interesting points of resemblance between this whole passage and the prayer to the Virgin composed by Boniface VIII. This is given by Nannucci in his Manuele delta Letteratura, i. p. 421. The resemblances can scarcely be accidental, and it is most curious that Dante should have thus imitated a composition of his bitter enemy. It should be noted however that this occurs in the one passage in which he speaks of him with sympathy. Compare especially the 1 I have sometimes thought that the following remark of K. O. Miiller on the style of Sophocles, in his History of Greek Literature, was peculiarly applicable to Dante. " It is distinguished. ..less by the choice of words than by their use and connexion; and by a sort of boldness and subtilty in the employment of ordinary expressions. Sophocles seeks to make his words imply something which people in general would not expect in them." This last statement is curiously like Dante's own declaration to the author of the Ottimo Comento, " ch' elli molte e spesse volte facea li vocaboli dire nelle sue rime altro che quello, ch' erano appo gli altri dicitori usati di sprimere " (On Inf. x. 85). COLLATION OF SELECTED PASSAGES. 397 four times repeated Veggio in Dante, with Vedeva and Vede similarly recurring at the beginning of four lines in Boniface's poem. Also 1. 89 here, ' Veggio rinnovellar 1' aceto e il fele,' with 'Vede 1' aceto ch' era col fiel misto 1 ' (1. 10). It is also interesting to observe that the very com- parison of the sufferings of the Pope with those of our Lord may seem to have been suggested by the Pope himself, who awaited the arrival of his persecutors sitting with calm dignity on his throne, and refusing to fly, saying, " If I am betrayed like Christ, I am ready to die like Christ." (See Milman's Latin Christianity, vol. vii. p. 151.) XXI. 19. parte andavam (or -van) EFGHKOQSTV 1 WXGl.^C^ J : J F///J/A r 6> 5f/FT24 > fiap-yT]KX(nrp\j/wFabdfgjlnpsvxzrt'^rt'M;r/7/w 3.7.8.9. 10. 12. 14.15.20.21.22.23.26.29.30.33.34.35.39.40.41.42.43.44.46.48.52.54.56.57.63.66. 68.70.71.73.76.77.81.84.85.88.89.90.91.95.102.104.106.107.109.110.111. 113. 119. [126] parte andava BCDPRUV 2 V 4 ZG72AAneivov X sehmortuwy^- 4-6- 1 To facilitate the comparison, as Nannucci's work may not be readily accessible, a few lines of Boniface's poem are reprinted here. [Nannucci says that this composition was found in a MS. in the Vatican Library, where it is stated that in the I5th century it was to be seen in the Church of S. Paolo fnori le mura, with the following curious inscription : "Santo Bonifazio Papa ottavo fece la infra scritta orazione, e concesse a chi la dicera liberazione di morte subitanea."] Stava la Vergin sotto della cruce : Vedea patir Jesu, la vera luce, Madre del re di tutto I' universo. Vedeva il capo che stava inchinato, E tutto il corpo ch' era tormentato Per riscattar questo mondo perverso. Vede lo figlio, che la guarda e dice : O donna afflitta, amara ed infelice, Ecco il tuo figlio : e Joan le mostrava. Vede 1' aceto, ch' era col fiel misto, Dato a bevere al dolce Jesu Cristo, E un gran coltello il cor le trapassava. Vede lo figlio tutto passionate Dicer colla Scrittura : e consumato, &c, PURGATORIO. 1 1. 13. 16.17. 19.24.32.36.37.38.47.49.5 1.55.58.59.64.67.69.72.74.75.78.80.82.96. 97.103.108.112.114. [69] e perch6 andava I. e perch6 andavan Tr. e pur andava A. e poi andava RV\lq 18.27.31.93. a parte andava L. e mentre andava k i. e parve andava 105. We have here another instructive instance of the disturb- ing influence of a misunderstood idiom. The true reading is certainly parte andavam (andavan being probably a mere variation of orthography, as commonly elsewhere). This reading is perfectly intelligible and plain, if parte be taken in the rather rare sense of an adverb : but it being otherwise quite unintelligible, a curious crop of variants, manifesting the endeavour in one way or another to escape the difficulty, has grown up. This use of parte, which occurs again in Inf. xxix. 16, is no doubt a Florentine provincialism, as is clearly pointed out by Benv. in his comments on both passages, which are in almost identical language. Here he writes as follows : " idest, et interim ibamus velociter, nee tardabamus illis loquentibus ; ita quod hie parte non denotat portionem, nee est nomen, imo adverbium, et tantum valet quantum 'in isto medio,' et est vulgare florentinum 1 ." Torelli (quoted by Negroni in his monograph on L 1 Avverbio ' Parte ,' &c. p. 15) cites a precisely similar use of parte from the prose of Boccaccio (Decani. Giorn. 8, Nov. 7), "parte che lo scolare questo diceva, la misera donna piangeva continuo." Negroni further cites (p. 17) instances from Cellini and Petrarch. The early Commentators, with the exception of Benvenuto and Buti, pass over the passage sicco pede, so that we cannot even tell what reading they adopted. Buti reads PercJie andate, but says in his Commentary, " altro testo dice, e ';/ parte andavan forte, cioe in quel mezzo che elli dicea noi andavam fortemente...pero pilli lo lettore quale vuole." The admis- 1 Lorini, in his work on the variants of the Cortona MS. ('-"') (to which I am indebted for my information about that MS), says in his note on Inf. xxix. 16, " Parte, in senso di mentre ^ intanto, & voce tuttora viva nel nostro contado," COLLATION OF SELECTED PASSAGES. 399 sion by Dante of a provincialism of this kind will remind students of the curious and virulent attack of Machiavelli upon Dante, on this and other grounds, in his ' Dialogo sulla Lingua] &c., where he maintains of Dante that "il suo parlare e al tutto Fiorentino " (' Ciceronem Allobroga dixit' !) but that out of his hatred for Florence he endeavoured to shew in the De Vulg. Eloq. that it was not Florentine, but "lingua curiale," &c. The present idiom however is not mentioned in the indictment, which is made to rest chiefly on Par. vi. 94 (morse], Inf. xix. 120 (spingavd), ib. 45 (zanca), Par. v. 64 (ciancia}. He also objects to introcque which Dante uses at Inf. xx. 1 30 ; and it is curious that Dante himself gives this word as an instance of a vulgar Tuscan provincialism in De Vulg. Eloq. I. 13. In Purg. xxxi. 51, Fanfani has very infelicitously proposed to read parte for sparte, supposing it to mean ora or adesso. Negroni's monograph is mainly devoted to combating this conjecture, and to establishing that parte means intanto, mentre, frattauto, &c., but not, as Fanfani would have it, ora; and also to explaining how the idiom arose. XXI. 25. perche lei ClP(^A z CDEMh.A. 2 ^\vo^Vni l (?) 5T]iKjnrpxsab 2 efgh*ikmnors*vyzrfi?o"//^r/'ze'7/ 1 . 3 . 7 . 1 1*. 1 2. 1 5. 17. 1 8.2 1.22. 23.26.27.31.32.33.34.36.37.38.39.40.43.44.48.49.52.54.56.57.58.59.63.64.66.67. 68.69.70.72.74.76.77.78.80.81.82.84.85.88 to 93.95.96.97.103.106.108.109.111. 113.114.119. [133] perche colei ADLRU-SSvdjtw^wv 6.10.14.16.19.24.35. 45. 47.51. 55^101. 105.107.112. [30] Ben-v. perche colui a. perche Lachesi U. perche color h 79. perche che colei che (sic] 117. This passage is given by Witte Proleg. p. xli. as a typical instance of the genesis of ' secondary ' readings ; and he traces the successive steps in the departure of the text from the original as follows ; The primary reading was no doubt, 400 PURGATORIO. ' Ma perche lei che dl e notte fila.' ' Lei' for the nominative case seemed to some copyist irregular, and he substituted ' colei l \ The next step was to restore the rhythm by read- ing 'per colei,' and then further to remedy the impaired sense of the passage by altering avea into era in the next line. This was one line of emendation. Another was as follows: A marginal or interlinear gloss 2 ' Lachesi' was mistaken for a var. lect., and substituted for lei die dl. Then the sense of the passage being wholly destroyed, a 'secondary' alteration followed, ( eke da le fila ;' a reading occurring how- ever in U only out of over 200 MSS. examined by me on this passage. Dr Witte cites this MS., but does not mention that he has met with the reading in any other. I believe the reading Virgilio for in ver lui in Inf. xii. 16 had a similar origin (see notes on that passage). Perhaps too the singular reading which I observed in 'u' at Inf. ii. 28, 'Andovvi Paulo vaso d' elezione,' may have arisen thus ; and possibly also ' Israel con Isaac ' for ' Israel con lo padre ' in ' D ' at Inf. iv. 59. XXI. 45. d' altro \|/sFa*begijkloprstuvwxyrrtk^wY 1.4.6.9. 1 1.16. 17.18.19.21.24.27.29.30.31. 77.79.82.90.91.93.96.102.104.105.109.110.113.114.119. [122J >//., Benv., Dan. d' altra CEGIKSTWX/^TiiKiivoirpo-^^dfhmnztfw 7.10.14.15.20.23.35. 58.64.66.68.69.73.78.80.81.84.85.97.103.107.108.112.117. [55] Butt, Veil. The reading altra is highly suspicious, as being in two ways much easier and simpler than altro: for (i) it simplifies the construction of the sentence 8 ; and (2) it has the air of rectifying an apparent false concord or supposed blunder of gender 1 This would be in fact giving practical effect to the comment of Buti : " Ma perche lei; cioe, Ma perch colei." " Precisely such an interlineation occurs here in ' 102.' 3 The constructions would be, (i) with altro : " cagione puote esserci di quel che il ciel . ..., e non d' altro": (2) with altra: '[alterazione] puote esserci di quel che il ciel ...... , e non d' altra cagione' (as Buti puts it, ' non d' altra cagione che della sopra detta'). COLLATION OF SELECTED PASSAGES. 40 1 (see similar cases in Purg. xix. 132, and Par. ix. 129) (see notes on those passages). The fact too of the more difficult, and prima facie more unattractive, reading being supported by a considerable numerical majority of MSS. is not to be overlooked. At the same time, as far as the MSS. I have examined go, altra has far more support than would seem to be recognized by Scartazzini's language h. 1. (" per dis- grazia sono troppo pochi i codici che recano tal lezione"), and certainly more than many undoubtedly good readings elsewhere can boast of. On the difficulties of interpretation of lines 44 45, Scartazzini's exhaustive note should be con- sulted. XXI. 61. sol voler fa lOQRACDMU3>pti\<>>Fa. l (?)dgnp\zaw* 72.8.9.13.20.21. 28.30.33.35.42.46.53.96.100.102.104.112.119. [45] Lan., Off., Anon. Fior., Benv., Butt, Land., Veil., Dan. solvers! (or solver si) fa EFGHKLO(w^.)PSTUWXZTr.A r OS > rAn2ftaYJ;ii9K|j.Trp(rTv4>xsa*d(;;zrtr'.) efhijklmorstvwy^deg/utmq'rtv 1.3.4.6. 7^10.11.12.14.15.16.18.19.22.23.26.27.29.31.32.34.37.39.40.41.43.44.47.49.51. 55.56.57.59.63.66.68.70.71. 74.75.76.78. 79.81. 82.84.88 /0 95.97.98.99*. 101.103. 105.107.109.111.113. [132] suo voler fa ECFA 2 vbuc 52.67.85. sua voler far (sic) E. sua voler fa 69.73. sol voler la prova D\|/ 24. suol voler far prova 48.108. solver fa sua prova 36.54 (suo) 80.106 (suo). solver(e) fa prova on 17.38.72.77.114. il sol ver si fa B. sol voler si fa hx. solver si s' aprova *. suo verso fa prova (?) 58. sol per ver si fa (with p erased) V. sol veder fa prova 1 10. Delia immondizia solversi fa (!) f 23.51. This passage exhibits the curious phenomenon of the loss of the true reading, ' sol voler fa prova] in all but a small minority of MSS., though it is preserved and rightly explained by all the old Commentators who notice the passage, without exception (viz. Lan., Oft., Anon. Fior., Benv., Bnti, Land., D. I. 26 402 PURGATORIO. Veil., Dan\ nor do they so much as mention any variation in the text. Notwithstanding this, the feeble and almost mean- ingless reading ' Delia mondizia solversi fa prova ' is found in the large majority of MSS. This is perhaps to be explained by the fact that the clue to the right understanding of the true reading ' sol voler fa prova ' was one easily lost, and depending on a knowledge of Aristotelian, and still more of the scholastic, teaching, which the copyists could not generally command. It is easy to imagine philosophical, moral, or theological objections arising to the statement of Dante here (viz., that the mere wish to pass upwards is proof that the soul's pur- gation is completed) if unqualified by a consideration of the technical teaching out of which it sprung. There is an opposition between voglia and talento the key to the whole passage which is excellently explained by Jacopo della Lana thus: " Qui fa 1' autore differenzia tra la volonta. e il talento;... la volonta sempre vuole 1' ultimo e perfetto fine, ma la giustizia di Dio vuole, acci6 ch' ella sia tutta intera satisfatta, che, siccome lo peccatore ebbe volonta di peccare e pecco, cosl abbia talento di stare, e stia, e a sua suggezione e purgazione ; sicche lo talento e volonta secundum quid, com' e detto." So again Buti : " 1' anima ae due volonta ; cioe, 1' una assoluta e semplice, e questo sempre vuole lo bene sommo e perfetto, ne non pu6 non volerlo essendoli mos- trato ; 1' altra volonta e respettiva, e questo nol vuole se non per giusto modo, e questa cosi fatta volonta e quella che fa prova de la mondizia, cioe quando non contradice a la volonta naturale; impero che se non fusse monda, contra direbbe; e chiamala 1' autore talento ; e pero finge che Stazio dica... ' non lassa 7 talento' cioe la volonta respettiva non lassa la volonta libera et assoluta voler quel bene, se prima non si sodisfa alia giustizia." Similarly, though more briefly, Benv, : " il talento non lascia ; idest, appetitus non permittit ; quasi dicat, quod licet velit, tamen non desiderat ascendere ante debitam satisfactionem." 1 I find in the unedited Commentary of Serravalle 'De munditia solvenda fit experientia, qu?e tota libera,' &c., which seems to imply the reading solver si fa prova. COLLATION OF SELECTED PASSAGES. 403 Hence we have a fine illustration of the principle so often insisted on by Dante, that each man (as Milton says of Satan) is 'punished in the shape he sinned According to his doom'; or, as we read in Wisd. xi. 16, "wherewithal a man sinneth, by the same also shall he be punished." This again is very well put by Buti : " come la volonta respettiva [ = talento, comp. its use in Inf. v. 39] fu contra la volonta assoluta a fare lo peccato (che la volonta assoluta non puo volere lo peccato e lo male, se non ingannata sotto specie di bene); cosi e contra a volere lo bene, se prima non e sodis- fatto a la giustizia." (This is very well illustrated by Ar. Nic. EtJi. in. iv. and III. v. 17, &c.) In other words: In this life volonta, but for talento, would choose rayaOov, but it allows itself to be misled by talento into choosing TO aivo- fj,evov dyaOov, " r/v Se, et OVTCOS TW%, KCIKOV " (Eth. III. iv. 2). Accordingly hereafter, in retribution for this, the volonta has again to submit itself to the talento, which now chooses pain and punishment (lines 65, 66), and until " the uttermost farthing " is paid, the volonta is compelled to follow the talento in choos- ing this purgation instead of il sommo Bene (rayadov). Thus we read of the spirits in the /th Cornice Poi verso me, quanto potevan farsi, Certi si feron, sempre con riguardo, Di non ustir dove non fossero arsi. Purg. xxvi. 13 15. When all is paid, then the volonta is at once set free to aspire again to its natural object, the talento now no longer opposing it, and the feeling that this is so is proof that the purgation is complete : " Delia mondizia sol voler fa prova? We might then in fact apply the language of Virgil in Purg. xxvii. 140, i- Libero, dritto e sano e tuo arbitrio, E fallo fora non fare a tuo senno. The passage of S. Thomas Aquinas quoted in Scartazzini's note on 1. 64 is a most instructive commentary on the idea of Dante here. To use another Aristotelian distinction, we 26 2 404 PURGATORIO. may say that this purgatorial punishment presents itself as actually desirable, ' avayKalws ', or ' e' U7ro#ecre&>?.' The enlightened understanding knows that it is now the essential condition of, and only road to, Happiness, and consequently the well-regulated will desires it. 1 son content! Nel fuoco, perche speran di venire, Quando che sia, alle beate genti.' In this sense Perez (Sette Cerchi, &c. p. 50) remarks : "Dacche ella [!' anima], posta nella regione della verita, vede che beatitudine non puo acquistarsi se non col patire, ella dee avere il talento del patire (come Dante lo chiama), dee volere il patire con quell' ardore con cui vuol la beatitudine : solo quando sentesi perfettamente rimonda non puo piii volerlo, non puo pur sentirlo, perche e gia beata in Colui al quale s' e perfettamente congiunta V The peculiarly audacious substitution of immondizia in the Brit. Mus. T (a MS. of some celebrity) and also in '23' and '51' is noticeable, and it is also found in one of the "First Four Editions." It introduces a moral sentiment, which, however sound in itself, is absurdly out of place here. 1 There is a difference of reading in 1. 63 (viz. voler and volar) upon which the above discussion bears. But voler is, I imagine, so clearly the correct reading, that I should scarcely have thought it necessary to discuss the question, except that Witte has strangely admitted volar into his text ; no doubt because it is very strongly supported among his 4 MSS. Volar appears to be absolutely unknown to any of the old Commentators ; while voler is distinctly explained by Ott., Benv., Buti, Veil, and Dan. It seems clear that voler stands in close relation to the context both before and after the line in which it occurs. Up to this point the voler had been checked and impeded, now it first becomes an 'cWpycm avfpno- Storoy,' and is exercised with pleasure (le giova). The MS. evidence which I have collected is not very large, but is as follows : voler BCEGH I KZ^Sf/Fabdfghilnorstuv. volar ADF^CZ^Fekmw. The eccentric MS. ' j ' has salir ! It maybe noted that the words voler and volar axe again interchanged at Par. xv. 72; where voler is clearly right, but the neighbouring word ali suggested the tempting alteration to volar by association of ideas. (See note on Inf. xxxiv. 99.) COLLATION OF SELECTED PASSAGES. 405 XXII. 5. n' avea chimnw 4.6. 10. 13. 16.17. 19.23.30.32.33.36.38.41. 42.45.46.48.49.5 1. 54.56.57.59. 64.67.71.72.74.75.78.79.90.102/0 112.119. [92] n' avean DEFGHlKLRSTWXAfiCDZFGtfSJTAttKwo-rfazf opyzadfgkqrtx 3.7.8.12.15.18.20.21.22.24.26.27.29.31.34.35.37.39.40.43.44. 52.53.63.66.68.69.70.73.77.81.82.84.88.89.92.93.113. Bat. 30.102. [90]. m' avean yr\vdev 11.14.87.96. m' avea CiiOsuv/ 1.9.47.58.76.85. ci avea Zv. Drieto (or dietro) dicevan ej. XXII. 6. Sitiois almost universal, it was found in about 170 MSS. The only exceptions were- Sitiunt A'f/45 1 . Static P\O 20.46^ Bat. 246 (auct. Scarabelli). silentio v 87. sientio 108. sentio 110. h has sitio with a space and erasure, as if some longer word like sitiunt had been original. This is a case rather converse to the last, in that the right reading (unless I am mistaken) ;/' avea is not found in any of the earlier Commentators, yet it has considerable support among the MSS., being found in about half of those examined. It should be remembered, however, that the difference between avea and avean (or avea) is a very slight one, and, as will be seen, the reading avean is on so many obvious and superficial grounds 'facilior,' that if it could be proved that the change had really been from avean to avea, one would be tempted to suppose its having originated in a clerical error, there being no obvious motive for such a change, but the reverse. Still on fuller and wider consideration both of context and parallel passages, avea will I think prove to be decidedly more appro- priate. In respect of sitiunt in 1. 6, the facts are somewhat dif- ferent. The reading sitiunt is almost entirely devoid of MS. 1 A friend has just informed me that Sitiunt occurs in V 1 , and Scara- belli reports it in Bat. 405. 406 PURGATOklO. authority 1 . I have myself seen it only once (in '45 '), and found it recorded besides only four times, out of about 185 MSS. which I have either examined or found reports of : nor does it appear that any of the early Commentators can be quoted as explicitly adopting it. As far as I can ascertain, Modern Editors are about equally divided between sitiunt and sitio. Now no one can doubt the unity of plan and method, not only throughout the whole poem, but also in each of its three great divisions. (For these and other points of symmetry in the Purgatorio, see Table VII. in my " Time- References in the Divina Commedia."} This unity of plan is nowhere more marked than in the circumstances of Dante's passage from one Cornice to another of the Purgatorio. In each case the angel in charge of the Cornice removes one of the seven P's that have been impressed on his forehead, leaving him lighter for his upward journey. In every case also, putting out of sight for the moment the present passage, i.e. at any rate in 6 cases out of 7, this act is accompanied by the recitation of one of the Beatitudes from S. Matt. ch. v. Further, in 5 cases of these 6, it is distinctly stated, or fairly implied, that the Beatitude was uttered by the Angel himself. In the re- maining instance (Purg, xii. no) the difficulty of saying to whom the 'voci' should be referred is so great, and the opinions of Commentators on the subject are so various, that some (as Scartazzini) suppose even there the plural (voci) to be used poetically (and no doubt for the sake of rhyme), for the singular. In that case, an Angel would be the speaker there also. Perez (Sette CercJii, p. 132) also attributes these words to the Angel of Humility, paraphrasing voci apparently by parole. Such would seem to be the use of voci, if my general interpretation of the passage be correct, in this very line, 1 In defence of the supposition made above that the true reading here is the one almost entirely devoid of MS. support, I would add the curious evidence afforded in other places by some of the earliest Commentators of the existence of readings, which (whether true or false) have almost entirely disappeared from existing MSS. See the instances given in the Proleg. p. vii. COLLATION OF SELECTED PASSAGES. 407 the same consideration of the conditions of rhyme probably accounting for it. The other five passages to which I have referred are : Purg. xv. 34 9 ; xvii. 66 9 ; xix. 46 5 1 ; xxiv. 148 154; xxvii. 6 9. We may then I think take it for granted, (i) that the quo- tation here is certainly a Beatitude ; and (2) that it is probably spoken by the angel guarding the ascent : and consequently we should certainly read sitiwit, and most probably also avea. Next as to the early Commentators in respect of (i) the reading ; (ii) the construction ; (iii) the source of the quotation, or quotations, in lines 4 6. i. Avean is found in Lan., Ott., Anon. Fior., Benv., Buti, Land,, Veil., Dan., Serr. (Pet. and Fals. Bocc. are not ex- plicit on this point). Sitio is the reading of Lan. (?), Pet., Buti, Land., Veil., Dan. and Serr. Sitinnt may have been the reading of Ott., Anon. Fior. and perhaps Benv. At any rate they all distinctly ex- plain the passage as a quotation of the Beatitude, and, with the exception of Benv. (see later), make no reference to the reading sitio. ii. With the reading l aveau,' quei cK hanno &c. would be the subject of avcan : = 'ii qui &c. nos declaraverant esse beatos ' ; (though different explanations are given of the persons referred to by 'ii qui' &c.). With the reading l avea] quei cti hanno &c. would be the object of detto avea: = 'eos qui &c. nobis declaraverat (sc. Angelus) esse beatos.' Now quei cJi hanno is taken by all the above Commentators (except the two who are indeterminate) as a description of the spirits undergoing purgation on the 5th Cornice, and they are followed in this interpretation by Blanc ( Versuch, &c. ii. p. 82) and by many other modern Commentators. But (a) this would be an exception to what we find in 5 certainly, if not in all the other 6 Cornici 1 : and (b) these spirits had (as Scart. points out) already been left behind, and had returned to their accustomed acts of penance (see Purg. xx. 142 4). 1 This is recognized as a difficulty by Serravalle. He says ' Ista litera est valde fords'; giving as a reason that Dante does not introduce an Angel here as usual. 408 PURGATORIO. Another interpretation adopted by some modern Com- mentators (e.g. Lombardi and Riagioli) is to consider the words quei cJi hanno as a description of the Angels generally. But besides the objection (as before) of the anomaly thus introduced in the case of this Cornice, it has been argued by Blanc and Scart. that such a description of the Angels would be inappropriate, since they have the full possession, and not the mere desire, of righteousness. The interpretation which I am advocating would, as I have explained, take the words quei cJi hanno . . .desiro as the object of detto n'avea, and as part of the quotation of the Beatitude, being in fact equivalent to ' qui sitiunt justi- tiam' in S. Matthew. So it is taken by Fraticelli, Brnnone Bianchi, Costa, Scartazzini, &c., but not, I believe, by any of the early Commentators, who do not in fact recognize the reading 'avea' at all. iii. As to the Scriptural quotation or quotations, (a) Lan. paraphrases ' Beati immaculati' as though the citation came from Ps. cxix. I. (/3) Buti (who is followed by Land, and Veil.} traces the quotation to Ps. xxxii. I ; " adjungendo al Salmo 'sitio'" which all three Commentators explain, "io ho sete dell' eterna salute"; Buti alone adding that the word has reference to our Lord's utterance on the cross mystically understood 1 . The brief comment of Lan. seems to imply the same general explanation of sitio as that of the three Commentators above named. It is, I think, a strong objection to this view of the origin of the quotation "Beati &c." (though I do not remember to have seen it urged), that this passage from Ps. xxxii. I is recited on another occasion, and in a different part of Purgatory, viz. at the entrance of the Earthly Paradise, when the last of the purgatorial Cornici has been surmounted, and all sins expiated, and their very memory is about to be effaced in the waters of Lethe. See 1 Philalethes also appears to suggest that perhaps ' sitio ' may stand in the ist person sing. pres. as a general way of referring to the verb ' sitio' occurring in the Beatitude as sitiunt, though he is unwilling to exclude a possible reference also to our Lord's use of the word. Serravalle agrees with Buti in the whole of his explanation, adding in reference to our Lord's word ' sitio,' ' desiderans salutem populi, sed isti dicebant de seipsis.' COLLATION OF SELECTED PASSAGES. 409 Purg. xxix. 3. The quotation is far more appropriate there than here, and in any case it would not have been likely to occur in both places. It is also an objection to both (a) and (/3) that, on either explanation, in the case of this Cornice alone there is no reference made to any of the Beatitudes. (7) The quotation is referred with, I think, undoubted cor- rectness to the 4th Beatitude in S. Matt. v. 6, by Off., Benv., Anon. Fior. and Dan. (though the last-named appears to read Sitio). Also by Pet., though he adds (referring to 1. 6) " se- cundando verba Christi in cruce dicta, scilicet, l sitio\" The comment of Benv. is as follows: " beati cum sitiunt ; idest, simpliciter cum hoc verbo sitio, quia dixerunt solummodo, 'beati qui sitiunt '." If we could be sure that the first italicised words were Benvenuto's actual quotation of the text before him, this would decide the question as to his reading; but such quotations, as we have often seen, cannot be trusted, and the language that follows seems to make it more probable that he read sitio, but did not consider it incompatible with the reference being to the Beatitude, as was apparently the case with Dan. This interpretation, he adds, 'magis videtur de mente autoris' than the other, which he describes as that of 'alii subtilius exponentes' ; Dan., though quoting the text as 'Beati con Sitio', explains it by simply quoting the Beatitude at full length. The reference to the Beatitude is here obscured by the somewhat awkward way in which the words ' Detto ri avea' break into the quotation ; also by the inversion of order by which Beati comes last instead of, as usual, first ; also by the free and altered form in which the quotation is made, and the mixture of Italian and Latin in it ; and finally by its frag- mentary character. On this last point a few words may be added in conclusion. In order to supply the required number of appropriate Beatitudes for the several Cornici, this one had to be divided, and a separation introduced between 'hun- gering' and 'thirsting' after righteousness. The former is reserved for the 6th Cornice, where it affords a natural contrast to the sin of gluttony (see xxiv. 151 4), while the latter offers an equally natural antithesis here in the 5th Cornice to the sin of avarice ; which is so constantly described as a 4IO 1'URGATORIO. ' thirst ' for gold that we are scarcely conscious of the metaphor. Hence it was necessary to quote the Beatitude more loosely and fragmentarily, with the result that by many the quotation has not been recognized. It may be ob- served that when the other portion of this Beatitude is quoted in xxiv. 151 4 it also is cited much more freely than in the other cases, and it is further the only case in which none of the original Latin appears in the citation. For another instance of a quotation cut short at a certain point comp. Purg. xxx. 84. The appropriateness of the love of giustizia being thus pointedly opposed to the sin of Avarice is not perhaps at first sight obvious. It would be illustrated by the argument of Dante in De Mon. i. 13 that a Universal Monarch is most likely to practise justice because, his power and dominions being unbounded, he is beyond the reach of cupidity, ' quod justitice maxime contrariatitr ctiftiditas'...and ' remota cupi- ditate omnino, nihil justitice restat adversum ' (with this we may compare Aristotle's statement (Nic. EtJi. IV. i. 23) Ato roi)9 Tvpdvvovs ov Xeyo/iez; ao-wrou? /c.rA.). Also Perez (Sette Cerchi, &c. p. 215) aptly quotes the language of Aquinas as follows : " Ubi intelligatur de justitia spirituali, quod est quod homo reddat unicuique quod suum est, con- venienter dicitur, 'Beati qui esuriunt et sitiunt justitiam'; quoniam esuries et sitis proprie avarorum est, quia nunquam satiantur qui aliena injuste possidere desiderant. Et vult Dominus quod ita anhelemus ad istam justitiam, quod nun- quam quasi satiemur in vita nostra, sicut avarus nunquam satiatur." (Comm. in Matth. c. v.) This passage of Aquinas may not improbably have suggested Dante's application of the Beatitude here. XXII. 97. antico ABCDEFGHKOPSTUWXZGl.Tr.^CZ?#/7^S^F*rAII S^a^KixirpoT^xfsabidefijmnopruvwyz^^z/^t/w* 1.3.6.7.10.11.12.14.15.16. 18.19.21.23.24.27.28.29.31.32.34 to 43.47.48.52.55.56.57.59.64.69.71 to 78.80. 81.82.84.85.88.89.90.93.95.96.102.103.105 . 106. 107. 109. 1 10. 1 1 1 . 1 13. [141] Benv., Dan. amico ILQREFGQy^) A$pY0iX-vv|/iob^gh*klstxrt^r/w 1 (?).r 4.8.9- 17.20.33.44.46.51.58.68.70.104.108.112.117.119. [55] Ott.)Buti,Land., Veil. COLLATION OF SELECTED PASSAGES. The MS. evidence largely preponderates for antico, though amico is not by any means uncommon. The former is most probably the true reading on other grounds also. The ex- pression nostro amico is so much more obvious and common- place, that, had it been original, nostro antico would not have been likely to have been thought of. Apart from the slightly unusual expression ' nostro antico', a copyist may have had a misty recollection of the use of amico (as between Virgil and Statius) twice, a little before (in lines 19 and 21). Though the reason for that is sufficiently explained by the context, there would be little or no propriety in applying it to Terence in relation to either of them. The only suggestion in ex- planation of it is the slender one that they were all poets ; which is made by Ott., Buti and Land., and supported by Biagioli with a reference to Inf. iv. 91. This might perhaps be thought to have had more force if they had belonged to fields of literature less widely separated than Epic and Comedy. Benv. and Dan. read ' antico', the latter protesting strongly against ' amico'. As in the case of nuovi and vivi in xx. 90, we unhesitatingly give the preference to the word which, on reflection, gives a definite and appropriate, though at first sight less obvious meaning, over that which is quite vague and commonplace. It is not without bearing on the reading here, as well as on a disputed point in the next line (viz. whether we should read Varro or Varid), to observe that Dante has apparently been guided in the selection of these names, Terence, Caecilius, Plautus and Varro (al. Vario) by the recollection of one (or perhaps both) of two passages in Horace. These passages are: Plautus ad exemplar Siculi properare Epicharmi; Vincere Ctzcilius gravitate, Terentius arte. Ep. II. i. 58, 59. and Ccedlio Plautoque dabit Romanus ademptum Virgilio Varioque 1 ? A. P. 54, 55. 1 Varius is mentioned twice again by Horace in conjunction with Virgil, viz. Ep. n. i. 247 and Sat. I. vi. 55, besides (which is curious in reference to the controversy noticed above) a passage in Sat. I. x. 44 46, where Varius, Virgil, and Varro (Atacinus) all occur together. 4t2 PURGATORIO. Either passage therefore gives three out of Dante's four names (if indeed we should read Vario). In favour of the former passage is the consideration that in that Epistle Horace is criticizing at considerable length (lines 18 78) the practice of admiring poets on the sole ground of their antiquity, and these writers are expressly given (among others) as instances of ' veteres poeta' (see 1. 64), so that the idea and expression ' nostro antico ' would be in keeping with the idea pervading this passage of Horace, for Terence certainly there appears (with others) distinctly as ' nostro antico' On the other hand, in favour of the passage from the Ars Poetica, and as against the reference to the Epistolce, is the fact that (as far as I am aware) there is no other evidence of Dante's familiarity with any of the works of Horace, except the Ars Poetica, which he definitely quotes at least five times (see Appendix on Dante's quotations from classical authors). It is true that the designation of Horace as ' Orazio satiro ' in Inf. iv. 89 (while presumptive evidence that Dante was not familiar with the Odes, on which Horace's title to immortality would more naturally be based), would seem to imply some ac- quaintance with the Satires. There is no evidence of this that I know of in Dante's writings, but the word may perhaps be taken in a wide sense to include all Horace's hexametral poems. Still the balance of probability, as I think, inclines to Dante's having had in view here the passage from the Epistles. I regret having overlooked the con- troversy as to Varro and Vario, and consequently not having collected information on the subject from the MSS. examined 1 . 1 The following is all the evidence that I have been able to gather : Varro AFGHIV 1 V 2 Z^4 J 5CZ>5f/F'v4>xadfmop^i8.i9.23.27.28.29.3i .56. 57.75.77.90.93.95.98.99.102.105.113.116.117. Bat. 146. Varo CDEKLFegijkn 76.106.119. Bat. 145.362. Varo V%1. Vario 63. Some of the varieties in the other names found here are amusing : Bj has Oecilio, Plato e Varo, with a u added to Plato and an r to Varo 2nda mann. 98.102 have also Plato with u added. C, had ' Plauto,' and the ' u ' has been erased. Plato is found in D|>fh (Plathro) 19.76.77.97.116.119. Prato is found in jl. g has Cecilio Pluto e Varo se lonsai (sic). COLLATION OF SELECTED PASSAGES. 413 As I understand, no MS. evidence is claimed for Vario. (See however note on last page.) At present it is merely an in- genious conjecture of Dr Witte, which has been adopted and advocated by Blanc and others. This will not affect the question as between antico and arnica, since the latter term, though it might have been appropriately used by Virgil in reference to either Varius or Varro, could not be adopted by Statius, who is the speaker here. XXIV. 61. pill a guardar CFGQ 1 RZ>7^^ AA * 2 P ^ ox l" 1)a i' ? 3.8.9.11.20.21.33.41.42.52.58.67.69.70.71.78.79.87.91.95.98.99.103.104.105. 109.114. Bat. 55. [66] piu a riguardar DOTr.ACFT&FdtvA 22.24.35.48.80.100.108.1 19. [21] piu a gradir BPQ 2 SUV 1 vn^Z67//ZFna?xsa*hknr^ 1.4.10. 16.17.18.19.23.27.28.29.31.32.36.37.39.49.51.54.55.56.57.59.75.82.85.90.93.94. 97.102.106.107.111.112.113. [66] piu a gridare w 38.45.64.72.74*. 101. [8] piu a gloriar A. a riguardar piu EHKT(?) VJX^KpTrpo-rmt 7.12.14.15.26.34.40.43.44. 63.66.81.84.88.89.92. [32] piu oltre a riguardar GLyfy 68.73. [6] piu oltr' al riguardar I. piu a guatare U. a gridir piu oltre S. a gridar piu oltre 76. a guardar piu oltre 77. piu al gradir 96. E qual riguardar piu Lv. E quale a riguardar oltre g 30. piu al gradare (sic) 1 10. T and 63 have a riguar piu (sic). There is, as it will be seen, a very great amount of varia- tions in the reading of this line. Passing over small changes a has Cecilio Plauto dimmi se lo sai. 97 has Vafio (sic). 95 has Cecilion Plusto (sic) e Varro. k has Cicero Plato e Varo. Among the Commentators, Ott., Benv., Buti, Land, have Varro ; Pet. and Anon. Fior. have Varo, probably referring to Alfenus Varus (see Hor. Satt. I. iii. 130), since Pet. says, " Varrum (sic) declamatorem tempore Caesaris"; and Anon. Fior., "Varo scrisse il libro delle declamazioni al tempo di Cesare," PURGATORIO. in the order of the words, which are probably euphonic (or would-be euphonic) in origin, the principal point is to de- termine the verb to be selected. The following are all found : 1 guardare 2 riguardare 3 guatare 4 gradire 5 gridare 6 gloriare Of these the last is evidently a conjecture, and is found only, as far as I know, in the Bodleian MS. 'A.' There is not much more to be said for gridare, which I have met with in only 8 MSS. out of more than 200 examined. Guatare and guardare are not unfrequently interchanged in MSS. (e.g. Inf. xvi. 78, Purg. viii. 96, &c.), but in this passage guatare may be dismissed as having the support of only one MS., viz. U. Gradire presents itself with a very considerable amount of MS. support. A few of those who have adopted it explain it in the very improbable and unusual sense of ' gradior' (as Tommaseo, Blanc, Philalethes), but it is generally taken in its more usual sense of ' to please.' In this way we have a comparatively easy, may we not perhaps say, suspiciously easy, interpretation of the passage ? Moreover, though I would not urge this too confidently, there seems to be a little redundancy in the use of both words piii and oltre with gra- dire in this sense, though not in that of 'gradior' I there- fore distinctly prefer guardare or riguardare. The former has much larger MS. support, and moreover (if such an argument be admissible) it suits the rhythm of the line much better. It is also not impossible that riguardare might have appeared to some copyist a preferable, or more familiar, word in itself. (Compare the interchange of posato and ripo- sato in Inf. i. 28.) Another and stronger argument for guardare, however, is that it would more naturally have given rise to the three or four other variants than riguardare would have done. The following arguments may be used in favour of one COLLATION OF SELECTED PASSAGES. 415 or other of the words guardarc or riguardare, as against the other three or four variants : (i) Pm and oltre have certainly no appearance of redundancy in this case. We commonly speak of looking 'further ahead/ or 'further in advance,' or ' further beyond,' &c. Pih oltre, besides being often used by Dante with verbs of motion, occurs again with veggio in Inf. xvii. 35, and xxxii. 64. (2) There is an appropriate and thoroughly Dantesque irony in the antithesis between guardare here and vede in the next line. ' He who sets himself to look further beyond, i.e. to look further ahead than that which his Master (Love) dictates, can no more see the difference between one and the other style.' Professing or endeavouring to see more, he becomes blind. His self-reliance makes him incapable of the true inspiration, and even of ordinary discernment. Preferring to ' follow his own imaginations,' he is given over (poetically speaking) to an ' aSd/a/xo? vovs.' Compare Inf. xx. 38, 39, which is closely parallel : Perch voile veder troppo davante, Diretro guarda, e fa retroso calle. For the idiom in 1. 62, ' dall' uno all' altro,' which seems to have been overlooked or misunderstood by most interpreters of the verse, compare Inf. xix. 113, E che altro e da voi all' idolatre? i.e. 'What difference is there between you and the idolater?' The following quotations from the early Commentators will shew not only how widely they differed in their inter- pretations, but also how difficult it is sometimes either to determine the reading which they adopted, or to trace any close connexion between their explanations and any of the readings. Lan. merely says, 'chi ad altro modo considera li stili non vede lo vero.' He probably therefore read guardar or riguardar. Ott. says vaguely, 'chi lauda quegli antichi dicitori.' A non. Fior. : 'chi vuole cercare se furono innanzi a frate Guccione &c., dicitori in rima piu famosi di loro, nolli trova.' Benv. is very vague and brief. He quotes 1. 61 with guardar, merely adding 'de hoc' (to explain 416 PURGATORIO. guardar); and 1. 62 he explains, ' non cognoscit plus... a tuo moderno (stilo) alP altro, scilicet, nostro antique.' Butt, reading riguardar t explains the line thus : ' cioe lo tuo dire e lo nostro'; and 1. 62 he explains: 'non vede piu di differenza dal tuo modo di dire al nostro, che quello che ditto e ; che tu vai stretto al movimento dell' animo e noi larghi.' Land, explains 1. 61 similarly, but on 1. 62 he says, 'non si puo vedere tutto lo spazio che e tra lui e chi rimane dietro'; quoting Virgil's 'proximus huic longo sed proximus intervallo.' Veil, reads riguardar, and explains the whole passage quite clearly and definitely in the sense which I have advocated above. He paraphrases pin oltre riguardare by " a volersi avanzare nello scrivere oltre a quello che li ditta amore"; and proceeds, " non vede piu oltre la differenza che sia dallo stile di chi egli li detta a quello di chi non." Dan. correctly inter- prets the idiom in 1. 62, and also clearly explains the text with the reading gradire, which however it is not difficult to do. It appears to me that Scart. makes needless difficulty about the interpretation of the passage, and in any case that his own unwarranted conjecture, in spite of the exul- tation with which it is introduced, need not be seriously discussed. XXV. 9. Che per artezza i salitor lRA*CDUIT2pi&Fdhn*pxer(marg:) w*x 20.21.35.46.91. [27] Anon. Fior., Veil., Bttti, Dan. Che per artezza il salitor OQF^gn^/i (F artezza) 7.11.40.73.90. 1 08. [15] Land., Benv. Che per ertezza i salitor HZ^^?) BE\^a\qr 6.18.19.22*27.31.32. 36.37.38.47.54.55.64.67.70.82.92.103.106.111. [34] Serr. Che per ertezza il salitor BCDEFGKLSTUWXZ^^TAa^virpo^ Xwbfjmrsuwzc*#>//t/ 4. 10. 13. 14. 15. 16.17. 22^?) 23.24.26.29.39.41.43.44.45. 48.51.56.57.58.59.75.77.81.84.93.101.107.109.112.113.119. [79] Che per altezza i salitor ^fiok; 1.30.33.42. [9] Che per altezza il salitor APGl.A4>0Tty^ 3.52.666.9.71.85.87.95.110. 117. [21] ertezza il salir 102.105. Perche F altezza i salitor Tr. Perchfc F altezza il salitor v 79. COLLATION OF SELECTED PASSAGES. 417 Perche 1' altezza al salitor fg. Perche altezza al salitor a 78. Perche artezza ai salitor 9. The following were imperfectly registered : eltezza il salitor 12 al salitor F. ertezza NO 74. altezza 80. NOTE. It is possible that some of the MSS. registered as ' Che per ' may have had ' Perche ' as my attention was not specially directed to this minor point at first. There can be no doubt that the true reading of this line is Che per artezza i salitor dispaia. There are however two distinct and independent forms of corruption, (i) ertezza, and (2) il salitor: and since many MSS. which escape one error fall into the other, the variants must be exhibited separately, though they may be discussed together. The ' fons et origo mali ' was no doubt the some- what unusual, or at any rate ambiguous, form salitor, for the plural salitori. This suggested 'il salitor' quite naturally, the resulting inappropriateness of dispaia being overlooked. Next it occurred to some one that the steepness rather than the narrowness of a path causes difficulty to a single climber, and numberless passages would offer themselves in which this is referred to by Dante himself. Hence there resulted a further substitution of ertezza (steepness) for artezza (nar- rowness). (The words arte and erte are found together in Purg. xxvii. 132.) It seems to have again escaped notice that dispaia (for which see Inf. vii. 45) is as meaningless with this change as with the other. Artezza may compel people to separate (dispaia)^, but not so ertezza. The same objection applies to the reading altezza, which bears the stamp of a con- jectural emendation still more plainly, though it is not uncom- monly found. The MSS. however reading ertezza are more nu- merous than those for artezza and altezza combined. Among 1 The corresponding word 'dispair' is quoted as occurring in Beau- mont and Fletcher, ' I have...dispaired two doves,' D. I. 27 41 8 PURGATORIO. the early Commentators, Anon. Fior., Benv., L and., ^V ell., Butt and Dan. all explain the passage correctly, though Benv. and Land, read il salitor without apparently noticing its inappropriateness. Dan. paraphrases, 'si stretto che non lo potevano salire accoppiati'; and Veil, explains dispaia by ' discompagna i salitori.' Butt notices the variant ertezza, and rejects it. So also does Benv., who was further aware of the prevalence of ertezza in MSS., since he comments thus : " et non dicas, per ertezza, i.e. altitudinem, sicut communitcr habent textus, nam altitudo scalae non impediret paritatem, etiam si pertingeret usque ad ccelum vel Deum." The expla- nation of Lan. seems to imply the reading altezza (or pos- sibly ertezza], and also to confuse dispaiare with disparire. It would seem therefore that here, as sometimes elsewhere, the evidence of the early Commentators and that of the MSS., as to the proportionate prevalence of rival readings, points in different directions; since ertezza among the former has no support, and scarcely any recognition. Serr. merely has 'per ertitutinem ascendentes dispares facit'; so that he read 'per ertezza i salitor! XXV. 31. veduta . Ott. (?), Anon. Fior. (?), Benv. [82] vendetta El KWX Ft T)Kir P T. [12] Buti, Land., Veil. (?),/?. verita D. vertti te eternal! (szc) G. vertii eternal j. veduta externa d. The MS. evidence here (so far as I have gathered it) is quite overwhelming for veduta?. On the other hand, about f Auct. Mussafia, though Scartazzini quotes this MS. for veduta. 1 Blanc ( Versuch, &c. p. 96) is therefore entirely wrong when he states the MS. authority for either reading is " fast vollkommen gleich an Zahl und an Gewicht." He is perhaps not altogether uninfluenced by his very strong preference for vendetta. He condemns veduta as "unbestimmt und matt." In his Dizionario he seems to prefer veduta, but is equally inaccurate in his statements of fact, since he states that "tutti i moderni COLLATION OF SELECTED PASSAGES. 419 two-thirds of the Editions which I have been able to consult read vendetta, and it seems to have larger support among the early Commentators than any other reading. At the same time one cannot but think that the majority of the MSS. have preserved the true reading in veduta. Plausible reasons might be given to account for either reading being original and having been displaced by the other. For though veduta is far more applicable to the whole tenour of the ' unfolding ' (dislego) in the following statement of Statius to which this expression refers, vendetta might seem more appropriate to the preceding difficulty (11. 20, i) by which that statement is elicited. Still I must say that though the latter argument may well have led an early copyist to substitute vendetta for veduta, the former is far more suitable to what is now pre- sumably in the mind of Statius (the speaker), and what is presently on his lips. It is very appropriate that Statius should modestly express reluctance to expound the ' eternal view' (veduta} in Virgil's presence (see xxi. 94 102; xxii. 57 seqq., 64 81); but there would be something like bad taste (see Purg. iii. 40 45) and an approach to 'things one would rather not have said,' in Statius declaring that when it comes to expounding 'vendetta divina' that would be more suitably left to Virgil. Among the early Commentators, Lan. is singular in reading vertute. In the Ott., the text is cited for explanation with veduta, but there is nothing in the note to make it clear what the Commentator actually had before him. Anon. Fior. in like manner is made to quote veduta, and then his comment proceeds : " s' io gli dispiego la verita, cioe la virtu di Dio." (It does not therefore seem certain that he did not read verita.) Benv. quotes vediita eterna, and adds " idest veritatem eternam hujus quaestionis." Buti,Land. and Dan. have vendetta, which they explain by giustizia, and justify the application of the former word to God's actions in a way which throws light on the probable motives for the substitution of the milder word in some passages of the text itself, to which I have hanno adottata questa lezione," which is certainly not the case. He says of vendetta eterna, " che sembra espressione poco conveniente alle pene del Purgatorio le quali sono a tempo." 27 2 420 PURGATORIO. called attention in the note on Inf. xi. 90. Indeed Vellutello quotes this passage in such a way as to make it seem pro- bable that he actually read giustizia for vendetta here. Giuliani reads veduta interna, purely e conj., as it would seem. At any rate I do not think I have ever met with internet in any MS. in this passage, nor does Giuliani profess to have done so, though the words are often confused elsewhere, e.g. Par. xvii. 9, xxiii. 115, &c. XXV. 131. Si tenne was found in about 1 50 MSS. Corse in one only, viz., I. Though the former reading is practically universal in the MSS., and is on other grounds undoubtedly the true reading, yet a very considerable number of Editors (see Scartazzini's note) have ventured to substitute Corse. This is one of the pas- sages in which (inter alia) the obvious imitation of Ovid helps us to determine the text. See further on this Appendix I., and also Scartazzini's note here, where the case for Si tenne is fully and clearly stated. It might be added that the action implied by this reading is more appropriate to self-restraint. As Benv. remarks, ' solitudo est maxima arnica pudicitiae.' It is possible that Corse was inserted by a copyist, whose memory was better than his judgment, from a reminiscence of Purg. xviii. IOO seqq., where the circumstances are quite different. XXVI. 41, entra was found in nearly 1 50 MSS. entrb in 9 only, viz., \EFm// 43.44.58.59.71.75.96. [12] NOTE. acti (or atti) cor occurs in npYtOKoirpty 10.14.15.18.19.27.31 (and perhaps others, since atti, alti and acti are scarcely distinguishable). negli altri lor tosto s' attuta 69. color tosto si muta LESVAwcev 38.72.76.97. color tosto muta CKZ(marg.) b u. negli atti e color tosto si muta d. Le quali negli atti il cor tosto si muta 3.27. nelli altri cuor tosto s' aruta (? = s' atuta) a. negli altrui cuor tosto s' attuta \ 23. nel glial cuor (sic) X. tosto si stuta (? = s' attuta) -y. negli atti suoi tosto si muta i\. tosto saluta (? = s ! atuta) C. j\ had tosto ebbe muta. The true reading here is certainly s' attuta, since its exist- ence will explain readily the other variants, but it could scarcely have arisen from them. The unfamiliar character of the word attutare will at once account for the substitution of the more obvious and common-place si muta. The word seems most naturally to connect itself with the Latin tutari (so Diez s. v. tutare) though other derivations have been sug- gested. Again, the word, though rare and likely enough to be unknown to some copyists, is found elsewhere ; e.g. the Vocab. Tram, cites instances from Boccaccio, Sacchetti, Giov. Villani, Tasso, &c. Its meaning in those passages is also 422 PURGATORIO. perfectly appropriate here, viz. to ' mitigate ' or ' calm down.' The further substitutions of color for cuor, and of attior acti for alti seem to be mere blunders ; but that of muta for attuta was no doubt the deliberate introduction of an easier word. The sentiment also in reference to 'alti cor' is a natural one, and is in harmony with that of ' II Filosofo ' in Ethics IV. iii. 29, where he says that the ^^aXo^fv^o^ is not Oavfiaa-riKos, and that of S. Augustine (quoted Ji. 1. by Tommaseo), ' Ignorantia mater admirationis.' (Compare just above, lines 67 9.) There is no trace whatever in the old Commentators of any of the false readings here. Benv., Buti, Land., Veil, and Dan. all adopt and rightly explain s' attuta. Lan., Ott. and Anon. Fior. pass over the passage. It is curious that si muta is found in the 'first four Editions' and also in the early I5th Century Catalan Translation of Febrer, yet it does not occur once in 25 later Editions of various dates which I have examined, s y attuta being found in all cases. XXVI. 75. viver A(marg.} ZtCDFdkp 27.93.95. Bat. 234.235.257 (The last three are given auct. Scarabelli). [14] Benv., Dan. morir AECDEFGHlKQV^V^V^ZAEGHf^QSfy^v^x abefghijlmnrt uvw/y 18.19.23.28.29.30.31.54.56.57.63.75.76.77.90.97.98.99.116.119. Bat. 362. [67] Buti, Land., Veil., Serr So also in 15 out of 19 Codici, the collations of which are published by Scarabelli. The very interesting variants morir and viver were unfor- tunately overlooked until the bulk of these collations had been made. Consequently I have comparatively little MS. evidence to offer. On critical grounds, apart from MS. evidence, viver appears to me preferable, partly on the ground urged by Scart., that it suits better than morir the sense of 1. 58 just above, "su vo per non esser piu cieco." To this passage we may perhaps add xxiii. 115 120. Also while ben morir is a common phrase, the same can perhaps hardly be said of meglio morir. Morir has further a sort of prima facie suitableness in the mouth of spirits in Purgatory, which may have led to its substitution, if such be the case. COLLATION OF SELECTED PASSAGES. 423 The only old Commentators who notice the passage are divided, as given above. Benv., reading 'viver\ adds : "Nee dubito quod poeta melius vixit, et melius mortuus est, per compilationem hujus operis.'' Land, and Veil, explain morir mcglio naturally as meaning " in maggior grazia di Dio." Buti rather curiously thus : "che non seresti morto se non avessi veduto 1' esperienza della nostra purgazione," which looks as if he understood morir in a spiritual sense as ' dead to sin.' According to Scartazzini, viver seems to be the reading approved by the vast majority of modern Editors and Com- mentators. This is confirmed by the evidence I have col- lected myself, since of the 24 Editions mentioned before, no less than twenty-two have viver and only two (viz. Witte and Camerini) have morir. Per contra the first four Edd. all have morir. It will be observed that Witte adopts morir against the weight of evidence provided by his four MSS. ; i.e. on the strength of the support of A against A(marg.)BC and D. XXVII. 4. da (or di) nona K?QCDEFHlJKUVt&$^\^<^(i} din^pw ab v dhmntvw 9. 14.20.21.24.33.35.42.44.46.52.71.73.75.79.85.88.91.94.99.100. 104.107.109.110.112. [65] Pet., Veil. da nova DGHLPSTV^ 4 X^ 2 SAi>ateK|j^pTvlmrtuxz6- 4.6.10.12.15. 16*. 1 7. 19.22. 23.26. 34. 36.38.40.43.48. 55j.59-64.66.67. 68.69.72.74.81. 89.92.106. [61] di novaEK (Pdinona) UWox^ 30.47.49.58.103.111.114.117.119. [18] da novo C*A r vekn 2 ^- 41.54. [9] di novo A\CyRV l ZG}.T:r.A l fiD(marg.} OTHERS c *b a fjvy7?r.r 1.3.7. 1 1 . 1 3. 1 8.27.29.3 1 .32.37.39.45*. 5 1.56.57.63.70.76.77.78.80.82.84.87 . 90. 93. 95 . 96.97.98.102.105.108.113. [65] Ott., Benv., Buti, Dan. dinuo/. da non variarsi 8. d (marg.} has al. nuovo. p (marg.} has al. di novo. h* has danhona riarse (sic}. Again we are able to determine the reading with full confidence. There can be little doubt that the easier and more obvious word nova (or by a further and more complete emendation novo] has been substituted for the less usual and 424 PURGATORIO. possibly misunderstood none?. A further reason is that nona when explained furnishes exactly the right sense for the passage, whereas nova or novo would give rise to a statement either meaningless or absolutely false, since the only assign- able meaning would be that it was sunrise in India at the same time as at Jerusalem. Nona is of course the description in ecclesiastical phraseology of mid-day. See Conv. iv. 23 (quoted by Scartazzini), " la diritta nona sempre dee sonare nel cominciamento della settima ora del dl." [Cf. also Conv. iii. 6.] It will be remembered that in Par. xv. 98 nona is again found. Now according to Dante's geographical system, the Ganges was the extreme Eastern limit of the earth, the Pillars of Hercules the extreme Western limit (cf. Inf. xxvi. 107 109) 2 . Jerusalem was the 'o/u-(/>aXo9 r^? 7775' half way between, with the Mountain of Purgatory at its antipodes. Hence when it is sunrise at Jerusalem (see lines i, 2), it would be sunset on the Mountain of Purgatory (1. 5), midnight in Spain (1. 3 ; since at the vernal equinox the sun being in Aries, the sign Libra will be 180 distant from the sun), and midday on the Ganges (1. 4) 3 . Thus all comes out perfectly clear and accurate in the description of these opening lines, and only gross ignorance or thoughtlessness can have ori- ginated the readings nova and novo, which have been as thoughtlessly repeated in the majority of MSS., Commenta- tors, and (I think) Editions, though among modern Editions 1 Another instance (probably accidental) of the interchange of nova and nona will be found at Inf. xxviii. 21. 2 This passage in the Inf. may be illustrated from B. Latini, Tesoretto xi. 11. 117 136, and Tesoro iii. c. 3. In the former, Brunetto, after a statement similar to that cited above from the Inf., adds that after the death of Hercules (!) people went beyond this limit, and colonized a fair and rich country, which by a comparison of the passage in the Tesoro, appears to be England. 3 This synchronism, the consideration of which explains four other passages in the Purg. (viz. ii. I 9 ; iii. 25 ; iv. 138, 9 ; xv. I 6), is exhibited by a diagram in my ' Time-References in the Divina Corn- media^ Table iv. I have also applied it to the explanation of the present passages, and the others above quoted on pp. 68 73 of that work. COLLATION OF SELECTED PASSAGES. 425 there is a good preponderance for nona. Among the old Commentators, only Pietro and Vellutello read and explain the passage correctly. Benv. (' di novo ' = de mane), Dan. ('nuovamente riardendo '), Ott. (' ogni dl lo riarde una volta), and Butt, all have di nnovo. Lan., Fals. Bocc., Anon. Fior. and Land, are vague or altogether silent. Of the ' first four Editions,' three have nova, and one has novo. Febrer's trans- lation also has novo. Of 25 later Editions examined, 15 have nona, and 10 have novo; but among the Editions published within this century nona is almost universally found. The familiarity of this way of referring to the hours of the day may be illustrated from the old Chroniclers, and inter alia from the Pisan chronicler referred to in my note on Inf. xxxiii. 26, where the battle in which Ugolino was defeated and captured is described as having lasted 'quasi da di po' Nona in fine a presso a di po' Vespero.' XXVII. 81. e lor poggiato serve CDEHIKLPRSTUWXZ/AWJ/AII2ai;Kn virpxsa*fikmnruxzd/ 1 r/z/ 1.3.4.6.7.11.12.14.15.16.17.18.19.22.23.24. 26.27.29.31.32.36.37.38.39.40.43.44.45.51.54.55.56.57.63.64.66.68.72.74.75.76. 79.81. 82.84.85 1 .9o.92.93.io3.io5.io6.ui. 113.114. [112] Veil. e lor di posa serve B*FGOOGl.Tr.^CZ>^67yrApYi\o^fdeghjlp tvviyagltftw 9.20.21.30.33.35.42.46.48.69.70.71.73.77.78.80.96.99.104.107.108. 109.119. [64] Benv., Butt. e lor di poscia A 10 (possia) 94 (do.) e lor con posa b. e lor poggiando Tj9. Appoggia se e lor dispensar serve 8. Ri)^! have Poggiando se. G has Poggiato sta. B! had diposato serve. In respect of MS. authority there is ample support for either of the two readings lor poggiato serve, lor di posa serve, though those for the former largely preponderate. I cannot but think that the arguments of Scartazzini h. /., and of Blanc (Versuch, &c. p. 107), are nearly conclusive in favour of the 426 PURGATORIO. former, though the coincidence of Witte's 3 MSS., A, C, and D, is a strong support for the latter. The words ' Poggiato s 2, e lor poggiato serve ' very graphically depict exactly what any one may see in the Roman Campagna, the flocks re- posing, and the shepherd watching them, 'leaning upon the top of his staff.' Moreover they imply that though the shepherd may thus rest himself, his watchfulness over his flock continues, even as did that of Virgil and Statius over Dante in this case. The picture is thus both graphic and appropriate. The words lor diposa serve however seem to give a false image altogether ; for, as Blanc remarks, the goats do not cease feeding and rest because the shepherd stands leaning on his staff: there is no connection between the two things, not to say that the very idea of his 'ministering repose to them ' is a very strange one, though it must fairly be admitted in reply to this, that in this case it has been already stated that the goats are resting (lines 76 79), and Dante, who is compared to them, is also resting. It may be added that if lor poggiato serve had been the original reading, it is conceivable that a pedantic objection may have been felt by some copyist to the repetition of poggiato, and the variety of expression introduced as a fancied improvement. In spite of Scartazzini's imposing list of authorities among the early Commentators, I can find only three who declare themselves here (in no case noticing a different reading from their own) viz. Benv. and Butt, who adopt lor di posa serve, and Veil., who has lor poggiato serve. It is quite useless, as I have often noted before, to quote the text printed with any Com- mentary, unless there is independent evidence of the reading to be found in the Commentary itself. The alternative read- ings here differ at any rate so entirely, that they are worth recording as an indication of relationship of MSS. XXVII. in. men lontani ;-^.) totmbefgkntw 7.9.10.12. 14. 1 5 . 20. 21 26.30.33.34.37.40.42.44.46.48.52.592.63.642.66.68.71.73.76.77.78.80. COLLATION OF SELECTED PASSAGES. 427 81.84.88.89.94.99.102.104.107.108.109.110. Bat. 30. [100] Benv., Veil., Bnti, Dan. piu lontani VCFOPSUV l V 2 V*ZGLTr.ACDEFGtfMQSAnS3>Va.t TlOv^ir<}>sa*bdeh*iklpruvyrt:^A//;^ 1 r?/A' 1.3. 4.6.8. 1 1 . 1 3. 16 . 1 7. 1 8. 19.2 1 .22.23 24. 27.28.29.31.32.35.36.38.39.43.45.47.5 1. 54.55.56.57.58.59j.69.yo.72. 74. 75 . 79.82.85.90.91.92.93.95.96.97.98.100.101.103.105. 106. ii 1. 112. 113. 114. 119. [122] Lan., Anon. Fior., Serr. j has alloggian men. Gregoretti reports the Venetian Codici as 5 having men and the rest (12 or 13 I suppose) having piu. The reading men appears to me to be unquestionably pre- ferable to piii in this passage, as introducing an idea more suitable to the context, and one in every way more in ac- cordance both with experience and also with the sentiments elsewhere expressed by Dante himself. (i) It suits, as I say, the context better. For in Dante's present case his wanderings were now all but ended, and the terrestrial Paradise was close at hand, and this thought is several times besides insisted on in this Canto. See lines 36, 54, 117, and especially, as embodying the precise sentiment of this line (with 'men lontani}, lines 122, 3, ad ogni passo poi Al volo mi sentia crescer le penne. (2) The word 'tornando* seems to lose its point if the greatness of the distance from home increases the longing of the traveller. See however Lan. quoted later. (3) It is surely far more true to nature to describe the home-longing as becoming tenderer and more intense as one draws nearer home, even as it is so on first parting from home 1 lo di che han detto ai dolci amici addio ' (see the exquisite passage in Purg. viii. I, seqq.}. Gregoretti, defending the reading men (App. p. 36), remarks: ' Non e chi non abbia piu o meno esperimentato viaggiando che quanto si e meno lungi da un luogo disiato, tanto piu cresce il piacere e 1' impazienza di giungervi.' (4) I should certainly attach great importance to the fact that the idea expressed by this passage with 'men lontani' appears, from its recurrence else- where, to have been a favourite one with Dante (see on this 428 PURGATORIO. argument, Proleg. p. xxxvii). See first Par. xxxiii. 46 48. Next, in Conv. iii. 10, after stating on the authority of 'il Filosofo' the general truth "quanto 1' agente piu al paziente se unisce, tanto piu e forte pero la passione," Dante exemplifies it in these remarkable words : " Onde quanto la cosa desiderata piu s' appropinqua [i.e. e 'men lontana'] al desiderante, tanto il desiderio e maggiore." And with equal distinctness of statement, though with some obscurity in the inference drawn from it, he declares in De Mon. i. c. 13 ; " Omne diligibile tanto magis diligitur quanto propinquius est diligenti 1 ." Then again, akin to this idea is the thought in Conv. iv. 28, where Dante describes in such exquisite language the tender feelings of the evening of life, when " la nobile anima...ritorna a Dio, siccome a quello porto ond' ella si partio, quando venne a entrare nel mare di questa vita" (compare with this generally the beautiful conclusion of Conv. iv. 12). (5) We may add, in reference to what has been called ' transcriptional probability ' as between the two readings, that the original men was probably altered into piu from the superficial completeness of antithesis thus secured between piu grati and piu lontani. Scartazzini feels himself bound, against his own better judgment, to adopt piu lontani out of deference to supposed overwhelming MS. authority. It will be seen however from the above collations that the MSS. which I have examined are, numerically at any rate, not very unequally divided between the two readings. Certainly, as far as MSS. are concerned, we may feel perfectly free to exercise our own judgment here, and I can scarcely imagine any hesitation as to what it should be. Among the early Commentators, the reading men has rather more support than piu. We find it in Benv. (' rever- tuntur domum et appropinquant patriae'), Veil., Buti ('meno di lungi'), and Dan. (' manco lontani '). Piu appears in Lan., Anon. Fior., and Serr. (with somewhat similar explanations), viz. that the further one is from home the more grateful it is 1 The inference is "sed homines propinquius Monarchal sunt quam aliis principibus : ergo ab eo maxime diliguntur, vel diligi debent." COLLATION OF SELECTED PASSAGES. 429 to turn one's face homewards a sentiment, though true in itself, quite inappropriate to Dante's present case. (I think too that any one who endeavours to read over the passage with this meaning, will feel that it lays too much stress upon tornando, and too little upon albergan) Pietro, Off., Fals. Bocc., and Land, are 'indeterminate.' As to the reading of Editions, men seems almost entirely to hold the field, if one can judge from 24 which I have examined, of various dates, since only two of these (Witte and Scartazzini) have pin. The 'first four Editions' however are equally divided. It may be added that interesting interchanges of pin and men occur also in Inf. xxx. 87, and Pnrg. ix. 17. XXX. 15. alleluiando ptwa&fgnrv 9. 10.12. 14.20.2 1.33.35.42. 45*46.52. 59-64*.69. 70.71. 73. 76*. 78. 85*.98.99.io2. 104.107. 108.1 10.113. [69] alleviandoBCDEFGHIKLOPQSTUV 1 V 2 V 4 WXZGl.Tr. J ff/'GA 7 C'AA n4 >x I' 1 HptT<(xv|oirpTajx.'J'wsF 1 a 2 befgij 1 mruvxyzfrfi?^z^;;^//jr 1.4.6.7.8. 11.13. 15. 16.17.18. 19.22.23.24.26.27.28.29. 3o.3i.32.34.36.37.38.39.4o 1 .4i.4344.47.48. 53-55-56-57-58.63.66.67.68.72.74.75.77.79.8o.8i.82.84.87.88.89.9o.9i.92.93. 96. 97.100.101.103.105.109.111.112.114.119. Bat. 10. [146] o-d/ have allevando. a and 3 have allei uiando (sic). So 54 (but in one word). 1 06 has alleuinando. Ott. has allegando. VOC6 was found in very nearly 200 MSS. carne in 15 only, viz., IV^ZSb^ 8. 29.56.57. ?>7.go.gi.g?>(marg.) 113. [15] veste o. vesta 53. There can be little doubt as to the choice of readings in the last word of this line, in spite of the preponderance of MS. authority. This has less weight since the two words allelniando and alleniando would be very easily confused, and the substitution of the latter for the former may possibly have arisen from an early clerical error, though probably the somewhat unusual 'ablative absolute' construction 'La...i>occ 430 PURGATORIO. alleluiando' may have prompted the change 1 . The imitation (as' Scarf, points out) of Rev. xix. i, &c. is almost conclusive for alleluiando. There can be no doubt that the substitution of carne for voce (which is extremely rare, though the case is certainly overstated by Scart. when he says, " cerca qui cerca la non sappiamo trovare che un solo codice legga 'carne alleviando '") is an obvious conjecture to avoid the supposed harshness of the expression rivestita voce, and is an unmis- takably ' facilior lectio! Possibly too Par. xiv. 43 may have suggested the alteration Come la carne gloriosa e santa Fia rivestita. Among the Commentators, Buti is the first who seems to recognize this reading. He is followed (as is often the case) by Land, and Veil. It is strange that carne is found in the majority of editions, if one may judge from 25 which I have examined, since of these only eight have voce, as against 17 with carne. The eight are all comparatively modern. The 'first four Editions' (not included in the above) are equally divided between carne and voce. The prevalence of the false reading in the printed editions makes it more necessary than it might otherwise have appeared to contend for the correct- ness of the reading voce. Dionisi (Bland. Fun. p. 136) cites in support of the reading voce, an expression from Lucan : ' defuncto in corpore vocem.' He does not give a precise reference, but the passage is no doubt P liars, vi. 631, which runs exactly thus : Pulmonis rigidi stantes sine vulnere fibras Invenit, et vocem defuncto in corpore quaerit. It will be seen that Lucan is speaking in reference to the physical organs of voice, so that the epithet rivestita would fall in with such a use of vocem, though surely it would not need any such justification. We might perhaps cite 1. 621 1 Blanc in his Grammatik der Italienischen Sprache (p. 655 med.) notes this use of the gerund instead of the present participle as especially common in the Venetian dialect. Passages illustrative of the construction la voce alleluiando as a sort of abl. abs. will be found in Inf. xvii. 1 1 1 and xxxii. 105. COLLATION OF SELECTED PASSAGES. 431 just before, when the witch desires to resuscitate a recently dead corpse, Ut modo defuncti tepidique cadaveris ora Plena voce sonent. This incident of the resuscitation' of the dead occurring in the same context with the other passage may also possibly have helped to suggest the turn of expression adopted by Dante here. XXX 73. Guardaci was found in 172 MSS. Guardami FIESl/Clw 7.2.10.36.48*. 54. 63. 76. 87. 106.107. 109. [19] Benv., Butt, Land., Veil., Dan. Guardati DV 4 ., Suit, Land., Veil., Dan. The following minor variations were noted : bene s' io son B. o. s' io son ben B. nu 10.76*107. ben ben s' io son S\Fg 77. ben ben se son Ev. se ben sem BELkz 42.71. se ben son DKfl 9.69. se ben se B. a. ben che ben sen 73. son ben son ben U. ben semo ben sum V. ben sem ben son v 46. bene si ben sien 52. 43 2 PURGATORIO. G. bene e bene si son 33. G. ben ben son ben B. 54. G. bene ben sem B. 43.44. G. ben ben sembri B. a. G. bene sesse (? se se') B. <|>. G. ben ben se ben se B. 108. A has Guardici ben sancta B. (!). The plural forms -ci and sem are certainly to be preferred here, partly as being the less obvious forms of expression, though perfectly intelligible and easily to be accounted for ; so that the natural course of correction would certainly be to alter the plural into the singular, and not the singular into the plural partly also because Dante (as has been observed) prepares us for this quasi-royal dignity of language in 1. 70 Regalmente... continue. (Compare ' Donnescamente' in xxxiii. I35.) 1 It is also to be noted that this ' difficilior lectio' has the support, both in respect of Gnardaci and sem, but espe- cially as regards the former word (which, as Scarf, observes, almost necessarily carries with it the other), of a very large majority of the MSS. The form sem for siamo appears again in Ptirg. xxxi. 106. Noi sem qui ninfe, e nel ciel semo stelle 2 . It is fair to draw attention to the fact that all the old Commentators who notice the passage explicitly (i.e. the five mentioned above) have guardami and son without recognizing the other readings. 1 The use of the plural both in the ist and 2nd persons ''honoris causa 1 is said to have arisen when the Roman Emperors accumulated in their own persons several offices of dignity. Dante refers to this tradi- tional origin of the custom in Par. xvi. 10. In accordance therewith it is employed by himself in the Commedia only in respect of Cacciaguida (Par. 1. c.), Beatrice (Purg. xxxiii. 92), and Brunetto Latini (Inf. xv. 30, &c.). 2 Witte here reads siam... siamo, but it is against the weight of evi- dence of his MSS., A l only having siam, while A^BCD have sent, which is the reading generally found. With precisely the same conflict of evidence, he follows A 1 alone, in reading son instead of sem in the present instance. Similar cases will be found at Purg. xxvi. 75, Par. xvii. 119, xviii. 75, 131, xxv. 73, &c. &c. COLLATION OF SELECTED PASSAGES. 433 XXXI. 29. degli altri. This, which is undoubtedly the true reading, is almost universal, being found in nearly 200 MSS. out of about 215 examined. The exceptions were as follows : delle altre Z?dgknpt 1.11^21.8^.107. [12] delli altre g. degli altre w* y 2 . del arte (?) 10. The rare and inferior reading delle altre seems to have arisen either from the idea that Beatrice's reproaches were inspired by the 'spretae injuria formae' in a very literal and common-place sense (with an eye no doubt to the traditions about Gentucca, the ' pargolettal and other pretended literal amours of Dante) ; or else, regard being had to the allegorical meaning of Beatrice as representing Theology, the feminine altre was thought necessary to indicate the other and inferior Sciences. The question of the interpretation of this re- markable scene, and the meaning of the reproaches which Dante puts into the mouth of Beatrice have been so often and so fully discussed that it is sufficient to refer to such sources as Scartazzini's Abkandhmgen ; Fraticelli, Dissert, sidle Poesie Liriche ; Witte, Dante Forschungen ; Fornaciari, Studi; besides Giuliani, Ruth, Carducci, d'Ancona, &c. &c. The majority of the early Commentators have altri, viz. Ott. (degli altri desiri), Buti (beni mondani ed imperfetti), Land., Veil., Dan. (beni fallaci e mondani). We find altre however in Benv. (aliarum dominarum, idest scientiarum), Lan. (in altra scienza che Teologia), and Anon. Fior. XXXI. 51. sono in terra sparte RDORGl.AEFriodgkvxyt>*efg/imw 13.14.35. 68.69.75.76.77.79.102.119. [35] Pet., Ott., Buti, Land., Veil. sono terra sparte ACEFGHIKPQSTUWXZTr.^CZXS-i/FAAIIS** apY?Ti0K\|j4'irp\|/wsfaehijlmnprtuwz<7Crt'/^/^r/'yr 1.3.4.6.7.10.12.15 to 22. 24.26.27.29.31.32.33. 37.38. 39.42.43.44.46.48.51. 54.s6.57.58.70-7i-72.74-8i.82. 84.87.90. 101. 105. 106. 107. 1 08. 1 10. 1 1 3. [125] in terra son sparte bf 1 1*.78. D. I. 28 434 PURGATORIO. son terre e sparte n 99.104. son terra o sparte 9. son terra asparte 30. son membra sparte vx 96 (So Fals. Bocc.}. e che son tre consparte (sic) ? v. I cannot but think that the true reading is that found in comparatively few MSS. ' sono in terra sparte! Not only is the reading sono terra sparte very harsh in construction (though this is a dangerous argument, if otherwise unsupported), but it may easily have originated from the other reading by a mere blunder. Thus son in might be written soin, so in or sd 'n, and then with the omission (as often) of the superposed mark, so 'n or son; as is pointed out by Scartazzini 1 . More- over the words in terra would be naturally suggested to Dante by his familiarity with many passages in the Vulgate ; e.g. Ps. cxlv. 4 ; Eccl. iii. 20, and xii. 7 ; in all of which the expression in terrain revertere occurs. We may compare also the well-known passage in Gen. iii. 19 (cited by Scart?), in pulverem reverteris. As to the word sparte, it is used again by Dante in a very similar manner in Pttrg. xii. 33, Mirar le membra de' giganti sparte. Again we might compare Petrarch Trion. Mort. i. i, 2 (which however, so far as it goes, rather resembles the other reading here) : Questa leggiadra e gloriosa donna, Ch' e oggi nudo spirto e poca terra. The same might perhaps be said of a passage something like this in Par. xxv. 124, In terra e terra il mio corpo. I cannot find any evidence at all of the reading sono terra sparte in the early Commentators. Thus Pet., Ott., Benv. (resoluta in terram sparsa sunt), Butt*, Land, and Veil, all 1 so 'n terra in fact actually appears in Butt. (See next note.) ^ButVs comment on this passage is very singular: "Ecco secondo 1' allegoria per le belle membra in che fu rinchiusa Beatrice sono li libri e li testi de la 'santa Scrittura, li quali contegnano la santa Scrittura come le membra corporali dell' omo contegnano 1' anima....r0 'n terra; cioe, sono in terra, sparte ; cioe seconda la lettera, sepolte ; ma allegori- COLLATION OF SELECTED PASSAGES. 435 clearly recognize the reading in terra: while Lan.,Anon.Fior. and Dan. are indeterminate. It is hardly necessary to consider the wholly unsupported conjecture of Fanfani that we should read 'sono terra parte' He would take parte in the sense of ora, and claims that it is the same idiomatic Florentine usage which we have spoken of at Purg, xxi. 19 (cf. Inf. xxix. 16). Negroni however in his monograph (already referred to) contends that the word cannot bear the sense of ora at all, but that it always signifies mentre or frattanto. camente sono libri e li testi de la santa Scrittura sparti per lo mondo in diverse parti ; ancora si puo intendere del divino officio che canta la santa Chiesa, nel quale la santa Scrittura e divisa si, che ne' luoghi con- venienti sono appropriate le suoe parti " (!). 282 PARADISO. O beati quei pochi che seggono a quella mensa, ove il fane degli Angeli si mangia. (Conv. i. i.) [NOTE. It is to be observed that g 2 and n l are not to be confused with those letters having the figures subscript. They denote separate portions of certain MSS., as is explained in the description of MSS. later on.] I. 135- L' atterra ) La terra ) TNabceglqrtvsw 1.2.4.7^ [53] L' atterra La terra xzr7*.ii8.H9. [33] L' atterr a ) La terra ) tolta So-]. tolto vyir. torce ^\\kr(marg.} La terra a torta Tr. a torto N.F. a tolto et and that Dante could not attribute this downward tendency to the ' impeto primo, cioe 1' istinto naturale [che] lo attrae al cielo ed a Dio.' To this I would reply not only that Dante at any rate certainly would do so, but that he would be stating a pro- found truth in thus declaring that 'the things that should have been for their wealth become to them an occasion of falling.' It is the impeto primo which supplies the motive power in any case, though it may be deflected from its right course by l falso 1 One of them (Y), it should be noted, is a text accompanying the Commentary of Buti, and presumably (though this is far from always being the case) conformable to it. PARADISO. piacere' or (as we read in Purg. xxx. 131) ' Imagini di ben seguendo false! Dante's views on this subject are very fully expounded in a beautiful passage in Conv. iv. 12, far too long to quote, though one sentence may be cited, in which, after a fine simile, Dante says, " cosi 1' Anima nostra incon- tanente che nel nuovo e mai non fatto cammino di questa vita entra, dirizza gli occhi al termine del suo sommo Bene, e pero qualunque cosa vede, che paia avere in se alcun bene, crede cJie sia esso." Thus it is clearly the impeto primo when mis- directed that carries us astray. So again in Purg. xvii. 126 seqq. he speaks of a form of ' amor ' Che corre al ben con ordine corrotto. Ciascun confusamente un bene apprende, Nel qual si queti 1' animo, e disira : Perche di giugner lui ciascun contende. Moreover not only the general principle of corruptio optimi pessima, but even this particular application of it would be familiar to Dante, both from Aristotle and still more from his special guide in subjects of this kind, S. Thomas Aquinas. Take such passages as the following: Summa I. Q. 49 Art. 3, "Effectus causae secundae deficientis reducitur in causam primam non deficientem, quantum ad id quod habet entitatis et perfectionis : non autem quantum ad id quod habet de defectu. Sicut quidquid est motus in claudicatione causatur a virtute motiva (compare impeto primo} : sed quod est obliqui- tatis in ea non est ex virtute motiva, sed ex curvitate cruris (compare torto da falso piacere}. Et similiter quidquid est entitatis et actionis in actione mala reducitur in Deum sicut in causam, &c." Again, I. Q. 103 Art. 7, "malum semper fundatur in bono...et ideo res aliqua dicitur mala per hoc quod exit ab ordine (comp. torto again) alicujus particularis boni." Again, II. i. Q. 18 Art. i, "malum agit in virtute boni deficientis, si enim nihil esset ibi de bono neque esset ens, neque agere posset :... homo caecus actu habet virtutem pro- gressivam per quam ambulare potest, sed in quantum caret visu qui dirigit in ambulando patitur defectum in ambulando." Much else might be quoted in these and other passages to shew how futile is the objection of Scart. to the statement that "impeto primo... 1' atterra." COLLATION OF SELECTED PASSAGES. 439 It is natural to ask on what principles of 'transcriptional probability' such perfectly simple words as A terra e torto, giving moreover a perfectly simple and appropriate sense, could have become changed to a reading embodying the somewhat uncommon word atterra ; and very improbable surely is Scartazzini's explanation of the supposed genesis of this reading from his assumed original A terra e torto. He suggests that some copyist omitted ' e\ another then prefixed ( L ', and a third afterwards inserted another '/'. Far more probable would it be, I venture to think, that L 1 atterra, or as it would generally be written, Latterra, was mistaken for La terra 1 , and that the obvious deficiency of a verb (as in other cases e.g. Inf. ix. 125 &c.) caused ' e' to be supplied, thus giving La terra e torta (torto being changed to torta, if the latter did not happen to be there before, as in Purg. xix. 132; xxi. 45 ; Par. ix. 129; and many similar cases). This giving no sense, the very slight change of La terra to A terra would inevitably follow. Dante uses the word atterrare 4 or 5 times, and on one occasion in a passage so remarkably like this that it seems to strengthen the probability of the word occurring here. Come foco di nube si disserra Per dilatarsi si, che non vi cape, E fuor di sua natura, in giu s' atterra. Par. xxiii. 40 2. My apology for discussing this passage at such length is the all but universal prevalence in Editions (as above noticed) of what I deem to be the false reading. I. 141. Come a terra auiete in FGOPRZCFQT$$ye'( l i]\oubdisvya&*cg-mytvw 4.i3.i8.27.29.3i.46.54.55.56.57.58.64 1 .8o.9o.ioi. 106.109.1 10.112.113.1 14. [56] Come a terra quieta 2 in A^cu?/ 33. 64*70. 108. i r i. [10] 1 In fact the difference between L atterra and La terra would in most MSS. be simply ' indeterminate.' See Pref. Note, p. li (c) and (d). 2 Probably a mere blunder for quiete, by a sort of attraction into agreement with terra. Cf. the reading altra in Purg. xxi. 45, tanta pianta in Par. ix. 129, c., c. 440 PARADISO. Come a terra quieta il Qe(el). quieto el (=il or e 1) I 68. quieto il Y^/. Butt. quieta foco x. quiete (or quiet' e) foco 73. Come in terra quiete in ADFp 8.21. quieta in D. quieto un 47 quiete (or quiet' e) foco 2g(see below)35. quieto foco 102. Come terra quiete (or quiet' e) in Bty. quieta il j. in C\rs 30. un IT. 1 C(h)omatera (Comaterra, &c.) quiete in B*L1 (quieta) rw 14.17.24.32. Come"| matera (or materia) quiete in HNSTUXTr.TI0Kfi.vpo-vsg 2 Com V hk^z'jri .2.7. 1 1. 1 2. 1 6. 1 9.23.26.28.34.37.39.43.48.63 . 66. 67 . 75 . 79. Con J 81.91.92.93.105.118. [50] matera (or materia) quieta in AKWfSVTa*mntux&n l io.2o. 38.42.44.51.71.76.77.84.95.107.119. [28] matera (or materia) quieto in KU 59. The following are some of the more irregular or blundering variants : Come ethera quiete f. Con matura quiete A. Cura matera quete (sic) E. Chome mathera cheta in z. Como in terra quetu e focu vivu g. Come A terra che te in (sic !) 69. Come a trovar quiete in 3(atrova) 82. Choma tranquieto e foco vivo (sic) 104. Comatera quieta in foco viso (sic) 103. Amidst the very great variety of readings here most of them obviously mere blunders, and more or less unintelligible the following would seem to emerge as most probably the original reading : Come a terra quiete in foco vivo. Come a terra is preferable to Come in terra, because it accounts for the origin of Come matera, Con matera, Come terra, &c., some of which (with minor variations) appear in a great number of MSS. These may be set aside as blundering 1 This probably is meant for Com 1 a terra, and is the intermediate step to Con matera, &c. COLLATION OF SELECTED PASSAGES. 441 reproductions of Comaterra or Comatera. I think too that it means not 'on the ground/ but 'on earth/ in contrast with 'in heaven/ where Dante now was (see 1. 91), and where a different set of laws operate, or rather perhaps the same laws freed from all earthly impediments (11. 139, 140): so that if he did not rise towards God, it would be as strange as absence of motion would be in a living flame on earth. Next, as to the choice between quiete in foco and quieto foco. The latter is preferred by Scart. rather dogmatically, and he boldly asserts that the former " non da per altro verun senso," and though he alleges that his reading is found in "parecchi codici," he does not specify any of them, and the result of my examina- tion of nearly 200 MSS. would indicate that it is practically devoid of MS. authority. Not content even with this, he declares, that ' all the Commentators adhere to this reading/ which (as will be seen below) is far from being the case. The reading quiete in is at any rate that of nearly all of the MSS. here examined ; and I think it (or at least 'in foco') seems to have been the reading to which the comments of Ott. and Lan. point. The former says, " come sarebbe maraviglia videre in fuoco vivo alcuna materia stare quieta." So that he probably read ' Come matera quieta in foco vivo.' Lan. says, 'in terra foco vivo che fosse in quiete/ which may either represent quieto, or be a paraphrase by a sort of inversion of quiete in foco, which latter seems more probable, else why not have said che fosse quieto, as in fact Veil, does, who para- phrases: 'vivo... fuoco se stesse quieto a terra'? The same remarks apply to Fals. Bocc., whose language here is almost identical with that of Lan. It is evident from the comment of Benv. that he read, Come matera quieta in foco vivo, in spite of the text being actually quoted in his MSS. as Come a terra quieto foco vivo. His explanation is 'sicut esset mirum videre aliquam materiam combustibilem stare quietam sine aliqua mutatione in igne accenso ardenti.' Buti has the reading quieto il foco vivo, which has next to no MS. support (Y being only a MS. of Buti's Commentary with Text), and is perhaps suspiciously easier. Pet. and Land, are indeter- minate, and Dan. omits all reference to the line in his Com- mentary. The reading ' quiete in foco vivo ' seems to me to 442 PARADISO. have the advantage of giving a natural antithesis with in te in 1. 139, and it is also the reading of the vast majority of MSS. Further it avoids the inelegance of the double epithets quieto and vivo 1 in Scartazzini's reading ; and so far from its giving no sense, as that Commentator asserts, it would rather seem almost open to the charge of being lectio facilior. The illustration itself, which is obviously suggested by Arist. Nic. Eth. II. I. 2, Phys. II. I, and similar passages, is rather a favourite one with Dante 2 . See another passage very like this in Purg. xviii. 28. Par. iv. 77-8 is a more direct imitation of the Ethics I.e. See also Conv. iii. 3 (which however is best illustrated by Par. i. 1 1 5), and De Man. i. 1 5 (al. 17) (lines 27 and 31). Once more, see Par. xxiii. 40-42 Come fuoco fuor di sua natura in giu s' atterra. II. 42. e Dio .32.33 35-37- 38- 3942.46.48.55-56-57-64-7i.73-77-8o.82.9o.9i.99.io2.io4.io6. 107.108. 110.111.113.114.119. [Ill] a Dio AIYFe 3.23.24.68. 105. [10] in Dio (or en Dio) CEFGHLKPSTWX^AS^vpcrTg%kls 1 u^//r 1.12.14.15.20.26.43.44.66.70.75.76.78.81.84.101.109.118. [52] e dio sia unio (stc\ 0. natura onde sonio (sic) v. et idio 79. etio s' onio (!) 112. m has e idio, and lines 41 and 42 are transposed. e Dio servio 72. Come a nostra natura Iddio s' unio j. Of the three readings between which we have to choose here (omitting mere blunders, or obviously conjectural changes such as that of ' j ') viz. a Dio> in Dio and e Dio, the first is prima facie the simplest form of expression, but apart from the 1 This objection may perhaps be obviated by punctuating, as in Tommaseo's Edition : Come, a terra quieto, foco vivo. 2 Pet. here quotes an interesting parallel passage from Boethius, which does not however help us in respect of the reading of this line. COLLATION OF SELECTED PASSAGES. 443 suspicion attaching to it on this ground, (i) its MS. support is exceedingly slight. (2) Dante would surely not have spoken of 'our nature uniting itself to God.' Compare his very guarded and theologically accurate language in a similar passage in Par. vii. 30-33. (3) This reading is really far less suitable than e Dio for the purpose of the context (a matter rarely taken into account by emendators). The mystery of two Natures in One Person is referred to in illustration of the minor difficulty how Dante, being still 'in the body,' could become as it were engulfed in the body of the Moon, which was " Lticida, spessa, solida e pulita " (1. 32), this latter not- withstanding ' permanendo tinita ' (1. 36), just as water which is penetrated by a ray of light. This is much more than uniting one thing to another (a Dio}. The reading e Dio taken in connexion with the singular verb s unio, though grammatically there are two nominatives, well expresses the absolute unity (of Person) resulting from the union of the two natures, and the illustration becomes most suitable to the difficulty in question. Of course Dante must not be suspected of Eutychianism, against which he is quite sufficiently guarded by Par. vi. 14 and vii. 32 &c. Scartazzini certainly goes too far in saying that the reading e Dio is "di quasi tutti i Codici," since in Dio or en Dio has very considerable sup- port, though the phrase s 1 unire in seems scarcely intelligible. Finally the prior existence of the (superficially) ' difficilior lectio' e Dio will account for the origin of the other readings, which were probably intended to remove the grammatical anomaly of a sing, verb with two nominatives. Compare the changes due to a similar anomaly in Inf. xx. 69. The sig- nificance of this idiom here (as above indicated) may perhaps be illustrated by Eurip. P/icen. 364 6, crnovbai re KCU cri) Triortr, fj p.* I have cited several instances of Dante's employing a singular verb with more than one subject (without any such reason as would make it specially significant here) at Inf. xx. 69. 444 PARADISO. Most of the early Commentators are indeterminate ; but Buti reads a Dio, and Benv. and Veil, have e Dio. Scartazzini is wrong in assigning a Dio to Benv. His words are: " nostra natura scilicet humana e Dio s' iinio et natura divina conjunctae fuerunt'simul." Veil, reads e Dio, and paraphrases s' unio by s' uniron. Both Land, and Veil, suppose this and the following line to refer to the mystery of Transubstantiation in the consecrated Host. II. 124. a me u^sabdefg 2 hjkmoprstuwxyM^/;j.r 1.2.7. 1 1 1 2. 14. 1 5. 1 7. 1 9.20.2 1.23.24.26. 30.32.33.35.37.39.43.44.46.48.55.63.66.69.70.71.72.75.79.81.84.102.104.105. 109. [107] omai DQRYZTr.^6TAA*S>I'T)virxFgilnvztf'/'z> 3.4.6.10.13.18. 27.29.31.38.42.54.56.57.64.68.77.82.90.97.106.107.108.110.112.113. [61] Riguarda omai a me OF 16. Biguarda bene omai Q. d (marg.} has ' al. omai '. f omits bene. This is not a variant of much importance, but I think a me is to be preferred to omai which is here adopted by Witte. Scartazzini supports a me, while admitting (unnecessarily) that omai "sembra infatti piu antica," and claiming for it (inaccurately) the support of the early Commentators, Benv., Buti, Land., Veil., Dan. As to the superior antiquity of omai, I have no idea on what grounds that assertion is made. A me has the support of far the larger number of MSS., and there is no reason that I am aware of to claim greater antiquity for the members of the minority. As to the Commentators, I suspect that Scarf, here, as elsewhere, has contented himself with looking at the printed text accompanying their Com- mentaries, which (as I have before pointed out) has no authority whatever, and is often flatly inconsistent with the language of the Commentary itself. In this case Land., Veil., Dan. pass over the passage altogether. Buti certainly reads omai, and I think that the language of Ott. also rather implies COLLATION OF SELECTED PASSAGES. 445 omai ; but Benv. reads a me. In favour of a me I would suggest (with Scarf.} that omai might very well have arisen from the occurrence of the word in precisely the same position two lines above, so that it may have caught the copyist's eye, while its repetition here would be pro tanto a blemish in style. A me also enhances the antithesis with sol in 1. 126, though it is not essential to it. Moreover there seems (as will appear in the discussion of the next passage) to be a special reason why Beatrice should call emphatic attention to the course she herself adopts (a me, si com io vado) in enlightening Dante's understanding, that he may be able to walk without assistance in the same path afterwards ; that method being to appeal to the experience and analogies of human life. II. 141. in lui was found in 146 MSS. in voi Cv^d/Tbfhvwxrtk^. 39.43. 54.106. 1 19. [19] in noi 58.77. in lei Sty. in vivi 108. The vicissitudes of the reading 'in voi' (which I take to be almost certainly correct) have been curious. It has very slender MS. support, but on the other hand it is backed by 6 out of 8 of the early Commentators 1 whose text can be determined, and it reigns almost supreme in printed Editions, if I may judge from the 30 which I have examined, in which I find voi in 26, and lui in 4 only, those 4 being the " First Four Editions 2 ". It would probably be curious, if one had the materials at hand, to trace how and when this complete change came over the Editions, and how lui, starting with over- whelming MS. support as well as that of the earliest Editions, came to be so entirely displaced. Scart. also notes that the moderns are ' quasi unanimi ' for voi. 1 For other instances of this divergence between the evidence of MSS. and early Commentators, see under Inf. xxvii. 21. 2 Compare the case of Par. i. 135. 446 PARADISO. The reading voi seems to me to be required from a con- sideration of the whole context from 11. 124 to 144. First of all Beatrice bids Dante mark well the method of her pro- cedure that he may be able to walk in the same path without assistance afterwards: 11. 124 6. (This has been already explained in the last note, and also how the admonition be- comes more pointed and emphatic with the reading a me in 1. 124.) She then proceeds to enlighten him on 4 points; her method throughout being to illustrate the information conveyed by an analogy or comparison drawn from human experience. I. The movement and 'virtue' of the celestial spheres is due to the angelic powers that move them 1 . So the artistic work of the hammer comes from the artist's mind (11. 127 132). 2. The One Supreme Mind is re- flected in all the different stars 2 . So the undivided soul of man works in his various members (11. 133 138). 3. Different angelic influences form, in union with the various celestial bodies to which they give life, different resulting compounds (lit. 'alloy,' lego). Even so does life in man [form and constitute different beings] (si come vita in voi} (11. 139 141). 4. The brightness of the stars comes from the inward joy that radiates through them. So does gladness beam in the pupil of the eye (11. 142 4). Thus in each case the instruction is conveyed by an illustration or analogy from human experience. If we were to read lui for voi, this uniformity of illustration would be lost. Again, si come vita in lui would be a feeble and tautologous repetition of what is already expressed in the words cJi elf avviva. Once more, the words 'si come vita in voi' have a suitable emphasis "'even as,' or 'just as' life in you"; but 'si come vita in lui' would surely be more naturally expressed by 'come vita in lui,' i.e. 'as 1 Comp. De Mon. I. ix. (al. xi), Et quum caelum totum unico motu, scilicet primi mobilis, et unico motore, qui Deus est, regulatur in omnibus suis partibus, motibus et motoribus, &c. 2 This is a familiar idea with Dante. See C0nv., Canzone to Tratt. II. 1. I (quoted in Par. viii. 37), 'and the comment on this in c. ii. "certe intelligenze, ovvero per piu usato modo volemo dire Angeli, li quali sono alle revoluzione del cielo di Venere, siccome movitori di quello." See also Chapters v. and vi. of the same Trattato. COLLATION OF SELECTED PASSAGES. 447 life in it ' (i.e. the prezioso corpo cJi elf avvivd). We do not require ' even as ' in that case, the effect of which would only be to emphasize an otiose repetition. It is also worth while to compare the very similar expres- sion occurring in Purg. xviii. 27, Che per piacer di nuovo in voi si lega. As to the Early Commentators : Buti and Veil, alone have lui; while Lan., Ott., Anon. Fior., Benv., Land., Dan. have voi. The first three paraphrase by identical words, ' nelli animali.' Benv. has, 'sicut anima in vobis hominibus ligatur cum corpore.' Land, has ' nel corpo umano.' Dan. has con noi, which may be a paraphrase of con voi in this passage, or he may possibly have had con noi in his text, as it does occur, though extremely rarely (see above), and moreover noi and voi are very easily confused [see Pref. Note, p. liv (z)]. At any rate he clearly had not ' lui! HI. 15- men tosto EFO(marg:) V l ZABCFILST^fli\yj\rvalqr 3.4.6.8.10.16. 18.23.27.29.31.35.56.57.59.80.82.90.93.97.98.106.107.108.112.113.119. [53] men forte ADEGHKLNOPRSTUWXYTr.Z^6^7AfA^KAIIZ ^ > il*ap-y8K\|xvioirpo"ru\|/wsFabdefgg 2 hiknopstuwxyz^^/^w 1 ^Z'jr 1.2.7.11. 12.13.14.15.17.19.20.21.22.24.26.28.30.32.33.34.36 to 48.50.51.52.55.63.66. to 73.75.77.78.79.81.84.85.88.89.91.92.93.95.99. loo. 101 . 102. 103. 104. 105. 1 10. 111.114.118. Bat. 30. [157] Non vien men alle (sic) C 64. Non men vien forte I. Non venne forte alle Q. Non vien si tosti d. Tornano dei nostri occhi le popille ( !)/ This passage illustrates very well the application of the principle ' difficilior lectio ' in combination with the scarcely less fruitful and important principle that that reading is to be preferred whose prior existence would account for the others, whereas their prior existence would not account for it. If we suppose Dante to have written forte, it is impossible to under- stand how tosto can have arisen either by accident or design. 448 PARADISO. If he wrote tosto, the seemingly incomplete antithesis to debili would make the substitution of forte almost inevitable. Biagioli very well explains the relation between tosto and forte as cause and effect. "Le postille dei nostri visi vengono all' occhio per quei mezzi che ha detto poco tosto, e siccome la celerita e proporzionata alia forza, egli paragona la poca forza delte une colla non maggiore della perla in bianca fronte, accennando per la poca celerita la poca forza dalla quale essa precede Adunque Dante confronta il venir debole delle postille col tornar poco tosto della perla, perocche il tornar debole precede dalla poca forza, siccome il tornar poco tosto dalla poca forza." I cannot find tosto recognised by any of the old Commen- tators except Dan. (who comments 'men tosto, cioe piu tardamente '). Forte is adopted by Ott., Buti, Benv., Land., Veil. (the latter observing "men forte, avendo detto debili"). Anon. Fior. and Lan. are indeterminate. It will be observed that forte is the reading of a very large majority of the MSS. This conjectural emendation was therefore probably a very early one. Is it not possible that in contemporary physical treatises some illustrative passages might be found in which the bright- ness or dimness of an image was connected with the velocity of its impact on the eye? Possibly the theory of 'visual spirits' and their movements as connected with sight may throw some light upon it. (See Vita Nuova, &c.) The substitution of ckiaro for caldo in Par. iv. 12 seems to have had a similar origin as offering a better correspondence to distinto. Par. xxvi. 78 might perhaps be cited in illustration of the association of swiftness of motion with strength of light. The intensity of the glance of Beatrice's eye is described by the words Che rifulgean da piu di mille milia. In any case, the association is quite sufficiently intelligible to justify Dante in adopting the less obvious word tosto in a poetic description. COLLATION OF SELECTED PASSAGES. 449 V. 88. tacere was found in 156 MSS. piacere CIKSWjV^^CZ-nSpeeXvpvx^wstu^r 2.37.(?)8. 10.18.29.31.32. 46.55.56.57.73.90.102.104. Bat. 136.444. [47] Note. Qn*6$* have ptacere (sic}, i has merely Lo suo (with rest of line blank). The reading piacere has very respectable support, though it is not numerically very large. Intrinsic probability however points in the same direction as the large majority of MSS., viz. : in favour of tacere. Among the early Commentators, I can find no trace of piacere, while Ott., Benv., Btiti and Veil, have tacere, and so have the ' first four Editions ', and a majority of the later ones also. I can only suppose that some matter-of- fact copyist thought that the silence (tacere) of Beatrice was a paradoxical reason for the silence of Dante, to whom it would rather give the opportunity for speaking, and that he preferred to account for Dante's silence by alleging the pleasure (piacere) of Beatrice as the cause of it Nor is it likely that the words ' suo piacerj having occurred in 1. 84, would be so soon repeated. VII. 21. Punita is almost universal, having been found in nearly 190 out of about 205 MSS. examined. The exceptions were as follows : Vengiata ~DV l ZAC&(marg.} w 2 (in later ink over space apparently left blank) 18.29.35.552.56.57.90.113. [16] Among the above the following variations of order occur : Fosse vengiata V 1 29.56.57.90.113. Fosse punita IY 68.98. It will be noted that a considerable number of those containing Vengiata have 'Vatican' affinities, though the reading is not found in B itself. This also is another case of Witte's four MSS. dividing into the pairs A and C against B and D. The very large majority of MSS., Commentators and Editors read Punita here ; yet I cannot but suspect that Vengiata is the true reading. The preponderance of MSS. for D. I. 29 450 PARADISO. Punita (as is shewn above) is overwhelming. Among the early Commentators Benv., Btiti, Land., Veil., Dan. all have Punita. It seems more probable that Lan. and Anon. Fior. read Vengiata, since they paraphrase by vendicasse, though it is possible that they are thinking of the language of the original passage (vi. 92) which is here being referred to. None of them can be quoted as explicitly adopting Vengiata. This consensus of MSS., and early Commentators is not to be lightly set aside, though it must be remembered that the introduction of the reading Punita at a very early date would sufficiently account for it. (See Proleg. pp. vi., xxvi., xliii., &c.). Among the Editions consulted, I find Vengiata in two only, viz. : Witte and Scartazzini, as against Punita in 28 ; these in- cluding the ' first four Editions ' and the very early Catalan translation of Febrer (1429), which should perhaps rather rank as a MS. than an Edition. In favour of Vengiata should be noted the obvious and distinct reference, almost amounting to a quotation, of the language of Justinian in the previous Canto, 1. 92 : Poscia con Tito a far vendetta corse Delia vendetta del peccato antico; and further that when this passage is again referred to in 1. 51 we find the words, che giusta vendetta Poscia vengiata fu da giusta corte. This I am quite aware would be a most dangerous and two-edged argument if unsupported. Indeed it will be seen that I distrust a precisely similar argument in the case of Par. xi. 26. But here there seem to be two other considerations to be taken into account: (i) there is in- dependent evidence of a systematic tendency to modify and soften down strong expressions of this kind in other passages. This has been noticed and illustrated by ex- amples under Inf. xi. 90. (2) Another possible motive for the substitution of Punita here might have been the notion of some pedantic or purist transcriber that the repetition vendetta... vengiata was inelegant. It seems probable that the repetition of persona in Par. xiii. 26, 7 was the 'occasion of COLLATION OF SELECTED PASSAGES. 451 stumbling ' which led to the substitution of sustanzia in the second line (see note /. c.). Perhaps Purg. xxvii. 81 (where see note) may be another instance of this tendency. VII. 114. una ...... altra ACDEGHKLOPSTUV l V*WXZTr.ACDFGJNOQAZ * 1 I f apY?10'Fd.e{g\\mn.rsvvrxyzabcefghkmn l qtsv 4. 6.7. 10. 1 1. 13. 16. 17. 18. K) i .22.(marg.) 27.28.29.30.31.33.35.36.38.39.41.42.45.46.48.51.52.54.55.56.57.58.59.64.67.69. 71.72.73.76.77.78.79.80.82.85.90.91.94.98 . 99^ IOO. 102 to 1 10. 1 12. 1 13. 1 14. [132] Lan., Anon. Fior., Benv., Veil., Dan. tutta quanta BEFGHILNPSTWXYZ^A^n^K^po-Txshjkopt^//?- x 1.2.3.8.12.14.15.20.21.22.24.26.32.34.37.63.66.68.70.75.81.84.88.89.92.93.97. 119. [68] Oft., Butt, Land. tanta pianta bu 23.50. tanto pianto a 95. tutta pianta 49. tanto plorata (1)47. There is little doubt that tanto pianta is correct here, and not tutta quanta, as advocated by Dr Barlow. The reason he gives in its favour tells against it, viz. : that it avoids the repetition of the same word pianta, though in a different sense, in two successive rhymes. But if the sense be different, this is by no means uncommon in Dante, and there is another and a still stronger case only four or five lines above, where pahna occurs twice, and it is not even a different part of speech, but the same substantive in both cases, though in a different sense 1 . It is not unlikely that the desire to avoid such repetition may have prompted the change to quanta here. Possibly too there was a blundering sense of an impropriety in the combination 'tanto pianta' as indicated by the slightly sup- ported variants tanta pianta and tutta pianta* . Tanto plorata in ' 47 ' is a specimen of the more audacious style of emenda- 1 See this subject discussed in note on Inf. viii. 78. 2 Compare Purg. xix. 132 and xxi. 45 for other cases of such ' attrac- tion ' in respect of gender. 454 PARADISO. tion. Doubtless Dante had in his mind the well-known passage in Wisd. ii. 24 (quoted by Scartazzini) ' Invidia autem diaboli mors introivit in orbem terrarum.' Also the idea of the ruin wrought in the world by the envy of the devil may be illustrated by the allusive reference to Eve in Par. xiii. 39, ' II cui palato a tutto il mondo costa.' Com- pare too Purg. xxxii. 66. The weight of authority among the early Commentators is, as appears above, in favour of tanto pianta. X. 4. per mente o per loco was found in 160 MSS. per mente o per 1' occhio in 21 only, viz. : 1 6Tab (luchio) 8.42.53.54. 58.68.71.76.80.106.108.115.119. J 5/.55.io2. M has o per occhio. 73 has per loco e per mente. 93 has per lume per loco. g has per mente o per focu. 1 10 has Quando la mente e per luoco. The scribe off has left a blank thus : per mente o pe si gira. Matteo Ronti's Latin Translation in the Borghese Library at Rome has ' mente vel intuitu,' so that he apparently read occhio. The substitution of occhio for loco, which is found in a small number of MSS., was probably intended to supply a better antithesis to mente. In fact Scarabelli's shallow note here may be said to betray the secret. He admits the over- whelming weight of MS. authority, but says that loco 'resta sempre inesplicabile non potendo essere il contrapposto di mente ma certo /' occhio che corporalmente vede sta a perfezione con mente, ch' e il mezzo onde vede lo spirito.' But the appropriateness of the thus acknowledged facilior lectio is merely superficial, and disappears on a closer exami- nation of the passage, for ' si gira ' suits loco much better than occhio. The ' eye ' is not the sphere in which the objects of the external world in any sense move or 'revolve' (si gira}. They revolve in space just as the objects of thought may be said to revolve in the mind. It is not therefore a question of the organ or instrument of perception, bodily or mental, but of the sphere of existence of the objects of sense or of thought. COLLATION OF SELECTED PASSAGES. 455 The antithesis is the familiar one between ra jevrjra (loco) and ra vorjra (mente), and the distinction thus briefly indicated is the same as is described with more pomp by Wordsworth in the lines : 'Whose dwelling is the light of setting suns, And the round ocean, and the living air, And the blue sky, and in the mind of man.' The change is a very slight one from loco or locho to /' occhio, if we recollect the common neglect of double con- sonants and of the divisions of words. These two words are in fact again interchanged at Inf. xviii. no. As to the Com- mentators: loco is found in Pet. (localiter et mente tenus), Ott. (cioe intellettualmente o localmente), Benv. and Serr. The last-named indeed is made to quote the text as per mente o per occhio, but his comment being "quasi dicat, quicquid potest cogitari mentaliter, moveri localiter in universe" he evidently read loco. On the other hand, occhio is the reading of Buti, Land., Veil, (who paraphrases si gira rather freely by si comprende], and Dan., who has the unusual reading si mira instead of si gira, and naturally therefore reads occhio. The others are indeterminate. In no case, I believe, does any of them notice the existence of any alternative reading. It is curious to note that loco, after appearing in the ' first four Editions ', and in the Ed. Nidob., and the early translation of Febrer, seems to have been almost entirely displaced by occhio in later Editions, if I may judge from my having found it in two only ( Witte and Scart.) out of twenty-five such that I have consulted here. X. 112. v' e 1' alta mente u' si A.Gl l ACFQTA*^r\sidfyfv 13.35.39.40.64.67.73. 78.80.100.108.109. [28] nell' alta mente un si CFGHNPRSUVWZ^narg.JTr.DmjKMN O /FA 1 (?)II2 1 I'iiapSeKXvioirp4/wsf gehiklmnprtuvwxz^//t7;;/r.y/.r 1.2.4.6. 7*.8. 10. 1 1. 14. 1 5. i6.i7.20*.2 1. 23.24.27. 28.30.32. 33. 36.37. 38. 42.43.46.48.5 1. 52. 54.55.58.59.66.69.70.71.72.75.76.77.79.82.84.85.92.93.97.98.99. 103. 104. 105. 107.111.112.118.119. [131] nell' altra mente un si IY. v' e T altra mente un si 106. 456 PARADISO. nell' alta mente u' si BK-yboc 47. nell' alta mente in si D 19. nell' alta mente si prof, a 1 10. nella mente un si -S* 3. un 102. v' e la mente u' si O. per T altra mente u' si j. nella mia mente un si ELA 12.22.81. nella nea (?) mente un si TWXji.o-26.88.89. nell' alta luce un si QV^Gl^Z-xd (marg.} q 29.31.56.57.1 13. u' si B. Entro nella cui mente si prof. T. Dentro dell' alta mente un si / Dentro nella mia mente un si 63. Tanto nell' alta mente u si n l (with entro in marg.}. Nella sua alta mente un si 68. nell' alta mente fu si profondo Saver fu, &c. v. Entrove 1' altra mente o si prefonda (sic) 95. Entrove 1' alta mente ove si 114. y has ne 1' alta. 22 has 'sua' in marg. Q and Z have l al. mente' in marg. There is thus an extraordinary number of variants in this line, most of them however obviously arising from the blunders or conjectures of copyists. Two only deserve discussion : (i) Entro nelt alta mente un si, which is that of the overwhelming majority of MSS., and (2) Entro v' e /' alta mente u' si which is almost certainly the true reading. The change is a very slight one, since v or u and n are often almost indistinguish- able, so that vel' alta would easily pass into nelt alta and then the further slight change of u to un would inevitably follow. Further (as Scarf, suggests), looking 3 lines back (which copyists seldom did), it becomes almost a necessity, after mentioning the fifth light, to describe (as the reading v e r alta..M si does) whom it enveloped. This is just what is done in the cases of those which precede and those which fol- low. The other reading breaks the symmetry which prevails through the whole context from 1. 103 to 1. 136, and this passage would stand alone, as containing no statement of the person to whom reference is made. Of course it is possible that the confusion of u and tin, or the supposition that u was a clerical error for un, may have been the first blunder, but since we COLLATION OF SELECTED PASSAGES. 457 find the intermediate reading neW alta mente u' si in a few MSS., the former supposition seems more probable. Buti adopts neir alta mente un si, and Pietrds words imply z/ e r alta, but the Commentators generally are indeterminate here. X. 119. tempi 2.6.7. 10. 1 1. 13 to 20.23.28.29.30.31.32.36.38.39.40.42 to 48.51.52.54.55.56.57.59.64.67.68. 69.70.72.73.76.77.78.79.80.85.90.91.92.95.97.99.102 to 114.118.119. Bat. 30. [158] Ott., Benv., Buti, Land., Veil, Dan. templi FGHLTWBCEGO&Vi\\>.p nat0K|j.|irptrTv\}/5/'ajlpO;/^;;i,'".)ruv\vxyz^^j / (' ) r //;;/;/V/7/.r 4. 6. 7. 1 1 . 1 2. 1 5. 1 6. 17.20.21.22.26.28-32.33.35.38.39.40.47.48.50.51.54.58.63.64.66.70.77.78.79.80. 81.82.88.89.92.95.99.100.104.105.106.108.109.1 10. 1 1 i.i 1 2.1 14. 1 1 8. [117] Ott. 14 has d' ingengni (? = di sengni). I*! apparently had sengni. 468 PARADISO. There is very strong MS. support for either scgni or sangne here, though there is a considerable numerical preponderance for the former. The early Commentators however (with the single exception of the Ott.) are all for sangue, and further they all (except Dan.) explain sangue as referring to the death of Christ, and martiri of the sufferings of the martyrs. The language of Benv. may be taken as a sample : " sancto sanguine Christi et aliorum sanctorum martyrum." (Dan. appears to take di sangue e di martiri as a sort of Hendiadys thus cioe, fu fondata sopra il sangue de' martiri). The in- terpretation thus generally accepted would redeem the reading di sangue e di martiri from the prima facie appearance of tautology, which most likely suggested the change to segni 1 . Another possible answer to this censure is suggested by Scarf., from the familiar combination ' death and passion ' of Christ. Looking at the whole context, we shall, I think, judge sangue to be much the more appropriate. Dante is denouncing the deadness of feeling and heartlessness of those who employ that spiritual edifice which it cost so much to build up nothing less than the death of Christ and the sufferings of so many martyrs merely for the purpose of avarice and traffic. The reproach is conceived in the spirit of Lam. i. 12 : "Is it nothing to you, all ye that pass by? (only that deliberate trafficking is worse than merely ' passing by ') behold and see if there be any sorrow like unto my sorrow." Compare also perhaps the appeal of I Pet. i. 18, 19: "Foras- much as ye know that ye were not redeemed with corruptible things, as silver and gold, but with the precious blood of Christ 2 ." If this be the point, the reference to miracles seems out of place, and certainly would much weaken the sting of 1 This seems to be confessed at any rate as the reason for adopting segni by the Edd. of the della Crusca. They state that they wished ' fuggire quasi un sinonimo.' (See Scart. h. 1.) It may be noted that in early MSS. the words might perhaps be accidentally confused, especially with the common archaic spelling settgni. 2 This is well put by Anon. Fior. "gli pastori che la impiagano e mischiano colli seculari esercizi, se avessono rispetto al costo de 1 suoi muri et a quello pensassono, egli non imbrigherebbono lo seculare reggimento." COLLATION OF SELECTED PASSAGES. 469 such a reproach. It is to be noticed that we have a precisely similar appeal (reading sangue) in Par. xxvii. 40 45 : Non fu la sposa di Cristo allevata, Del sangue mio, di Lin, di quel di Cleto, Per esse ad acquisto d' oro usata; Ma per acquisto d' esto viver lieto E Sisto e Pio e Calisto ed Urbano Sparser lo sangue dopo molto fleto. See further lines 58 60. Here we have distinctly both sangue and martiri, and again set in contrast with the avarice of later times. The following further points may be borrowed from Scartazzini's note: He urges (i) that segni is not used by Dante elsewhere in the sense of miracles ; (2) that there seems an obvious reference to Acts xx. 28, 'the Church of God which he hath purchased with his own blood;' (3) that the building up of the Church by miracles is a " concetto, ne biblico, ne patristico, ne scolastico." In reference to the in- terpretations above given of the older Commentators, Scarf. objects that the Church is never said to have been built up by the sufferings of the martyrs, and would take both sangue and martiri as referring to Christ Himself, the latter word corres- ponding to 'passion' or 'sufferings 1 .' But it is surely pedantic to require ' precedents ' for an expression so obviously true, as that the Christian Church, which was then profaned by the avarice of its rulers, owed its actual growth to the sufferings of Christ and a long succession of martyrs. Buti very well paraphrases thus: "intende del tempio spirituale, come della santa Chiesa, che e fondata nel suo corpo et murata col suo sangue e dei santi martiri, et edificato et fatto grande." It is true of course that " other foundation can no man lay than that is laid, which is Jesus Christ" (i Cor. iii. u), but it is Scriptural also to allow that Apostles and Martyrs might " fill up that which is behind of the afflictions of Christ for His body's sake which is the Church " (Col. i. 24). Moreover the passage from Par. xxvii. 40 seqq,, quoted above, embodies precisely the idea which is thus impugned. 1 Serravalle comments thus : sanguine, scil. Christi et suorum mar- tyrum, et martyriis. 470 PARADISO. Those who prefer segni may possibly find some support in the language of Conv. iii. 7: "conciosiacosache principalissimo fondamento della Fede nostra siano i miracoli fatti per Colui che fu crocifisso, &c.," or in the interesting argument for their necessity given in Par. xxiv. 106 108. The history of this passage in Editions is curious. I find sangue in Febrers translation, Edd. Mant. and Aid., and three sixteenth century Editions ; but in no Edition that I have examined later than 1575. No less than twenty-four of the thirty Editions before referred to have segni. In fact since the Cruse, introduced segni, it seems to have been universally accepted. Scart. prefers sangue, though retaining the generally received segni in his text. XVIII. 131. Paolo, Paol, or Paulo is practically universal, being found in more than 180 MSS. as against 4 containing Polo, viz.: AFz 101. The very fact of the reading ' Polo n in this line being all but non-existent among the vast number of MSS., whereas had it been original the recurrence of the form in 1. 136 would have protected it from alteration, or at least from all but universal alteration, seems quite fatal to it. The general tendency is to the assimilation of forms, as we have often seen, and this has been probably the principle operating here with those who read Polo. Moreover the fine irony implied by the adoption of the regular forms Pietro and Paolo (or Paulo) in this line, in Dante's own mouth, and the substitu- tion of the familiar and colloquial Polo side by side with the rather common-place Pescator for Pietro, in the mouth of his imaginary respondent in 1. 136, fully explains the purpose of maintaining the difference of forms. Besides, I think it might be added that there is a marked flippancy and want of respect in the reference to John the Baptist also, both in the terms of that reference he who by dancing was led to martyrdom and still more in its purport, since the Pope's alleged exclusive 1 A form familiar to us in the name of the great Venetian traveller Marco Polo, i.e. Marcus Paulus. COLLATION OF SELECTED PASSAGES. and absorbing devotion to that Saint merely stands for love of the money (florins) which bore his image. (Comp. the reference in Par. ix. 130.) See by all means Scartazzini's interesting note. Such a 'finezza sublime d 1 ironia ' as this would never have been so generally appreciated as to cause Polo to be almost universally changed to Paolo or Paulo, had Polo been original. Witte adopts Polo, again following (unfortunately as I think) A^ against A v B, C and D. (See note on Par. xviii. 75.) XIX. 141. male a visto was found in 118 MSS. (in some cases, however, subject to the doubt explained below). Buti. male aggiustd in C^?) 19.69.72.75. Lan., Anon. Fior., Benv., Veil., Dan. mal a viso a 20. male advisto (adiustb ?) BY. male a iusto (sic) I. Che molto ha vista (sic) il comune di Vinegia (!) 14. This passage has already been mentioned in the Prefatory Note p. Iv. (which may be referred to for other similar cases) as one in which a combination of several of the common causes of ambiguity in MSS. leaves the rival readings almost indeterminate. Aggiustb would often be written aiusto, and this is scarcely distinguishable from avisto, or a visto. No doubt male a visto (i.e. 'has seen to his sorrow') is correct, and the other reading ' agginstb ' arose either accidentally, in the manner above suggested, or more probably from a misunder- standing of the idiomatic use of male, which has led to such numerous alterations in other places, as in Inf. ix. 54 (where see note, with a list of such passages). The certainly inferior reading (as I take it) aggiustb has most support among the early Commentators, Buti alone (of those who are explicit) having a visto, which he explains, ' cioe, mal per lui a veduto.' 472 PARADISO. XX. 117. giuoco (or gioco) was found in 139 MSS. Also in Lan., Benv., Land., Veil., Dan. loco Gy r A*U^t^-jrr^^Ffh)kmprvc^gr 1.2.4.6.13.18.19.20.21.24.27.29.32.33.35.39.47.56.57.63.68.70.80.84.90.113. 1 1 8. [65] fatto ritrorso DSo/ 16.104. fatto 6 (or fatt' 6) ritrorso Zr\ 43.75. volto a ritrorso Ay. XXII. 95. il mar fuggir Tra~nJ/ci>sdefghijlmn*prvwx b^fgikn^qrtvx 6.7. 10. 1 1. 1 2. 1 3. 1 5. 1 9.20.2 1 . 24 . 29 . 90.99.102.107.112.113.118. [112] COLLATION OF SELECTED PASSAGES. 475 il mal fuggir CSAVtQ&xa.bktuafyeMm 1.2.3.4.14.16.17.18.27.30.37.39. 48.78.79. (05. 108.1 1 1. [39] al mar |x. Piu fe' il mar fuggir poi idio volse o. Piu fe' il mar fuggir quando A/: quando volse q. Piuttosto al mar fuggir E. This is an instance of a short-sighted alteration, such as we have often noticed, made to complete the construction within a line, whereas it originally extended over three. The state- ment is very simple: 'The turning back of Jordan and the drying up of the Red Sea at the will of God were more startling miracles than a divine intervention to remedy these present ills would be.' As Benv. very clearly puts it "quod miraculosius fuit Jordanem converti retrorsum et mare rubrum aperiri per medium, quam si Deus succurreret et provideret istis malis." In other words, however bad things have become, " the Lord's arm is not shortened that it cannot save" (Is. lix. i). A close parallel might be found in Is. li. 9, 10: "Awake, awake, put on strength, O arm of the Lord ; awake as in the ancient days, in the generations of old... Art thou not it which hath dried up the sea, the waters of the great deep ?" &c. This and other familiar scriptural references might well be in Dante's mind here. The sense is not very different if ( e' be omitted in 1. 95, provided of course that volto ritrorso, and not volf ritrorso, be read in 1. 94, which is all that I am insisting upon as vital to the primary as contrasted with the secondary reading of the text. The only difference is that the subject of the sentence then becomes Giordan volto ritrorso fuggire il mar instead of merely Giordan volto ritrorso, and one miracle only instead of two will be referred to. This is Vellutello's expla- nation, but the construction is a harsh and awkward one, and I have no doubt that ' e ' is to be retained in 1. 95 before ' il mar.' Looking at the so-called ' transcriptional probability ' of the passage: (i) the insertion of e in 1. 94 was most probable, I should even say that it was inevitable that it should be inserted in some MSS., since we have seen over and over again the tendency to round off a construction within a single line, and 1. 94, in what I venture to call its genuine form, 'Veramente Giordan volto ritrorso,' seemed to want a PARADISO. main verb to complete the sentence : (2) the loss of 'e' in 1. 95 in many MSS. would be equally inevitable, for (i) those who had inserted ' e ' in 1. 94 would have to get rid of this word ; (ii) apart from this, some MSS. (e.g. the Bodleian L) habitually omit both 'e' and f e\ however necessary they may be to the sense; (iii) many others (e.g. the Bodleian T) regularly write el both for zY, and also for e il, so that such MSS., when intend- ing 'e il\ would be liable to be misrepresented by //in the copy. Thus, granting the reading I am advocating to be original, the aberrations are very easily accounted for. This reading has the support of Lan., Anon. Fior. and Benv. (see above). The read- ing volt' e rilrorso is adopted by Ott., Butt, Land, and Dan. It should be noted that Buti and Dan. suppose that the line merely amounts to a metaphorical way of stating that all has gone wrong (see the previous line), "the times are out of joint'' or, as in the similar metaphor, ' avappeova-iv at Trayai.' This may indeed be the meaning of Ott. and Land, also, but it does not explicitly appear. In any case a metaphorical and quasi- proverbial reference to the Jordan side by side with a literal reference to the similar miracle in connexion with the Red Sea is very unlikely and would be extremely inartistic 1 . I venture also to think that the obvious reminiscence of Ps. cxiii. 5 (' Quid est tibi, mare, quod fugisti, et tu, Jordanis, quia conversus es retrorsum ? ') excludes this disjointed ex- planation of the passage, and moreover strongly supports the reading which I have advocated, by which the parallelism of the two cases is preserved in Dante's sentence, as in the original passage. In the next line, mat is found instead of war in a con- siderable number of MSS., but it is so common-place and pointless that one is surprised at its comparative frequency. A similar substitution of mat for mar is found at any rate in one of the MSS. collated in Inf. xxx. 19, and conversely mar for mat in Inf. xxvii. 47. For the common interchange of / and r in other words, and especially in Tuscan MSS., see note and illustrations at Inf. xxv. 1 12. 1 A somewhat similar difficulty arises in reference to the explanations sometimes given of two well-known passages in Ar. Poet. c. iv. 13, and c. xxii. 6, in respect of the words ffjMmrywMargii and ^trpov respectively. COLLATION OF SELECTED PASSAGES. 477 XXIII. 115. interna EGHINPQRTUWXYTr.^ n-pcrrabfghimnvxyz&^^/^wVTAr 3.7.10. 11.12.13.15.20.22.30.33.38.42.43.66. 68.70.75.77.78.80.82.84.99.107.108.109.110.114.118.119. [88] Lan., Anon. Fior., Off., Benv., Suit, Land., Veil., Dan., Serr. eterna B CD FO SZA BCDFyPnZ&VpWX&xfysdeiklprstuwafyhlmgr 1.2.4.6.14.16.17.18.19.21.24.27.29.31.32.35.36.37.39.46.47.48.54. 55. 56. 57. 63. 69.7 1.79.85.90. 102. 105. 106. 1 1 1. 1 1 2. 1 1 3. [85] A has 1' enterna. /* have 1' intima. 64 has T intera. A has Vedea sopra di noi lanterna viva (!) The difference between eterna and interna (i.e. iterna or iterna) is very slight, and several cases occur of the inter- change of these two words (e.g. Par. xvii. 9, &c., and even in Inf. iii. 2 (!) auct. Witte Prol. p. Ix). There can be no doubt that interna is correct. It has the support of all the old Commentators (see above), who have no difficulty in giving a clear and consistent explanation of it. Benv., for instance, explains : " idest concavitatem ; nam pars interior sperae appeilatur concavum, pars vero exterior appellatur convexum." It thus adds a precise and very definite matter of fact to the description, whereas eterna is a mere common-place ' epitJieton ornans' as much as jd\a \evicov, or 'heaven's eternal arches' of the modern hymnologist. It need hardly be said that the former is exactly in Dante's manner and the latter is not. No student of Dante need be reminded of the precision and definiteness of Dante's local descriptions in such passages as Inf. xv. 10, xviii. 10, xix. 16, xxx. 86, xxxi. 59; Pnrg. xii. 100 8, xiii. 6, &c., &c. (For this characteristic of Dante see Ruskin, Modern Painters, vol. iii. c. 14, especially the extract printed by Longfellow vol. ii. p. 478, &c. ; also Macaulay's Essays cited ib. p. 470, &c.) XXV. 3. pill anni BDFGINV'Z^CTTcgef/; 1 8.13.16.17.18.19.22.27.28.29.30.31. 32.35.36.37.38.39.41.45.54.55.56.57.68.73.76.77.78.79.80.82.85.91.93.95.97.98. 102.103.104.105.106.108 to 113.115.119. Bat. 129.440. [71] Oft., Biiti, Veil., Dan. PARADISO. molti anni ACEHTWZWgpofabdghijklmnp 10.11.12.14.15.20.21.23. 24.26.33.34.40.42.43.44.46.47.48.51.52.59.63.64.66.67.69.70.71.81.84.88.89.90. 92.94.99.100.107.114.118. Bat. 10.30.55. [68] Ben-v., Land., Serr. le piti volte macro B. per voltarmi macro 75. E che molti anni m' ha facto star magro j. This passage is recorded because of the distinct separation of the two readings in form, though they are identical in meaning, so that one or the other must have arisen from a deliberate alteration of the text. The MSS. are as nearly as possible evenly divided, so far as my collations go, but as this passage was added comparatively lately to my list, a smaller number of MSS. than usual has been examined here. It is difficult to decide between these alternative readings, but perhaps on the whole piu is rather more likely to have been altered into molti than vice versa. The same variants piu anni and molti anni occur also at Inf. xxxiii. 137. Benv. quaintly justifies the sentiment of the text thus : " Nee mireris, lector, si auctor diu laboravit, et si labore macruit in hoc opere altissimo componendo, quia mihi simile accidit in ipso exponendo." XXV. 14. spera was found in 161 MSS. schiera DOSUV^ (mar^.)BFLIL^l^ud(mar^.){kryr- 1.16.18.19.24. 27,28.29.31.37.39.56.57.70.78.90.95.103.1 13. [41] A and 47 have Di quella onde uscl (omitting the word altogether). 8 has Di auellopera (sic). The reading spera, in preference to the common-place schiera, which latter is found in the Vatican family of MSS. and others, seems to be conclusively established as the right reading, as well by the overwhelming evidence of MSS., as by the reason advanced by Scart., viz. the use of the word in precisely this sense by Dante twice in the last Canto, lines II and 3O 1 , to which no doubt this passage refers, 1 In xxiv. 30, Giuliani, that consistent advocate of ' easier readings ' and careless conjectures, boldly substitutes schiera, though I am not aware that any authority whatever is, or can be, cited for it in that COLLATION OF SELECTED PASSAGES. 479 though it is such a reference as might well be overlooked by copyists 1 . Apart from this, the meaning of spera is not obvious, and hence it would be natural to substitute the safe and easy word sckiera. The preponderant authority of the early Commentators, viz. Lan., Anon. Fior., Benv., Buti,Land. y Serr., is also in favour of spera. In the Ott., both readings apparently occur in the MSS. of the Commentary. Dan. does not refer to either word. Only Veil., as far as I can see, adopts sckiera. Benv. explains, 'scilicet de collegio vel circulo Apostolorum ; ' Butt ' cioe di quel tondo, ovvero cerchio ; ' Lan. ' Nota che li santi in la predetta congregazione pone essere in ispere, ciascuna condizione per se, cioe apostoli per se, martiri per se, confessori per se, ect, e perche San Jacomo fu apostolo, per6 era in quella spera, nella quale prima era San Piero.' It is more necessary to insist upon retaining spera, because nearly all the Editors have banished it from the text. I find scJiiera in twenty-two out of my thirty Editions, for though spera occurs in three of the ' first four Editions ' and in the ancient Catalan translation of Febrer, I do not find it again till we come to the modern Editions of Foscolo, Witte, Scartazzini and Camerini. XXV. 29. larghezza ABDEFHINPQRUWX CDEFGHIJLMQSVTLi&vm irpvx/'abde 1 fghijlmnpuxz^*^^ftapeX|Aa-T4uose 2 kortvvvyrt/b7; ^1*3.4.6.12.14.16.17.18.22.27.28.29.30.31.32.36.37.39.41.45.46.47.48.54.55. 56.57.68.70.76.77.79.81.88.89.90.102.104.105.106.108.113.119. [88] The difference of reading is again interesting here on the ground that one reading (larghezza} has a precise and distinct reference, and the other (allegrezza] is altogether vague and passage. See other instances of this habit of Guiliani criticised in my notes on Inf. xvii. 95 (p. 317) and Purg. xxv. 31 (p. 420). 1 When such a reference is very obvious, the argument would often point the other way ; e.g. perhaps sperent instead of sperino in xxv. 73, the former being perhaps suggested by 1. 98. (See Scarf., note on 1. 73) ; or again, surse instead of nacque in Par. xi. 26 (see my note on the passage). 480 PARADISO. general. At the same time the appropriateness of that refer- ence is not superficially obvious, and consequently here, as often elsewhere, a safe and colourless word has been sub- stituted. Among the old Commentators, I can find no recog- nition of allegrezza except in Butt, who mentions both readings, and apparently indicates some preference for allegrezza by describing larghezza as found in 'altro testo,' though he supports it by the citation of James i. 17. Lan., Anon, Fior., Ott., Benv., Land., Veil., Dan. and Serr. all have larghezza, and justify it by various citations from the Ep. of S. James. The most obvious passages are of course James i. 5 and 17, but Lan. and Anon. Fior. (identical here) add iv. 3, and Dan. quotes also ii. 13 thus: "la gran liberalita e cortesia della misericordia di Dio, della quale il medesimo nelle sue Epistole (sic), 'Superexaitat autem misericordia judicium.'" He adds that the frequent allusions to Hope in S. James would also justify 'larghezza', "come si suol dire volgarmente, 'Tu mi hai con questa speranza slargato il cuore'," and he quotes especially in illustration, James i. 2 4, v. 7, 8 1 . Benv., after paraphrasing larghezza by ' ampla largitas vel amplitudo largitatis,' justifies it by quoting ' Omne datum optimum' &c. (James i. 17), to which he curiously adds, 'et hilarem datorem diligit Deus/ forgetting that this is not from S. James but from S. Paul (2 Cor. ix. 7). The interesting point is the general recognition on various grounds of the appropriateness of largliezza in reference to the writings of S. James. There is not a single passage in the Epistle that refers to the joys of heaven (allegrezza}. It ought not to be forgotten that at the beginning of the De Mon. (I. \.fin.} Dante himself quotes James i. 5, "Arduum opus et ultra vires aggredior non tarn de propria virtute confidens, quam de lumine Largitoris illius, ' qui dat omnibus affluenter, et non improperat *'." 1 This is quoted by Dante also in an interesting passage in Conv. iv. 2. 2 We might perhaps quote Conv. ii. 1 1 as an illustration of the asso- ciation of larghezza with basilica (' the courts of heaven,' comp. 11. 42, 43) here. Dante, protesting against the vulgar notion thai larghezza and cortesia are identical, whereas ' larghezza e una speziale e non generale COLLATION OF SELECTED PASSAGES. 481 It will be observed that Dante (see lines 18, 33, &c.) com- mits the error of attributing the General Epistle to S. James the Greater. So also does Brunette Latini, for when giving a separate description and short biography of each of the 12 Apostles, he says nothing about the Epistle in reference to James the Less, but of James 'the brother of John,' he says, "Elli scrisse le epistole (compare the language of Daniello quoted above) alii xii lignaggi, che sono nella diversita del mondo." This evidently refers to James i. I. Thirty editions which I have examined are equally divided between larghezza and allegrezza. The 'first four' however have allegrezza. XXV. 33. Quante BE H PR STU WXYZ GHIUT ApYKXixvoirpo-x / 1 J/2/.r 2. 7. 1 1. 12. 1 3. 1 5. 20.22. 26. 29. 30.37. 38.42. 43.44.46.56. 57.63.64.66.67.69. 70.77.8 1 .85*.88.89.9o.99. 107. 109. 1 1 3. [85] Quanto ACD FGI N OQTr.^ BCDEFLPQS FA2*fleTv S bdefhklpr tuvwxybidehlmqr 1.3.4.6.10.14.16.17.18.19.21.23.27.28.31.32.33.35.36.39.41. 45.47.48.54.55.68.71.75.78.79.82.84.94.102.105.106.108.112.119. [96] Quando n\|/ 104. 9 has ad te (cf. the variants in xxxiii. 128). The reading Quante is certainly to be preferred here, being almost necessitated by the preceding tante (as Scarf, points out). It is supported by Benv., Buti, Land., Veil., Dan. On the other hand, Ott. and Lan. (with whom as often Anon. Fior. is verbatim identical) support Quanto, explaining, " che tu la figuri cotanto piu chiara che nullo altro, quanto Cristo si mostr6 piu chiaro ai tre." This is very lame both as regards tante volte and Quanto. We may compare the confusion of Quante and Quanto in Inf. xxvi. 25, where a somewhat longer construc- tion occasioned the same alteration. Scarf, seems needlessly troubled about deficiency of MS. support for Quante. The two readings enjoy in fact a nearly equal degree of MS. cortesiaj notes that the latter is appropriately derived from ' CorteJ ' perocch nelle Cord anticamente le virtudi e li belli costumi s' usa- vano;' adding with characteristic bitterness, ' siccome oggi s' usa il contrario.' P. I. 3 1 482 PARADISO. authority, numerically considered. The large majority of modern editors read Quante. Witte is an exception, since all his four MSS. have Quanta. In regard to carezza and chiarezza at the end of the line, I have not thought it worth while to reproduce the result of my collations in detail (93 for carezza and 89 for chiarezza) since the words are very similar, and easily confused (e.g. Inf. xv. 83, Purg. \. 75, Par. xxii. 146, &c.), and it is not easy to decide on critical grounds between them here. Among Com- mentators, Lan., Anon. Fior., Ott., Butt, Veil., Dan. are for chiarezza (generally influenced as it seems by the reference to the Transfiguration as one of the ' tante volte '), and Benv., Land, and Serr. for carezza, which they paraphrase by fami- liaritatem or familiarita. Dante's reference to this selection of the three Apostles in Conv. ii. I is perhaps slightly in favour of carezza. He is illustrating moral interpretation (as contrasted with the literal, allegorical and anagogical) and he refers to the Transfiguration, when Christ " delli dodici Apostoli ne meno seco li Tre: in che moralmente si puo intendere, che alle secretissime cose noi dovemo avere poca compagnia." This seems to suggest ' carezza^ or 'familiarita as Land, paraphrases it. XXVI. 96. Non la dico .7. 10. 1 1. 12. 13. 15.17.19.20.21.26.30.33.34.35.37.38.40.42.43.44.46.54.59.63.64.66.67.68.69.70. 7i -73- 75 -?6. 77 -?8. 80. 8 1. 84.85.90.98. 103. 104. 106.107. 108.109. 1 10. 1 1 8. 1 19. Bat. 146.362. [135] Ott., Benv., Butt, Veil., Dan. la ti dico CV^Zn^seixsekrto/zz'/;^ 1.4.6.14.16.18.22.27.28.29.31. 32.39.4i.4748.55i.56.57.79.9i.97.io5.il 1. 1 12.113. Bat. 129. [50] tosto la dico S 24. non la 'ndico 82. neladico (sic) ux. non la ridico 3. nonlo dico 8.102. Non la dico is certainly the correct reading here. La ti dico is not only less suitable to the line itself, since when a person's wants are known (note ' tu vedi mia voglia') it is COLLATION OF SELECTED PASSAGES. 483 obviously not speaking (la ti dico) but refraining from speaking (non la dico} which secures their speediest satisfac- tion (per udirti tostd). But if the shortsighted copyist who introduced la ti dico had looked a little further on, he would have found these words absolutely inconsistent with the context. Adam, after explaining that he can read Dante's unexpressed desires (1. 103), and how it is that he can do so (1. 106), proceeds to enumerate four of them, viz. to know (i) how long it is since the Creation of Man (1. 109) ; (2) how long from the Creation of Man to the Fall (1. 112); (3) the precise nature of Adam's sin (1. 113); (4) what was the language spoken by Adam (1. 114) and then to satisfy them successively (lines 115 to 142). But as no hint has been given of any one of them by Dante, la ti dico would be altogether untrue, and non la dico precisely accurate. There seems to be no trace of la ti dico in any of the early Commentators. The differences of reading in Inf. ii. 81 (q.v.} involve somewhat similar con- siderations. XXVI. 104. Da te 17.18.19.21.22.23.24.27.28.31.32.36.42.45.47.49.53.55*. 59.79.91.93. 94. 95. 97. 98.102.105.107.109.111.114.116.117. Bat. 55.65.114.129.416.440.444. [77] Benv., Veil., Serr. Dante BDEFGINPQRTUV 1 V 2 WXYZGl.Tr.^C 1 J FZA^(9 J P25FrA ASflapK\[ivopz 2 V?y 8.10.11.13.17.21.30.32.33.36.38.41.42.43.44.46.54. 64.67.69. 70.7 1. 73. 75.76.85*. 90. 94. 99. 102. 103*. 104. 106.107. no. 111.116. 118.119. Bat. 302.146.429.440.444. [79] Y s' appellava v i.5i.59*.8o. [As_y is often written for /,this reading is probably the same as above.] Ja s' appellava O. UU.ll On in. 28*. Una 9. U v 39.92.101 V(?) 23- A un (?) 1 08. Gik 1 1 5. Bat. 428, COLLATION OF SELECTED PASSAGES. 487 1 appellava DA. E 1- v. L Z?. El U. E appellava C. Bat. 10. .1. (sic) se chiamava 73.109. S' appellava (with no sign of omission) at^u^ 4.14.114, but b and b have I (znda mamt) in i/iarif. S' appella (with no sign of omission) a. Several of the above MSS. have.J. (sic), for which see notes below, e.g. Gl.xj/F/^ u. 13. 17. 30.32. 33. 38.42. 44.46.67. 73.76. 85*. 90. 99.118, and a few others. Un s' appellava BEFGHINPORSTUV^XYZ^CZAW^n-t^a^K X(i|oirpo-Txwde 1 (?)hkortwxz(:^%//w 1 (?)^r.i- 1.2.4.6.7. 12. 15. 1 6. 1 8. 19.20. 22 . 24. 26.27.29.31.34.35.37.40.45.48.49.50.55.56.57.58.63.66.68.77.78.79.81.82.84.88. 89.91.93.95.97.98 [.i. in marg,~\ ioo.iO3 1 (?). 105.1 12.113. 117. Bat. 55.102.114. 129.132.136.142.145.362.426.431. [126] Among the above, the following have Uno, O&i-wor 6.24.35.45.55.112. 117. Bat. 426. XXVI. 136. El si chiamb poi (or poscia, or da poi) ADEFGOORSTUV-WXTr. mff)qrsti>x 1.3.4.6.7.8.12.15.16.17.18.20.21.22.23.24.26.27.28.30.31.33.34.35. 36.37.39.40.41.43.45.47.49.50.51.54.55.58.63.66.70.75.77.78.79.80.81.82.84.88. 89.91.92.93.95 (see below) 97. 98. 99. 100. 101.103. 1O 4- 105.106.109. 1 1 i.i 12.1 14. 115.116.117.118.119. Bat. 10.30.55.129.132.145.431. [158] Et El si chiamb BHiw 69. E .1. si (sic) 1 10. Elli si 14. Et Elli si w. E si C 11. Essi Bat. 142. Eli (Ely or Hely) si chiamb WLPUVLh*wcmit 2.19.64.67.68.71. 102.108. Bat. 102.146.362.444. Etli si chiamd I. Et L si np\o 46.76. Bat. 136. Et si chiamb poi 32. L si chiamb MFS (chiama) z 42.59*-73- 8 5*-94-io7. Bat. 65. .L. (sic) si chiamava 10. El si chiamava Bat. 440. He el (sic] si chiamb poi 48. El poscia si chiamb 7.29.57.90. 1 13. Et poscia si chiamb V J . Eli si chiamb e poscia si conv. P. 1 s' appellb poi 13. 488 PARADISO. Poscia si chiamb e ci6 conv. 56. E chiamossi L poi e cib conv. . . El si cMamd e cib convenni 38. El (? error for Et) si chiama puoi Eli s. El ci chiamb e poi el ci convenni 95. Et tre si chiamb (!)j. .L. si chiamb (sic) M. We have here a great number of variants, but all of them seem to be easily accounted for through a misunderstanding of an original '/' or '/', standing for the Hebrew letter ' Jod\ and representing either the 'Jah' or 'Jehovah' of Ps. Ixviii. 4, &c., or still more probably (since Adam says that his primi- tive language, of which this would be presumably a fragment, was ' tutta spentaj 1. 124), the letter ' Jod\ regarded as the cabalistic symbol for God. (On this point Scartazzini's full and interesting note should be consulted.) It is especially noteworthy that not only this letter itself was a sacred cabalistic symbol, and that its pronunciation was, from motives of extreme reverence, substituted for that of the full name of which it was the initial letter (see Lubin's note //./.), but also in particular that the letter with a point on each side (thus, }) had a special significance. This is precisely the form which is found in many MSS., and it is curious that the letter L is similarly treated (-L-) in one or two MSS. which have that false reading. The statements in the text become thus at any rate quite intelligible. God was first known in the primitive and early extinct language of Adam (1. 124) by the sacred and mysterious symbol '/' or ' J' ; next by the term l Ef (1. 136). In this latter word at any rate we re- cognize a distinct reference to Hebrew, which, in a passage very hard, if not impossible, to reconcile with 1. 124 above, Dante plainly declares to be the primitive language of man, and to have been spoken by Adam himself. See De Vnlg. Eloq, I. vi (fin.} " Hac forma locutionis (soil. : certa forma locutionis a Deo cum anima prima concreata) locutus est Adam, hac forma locuti sunt omnes posted ejus usque ad aedificationem turris Babel, hanc formam locutionis hereditati sunt filii Heber, qui ab eo dicti sunt Hebraei...Fuit ergo Hebraicum idioma id quod primi loquentis labia fabrica- COLLATION OF SELECTED PASSAGES. 489 verunt." It seems then that this must be added to the very small list of inconsistencies between Dante's several works 1 , on which I have spoken already in Proleg. p. xxxix (note] and shall add a few words presently. Now as to the origin of the other readings. These may be, I think, easily accounted for on the supposition that '/'or ( y was the original reading. First, '/' may have been very easily mistaken for the numeral, and hence at an early period the reading Un or Uno, now widely spread, may have been substituted for it. Next '/, ' or '/' might have arisen by a misreading of'/,' though this reading is scarcely worth dis- cussing, since it is quite unmeaning, and is extremely rare, being found in D only out of all the MSS. which I have examined here (about 240). It is very strangely adopted by Witte, against the evidence of three (and I should say the best three) of his four MSS. The reading ' ET also appears in one MS. (U), and is adopted and defended by Dan. (who is followed by many later Commentators and Editors) by the dangerous argument that in the De Vulg. Eloq. I. iv Dante states that the first sound uttered by Adam was the name of God, and that that sound was 'El*! (This argument would rather point to the probability of the reading being a deliberate rectification of the text in order to remove an obvious difficulty). If we were thus to adopt El in 1. 134, we must of course read Eli 1 Giuliani in his note onDe Vulg. Eloq. 1. 4 strives hard to explain away the discrepancy by citing such passages as lines 137 8 of this Canto, and still more De Vulg. Eloq. \. 9 and Conv. 1. 5, where Dante speaks of the rapid changes of language even in 50 years, whereas Adam lived more than 900. But (i) there is a vast difference between "mold vocaboli essere spend e nati e variati," as Dante says in Conv. I. 5, and ' tutta spenta] as he says here ; (2) the survival of a word of such momentous importance as El would of itself jar with the unqualified 'tutta spenta'; (3) the passage above cited from De Vulg. Eloq. I. 6 very distinctly identifies Hebrew with the actual language spoken by Adam, which could not possibly be the case if the latter were ' tutta spenta? Giuliani is clearly ' dfcrtv 8ia(f)v\aTT(ov.' Philalethes (like Scartazzini) maintains that we have here a clear (and moreover intentional) contradiction of the De Vulg. Eloq. This would be in fact one of Dante's 'Retractations,' on which subject see further, Proleg. p. xxxix (note). 2 Another reading found here is ' Eli' 49O PARADISO. in 1. 136, but this also has very slight MS. evidence compared with the overwhelmingly strong attestation of El in that line. Blanc (Dis. s.v. '"/') and Giuliani (on De Vulg. Eloq. I. iv.) quote what looks like a very remarkable parallel passage from S. Isidore thus : " Primum apud Hebraeos Dei nomen El dicitur, . . .secundum Eloi." But on reference to the original, the resemblance will be found to be much less striking and conclusive. It occurs in the Origines B. vii. c. I. It is there stated that there were ten successive names of God 1 , viz. (i) El, (2) Eloi, (3) Eloe, (4) Sabaoth, (5) Elion, (6) Eieie, (7) Adonai, (8) Ja, (9) Ja Ja (or the ' Tetragrammaton'}, (10) Saddai. This scarcely suggests a conscious imitation, since the precise form Eli does not occur in Isidore, nor does Eloi occur (as far as I know) in a single MS. of Dante. It may be added that S. Isidore in ix. I I accepts the common theory that Hebrew was the primitive language, but this was so generally believed and so often repeated that such a coincidence would prove nothing. Dante may well have known the works of S. Isidore, but he never quotes him by name, nor is there any proof whatever (unless I may have overlooked it) of any acquaintance on the part of Dante with his writings*. Such negative evidence is more significant because the De Vulg. Eloq. travels over ground in which Dante would naturally have referred to S. Isidore had he been a familiar author. He is of course mentioned in Par. x. 131 together with Bede and Richard of S. Victor, &c. It becomes therefore an interesting question whether we are to reject overwhelming MS. evidence or to acknowledge a clear contradiction of Dante's own statements elsewhere. I certainly incline to the latter supposition, because Dante does certainly contradict in 1. 124 (where no difference of reading is in question), what he says in De Vulg. Eloq. I. vi., and we can only vindicate his consistency with De Vulg. Eloq. I. iv. in this passage (1. 134), at the expense of 1 This number is also mentioned by Pet. in his Commentary h. I. * On De Man. II. 8 Witte quotes a passage of S. Isidore which illus- trates strikingly a statement of Dante ; but they are both referring to a ' locus communis ' of symbolism, so that there is no direct proof that Dante derived it from that source. COLLATION OF SELECTED PASSAGES. 491 an inconsistency between lines 124 and 134. An incon- sistency between two different works separated by some years is more probable than an inconsistency in two passages ten lines apart. It is well known too that in two passages at least in the Commedia, and both occurring in the Paradiso, Dante does distinctly refute opinions expressed by himself elsewhere, viz. Par. ii. 61 compared with Conv. ii. 14 (the cause of lunar spots), and Par. xxviii. 98 seqq. compared with Conv. ii. 6 (Orders of Angels). It is not improbable then that this may be a similar case. Indeed his language here is so explicit that Scart. observes, " Pare che Dante dettasse questi versi appositamente per retrattare quanto in proposito aveva asserito nel De Vnlg. Eloq!' (p. 714). Another incon- sistency (real perhaps, but less striking) which is sometimes pointed out is the admission of ancestral distinction as a condition of nobilta in De Mon. ii. 3, and its repudiation in Conv. iv. 7 and Par. xvi. I seqq. In line 136 again there is a great variety of readings, but in spite of minor variations there is a general consensus for El or Eli or some form that has evidently arisen from a mis- understanding of these ; 1 EI' however actually having almost overwhelming support of MSS. The older Commentators may be classified as follows in reference to the two names of God which they adopt in lines 134, 136 : I ... Ely Lan.) Anon. Fior. I ... El Ott.,Serr. El ... Eli Benv., Dan. Un ... Eli Buti, Veil. El seems to be implied clearly by the language of Pet., and Eli probably by that of Land., as the original name, but both are silent as to a second name. I cannot therefore understand the way in which these two Commentators are cited by Scartazzini h. I. Dionisi (B. F. p. 19) hazards the strange conjecture that the true reading is El i si chiamb, i'=ivi, i.e. in terra. In the same passage he contends for the reading /or J in 1. 134, among other reasons (some good and some bad, including among the latter that / or J is the initial letter Qijustizia, imperio, imperatore, &c. !) suggesting that the 492 PARADISO. form of / or Jod is well adapted to express the simplicity of the Divine Nature, which Dante elsewhere symbolizes by an intense /w'#/ of light, as in Par. xxviii. 16 21. The 30 Editions which I have before referred to are divided as follows : Un ... El The first four Editions.' Un ... Eli Nine. I ... El Three. I ... Eli Four. El ... Eli Nine. L ... El Witte only. XXVII. 57. difesa in 98 MSS. giudizio Dp 21.33. vendetta p\o 46.59.68. The three readings difesa, giudizio, and vendetta are re- corded here chiefly as illustrating the tendency which we have noted elsewhere to change words or expressions into others either clearer in meaning, or perhaps of mitigated severity. Most likely difesa is original, since (apart from the strong support which it receives from the MSS.) it is rather more out of the common than the other words, and therefore less likely to have been substituted for one used in an obvious or ordinary sense such as giudizio or vendetta. Moreover either giudizio or vendetta may have arisen most naturally as marginal glosses on the somewhat unusual difesa. The use of difesa here is very like that of - soccorso in Par. xxii. 96, the same intervention being regarded either as difesa or vendetta, al parer di colui Che disiando o temendo 1' aspetta. (Par. xxii. 17, 18.) Note also the occurrence of difese and soccorrd in lines 62 and 63 just below, which seems to me strongly to support difesa here. Compare too the interchange of vendetta, giustizia and potenzia in Inf. xxiv. 119, and of giustizia and vendetta in Inf. xi. 90, and the notes and illustrations on the latter passage. A discussion of these three readings in Foscolo's Discorso, p. 411 may be referred to. COLLATION OF SELECTED PASSAGES. 493 XXVIII. 23. Alo (flrHalo) cinger CNQRV 2 Gl.Z?(?)/P.9FAai;eiinKabegknort acglqv i. 2.6.7.10.1 1. i6.i8.i9(?). 24.27.3 1.37.38.47.48.55. 64. 77. 104. 107.108. [59] Allo cinger DFSC"6TIISvdfiuvwxyz;;2/ 3.35.39.5 1.54.78.79.80.82.91. 102.105.106.109. [36] Al cinger EGHPTV 1 WXZZ>(?)4YiiKnpo-Tuxshlmp/r^ 4.15.20.21.26. 29.30.33.40.43.56.57.63.66.67.70.75.76.81.84.90.112.113.118. [54] Ad lo cinger A. A lo cinger IYTr.' 17. Alor cinger U. A cinger 14. A cinger 11 la luce . . II cinger la luce k. II cinger della luce 32.42.71. Al cinger quella luce 68. Alo al cinger OF 119. Alo accinger A. Lo giugner della luce d. A conciger (?) B. A contigiar j. Al congiugner Ap\o 46.69. TZ>1 and 75 (and perhaps others) have della luce. Note. D has Al cinger according to Witte, Halo cinger according to Sicca, but the former is probably correct since della luce follows. The form 'Alo,' being somewhat unusual for 'A/one 1 ,' will account for the existence of almost all the variants found here as having arisen from either misunderstanding or deliberate emendation. Indeed it is almost the only reading which gives any sense at all, and it gives one which is perfectly distinct and appropriate. Moreover in Par. x. 67 Dante again draws a comparison from the same phenomenon. As Scart. observes, ' Alo 1 and 'Allo' are mere differences of orthography. The same might almost be said of A lo, re- membering the constant irregularity in MSS. in respect of the 1 Such forms occur occasionally elsewhere in Dante, as we should expect in his archaic and semi-classical diction : e.g.polve ii. 133, turbo two or three times (besides once apparently as z.&). = Par. ii. 148), image Par. xiii. 2, erro Inf. xxxiv. 102, and some have sug- gested (very questionably) mile for milite in Inf. xvi. 102. 494 PARADISO. divisions of words. Thence would naturally arise, from mis- understanding, the readings Ad lo and Al. Next would follow deliberate emendations to improve the sense, such as A, //, or Z0,or the change of the verb, and so on. Among Commentators, Lan., Anon. Fior., Benv. and Buti read Alo, and explain the passage correctly. Dan., and apparently Land, and Veil., manage to extract a similar meaning out of the reading A lo cinger ; Dan. paraphrasing ' quanto la luce che lo dipinge pare appresso a lo cinger, cioe ad esso cerchio ' &c. XXVIII. 50. volte BCFGINQYZ,4 ( egijkloptuwzrtM//yz/ 1 . 10.1 1 . 14. 1 7. 1 8.2 1 .24.27.29.30.3 1 .35.37.38.42.46.47. 56.57.58.59.64.69.71.77.79.90.101.104.105.107.111.113. [90] cose AEO P WXZ}(?) J F/ ) A*Y0K|A|porTve (marg.}\\tmxcikrt 2 . 3 . 4*6 . 7. 1 6. 20.26.28.33.39.43.44.51.55.63.66.67.68.70.75.78.81.82.84.99.102.108. 109. 1 18. 1 10. [61] Off. rote HAJn?2vvydgr(marg-.)x8(marg.i ma manu)$4.76.8o.io6(?se rote). [17] viste 32.112. luci 48. note (or vote ?) n. le colse (sic) ? 4>. le vostre D. Le volte veder o. Tanto pill festine is the reading of F A (marg.) D (marg.} {JXp 21.46. and possibly others. r has a blank space between le and tanto without any signs of erasure, as if the scribe had been in doubt which word to insert. D has cose according to Witte, volte according to Sicca and Scart. The reading volte is certainly to be preferred here (i) as giving to the passage a perfectly appropriate sense, as well as one very definite and technical in reference to the Ptolemaic system of astronomy ; (2) as being a word employed else- where by Dante in precisely the same sense, viz. of the heavenly ' spheres,' in Purg. xxviii. 104, L' aer si volge con la prima volta ; (3) as being just sufficiently removed from common use to suggest either the substitution of a more obvious word, or at any rate the addition of a marginal gloss which would be COLLATION OF SELECTED PASSAGES. 495 easily mistaken for a var. lect., and then introduced into the text. The word rote is just such a word. Cose again is one that might recommend itself by its safe vagueness to an annotator who did not understand the passage, just as seems to have been the case at Par. xiv. 72. Viste and luci are obviously conjectural, and are very rarely found. On the other hand, with rote or cose as original, it would be diffi- cult to explain the occurrence of volte. Witte (Prol. p. Ixv) assumes volte as obviously correct, and cites cose as an instance of an indubitably false reading (inter alia] in the Ottimo Commento. Besides Ott., it would seem clear that Lan. and Anon. Fior. (identical, as usual hereabouts) are commenting on cose. Buti and Land, have volte, the former explaining by ' li giri e le revoluzioni,' and the former by ' i giri che fanno i cieli.' Benv. comments 'id est, circulos vel orbes,' which may be a comment either on volte, or on rote, which latter word he is made to quote in his printed Commentary, this not being necessarily any evidence as to the text which he adopted. In regard to the variants divine and festine at the end of the line, it is scarcely necessary to point out that the latter is an obvious case of the substitution of an easier word, very likely first appearing as a quite natural marginal gloss, and then mistaken for a variant. There is no trace of any other reading than divine in the early Commentators who notice the word at all; viz. Ott., Lan., Anon. Fior., Land., Buti (piu divine, cioe piu veloci), and Dan. (piu divine, tanto piu pcrfettc e veloci). The last two explanations shew how easily such a gloss as I have supposed may have arisen. XXIX. 4. zenit (cinit, &c.)inlibra (or in libra) C*GHINPQRV 2 YG1.^CZ I G// ) w*( to 2.8.10.13.19.20.21.33.35.38.39.42.43.44.46.51. 58.59^.64.66.67 .68.70.71. 76*. 77.78.80.94.99.100.102.104.107(11 libra). 109.1 19. Bat. 146.362. [98] Lan., Anon. Fior., Ott., Buti, Veil., Benv., Serr. tiene in libra B*EFOSTUV 1 WXZ,4 (marg.) BFJPQW) IL^^l^y. v|o-TX)x4'sab(il librajkmortxa^?)//^'//;^.*- 1.3.4.6.7.1 1.12.14.15.16.17.18.22. 23.24.26.27.28.29.31.32.36.37.40.47.48.55.56.57.63.69.75.79.81.82.84.85.88. 496 PARADISO. 89.90.91.92.93.95.97.98.101.103.105.11^112.113.115.116.117.118. Bat. 145. [108] Dan. il zenit e in libra A. che tienne a libra 1 14. intiene in libra D. viene in libra A. che dal centre in libra d. che '1 centre inlibra Tr. SI 54.106. che 1 cerchio dilibra j. che 1 tenit (sic) S. ch' el tenete in libra 1 10. ch' el genito inlibra (sic) 108. Quando e del punto ch 1 ei rimiri in libra 30. Z has ch' egli tiene, which probably = che li tiene. In 'i' the line is left blank after /punto' as though the scribe were puzzled. M has che '1 Cenit in Libra (sic). There are four principal points which call for discussion in this very difficult passage : (i) The reading; (2) the meaning of the unusual words inlibra and dilibra ; (3) the translation ; (4) the general interpretation. (i) As to the reading, there should be, as it seems to me, very little difficulty in choosing zenit inlibra and rejecting tiene in libra ; though the MS. support for either reading is nearly equal. For (i) zenit inlibra is prima facie more difficult, since it contains two words which are uncommon, and therefore liable to be misunderstood, zenit (or cinit, &c., see below), and the cnra^ Xeyopevov inlibra. (ii) The supposition of this reading as original readily accounts for the existence of the other. Indeed there are two or three ways in which tiene in libra may naturally have come to supplant zenit inlibra. Most probably (as Scart. suggests), the former words were a marginal gloss explaining (accurately enough) the unusual word inlibra, and they would not unna- turally be mistaken for a var. lect. of the equivalent length of text, 'zenit inlibra' Or again, it is not impossible that if inlibra were written divisim, i.e. in libra (and this was likely enough either by accident [see Pref. Note p. li], or as substi- tuting two very common words for a difficult a?ra \ey6fievov), the want of a verb would be supplied either by a misunder- standing, or by a conjectural alteration, of the unfamiliar COLLATION OF SELECTED PASSAGES. 497 word zenit. This word might be not only unfamiliar, but even unintelligible, when written, as it frequently is, cenit, cinit, or cienit. The change to tiene (conjectural or otherwise) becomes still more natural when we recollect that in some MSS., especially those of early date, c and t are almost indis- tinguishable, so that there would be a very slight difference indeed sometimes between cienit and tiene. In fact, I find the hybrid form tenit reported in one MS. (5). Among the early Commentators, the only one, as far as I can see, who adopts li tiene in libra is Dan. He expressly says " cosi ne' buoni antichi testi si legge, e non cinit." (His explanation of 'punto' is curious : " per il ' punto ' adunque intenderemo quello stile o linguella...che tien giusta la bilancia.") On the other hand, Lan., Anon. Fior., Ott. (who speaks of "ritta e ceniti- cata oppositione "), Benv., Buti, Veil, and Serr. all adopt and explain zenit or cinit. Land, is indeterminate. (2) In the next place, I cannot but think that the bold coining of two such antithetical words as inlibra and dilibra (both, be it observed, being aira^ \ey6^eva) is thoroughly in Dante's manner. The number of these specially-coined words compounded with c m' may be counted almost by dozens, and it is remarkable that they nearly all occur in the Paradiso. No doubt these words are in the large majority of cases used in the passive constructed reflexively. But this is by no means always the case, and the following are some instances that I have noticed (and there are likely to be others) of the ordinary transitive use, intuare and immiare Par. ix. 81 ; ingemmare Par. xv. 86, xviii. 117, xx. 17; incielare Par. iii. 97 ; imparadisare Par. xxviii. 3 ; invogliare Par. iii. 84, Purg. xiv. no; inanellare Pnrg. v. 135 ; insaccare Inf. vii. 18. We may add a few words about its companion word dilibra. This is constructed in the usual reflexive way, and its formation may be compared with that of the following words, which are also, like itself, awa^ Xeyo/^e^a divimarsi in 1. 36 of this Canto ; dismagliarsi Inf. xxix. 85 ; and disinalare Purg. xiii. 3. The analogy of such words, as well as the obvious antithesis with inlibra (if that be accepted), would certainly point clearly enough to the meaning of dilibra, viz. to break the balance or destroy equilibrium, just as inlibrare D. i. 32 498 PARADISO. is to form a balance, or constitute equilibrium, and this antithesis again would reflect back more probability on inlibra, as against in libra, in 1. 4. I mention this chiefly because dilibra has sometimes been taken as if equivalent to dilibera, as, for instance, by Buti ('cioe, si dilibera, escendo fuora di quello,' soil, orizzonte, or cinto), and Veil, (' si deli- bera, divide e parte da quel cinto'), and apparently Ott. also, since his text reads si delibera dal zenit, as is admitted by the Editor (Ed. Pisa, 1829) who has altered it to dilibra. (3) Next as to the translation. Here three points present themselves. (i) Is inlibra transitive or neuter ? (ii) Consequently on this, is che a relative introducing the object of inlibra, or a sort of temporal particle, as in Inf. x. 108 and xxxii. 125? (iii) Does /?/;//# refer to time or space? I think there is little doubt that inlibra, after the analogy of similar compounds, is transitive, though even were it natur- ally so, it might perhaps be justifiable to use it intransitively here (if it were thought necessary) on the analogy of the use of rinnovella in Par. xxvi. 128. In the latter case, the transla- tion would be something of this sort: "From the point or moment when the zenith forms equilibrium, until &c." Besides the passages above cited for this use of c/te, several others are given from Boccaccio in Vocab. Tram. (s. v. Che 13, No. 18). I do not however see any difficulty in Mr Butler's objection to taking inlibra transitively, viz. that ' the zenith could not be said to keep the point, whether of place or of time, balanced,' since it could quite well be said to hold balanced the situa- tion, or point, when the sun and moon are directly opposite to one another. I would therefore take inlibra transitively ; che as referring to punto ; and pnnto as probably describing the situation of exact opposition of sun and moon. The //m^-comparison for which the whole passage is introduced is sufficiently indicated by Quanta and Tanto in 11. 4 and 7, without its needing to be helped out by giving a tem- poral sense to clie also. Thus the general meaning (freely rendered) would be : " As much time as elapses from the exact point which the zenith poises 1 , when the horizon forms 1 Or ' equilibrates,' if we might imitate inlibra by a similarly unusual word in English. COLLATION OF SELECTED PASSAGES. 499 a common girdle for the sun and moon (i.e. when they are in direct and exact opposition), until the balance is broken by one passing up into one hemisphere and the other down into the other, for so long a time was Beatrice silent, &c." It might be worth while, for the general idea of the passage, to compare De Mon. I. xi (al. xiii), where Dante, after quoting Aristotle, " ut Philosophus inquit, ' neque Hesperus ncque Lucifer sic admirabilis est,' " adds " Est enim tune Phcebae similis, fratrem diametraliter intuenti de purpureo matutinae sereni- tatis." (4) Lastly, as to the general interpretation of this com- parison what amount of time is intended to be described by it? On this there are two opinions, viz. that it is (i) an absolutely instantaneous moment ; (ii) a moderate interval, such as would be occupied by one of the two orbs rising clear of the horizon and the other sinking below it. The former appears to be the view of Lan., Anon. Fior. (' in instanti'), Fals. Bocc, (' di subito '), Oft. ('quasi in uno mo- mento '), Pet. ('subitissime '), Land. (' subitamente '). The latter is maintained by Bcnv. and Dan. Benv. says, " Sicut enim in posito casu lima quo; recipit lumen a sole modicum teinpus stat sibi opposita in recta linea": and again, "quia sol in ariete ascendit hcmisperium nostrum, et luna in libra descendit cito." Dan. explains thus : " cioe dal cominciar a levar del Sole infino a tanto che sara tutto levato ; il che avvicn della Luna ancora, ma per il contrario "; and he adds that " quello spatio di tempo e pochissimo." Veil, says ' brevissimo tempo,' which is perhaps capable of bearing either interpretation. The explanation of Scrr. seems to waver be- tween the two interpretations. His comment is " mutando emisperium quod tune utcrquc habet ; scilicet subito propter motum continuum cceli mutant emisperium : sc dcliberat, i.e. rccedit ab illo orizzonte et quasi in instanti" But on I. 7 he writes " Tantum, supple modicum, cum vultu, &c." The former interpretation no doubt would correspond to the strict astronomical fact, but the addition of the words camlriaudo emisperio seems to imply that Dante is rather appealing to a visible and popularly intelligible phenomenon. This, more- over, I believe (as I have elsewhere maintained) to be Dante's 322 500 PARADISO. usual practice when referring to the movements of the heavenly bodies for data of time. XXXI. 20. plenitudine AC D EO STUXZ/i 5CZ An2*^pSrieX|x| cfhlmqr(marg.} 1.2.4.6.14.16.17.18.19.23.24.27.28.29.31.32.35.37.39.46.47.48. 55.56.57.63.67.78.79.81. 90.93.102.105. 109.111. 112.113.118. [87] Veil., Dan. moltitudine BFGH I N VQKWYTr.DEFGHIJKMSUVT&a.vKvoTrp v-^v>f&g\)\)\vvp\.vyzd'egkn l rlvx- 3.7.8.10. 1 1. 15.20.21.26.33.38.42.43.44.50.51. 54.64.66.68.69.71.75.77.80.82.84.92.99.100.103.104.106.107.108.110.119. [97] Benv. (?), Butt, Land. (?), Serr. 30 has splenditudine. * has plenitudine obstante. The two readings /&#*Vzftft#i and moltitudine are registered as being at any rate unmistakably distinct, and moreover, as Scarf, observes, moltitudine seems clearly to be a gloss on the obscurer word plenitudine, which is probably therefore to be preferred. In any case, if moltitudine were original, it is scarcely possible to imagine how the other reading could have arisen, except indeed on one supposition, which perhaps in some cases like this may hardly have been sufficiently taken into account, viz. whether Dante himself may not have hesi- tated between the two words. We know that even in these days of printing and publishing, authors frequently introduce variants (e.g. notably Wordsworth and Tennyson), and not by any means always improvements. How much more easily would this be done, and how much more freely would it be likely to be done, when only a few copies existed in MS., and especially when, as we gather from internal evidence, the Corn- media was kept constantly at hand by its author, and subjected to alterations and additions. (See on this notes on Par. xxvi. 96 and Proleg. p. xxi.) Ugo Foscolo in his Discorso insists much upon this, and considering the enormous number of variants on every page, and the frequent difficulty on critical grounds in suggesting any adequate motive for at any rate an outsider making an alteration, I cannot but think that this con- sideration of author's own variants may not have been suffi- ciently allowed for. In such a case two readings would have COLLATION OF SELECTED PASSAGES. 50 1 nearly equal authority, like a picture and its replica, with some variations, by the original artist. Among the Commentators, Veil, has 'di tanta plenitudine e moltitudine d' angeli.' It seems clear from this that his actual text read plenitudine, and his employment of moltitudine as an explanatory gloss shews how easily it might have been received as a variant. XXXIII. 57. la memoria is almost universal. la materia occurs in V 1 Z/?Zx(?i8. 27. 28.29.3 1.56.57. 90.93. 113. The reading materia is characteristic of the Vatican family of MSS. (see Appendix), but there can be little doubt of the correctness of the generally received reading memoria. Apart from its intrinsic preferability (which will scarcely be disputed), it expresses an idea which occurs two or three times elsewhere in Dante, without conveying it in such identical terms as to suggest the suspicion of intentional imitation. See Par. i. 4 Q 1 , and especially note the com- ment by Dante himself on these lines in the dedicatory Epistle to Can Grande, 28 " intellectus humanus in hac vita propter connaturalitatem et affinitatem quam habet ad substantiam intellectualem separatam, quando elevatur, in tantum elevatur ut memoria post reditnm dcficiat, propter transcendisse humanum modum." He then proceeds to quote the well-known language of S. Paul in 2 Cor. xii. 3, 4. See further Dante's comment in 29 of the same Epistle to Can Grande on Par. i. 6 cose che ridire Ne sa, no puo chi da lassu discende : "'nescit et neqnit ;' ncscit, quia oblitus ; ncquit, quia si recor- datur et contentum tenet, scrmo tamcn deficit." There is no trace of memoria in any of the old Commentators. 1 It is perhaps to be noted that the expression 'materia del mio canto 1 occurs in the same context 1. 12 ; but whether or not that may have suggested its introduction here, it is clear that the ottraggio would not affect the maieria, though it would the memoria. 502 PARADISO. XXXIII. 89. Quasi was found in 144 MSS. Tutti lW2J3Lttffar(marg.) 1.2.18.19.27. 29. 31. 55*. 56. 57. 70. 90. [25] Tanti conflati NSic 16.26. Tanti conflanti (sic) x. Tanti construtti 118. Tanto conflati 22. Quasi colcati (?) 3. Conflati insieme quasi 0115. Quasi confinti insieme 114. Some MSS. have confrati, e.g. EFSrtuv/z 16. 48. 84. 95 (confratte). 1 10, &c. This however is only a case of the interchange of r and / which is found in many other words, and is not properly a 'variant.' The reading Quasi is that of the vast majority of MSS. here. Tutti is characteristic of the Vatican family and a few others. The former is probably correct, for it is natural that the strong metaphor conflati should be thus qualified ; and further quasi well indicates the sense of a mystery which the poet could not adequately explain or express (so Scarabelli quoted by Scarf, h. /.). XXXIII. 126. te ami ed arridi CFORS (e te arr.) VJACDEFGHIJKLQSUVTK* rg.^&Sh.ilmsvybtfgkt 6.7. 10. 1 3. 1 6. 1 7.20.22.33.35.42.43.44.48.49.5 1. 54.55^64.70.71. 73. 76.78.85.94.99.100.102.106.107.108.115. 119. Bat. 129. [78] tea me arridi AB*DEGHINPQTUXYZTr.^/ > A2^aYS\jxopTvx>|/ws Fabegklrtuwz^^?)!?/^//*;;^ 1 ^^-*" i -2.3.4.8. 1 1. 1 2. 14. 1 5. 1 8. 1 9.2 1. 23.24.26.27.28. 29-3o.3i-32.38.3946.56.57-59-67-68.72. 77-79-82. 84. 88. 89. 90. 92. 93. 97. 103 (intendendo). 104. 105. 109. 1 10 to 114.118. Bat. 444. [Ill] a me arrivi (sic] p< l degr 4.11.18.21 (marg. 108.113.119. [33] 14 has Pare a te. has te, but the mark over seems later. 90 has Pareva in me. In this case an overwhelming majority of MSS. has pre- served the correct reading in tc, though, as we have seen else- where, this consensus would not by any means suffice to prove it to be so. The reading in trc, which has a vague and super- 1 Since writing the above I find the following very similar remarks on the rci'tewrg a me arridi in Dionisi Ancdd. IV. p. 175 : " le quali parole potrebbon dirsi a Dante ingiuriose, e in ccrto modo anche a Dio : poiche per csse si turba il mistero della Triade, togliendosi da questi sacra- tissimi vcrsi la terza clelle divine Persone, e si finge che : 1 Poeta sapi- entissimo la teologica suaorazione interrompa (perche mai?; per dire a se stesso arridente la Santissima Trinita ! " 504 PARADISO. ficial appropriateness in an address to the Trinity, is really inappropriate, since Dante's language is never, and least of all on such a subject as the present, vague or inaccurate. This reading would be both, for it overlooks the very definite limitation implied by " Quella circulazion, che si concetta pareva...come lume reflesso" language which distinctly limits the passage to the Second Person of the Trinity. This is also required by what follows in 11. 130, 131, and for this purpose the expression ' come hime reflesso' is accurate and suitable. See lines 118, 119, where a similar expression is used, the word riflesso no doubt representing the idea of ' airav^aa-^a r?79 o?79,' ' T09,' &c. Te and tre are also confused in Inf. iv. 87. All the early Commentators appear to have read te 1 . 1 In the printed Ed. of Anon. Fior. (Fanfani, Bologna 1874) we read, it is true, ' Cioe nelle tre persone che m' apparevano, &c.,' but as this note, as well as this Commentary generally in the Paradise, is identical with that of Jacopo della Lana, in whom we read ' Cio& nelle trie persone, &c.,' it seems probable that tre is a misprint, or perhaps a MS. variant, for tue. LIST OF PASSAGES DISCUSSED. INFERNO. Page Page I. 28 257 XII. 16 300 42 259 66 302 48 263 125 302 94 264 XIII. 20, 21 303 II. 12 265 63 304 22 265 XIV. 48 307 . 55 266 126 307 60 270 XV. 39 311 81 273 XVI. 3 312 HI. 8 275 14, 15 313 3i 275 104 315 36 276 XVII. 20 315 59>6o 278 95 315 114 278 XVIII. 12 317 IV. 23 279 42 319 68 279 55 321 95 280 91 321 141 282 XIX. 3 323 V. 38 283 23 325 59 285 64 ,...325 102 286 XX. 30 326 ,, 126 290 69 327 135 290 XXI. 14 329 VI. 43 291 46 329 VII. 108 292 93 330 VIII. 78 293 112 331 125 294 113 332 IX. 54 294 XXII. 101 333 7 296 142 333 125 298 XXIII. 83 335 X. 65 298 XXV. 8 336 XI. 9 299 120 337 5o6 LIST OF PASSAGES DISCUSSED. Page Page XXVI. 25 337 XXXII. 47 355 XXVII. 21 338 11 136 356 XXVIII. 10 340 XXXIII. 26 357 83 343 11 72 362 135 344 11 74 363 XXIX. 46 , 351 11 78 364 XXX. 18 352 XXXIV. 13 365 11 44 354 11 53 366 11 115 354 11 99 366 XXXI. 75 354 PURGATORIO. i. 108 368 XVIII. 83 39 r ii. 13 369 XIX. 34,35 393 26 371 11 132 394 11 44 372 XX. 9 395 93 373 XXI. 19 397 118 375 11 25 399 IV. 22 375 11 45 400 11 72 376 11 61 401 125 377 XXII. 5,6 405 V. 38 378 11 97 410 i) 88 379 XXIV. 61 413 136 380 XXV. 9 416 VII. 15 381 11 31 418 51 382 )1 131 420 VIII. 64 384 XXVI. 41 420 11 94 384 11 72 421 94 (to).... 385 11 75 422 IX. 17 385 XXVII. 4 423 X. 14 386 J) 81 425 11 3 386 11 in 426 XIII. 2 388 XXX. i5 429 11 154 389 11 73 431 XIV. 126 390 XXXI. 29 433 XVI. H5 391 11 51 433 PARADISO. , i. 135 436 II. 124 444 n HI 439 11 Hi 445 II. 42 442 III. 15 447 LIST OF PASSAGES DISCUSSED. 507 V. 88 VII. 21 Page 449 449 IX. 129 ......................... 453 X. 4 ......................... 454 112 ......................... 455 "9 ......................... 457 XI. 26 ......................... 460 82 ......................... 462 XIII. 27 ......................... 463 XIV. 72 ......................... 464 XVI. 47 ......................... 464 XVII. 9 ......................... 465 XVIII. 75 ......................... 466 123 ......................... 467 131 ......................... 470 XIX. 141 ......................... 471 XX. 117 ......................... 472 XXII. 17 ......................... 473 Page XXII. 94,95 474 XXIII. 115 477 XXV. 3 477 U 478 29 479 33 48i XXVI. 96 482 104 483 134- 486 1* 136 487 XXVII. 57 492 XXVIII. 23 493 So 494 XXIX. 4 495 XXXI. 20 500 XXXIII. 57 501 89 502 126 502 128 503 ACCOUNT OF THE MSS. EXAMINED OR COLLATED. In order to save space I have not generally printed at length passages the record of which will be found in the Collations above, unless there appeared to be some special reason for doing so. Wherever any peculiar reading of a MS. is described by a mere reference to the passage, it is to be understood that the reading is to be found already recorded, either among the Selected Passages, or in the full Collation of the 17 MSS. for the Inferno. If a reference is given without the Cantica being specified (e.g. xii. 36) it is to be noted that the Inferno is referred to. A (Bat. 491). [In the case of the Bodleian MSS., it is unnecessary to give details as to titles, colophons, &c., since they are all fully described in Mortara's excellent Catalogo dei Manoscritti Italiani, &c., Oxford, 1854, which may be consulted throughout.] In the Bodleian Library (Canon. Ital. 105, 106, 107). A fine MS. on vellum, in large folio (i$f x u), consisting at present of three volumes, but perhaps originally in one volume, as the numeration of the pages is continuous. Its date is probably of the early part of the I5th century. The orthography shews beyond a doubt that it was written in the Venetian district. It contains an Italian translation by an unknown hand of the Latin Commentary of Benvenuto da Imola till lately unpublished, but recently given to the world through the munificence of the Hon. W. W. Vernon (Florence, Barbera, 1887). Mortara states that he knows of only one other MS. containing this particular version of Benvenuto's Commen- tary, viz. one in the National Library at Paris, no. 7002, 2, Fonds de reserve [now " Ital. 73"], numbered 429 in Batines. It should be noted that the readings in the text frequently fail to correspond with those in the Com- mentary. A remarkable instance occurs in Par. xxvi. 104, where the Commentary assumes and defends the reading Da te, while the text has the almost unique reading La voglia tua discerno io meglio, omitting both the disputed readings Da te and Dante 1 . From the cramping of the letters of the text in one or two places, e.g. Inf. xv. (init.\ it seems as if the text had been filled in after the Com- mentary 2 . It may be observed however that the dialectic peculiarities of the Commentary and text resemble one another. 1 It is very remarkable that this same form of the line occurs in only one other MS. out of nearly 240 which I have examined on this passage, and that is precisely Bat. 429, in which alone this version of Benvenuto's Commentary is known to occur (see above). This raises the presumption of the close relation of the texts of the two MSS. also. Unfortunately that is a MS. which I have not examined closely. I find however that out of 20 passages noted the readings in all cases except two agree with those of ' A.' 2 It is extremely common both in MSS. and in printed Edd. to find that the text and Commentary have totally different readings. I have had occasion fre- 512 ACCOUNT OF THE MSS. The following may be noted as some of the most characteristic pecu- liarities of orthography. 1. Unassimilated and distinctly Latin forms of words are very common, though not nearly so universal as in some MSS. (e.g. notably ' Z '), e.g. crucifixo, obscuro, judicata, fixo, pericidoso, &c. This (as we have seen) generally affords a presumption of an old text (see Proleg. p. xii). 2. The dialectic forms are very strongly marked in this MS., and are unmistakably Venetian, or Lombardo-Venetian 1 , e.g. (a) Softened forms such as s for sc, and z for c and g, e.g. lassare, basiare, ussito, simia, sisma, sempio (x. 85), sipa (vii. 21), Pressiano (xv. 109) ; and again, zaschun (passim}, zenno, zerto. In xvii. 5, Ed accen- nolle appears as Ed a zenolle, and in Purg, xxiv. 125, Gedeon as Zedeon. Conversely g is sometimes found for z, since the soft Venetian g would give nearly the same sound as z, e.g. batigiatori (xix. 18), da Regio d' Arezzo (xxix. 109) 2 . (/3) Abbreviated forms of words, and especially the suppression of gl between two vowels, e.g. fiastro, fiol, meio, mior (xiv. 67), boiente (xxii. 141), Pitiese and Taialcozso (both in xxviii. 17), taio (ib. 38). Conversely we have Troglia, noglia and croglia in xxx. 98, seqq. The explanation is the same, viz. the evanescence of the sound of gl rendering its insertion or omission indifferent. (y) Fondness for the letter x, e.g. x for g, malvaxio, raxion, buxiardo and boxadro, abexognava (abbi- sognava), &c. x for c, camixia, bruxiato, nemixi, &c. xior sc, choxie (xxv. 55), bixie (xxv. 20), stroxio (xvii. 119), &c. (8) Preference for the vowel e, e.g. e for a, doverien, serebbe. e for i, ordene, anema, el for // ( passim}, graveda, simelc, vence, lagrimabel, cardenali, ogne (passim}. e for o, salutevel, cagieno (vii. 14), odeno. (e) Disregard of double consonants, in the way both of omission and insertion, e.g. luto (lutto), Nillo (Nilo), nerre, girrb, ally (ali\ &c., &c. () Most of the following minor varieties are found, 1 believe, in the same district of Italy. fo for f^l ; como for come ; chussi for cost ; doy for due, and con- quently to point out the inaccuracy of citations of Commentators (especially by Scartazzini) in consequence of their being reported merely from the accompanying text and not from their own words. 1 See Contarini, Dizionario, p. 14; Blanc, Grammatik, &c. p. 652 seqq. Also a great many of the same dialectic forms will be found noted under ' C.' * Compare megio for mezzo in E and Q, and vidi egiuro for vidi azzurro in E at xvii. 59. Both these are strongly Venetian MSS. Add the interchange of gire and ire noticed in the Collations. EXAMINED OR COLLATED. 513 versely nuy for not; elgli for egli ; torgier (forcer); dil ( = del). Strella for stella appears four or five times, and this is, I believe, a common pro- nunciation about Modena and Parma. Note further the curious forms rivardavano and Vanni in xxv. 67 (where see note) and xxx. 32. Cha for casa (xxiv. 10) is familiar to us from the 'Ca d' 1 oro' 1 at Venice. 3. Under the head of the ' personal equation ' of the writer we may noteyf0 for infino (passim); su for suso (e.g. xxxii. 138; xxxiii. 90; xxxiv. 136, &c.); also a tendency to omit the definite article (e.g. di altri for degli altri\\\ x. 120). [This however is not nearly so common as in L, where it is very frequent] There is also a general tendency (still more marked in I) to omit a final / when following another vowel, e.g. oma for omat, cJii fu for cK io fui (xxxii. 68), mie for miei (e.g. mie desiri, x. 6), suo for suoi (e.g. tutli suo seguaci, x. 14), tu sara (iv. 15), po for poi (ix. 132, &c., &c.). 4. These may be taken as samples of the unintelligent blunders which sometimes occur, la tyopia = T Etiopia xxiv. 89, di Ritusa = d' Aretusa (cf. da Rethusa in B), xxv. 97, alquanto divenuti for al quarto dl venuti (possibly an oral blunder), xxxiii. 67. Occasionally careless omissions or inser- tions mutilate the rhythm of a line, e.g. ix. 60; xxii. 62; xxiii. 65; xxxiii. 36, &c. This however is much more frequent in some other MSS. The following are suggested as test-passages which may be useful in tracing relationships with other MSS. They are selected as exhibiting some marked or rare peculiarity of reading. In this and the other cases of MSS. collated throughout the Inferno in full, such lists might be very largely increased. Indeed in the case of some MSS. there is scarcely a page which does not yield some examples of unique or peculiar readings. Inf. i. 115; ii. 102; iv. io; v. 56, 142; viii. 52 ; ix. 9, 60; x. 84; xi. 55; xiv. 26, 48; xvi. 72; xviii. 135; xix. 117; xx. 17; xxviii. 36; xxxiii. 67, 89; xxxiv. 79. Purg. ii. 118; viii. 94; xix. 35; xxi. 19, 61; xxiv. 61 ; xxvii. 81. Par. ii. 42; xxii. 94; xxv. 14; xxvi. 104; xxviii. 23; xxix. 4. Under this head may be added cases in which lines have been omitted or transposed in error, as such errors are very likely to be reproduced 1 , and thus may afford indications of relationship. Such cases will be found at Inf. v. 97100; xiii. 71, 2 ; xxxii. 36, 7. Speaking generally, A has good readings and comparatively few arbitrary alterations of the text. 1 e.g. in ix. 80 A and E omit cos}, which was probably due to a clerical error in their copy or copies. In D it was also omitted at first, but inserted above by apparently the same hand. There are other clear indications of relationship in the case at any rate of A and E. 33 514 ACCOUNT OF THE MSS. B (Bat. 493). In the Bodleian Library (Canon. Ital. 109). This is a very beautiful MS. in folio on vellum (13x9^), with very large margin, as it has single columns and the writing though large and clear is much compressed. It appears to be late in the i4th century, or perhaps early in the I5th from a comparison with Bat. 432 (Paris), dated 1411. The titles of the Cantos are in brilliant red, and the initial letters (alternately blue and red as is very commonly the case) illuminated. This MS. has been sub- jected to very frequent erasure with a penknife, indeed on almost every page, and corrected readings are written over the erasures. This is espe- cially frequent when z (or f) occurs. Perhaps s was the original letter, since we have (uncorrected) such forms as pusso (ix. T,\),posso (xviii. 5), lassi (xv. 65), condission (xvi. 53), sossa (xvii. 7), stissosamente (viii. 83), guissavan (xix. 26), sensa (frequently), dinansi, &c. ; possibly sometimes it may have been /, since we find in viii. 56, seqq., satio, stratio and ringratio. Also, mio has often been altered to mi by the 8iop6a>rijs, e.g. xiii. 21, 75, &c. Compare su for sue, xx. 94. In many cases a dot has been placed under an erroneous letter, without further correction, as often in the case of ultra when alta should be read (e.g. xviii. 8). But besides the occa- sional confusion of altro and alto, the scribe has a curious habit of some- times writing altro thus, alt . So also diet for dietro, &c. For the title see Mortara's Catalogo, as also for a description of the 'Capitolo' of Jacopo, which it contains, together with the 'Proemto' to his Commentary, without however going on to the Commentary itself. Among peculiarities of orthography we may note : 1. Distinctively Latin forms are preferred, and unassimilated conso- nants generally, e.g. justitia, scripto, obscuro, factore, victoria, sospecto, sompno, dampnati, autumpno, and we also have such gratuitously unassi- milated combinations as Brunecto, ricecto (for ricetto, xvi. 102), mecte (xx. 76), sobmecton, &c. These however are not uncommon elsewhere. 2. There is a remarkable absence of distinctively dialectic forms. So far as I can trace any, they are Tuscan or Florentine, and the scribe was probably familiar (like the writer of Z) with the ordinary Florentine Italian of educated society. I have noted sprende for splende in x. 102, and aro for avro frequently. These are Tuscan forms. Also -evil for -evol, e.g. orrevil, salutevil, which again is Tuscan, and especially in the neighbourhood of Lucca. The archaic forms perdeo, sentlo, aprlo, &c. &c. are also, I believe, Tuscan. 3. Personal peculiarities of the writer again are not very marked the most so perhaps being the irregular observance of the second concord in respect of possessive pronouns e.g. suo parole (xvi. 72), suo cose (xxii. 51), suoi arti (xx. 86), and /'/ mie tesoro (xv. 119); indeed mie for mio is quite usual. EXAMINED OR COLLATED. $1$ Also three or four times he adopts the curious form ambur or ambor for ambo (e.g. xiii. 58 ; xix. 124; xxxiii. 58). This seems to be a Provengal form, and one which occurs frequently in the Chanson de Roland, &c. Speaking generally, this MS. seems to have remarkably good readings, and few variations, either careless or conjectural, while unintelligent blunders are scarcely to be found. The following are some of the very rare peculiar readings that I have been able to discover. Inf. iii. 41 ; iv. 45, 52 ; xiv. 83 ; xix. 36; xxvi. 57 ; xxvii. 37. Purg. vii. 15, 51 ; xiii. 2; xxi. 61 ; xxvii. 81. Par. xxvi. 136; xxviii. 23. These it will be seen are of very slight importance or significance, and this is evidence of the scrupulous accuracy of the scribe, as well as of the merit of the exemplar text. This MS. at present has Romani in Inf. xxviii. 10, but it is obviously a correction. I have gained a general impression that the text of B is related to that of Witte's A. This would be confirmed by the rarity of peculiar readings as above noticed, since Witte's text (from which variations are registered) is professedly based mainly upon this MS., which indeed Witte has often followed even when B, C, and D agree in supporting some different reading. C (Bat. 492). This is another beautifully written MS. in the Bodleian (Canon. Hal. 108) on vellum, described by Mortara as " of semi-Gothic character and I4th century." I should judge not later than 1380 from a comparison with one or two other dated MSS., and especially the Brit. Mus. 26836 (o). It is in double columns, the vellum is much discoloured, and there are curious pen and ink illustrative sketches at the bottom of the pages (size 14^ x 10). The portion from Inf. xxvii. 31 to the end of Canto xxxiv. has been written by a different scribe whose writing and ink are both inferior. The break occurs in the middle of a page. At the beginning of the Purg., the former scribe seems to have begun again with the old ink. Many of the corrections over erasures in the preceding Cantos seem to me to be in the hand- writing and ink of the second scribe. It ends with this colophon : Compiuto il paradiso di dante alleghieri di firenze. deo gratias. Amem. A mem. A mem. The MS., like B, has been subjected to a considerable amount of correction and erasure, sometimes with a view to the substitution of easier or more obvious expressions, e.g. di quei and da not for tra quei and per not in Inf. xiii. 26, 27, &c., &c. We note the following dialectic and other peculiarities, the former pointing obviously to a Venetian origin. 332 5l6 ACCOUNT OF THE MSS. (a) The reckless manner in which single and double consonants are interchanged is perhaps its most marked peculiarity. This prevails to a greater degree than in any other MS. I have met with. Take as an ex- ample Inf. iv. 93 Fanomi onore e de$o fano benne, and again popollo maligno (xv. 61), sono (iv. i, though sonno in previous line), dritOyfosi, eser,fato (xxiii. 74), dise, basa, pasa, coren (xv. 122), elleto, malle (ii. 16), parllar, nobelle, comfratti (for conflati or confrati, Par. xxxiii. 89), &c., &c. (/3) sc is regularly softened to s. Disese for discese (frequently), siencia, lasando for lasciando, creseranno, sender and siender for scender, sipa for scipa (vii. 21), basomi (viii. 44), sempio (x. 85), disente for discente (xi. 104), pesi for pesci (xi. 113), sema for scema (xii. 128), &c., &c. (y) A preference for e instead of i or a 1 , e.g. anema,femene (iv. 30), benegno, biasemo, serb, serai, &c. ; aV for a 7 / (passim], someteno (v. 39). (5) Among minor peculiarities (mostly, it will be seen, Venetian) may be added caxion for cagion, nut for not (frequent), como for come (frequent), fiol (iii. i2.\\Jiollo (vii. 61), aduto for adotto (xxxiii. 44). (Comp. perhaps aduna for adona in vi. 34.) As illustrating the 'personal equation' of the writer I have noted: (a) A tendency to omit the definite article and the copula (a tendency still more strongly marked in L), e.g. viii. 1 1 1 ; ix. 30, &c., &c. (/3) Frequent omission of io after quando, che, &c. (e.g. viii. 39, 108 ; xxxiv. 23, &c., &c.). (y) Quelli is generally substituted for egli. (8) Final m and n are constantly interchanged, e.g. Faren not a Chirom (xii. 65). Guardaci bem bem sem bem sem (sic), Purg. xxx. 73. () k is often written for ch, e.g. guarda keno meti (sic), (xiv. 73), kemi largisse (xiv. 92), kame = che a me (xiii. 83), so ki for chi (passim}. This is not uncommon elsewhere, and is an indication of archaic ortho- graphy, and according to Batines, of the first half of the I4th century (see n. p. 100). () Adverbs in mente are generally divided into two words and the latter part written mentre, thus canina mentre (vi. 14), egual mentre (vii. 76), simelle mentre, &c. (77) Ignorant and unintelligent blunders are very frequent indeed, in the matter of false concords, meaningless orthography, and absurd divisions of words. These are often so startling as to suggest even ignorance of the language on the part of the scribe, or else copying from dictation by a very ill-educated and unintelligent person. Among false concords take the following : lunga tema (iv. \^G),fangosi genti (viii. 59), le mure (viii. 78), la cilgia (iv. 130), la strida (v. 35), plena la pugna (vi. 26), lifronti(v\. 70), li maschili penne (xx. 45). 1 See under ' A ' supra, EXAMINED OR COLLATED. The writer evidently thought that anything terminating in a must be fern, sing., and in i, masc. plur., and made alterations to suit his theory ! The following are specimens of unintelligent divisions of words and other blunders suggesting, in some cases at least, a mere phonetic repro- duction of a misunderstood sound : Inf. ix. 75 -127 x. 77 97 129 xii. 51 - 93 xxi. 20 xxii. 85 128 xxxii. 134 fiume piu chieresi archi s' elli in quella sebe nodo a te da cui ci voile nesciamo ma quelle bolgie egliasolle cogliando sotto tu m' ingagni for fummo e piu. gli eresiarchi. s' elli (h)an quella. se ben odo. attendi qui. c' immolle. noi siamo. ma' che le bolle. e lasciolli. quegli ando sotto. tu ti mangi. These and many other like blunders, which occur frequently, strongly suggest the probability of the MS. having been copied orally. (See further note on Inf. xxii. 89.) The following may be useful as testing passages for comparison with other MSS. Inf. ii. 23 ; iii. 116, 121 ; vi. 48; xviii. 109 ; xix. 29 ; xx. 25 ; xxii. 69 xxxii. 62 ; xxxiv. 75, 85. Purg. iv. 72 ; x. 14 ; xxi. 61 ; xxii. 5 ; xxvi. 72. Par. i. 135 ; iii. 15 ; xxii. 17 ; xxvi. 136. In conclusion, I should judge that C had a very good foundation text, but a very unintelligent and ignorant scribe ; and further the MS. has been very freely erased and corrected. xiv. 66, 75, 104; xvi. 45 ; xvii. 14; ; xxiv. 83, 92 ; xxvi. 19; xxix. 132 ; D (Bat. 487). In the Bodleian (Canon. Ital. 97). This is a MS. on vellum in 4to (9 x 6) with single columns. The writing seems to me to be certainly early in the I5th century, but Mortara describes it as of the I4th. It has no title or colophon, and the Cantos also have no titles. The similes have been marked throughout, and the word copo usually added in the margin, apparently by the same hand, or at any rate an early one. The most curious point about this MS. is the evidence which it affords as to the authorship of the celebrated epitaph still existing on the tomb of Dante at Ravenna and traditionally assigned to himself, consisting of six rhyming hexameter lines, and commencing Jura Monarchic, Superos, Phlegetonta, Lacusque. 5l8 ACCOUNT OF THE MSS. They are given at the end of this MS., and distinctly attributed to one Dominus Bernardus de Canatro. A sonnet follows ' De laude dicti Domini Bernardi] and then another entitled 'Responsio dicti Domini Bernardi? See further Mortara's Catalogo p. in, and also a full dis- cussion of this and other epitaphs on the tomb of Dante in an article by myself in The English Historical Review. Following our usual plan we will indicate : (1) Dialectic peculiarities. These are not so strongly marked as in many other MSS., but seem to point to a Tuscan origin. Of course the peculiarities of this dialect are in any case not so strongly marked as those of the provinces more distant from the centre of Italian proper, many of what would otherwise be peculiarities of dialect being admitted as features of 'the language.' (a) aro, arei, arebbe, &c. are distinctively Tuscan forms which occur very commonly. (j8) The confusion of r and / points in the same direction, e.g. grolia for gloria (iii. 42; xxxi. 116); grorioso (xv. 56); obriando for obbliando (xxviii. 54); archemia for alchemia (xxix. 119, 137). Perhaps we may add intralle for entrarle in xxv. 112. Contradio for contrario (xxxii. 94) is also Florentine. Comp. conversely Gereon for Gedeon in Purg. xxiv. 125. (y) i is prefixed to words with extraordinary frequency (as also in *F'). This is, I believe, a Florentine habit. E.g. isvolvi, istrida, ischietti, istroscio, ispose, &c., &c. Every page affords numerous examples of this. (8) The forms mentlo, penteo, aprlo, &c. point in the same direction 1 . So also//fl/# for pieta (frequently). (2) Among 'personal peculiarities' are : (a) Constant interchange of m and , especially when they are final letters (compare 'C' in this respect), e.g. um poco (passim\ bem, falcom, facciam (xv. 73), Menbrotto (xxxi. 77), passavan (iv. 65), ponevan (vi. 35), &c. (in fact this form of i plur. is very common), uon (for uom), /' on for /' uom frequently. (/3) Final e for i : quince, qiiinde and ogne are very common forms. So gente, parte, frate frequently stand for the plural ; cf. esse for essi (iv. 62). Rarely the converse also is found (as might be expected sometimes, the interchange being once permitted), e.g. falconieri (xvii. 129), grandi (xxxi. 132). Similarly e for i in subj. tenses is usual ; e.g. iv. 64 (dtcesse); 71 (discemesse) ; vi. 45 (vedesse}\ &c., &c. (Consequently readings (such as conoscesti and conosceste in v. 120) which depend on the distinction between e and / are practically in- determinate in this MS.) . (y) There is a tendency to double certain consonants, especially after che, e.g. chellasiche lassi (xi. 18); chellattende (xi. 97); so/, /, and s are regularly doubled after che. 1 See supra under ' B ' and infra under ' G.' EXAMINED OR COLLATED. 519 (Consequently // is not conclusive evidence that the definite article, or a pronoun, is intended to be inserted.) (8) Another very singular peculiarity of orthography in this MS. is the use of c and g soft before a, o, and u 1 . Thus we findposca (xxx. 17), marca (xxx. 122), lanca &c. (xxxi. 2, 4, 6), facca &c. (xxxi. 44, 6, 8), oca = oncia (xxx. 83), &c., &c co = cib (xxx. 148), usco (xxxiii. 46), fasco (xxxi. 135). gu-giu (xxxi. 48, &c.), gochogoccio (xxix. \\2), gallagialla (xxxiv. 43), gunto=ginnto (xxxiv. 112). The text of this MS. is very highly (I think too highly) praised by Mortara, and the copious supply of 'belle variant? to which he refers is a questionable merit. In fact it seems to be only another way of saying that the writer has indulged in reckless license of conjecture, as the following samples will shew, where many lines will be found entirely re- constructed, often with an obvious motive, so that its list of 'peculiar passages' is exceptionally large. The following are examples : Inf. iii. 6 e '1 primo honore. (Cf. e '1 sommo honore in /3.) 85 Non sperate giammai veder. (So p.) iv. 59 Israel con Isaac. 141 Tullio e Merlino. ix. 72 le bestie e lo pastore. (The tame bestie being evidently thought to be in more natural connection with pastore than fiere would be.} xvi. 1 8 a lor che a te. (A curious and clumsy compliment to Dante, making nonsense of the passage.} xix. 117 Ch' ebbe date. (Apparently to indicate the voluntari- ness of the ' donation of Constantine?} xx. 8 piangendo e lacrimando. (piangendo perhaps thought more suitable than tacendo with lacrimando.) xxx. 86 dodici for undici. (A curious change for which I cannot see the reason, and which I have never been able to meet with elsewhere?) Purg.xxxii.io2 Di quella terra. We may add : Inf.\\. 139; iv. 27 ; xvii. 6 1, 1 20; xviii. 36, 1 1 2, 113; xxiii. 62, 63; xxiv. 14; xxv. 1 6, 74; xxvii. 46 ; xxxiv. 26, 68. Purg. ii. 13, 44; vii. 15, 51 ; xxi. 61 ; xxv. 31 ; xxx. 73. Par. vii. 21 ; xxii. 94; xxvi. 134; xxviii. 50; xxix. 4. There are also very many lines in which single words have been changed, e.g. vii. 21 vita for colpa, xviii. 31 canto for lato, xix. 93 disse for chiesc, xxii. 20 sommo for area, xxiv. 116 stordito for smarrito, xxvi. 25 al pocho for al POZZO, xxviii. 27 stereo for merda, xxxiii. 71 morir for cascar, and many more. 1 This peculiarity occurs chiefly, if not entirely, in the last few Cantos, in the case of this MS. It is found also in ' G,' and I noted gorno at Purg. vii. 69 in 'm.' 520 ACCOUNT OF THE MSS. (e) A few careless or unintelligent blunders may be noted, such as tuo stella (xv. 55), quello Eriton (ix. 23), d' un peccali (xv. 108), per le qua (ix. 1 19), donzella for dolcezza (xxvi. 94), farse innata for forsennata (xxx. 20). Unusual words and proper names seem to have given rise to strange blunders, as egli intropia for elitropia (xxiv. 93), e dare tu se (for e d'Aretusa !) (xxv. 97), quive teode (for ove or dove Eteocle) (xxvi. 54), chorde lero (for cordellierd) (xxvii. 67), alia terano (for al Laterand) (xxvii. 86). E (Bat. 489). In the Bodleian (Canon. Ital. 103). A MS. on paper, in large folio (16x11), of which the date is given in the following colophon, Deo gras Amen. Finite adi 15 Febrar 1443. (For title, &c., see Mortara, Catalogo, as before.) This MS. seems to have had a good foundation text, and has compara- tively few variants, except such as are due to careless and barbarous orthography, dialectic peculiarities, or want of intelligence on the part of the copyist. He does not however seem to have set himself (like the scribe of F, and to some extent those of D and I) to re-edit or amend his text. This MS. has a special interest from its containing the remarkable interpolation of 18 lines in Inf. xxxiii., first noticed in 1874 by Dr Greg. Palmieri [see Appendix ill]. I have suspected, more strongly than in almost any other MS., oral dictation in this case. This seems to be indicated by i. The very great frequency of dialectic forms and peculiarities, which would at least be likely to be multiplied by such a method. ii. Strange and unintelligent blunders in specially difficult passages, proper names, or unusual words, e.g. xx. 9; xxiv. 100; xxvii. 30, &c. iii. Lines are frequently defective from the omission of words, especially in the middle of the line, but without any indication of omis- sion, as if the copyist had not always kept up with the dictator, e.g. xix. 114; xx. 73; xxii. 33 (and very many more). iv. Extreme irregularity and variability of orthography. v. SonTe special blunders which seem most naturally explicable on the theory of dictation, e.g. vii. 121 de contristi for dicon tristi, x. 92 tonri a Firenza, xxiv. 89 la thiopia, xxv. 58 Ela era abbarbicata for Ellera abbar- bicata, xxviii. 122 Pur sol for Peso/, xxx. 49 del tuto for di liiito ; and see note on the curious blunder in iii. 117, &c., &c. The following are some of the obvious Venetian peculiarities. i. Neglect of double letters ; mose and moso, pasa, groso, speso, (era, baso, vale, tore, quela, ela, nula, belo,foso,poso, sasi, &c., &c., and on the other hand, oribilli, pianetta, turbatta, malle, loquella, futtura, cossa, EXAMINED OR COLLATED. 521 solla, golle (vi. 14), nillo (xxxiv. 45), tintta, allto, scalla, subitto, notte (xix. 1 1 8), velle (xxvii. 81). In xix. 62 colui and collui both occur, and in xxiv. 33 we read di ciappa in chiapa ! ii. Prevalence of soft sounds, such as (i) s for c, sc, g and z; (2) c for ch ; (3) z for c; (4) u for o ; (5) e for /. E.g. (i) sendt(vi. 87), lasati (x. 12), pest (xi. 1 1 3) and pessi (xxix. 84) (both for pesci), cose (for amor &c. The text has a relation (though more distant than that of F) to the Sienese type 1 . It has also a distinct relation to F, and though the text has been far less frequently tampered with than in that MS., it has a very large number of peculiar readings. It is, I think, undoubtedly of Florentine or Tuscan origin, e.g. 1. The form of the imperf. in o for a is very common, and so in the plur. ono for ano, e.g. seguivo (xvi. 91), aveo for aveva (vii. 36), discerneo, sapeo, stavo, &c., also gridavono, erono,facievono t mirono (iv. 133). This is good Tuscan. 2. So again ard, arei &c. for avrb &c. commonly. 3. It is, I believe, rather vulgar Florentine 2 to use suo or sua &c. indeclinably as is very common in this MS., -e.g. e ( = *) sua eterni danni (xv. 42). Cf. e sua omeri (xvii. 42), e suo compagni (xv. 102), le sua spalle (i. 1 6), suo -vista (i. 53), parenti mia (i. 68), suo figura^ tuo vita, tuo stella, sua foglie, gambe tua, sua meschite, mia mam, mia desiri, mia primi, &c., &c. Among other peculiarities, some perhaps dialectic and others per- sonal, may be noted : The forms drieto, drento, anbedne, duo? (generally for diie or dud) (Florentine?), contro a (for contro often). The masc. article is el in sing, and e instead of i for plural, regularly. The ist plural of verbs is written generally ano instead of amo. This may be only a special instance of the frequent interchange of n and m, e.g. linbo, onbra, tenpo, rinbonbo, tronba &c., &c. a occurs for e in several words, as piatosi, state, dispiatate. (This is, I believe, Tuscan. Comp. 'F'.) The past part, is generally used indeclinably with the auxiliary avere &c., e.g. viii. 120; ix. 96; xxix. 44. This writer pedantically alters /' occhio into gli occhi regularly. He generally writes words at full length, rhythm or metre notwithstanding, e.g. pare nel duolo (iii. 33), che gia /' usarono (viii. 125), vereno verrem (xiii. 103), &c. The peculiar readings are extremely numerous, and in very many cases the reason for the alteration can be easily traced, e.g. Inf. v. 1 13 pensier changed to sospiri to correspond with 1. 1 18. 1 See Witte, Dante Forsch. I. no. xvi. 2 See Blanc's Versuch, &c., p. 203. See also supra under G. 3 Compare G. a i. 34 530 ACCOUNT OF THE MSS. Inf. xix. 19 non sono molt' anni for non <<* (pedantic grammatical accuracy). xx. 72 lama, for riva /0 J/V /. 79. xxi. 28 fuggir (see /. 26) substituted for veder _/>w# /- apprehension of the sense. xxvii. 82 erapiacuto/^rpiaceva (# supposed more accurate tense]. xxx. 40 con padre for con esso (a needless gloss). xxxiii. 43 trapassava y^r appressava (evidently supposed to heighten the effect}. xxxiv. 42 sommofor loco (probably the same}. Purg.\\. 118 andavan (al. eravan) is altered to sedevan as more suitable to fissi ed attenti, and to the rebuke of Cato I. 1 20 &c. vii. 15 e nel minor s' appiglia. x. 74 Del Roman principe, il cui gran valore. xviii. 57 E del primo appetibile. We may add : Purg. ii. 13, 93; iv. 72; x. 30; xix. 34; xxi. 19; xxiv. 61 ; xxv. 131; xxvi. 41 ; xxx. 15. Par. i. 135; ii. 42; iii. 15; vii. 21 ; x. 112; xx. 117; xxvi. 136, &c. There are many as arbitrary as those of F ; e.g. Inf. ii. 86 pienemente^r brevemente. vi. 58 lo cominciai/0r lo gli risposi (comp. Rispose/or Disse viii. 20, and also gridava inserted in xiii. 118). ix. i vilta nel core/or v. di fuor. xiv. 65 null' altra penafor nullo martirio. xv. 122 il palio verdeybr il drappo verde. xvi. 125 la boccafor le labbra. xviii. 9 dieci part\/or dieci valli. xxiii. 83 innell' atto del viso (comp. 86). xxiv. ico presto for tosto. xxv. 113 bestia/0r fiera. Many similar arbitrary substitutions of words occur elsewhere. Lines or sentences reconstructed will be found at Inf. v. 91 ; vi. 88 (cf. F) ; xix. 115 7; xxi. 19, 130; xxviii. 8, 9, &c. The following are interesting and, I think, unique : Inf. iv. 141 Tullio e Plinio; Purg. xxv. 131 Corse Diana (see Scartazzini h. I. who mentions several Edd. and Commentators (modern) who adopt or defend this reading, but he gives no MS. authority for it 1 ). In Inf. xxiv. 12 the easy riguadagna is substituted for the more technical ringavagna, but I have not met with this in any other MS. This MS. seems occasionally to shew affinities with the Vatican family, ' Scarabelli quotes it from Cod. Lamli (A). EXAMINED OR COLLATED. 53! and it holds a sort of intermediate position between it and the Sienese, having some relationship to both groups. It is certainly closely related to the Turin MS. '86' (q. v.\ but un- fortunately, not having my notes from ' I ' with me at Turin, I was unable to compare them at all thoroughly. J (Bat. 494)- In the Bodleian Library (Canon. Ital. no). This is a comparatively modern MS. on paper (late i6th century according to Mortara, but I should have thought even later) which has not been collated because it is evidently subsequent to many printed editions, being regularly punctuated, either (as Mortara suggests) in preparation for the press, or perhaps as copied from a printed edition. I had noted from the comparison of a few test-passages its evident connection with the 'Vatican' family of MSS., but I have not thought it worth while to enter them in the published col- lations, having since found it already noted by the indefatigable Dr Witte that this MS. is an almost exact reproduction of Bembo's autograph MS. in the Vatican collection, and consequently of the first Aldine edition of 1502 (Proleg. p. xiii fin.). K (Bat. 496). In the Bodleian Library (Canon. Ital. ill), dated Nov. 7, 1445. This is a palimpsest (13! xg), the original MS. having contained some Latin work in prose written in double columns, and with somewhat frequent paragraphs commencing with red or blue letters which can be traced in several places. At fol. 55 the later scribe has accidentally turned over two pages, so that the old writing can be easily traced, though rubbed down. To judge from the following fragment, taken at random, the loss of the original would not seem to be serious : " Est autem idem homini animal gressibile bipes. Quare animal gressibile bipes est. Sed non propter hoc homo(?) inconveniens aliquod accidit. Non enim de animali gressibili bipes ponitur. Sic enim, &c. " Quam foetus tamen bipes. Aut et in concupiscentia. Non enim de appetitu ponitur delectatio sed de universe." The contents of the volume are of a miscellaneous character, a full account of which will be found in Mortara's catalogue. The following is a brief summary : 1. Four sonnets. 2. Canzone (with Commentary) of Antonio de' Beccari da Ferrara on the falsely reported death of Petrarch. 3. Reply of Petrarch to the same. 4. A collection of 60 anonymous Sonnets, Ballate, &c. 342 532 ACCOUNT OF THE MSS. 5. La Divina Comedia, which breaks off however at Par. xi. 123. 6. A few pages of anonymous moral aphorisms identified by Mortara as Trattato delle Virtu Morali di Graziuolo de' Bambagioli, Cancclliere di Bologna (who is also known as the author of an almost contemporary Commentary on the Comedia quoted by name in the Ottimo Commento}. There is evidence in the 'gatherings' of the parchment that the rest of the Divina Comedia is not lost but that it was left unfinished, since the leaves on which the last-named fragments are written are of one piece with those containing the last few Cantos written of the Paradiso. Appa- rently therefore the work of the scribe was by some means cut short, and another scribe finding a few blank pages already prepared filled them in with the fragments in question. These pages contain two dates, 1429, and (later on) 1409, which are (see below) earlier than that given in the colophon to the Purgatorio. It should be observed however that the Paradiso is by a different scribe, the writing being much larger and more clumsy, than both that which precedes and that which follows. At the end of the Purgatorio the following curious colophon occurs : MCCCCXXXX V. die septima novembr. Explicit purgatorius liber per blasium raguxen qui stetit in domo dni pauli lauredano ad laudem dei scriptor scripsisset. et melius si voluisset certe. This is (with the exception of L) incomparably the worst MS. in the Canonici collection, in respect of barbarous spelling and of the frequency and recklessness of its alterations. Also the Venetian dialect is more strongly marked than in any other, which accords with the indications of the colophon just quoted. We find in fact all the characteristic peculiarities of this dialect that have been already noted, in their most exaggerated form, such as i. Neglect of double letters, piettatte, afanatta, altto, alttro, schortte, ttalentto, ttiio, tolssi, ssocessor ( = successor, ii. 24), cole (i. 13), colo (viii. 43), and even ttera= terra in ii. 2. ii. The very common substitution of x for s\ z for c, ^and/; s for g and sc\ e.g. perigloxa, quaxi, rispoxe, pexi=pesi (vii. 27), mixere = misere (xiv. 41); fazzeface (i. 56), pazze=pace, zi=ri\ also si-ci (see note on xvii. 34), &c. ; ziaschuno, zoia (i. 78), zia =gia, zu =gtii, zielo=gielo, zenttil, zusso=giuso (xxii. 74, &c.), zientte = gente ; busiadri, rasione, adasia, mal-vassio, desiesi=discesi, usito = uscito, siencia = scienza. Even suoiior scuoi (xxii. 41), &c., &c. iii. There is also the usual Venetian dislike for the g sound, especially in combinations such as gl,gn,gu, &c. ; e.g. v is found for gu, as -vera= guerra (xxvii. 86; see note on xxv. 67) ; vardiguardi (xxxiii. 51). ogni is frequently written one, figlio appears as fio (iv. 56, &c.). Cf. fiol (iii. 121, &c.), and fioli (xxxiii. 38). scoglio is written schoio (xviii. ill). So also isgaiarda = sgagliarda (xxi. 27) ; simiiente = simigliante (xxx.. 147); maori=maggiori (xviii. 27, &c.). Comp. pare ior padre (xxvi. 95). On the other hand, aglia is found, in apparent spite of rhyme, for aia in xxi. 6o ? EXAMINED OR COLLATED. 533 but the evanescent dialectic pronunciation of the gl would satisfy the rhyme 1 . Nut and vui generally occur for not and voi. iv. Peculiarities which are perhaps rather personal than dialectic are cridare for gridare (passim} ; qui giu for quaggiu (passim] ; plu for piu (generally, but not quite always); and a curious tendency to substitute compounds with in for the simple verb, e.g. s' inpente (xxvii. 118); s' intrulla (xxviii. 24) ; s' inchiude (xxx. 27). On the other hand, fino (usually written find) is generally substituted for infino, and su for in su. Also tito, suo, &c. are used as it were indeclinably passim. (This is, I believe, rather Tuscan than Venetian.) Inf. v. 56 illustrates well the curious orthography, Che libittofe lecitto in suo lezie \ The following is a selection, for purposes of comparison with other MSS., out of a vast number of peculiar readings with which almost every page abounds : Inf. i. 122, 130 ; ii. 81, 93 ; iii. 28, 79; v. 94, 123 ; xxi. 60 ; xxii. 149; xxiii. 72; xxvi. 78 ; xxx. 51 ; xxxii. 62; xxxiii. 34, 59, 68. Purg. xviii. 83 ; xxvi. 72 ; xxx. 1 1 1 (secondo che le bestie !). The following may be added as cases where a change has been made in the supposed interest of clearness or simplification : Inf. viii. 102 ; x. 67; xiii. 92; xxi. 9, 57, &c., &c. The following lines or terzine are omitted altogether : Inf. ii. 112 114 ; viii. 49 51 ; x. 136. The relations of the text are most capricious. It seems connected strikingly with different MSS. in different Cantos, but with none, as far as I can trace, consistently, or for long together. L (Bat. 490). In the Bodleian Library (Canon. Hal. 104). A I5th century MS. on paper (15 x lof) inscribed 'di Piero di Nero 1591,' written with extreme carelessness. It often presents nothing but ' un accozzamento di parole vuote di senso ' (as Witte says of some other MSS., Proleg. p. Ixxiv), e.g. Inf. xxiv. 87 reads thus Si riduce incierti chonafi si bena ; xxiii. 117 Porte vemeni a martiri. Its spelling is most barbarous, proper names especially being in the most hopeless confusion 2 (e.g. ix. 46 ; xii. 134 5 ; xiv. 116; xxviii. 59, 89 ; xxix. 134 ; see Collations for these and many others). 1 See supra under 'A' (/3). 2 These blunders are so frequent and unintelligent (and the list might be very largely amplified by similar cases when any words are at all difficult 01 uncommon) as to suggest that the MS. was copied from dictation. See also supra under C and E. 534 ACCOUNT OF THE MSS. There are endless blunders of omission, causing many defective lines both as to sense and rhythm (e.g. i. 28 ; ix. 3, &c., &c.) ; also bar- barous linguistic blunders, as though the writer were unacquainted with the language, e.g. dela pianeta (i. 17), pel maculata (i. 33), furon che (xiv. 44; fuor having been apparently mistaken for the archaic form of furon}. Similarly in xv. 18 he has mistaken suol (verb) for sol (noun), and adding the article has written il sole \ See note on xiv. 44 for other similar blunders, and some further instances noted below. Inter alia the writer generally omits the article and the conjunction e; often too the verb ^, which he perhaps mistook for e, a word which is habitually neglected (e.g. xxiii. 99, &c., &c.). Rhymes are frequently violated by blunders or deliberate alterations, e.g. iv. 149 ; v. 86 ; xii. 84 ; xxiii. 8 ; xxviii. 86, &c. As it stands, the MS. has very little value or authority, though it repre- sents the mangled remains of what I should suppose to have been a good foundation-text, of which it occasionally preserves some rather rare good readings, and which seems often to exhibit a remarkable connexion with that of the celebrated Egerton MS. of the British Museum (' a')- The MS. is imperfect, as it breaks off at Par. xi. 120 (almost at the same point as K). Also Inf. Canto xviii. is altogether omitted, without any indication of omission. The MS. has no title. The dialectic varieties indicate a Tuscan origin : e.g. (1) rand / are interchanged, grolia (xxxi. 116, &c.); groria (iii. 42, &c.) ; glroria (Par. i. i) ; sprendore, risprendea, delitani=diretani (xxv. 55); Giuria= Giulia (iv. 128); obria-oblia (xi. 61) ; Pluto is written Pruto (vi. 115), and Plruto (vii. 2). (2) mza, sua, c., are used quasi-indeclinably. (Comp. ' I,' c.) (3) The initial i in such forms as istabilito, istrada, &c., occurs passim. (4) Such verbal forms as parlavi, tenevo, arei, occur occasionally. (5) I believe also that the following forms are Tuscan vulgarisms : grede, grudele, gruda (for crede, c.), ginchiato ( = cinghiato t iv. 107) ; piata or piatta=pieta, &c. The following are a few samples of the ignorance and barbarism of the orthography: nosstri pasi(m. 77); gguai(\\.\. 22); auntuonno (iii. 112), monlta (vi. 66) ; orro and orre (for oro and ore}. Some peculiarities of a more personal kind are the substitution of n for m, e.g. di Rona e di suo inpero (ii. 20) : neglect of final letters, not only vowels, as ave avea, c., but also /, n, r, c. ; i re (iv. 125) ; pe la (vi. 53) (the omission of n often leading to apparent false concords in verbs *) : furon or furono is always written for fur, even violating rhythm, and by curious blunders for fuor in xiv. 44 and even far in xvi. 22 : in su la is regularly substituted for sulla, sopra a and contra a for sopra and contra, and this even in violation of rhythm sometimes (e.g. 1 This habit has led him into the converse error of mistaking i for in and writing in accordingly in xvi. 77, and similarly in ii. 3 he took sol utto for so /' uno and this for son V uno and thus writes in full sono V uno. EXAMINED OR COLLATED. 535 v. 98) : che is generally substituted for chi or cui, qual and quel are inter- changed promiscuously, and generally the wrong word chosen (e.g. xiv. 68, &c.). occhio is usually changed to occhi [cf. ' I '], and so on. Besides the larger defects mentioned above, the following omissions occur: Inf. xx. 64 6; xxiii. 64 6, 127 9; xxv. 21 2; Purg. iii. 133 145. The lines Inf. xvii. 72 75 are transposed (see note /. c.). It would be equally endless and useless to give a list of the vagaries and peculiarities noted in this MS. The following is a small selection which may serve for establishing relationship (if any) : Inf. iii. 30 ; iv. 141 ; vii. 30 ; x. 27, 43; xvii. 78 ; xxii. 70 ; xxiii. 46 ; xxiv. 62, 3 ; xxv. 34 ; xxx. 83, 148 ; xxxiv. 48, 54. Purg. iv. 72; xiv. 126; xviii. 83 ; xxi. 19; xxiv. 61 ; xxvi. 72. Par. x. 112. Add Inf. xx. 24 bagnava per se stesso. (Is this perhaps an 'expur- gation ' ?) Purg. xxx. 112 Ma per grandezza. Several remarkable agreements will be noticed between 'L' and 'a,' whose texts seem to me undoubtedly related. In ii. 3 and xi. 56 they share in common blunders. In the later Cantos L and D seem to have some connexion. Very often too L seems to agree with F, and I think generally in places where F and I differ. The relation between M and L is also very noticeable, e.g. iii. 27 ; viii. 29, 63; ix. 59, &c. M (Bat. 498). In the Bodleian (Canon. Ital. 115). This is an early i5th century 1 MS. on paper (i \\ x 8J) containing the Inferno only, but it so closely resembles N in respect of paper, writing, arrangement of the Commentary, and the preliminary contents (see below), that it seems evident that they are parts of the same work, the intermediate volume of the Purgatorio being lost. The Commentary also itself in both volumes is, according to Mortara, an Italian version of that of Jacopo della Lana. This volume contains (i) A ' Capitolo'' in terza rima of Minghino da Mezzano, entitled Dno mengino mezzano super Infermtm. Mortara states that he was Canon of Ravenna and a friend, though much younger, of Dante himself 2 . (2) The Capitolo of Jacopo di Dante headed Compilatio tolius operis inferni. (3) Breve raccoglimento di cib che in se superficialmente contiene &c. beginning Nel mezzo del camin di nostra vita Smarito in una valle /' auctore Et la sua via da tre bestie impedita &c. 1 The writing is very like that of the dated (1408) MS., Bat. 261, described below as '13.' ' 2 Some further details about him will be found in Guerrini e Ricci, Polemiche Dantesche pp. 5, &c. He is there said to have been a pupil of Dante and a friend of Petrarch, Boccaccio, &c. 536 ACCOUNT OF THE MSS. commonly attributed to Boccaccio. (4) Comedia prima dil primo libra di Dante nomato clamato (sic) Inferno con le gloxe. These ' gloxe' are in fact the Commentary of Jacopo della Lana, and finally before the ist line of Inf. we have the following title : " Incomincia la comedia di dante alegieri de firenze nel quale tracta di le penne (sic) et punimenti di vicii et meriti et premii di le virtute. Canto primo basare, cortexe, giuxo, dixio (iii. 126); vixo ( = vtso),&c., &c. i and e are freely interchanged : e.g. dil=del (usually), riprisi, vinni, tinni,$titti, tenite (xxiii. 77); vira = vt:rra (xix. 77). On the other hand, el is usual for zY, and ;,F,r,iig.) Fesso. (So F.) gambe tirando collo trafito Fesso Rotto conversely Ciascuna substituted for Ognuna. (So F.) It need hardly be observed how conclusive is the evidence here afforded of the affinity of the texts of O and F, and this is a good illustration of the practical use of these tests of ' peculiar passages '. (See further evidence below.) Purg. xvi. 145 Cosi spari. (So only Bat. 428.) Par. ii. 124 Riguarda omai a me. (So F and 16. A sort of ' conflate' 1 reading.} Add perhaps Par. x. 112; xi. 26 ; xxvi. 1 34 ; xxxiii. 89. This, instead of being (as often) a small selection from a large number, is a nearly exhaustive list of peculiar variants of any importance that I have noticed. In Inf. xx. 9 a curious blunder occurs of a kind very rare in this MS. fanno le Tanie. This MS. has a few marginal variants apparently prima manu (mostly in the Purgatorid), some of them curious. The following are all that I noticed. Inf. xvi. 1 02 al. de per Emiliay^r dovea per mille. (So F.) xviii. 12 al. sicura. (So F.) xxiv. 69 al. ad ire. (So F.) Purg. i. 15 text Del mezzo. (So F.) Marg. Del acre. vii. 5 1 text o non sarria. (So F.) Marg. o ver sarria. 99 text molta. (So F.) Marg. monta. xi. 36 text uscir de le scalate. Marg. salire a le stellate. (So S, F.) 98 marg. del parlare. xiii. 1 54 text perderanno. (So F.) Marg. metteranno. xiv. 87 text has et divieto. (So F.) The marg. the curious confession "alibi non est 'et', sed hie posui quia est in capitulo sequente"! This refers to xv. 45 540 ACCOUNT OF THE MSS. where et is appropriate, while here it is clearly out of place. Purg. xxi. 6 1 text sol voler fa. Marg. solver si fa (an inferior variant}. (F has suo voler here.} xxvii. 4 text da nona. (So F.} Marg. da novo a riarse. xxix. 60 text vinte. (So F.) Marg. giunte. xxx. 15 text alleviando. (So F.} Marg. alleluiando. Par. xiv. 57 text che tutto di. (So F.} Marg. che tutto il di. xvi. 38 text tre. (So F.} Marg. trenta. xxi. 15 text misto. Marg. mesto. (So F.} xxx. 114 text Quanto da noi. Marg. Quanto di noi. (So F.} - 124 te.r/giglio. (So F.} Marg. giallo. In these cases it will be seen that sometimes the text of O and some- times its marginal reading corresponds with F, but generally one or the other. The title is Incipit comedia Dantis de Allegheriis Poete Florentini. Capitulum primum et que prohemizat ad totum opus. There is no colophon to Inf. or Purg., but at the end of Par. we read : Explicit comedia Dantis de Allegheriis de Florentia poete nobilis- simi. F (with which this MS. so often agrees) has no title or colophon. P. In the University Library, Cambridge (Mm. 2. 3), and forming one volume with the last MS. The two MSS. evidently, I think, belong to the same family, but the readings in this MS are much inferior to those of O, though of a very fair average value. It is on vellum, in bold Italian Gothic of the I4th century, in two columns, and after each Canto is the Commentary of Jacopo della Lana in smaller characters. The title is : Qui comincia la Comedia di Dante Alleghieride Firenze. Capitolo primo dillo Inferno. The colophon is : Qui finisce la prima parte di questa comedia di Dante Aldighieri defirenza. Sit laus et gloria Christo. Amen. The Purg. has a brief title and colophon, and the Par. a longer title (all printed by Barlow, Contributions &c. p. 67). There is a long colophon at the end of the Paradiso in which the anonymous author of the Commentary submits himself and all his opinions to the authority of the Church as follows : La sovra scripta expositione, giose (sic) o vero pastille io si o scripto secondo che a me minima intendente pare che fosse lo intellecto dillo autore. E percib ogni exemplo, argomento, opinione, conclusione, alle- goria, sententia o vero alcuno dicto che in essa s). o scripta, intesa, -vel signata, se ello si confirma e somiglia allo senso e allo tenore dilla sancta EXAMINED OR COLLATED. 541 madre Ecclesia Romana, approve affcrmo et o per bcne dicto. E si deviasse, discrcpasse, o vero contradicesse allo predicto senso e tenore dilla prcdicta sancta Ecclesia io si o per vano e per nan bene decto. E perb il casso et evacuo e tegnollo per da nessuno valore. S\ come Cristiano puro fedele e verace che Credo in una sancta Trinitate Gr'c. The dialectic peculiarities are much more strongly marked than in O, and leave no doubt of the Venetian or at any rate Lombardo- Venetian origin of the MS. o and u are frequently interchanged e.g. sul for so/, urecchie regularly for orecchie, humeri (xvn. 42), tumultte, spirittt. The converse change however is more common, e.g. fogir {=fuggir}, ponir, crocciata, ponto (=puntd), crodel (3 or 4 times), roscello, toa, soa, &c. fuoco is generally written fuogo. Comp. fatigoso and fatiga. So again we have such forms as lassato ( = lasciato) and cosse ( = cosce). Dil and dillo (passim], andfo for/u are also, I believe, Lomb.-Ven. peculiarities. The form of the future &c. in arb (for ero) is very common, e.g. parlarb, usarei, -vedarai, &c., &c. and this, according to Gaiter, // dialetto Veneto nel secolo di Dante, p. 9, is a Venetian peculiarity. Speaking of a Venetian Chronicler of the I4th century he says, "Alia prima conjugazione trascina quasi tutti i verbi, come allora usarono i dialetti Veneti." Under the head of 'personal equation' of the writer I would note : (r) the general substitution of fra for tra. (2) his fondness for regularly inserting the pronoun io, e.g. Quand' io, Vid' io, &c. (compare H and others). (3) a curious dislike for tt, which seems invariably to be written ct, e.g. socto,fiorecti, quactio, Scocto (xx. n6\pocti=pote' (viii. 112), dirocti (ii. 86), and even checti=che ti (ii. 135). (4) the frequent false concords are curious, especially in the earlier Cantos, e.g. degli altre (i. 9), ver- gognoso fronte (i. 81), le dispietati (i. 114), degli altra (iii. 113), le strali (xxix. 44), &c. There are a great many petty alterations of single words and changes of order in the supposed interests of simplicity, euphony or rhythm : e.g. Inf. i. 108 ; iii. 89, 109 ; iv. 45, 59, 114 ; v. 65 ; vi. 72, 105, 108 ; viii. 31, 102 ; xi. 103 ; xiv. 70 ; xviii. 63 ; xxv. 16, 27, 131 ; xxxiii. 15, 47, 71. Purg. vii. 51 ; xviii. 83; xix. 35 ; xxvi. 72. Par. xxvi. 136. It has already been noted that O and P sometimes exhibit striking agreements ; and also that several readings are attested by E, H, P in common. Q. In the University Library, Cambridge (G. g). A I5th century MS. on vellum in two different hands, one beautifully clear Italian Gothic (closely resembling *, r, and m) on the first two pages of the Inferno, and also on pp. 15 and 16, the explanation being, as pointed out to me by 54 2 ACCOUNT OF THE MSS. Mr Bradshaw, that these were the two outside sheets of the first 'gathering' of the parchment. These were left to the professional illu- minator by whom no doubt the elaborate and heavily gilded illuminations of the first page were executed. The same occurs at the beginning of the Purg., where the first page of the Purg. and the last page of the Inf., forming one leaf, are similarly executed, and so is the other outside leaf of the 'gathering', containing Inf. xxxiii. 94 153. At the beginning of the Par. the professional scribe appears to have written only the first 2 J lines. The rest is written by a very inferior artist, and abounds with careless blunders of all kinds, unintelligent divisions of words, omissions of letters, words and lines, though such omissions are generally supplied above or in margin, and often prima mantt. The text is however at the bottom a good one, and exhibits remarkably few variations from Witte's standard text. I think that O, P and Q belong to the same class or family, though the relation is closest between O and P. The Venetian dialect is very strongly marked. We may note : 1. Interchange of u and 0, and especially fondness for u, e.g. ripu- sato, mustrando, luco ( loco), buno ( = bond), cului, busco,Jigliolu : also hi for lo occurs very frequently, e.g. lu fundo (xviii. 109), lu puco (iii. 75), pututo ( potutd), cruciatu. Per contra sometimes o for u, e.g. jodicio, ponito. 2. The parasitic insertion of u before o, so very remarkable in n, and also in s (both by Betinus de Pilis), is here very common, e.g.puoco, fuosso (xviii. 112), cuosa, puoi (=poz\ suon = son, tesuoro, cuost, puosa, puolo, nuoi (nut is also common) ; and even diwro for d' oro in xix. 112. 3. Interchange of e and /, e.g. vinendo, -vignon, udirimo,vidisti(-&.\\\\. 77), tivo = tevd (vi. 38), vincirb, ctnno, medismo ; even pina=piena (ix. 1 1 1), viro = verro. U often occurs for le (sometimes in O also), e.g. // bructe (xiii. 10), li fronde (xiv. 2), li volsi (xv. 52), U piace (xv. 96). So quil, quillo, quisto, quilgli, &c. are common. i before e is commonly omitted (comp. M), e.g. lumera, vohtnteri, pensero, &c. 4. Another curious Venetian peculiarity is the evanescence of g (though in many of the following cases g has been inserted secunda manu). Thus we have ia=gia, iu=gtu, ir=gir, taccio, irando (iii. 52), iunto (vii. 34, &c.), verra-=guerra (xxviii. 10). Conversely we have gire for ire in xxxi. 124, being in fact of no importance in pronunciation. See above on A (p. 512). The forms bangia (ix. 1 14), volgle, melglo, elgle, casione, spresgiando, rasgionam, fresgi, &c., &c., and megio = mezzo in xvii. 83 may also be noted. 5. Among minor indications we may note dote, doy and duoi (for due\ ambedoy, the fut. indie, in arb (cf. P), &c. It may be added that Latin and unassimilated forms of words are very common, indicating an archaic diction. The writer constantly interchanges datf and delF, and frequently EXAMINED OR COLLATED. 543 inserts an initial h, as Horfeo, Habraam, Horlando (xxxi. 18), &c. He also uses rather violent contractions, which sometimes render the reading difficult to determine, e.g. mo=mondo, modo, or moto\ ch = chi or che, na = na/ura, qu=quando, ee = esser, &c. The MS. also abounds with erasures and corrections ; e.g. a 8top$wr^f has systematically altered the archaic forms of the sing, such as guei, questi, into quel, questo ; he has inserted the initial g (see above) ; and has also made occasional conjectural emendations of text, as xv. 65 6, xix. 46, xxii. 12. After the Paradiso is the colophon : Explicit tcrtia et ultima Cantica comedia (sic) Dantis. Amen. In the epitaph of Dante which follows we read in line 5, patrius (sic) exter* ab oris (i.e. exterus, as in U, e, m, and several other MSS.). Then follow the well-known Capitoli of Jacopo and Busone. Then (in a much later and comparatively modern hand), Tre Sonetti di Piero Alighieri fightolo di Dante che sono come argomento di tutte le tre parti del la sua Comedia. Then 11 hexameter lines, and afterwards this curious entry " 1447- 17. Januarii. D ns Bianzardinus et Cola de Castillione in avinione (Avignon) dormierunt simul et emerunt aromata, &c." Also in the margin of the hexameter verses we read " Urbanus Papa V pro magno munere misit Imperatori unum agnum cum his versibus." The following is a selection of some of the peculiar or characteristic readings of this MS., most of them obviously conjectural emendations : Inf. i. 29, 37, 63; iii. 31, 112; iv. 73, IO3 1 ; xii. 125 ; xxi. 64 ; xxvi. 122, 131 ; xxix. 137; xxxiii. 90; xxxiv. 8. Par. iii. 15; x. 112; xviii. 75. There are also a great many instances of single words being changed into others either nearly synonymous [e.g. gridb tor par Id (xiv. 6i},grandi for alti (xii. 102), giri for volgi (xxx. 86), &c., &c.] or such as are supposed to be more appropriate to the sense [e.g. viii. 9 ; ix. 7 ; xxv. 19, 62 ; xxx. 129, &c., &c.]. The text (as in some other cases already noted) has varying and capricious relationships to different MSS. in turn. R (Bat. 246). This very valuable MS. is in the Ambrosian Library at Milan. It is beautifully written in clear and very upright Gothic, reminding me (speak- ing from memory) of the British Museum MS. ' a ' (943 Egerton). This would accord with the estimate of Batines that it belongs to the middle, or possibly even to the earlier half, of the i4th century. As to the orthography, it abounds with the unassimilated forms common to all the oldest MSS., e.g. tucto, actese, aspecta, Neptuno, descripto, Piectola, &c., &c. 1 A similar substitution of the gerundial construction occurs in viii. 44. 544 ACCOUNT OF THE MSS. It is written in two columns, one containing the text, and the other Latin notes. A note on the initial fly-sheet (quoted at length by Batines), written in 1609, gives an account of the former owners of this MS. which is described as " diligentissime scriptus et notis antiquioribus tllustratus." The latter statement is supported by the note on the well-known passage Inf. xiii. 146, u ubi est caput statue /"(?) ydoli Matis" which would indicate a date before 1333, at any rate for the original composition of the note. Between the Inf. and the Purg. we have a page of medical prescriptions, chiefly of 'Emplastra? Of one, after minute directions for its composition, it is stated " -valet ad sanandum omnes plagas novas et antiquas, &c., &c et valet etiam ad infinita alia" ! Then we have the prescriptions for Emplm dyaquilon magnum and Emplm dyaquilon parvum, and several others. The following readings may be noticed : Inf. i. 28 ; iv. 141 ; xii. 16, 66 ; xiv. 126 ; xv. 39 ; xvi. 3, 14; xviii. 12 : xix. 3 ; xxviii. 10. Purg. iv. 22 ; x. 30 ; xxvii. 81 ; xxx. 73. The agreements in several of these passages with 'L' are noticeable. The general character of the readings seems to me to be decidedly good, but there are several instances of distinctly secondary readings. S (Bat. 249). In the Ambrosian Library at Milan. A sort of 4to, on vellum, its writing good, clear and very upright, though somewhat rounded, and with rather long tails and flourishes ; something like Bodleian C in this respect, but decidedly later, probably (I should think) late in the I4th century. It has no notes, summaries of Cantos, or colophon. The readings call for no special remark. The following are peculiar, and as obviously bad, being clearly of a conjectural type : Inf. xiii. 20; xiv. 126; xv. 39; xxix. 46; xxx. 18; Purg. xviii. 83; Par. i. 135 ; xvii. 9 ; xxvi. 96 ; xxxiii. 89. In Par. xi. 91 we have ^dritta intentione.' (So F, b, 1.) In reference to the connexion between this MS. and the Holkham MS. 't,' see under the latter. T (Bat. 251). A very beautiful MS. in the Brera at Milan, described, not without reason, by Batines as " stupendamente bello." The vellum is very fine and white, and in this respect, as well as in the extreme beauty of the writing, and also in the peculiar long tails to such letters as b, h, d, /, g,f, s, &c., it resembles the Bodleian ' C,' Lord Ashburnham's f, and the British Museum 'o.' I should suppose all these MSS. to be about 1370^80. The EXAMINED OR COLLATED. 545 following modern MS. note on the title-page assigns (though without any reasons alleged by which such a judgment could be tested) a much earlier, and indeed (as I should have thought) an impossibly early date. Dovrebbe essere scritto da Ser Nardo da Barberino a giudizio del bibliqfilo M. Girolamo d y Adra. Del suddetto Amanuense altro codice esiste nel Bibl. Trivulz. e precisamente del 1337. Una della Laurenziana pub essere (why?) del medcsimo Ser Nardo e del 1347. Questo dei? essere o perb del tempo di mezzo, o in qual sia caso d' intorno alle due date. (signed) L. L. (? Longari). I must say the writing of this MS. is remarkably like that of the MS. which bears the date of 1337 (r). The initial letter of each terzina has a dash of thin gilt or silver, like the Bodleian D. The following readings are distinctly secondary and inferior, though the text seems generally a very correct one. Inf. i. 28 ; xxx. 18 ; xxxii. 47 ; Purg. xviii. 83; Par. x. 112, 119, &c. I observe that in all except the last of these passages X (which is very similar in writing) agrees with T. Following up this clue, I find in my collations very remarkable evidence of the singularly close relationship of these two MSS. Unless I am mis- taken, there are only three passages out of the whole number (170 to 180) in which these MSS. are registered as differing (viz. Par. x. 119; xvii. 9 and xxv. 29) ; and in all these cases the difference is very slight. Both MSS. are quite remarkably free from ' peculiar readings,' and this, it need hardly be said, is a point in favour of the purity of the text. In Par. x. 1 12 they agree in the case of a curious blunder. U (Bat. 252). A large folio MS. on vellum, in the Brera Library at Milan. On the fly-sheet is a strange note signed ' L. Scarabelli'' "Questo codice e fratello carnale dell' altro in Brera segnato AN. xv. 17" [i.e. the one last described (T)]. To this is appended the remark "per carattere no di certo. L. Longari" The fact is that it is not the least like in writing. This MS. is more in the style of the sample pages of the Cambridge MS. ' Q,' though more beautiful and (as I should judge) older. The characters are clear, large and upright. Later on however, on fol. 41 (i.e. from Purg. vi. 42), both the ink and the scribe are different and much inferior. Nor, as far as my evidence goes, is there any trace of relationship between the texts of the two MSS. The following passages may be worth noticing : Inf. ii. 55 quanto la stella. (So *, i, iv.) iv. 49 alcun huom per suo merto. (So $.) x. 65 gia di costor si eletto. Add Purg. xix. 34 ; Par. xxviii. 23. In the well-known lines "Jtira monarchic?" &c. which follow the poem we find the reading " patriis exterus ab horis" (sic) in 1. 5. D. i. 35 546 ACCOUNT OF THE MSS. V 1 (Bat. I84) 1 . In Lord Vernon's Library at Sudbury Hall (No. 2243 m Catalogue), but formerly in the Library of the Marchese Rinuccini, under which head it is fully described by Batines II. p. 100, to which reference should be made. It is on vellum (iof xyj), and the date is probably about 1400. The orthography is archaic in respect of unassimilated letters, e.g. subcedecte, scripto, Sogdoma, dilectione, &c., &c. The following are peculiarly arbitrary and audacious changes : Purg. \v. 72 La qual non seppe. (So however Z and several others.} vi. 48 libero e felice. xxxi. 43 piit vergogna. (So Z, d, and three or four others.} Add Purg. ii. 13 ; Par. vii. 21 ; xvii. 9; xxvi. 136. In Inf. ii. 12 V 1 has ' Anzi deW altro passo'; and in Purg, xx. 104, both V 1 and Z have ' et latro parricida.' This MS. is interesting as exhibiting the "Vatican" text (see Appendix n.) very strongly, and as being especially in close relation to Z. V 2 (Bat. 183). Also now in Lord Vernon's Library (No. 2242) and formerly in that of the Marchese Rinuccini. It is a folio (12^x9^) on paper. The character is Gothic, and the MS. is judged by Batines to be about the middle of the I4th century or even earlier, though here, and sometimes elsewhere, I am inclined to think his dates too early. He cites in proof the frequent use of k for c or ch. This however is fallacious, since the orthography in this and other respects might be faithfully reproduced, even though obsolete, in a later MS. The colophon is Explicit paradiso e purghatorio e ninferno di Dante Allighieri di firezze (sic). Deo gratia. Then follows the Capitolo of Jacopo. I examined this MS. some years ago and very imperfectly, so that I do not feel able to speak as to the character of its text. On the last page, in an old but much later hand, occurs this note "questo Dante e fato ant 100 per mano di oratio ri- nuca'nt." I noticed l alla perduta gente' in Inf. iii. 3, which is, I believe, quite peculiar; and ' Poi si restrinser' in 1. 106 of the same Canto is very rare. (So B, 6, and i.) 1 In the case of V" 1 , V 2 , V 3 , V 4 and a few others (such as g 2 and n 1 in the Paradiso) the figure when thus added is not to be confused with the figures i or 2 subscript, by which readings i'"" or i tuia manu are always indicated. EXAMINED OR COLLATED. 547 V s *. In Lord Vernon's Library (No. 244). It is a sort of 8vo. (9^ x 6^) on vellum, and contains the Inferno only. The first page is very elaborately illuminated. On a fly-sheet opposite in large but plain letters and in different colours " Incomincia la prima Comedia de Dante dove esso tracta delle pene de lo inferno. Capitolo primo? The name of Albizi appears below with a coat of arms. There are small but beautifully illuminated letters to each Canto. The writing is singular and artificial, but beautifully clear, being in imitation of printed Roman characters, (comp. d, h, i, &c.), and is pro- bably not earlier than the middle of I5th century. Colophon: Explicit pars prima operis dantis poete Florentini: que vocatur Inferni : Deo gracias. Then there is a little slip of vellum pasted on with the date MCCCXXVll. This is more modern than the preceding colophon, and certainly of no authority, for of course the date is quite an impossible one for the MS, by a century or more. Perhaps, as in one or two other cases which will be found noticed, a *c' may have been omitted. It abounds with unassimilated forms e.g. rocta, condocta, &c. In iii. 112 there is a curious blunder (evidently ocular and not oral in this case), " Como d' aventino? V 4 (Bat. 191). In Lord Vernon's Library (no Catalogue number). A small volume (65x4^) on vellum. It is clearly written, but in a roundish hand which becomes more sloping as it proceeds, and is probably somewhat late in the 1 5th century. It is defective, beginning at Inf. i. 28 (Pot cK ebbi &c.) and ending with Purg. xxxi. 42 (taglio la ruotd). There are a great many marginal variants in a small and neat hand very similar to that of the text, but in much smaller characters. It is also full of corrections in the text itself in darker ink by a late BiopdaTJs : some very poor, e.g. that in iii. 38 cK a dio furon ribelli. Some of the marginal notes are of the nature of glosses : e.g. iii. 60 (il gran rifiutd) the marg. has "di Giuda" ; and just below in iii. 62 the marg. has the equally mal a propos note "z traditori" In ii. 53 the donna beata e bella is explained to be La Teologza, and this note is repeated at ii. 85. It is further noticeable that in most, if not all, of its leaves it is a palimpsest. The original is generally quite obliterated, but on fol. 36 it is obvious that the original was Latin and in a strong clear Gothic hand. It also appears that it is not in its original form, a larger MS. having been cut up into these small sheets. 1 I have not been able to identify this MS. in Batines. 352 548 ACCOUNT OF THE MSS. The few specimens of orthography noted (so far as they go) point to a Venetian origin ; e.g. lassiare, suspiri, cului, profundo, &c., xonno ( = sono) &c. The following are noticeable readings : Inf. iii. 20 Con lieto -visa. (So Z.) xxiv. 12 Poi riede a la speranza e ringavagna. Add Inf. v. 38 ; Purg. xxx. 73. w. This is a very beautiful I4th century MS. on vellum, in excellent pre- servation and with singularly black ink. The size is 13^x9^, and it is in two columns. It was until lately in the Library of Thenford House, near Banbury, and through the kindness of the late Mrs Severne I was allowed to consult this MS. and also that of the Commentary of Serravalle, which will be found described elsewhere. This very valuable Library was collected by Mr Wodhull 1 , who has entered on the fly-sheet of this MS. that it was bought by him on Ap. 26, 1799 for f i \. 3^. (!) at the sale of G. Mason, Esq. At the beginning of each part is an illuminated letter. That to the Inferno contains Dante seated writing at a desk, his cap and robe being blue and the sleeves and collar red. At the beginning of the Purgatorio Virgil and Dante are seated in a boat. At the beginning of the Paradise we have Beatrice addressing Dante, who stands below her with his arms reverentially folded. The writing is very beautiful, but of a type not uncommon. It often seems very much to resemble T and X. The text is also closely related to that of these MSS. The title is Qui comincia la prima parte de tucta la Chommedia di dante Alleghieri di firen$e. Et prima inchomincia lo primo canto della prima Commedia del primo libra dello inferno nel quale elgli fae prohemio di tucta V opera sua, come qui di sotto si contiene per loinfrascripto ordine e Capitoli, e pa (i.e. prima}. Colophon Explicit liber totius voluminis Dantis Allegherii. In- ferni, Purgatorii, et Paradisi. Deo gratias. Amen. As to the orthography, there seem to be no strongly-marked dialectic peculiarities. The archaic ke for che occurs throughout, and both Latin and unassimilated forms of words are common. The text is about an average one, containing some undoubtedly inferior readings [e.g. Inf. xv. 39; xxx. 18; xxxii. 47; Purg. xviii. 83; Par. xvi. 47, &c.]. I think there is evidence of some relationship between T, W, X, Tr and /ti, which it might be worth while to investigate further, but that between the two MSS. T and X (as already pointed out), is much closer than that of any of the others ; ' Tr ' especially having several peculiar 1 This Library was sold by auction in Jan. 1886, and this MS. and that of the Commentary of Serravalle were happily secured for the British Museum. EXAMINED OR COLLATED. 549 readings of its own, and occasionally parting from the group in other cases also. In Par. xviii. 75 W, Tr and /x have the common error of alta for altraj and in Par. x. 112 T, W, X and p. have the common blunder, nella nea mente. The following are also noticeable in view of this relationship, Inf. i. 28; xxx. 18; xxxii. 47; Purg. ii. 44; xviii. 83. These are more peculiar or distinctive : Purg. xxiv. 65 fan nel acre schiera (W only?). In Par. xx. 14 the unintelligible word sfilli occurs. X (Bat. 484). An extremely beautiful MS. in Eton College Library. It is a large folio (i 5f x 1 1 1) on vellum, with very clear and beautiful writing resembling C, o, , &c., and therefore I should suppose about 137080. It has fine illuminations on the first page of each Cantica, rubricated introductory notes to each Canto, and illuminated initial letters. It is in fine preserva- tion, and has (I think) no notes, corrections, or erasures. There is nothing marked about the orthography to indicate dialect. With the exception of // (for which ct is generally found), double letters are generally assimi- lated ; but such forms as facto, dilectoso, Burnecto, decto, ecterno, succedecte, &c. are common. I noted also ke for che occasionally. The lines Inf. xxvi. 136 8 were omitted at first, and are added by a later hand in the margin. The readings appear to call for no special remark, and to be of an average and ordinary type, except for the obvious relation existing between this MS. and T and W and to some extent Tr also 1 . In Inf. iii. 7 Dinanzi a not is unusual (so EHW/ir 15). The title is : Incominda la commedia di dante Alleghieri di fire$e (sic) nella quale tracta delle pene e punimenti de' vitii et de* meriti et premii delle virtu. Canto I dela prima parte la qual si chiama inferno nel qual I' autore fa prohemio ad tucta /' opera. No colophons whatever occur to any of the three Cantiche, Y. This is a 4to MS. on paper (n|x8J) with 251 pages, bought a few years ago by the Curators of the Taylor Institution, Oxford. It evidently belonged to the Library of Pio VI. (Braschi), whose well-known arms (which formed the subject of one of Pasquin's Epigrams) appear on the binding. It contains the Paradiso only, accompanied by the Commentary of Francesco da Buti. It is probable that it had companion volumes for the /;// and Purg. These may still exist elsewhere, the Library of 1 See supra under T and W. 550 ACCOUNT OF THE MSS. Pio VI. having been dispersed by the French in 1798 or 1799*- It has two illuminated and heavily-gilded initial letters. There is no title, or colophon 2 . I should judge it to be early i5th century. The text stands in very remarkable affinity with that of the Bodleian ' I.' Out of 53 passages examined I found agreements in 46 and differences only in 7. The latter are comparatively slight, whereas many of the former are very significant, e.g. Par. vii. 4 rota for nota (rare). xi. 105 Tornossi (rare}. xiv. 72 cosa for vista (see sup. p. 464). xx. 117 loco for gioco (see sup. p. 472). xxviii. 23 A lo cinger (see sup. p. 493). Add (see collations) Par. i. 135 ; vii. 21 ; x. 112. ' z. This is a very beautiful little MS. on vellum, of a rather unusual form (6| x 3). The MS. is in excellent preservation throughout, and apparently dates from about 1400, possibly 10 years earlier or later. It belongs to myself, and was purchased by me from the Fratelli Bocca of Rome in 1880. I have not been able to identify it with any MS. in the Catalogue of Colomb de Batines, or to trace its previous history. 1 See also below under the description of ' 7.' 2 Since writing the above I find that though there is no separate colophon, the concluding words of the Commentary are as follows: " Et qni si finisce el chanto 33 della terza Chanticha della chomedia di dante Aldighieri et la sua lectura facta per me franchescho di bartolo da btiti et chompiuta el di della festa di santo barnaba, cib e a di n di giungno nel 1395. Indictione sechunda. Della qualecosa io rendo divoiamente qtianto piu posso allo omnipotente idio patre figlizwlo et spirito sancto, et alia gloriosa Virgine et Madre del nostro singnore Jesu Christo, Madonna sancta Maria, et al prefato appostolo et alia chorte tucta del paradiso gratie immense et debite. A quali sia sempre honore et gloria per infinita sechula sechulorum. Amen. These words might lead us to suppose that the MS. was the autograph of the Commentator, but they may only be a literal copy of what the amanuensis had before him. And this seems probable because a somewhat similar colophon is quoted by Batines n. p. 326 as occurring in a MS. in the Laurentian Library, but with the erroneous date, 1495. So again another occurs in a Magliabecchian Codex \Palch I. 29] containing nearly the same words (see Bat. II. p. 58) and the date 1385, with the mistake of substituting S. Bartolommeo for S. Barnaba, though the nth June is mentioned as the day of completion. It is singular that in the MS. in the Riccardiana (1006 8) forming the basis of the Edition of this commentary published at Pisa in 1858, S. Bernardo is erroneously substituted for S. Barnaba, though the date of n June, 1395 is given as in the Oxford MS., and the whole colophon corresponds verbatim in the two MSS. with the single further exception of maestro (for the word me) in the Florentine MS. [This statement is made on the authority of the printed Edition referred to.] EXAMINED OR COLLATED. 551 The only title is Comedia seu cantica prima incipit: and the Paradiso has the title Paradisi tertia cantica seu comedia feliciter incipit. (The Purgatorio has no title.) The colophons are as follows : Inferno. Clarissimi poete florentini prima explicit comedia inferni. Incipit secunda feliciter. Purgatorio. Dantis alighierii (sic) feliciter Explicit secunda Cantica seu comedia purgatorii. Xo (?) gratias. Paradiso. Clarissimi poete Dantis Alighieri florentinique tertia et ultima cantica explicit. The writing, which is very beautiful, clear and regular, is rather un- common in character, and appears (in the opinion of more competent judges than myself) to be that of an educated man rather than that of a professional scribe, though his emendations shew him to have been singularly devoid of taste. This is also consistent with the fact that there is a marked absence of vulgarisms, dialectic forms 1 , or unintelligent blunders. I noted however a certain number of clerical errors, e.g. vii. 128 ; xix. 123 ; xx. no; xxix. 79 ; xxx. 74, 78 ; xxxii. 76, &c. (see the collations for these). It abounds with Latin and unassimilated forms of words ; an indication of a text archaic in date or at any rate in type, e.g. homo, dilectoso, ad for a, and et for e (both passim\ affection, exilio, nocte, dextra, Captolica (xxviii. 80), &c., &c. This MS. has a special interest from the obvious relation of its text to that of Witte's B. It shares the merits and defects of that valuable and interesting MS. This text is a somewhat typical one, since it appears, at any rate in some of its most characteristic deviations from the textus receptus, in a sufficient number of MSS. to constitute a ' family.' (This is discussed at length in Appendix II.) Z seems also to have a few significant agreements with 'I,' e.g. x. 89; xxvii. 8 ; xxix. 54. I forwarded the late Dr Witte a long list of readings from this MS., from an examination of which he expressed a very high opinion of the value and excellence of the text. The following readings seem noteworthy (beyond the remarkable coincidences with B already referred to), many of them displaying an un- warrantable freedom of conjectural, and sometimes very injudicious, alteration, seriously lowering the value of a very good foundation-text 2 . Inf. vii. 119 al fondo. (A reckless alteration violating rhyme, appa- rently to suit sotto 1' acqua in previous line.} xi. 84 piu biasimo. (Obviously wrong, but why altered?} xiv. 36 si spegnesse. (So f.) 1 El is used for ei regularly (e.g. xxix. 24, 34, 36, &c., &c.). I observed arei once (xv. 60). Maiorica occurs in xxviii. 82, and razo hrraggio in xxxiii. 55. Also j bitrchi for i burchi (xvii. 19). I have scarcely observed any other unusual forms throughout. 3 See on this Proleg. pp. xxii, xxiii. 552 ACCOUNT OF THE MSS. Inf. xiv. 124 ch' e lungo e tondo. (Similarly in I. 138 we find ^ pentuta e rimossa.') xxi. 113 Mille dugento un. (A curious change found also in M, where see note, and also above p. 332.) xxvi. 120 Ma per aver, (f has per usar here.} -125 al nostro volo. (A very tasteless change?) xxx. 98 Sinon, credo, da Troia. (Query, did the scribe suppose ' Greco da Troia' to be an impossible combination?} xxxi. 47 e degli occhi gran parte. (A very feeble change.} Purg.i.iK) alia solinga strada. (But solinga occurs in the previous line also.} xx. 104 et latro parricida. (So V 1 .) xxiii. 37 gik era un animal. (A curious blunder} xxv. i Hora era del salir ne volea storpio. -52 In una fatta. xxix. 67 L' acqua prendeami. (The readings mi prendea, mi pendea, mi pendeva, as well as imprendea, &c. occur not unfrequently , and D has pendeami J .) Par. iii. 47 ben mi riguarda. (So ' 10.' B has 'ben ti riguarda.') Add further: Inf. ii. 20; iii. 20, 72 ; x. 85, 103; xii. 57; xiii. 91, 145; xiv. 75 ; xv. 85, 87 ; xviii. 78 ; xix. 54 ; xxi. 50 ; xxiii. 44, 73 ; xxiv. 59 ; xxvi. 117; xxviii. 8; xxxiv. 73; Purg. iv. 72; x. 30; xx. 90; xxii. 5; Par. x. 112 ; xiii. 27 ; xxii. 94; xxvi. 136; xxix. 4; xxxiii. 57. Gl (Bat. 500). A folio MS. on paper in the Hunterian Museum in the University of Glasgow. It is very clearly and well written, and its date I should think is early in the I5th century. It has very few corrections either in the text or margin; the general average of the readings appears good. Colophon : Qui finis ce el paradise che fecie dante Allighieri (sic) di firenze scritto per me montuccio di franchiescho difirenze. The scribe has a tendency to insert n constantly, e.g. anme a me. So also ongni, insengna, vengno, ingnudi, c., &c. The Florentine form grolia occurs in Par. \. i, but is not usual. There appear to be very few peculiar readings, but the following may perhaps be noted : Inf. viii. 125 ; Purg. xxiv. 61. Also at Inf. i. 28, while the text reads ' Cont io posato un pocoj we have a marginal note, ' al. Poi posato ebbi? In Purg. xiv. 133 we have the very unusual reading 'qualunque mat prende.' 1 The lect. vulg. is splendeva or splendea, and it seems possible that the dialectic (Tuscan) confusion of / and r may have led to the interchange of splendere and prcndere. EXAMINED OR COLLATED. 553 Tr (Bat. 283). This MS. is preserved in the Bibl. Municipale of Treviso. It is in 410 on vellum, with splendid illuminations. The writing is very upright, but somewhat rounded and free, letters like f, s, I, h, &c. having very long tails, the letters are very close together, and the words also generally. The parchment is somewhat coarse and the pen thick, so that the effect is rather clumsy. Speaking from memory, I should say the writing resembled that of Brit. Mus. 3488 Harl. (referred to as 'c'), and I should judge it to be of about the same date, viz. c. 1360 70. Great interest attaches to this MS. from the connexion of the family of Dante with Treviso, and the fact that the MS. is, so to speak, an in- digenous one, i.e. the dialectic peculiarities shew it to have been written on the spot or in the neighbourhood. Batines (quoting the Proleg. of the Udine Edition) mentions that Dante's family began to reside at Treviso before 1326, and that in 1391 Lorenzo, son of Simon Alighieri, was inscribed among the nobili of Treviso, as appears by the records still preserved there. Hence it has been conjectured that this MS. may not improbably have belonged to the family of Dante in the I4th century. In respect of orthography, indifference as to double consonants [e.g. pallude, collni, luccano ( = lucevano ii. 55), cossl, pillose, ducca, pollidorre (xxx. 1 8), possa (Purg. xxvii. 81), vappori, tebbe (=Thebe xxxiii. 89), giellata; and on the other hand, freda, note ( = notte), gtto, tuta ( = tutta), temeti (xxi. 93), &c., &c.] and the frequency of such forms as puoco, puose, volglia, sengni, ongni and onni (ogni], cum ( = con\ &c., &c. point to a Venetian origin. Viviani in his Proleg. to the Udine edition says that "alcune voci provinciate Trevigiane che si trovano in esso" still further localise its origin. The verification of this would require more minute knowledge than I possess, and a more thorough examination of the MS. than I was able to carry out, and unfortunately Viviani gives no examples of these alleged peculiarities. The following readings may be noted : Inf. ii. 55 ; xvii. 19 ; xxi. 93 ; xxvi. 25 ; Purg. vii. 51 ; xxi. 19 ; xxv. 9; Par. xvii. 9; xxix. 4. Inf. xxii. 64 reads thus : 'Lo ducca donque del gli altri rii,' and in Inf. xxxiii. 78 we have ' Che foro 1' osso ' (omitted supra p. 364). It is by no means free from secondary and inferior readings, though, as Dr Barlow says, it appears to be generally a good text. The terzina Inf. vi. 46 8 was at first omitted in error ; and in x. 69 name occurs in error (o/notoreXeura) for lome. The MS. commences with the Capitolo of Jacopo. [The MSS. denoted by italic capital letters are those which I have not examined myself. My information is derived from various published 554 ACCOUNT OF THE MSS. editions or monographs referred to in each case in the description of the MS. In several cases the information is very casual and imperfect, and extends only to a portion (sometimes a small portion) of the whole list of selected passages. In other cases it is quite complete as ABCDEFSUV\ A BCD. The four MSS. included in Witte's Edition are sufficiently described by himself in his Prolegomeni, and also by Colomb de Batines. A few general remarks and points of comparison may be added. As far as I can judge from applying the tests afforded by the most important of the selected passages and from other considerations, I should consider A as being certainly the highest authority, C being not very far behind it, and both decidedly superior to the other two. B and D not unfrequently agree and have, I think, a distinct though distant relationship. They both abound with startling variations, often in my judgment bearing on the face of them a secondary and conjectural aspect. Any one who will take the trouble to go through the marginal variants given in Witte's 4to edition which are marked as peculiar to B only or to D only (or not unfrequently to .5 and Z? together) will I am sure be convinced of the conjectural origin of a large number of them. Many, especially in the case of B\ consist in the substitution of single words of a more common-place and obvious character for the alternatives given in the text. But after deducting or allowing for the numerous alterations of this kind, I believe B will be found to have a purer foundation-text than D ; i.e. in test-passages properly so called, individual aberrations apart, D more often exhibits a secondary type of readings. Indeed if abstraction be made of the class of passages referred to, I should consider the foundation-text of B as little if at all inferior to that of C. It is perhaps interesting to note that B was the original from which Bembo copied (with alterations) in his own hand Cod. Vat. 3197 {Bat. 341), from which again the celebrated first Aldine edition (1502) is an exact copy [see Batines, II. p. 180; Witte Proleg. p. xiii]. In regard to A In addition to Witte's account of this MS., a full dis- cussion of it will be found in Dionisi's Aneddoti V. c. vi xi (pp. 43 66). He also, like Witte, considers it the most valuable MS. in existence, though repudiating the tradition that it is by the hand of Filippo Villani himself, and acknowledging (p. 51) that the pretended date of the colophon (1343) is an error for 1443. He traces 3 or 4 hands at work in various ways in the MS. The bulk of the work is, he thinks, in the writing of a skilful pro- fessional hand, not that of a private scholar (p. 52), and the scribe was most careful and conscientious in the reproduction of his original. Later hands however have introduced numerous corrections and erasures, some marginal and interlinear glosses, punctuation and accents. In- 1 Mussafia (Sul Testo ddla D. C. p. 5) says of B " che a lungo godette molta rinomanza, ma e certo ben lontano dal meritarne tanta." EXAMINED OR COLLATED. 555 dications of fidelity to an archaic text 1 are found by Dionisi in (i) words being written in full, though the metre requires an abbreviated pro- nunciation : e.g. Ambo le mani per lo dolor mi morsi (Inf. xxxiii. 58), where a 8iopdo>T^s has put a mark under the i of mani. Again, Ond' io per lo tuo meglio penso e discerno (Inf. i. 112), where similarly the last four letters of meglio are dotted below. (2) The forms of proper names, which tend to preserve the Latin type such as Dante would himself use in prose as well as in verse e.g. Ipocras, Flegeton, &c., Enrico (not Arrigd]. See Par. xvii. 82; xxx. 137. Add Fesule Par. xv. 126; xvi. 122. (See further my Proleg. p. xii.) (3) The preservation of archaic words or forms of words (inter alia, auct. Dionisi p. 46, we never find dietro or di dietro, but always retro or di retro in this MS.), and also archaic syntax, e.g. compie for compier in Inf. xxi. 114. In reference to the peculiar text of B, the archetypal MS. of the 'Vatican' family (as I have called it), a fuller discussion will be found in App. II. E (Bat. 215). This is the celebrated MS. in the Biblioteca dell' Accademia Etrusca at Cortona. It is fully described by Colomb de Batines, to whose work reference may be made. It is also the subject of an elaborate monograph by Don Agramante Lorini, the Librarian of the Museo dell' Accademia Etrusca (Cortona 1847) in which he has registered its variants from the edition of the Quattro Fiorentini of 1837. To this source I owe the in- formation embodied in my collations, which I have been able to make tolerably complete, assuming that the MS. has the readings of the edition cited when Lorini does not note to the contrary. Lorini (I cannot but think) much exaggerates the antiquity of the MS., if one may judge from the few lines given by him in facsimile. He would fain place it about 1330, and some enthusiastic admirers have even supposed it to have been written in the lifetime of Dante. Colomb de Batines (n. p. 112) thinks that it is not earlier at any rate than 1350. The following readings are noticeable, and in many cases not only unique, but bearing such obvious marks of audacity in conjectural emen- dation as to damage seriously the character of the MS., whatever its age. Inf. v. 1 20 Che conducesse a dubbiosi martiri(!). xiv. 36 Piu si stringea. xxv. 1 6 El si parti. (So D, P.) Purg. xxx. 74 di cedere al monte. 112 Ma per altezza. (Comp. L, which has grand ezza.) Par. iv. 140 tanto divini. xvi. 37 Al sol leone. 1 " Fra tutte le copie del divino Poema la piu antica e piu tenace della lingua Dantesca" (p. 43). 556 ACCOUNT OF THE MSS. Add further : Inf. i. 28, 42, 48 ; ii. 55 ; xiii. 20 ; xvii. 95 ; xxiii. 83 ; xxx. 18 ; xxxiii. 26, 72 ; xxxiv. 53 ; Purg. xxvi. 41 ; Par. vii. 114; xvii. 9; xxii. 17, 95. F (Bat. 409). This is the Codice Cassinense, which was magnificently edited in folio by the Monks of Monte Cassino in 1865. In the elaborate Prolegomeni its probable date is discussed at great length and with vast erudition. The result is that it is probably about the middle of the i4th century. This is confirmed by the internal evidence of some of the marginal notes. These are by various hands, one at least of which is judged to be contemporary with the text. Among other indications of date (for which see the Prole- gomeni in question) is that of Purg. xx. 69 (p. xii) where the ' Chiosatore ' says that the bones of S. Thomas Aquinas are still in the Badia of Fossanova where he died. As they were translated to Tolosa in 1368, the note was anterior to that date. Other indications are noted to shew that these Chiose were before Benvenuto da Imola (c. 1379). Colomb de Batines however describes the MS. as of the I5th century. Others have endeavoured to claim for it as early a date as 1328. The following are a few peculiarities of orthography noted, cz is generally written for 22, e.g. meczo, driczo, chiarecza, spcczava, &c., and ct for tt, e.g. tucto, pecto,factendo, tracto, &c. The following forms are also common busgiardo, rasgion, casgion, menusgia, indusgia, &c., &c. [which the author of the Prolegomeni says are Florentine (?)]. Cusl, cului, cum, possato, &c. would rather suggest N. Italy. So also would iunse for giunse, which I noticed in Par. xvii. 94 (see supra under Q). The following readings are peculiar or noticeable : Purg. xxiv. 65 in aera fan no riga. (Unique (?) and violating rhyme. Apparently due to an unlucky reminiscence of Inf. v. 47.) Par. xi. 91 dritta intentione. (Very rare, only S, b, 1.) 105 Rendesi. (Do., only , t, 47.) xvi. 38 E tre fiate. (So Q and 119. Comp. 68.) Add further : Inf. vii. 108 ; xx. 69 ; xxviii. 10 ; Purg. xxvi. 41 ; Par. i. 135 ; ii. 124 ; xiii. 27 ; xxvi. 136 ; xxviii. 23. The following, though not uncommon, are distinctly inferior, and tend to discredit the very exaggerated value that has sometimes been assigned to this MS. (as is also the case with some of those mentioned above) : Inf. ix. 54 ; xxi. 46 ; xxvi. 25 ; xxxiv. 53 ; Purg. iv. 72, c. But I am far from denying that its text has a very high character, as has also that of O, to which, as I have already pointed out, it has a remarkable affinity, though they are by no means identical. (See supra under ' O.') EXAMINED OR COLLATED. 557 677/7. These four MSS. are at Padua. They are described by Batines under the numbers 279, 280, 281 and 282 respectively, to whose description reference can be made. They are also described by Sicca, Rivista delle varie Lezioni della Div. Com. (Padua, 1832), to which work I am indebted for the references here made. I strongly suspect Sicca's accuracy, partly because I find him sometimes differing from Witte and Scartazzini in his record of the reading of a MS., and partly because in some cases a reading extremely rare elsewhere is registered as existing in a far larger number of his MSS. than seems at all probable on general principles of proportion. The passages selected by Sicca seem very arbitrarily chosen, many of great interest being unnoticed, and per contra a large number of trivial and not in any way significant variants being recorded, e.g. scura and oscura ; gaetta and gajetta ; cff io perdei and ctt ? perdei ; and even such as are quite indeterminate, as cK ivi trovai and cK z' vi trovai, &c., &c., and so on on every page. Another defect is that even in the passages selected the number of MSS. quoted is most capricious and variable. The colophon of / is curious in describing Dante as "poeta egregius orator lucidissimus? K (Bat. 307). This MS. belongs to the Biblioteca Florio at Udine. It is praised very highly by Batines for the beauty of its execution and the excellence of its readings. He says that it is in frequent agreement with the more cele- brated Codice Bartoliniano at Udine, denoted in these collations as U. I have not noticed any remarkable agreement between them myself, but the references that I have been able to make are very scanty, and such as they are they are given on the authority of Sicca, " Rivista? &c. Peculiar readings will be found noted at Purg. iv. 22 ; ix. 17 ; xxii. 6. Also at Inf. xxxiv. 50 we have 'in su lanciava.' (So J/ and U.} L (Bat. 293). The passages collated from this MS., marked " Cod. Marc. No. LV.," in the Library of S. Mark at Venice, were kindly obtained for me by the late Mr Shute of Christ Church, Oxford. The MS. is interesting as being one of the 26 MSS. selected by Dr Witte (Proleg. p. Ixxvi) as having, in his opinion, a special value and importance, thus forming his provisional ' ist Class,' though he told me himself that he had since found reason to be dissatisfied with that selection, especially in the way of omission. The MS. also exhibits (like several others of those in Dr Witte's ist Class), in a very marked degree the peculiar readings of the ' Vatican ' type of text throughout all the three Canliche (see Appendix n.). 558 ACCOUNT OF THE MSS. We may note as peculiar : Inf. xvii. 96 Con le braccia mi prese. xxvii. 65 Non torno mat alcun. xxviii. 71 E cui vidi in su. M (Bat. 227). This is the celebrated MS. in the Ducal Library at Modena, registered No. vill. G. 6, and commonly known as the ' Codice Estense ' par excel- lence. The readings recorded were supplied to me by the kindness and courtesy of Dr Antonio Cappelli, the Vice-Bibliotecario. I must confess myself somewhat disappointed in the character of the text as judged by these selections. It has a considerable number of distinctly inferior readings, such as will be found atfnf. iii. 31, 36; vi. 43; xxx. 18 ; Purg. iv. 72; x. 30, &c., as well as a very peculiar version of Par. iv. 121, 'Non 6 la voce mia tanto profonda' (!) The following are also noticeable : In Inf. xii. 125 it has (with y only as far as I know) the order pur cocea. In Par. xiii. 27 we have the rather rare (and doubtless false) reading sus/anzia, and in Par. x. 4 instead of the usual o per loco we have o per occhio (not /' occhio). In Inf. xxviii. 135 we find the probably correct and not very common reading ' re giovine ' (sic). Add Par. xxvi. 136; xxix. 4. N (Bat. 534). A MS. in the Royal Library at Copenhagen (" Thottske Samling, 411 fol."). The Library being closed on the occasion of my visit, I was unable to collate the two MSS. there, but I have since obtained a full description of them, and the collation of more than 100 selected passages, through the kindness of Mr York Powell of Christ Church. His name is a suffi- cient guarantee for the accuracy and completeness of the work done. As however these two MSS. were not personally inspected by myself, I have registered them (as in other cases) by italic capital letters, viz. A 7 ' and O. This MS. is in folio on vellum, the writing clear and beautiful, with commentary surrounding the text. This Commentary is stated by Batines to be that of Jacopo della Lana, and this also appears to 1 be the case from an extract kindly made by Mr York Powell. It came to the Royal Library in 1 704 with the rest of the collection of Dr Thott. A fac- simile of a line would seem to indicate a Gothic hand of the I4th century. The readings appear to be of an ordinary type. The only passages that I have noted as peculiar are Inf. ii. 22; xii. 125; Purg. iv. 72. O. This MS., like the last, is in the Royal Library in Copenhagen, and came to it in the I7th century ; the first description of it occurring in Petals Scauerinus Designatio Librorum...i665 Hafniae, Christ. Wering EXAMINED OR COLLATED. 559 Acad. Typog. p. 368. An earlier owner appears from a note on the fly-sheet to have been one Giuliano di Pergolo. A facsimile of a line seems to indicate a I5th century origin. A brief commentary has been added by a later hand. This however gradually gets less, and ceases altogether in the Paradiso. The readings seem to be of an average kind, but though generally agreeing with those of the last MS. (AT), it is frequently superior when they differ, e.g. Inf. i. 42 ; ii. 22, 81 ; Hi. 36 ; xiv. 126 ; xvii. 95 ; xxxiv. 13 ; Purg. xiii. 2. On the other hand, N has the advantage in Inf. ix. 70 ; Purg. v. 88 ; vii. 51 ; Par. x. 119. Both are wrong (but O worse) in Purg. iv. 72. As in the case of N, my information about this MS. is due to the kind assistance of Mr York Powell. . P. This letter refers to the " Frammenti Palatini " commencing at Par. x. 31, and including (according to Witte Proleg. p. Iv), 3240 verses, i.e. only about j 3 .j of the Divina Commedia. It is generally known as " II Quinterno" the name given to it by Borghini, who first drew attention to its import- ance. Signer Palermo published it in 1860, and claimed to recognize in it the handwriting of Petrarch, and also judged the date of it to be about 1360. Witte Proleg. (I. c.) praises very highly the correctness of the text, but denies the probability of its having been written or annotated by Petrarch. Scartazzini, without giving a positive decision, inclines to agree with Witte (see note on Parad. x. 31). See, for a fuller discussion, Witte Dante Forsch. n. p. 356, &c. "Abdruck der Palatinischen Paradieses- Fragmente." Q (Bat. 407). This is the celebrated Codice Filippino in the Library of the Bibl. Oratoriana at Naples. A minute account of it will be found in the splendid reprint of the Codice Cassinense (1865) pp. 583 &c. The variations of this MS. are registered in the notes to the text of the Cod. Cassinense, but in many cases not sufficiently clearly for me to be able to determine with certainty the readings intended. Still the list of variants which I have recorded is tolerably complete, and I owe them to this source, as I do also the following details extracted from the description above referred to of the MS. It is interesting to note that it was acquired by the Fathers when they purchased the Library and many archaeological treasures belonging to Giuseppe Valletta on the recom- mendation of the celebrated Giambattista Vico, in 1726. Batines states that there are two Codici in this Library, but I can find no trace of any but this one, Batines having been probably misled by a statement that Valletta's Library once contained two MSS. This MS. contains a great number of interlinear and marginal chiose in Latin, but in a different 560 ACCOUNT OF THE MSS. hand and ink from the text. Two leaves of the Purg. are missing, viz. from xxviii. 124 to xxix. 106 ind. At the end of the Purg. are the two well known Epitaphs on Dante beginning " Theologus Dantes" (14 lines) and "Jura monarchies'''' (6 lines). After the Paradiso there is a curious epitaph which I have not met with elsewhere, and the author of the description of the MS. from which I quote states that he has found it here only. It consists of two lines as follows : Comicus hie Dantes jacet Excelsusque poeta Non solum Comes Satirus liricusque tragedus. (!) The date of the MS. is estimated by Sig. Mandarini, the Librarian and author of the note from which I quote, as being about 1350. Witte includes this as one of his 'ist Class' MSS. The readings I have been able to trace seem on the whole perhaps good, but I feel that I have scarcely the materials for forming a decided opinion. In Par. ii. 124 " Riguarda bene omai" seems quite peculiar. 5 (Bat. 523). This is the MS. in the Royal Library at Stuttgard, which together with V forms the subject of Mussafia's very careful monograph published at Vienna 1865, and deservedly commended by Witte. Mussafia thinks it was written in North Italy, though its dialectic peculiarities are not so strongly marked as those of V. Both he and Batines assign it to the i4th century. According to Dr Witte (quoted by Batines II. p. 271) it is "di lezione non troppo corretta." For a full account of the MS. the reader is referred to Mussafia's monograph, to which I owe the information con- cerning its readings which is embodied in these collations. Also I have followed Mussafia in the two or three cases in which his citations differ from those of Scartazzini. The following seem peculiar : Inf. ii. 15 Fu ed ando. x. 39 Le tue parole. xx. 75 fiume per diversi. xxvi. 64 S' io posso. (Comp. Se posso in M.) xxvii. 46 nuovo e '1 vecchio. (So D.) xxxiv. 50 in su lanciava. (So K and U.} Par. ii. 115 ha tutte virtute. The following may also be worth referring to : Inf. iii. 114; Purg. vii. 51 ; xxiv. 6 1 ; Par. i. 135 ; ii. 141 ; xviii. 75 ; xxvi. 136; xxix. 4. These may be largely added to from Mussafia, but the peculiar readings in 5 are much less numerous than those in V. My impression is that though Fand .5" are related, as Mussafia points out with ample and curious illustrations (p. 14), and both belong as I should say to the Vatican (B] rather than the Florentine (A) type [see however below (4)], yet EXAMINED OR COLLATED. 561 (1) Fseems to be more closely related to />'than S is [e.g. such passages as Inf. xi. 90; xii. 125 ; xviii. 42; xix. 64 ; xxix. 48 ; xxxiii. 26, &c., &c.]. (2) V has many more distinctly inferior readings than S, and many more that are peculiar to itself, as though the scribe had indulged in much latitude of conjectural emendation (see passages collected under V}. (3) (As observed by Mussafia) the scribe of 5 was much better educated than that of V. (4) The connexion of 6" with the Vatican type seems limited to the Purgatorio, in part of which it is rather close and striking, though it is curious that even here in the first dozen Cantos there seems no trace of it. It is also singular that '54' in this respect corresponds in all cases with V, and in nearly all with S. V has scarcely any (perhaps two or three, and those by no means very distinctive) of the Vatican readings in the Inferno, and S, I think, none at all. In the Paradiso both MSS. seem again out of all relation to this family. U (Bat. 306). This is the celebrated ' Codicc Bartoliniano'' preserved at Udine, and reprinted by Quirico Viviani at Udine in 1823!. He claims for it a very high antiquity, and also a special value from the alleged sojourn of Dante at Udine under the protection of Pagan o Torriano, one of the Patriarchs of Aquileia, for which Viviani in his preface cites various documentary authorities. He observes that it contains many words of the dialect of Friuli clearly indicating its local origin, but unfortunately he does not give a single instance of this 2 . Witte (Prolog, p. xl) considers the MS. to be a decidedly good one, but its original text is very frequently obscured by the numerous alterations and erasures to which it has been subjected. The following are rare and generally speaking inferior readings : Inf. i. 28; xv. 39 ; xix. 64; xxvii. 21 ; xxviii. 10 ; Purg. ii. 13, 93 ; xxii. 6 ; xxiv. 61 ; Par. vii. 114 ; xxvi. 134. The following are especially noteworthy : Inf. xxxiv. 50 in su lanciava. {So A', A'.) Purg. ix. 17 Men piii. (Sec on tliis Wittc Prolog, p. Ixvi and supra/. 385.) xxi. 25 Ma perche Lachcsi. (On Ihis sec ]Vittc Proleg. p. xli ; and also tlic discussion supra, p. 399.) In Pnrg. xxxiii. 107 we have schiera for gentc, which I have only found besides in 11, out of about 100 MSS. examined here. 1 Witte, Proleg. pp. xxxviii, &c., expresses a very unfavourable opinion of Viviani's accuracy in this work. - See a similar case supra under ' Tr.' D. I. 36 562 ACCOUNT OF THE MSS. V (Bat. 522). The MS. at Vienna known as the ' Eugenian ' and fully described and collated by Mussafia in his monograph already referred to under S. Its Venetian dialectic peculiarities are very marked, as may be seen at once from the specimens given by Mussafia, p. 7, and the scribe was, as the same scholar observes, rather ignorant and uneducated. I owe the refer- ences in this MS. to Mussafia's work, and, as in similar cases, I have assumed the agreement of the MS. with the standard text, when no varia- tion has been noted by Mussafia. The following seem peculiar, and may no doubt be largely added to from Mussafia's monograph by anyone wishing to compare this MS. with others : Inf. x. 85 lo strazio grande e scempio. 136 spander suo lezzo. xi. 95 la dove dici che s' offende. (So D, K, Q.) xii. 92 si malvagia strada. xiii. 147 Rimase ancor. xviii. 1 10 L' occhio a veder. (Cf. A, I.) xix. 89 Con questo metro. xxiii. 98 su per le. (So C, G, Q.) xxvi. 31 D' ardenti fiamme. xxxiv. 50 suso alzava. (Compare several variants registered in collations. See also under U.) 118 Qui e doman quando la giu e sera. Add (from the collations supra) Inf. xvii. 95 ; xix. 23; xxi. 112; xxviii. 10 ; Par. xviii. 75. T (Bat. 257). A very beautiful and celebrated MS. in the Library of the Marchese Trivulzio at Milan, through whose courtesy I was allowed to consult this and the other MSS. of this magnificent collection, which appears to possess (since the dispersion of Lord Ashburnham's Library) the largest number of MSS. of Dante in any private Library now existing. This MS. bears the date of 1337, the genuineness of which has not, as far as I know, been disputed. For the distinction of being the oldest dated MS. in existence it has a rival in the celebrated Codice Landi at Piacenza (A), dated 1336. (The genuineness of the date 1335 in one of Lord Ashburnham's MSS. is very much disputed, and moreover it does not occur as a regular colophon but on a fly-sheet.) There is no doubt at all, I think, in this case that the colophon (for which see Batines 1 1. p. 138) is by the same hand and of the same date as EXAMINED OR COLLATED. 563 the text, and if the MS. be later, the statement of the colophon must be deliberately false 1 . This MS. is most exquisitely written, somewhat resembling the Mala- spina MS. in Lord Ashburnham's collection, but is even more beautiful. The initial letters of the terzine are lightly touched (as frequently else- where) with yellow. I did not notice any varieties of a dialectic character in the ortho- graphy, but it has the usual archaic characteristics [e.g. racto, puncto, mactino, quacto (quatto), actese (attese), Piectold\, except that che, not ke, is found. The readings are very remarkably good, and I think decidedly better than those of its rival at Piacenza. It has however a few which I think can be safely described as secondary, e.g. Inf. xx. 69; xxxii. 47; Purg, iv. 125; xxv. 9; xxvii. 4. In Inf. xxi. 113 (auct. Scarabelli) it has dugenf uno. See further supra, p. 332. On the whole I regard this as one of the highest authorities in respect of purity of text that I have met with. It would well repay, I imagine, a much fuller collation. Vivarini has given a facsimile of a few lines of this MS. in his reprint of the Codice Bartoliniano, Vol. I. Inf. viii. 90 we have ' si sicuro entro ' (auct. Scarabelli). So M. A (Bat. 248). In the Ambrosian Library at Milan. A folio MS. on paper dated 1399. The writing being cursive, rounded, sloping and very coarse (though clear) would certainly have suggested a later date. For a full description Batines (ll. p. 133) may be referred to. The orthography is archaic, the consonants being unassimilated, and the carelessness about double letters and the substitution of s for z (though both these peculiarities are only occasional) imply a slight Vene- tian tinge. The following are illustrations : gaecta, bructo, racto, ribac- tendo, ecterna, adcatta, dapnat, dopno (donno), subcedecte ad Nino (Inf. v. 59) ; viddi ed cognobvi /' onbra (Inf. iii. 59), altessa, semensa, saglita (salita), &c. It has an unusually large number of peculiar readings, especially in Pitrg. and Par., many being found (as far as I know) in this MS. alone. The following samples, taken in connexion with the large number of inferior readings registered in the collations, seem to me to deprive this MS. of much weight as an authority. Inf. i. 28 ; viii. 125 ; Purg. i. 108 ; ii. 13, 44 ; iv. 22 ; vii. 51 ; x. 14, 30; xxx. 73; Par. x. 112; xi. 82; xxii. 95; xxiii. 115; xxviii. 23; xxix. 4 2 . 1 On a second visit (in 1884) I am equally convinced of this. 2 The reading of A in Inf. xxi. 113 is wrongly entered supra p. 332. See Table of Errata. 362 564 ACCOUNT OF THE MSS. To these we may add : Inf. vi. 86 Diverse pene. (So G, 10, 109.) Purg. xxiv. 65 fan di loro schiera. (So s.) xxx. 74 d' ascendere al monte. (Comp. E.) xxxi. 43 Tuttavia mo perche. (So i only.) (Bat. 247). In the Ambrosian Library at Milan. A folio MS. on paper containing the Inferno only, and dated 1398. See Colomb de Batines II. p. 132 for a full description. The writing is very cursive, but not so much so as in A. The following few notes of orthography and text may be added : tallento, cossl, collui, malligne, golla, dollente, parolle, dibattcrro, smart to, mezo, camin,ponto (=punto). Inf. i. 28 Quand' io ei possato un puocho. (Comp. A.) iii. 59 Conobbi e vidi. (A curious and unique (?) inversion.') 74 parer al trapassar. iv. 95 quel signori. (So "2,. ) xxxiii. 74 E due d' elli chiamai. (Inaccurately entered supra P- 363). A (Bat. 237). This is the celebrated Codice Landi, now in the Biblioteca Comunale at Piacenza. It has a carefully written, or rather engrossed, colophon (see Batines II. p. 124, where however ' Paradisi' should be ' Paradasi'), with singularly long tails to such letters as b, d, /, s, &c. Here the date 1336 is clearly given. The writing of the colophon being somewhat artificial and the pen finer than that used for the text, and special pains having evidently been spent on the execution of it, it is difficult either to affirm or deny that it is written by the same scribe as the text, but there seems no reason why it should not be. The writing generally is somewhat larger and thicker than that of its Trivulzian rival r (q. v.), and becomes coarser as it proceeds. At fol. 17 (Inf. xx. 31) a different scribe seems to have come on, and in the later part (especially Paradisd) the writing is decidedly coarse and less upright. Speaking from memory, the writing reminded me somewhat of the Brit. Mus. ' a,' but such comparisons are scarcely to be relied on unless two MSS. can be seen together. There are a great many erasures by a 8iop#a>rj}? , and, as will be seen from a reference to Pallastrelli's monograph (mentioned below), a large number of peculiar or unusual readings have been thus introduced, which were not in the ori- ginal text. The following are a few orthographical peculiarities noted : direta (diritta}, graveza, me$o, selvagia, ogne, gaecta, facto, temecti, forsse, quac- tro, sexto, huropa (Par. vi. $),grolia for gloria (Inf. xxxi. 116), &c. Che (not ke) is written. EXAMINED OR COLLATED. 565 There appear to be no very marked indications of dialect. Its readings are often in agreement with those of r, though the latter has, I think, on the whole a decidedly purer text. There is a note : " Questo libro % di me Artaserse Bayardi...donatomi dal fu Sig. Cav Orazio Pencolini" &c. It has the title : "Incipit primus Cantus prime Cantice Commedie Dantis Alagherii florentini." For a full account of this MS. and its history, Pallastrelli's monograph (// Codice Landiano della D. C., Piacenza, 1865) should be consulted. Among the comparatively few 'peculiar readings' to be found (except- ing those introduced by the Siop&ur^?, as explained above) I noted the following : Inf. ii. 10 lo dissi a lui, Poeta. ix. 68 per li diversi ardori. The following bear marks of correction : Inf. i. 35 Ma impediva. iii. 3 si va nella perduta. xxi. 113 ducento uno con. See also Inf. i. 28. n (Bat. 402). In the Biblioteca Classense at Ravenna. A small MS. (6^x4) on vellum, dated in the colophon (see below) 1369 70, and written by Betinus de Pilis, who was also the writer of s (1368) and 2 (1351) (Bat. 187 and 414). The shape of the MS. is unusual and is something like Z, but the writing of this MS. is certainly older, and though rather minute is beautiful and clear. It changes to another hand at Par. xxiii. 19 [notxxii. as ap. Batines] and is defective at the end, the last line being Par. xxxiii. 1 8. The colophon to the Inferno is as follows : De suo betinus de pilis scripsit in usu 9 kal. Januarii anno domini 1369 (the last two figures seem to have been touched up, but there is no sign of alteration or erasure) cujus anima in Deo pace quiescat utinam. Facto (?) fine in hujus prime partis. That to the Purg. is : Deo gras. Scripta hac secunda parte per me betinum de pilis 4 kal. Februarii 1370. The orthography in this MS. (as in the other two by the same writer) is unusually strongly marked by Venetian dialectic varieties. The following speak for themselves, and should be compared with the similar forms noted in ? and 2 : dillattata, Senecca, ducca (sometimes), gellata, cossl (sometimes), Pissa, lumme, scemmo,fioccho, &c., &c., and on the other hand mezo, camin, vilan, ochi, &c. We find also the parasitic u which is so common in s, e.g. nuon, cuon y puoi ( =pot),guote ( =gote, Inf. iii. 97), nuoi and nut for not 1 . The following 1 See also supra under ' Q.' 566 ACCOUNT OF THE MSS. are also Venetian peculiarities : cupidisia, pillia (piglid), iwllia, mundo, ingumbra,filli (lorfiglio in Inf. x. 60). The readings of the Inferno seem to me to be as good as in any MS. that I have met with, but the text of the Purg. and Par. (especially the former) is far inferior. Several other cases have been noted in which the character of the text of the several Cantiche differs considerably 1 . No doubt different originals were employed, and probably in the earliest times Codici containing the Cantiche separately were more common. This would accord with the traditional separate dedication and ' publica- tion' (so far as such a word is applicable at all) of the three parts of the poem. 2 (Bat. 403). In the Biblioteca Classense at Ravenna. It is a small MS. on vellum (size 6 x 4j), written in a very minute but clear hand, and with two columns on a page, and is probably (as Batines says) somewhat later than the last named MS. This is the MS. locally, but surely falsely, assigned to Pietro Dante as the writer. It abounds in Latin and unassimilated forms such as node, dixi, neptunno, &c. The text is far inferior to that of n in the Inf., but better, I think, in the other Cantiche, or at any rate in the Purg'. 2 The following readings may be noted : Inf. ii. 22 ; xxi. 93 ; Purg. xxx. 1 5 ; Par. xxii. 94. There is a note at the end as follows : Nota che Dante natus fuit anno 1365 sedcnte Urbano quarto, vacante Romano Imperio per morion Frederici secundi die ocio Marcii, et obiit anno Incarn. 1321 die 14 Sept., et sic vixit annos L VI, &c. There is, I imagine, no authority or probability for the day here given as that of Dante's birth. In the Biblioteca Comunale of Bologna (Cod. 223 in Catalogue), where I found three MSS. not mentioned by Colomb de Batines. (He records two as existing in the Biblioteca dell' Universita, but this was unfor- tunately closed at the time of my visit.) This is a small folio (HJ x 8) on very fine vellum, and very beautifully written. It seemed to me (speaking 1 e.g. Comp. what is said of and x> &c. and see other examples of this in the Appendix on the 'Vatican family ' of MSS. 2 I cannot agree with Witte either in his depreciatory estimate of these MSS. (so far at any rate as relates to II), or in his statement that they do not essentially differ (Proleg. pp. xlv vi). It should be remembered however that his inferences here and elsewhere are based on the collations of a single Canto, which, as I have pointed out in my Proleg. (pp. xxxiii iv), would afford very insufficient, and possibly misleading, data for any such judgment. EXAMINED OR COLLATED. 567 from memory) to be almost a facsimile in handwriting of the sample page of the Cambridge MS. Q, and of Lord Ashburnham's r. The illuminations at the beginning of each Cantlca also seemed to me like those of Q. It is written in one column, and has a very broad margin. There are no titles to the Cantos. The initial letters are beautifully illuminated, though on a small scale. There are no corrections or notes of any kind, and the MS. is in a beautiful state of preservation, except that it is unfortunately defective from Par. xii. i to xvii. 9 (incl.}. The following are a few notes of orthography : cusl and chossl (for cosl\cum, sum ( = sono), talle (tale), temmo, huoppo (uopd), bagiato, bascio, retcnutti, presura, sollo (solo], erteccia (ertezza), simele, plenitudene. Que is sometimes written for che and qual for cK al. The following readings seem peculiar : Inf. xx. 87 suo corpo umano, auct. Scarabelli. (So G.) Also Inf. ii. 55 ; Purg. i. 108 ; ii. 13 ; iv. 125 ; xiii. 2 ; xvi. 145 ; Par. xxviii. 50. Also in the Biblioteca Comunale at Bologna (Cod. 200 in the Cata- logue). A I4th century MS. on paper (iif x 8), the writing clear and upright and with rather long flourishes. The colophon gives the date 1380, and is as follows (in a sort of 'engrossing' hand very much like that of A) : Explicit liber tertius sapien poete Dantis de Aldigheriis de Florencia. In quo tractat de Paradixo. Scriptum per me fratrem SANCTUM de Abbacia Vangadide tempore ven patris e dni D. Antlion ejusdem abacie Abatis milesimo ill lxxx die xxviif Octubr. After this follow : (i) one page of verses ; (2) four pages (eight columns closely written) of prose, commencing, " Inquit in Ecdesiastico Salomon Sapia abscondita et thexaurus^" &c. ; (3) five pages of com- mentary beginning with the citation of Inf. i. i. The orthography is of the most strongly marked Venetian character : e.g. we have x for .$ and for g constantly ; u for o ; initial g evanescent ; double letters completely disregarded, &c., &c. The following will speak for themselves : raxone ( ragione~], boxardi (bugiardi}, imprexa (impresa\ ofexa (offesa), dessixi (discesi), spoxo, opoxito, cului, conubi, chunio (conio\ luntana, nut, cossl and cusl,possa(=poscta], semo (scemo), iacere (giacere), cu iochi ( = con gli occhi, Inf. iii. 79), gotta (goto], cita, note (notte], sono (sonno\ &c., &c. The text exhibits more than an average number of peculiar readings, and also of distinctly inferior readings among those more commonly found, preserving however a few that are good but comparatively rare, (e.g. Inf. iii. 36 [prima manu~\; xxviii. 135 ; Par. iii. 15 ; xxvi. 104, &c., though the excellence of some of these may perhaps be disputed). 1 These words form the commencement of the Commentary of Pietro Allighieri. 568 ACCOUNT OF THE MSS. The following are more or less peculiar : Inf. i. 28 ; v. 102 ; xi. 90 ; xii. 66 ; xiv. 48 ; xviii. 12 ; xx. 69; xxx. 18 ; xxxiv. 13; Purg. vii. 51 ; xxi. 19, 61 ; Par. ii. 141 ; xi. 26; xxxi. 20. a In the Biblioteca Comunale of Bologna (Cod. 700 in the Catalogue). A MS. on paper (8| x 5^) written in one column, the characters rounded but very upright. The date probably late I4th or early I5th century. It commences with six pages of prose introduction. The following peculiarities were noted in a rather hurried examination : Inf. iii. 123 Tutti vegnon qui 1 . (So only w.} xxi. 138 verso il duca mio(!). (anct. Scarabelli. See note under M.) xxiv. 62 ronchiosa stretta. (So L and f.) Purg. viii. 64 1' altro a lui. Par. iii. 47 bene in s guarda. (auct. Scarabelli.) Add Inf. xxvi. 25 ; xxxiii. 26 ; Purg. xxx. 73; Par. xxvi. 136; xxix. 4. These are nearly all clearly inferior readings. [The 27 MSS. marked with small Greek letters belonged to the cele- brated Ashburnham Collection, which was lately purchased by the Italian Government, and is now added to the already unrivalled collection in the Laurentian Library, raising the number of MSS. of Dante in that one Library alone (to say nothing of others in Florence) to considerably over a hundred. I wish to acknowledge here very gratefully the kindness and hospitality of Lord Ashburnham, by which I was enabled on two occasions to examine this valuable collection at leisure and without restraint. I regret that I omitted at that time, in many cases, to register ' peculiar readings ' of these MSS.] a (Bat. 450). This interesting and celebrated MS. belonged till lately to Lord Ashburnham's collection, and was formerly in that of Signor Libri. It has been called Z.' Antichissimo' i , bearing as it does (on a fly-sheet) the date "...dogosto 1335." The day of the month has been ob- 1 This is clearly erroneous, as it will not scan. Convegnon would often be written* cvegnon or yvegnon. This first letter or symbol has been apparently omitted or overlooked in this case, and has given rise to the reading ci vegnon in some MSS. (e.g. e, 17, x)- 2 This and the other descriptive names for the Libri MSS. were given by Montani. EXAMINED OR COLLATED. 569 literated, the year is quite distinct, though its authenticity is naturally disputed, for it may have been added at any time and by any hand. There is no colophon at the end of the Paradise itself. The MS. is undoubtedly of great antiquity, and at any rate probably within the 1 4th century, though it does not appear to be all by the same hand. It is a small folio (12^x9) on vellum, and the writing is large and clear. Erasions and alterations by a diopdvTTjs are not uncommon. The Capitolo of Jacopo follows the Paradiso. The following and similar archaic forms are common : cicta, tucto, grocta, f actor e, sospecto, justitia, ogne, ke ( che}^ Ghaleocto (v. 137). In v. 134 we find bagiato, but bascib in 1. 136 ; li for gli occurs passim. I also noted piova (xiv. 48), piei (xix. 64), Netupno (xxviii. 83). Whatever its age, its readings, as given in the selected list of test- passages, are sometimes such as fully to bear out Witte's description of it (Proleg. p. Ixviii), "II testo non e esente di errori indubitabili, e di lezioni moderne." The following may be noted in the collations supra : Inf. ix. 125 ; xiii. 21 ; xxiii. 83; Purg. i. 108 ; xiii. 2; xix. 34; xxi. 25 ; Par. xvii. 9 ; xix. 141. Add Purg. xxiv. 65 fanno di se schiera. (So '39' only, out of about 1 60 examined.) In the Ashburnham Collection (Libri 406). A MS. entry on the fly- leaf states that it was the property of Count McCarty and was purchased at his sale in Paris, (signed) Cyrus Redding, Oct. u, 1817. It is a small folio (i ij x 7^) on paper, probably early in I5th century, clearly and beau- tifully written (though the letters are rather rounded) with a broad pen or reed. There is only one column on a page, and the margin is very broad. The orthography of words is archaic : e.g. pecto, ad me, descricto, cro- cifixo, maladecto, ecterna, impecto ( = itnpeto, Par. i. 134). The following forms are perhaps noticeable: Nexo ( = Nesso,Inf. xiii. i), Fregetonta (xiv. 131), aresti (Par. xxii. 109). In the headings to the Cantos ociavus is written optavus. Proper names are written with capital letters, which is, I suppose, a sign of somewhat late date. It contains several bold and strange variants : e.g. Inf. i. 48 SI che parea che '1 mondo ne tremasse. [The three rhymes are desse, venisse, tremasse, and the two last are underlined by some one who noticed the violation of rhyme.] iii. 6 e '1 sommo honore. (Cf. D.) xvi. 136 si ragrappa. Add Inf. iii. 31 ; xii. 125 ; xvii. 95 ; xxxiv. 99 ; Purg. xxii. 6 ; Par. xi. 82 ; xvii. 9 ; xxvi. 136 ; xxvii. 57 ; xxviii. 23, 50; xxxiii. 126. 570 ACCOUNT OF THE MSS. These and other peculiar readings of this MS. are further discussed in Appendix II., where it is shewn to belong to a small but well-marked group. Title : Incipit primus Cantus Inferni. Colophon : Explicit tertia et ultima Comedia 1 Dantis Aldegerii de florentia, q dicitur Paradisus. Deo gras. Am. Amen. Then follows the Capitolo of Jacopo with colophon : Explicit qua;- dam expositio super tribus libris Dantis edita a filio suo. Deo gras. Amen. y (Bat. 458). In the Ashburnham Collection (Libri 836). A small folio (n|x8), on paper, somewhat coarsely written, but the letters though much rounded and sloping are large and clear, and mezzo-gotico in character. The MS. is clearly of the I5th century. Inf. vi. lines 55 87 are defective, owing to the loss of a page. The MS. is much damaged and torn, but has been carefully repaired. Title : Qui comincia la prima Cantica ? 2 cominciamcnto del libra del venerabile auctore dante Aldigieri (sic) di Firenze intitulata inferno nella qual poeticamente fingendo pertracta particulare (sic) e divisainentc de le pene e punitioni di peccatori. Colophon: Qui finisce la terza et ultima cantica intitulata Paradiso del venerabel (sic) poeta dante Aldigieri difirenze. Dio gratia. Amen. Then (in a different hand) the terza rima poem of about 80 terzine commonly known as the ' Credo ' of Dante, beginning lo scrissi gia d" 1 amor piu volte rime, &c. It has a singular peculiarity that the final vowels are constantly omitted, where the scansion requires elision, and where we should put an apostrophe, e.g. iscript, lasciat, dentr, tumult, concedet, succedet, &c., &c. A few peculiarities of orthography or reading which I noticed were : ecterna, habondante^fixi (fisst), diddo (Inf. v. %$},bagiato (Inf. v. 134). Inf. xxxiii. 82 reads, ' Muovasi la cravara e la gargona ' ; and Inf. xxxii. 34, 'Si vid'ioli la' (so G). To these we may add : Inf. xii. 125 ; xvii. 95 ; xx. 69 ; xxxiii. 72 ; Purg. ii. 93 ; xxiv. 61 ; Par. i. 135 ; xxii. 94. 8 (Bat. 460). In the Ashburnham Collection (Libri 837), known, according to Batines, as "/'/ Verrezzano^ from the name of its former owner, as appears from the note on fly-sheet "del cavaliere Andrea da Verrezzano." It is a paper 1 This is rather unusual, but not so uncommon as is implied by Batines II. p. 239. Comp. e, IT, &c. 8 The MS. has been torn and repaired and is illegible here. EXAMINED OR COLLATED. 571 MS. (i i x 8j) of the 1 5th century, and perhaps somewhat late in that cen- tury, as the writing is much rounded, coarse, and slightly sloping. The letters too are separate and wide apart. Still it is very clear and neat. It contains the Inferno only (74 leaves), and then an Italian Commentary (8 1 leaves), anonymous 1 , very clearly though somewhat flowingly written. Though the MS. is somewhat late, the forms of words are archaic, e.g. obschuro, scripte, ad me, ad le, ad quel, &c. (passim}. Suo, &c. seem to be treated as indeclinable. In xiii. i Nesso is written Nexo, and in xxxii. 134 we find the vulgarism Hodio. Inf. xxiv. 40 43 was accidentally omitted at first, and added, by the same hand apparently, in the margin. The MS. has a special interest from its presenting the 'Sienese' type of readings distinguished by Witte in Canto iii. 2 ; and moreover keeps more closely to them throughout than any other MS. I have seen, having 1 8 out of the 20 passages given by Witte as tests. Speaking generally, it appears to me to have excellent readings and a remarkably pure text. I have noticed in other ' Sienese ' MSS. also evidence that the foundation text is a very good one. Title (very similar to a-} : Comincia la Chomedia di dante Allighieri dajirenze ilquale tratta delle pene epunizioni de" vitii e de' meriti e premi delle virtu. Cap" primo della prima parte di questo libro lo qtiale si chiama inferno, el quale /' autorefa proemio ad tutto el trattato del libro. Colophon : Explicit liber primus inferni dantis Elegherii (sic) de florenzia (sic). Deo grazias (sic). Amen. The following readings may be noticed : Inf. xxi. 93 ; xxiii. 83 ; xxviii. 83. In the Ashburnham Collection (Appendix No. 185). A MS. on paper (10^x8) of 210 leaves. The writing is large and clear, but the letters much rounded, very sloping and clumsy. It has no title. The date is given by the colophons : Inferno. Explicit comedia diltis Aldegherii de florentia que dicitur infernus. deo gratias. Amen. In fighino anno domini 1435 die octava Ottubris in ora vespertind. Purgatorio. Explicit &c. die xxviiif Ottubris. Deo gratias. Amen (repeated 6 times). Paradiso. Explicit liber paradi 'si dantis ad"Sgherii de florentia consu- matus per me Nazarius (sic) laurentii dadi de spinellis de Santo Gemi- 1 This appeared from the identity of a few lines at the beginning to be that of Jacopo della Lana. I found however on comparing several passages in different places that it did not correspond with that Commentary as published at Bologna 1866, but in some places there was a general agreement in substance, though not in language. 2 See Witte, Dante Forsck. I. p. 280; or Prol. p. Ixxv. 5/2 ACCOUNT OF THE MSS. niano die domenico die xiiij. mensis Ottubris hard xiiij sub anno doinini 1436 (in full Latin words). Three pages of Italian and Latin verses (fragments) follow at the end, texts, quotations from Chrysostom, Jerome, &c., also fragments of ac- counts, and portions of a copy-book. Suo, &c. are used indeclinably. Sometimes mi for into, dixe for dtsse, &c. On the remarkable relation of e to /3 and other MSS. see Appendix II. We may note : Inf. ix. 54 ; xx. 69 ; xxvii. 21 ; xxxiv. 99 ; Purg. xxii. 6; Par. xxxiii. 126. (Bat. 464). In the Ashburnham Collection (Appendix 181). This is a most beautiful and highly interesting MS., and has excellent readings. The vellum is singularly white, and the writing exquisite, though not of an uncommon type. It resembles in character the Bodleian 'C' and Br. Mus. 'o' and also a MS. at the Brera at Milan (designated T) and may perhaps date from about 1360 70. The writing is also extremely like that of , and the initial letter to the Inferno is very similar. The writing is distinguished by light flourishes, and long tails to letters like d, I, h, g, s, &c. It contains 98 leaves, and the size is I2f xg: there are two columns on each page, and the Cantos (as is often the case) have blue and red initials alternately. It contains the two poems known as the Capitoli of Jacopo, and of Busone (Perb ke sia &c.). It came from the Bourtourlin Collection [No. 89], and belonged to the Marchese Manfredi Malaspina and to other members of that noble family. The Catalogue 1 suggests that, as the Purgatorio was dedicated by the author to Marcello Mala- spina, this MS. from its unusual beauty may have been the presentation copy, or at least a 'copie de luxe'' made from it by Malaspina. The former supposition would be excluded by the date as implied by the writing (even allowing a wide margin for error of judgment) ; also by the internal evidence of the text, which, though generally excellent, contains some undoubtedly secondary and derivative readings ; nor can we suppose that Dante, the exile whose poverty is so pathetically described in a celebrated passage of the CoHVito\ could have commanded the means for the execution of so costly a work. The latter suggestion, viz. that Malaspina may have caused a splendid copy of the work to be executed at an early date, does not seem improbable, and its possibility gives the MS. a peculiar interest. But in any case textual considerations (good 1 I find that this suggestion occurs also in Batines, where I have identified this MS. since writing the above. 2 " per le parti quasi tutte, alle quali questa lingua si stende, peregrine, quasi mendicando sono andato portato a diversi porti e foci e liti dal vento secco che vapora la dolorosa poverta." Tratt. I. Cap. iii. EXAMINED OR COLLATED. 573 though the text is) exclude the possibility of its being a literal and exact copy of the orginal. Title : Incipit comedia Dantis Allegherii deflorentia. Incipit primus liber inferni. Capitulum primum. Colophon : Explicit comedia Dantis Al. (sic) deflorentia. Do gr. Among other indications of archaic orthography we have ad for a constantly, and ke for che (see Colomb de Batines II. p. 100). The lines Inf. vi. 69 71 (Con la forza &c.) and Par. x. 4 7 (Quanta per &c.) were omitted at first, but have been added in the same hand at the bottom of the pages. On a blank page (fol. 64) between Inf. and Pitrg. is written 'Epitaphium Dantis,' Jura Monarchic &c. This is in a much later hand, and probably (as suggested in an anonymous modern MS. note in Italian on the fly-sheet) about 1500. In 1. 5 we read patriis extorris ab oris. In the Ashburnham Collection (Appendix 182). A folio (i2f xg) on vellum, of 133 leaves. The letters are very large, and in clear print- like Gothic, though somewhat rounded, and the MS. is one of the most beautifully written I have seen anywhere ; but, though no doubt well within the I4th century, it abounds with bold and peculiar readings, in most cases bearing unmistakable evidence of their conjectural character. It is defective in the Paradiso; there being a hiatus from xxv. 29 to xxvii. 49, and the MS. breaks off altogether at Par. xxxii. 138 (ciglia being the last word). There is also a page missing after Inf. xxii. 18, causing the loss of Inf. xxii. lines 19 to 123. It is written in double columns, but the second column is occupied by the Commentary of Pietro di Dante, who is described as "doctor in decretali et scientifico huomo." Title : Comincia il primo capitolo della comedia delF alto poeta dante Alleghieri de firenze nella quale tracta delle pene de 1 vitii e premi dclle virtu. The following are some of its bold and obviously conjectural al- terations : (It should be noted that in the majority of these cases a later Siop&or^r has added in the margin the ordinary readings.) Inf. iii. 42 Ch' alcuna gloria non arebbor. vi. 36 Sopra lo dosso a ciascuna persona (!). vii. 119 borbogliar questa. 124 bolliente negra. viii. 78 che foco fosse. - 79 far lunga giornata. xi. 12 Al tristo senso. ,, xiv. 1 1 1 piu che altro. xvi. 93 ancora uditi. 574 ACCOUNT OF THE MSS. Inf. xvii. 8 Soave arrive la testa. xx. 32 S' aperse gli occhi. xxiii. 143 breaks off after 'Del Diavol' without any sign of erasure. xxx. 19 Del mal. (So P, 6, and perhaps others^ mar and mal being not unfreqtiently interchanged. See note under 6.} Purg. xxi. 21 tanto forte scorte (line too long). xxviii. 92 Fe' 1' uomo buono e ad ben questo loco. xxxi. 43 Tutto perche mo tu vergogna porte. Par. vi. 117 non vivi. These examples, taken, as it may almost be said, at random, being met with merely in searching for the occasional test-passages, enable us to judge of the wholesale manner in which the text has been tampered with. The critical value of the MS. as a witness to the text is con- sequently small. To these we may add : Inf. iii. 114; v. 38 ; xvi. 3 ; xix. 3 ; xxxiv. 99 ; Purg. iv. 22; v. 38; xxvi. 72; xxvii. 81 ; xxx. 73; Par. xvii. 9; xxii. 94. e. In the Ashburnham Collection (Libri 405). A MS. on paper (ujx 8) of 121 leaves, with the usual alternate red and blue initial letters to the Cantos. It is written very clearly in large roundish Gothic, and is of the latter part of the I4th century. On the first page is the Capitolo of Jacopo, and after the Paradiso is that of Busone. This MS. is noted in the Catalogue as "Important"; and on a fly-sheet in oldish writing we read " Opus rarissimum, imo Cimelium Bibliothecarum." There is also on a piece of paper signed H. C. B. [Dr Barlow] Aug. 3, 1876 "This is a very remarkable Codice, and deserves to have all the varianti carefully copied out." The MS. has apparently been extensively examined by ' H. C. B.' as pencil memoranda thus signed on slips occur here and there. It is indeed most 'remarkable' and highly curious, but certainly not pro- portionately valuable. Indeed for purposes of textual criticism it is almost valueless, since it abounds with the most extraordinary and ex- travagant alterations, amounting sometimes to a deliberate refacimento of the original, as reckless as the specimen of Bentley's treatment of Milton referred to Proleg. p. ix. There ought to be no difficulty in tracing its relations among other MSS. if it has any. It has the title : Incipit Comedia liber sapientissimi poete Dantis Aldigieri dejlorencia. pars prima Inferni. Capitulo primo. The colophon is : Explicit liber sapientissimi poete Danti (sic) Aldi- gieri dejlorencia. Dei (sic) gracias. Amen. The dialectic varieties are scarcely less singular and well-marked than those of the text. The majority of them point clearly to a Venetian origin. EXAMINED OR COLLATED. 5/5 We have u for o constantly, e.g. am, ail ( = con, &c.), doloruse, multo, ultra, pro/undo, cutanto, lucho (loco), duve (dove). So i for e, even in final letters ocasionally, e.g. aspcctati (iii. 85), parir, vidir, qnisto, quillo, avissi, Maistro, virgognusi, illo and ilia Conversely dc, me, se are found for di, mi, si. Add casone (cagionc), baso, basato (bacio, &c.), so for siio (iii. 129), angossa (angostia), suzcsson (succession), zoclie se vieole (iii. 96), zascha- duno, nassimenti, tnalvasa, bitgia (=buia, iii. 1 30), piczola (Par. xxiii. 67). All these are clearly Venetian. As a specimen of orthography we may take iii. 123, Tucti ci vegneno qua de oiie paise. The following are some samples of its many curious vagaries. (I have preserved in some cases the strange dialectic orthography.) Inf. iii. 8 Lo non eterno. - - 23 Risonando. --25 Dentro so (?son) lingue de orribili faville. --47 Che lor. ,, -- 49 al mondo. (There arc numberless such little changes of single words.) -- 73 sappia, chi son quisti. ,, - - 80 lo mio parlar. --85 Non aspectati mai vidir el celo (sic). --94 non ti corn/are. Cf. corrucci, Inf. xxiv. 129. ,, - - 100 L' anime ch' erano quivi lassc e nude. - - 106 restrinser. (So B, V- and i.) ,, v. 55 Ad vicio de luxuria fo corructa. vi. 105 O fion minori o fino (sic) si content!. ,, xx. 30 Ch' al judicio de dio passion porta. ,, xxx. 19 Del mal 1 . Par. xi. 91 pura intentione. xvii. 76 Colui vedrai colui in presso fue. At Inf. xxiv. 50, c., a curious muddle occurs with gaps and endings of the lines, thus ; Cotal vestigio T chagla Oual fora i aire aqua ischiuma I'iii lunga schala essi sagla. It will be observed that a tcrzina lias also been omitted between the second and third of these lines. This looks as if the MS. had been written from dictation, and the scribe had here become sleepy or other- wise confused. The endless dialectic variations and strange orthography 1 Is this a case of the not (infrequent practice of introducing a correction with regard to the construction of the single line in which it occurs (see Proleg. p. xvi)? It is to he noted however that mar and mal are not unfrequently interchanged elsewhere. (See sup. p. 4^6.) 5/6 ACCOUNT OF THE MSS. are also in favour of the theory of dictation, though the unbridled license of conjecture in the text would be adverse to such a theory, unless these variations existed in the exemplar MS. and were not first introduced in this. See further the readings registered at Inf. \. 28 ; iii. 31 ; ix. 125 ; xii. 16; xvii. 19 ; xix. 3 ; xxi. 93 ; xxiii. 83 ; Purg. viii. 94 ; xiii. 154 ; xviii. 83 ; xix. 34; xxvii. 81 ; Par. xxv. 33 ; xxvi. 134. The lines Purg. ix. 8 and 9 (Fatti avea, &c.) are accidentally omitted. (Bat. 453). In the Ashburnham Collection (Libri 832, l 'L' otttmo"}. A large folio (14x9!) on vellum, of 135 leaves, late in I4th century. The writing is large but roundish Gothic, something like Bodleian A in size and cha- racter, but larger and, I think, older. Some of its abbreviations, e.g. those for con, et, &c. are also similar to those in A, and the Commentary goes round the margins of the text in a similar arrangement. The Commentary however in this MS. is that of the Ottimo (according to Batines), and on the margin outside this again in a later hand, flowing and difficult to read, is an anonymous Latin Commentary. This is very brief and fragmentary, and many pages are wholly without it. The MS. contains the Inf. and Purg. only. The readings generally seem very good and correct. The following are a few orthographical varieties noticed : secrata (segreta), durara, familglia, ongne, como, lecto, cusl, and dil for del occa- sionally. These point, so far as they go, to a N. Italian origin. The title is : Incipit ca p inferni, and the colophon : Explicit secunda cantica Comedie Dantis in qua tractatur de purgatorio. Deo gracias. Am (three times). Very few peculiar readings occur. I noted only Inf. xiii. 20 ; xix. 3 ; Purg. xix. 34. K (Bat. 455). In the Ashburnham Collection (Libri 831, "// nobile"}. A large folio (14^ x 9^) on vellum, of 90 leaves. The writing appeared to me somewhat similar to that of the Bodleian 'C,' and therefore probably c. 1380. It also closely resembles f and , but is slightly inferior, and perhaps a little later in style. The initial letters are beautifully illuminated. It has (like 'C') the archaic ke for che, and also such forms as elgli, colgli, quelglt, ongni, &c. It has, I think, remarkably good readings throughout. It concludes with the epitaph of Dante, beginning Inclita fama cujus universum penetrat orbem, &c. This is in a later hand 1 . 1 See also the Paris MS. '6,' where this epitaph is attributed to 'filius suns.' EXAMINED OR COLLATED. 577 The handwriting changes at Inf. xxx. 136, and becomes finer, but both before and after, though similar to f and |, it is decidedly coarser than that of those MSS. Title : Incomincia la comedia di Dante Allaghieri di Firenze nella qual tratta dele pene e punimenti de vicii e de meriti e premii dele virtu. Canto primo della prima parte detta Inferno nel quale /' autore fa pro- hemio a tutta /' opera. Colophon : Qui finisce la terza et ultima cantica della comedia di dante alleghieri difirenze chiamata paradiso. deo gras. Amen (four times). The only peculiar readings noted were Inf. xxiii. 83 ; Par. xxxiii. 89. A, (Bat. 457). In the Ashburnham Collection (Libri 830, "// magnifico"}. A splen- didly illuminated late i4th century MS. (according to Batines rsth century but??), on beautiful vellum (i3|x8|) of 243 leaves. It has very large clear Gothic characters fully inch long. The text appears to me to be a very good one. It has no titles or colophons. The following are a few specimens of its orthography : malvasgia, casgione, rasgion, palasgio, Matestro, ongni, sengna, bangna, Kerubmt, secta, captivi, tucto, &c. The terzina, Inf. ii. 97 99 (Questa chiesi, &c.) is omitted, and supplied at the bottom of the page in a comparatively modern hand. The MS. abounds in erasures and alterations. On the connection of this MS. with /3, t, o, &c. see Appendix II. The following readings are noticeable : Inf. iv. 23 ; xii. 66 ; xxix. 46 ; Purg. ii. 93 ; xxii. 6 ; Par. xi. 82 ; xvii. 9; xxvi. 136; xxvii. 57; xxviii. 23, 50. fi (Bat. 459). In the Ashburnham Collection (Libri 833). An early i5th century MS. on paper (njx loj), the writing round, but clear and upright. The first page is lost, so that the poem begins at ii. 10 (lo cominciai, &c). It has the colophon : Finito il libra di dante allighieri di Firenze, il quale morl nella cipta di Ravenna il dl di santa croce a dl xiiij al mese di Sett~bre. Anni (sic) Domini MCCCXXI, lachui anima requieschant (sic) in pace. Deo gra- cias. Amen. Then follow : (1) O Sommo Eterno et infinite bene, in terza rima, about 100 lines. (2) I' scrissi gia d' amor piu volte rime, &c., about 150 lines, ending (apparently unfinished) with the line a rifermar questo creder medesimo. (This is a portion of the so-called ' Credo 1 of Dante.) D. I. 37 578 ACCOUNT OF THE MSS. (3) Five or six pages in modern hand of prose and verse (Italian). (4) Nine leaves of vellum containing the whole of a very brief Latin Commentary 1 on the Inferno ; about half a page to a Canto. Among other evidences of a later date are the frequent (though not invariable) assimilations such asfatto for facto, &c. The text seems in close relation with the group noticed elsewhere, T, W, X, Tr, &c., and to be one of a good average character, no unusual readings having been noted except Par. x. 112 ; xxii. 95. v (Bat. 452). In the Ashburnham Collection (Libri 827, " Z,' elegantissimo"'). A small MS. on vellum (size "]\ x 5). The writing is very neat and beautiful, but the Gothic characters are rather rounded, and I should have thought it could not be placed at so early a date as that assigned by Batines (viz. middle of the I4th century). It has very graceful illuminations, and richly gilded initial letters to each Canto. There are many marginal varr. lectt. by a later hand. The following are indications (in respect of orthography) of an early date for the text. Latin forms of words occur, such as tracta, rapta, antique, but assimilated consonants are also not uncommon. The archaic k is frequently found, as kalle (i. 18), k' z' ora (xxviii. 3), &c. : g for c occurs in gonflati (Par. xxxiii. 89), and gingier = ringer (Par. xxviii. 23) ; also the forms piatosa and groria. This looks like Tuscan. It has a remarkable number of mutilated or half-finished lines, which may have arisen from an imperfect exemplar, or from ill-understood dictation : e.g. Inf. xviii. 114 Che dagli uman xxii. it Cavalier vidi 21 di campar lor legno. 33 Una remenare e 1' altra spiccia. 58, 100, 1 14 the words gatte, stien and galoppo respectively are omitted, and in xv. 92 garra is omitted. Such blanks are very numerous. This peculiarity occurs also in $, and in the British Museum 'i,' but in the latter case it appears generally to have arisen from doubtful read- ings, which is not the case here. Compare also w, 4 and 8, for a similar peculiarity. It has no title, and the colophon is merely Deo gracias. Amen. It has on the whole, I think, a poor text. 1 The Commentary is of a very early date, as appears from the note on Inf. xiii. 147, 'et quod nisi forte quedam ejus statua qnam adhuc habent florentini super pontem ami frustra fuisset hedifficata (sic!} quum earn Attila tradidit incendio.' EXAMINED OR COLLATED. 579 The following readings are noticeable : Inf. iii. 117 com' augelli. (So r.) Purg. xxxi. 43 Tutto pero che mia vergogna porti. To these we may add : Inf. xvii. 20, 95 ; xviii. 12 ; xxxiii. 26, 78 ; Purg. ii. 44 ; iv. 22 ; xviii. 83 ; xxiv. 61 ; xxxi. 51 ; Par. vii. 114 ; xi. 82 ; xxii. 17. f (Bat. 454)- In the Ashburnham Collection (Libri 829 "// correttissimo"}. A I4th century MS. on vellum (size I2f x8|). The writing is very beautiful, and indicates, I should think, the date of c. 1370 1380. It very closely resembles f, and also 'o' in Brit. Mus. This MS. is very imperfect (see a list of defective portions in Batines n. p. 255), a misfortune the more to be regretted since its readings appear to be singularly good. I should be inclined to consider it a very high authority for the text. It is also curious that, like < and x, while closely related to the Vatican family in Purg. and Par., it bears no relation whatever to that family in the Inf., as far as can be judged from the fragments that remain. Title : Iiicipit Infernus Dantis Allagherii defloretia. Cantus i. There is no colophon to Purg. or Par. and the end of Inf. is wanting. I did not notice any peculiar readings whatever, which tends to con- firm the high character above attributed to the text, as well as Montani's epithet of "z7 correttissitno? o (Bat. 451). In the Ashburnham Collection (Libri 834). A late I4th or early I5th century MS. on paper, of 218 leaves (size iiJxS). The colophon (quoted at length by Batines II. p. 253) gives the date March 10, I387 1 . It is defective in places and much damaged by damp. The first leaf is gone, so that the poem commences at i. 112. There is also a leaf lost which included Inf. xxxii. 127 to xxxiii. 46 (incl.}. The following are some orthographical peculiarities noted : Lassate, omni for ogni (passim}, avaria, vissaro, avrebbar, longa tracta (iii. 55), basgiato (baciatd), Robbaconte (Purg. xii. 102), etade, pietade, &c., talliata, pillia, 8f.c.,grisge (vii. 108), cupidisgia, casgione, &c. 1 This is, I think, an impossible date. I should have said certainly isth century. The writing is rather coarse and rounded, though clear, and with a slight tendency to slope, and to run letters together. I am confirmed in the rejection of this date by Mr Maunde Thompson, who says that both the writing and the paper are not before 'early rsth century.' 372 580 ACCOUNT OF THE MSS. Peculiar readings will be found noted supra at Inf. ix. 54; xxix. 46; Purg. ii. 44, 93 ; xxi. 61 ; xxii. 6 ; xxx. 15 ; Par. vii. 114 ; xi. 82 ; xvii. 9 ; xxii. 94, 95 ; xxvi. 136; xxvii. 57 ; xxviii. 23, 50. See also the relationship of this MS. with a small but well-defined group discussed in App. II. 7T. In the Ashburnham Collection (Appendix 184). A MS. on paper of 258 leaves (size lof x 7f ). At the beginning are the Capitoli of Busone and Jacopo. The title is a sonnet beginning Quivi comincia /' alta comedia and ending DelP avaritia malre di gran mali. The Purg. and Par. are each similarly introduced by a sonnet. The colophon is as follows : I$a finixe la ter$a et ultima comedia del Paradiso di Dante aldicherio de fioren$a. Scripto per me Jacubo de cauallo neir anno MCCCCLXXlllj die xxiiif Augusti vii Indictionis. Feliciter. Amen. Then follows Dante's 'Credo', *Io scripsi' &c., ending not la faccia (and hence deficient by about 80 lines). The writing is rounded and sloping and somewhat coarse though clear. There is a MS. note in pencil by H. C. B. (Dr Barlow) "copied from a very good Codice." (I am inclined to doubt this.) This MS. seems to have been used as a sort of Family Bible 1 . On the fly-sheet before the Inferno we read : "A dl xiii" d' Ottobre \ 5 1 3 in la citta di cotrone in casa mia a vi hore de notte et un quarto fino li sette h. % nato gaietto Viengeri (or perhaps Viringeri) mio figlio. Con bona ventura el nostro S. ydio li dona longa vita." Among peculiarities of orthography may be noted lassar (but lasdate in iii. 9), diserno, como, cossl and cussl, captivi, vagnia ( = bagna iii. 132), gaecta (i. 42), basb and basciato, magior, pietade, desende (dtscende), ultre, piatosa, pro/undo, mundo, lassivan (iv. 64), ropta (rotta), erbecte (Purg. xxix. 88), conflacti (Par. xxxiii. 89), alu (alo Par. xxviii. 23), multi sun (Inf. i. 100), Dirrb come collui (Inf. v. 126). Calca for carca in i. 50 is curious. Most of these, like the orthography of the colophon, would suggest a Venetian origin. I did not observe any peculiar readings except 'tre di lei' at Inf. xxxiii. 74. We may perhaps notice also Par. i. 135; xxvi. 136. 1 The same is the case with a MS. at the British Museum ('b'). The proper names are rubbed and difficult to read. EXAMINED OR COLLATED. 581 In the Ashburnham Collection (Libri 404, " // gentile'"}. A MS. on vellum (size iOjx6|) of i5th century (and probably late in it). It is written with a very finely pointed pen, and in a round and flowing and very sloping hand, but quite clear. The words as well as the letters are sometimes run together. The Paradiso is followed by the Capitoli of Jacopo and Busone. Both these seem to be more carefully written than the text, and the characters are more like printing, and in the style of v. The first page is splendidly illuminated, and there are also finely illu- minated initial letters to each Canto. It contains the book plate of W. Combs, and on the fly-sheet is a MS. note by H. C. B. (Dr Barlow). " This is a very correct and carefully written text, and may be referred to as an authority. It is evidently a selected text." I again differ entirely from this estimate. The average merit of the readings does not appear to me to be high, and there are many that are very inferior. The following examples would indicate that the orthography of the text is archaic, and has no relation to the date of the MS. itself. Tincta, secta, captivi, anttquo, damnati, prompti, giuncto, muglia (Inf. v. 29) ; speze = specie (iii. 104). The title is much obliterated and scarcely legible. There is no colo- phon but " Finis" The following may be noted : Inf. iii. 85 Non sperate giammai. xii. 28 via per lo scarco (omitting giu). Par. xi. 21 onde ragion. (So P, a, 66.) Add (see supra) Inf. i. 28 ; ii. 22 ; x. 65 ; xviii. 12 ; xx. 69 ; xxi. 93 ; xxiii. 83 ; Par. xviii. 75. In Inf. xix. 53 the following confusion occurs : Et il grido se' tu gia cost! ricto Diparethi anni mime Diparethi anni mimenti lo scripto. As / and c are often confused, Diparethi evidently means Di parecchi. This MS. stands in remarkable relation to '66' (q.v.}. It is singular that Inf. xix. 52 is entirely omitted in that MS. This may possibly have been to spare Bonifazio, or, if this or a similar MS. were the exemplar, it may have been due to the accident of such a defect as this. a- (Bat. 463 ?). In the Ashburnham Collection (catalogued as " Barrois 23"). A I4th century MS. (according to the catalogue) on vellum, of 129 leaves (size lox 7). On p. 109 begins Boccaccio's Commentary. The writing of this MS. is very minute, but very beautiful and clear. It is roundish Gothic, and probably late in the I4th century. $82 ACCOUNT OF THE MSS. It contains the bookplate of Hon. Fred. North, and has an anonymous pencil note " Bourtourlin" on fly-sheet. Title : Comincia la chomedia di dante Alighieri da firen$e il quale (sic) tratta delle pene e punimenti de> vita e de 1 meriti delle virtu. Cap" primo della prima parte detto iferno, nella quale V autore fa proemio a tutta V opera Cap j. (Compare the title in d.) It begins with a table of headings of each Canto, "Tavola de 1 Chapi- toli," &c. Colophon (on p. 93) : Finito il Paradise, &c., and then Qui finisce tutta la comedia, &c. Deo gratias. Amen (three times). Then follow the Capitoli of Jacopo and Busone. On p. 95 (reverse) Comincia il trattato che fecie lo illustro (sic) poeta Messer Giovanni di Bocchaccio da Certaldo sopra Dante. Arghumentum super tota prima parte, &c. Nel mezzo del chamin di nostra vita Smarrito in una valle /' autore, &c., &c. concluding on p. 107 with the colophon Finito il Dante piccolo doe certe spositioni d' esso Dante fatte per Mess. Giovanni Boccaccio da Ciertaldo. Amen (four times). On p. 109 begins an Italian prose Commentary (anonymous) very neatly written, commencing with Inf. i. 4 and breaking off at Par. vi. 50. A few extracts made at different places in Inf. i. seem to correspond with the Ottimo. It has some of the typical 'Sienese' variants in Inf. Canto iii., and has apparently some relation to this class. Its orthography is archaic, e.g. cipta, obscuro, scripto, ad me, adcorta, &c.: suo &c., are often indeclinable. Bracia for bragia in iii. 109 is peculiar. It seems to have a tendency to omit the conjunction ' e] e.g. iii. 87 (in caldo in gelo), &c. Perhaps this may account for the otherwise inferior reading in Purg. viii. 94 where e is omitted before Sordello, which is rare and certainly wrong. The following readings may be noted : Inf. xvi. 3; xxx. 18; Purg. x. 30; Par. i. 135; vii. 114; x. 112; xxxiii. 126. r (Bat. 1 88). In the Ashburnham Collection (App. 240). A splendid MS. on vellum (13^x9!), written in very black ink and in a beautiful upright Italian Gothic hand closely resembling the sample page in the Camb. MS. Q, and the Bologna MS. *. It is written in one column, and with a singularly broad margin. There is a fine illumination on the first page of each Cantica. It was bought at Baron Kirkup's sale in Dec. 1871. (For further details see Colomb de Batines n. p. 103.) The writing indicates the later part of the I4th century. EXAMINED OR COLLATED. 583 Title : Incliti vatis e poete clarissimi Dantis A llegherii civis honora- bilis florentini. Prime partis ejus comedie que infer nus dicitur Capitulum incipit primum fcliciter dante Deo. Colophon : Explicit feliciter. Amen. Up to the end of Inf. ix. 1 14 there is a Latin Commentary filling the margin, in very neat and clear though minute characters. The terzina Inf. vii. 31 33 (Cosl tornavan, &c.) is omitted. The following are peculiar or noticeable : Inf. iii. 64 Questi dolenti. 117 com' augelli. (So v.} Add Inf. i. 28 ; ii. 81 ; xxix. 46 ; Par. x. 112 ; xi. 82. u. A remarkably interesting MS. in the Ashburnham Collection (App. 219). It is on vellum (iox6f), and is written in a sort of print-hand, very beautiful and clear. This is not an uncommon type of writing, and is very like that of 'd' and 'h' in the British Museum, and '^' at Florence, the first and last of which it very closely resembles in every way 1 . It is of the 1 5th century, and perhaps about the middle. In the Catalogue it is stated that the writer was certainly a Tuscan. Such orthography as sprendea in Purg. xxix. 67 is an indication of this. It has exquisite illu- minations at the beginning of each Cantica, especially in the Inferno. The MS. begins with the following title : Brieve acchogliamento di cib che in se superficialmente contiene la lectera delta prima parte della cantica overo comedia di dante Alighieri di firenze Chiamata Inferno [the same (mutatis mutandis) for the Purg. and the Par.]. Then follows in each case a summary, that of the Inferno being the Capitolo of Jacopo. At the beginning of the Inf. is the following title : Comincia La prima parte della Cantica. Overo comedia. Chiamata Inferno. Del chiarissimo Poeta Dante Alighieri. Difirenze & di quella Prima Parte il Canto Primo. Videlicet. Colophon : Deo gras. Amen. Finis. The following are some samples of its orthography, which is archaic, and abounds in unassimilated letters. scripto, obscuro, captivo, cipta, ecterno, mactino, decto, ricto, tucto, &c., &c., even vencto (for vento], spavencto, talencto, &c., adnumptio (=an- nunzio, Inf. xiii. 12) : also et and ad, &c. occur passim. The vulgar initial h is very common, e.g. hughualmente (sic) in Par. xxxiii. 144. In Inf. xv. i this superfluous initial h is the subject of a fine illumination ! In the matter of its readings it appears to me to be a remarkably 1 I can say this with confidence, as I had an opportunity of placing the MS. side by side with the British Museum MSS. in May 1883. 584 ACCOUNT OF THE MSS. high authority, in spite of numerous and very serious vagaries. Its foun- dation-text seems to have been an admirable one, and, allowing for the above peculiarities, of quite exceptional value 1 . Another interesting point about it is its general agreement with what I have called the Vatican family of MSS. [see Appendix II.], which however it very curiously deserts for five or six Cantos at Inf. xxvi., and again similarly at Par. xxiii. (see the Tables in App. II.). The following are some of its peculiar readings (many being obviously mere blunders) : Inf. iii. 72 dissi al Maestro. (So only Z.) x. 85 lo stratio e '1 crudo scempio. (So also Z.) xv. I Hora comporta. Purg. xx. 1 14 Enfamia (sic) a tutto '1 inondo che si gira. xxiii. 37 Gia eravamo in ammirar. xxix. 67 La qual sprendea. xxxii. 1 02 Di quella remo (sic) onde Cristo era mano (!). Par. x. 98 Frate vicino fummo. 119 degli empi Cristiani (attributed supra p. 457 to 'v' in error). 1 36 de mestieri (for di Sigieri}. xxiv. 119 Con la tua bocca la mente. Add further Purg. xiv. 126 ; xviii. 83 ; xix. 35 ; xxii. 6 ; xxiv. 61 ; xxx. 73 ; xxxi. 51 ; Par. ii. 42 ; x. 112; xi. 26 ; xxvi. 134. $ (Bat. 517). This is another singularly interesting MS. in the Ashburnham Col- lection (App. 242). It is known as the ' Codice Wellesley,' having been purchased at Dr Wellesley's sale in 1866, and it formerly belonged to Dr G. F. Nott, of Winchester, in 1827 (as recorded on fly-sheet). It is unfortunately defective, since the MS. breaks off at Purg. xxxii. 141. The writing is very like that of the Bodleian MS. ' I,' but it is inferior to it, and more flowing and sloping, and sometimes rather difficult to read. The paper on which it is written in quality and size (8^x5!) also resembles that of 'I.' It is certainly written by a Florentine, or at any rate Tuscan, scribe (as also is 'I'), e.g. such forms as arebbon, groria, &c. occur, and tuo &c. are used indeclinably. On the fly-sheet Dr Barlow has described it as "Rarissimo Mano- scritto," and otherwise extolled its value very highly. I think its value has been very greatly exaggerated. It appears to me that its readings are generally not very good, and it contains many bad blunders and auda- cious changes (' belle varianti] as the phrase is !), and there is certainly no artistic beauty about its execution, but rather the reverse. 1 See Froleg. pp. xxii, xxiii. EXAMINED OR COLLATED. 585 Title : Chomincia la prima parte overo Chanticha chiamata inferno ddla comedia del vcnerabile (comp. y) pocta dantc aligliieri nobilc citta- dino fiorcntino. Chapitolo p inferno. Colophon at end of Inf. : Finita la prima cantica dctta Inferno del nobilc poeta dante Alighieri fiorentino. Finis. Qni comincia la scconda parte (fella chomedia di dante chiamata Purgatorio. A curious point about it is that it has a strong relationship with the ' Vatican ' family in the Purg. and none whatever in the Inferno. It is also evidently closely related to x, which exhibits the same peculiarity. See Tables in Appendix n. Another curious feature is found in the frequent omissions of words or whole lines, as if the scribe had an imperfect copy, or failed to catch the words if writing from dictation 1 , e.g. Inf. xx. 77 Non piu benacho si chiama. xxvi. 6 grande ne nu sali. xxxi. 36 stipa (no more tlian this). - 41 Montereggion di torn - 71 Tienti col corno disfoga. In xxxiii. 27, squardb il is omitted, and so also the names Sismondi in 1. 32 and Uguccione in 1. 89. Many similar omissions occur throughout. After Inf. xxxii. 66 there is a blank space left at the bottom of the page equal to three terzine, but there is no actual omission, and xxxii. 67 begins at the top of the next page. Also the tcrzina Inf. x. 109 n is left out without any indication of an omission, and it is very remarkable that this is also the case in the MSS. 54 and 106. The following readings are peculiar or noteworthy : Inf. ii. 53 cortese e bella. (So 54, 106 and some others.) 81 aprire lo mio talento. (Comp. ' 106.') iii. 40 Cacciarli i cieli per esser men che belli. (Comp. 106.) 68 Mischiate eran di. (54 and 106 have Che mischiato as usual.) - 75 per lo ciecho lume. (Not so 54 or 106.) ,, v. 118, 120 desiri (1. 118) and sospiri (1. 120) are interchanged thus. (Not so in 106.) Also in 1. 120 luc have tJie curious insertion 'non conoscessi" 1 (sic). 77n's is found also in 106. ix. 54 Ma non il vegiamo atteso all' assalto (?). xiv. 52 Se gia ne stanchi il suo. xviii. 12 dove '1 sol rende fotura. (Compare 54 and 106.) xxi. 91 Perch' io mi volsi allui e vcnni ratto. xxi. 112 I'iu oltre cinque ore che quest' otta. (So only c/>, 54 and 1 06.) 1 See note under v. 586 ACCOUNT OF THE MSS. Inf. xxviii. 71 E cu' io vidi in sua terra. (So A and 106.) 138 E diedi a lui chon malvagi. (So x but not 106.) xxx. 1 8 Vide '1 suo Polidoro. (So x-) Add Inf. xii. 16 (so 54 and 106) ; xvi. 3 ($ only) ; xxx. 44 ; xxxiii. 72, 78 ($ only). Purg. ii. 99 e terra con tua pace. (Compare 54 and 106 'which have 'terra con tutta pace.') xviii. 58 che sono a voi. xx. 1 14 tutto il mondo gira. (So 54 and 106.) See further supra under Purg. iv. 72 ; xviii. 83 ; xxii. 98 ; xxiv. 61 ; xxvii. 8 1 ; xxx. 73. It is obvious that this MS. has a very close and remarkable relationship with 54 and 106, at any rate in the Inferno. See further under the head of those MSS. It must be remembered that not ever having had the three MSS, or even any two of them, simultaneously before me, I have not been able to follow up fully the interesting indications here given. X (Bat. 456). In the Ashburnham Collection (Libri 835). It is on paper (io|-x 74), and contains 226 leaves. It is of the i5th century, and is written in different hands, which are however similar in character. It is imperfect at the commencement, beginning with i. 62 Dinanzi agli occhi, &c. There are no colophons to Inf. or Purg. After the Par. as fol- lows : Explicit liber Paradisi dantis allegherii deflorentia. gratia salvator totus visibilia et invisibilia (sic). Then follow several pages of terza rima viz. (i) Capitolo of Jacopo. (2) That of Busone. (3) The poem known as the Credo of Dante. (4) A short poem of about 20 lines giving a brief life of Dante. The orthography is occasionally curious, e.g. bacata for baciata (Inf. v. 134), gu for giu (Inf. xii. 28) [compare Bodleian D], spoxe for spose (Inf. xix. 3). As has been already noted, this MS. has close affinity with $, and (see page 585) shews unmistakable relationship to the 'Vatican' family in the Purg. and also in the Par. [which is wanting in <], while there is no trace of such connexion in the Inf. The following readings are to be noticed : Inf. ii. 53 gentile e bella. iii. 72 Perch' io a lui, Maestro. - 123 ci vengon. (So f, 77, 6. See also note under Q..} xvii. 96 m' accinse. xxviii. 138 E diedi a lui chon malvagi. (So <.) Add further : Inf. ii. 81 ; iv. 141 ; x. 65 ; xvi. 3 ; xviii. 12 ; xxx. 18 ; xxxiii. 78; Purg. xxvi. 72 ; xxxi. 51 ; Par. xiii. 27 ; xviii. 75. EXAMINED OR COLLATED. 587 In the Ashburnham Collection (App. 183). A small MS. on vellum (5l x 3f) an d> according to the Catalogue, of the I5th century. The writing is extremely minute, the letters rounded but separate from one another, and the ink very pale, so that it is often extremely difficult to read. The first page of each Cantica is illuminated, and so are the initial letters of each Canto. It contains the autograph of Benedetto Varchi to whom it formerly belonged. Title: Canto primo della prima parte della Comedia di dante alighieri indito poetafiorentino detta Inferno &*c. Colophon : Explicit tertia Cantica Comedie Dantis. Alleluia. Then follow the Capitoli of Jacopo and Busone. A peculiarity about this MS. is that it is practically identical with another (&>) in this collection. The differences are rare and mostly unimportant. The two MSS. have all the appearance of being in the same handwriting. This is interesting, as it is almost the only instance in which I have found any approach to a replica, even in cases where MSS. are signed by the same scribe 1 . The following peculiar readings were noticed : Inf. viii. 125 ; xvi. 104; xxxiv. 99; Purg. xxi. 61 ; Par. xxv. 33. a) also has these, except Inf. viii. 125, and Purg. xxi. 61. to. In the Ashburnham Collection (App. 241) on vellum (5|x3|). The writing and text closely resemble // (q. v.) and, like >//, it is very hard to read. It is a I5th century MS. and formerly belonged to Baron Kirkup. Title: Qui comincia la prima (seconda, &c.) parte, &c. The Inferno has no colophon. The colophon to the Purgatorio is as follows : Explicit secunda Cantica Comedie D. A. 2 de Florentia. That to the Paradiso is : Explicit tertia Cantica Comedie D. A? Alleluia. Deo Gracias. Amen. After three pages of Italian poetry comes the Capitolo of Jacopo, all this being in the same hand. Its peculiar readings mostly correspond with those of \^ (q.v.}. In Purg. xv. 73 it reads discende, or distende, and in Par. xi. 105 it has Rendesi (with F and t and 47 only). In Inf. vii. 108 it has the singular reading marine piaggie grige 3 , and in Par. xxiii. 115 1' intima riva. These four peculiarities are not found in ^. 1 Other cases will be found noticed below in Nos. 56 and 57 in the Chigi Library at Rome, and Nos. 88 and 89 in the Laurentian Library at Florence. 2 My notes do not contain the orthography of the name. 3 This reading is by a misprint attributed to W, supra p. 291. 588 ACCOUNT OF THE MSS. 5 (Bat. 187). In the Ashburnham Collection. A dated I4th century MS. (1368) on vellum (5fx3f), formerly in Baron Kirkup's Library, by whom there is the following note : " This Book I bought of Piatti the bookseller, to whom it was sold by Sig. Galeotto Corazzi of Cortona, who told me that it came from Genoa. S. K." The first seven Cantos are missing, the MS. beginning at vii. 127 Cosi girammo, &c. It is by the same scribe as the Paris MS. Bat. 414 (numbered '2' infra), the first seven Cantos of which were transcribed in facsimile by Audin on vellum and sent to Baron Kirkup, and they are now preserved in the same cover with this MS. The scribe was Betinus de Pilis, who, besides the Paris MS. already referred to, wrote the interesting and important MS. (n) at Ravenna. The dialectic peculiarities (Venetian) of this MS. are more than usually violent, and correspond remarkably (as will be seen) with those of the Ravenna MS. The description of this MS. by Colomb de Batines (n. p. 101) is so full and minute that it will be needless to repeat the details here. The word printed decimoceto in the final colophon should be decimo ocio. The following are some of the strongly marked Venetian peculiarities. (Compare the notes on n.) Ducca, lumme, scemmo, sposse (xix. 3), remmi (xxi. 14) (for duca, lume, &c.), and on the other hand, pasion, lasia, &c. (for passion, lascia, &c.). Also nuy for noi is common. The insertion of a parasitic u occurs in almost every line (comp. the Cambridge MS. *Q'), e.g. suon (sono),puoscia, cuon, puoco, nuon (for non usually), luoro (lord), muo ( = mo Purg. xxxi. 43). This is, I believe, common in the neighbourhood of Bergamo and Friuli. The readings are, I think, on the whole rather inferior, and some especially bad will be found. These can be traced in the lists of test- passages. The following are peculiar : Inf. xix. 23 ; xxxiv. 99 ; Par. xxvi. 136. The Ravenna MS. however by the same scribe is (in the Inferno at any rate) one of the best texts I have met with anywhere. F. In the Ashburnham Collection (Libri 408). A small MS. (4! x 3^) on vellum, with writing very minute but beautifully neat, very clear and in very black ink. It is Gothic in character, slightly rounded but quite upright. There occur however occasionally pages of very inferior writing. I should think its date is on the borders of the I4th and I5th centuries. The text has been frequently altered by a 8iopda>Tf)s and usually for the worse, the general character of the readings prima manu being I think decidedly good. It has generally long introductory head- EXAMINED OR COLLATED. 589 ings in red to each Canto, but none to Inf. i. which begins without any title or introduction. Colophon at end of Paradise merely Deo Gratias; and on the last page are 1 1 hexameter lines beginning Gloria jam felix, and ending regnas units per secula cuncta. Amen. After Inf. xxxiii. are 6 hexameter lines : Jamque domus Stygias (?) et tristia regna silentur and ending Flumina et horribiles. These are repeated (eight in number now) at the end of Canto xxxiv. There are again 10 hexameter lines at the end of the Purgatorio. It seems to have belonged to the Spinelli (a Florentine family), as the inscription shews " De Tommaso Spinelli, MCCCCCLXI? with appa- rently some further numerals obliterated. The following readings may be noted : Inf. i. 28; xii. 16; xxxiv. 13; Purg. vii. 51 ; Par. vii. 114; xvii. 9 ; xviii. 131 ; xxii. 95 ; xxiii. 115. ? (Bat. 461). In the Ashburnham Collection (Libri 838). A I5th century MS. on paper (i \\ x 8). The writing is rather rounded but upright, and the cha- racters are very large, generally nearly \ in. high, and sometimes fully this. The initial letter of each line is red, and that of each Canto is illu- minated. It contains the Inferno only, and is very fragmentary and imperfect. It begins at Inf. i. 43. There is a page lost at i. 89, and again at ii. 43, and there is also a gap from iii. 35 to iv. 81, &c. It breaks off at xxxiv. 127 (remoto). It is of no value or authority, and its text is one of the worst I have met with. The word gro lioso for glorioso, which I find noted, would seem to indicate a Tuscan origin. The following peculiar readings (some very bad) may be noticed : Inf. ii. 22 ; ix. 125 ; xiii. 21 ; xix. 3 ; xxv. 120 ; xxxii. 47 ; xxxiii. 26. a (Bat. 537). This very beautiful MS. in the British Museum (943 Egerton), the gem of the whole collection, dates from about the middle of the I4th century. It is a large folio (15! x io|) on vellum, and has splendid illuminations, and is written in a beautifully clear upright Gothic hand throughout, though on some pages the ink has scaled off so as to make it occasionally very diffi- cult to read. It has had an active 8top6a>Tris, and exhibits numerous erasures and corrections. The older readings have generally been erased with a knife, but sometimes words or syllables have been merely under- lined to indicate a correction to be made. The readings are on the whole very good indeed, and, speaking generally, I believe the corrections mostly 59 ACCOUNT OF THE MSS. exhibit later types. As an extreme case see Inf. v. 97. It contains a very brief anonymous Latin Commentary, terminating at the end of Paradiso xi. It would seem to be a very early one, from the gloss on the well-known passage referring to the statue of Mars at Florence, Inf. xiii. 146, "et nisi foret quaedam ejus statua quam adhuc habent Flo- rentini super pontem Ami 1 ," &c. This statue was destroyed by an inun- dation in 1333, and unless (which seems possible) adhuc habent 'is merely the equivalent of Rimane in the text, it would indicate a date anterior to J 333 f r the Commentary. The passage however must be admitted to be inconclusive. The well-known Capitolo of Jacopo follows at the end of the Paradiso. The title is : Incipit primum Capitulum primfiammette is written fianbecte. These seem to indicate a Venetian origin. The following are peculiar : Inf. xxxiii. 78 ; xxxiv. 99 ; Purg. ii. 93 ; xviii. 83 ; xxiv. 61 ; xxvi. 41 ; Par. i. 141 ; x. 4. After the end of the Paradiso we read, Explicit liber Paradisi Dantis. Deo gracias. Amen. Then Incipit liber acerbe vite &... .by "Cichtts de Esculo'" [Cecco d' Ascoli], about 16 closely written pages with double columns. Then the Capitolo of Busone. Next that of Jacopo. Then follows another copy of part of the Paradiso, in a different hand [called g 2 in the collations], and free from the dialectic varieties which are so striking in the rest of the MS. This breaks off suddenly at Par. viii. 80 in the middle of a page. h (Bat. 480). A small MS. on vellum in the Brit. Mus. (3513 Harl.) (yfx 5), of the first half of the I5th century, in a careful and somewhat artificial hand something like that of 'd,' but of earlier date. It has been much cor- rected secunda manu. At the end we read : Qui finisce la terza e ultima Cantica di Paradiso della Comedia del clarissimo Dante Poeta famosissimo della casa delli Alinghieri (sic) nobile famiglia della cicta de Firenze. Deo gratias. The readings are, I think, somewhat inferior, and include some very daring alterations, e.g. Inf. xii. 134 Quel Thotila 1 . xxx. 86 Con tutto ch' ella^izrz. (So Q and'}.) Par. xi. 21 ond' io ragione apprendo. Add further: Inf. xii. 16 ; xviii. 12 ; xxiii. 83 ; Purg. iv. 72 ; x. 30; xviii. 83; xxii. 98; Par. i. 135. Inf. xxxiii. 87 is written thus porre a tal >fr. Unassimilated forms abound, such as pecto, saecta, lancilocto, scripse^ &c., &c., and the vulgarism of the inserted initial 'h' is common. After the Paradiso comes the Capitolo of Jacopo with the words Respice Finis (sic). Laus Deo. Amen. Then follows : " Comincia il libra della -vita e studi et costumi di Danthe poeta clarissimo composto novissimamente da lionardo qui nuccupabatur (sic) dns leonardus Aretinus." 1 This would-be correction is perhaps due to the scribe's acquaintance with Giov. Villani who in B. n. c. i. erroneously speaks of " Tattle flagellum Dei." 382 596 ACCOUNT OF THE MSS. This curious life (in which that of Boccaccio is severely criticised from time to time) has been several times printed in editions of the D. C., beginning with the Paduan edition of 1726 27. It was translated into English by Boyd in 1785. i (Bat. 478). In the British Museum (3460 Harl.). A folio MS. on paper (13x9). The writing, which is in a kind of print-hand, is rather sloping and rounded, with a cursive tendency in some pages, while in others the letters are isolated. The colophon gives about the date one would have expected. It is as follows : " Explicit tercia pars comedie Dantis tractans de Paradiso, scripta et finita per me Martinum de bonsegnoribus de archuli laudensis (? Lodesano) ano 1469 die xx mensis Ottobris in die Veneris. Deo Gras. Amen." Then follows the ' Credo' of Dante, " lo scripsi" &c. as far as " messo alfondo" It begins with the Capitoli of Busone and Jacopo, and it has no title except " Infernus prohemium." The readings seem to me to be very poor, including many of a secon- dary character, and also not a few audacious conjectural alterations ; and the MS. consequently has apparently little value as an authority. One peculiar feature of this MS. is that it contains several half- finished lines, and lines with a word or two left blank, especially in the Paradiso, as though the exemplar MS. were defective or illegible, or possibly the scribe, writing from dictation, may have missed portions occasionally \ Par. v. 88 and xxix. 4 (for which see supra, pp. 449, 496) are instances which are possibly to be explained by the difficulty and uncertainty of the reading. Par. vi. 117, 120 (the former stopping short at amore and the latter at non It) and ix. 63 are defective without any such apparent cause. In Par. ix. 44 and 81, the unusual words may have puzzled the scribe. In Par. xi. 5 I suspect that the words seguendo sacerdozio may have been omitted for theological reasons, as in the well-known case of Inf. xi. 8. In Par. xi. 82 the scribe, or a Siop&or))?, seems to have hesi- tated between the readings ferace and verace, for at present the MS. has only erace, but with marks of erasure. Purg. xxviii. 131 is left blank, only the initial letter 'e' being written. Probably the scribe was puzzled by the word ' Eunoe? The following are a few peculiar readings noted : Inf. ii. 22 Del quale al quale. (So 9 and /.) 55 quanto la Stella. iii. 59 Guardai e vidi. (So only b and I.} 79 Poscia cogli occhi. (So /.) 1 A similar peculiarity occurs in v, (f>, 4, 8, and 95. EXAMINED OR COLLATED. 597 Inf. xvii. 95 altro periglio. (So only I and 77.) xix. 64 spirito storse allora i piedi. Evidently a conjectural alteration to avoid the supposed difficulty of tutti as applied to piedi. \So also U, d (marg.) and I. Com- pare 77 which has ' sparse allora. '] Purg. xxi. 19 poi andava. (Cf. B, I, q and a few more.) xxxi. 43 Tuttavia mo perche. (So A.) Par. xxiii. 67 Non e parecchio. Add to these : Inf. xiii. 21 ; xviii. 42 ; xx. 69 ; xxiii. 83 ; xxx. 1 8 ; Par. xxvi. 136. Many of the readings here noticed are peculiarly audacious conjectures. This MS. has most of them in common with '/,' with which MS. it is evidently connected, as indeed Dr Witte has already noted in his Prol. (p. xiv). Dr Witte says (Prol. pp. xiv, Ixxv) that this MS., Batines No. 133, and Batines No. 243 (at Mantua) are nearly identical. The last named I have not seen, nor have I had the opportunity of com- paring the other two (i and /)* more minutely than by the passages above collated. These shew most remarkable and frequent coincidences, but by no means an identity, between the two MSS. I suppose Dr Witte's inference was based on an examination of Inf. Canto iii. only. Such data would be very insufficient, as I have shewn in my Proleg., pp. xxxiii, xliii. Indeed in the Paradiso I do not trace much resemblance between the two MSS. (See Collations.) J- In the British Museum (2085 Egerton). A 4to. on paper (iijxSf), with clear but ugly writing, and very slightly sloping, something like Bodleian F, but inferior, and in any case of the I5th century. It is defective in places. It begins at Inf. i. 71 (' vissi a Roma''}. Inf. vi. 24 to vii. 72 and Par. xxxiii. 10 to 99 are also wanting. It abounds with peculiar variants, bold alterations, and very dis- tinctly conjectural readings. The most startling is perhaps the audacious alteration of Par. xxvi. 136, where, in order to vindicate Adam's orthodoxy on the doctrine of the Trinity, El si chiamb da poi has been altered to Et tre si chiamb pot \ The following also may be worth recording : Inf. ii. 22 II quanto e '1 quale. (So p and 66.) v. 104 di costei. xi. 8 d' Anastasio Papa quarto 2 . 1 / is in the Bibl. Riccardiana at Florence. 2 Probably a mere blunder for gitardo, since, though 'Anastasio primo' 1 is said by Dr Barlow to occur in one MS., there could be no point in introducing 'Anastasio quarto' 1 who was Pope 1153-4. ACCOUNT OF THE MSS. Inf. xxii. 6 e far lor giostra. xxx. 86 Con tutto ch' ella giri undici miglia. Purg. vii. 15 il maggior s' appiglia(I). xxiv. 65 di se fanno schiera. xxv. i Gia era uve il salir. xxvii. in alloggian. (A commonplace substitute for albergan.) Par. xxvii. 100 Le parti sue illustrissime. (So ' e.') See further: Inf. xviii. 12; xxi. 14, 93; xxv. 120; Purg. ii. 93; v. 38; xix. 35, 132 ; xxii. 5, 98 ; xxv. 31 ; xxvi. 72 ; Par. i. 135 ; ii. 13, 42, 141 ; x. 112 ; xi. 82 ; xxv. 3 ; xxviii. 23 ; xxix. 4. It is evident that the writer (like the scribe of the Bodleian MS. 'F') allowed himself almost unlimited license of conjectural alteration. k. In the British Museum (932 Egerton). A large folio on paper (15! x iij), late in the I5th century. It is called on the binding "Colonna MS." It has three columns to a page and the text is broken up into irregular fragments, between which is inserted by the same hand, an anonymous Italian Commentary, which is apparently later than 1439, since that date appears in the remarks on Purg. viii. (Catalogue). The title is : " Qui comenza la comedia di Dante Alighieri di Firenze, ml quale tracta dele pene et punimenti deli vitii et di meriti et premii dele virtu. Comincia il Canto primo de la prima Chomedia chiamato Inferno." There is no colophon. In the matter of orthography I noted : drento and di custu (Inf. v. 104) and many such forms as vo Ilia, pillia^ accollia, &c. In Inf. iii. 1 14 it has the very rare (and bad) reading Rende. We may add as also peculiar readings : Inf. iii. io6f si ristrenser. - 1 08 ciascun che die (sic). nyt Per cenno. xxxiv. 13 Altre stavano a giacere et altre stan recte (!). Purg. iv. igt Minor aperta. viii. 64! A me s' accolse. xxxi. 43 1 men vergogna. Add further: Inf. ix. 54t; xii. i6f; xvii. 95; xxi. 93t; xxxiii. 26f; Purg. xxi. 19 1; xxxi. 29 1; Par. xvii. 9t- t In all these passages 'k' and ' i' coincide, and it is clear that a very close relationship exists between these two MSS. It may be noted that both agree in preserving the rather rare, and I believe correct, reading issa in Inf. xxvii. 21. EXAMINED OR COLLATED. 599 1 (Bat. 477)- In the British Museum (3459 Harl.). A folio on paper (13x9^) of which the writing is thoroughly tondo and cursive, and abounds with flourishes and abbreviations, making it very hard to read. It is pro- bably quite late in the I5th century, and perhaps about 1470. On a fly- leaf at the beginning we read the following note : Questo libra do per /' amore di Dio al luogho de' frati di S"co [several words obliterated here] di firenze. Bartolomeo di Giovani torniaio el quale disse voleafusse per elemosina per /' at a [ammo] de dec to Giovani suo padre e vole sia messo o vo \pwero~\ posto in libreria [ob- literation] de detti frati accib preghino dio per lui e per Fata de decto suo padre e deli altri suo (sic) proximi. E questo do a d\ 9 de Marzo 1487. The title is *i< Incipit Liber dantis alaghieri, and the colophon Finito il terzo libro detto paradiso. Laus X'. Amen. Then follow the Capitolo of Jacopo di Dante, a table of contents from the Paradiso, and sundry canzoni. There is an Italian Commentary throughout. The MS. contains a large number of thoroughly secondary and inferior readings, and not a few of the most audaciously conjectural character, but it is curi- ous that the readings in the test-passages in the Purg. seem remarkably good. The scribe was evidently an ill-educated person, as he inserts the initial letter 'h' constantly, e.g. kottavo, hombra, hopra, hornata, horfeo, hongni (pgni), honde, ho (both for the conj. and interj. commonly), hudito, &c., &c. The following are some of the distinctly inferior readings : Inf. ii. 55 ; vi. 43; xvii. 20; xviii. 12; xix. 23; xxv. 120; xxviii. 83; xxx. 18,44; xxxii. 47 ; xxxiv. 53 ; &c., though none of them are uncommon. This indicates an inferior foundation-text at any rate for the Inferno. These are some of the conjectural type, or otherwise peculiar readings: Inf. xviii. 43 Mentr 3 io affigurarlo. (See I. 40.) xxiii. 82 Volsimi e vidi. xxx. 86 Con tutto che la bolgia he (i.e. e) undici miglia. Purg. i. 33 Ch' al padre piu non dee nessun figluolo. ii. 26 Finche i primi. xiv. 133 qualunque m' offende. xxvii. 8 1 S' e appoggiato. Par. xi. 91 dritta intentione. xxvii. ico velocissime. (So t also.} Add further : Inf. ix. 54, 70 ; xii. 16 ; xiii. 20 ; xvi. 104 ; xxxiii. 72 ; Purg. ii. 44; xvi. 145; xix. 34; xxii. 98; Par. xxviii. 23. Par. xi. 19 21 is omitted without any indication that the omission was noticed. 600 ACCOUNT OF THE MSS. m (Bat. 481). In the British Museum (3581 Harl.). A MS. on paper, in small folio (i I \ x 8) bearing its date in the colophon thus : Finita la terza et ultima Canticha della Comedia di dante Alighieri excellentissimo poeta fiorentino scripta per lo sopra detto Geri. Finito a d\ 4 d' Ottobre a. d. 1464. [' Sopradetto ' refers to similar colophons to Inf. and Purg.] Then follow the two lines : Qui scripsit scribat semper cum Domino vivat; Vivat in coelis semper cum Domino felix. Then ' Questo libro % di Geri di Gherardo Risaliti, il quale iscrisse di sua propria mano. E finillo di scrivere a d\ iiij" d' Ottobre e ore xxij il dl del glorioso Messer Santo Francesco. Correndo gli anni Domini MCCCCLXIIIjV Next we read as follows : 'Versus qui stant super sepulchrum Dantis Jura monarchies] &c. In line ^propriis was first written, and altered by the same hand to patriis. Then we have the Capitoli of Jacopo di Dante and of Bosone, and after them a poem of 47 terzine entitled 'La morte di Dante Alighieri,' ending Et era Ghuelpho, onde (?)fu Ghibellino. Laus tibi Xto. Finally two sonnets (i) Sonetto di Messer Cino da Pistoia biaximando Dante che non puoxe la sua manza 1 in Paradixo, et mandalo a Messer Busone da Ghobbio cioe Selvaggia. Entra gli altri peccati del libello Che mostra Dante signior d' ogni rima Som doi si grandi &c. (2) Risposta di Messer Busone a domino Cino. The title to the Inferno is : Incomincia il primo canto della Comedia di Dante Alighieri poeta fiorentino. Nel quale si diinostra come voleva pervenire alia cognitione della virtu. Et percib conoscere li appariscono le tre furie, e de procmio a tutto F inferno. Incipit feliciter. Laus Deo. The readings generally seem to me to be remarkably good, though Inf. xxx. 1 8 ; xxxii. 47 and Pnrg. xviii. 83 are striking exceptions. The following seem peculiar : Inf. xxi. 93; Purg. ii. 44 ; xix. 35 ; Par. xxvi. 134. Add Inf. xxi. 14 Altri fa remi e chi rivolge sarte. xxiii. 63 per li monaci fanno. I noted in the matter of orthography suo sposa, drento (for dentro}, ga =gia (xxxiii. 26) (compare D passim} ; also final m and n are freely interchanged (as again in D). 1 An archaic form of amanza. EXAMINED OR COLLATED. 6oi n (Bat. 505). In the British Museum (2567 Egerton). This MS. (12 x8J), successively the property of Roscoe, Ugo Foscolo, and Panizzi, has a considerable celebrity. It is especially praised by Dr Barlow as " among the very best he has ever seen for the correctness of its orthography and the excellence of its readings." This is scarcely borne out by the samples which are quoted by Dr Barlow himself. I quite agree that its text is generally fairly good, but it has been very much altered by a Stop&ori)?, and some of its readings are very distinctly secondary. It is amusing to compare with the extravagant praise of Dr Barlow the as unduly severe condemnation of this MS. by Foscolo, Disc. pp. 427 8. The latter however is endorsed by Dr Witte, Proleg. p. xliv. The MS. is on paper and is dated " 27 Feb. 1379," The colophon is Deo Gras. m'cc^lxxix" feraie xxvij die februarii. The following are a few samples of orthography and readings, indi- cating North Italy as its locality, consistently with Ferate (i.e. Ferarie) of the above colophon : fuor (i.e. for, or fur) ; rasgion (v. 39) ; rasigion (xxxii. 136) ; casioni (Par. xi. 21). So indusgio,pertusgio,palasgio,Anastasgio\ even basgiato (for baciato}, &c. Also chesto, chei, and chil are written for questo, quet, and quel. Distinctly Latin and other unassimilated forms occur : e.g. cognobbi rapto, profundo. The following may be noted as peculiar : Inf. xxi. 113 ; xxiii. 83 ; Purg. xxx. 73; Par. xviii. 75 ; xxviii. 50. O (Bat. 255). This is a I4th century MS. in the British Museum (26836), about 1380 (Catal.), (size lof x8, but shrunk by fire), on fine vellum and very beauti- fully written. The initial letters to the Cantos are finely illuminated, and the titles are in red (compare Lord Ashburnham's f and ). It is unhappily very much damaged by fire, though carefully repaired as far as possible, and it is also very defective. It begins at Inf. ii. 27. The Purg. begins at ii. 25, and the following portions (besides minor gaps elsewhere) are missing: Purg. xxii. 135 to Par. ii. 24 ; Par. xiii. 18 to xxv. 13. The MS. ends at Par. xxxii. 53. It is described in Batines (No. 255) as belonging to the Library of the Marchese Archinto at Milan. The following are noteworthy : Inf. iv. 141 ; ix. 54 ; xii. 16; xxiii. 83 ; xxx. 44 ; xxxiii. 78 ; xxxiv. 53. 602 ACCOUNT OF THE MSS. A MS. on paper (lof x8) in the British Museum (26771), said in a MS. note on a fly-leaf to be of the end of I4th or beginning of i$th century. This is undoubtedly too early. It is probably (as estimated in the Catalogue) not before the middle of I5th century. The letters are rather wide apart, slightly sloping, generally rounded in form, and have somewhat the character of printing. The z's are regularly dotted. The writing becomes more flowing and sloping as it goes on. The MS. is defective in several places. The first two Cantos of Inf. are missing. Also Inf. xv. 73 to xvi. 77 ; Par. xvi. 55 to 115 ; xxiii. 123 to xxiv. 54 ; and xxxii. 148 to xxxiii. 129. It has a marginal note (? in Libri's hand), " Antaldino II 1 . Lo acquistai in Bologno (sic} dal Librajo Guidotti 1' anno 1814." It has also a fair number of marginal variants apparently prima manu, I noticed very few peculiar readings, and many that were very good and primitive in character. It seems to have a very good text at the bottom, and it is in frequent and marked relationship to Witte's '/?.' The following may be noted : Inf. iv. 141 ; ix. 54; xviii. 42 ; Purg. ii. 93; Par. xxvii. 57; xxviii. 50. Also at Purg. vi. 109 we have the rather uncommon reading oppressura. After two colophons of ordinary type, viz. (i) Finito libra isto sit laus et gloria Xto ; (2) Explicit tertia cantica, &c., follow the two lines Qui scripsit, &c., as in ' m ' (q. v.\ with the addition Qui michi furatur ut reddat ut moriatur. (Qy. \sut=aut?} On the fly-sheet is a sonnet in a later hand, entitled ' Sonetto facto per 1' amore di margarita da gente busini. Mori a di xj d' ottobris 1527 la nocte seguente a hore v.' After the sonnet the word Valete, and then Epitaphium ad idem. There is appended a modern note stating that the MS. belonged to the Library of the PP. Cappucini di Lugo. This remarkable treasure was secured for the British Museum at the Sunderland Sale in 1882 (registered in Catalogue as 31918). Its singular value depends on its containing the extremely rare Commentary of Guido Pisano. It was apparently unknown to Colomb de Batines. At any rate he mentions only one copy of this Commentary, and that as existing at Milan in the Library of the Marchese Archinto (Batines, No. 256). I was informed recently at Milan that the library in question was sold some time ago, and I was unable to trace this MS. From the description 1 This is to distinguish it from T commonly known as 'Antaldino I'. See Colomb de Batines n. p. 215. EXAMINED OR COLLATED. 603 of the illuminations given by Batines it is evidently different from the MS. now under consideration, though it will be seen that the contents of the two MSS. (including the voluminous terza rima additions at the end) are very similar. The writing is extremely beautiful, very upright, sometimes almost sloping to the left. The paper is very fine, almost like parchment. (It was indeed erroneously registered as ' on vellum ' in the Blenheim Catalogue.) Its date is probably early in the I5th century. The margin is very broad, and the paper unusually large (ijfx nj). It contains the Inferno only, which occupies about \ of the volume, the rest being the very voluminous Latin Commentary above mentioned. It commences thus Incipit prima cantica Comedie excellentissimi poete Dantis Alagherii florentini distincta per xxxiiij Cantus. Incipit primus. The colophon is : Explicit prma cantica comedie Dantis. The Commentary is headed : Expositiones et glose super comediam Dantis facie per fratrem Guidonem Pisanum ordinis beate Marie de monte Carmelli ad nobilem virum dmn Lucanum de Spinolis de Janua. Incipit prologus. Colophon : Expliciunt expositiones et glose super primam canticam comedie Dantis alagherii excellentissimi Latinorum poete. Deo gras. Amen. After this five leaves of poetry with minute marginal notes. " Incipit declaratio^ &c., as given in Batines II. p. 137. Then after the Prologo (as in Batines), eight Cantos in terza rima of about 24 or 25 terzine each. Then the colophon : " Explicit declaratio super primam canticam co- medie Dantis." This is not the place to describe in detail this very interesting and generally unknown Commentary of Guido Pisano. Suffice it to say that from internal evidence it seems to have been composed within about ten years of Dante's death, and therefore ranks among the four or five earliest existing Commentaries. It is to be hoped that it will not long remain unedited. In the text of the Inferno I have noted the following among peculiari- ties of orthography : Elli and li are written regularly for egli and gli, and similarly pillia, ricollie for piglia, ricoglie, &c. ; comi, percht, &c. for com' z', 47.) xxvii. 100 velocissime. (So only 1.) Par. xv. 67 reads thus La tua... (the rest blank}. Add further: Inf. xiii. 21 ; xxix. 46; xxxiii. 26 ; Purg. v. 136; vii. 51 ; xix. 34; xxxi. 29; Par. i. 135; xi. 82; xxii. 17. There seems to be a remarkable connexion between S and t, at any rate in the Inferno, and, besides the passages given above, many other instances of their agreement will be found in the collations supra. In fact in the Inferno, in 63 passages in which the readings of both were recorded, I found 54 agreements and only 9 disagreements (omitting several passages in which nearly all MSS. agree). In the Purgatorio on the other hand, in 40 passages there were 20 agreements and 20 disagreements, many of the latter being of a very striking kind (as may be seen on reference to the test-passages) ; so that one would say on the whole there is no evidence of connexion between the texts in the Purg. In the Paradiso the disagreements, both numerically and in individual importance, exceed the agreements. This is another illustra- tion of the fact noticed elsewhere, that the character of the text in the several Cantiche in the same MS. is often very different. EXAMINED OR COLLATED. 607 u (Bat. 512). Also at Holkham (No. 516 in the Catalogue). A MS. on paper (iix8J), written in one column, and probably of the first half of the 1 5th century, and perhaps not very different in age from E (dated 1443), whose writing it in some degree resembles. Title : Incipit prima cantica Dantis in q. df infernum et di-vidif 1 xxxiiij ca x , 1 hoc tfo (sic ?) p" c ponit dispositionem sui (sic). Colophon : Exspilit (sic) liber terzie comedie dantis Allegherii, et pe qus (? paries cujus) omes tres in hoc volumine sunt explete. Deo gras. Amen. The orthography is archaic, and also apparently Venetian in dialect. Inter alia, l ke' is found for che frequently. Hence it was probably copied accurately from an older exemplar, if the opinion of Batines (II. p. 100) is correct, that this spelling was abandoned about the middle of the I4th century. It may be noted that the much earlier MS. C, which this MS. seems connected with (see below), uniformly has ke. The following are some of the peculiarities noted : caxione, raxione, tucto, racto, ecterno, baracta, &c. ; dilletto, efetti, magior, salvagia, conubi, lassavan, crucifixo, pinsiert, parllava, il to talento (ii. 81). There is a distinct relationship, I think, between this MS. and the Bodleian C, especially in the Purg. and Par., where they seldom differ, and most remarkably in the former. In the Purg. I noticed at least 50 cases of agreement (some no doubt proving little or nothing, but some very striking), and only five cases of difference, and these extremely slight and insignificant (viz. viii. 94 (bis) ; xiii. 2 ; xxi. 45 ; xxvi. 72), with the exception of xxx. 73 (bis), where 'u' has a reading very rarely found elsewhere, viz. s 1 io son ben Beatrice. Among the striking agreements are : Purg. iv. 72 Che mal seppe. (So only C, F and 8.) vii. 51 u' non saria che non. (So C only.} x. 14 lo scender della luna. (Ditto.} xxi. 61 suo voler fa. (So C and a few others.) xxii. 5 M' avea. (Ditto} xxv. 31 displego. (So only C and A*.) It is also instructive to observe that in four other places where C and u differ, the text of either one or other bears marks of erasure ; e.g. Purg. viii. 64 ; xiv. 26 ; xviii. 83 ; xxvii. 4. In the Paradiso there are only seven or eight cases of disagreement recorded, all (as will be seen on reference) very slight differences, except possibly xxvi. 96, where C has la ti dico, and u (with x only) neladico (? = non la}. In the Inferno the relationship still seems to hold, but is much less close, some few of the 1 6 or 17 differences recorded being 608 ACCOUNT OF THE MSS. rather noticeable. So however are some of the agreements, as the lect. fals. f si mosse ' in iv. 23. ii. 28. ' Andovvi Paulo vaso dilectione ' is I think peculiar. So also is Purg. x. 30, dritto di salir (where comp. o-). Unusual readings will also be found noted at Par. ix. 129 ; xxvi. 134, in which cases however C differs. v (Bat. 514). In the Library at Holkham (Catalogue No. 517). Put by Roscoe towards the end of i $th century ; but I should have thought about the middle, as it somewhat resembles the dated Bodleian MS. E (1443), though with more tendency to a print-like hand, and so probably later (cf. v, &c.), but still distinctly earlier, I should think, than the next MS. 'w,' which is dated 1474. It is very clearly and carefully written, but it has a very late and poor type of readings. It is on paper (size 13! xg), and written in double columns. It contains the Capitoli of Jacopo and Busone after the Paradiso. The following are peculiar : Inf. viii. 78 che terra fosse. Par. xxvi. 134 E i s' appellava. xxvii. 100 Le parti sue verissime. Add Inf. ix. 54 ; xxiii. 83 ; Purg. xxv. 9 ; Par. i. 135. gligiegrige in Inf. vii. 108 may perhaps be noted. Also the vulgar initial h is frequent, e.g. hor, hodii, &c. W (Bat. 515). Also at Holkham (Catalogue No. 518). A 4:0 MS. on paper (jof x 8), with rather flowing writing, and dated 1474, as appears from the following colophon : Questo Dante Z di marabottino di Tuccio Manetti scripto di sua proptia mano. Chillo leggie prieghi Dio per lui. 1474. On the reverse of the last page are 52 indifferent lines by the tran- scriber beginning A chi in questo mondo pecora si fa. Signed Fecie Marebottino Manetti. In several places lines have apparently been left blank, generally, as it would seem, where varr. lectt. occur, and words or portions of lines are often filled in later with paler ink (comp. v, , i, 4, and 8 for a similar peculiarity), e.g. Purg. xiii. 154; xvi. 145; xxi. 61 ; Par. xvii. 21, &c. The MS. is defective from Inf. i. 55 to ii. 81 inclusive. EXAMINED OR COLLATED. 609 The following are some notes of its orthography : cipta, giptb (=gttto, Inf. vi. 27), ripto = ritto in Inf. xix. 52 (yet fitto in 1. 50), Sogdoma ; and the vulgar initial ' h ' passim, e.g. hopere, hocchi, homai, honde, hombre, hornate, &c., &c. The following readings are noticeable : Inf. xviii. 91 ; xxx. 18 ; Par. xiii. 27 ; xxvi. 136. N.B. The Library at Holkham also contains a beautiful MS. of the Convito, on vellum, in folio (13^x9^), and early in i5th century, as I should suppose. It has no title, beginning abruptly, " Si come dice il philosapho" (sic), &c. It has however the following colophon in large capitals : Explicit. Deo Gratias. Amen. Per me Antonium Marii Frandsci Nini. x (Bat. 508). In Sir Thos. Phillips' Library, now belonging to Rev. J. E. A. Fenwick, Thirlestaine House, Cheltenham (Catalogue No. 247). A most beautiful MS. on vellum (12^x9) with single columns, and in a very large, clear and upright Gothic hand. The colophon (see below) gives the date 1412. It is in beautiful condition, without erasures, corrections or notes. The orthography has no archaic or dialectic peculiarities ; the scribe was, I should judge, a well educated man, and he had a very good foundation- text, though I suspect he made some alterations and probably modernized the orthography. It contains the whole Comedia, though stated erro- neously by Batines to contain the Inferno only 1 . On the first six pages we have ' Rubriche di Dante 1 i.e. contents of the Cantos, in vermilion, and beautiful Gothic writing. On page 7 : Qui comincia il i" 18 Canto delta prima Cantica di Dante Allighieri cictadino di firenqe nel quale /' autore fa prohemio a tucta la comedia universal- mente e generalmente, &c., &c. (about half a page). On page 8 : Prolago (sic) sopra la prima Cantica della Comedia di Dante Alligheri cictadino fiorentino. Dante poeta sovrano corona e gloria della lingua latina di natione et di costumi (this is curious) fiorentino, uerga gentile della antica e nobile pianta, &c., &c. (8 pages). Then a splendidly illuminated page with yet another title : Incliti vatis et Poete celeberrimi Dantis Allegherii cims honor ab His florentini Comedie prima pars que Infernus dicitur feliciter incipit, ac ipsius partis primum et pro hemiale capitulum. 1 Batines is also in error in his account of the other MSS. in this collection : (r) in No. 506 in omitting to mention the Purgatorio, unless it were on the grounds of its being of a different origin to the rest of the MS. [see under 'y']; and (2) in registering No. 509, which does not appear to exist. Possibly he may have written Paradiso for Purgatorio in the description of 509. In that case the former error would also perhaps be explained, the Purgatorio now bound up with 506 being numbered as a separate MS. D. L 39 6lO ACCOUNT OF THE MSS. After the Inferno we find only the words : Explicit Infernus. Then heads of Cantos as before, and another splendidly illuminated page and title as follows : Miserationis divine indagatoris eximii Dantis Allegherii prelibati. Secunda pars ipsius Comedie que Purgatorium dicitur, Colophon : Explicit Purgatorii (sic). Then heads of the Cantos of the Paradiso and magnificent illumina- tion again. Colophon : Scrip to per mano di me Paulo di Duccio tost di Pisa 1 . Negli anni Domini MCCCCXIJ. A dl xj di Gennaio. Deo gras. Then follow the 'Credo' of Dante and the Capitoli of Busone and Jacopo. The following readings are to be noticed : Inf. iii. 1 06 Poi si rivolser. ,. xvi. 122 il pensier tuo. Purg. i. 119 Come gente che torna a suo cammino (curiously violat- ing the rhyme, from a recollection apparently of Purg. ii. n). Add further: Inf. ix. 54; Purg. iv. 72; Par. xxvi. 96; xxxiii. 126. The terzina, Purg. xii. 52 4 is omitted, the space being left blank. Also the two lines Inf. xxiv. 84, 5 are omitted with space left. y (Bat. 507). Also in the Phillips Library at Cheltenham (Catalogue No. 9589) and formerly in that of Lord Guildford (see Batines). A MS. on vellum (8Jx6f), with 244 leaves, single columns and wide margin. It is evi- dently made up of two different MSS. the Purgatorio being by a different scribe, and not nearly so well written, the letters being more rounded and sloping and the ink inferior. The writing of Inf. and Par. is small, but very clear and upright, and the ink is very black. Moreover the Inferno and Paradiso are a palimpsest, not so the Purgatorio. Some pages are seriously damaged by damp and some by fire, and almost illegible, sometimes entirely so. On the first page is an illuminated initial letter and at the bottom a coat of arms with two cherubs as supporters. Title : Incomincia la commedia di dante aliighieri (sic) di fioren$ a nella quale tratta, &c. Colophon: Finita la prima cantica, &c. The Purg. has no title. Leaves are cut out at the end of Purg., but only one line is lost, as it ends with novelle fronda (sic). 1 There are 3 other MSS. recorded in Colomb de Batines as written and signed by this amanuensis (Batines omits to mention the fact in this case), viz. 431 (Paris) dated 1403; 261 (Milan) dated 1405; and 144 (Florence) dated 1329, an obvious blunder, and probably intended for 1429. Cf. A and V 3 for similar errors, and also Witte Proleg. pp. Ixvii and Ixix. EXAMINED OR COLLATED. 6ll The Paradiso has the title : Cominda la iij* Cantica della comedia di Dante Allighieri, &c. ; and the colophon is : Finita la terza Cantica, &c. (as in ////".) There is no date, but I should suppose the MS. to be early in the I5th century in respect of the Inf. and Par., the Purg. being somewhat later. It should also be added that the Purg. displays dialectic peculiarities which are not found in the other two parts, especially in respect of neglect of th^e double letters, e.g. prendette, come vene -velloce (ii. 51), fratte (frate), presura, Vergillio, drito, filla (xxi. 25), s 1 atutta (xxvi. 72), possa (xxvii. 81), &c. There is a page lost including Purg. xviii. 79 138. There is an occasional marginal Commentary, by the same hand apparently, though in smaller characters. This does not extend however beyond the middle of Canto ix. of the Inf. The following readings may be noted: Purg. iv. 72; xxiv. 61 ; Par. i- 135- z (Bat. 506). Also in the Phillips Library (Catalogue No. 8881). A MS. on paper (iijx8j) and in single columns. It begins with i CapituW of the several Cantos, and those to the Purgatorio end 'Anno Mccclxiij? The writing is not unlike that of Lord Ashburnham's Malaspina MS. () and therefore probably c. 1370 80. After the Inf. is the colophon: Explicit liber primus dantis. Incipit secundus. Somewhat similar colophons occur after Purg. and Par. It is frequently corrected by erasure. A page is lost including Inf. xxi. 25117. There are occasional fragments of a marginal Latin commentary. The orthography is very peculiar, and it seems to indicate (as is so very common) a Venetian origin, e.g. mezo, pasai, afanata, eror, posente, presura, cita, credete, cala (=calla in Purg. iv. 22), ripossato, gironne, dipella (Inf. xxv. 120). The dislike for the g sound and preference for s and z is remarkable, and points in the same direction : e.g. mazor (also magior), rasion, casone (Purg. xi. 21), si zorna (sic) = soggiorna (Par. xxxi. 12), rizunse=^rigiunsc (Purg. x. 15), conzunta = congiunta(Par. ix. \\f^],ra^o = raggio, &c. Add further cumpunto, fuosse (= fosse), lassate ( = lasciate\ siemo = scemo (Purg. x. 14). Archaic and Latin forms of words are also common, e.g. dixi, ripa (Inf. iii. 86, in spite of rhyme), &c. z has Polio both in 131 and 136 of Par. xviii. (where it has also Pierd) though the usual Paolo occurs in Inf. ii. 32. In Par. xxxi. I dunque is written donqua 1 . The readings generally are of an ordinary type. The following may be worth noticing: Inf. xii. 66; xiii. 63; Par. xviii. 75, 131 ; xxvi. 136. 1 Hence we see the easy connexion between the two variants in Purg. xxiii. 36 E quelli dunque and E quel c/' tin' acqua. 392 6l2 ACCOUNT OF THE MSS. a (Bat. 163). This is a very celebrated and interesting MS. in the Magliabecchian (now 'National') Library at Florence. It is generally referred to as 'il Codice Poggiali,' being the most important and best known of that valuable collection. Its present Catalogue description is ' Palat. 313' [formerly No. 178]. It is on vellum (size ufxSf), and is much damaged by age and damp, which gives a prima facie plausibility to the extremely early date which has frequently been claimed for it. The writing is large and clear, but very rude and coarse. Each line is divided into two parts (a peculiarity which I have never met with elsewhere) thus : Poi chei posato un po cho il chorpo lasso Ripresi via per la pia gia deserta Si chel pie fermo era sem pre il piii basso. The text is surrounded by a Commentary, written by the same hand as the text. A full account of the MS. is to be found in Batines (n. pp. 88, &c.); in the Catalogue of the Esposizione Daniesca of 1 865 ; and also in the 'Manoscritti Palatini' of Francesco Palermo 1 . But the most complete discussion of the MS. and of the various questions that have been raised about it, and one which by its thoroughness has quite superseded all that has been previously written, is contained in a monograph lately published (1888) by Prof. Luigi Gentile, whom I have to thank for courteously sending me a copy of this interesting pamphlet. His very minute and careful examination of this MS. has thrown an entirely new light on the controversy as to its age and character, as will be seen from what follows. Very extravagant claims have been set up for its high antiquity, and it has frequently been regarded as dating from 1332, 1330, or even 13292, and therefore as being the most ancient now existing. This has been argued chiefly on three grounds : (i) The broken lines, which manner of writing is said to be an indi- cation of extreme antiquity. This however is nothing but a mere guess, quite unsupported by evidence, and most effectually refuted by the fol- lowing very curious discovery of Prof. Gentile. Some of the leaves of this MS. are palimpsests, and five of these in particular are a palimpsest of a previous copy of the Divina Commedia itself. Moreover the earlier 1 Facsimiles of some portions of the MS. will be found in Palermo's Manoscritti Palatini, and also in the ' Collezione Fiorentina di facsimili paleografici.'' 2 In particular it has been thought to be a MS. which has been long lost, but which is known to have belonged early in the i6th century to one Luca Mar- tini, and to have borne the date of 1329. Professor Gentile (pp. 17 24) clearly shews this at least to be quite impossible. EXAMINED OR COLLATED. 613 copy which has thus been destroyed is in the handwriting of the same scribe. Further, it was written in two columns, and the lines were not divided. This entirely disposes of the notion of the hemistich writing of the lines being a proof of antiquity. There is no doubt (as Prof. Gentile suggests) that the scribe adopted this unusual method to allow more room for the Commentary, which in the original copy was very incon- veniently compressed into a narrow margin. (2) Another argument has been based by Poggiali himself and others on the extreme rudeness of the illuminations, which is said to indicate a very early period of the art. Apart from the uncertainty of such an argument resulting from the difficulty of distinguishing the personal in- competence of the artist from the general condition of the art (Gentile, p. 8), the careful researches of Prof. Gentile (into the very curious and interesting details of which we have not space here to enter) have clearly shewn that there are two distinct classes of illuminations. One of them is fairly good and contemporary with the MS., since the text and com- mentary adapt themselves to the pictures, and probably by the hand of the scribe himself, since one or two that are only gummed in have on their backs portions of the two-column original MS. already referred to. These do not indicate any rude antiquity or want of skill. The others are, it is true, very rude and barbarous, but they are posterior to the text, since they conform themselves to the space left by it, and not vice versa. Why these Cantos, since they are not at the end, were left origin- ally without illuminations one cannot say. It is clear however that the rudeness of these illuminations, under the circumstances explained, is due not to antiquity but to want of skill in the artist. This argument then also, being both uncertain in principle and inaccurate in its alleged facts, falls to the ground. (3) The third argument is based on a very interesting passage in the marginal Commentary on xiii. 146, that well-known touch-stone of the age of Commentaries, according as they speak of the Statue of Mars at Florence, finally destroyed by the inundation of Nov. 1333, in the present or past tense. The passage in question is given by Batines, II. p. 89, and is full of peculiar interest in relation to other references to the same event by Chroniclers, or by other Commentators on Dante. It suggests a very interesting and rather complicated problem for discussion, but I forbear to enter upon it here, because (i) the passage has been so variously interpreted that some have argued from it that the Commentary \spriorto 1333, and others (e.g. Batines) that it is posterior to that date: and (2) in any case it can only serve to prove the date of the composition of the Commentary, not that of the execution of the MS. We are therefore thrown back upon the ordinary palaeographical considerations for forming an opinion of the age of the MS., and the deliberate judgment of Prof. Gentile is that it belongs certainly to the second half of the i4th century, and not improbably comes somewhat late in it. The text is not one of any special excellence, and certainly has 614 ACCOUNT OF THE MSS. many distinctly secondary readings. There is assuredly nothing here to imply an exceptionally early or pure source of derivation. The orthography of the text has but few marked peculiarities. In the Commentary we find ztot for riot, and the scribe there writes piazia, though the same word appears as piaggia in his text. His double letters are also somewhat irregular 1 . He does not, I think, use the archaic form ke for che, as was usual according to Batines in the first half of the I4th century (see vol. II. p. 100). The MS. is defective, leaves being lost in three places, with the result that the following passages are missing: Inf. iii. 108 iv. 16; xxxi. 37 102 ; Purg. xxxi. 79 xxxii. 64. (I take these from Prof. Gentile's pamphlet, p. 5.) I also noticed the following defective lines : Purg. iii. ends 2\spento (1. 132), and then follows l Qui comincia il caplo mi' without any space or sign of omission (Query, Was this omission 'a pious fraud' prompted by the nature of the contents of the omitted passage ?). Par. xxxi. 20 ends with the word l plenitudine? Prof. Gentile (p. 8) also notes a defect in Par. i. 32 thus : Del detta dovria la froda. This was apparently due to misunderstanding or misreading the ex- emplar text. It is curious to note that this same defect occurs both in the palimpsest copy and in the original MS., in which latter this pas- sage can be traced. The following peculiar readings are noticeable : Purg. ii. 13 E echo qual el sol presso dal 2 . x. 74 Del human principato. (A mere blunder?) xiv. 133 s' apprende. xxv. 9 Perche altezza al salitor. (So 78 only?) Add Inf. xvi. 14 ; xxxii. 47 ; Purg. vii. 51 ; xix. 34, 35 ; Par. xi. 82 ; xxvi. 134. Also Par. vi. 129 Fu 1' ovra grande e 1' ovra mal gradita. xxiv. 1 1 8, 1 19 Incomincio, La grazia che vedea Con la tua mente Ik over t' aperse(P). b (Bat. 98). In the Magliabecchian Library at Florence. A I4th century MS. on vellum (13x9), the writing very good and upright, with rather long tails and flourishes. I should suppose about 1370 80, or possibly earlier. It rather resembles the Bodleian C. The alterations by a 8iop6corfis are extremely numerous, k for ch is almost universal, e.g. ke, soverkw, &c. 1 Perhaps the lect. fals. notti for note in Purg. xxx. 93 is due to this. 2 This is by a misprint in the collations (p. 369, line 20) assigned to 'a' instead of '.' EXAMINED OR COLLATED See further the description in Batines. On fol, 100 we find "Pathaphium (sic) Dantis, quod ipse fecit" and in line 5 there is the curious variant ' longis externus ab oris.' Then an epitaph on Petrarch (3 lines) begin- ning Frigida Francisci lapis hie tegit ossa Petrarchce. Then follows an epitaph (4 lines), dni Jofils Andree. Peculiar readings will be found noted at Inf. i. 28 ; ii. 81 ; xvi. 14 ; Purg. xx. 90 ; Par. xxvi. 134. c (Bat. 7). In the Laurentian Library at Florence, known as u Codice Tempiano Maggiore" A folio MS. on vellum (14x9!), with beautiful and clear mezzogotico writing very like the Bodleian A, and dated 1398. (See Batines II. p. 12 for full description.) It abounds in assimilated and un- Latin forms of words, yet the archaic ke for che occurs, and also ka, as in Inf. ii. 55 piu ka la stella. The text of this MS. is rated very highly by Witte (Proleg. p. Ixvii): "II Tempiano e senza dubbio uno dei testi piu corretti del Poema," and certainly I found no peculiar readings what- ever (a good sign as has been noted before) except perhaps Purg. ii. 93, and that is one of very slight importance. d (Bat. 22). Also in the Laurentian Library. A folio MS. on paper (12^x9^), the writing tondo, and I should have said decidedly of the I5th century, though it is described by Batines as of the I4th. For the title, see Batines II. p. 21. There is no colophon. This MS. is interesting as being one of those which exhibit nearly all the 'Sienese' readings given by Dr Witte in Inf. Canto iii, see Dante-Forschungen I. No. 16. It is in very frequent, though by no means invariable, agreement with the Bodleian F, and the Ashburnham MS. 8, and occasionally too with I, another Sienese MS. It will be seen that there are many readings (often mere blunders), which it has in common with F only, out of over 200 MSS. examined. Inf. vii. 32 Da ogni parte. (F only ; see note under ' F.') 1 08 malvage coste grige. (Ditto.) xi. 34 toilette dannose. (An obvious blunder, corrected secunda. manu in d, but standing in F.) M 91 faccia turbata. (Also corrected in d secunda manu, but not in F.) xiv. 126 A man sinistra. (F only.) xvii. 20 Che stanno parte. (Do.) 95 Ad alto tosto ch' io montai. (Do.) \The omission or transposition of forte seems to characterize most of the ' Sienese ' family I\ 616 ACCOUNT OF THE MSS. Inf. xxv. 1 20 Per 1' una...e per altn?. (F only.} xxxi. 75 E vedi Ik giu. (Do.} Purg, v. 135 E disposata. (So F and']'] only} Per contra at Inf. xvi. 104; xviii. 55 ; Purg. ii. 44; xvi. 145 ; xxiv. 61 ; xxvii. 81 ; Par. ii. 141 ; xxix. 4, and other places it will be found that F and d are not in agreement. Where d differs from F, it is often found in marked agreement with I, e.g. Inf. xviii. 91 Ivi con cenni. (I, d and 86 only} Purg. iv. 72 La qual mal. (I, <9, d and 77 only} xxx. 15 La rivestita carne. (A rare lect. fals.) Par. i. 141 Com' a terra quieto il foco vivo. (I has 'el,' but so it has commonly for ' il.') xi. 105 Tornossi. (So I, d and 10 or 12 more out of about 130 examined here} This MS. (as also F and I) abounds with peculiar readings, and cannot therefore be considered as a high authority. The following are some of them : Inf. iii. 105 Di lor parenti. 114 Rivede alia terra. Purg. iii. 103 E un di di loro subito (later) grido chiunque. Par. xxvii. 100 Le parte sue vicine. All the above are, as far as I know, unique, and so are the following (for which see the collations) : Inf. ix. 54 ; xii. 16 ; xiii. 20 ; xx. 69 ; xxvii. 21 ; xxviii. 83 ; xxix. 46 ; xxxiii. 26; Purg. vii. 51 ; xxvi. 72; Par. iii. 15 ; xxviii. 23 ; xxix. 4; xxxiii. 126. The following are also very unusual : Inf. iv. 23 ; xvi. 14 ; xxiii. 83 ; xxxiii. 72, 78 ; Purg. ii. 93 ; xxx. 15 ; Par. i. 135 ; xi. 26. e (Bat. 3). In the Laurentian Library at Florence. A MS. on paper of somewhat unusual shape (u^x 5^), the writing of which appeared to me to resem- ble that of a dated MS. in the British Museum (1379), but the actual date given by the colophon is 1355 (or rather, by clerical error, 1255). See Batines for a full description. Like so many other MSS. it seems to ex- hibit a Venetian colouring in its dialectic forms, e.g. unde, mundo, quantu (Par. xxv. 33), puoche (poich^), suono ( = sono}, forri (=fori, Inf. xxix. 46) ; solvar(Purg. xxi. 61) ; essare 1 (Inf. xix. 3). In the epitaph, 'Jura Monarchic,' &c. line 5 reads ' patriis extents ab oris.' It is also very remarkable that this epitaph is described as 1 According to a monograph by Luigi Gaiter Sul Dialetto Veneto nel Secolo di Dante (Bologna 1877) the preference for ist conjugation inflexions is a character- istic of the Venetian dialect (p. 9). EXAMINED OR COLLATED. 617 already existing on the tomb at this early date. " Hi versus sunt scripti ravenne in tumulo dantis in introitu ecclesie beati francisci a sinistra parte parve porte ipsius ecclesie pro ejus epitaphio." This seems very definite and circumstantial 1 . The text, notwithstanding several peculiar or secondary readings, appears to me to be an excellent one. The following are some of its peculiarities : Inf. vii. 32 Da ogni lato. (? unique.} ,, xxiv. 12 li gavagna. Par. xi. 91 Sua buona intentione. (? unique.} xvii. 8, 9 Si che 1' esca Sia ben segnata dalla eterna stampa. Add to these: Inf. ii. 60; xvi. 14; xviii. 12; xxi. 93; xxxiii. 72; Purg. iv. 22 ; vii. 51 ; xvi. 145 ; Par. xvii. 9 ; xxvi. 136 ; xxviii. 23. Inferior but not otherwise uncommon readings will be found at Inf. xxvi. 25 ; xxviii. 83 ; Purg. xviii. 83 ; Par. xi. 82 ; xvi. 47 ; but these are not sufficient to detract much from the high value of the text. / (Bat. 9). Also in the Laurentian Library. A very large folio (15^x10!) on vellum, with huge letters nearly \ in. long. (See Batines, II. p. 15, for further description.) The orthography again exhibits Venetian peculiari- ties, as the following few samples will shew -.possato, possia (poscia), anghossia, bassciato, esscon, chossl, dipella (for dipela), sallita, triunfa, meschiado, &c. The text seems quite ordinary, and calls for no special remark. The following peculiarities are noticeable : Purg. xxxiii. 63 is omitted and space left blank. Inf. xxxii. 136 reads 'di cib ti piangi.' Add further : Inf. i. 28 ; xvii. 20, 95 ; xix. 3, 23 ; xxi. 93 ; xxix. 46 ; xxxiii. 74; xxxiv. 13 ; Purg. iv. 72; viii. 64; xxv. 9; xxvii. 4; Par. iii. 15; x. 112. g (Bat. 72). A magnificent MS. in the Laurentian Library (i2|x7|), with very beautiful large mezzo-Gothic letters, fully J inch high, the initial letters of the terzine gold, blue and red in regular sequence ; large margin and splendid illuminations. I do not understand Colomb de Batines (vol. II. p. 48) describing this as a I5th century MS. I should have thought it certainly Hth, and not very late in it. Further, I consider it to be one of the very best texts I have ever met with. Among the very rare in- stances of peculiar or inferior readings I noted : Inf. xviii. 42 ; xxiii. 83 ; Purg. vii. 5 1 ; xxv. 9. 1 I have discussed this in an Article on 'The Tomb of Dante' in the English Historical Revieiv for Oct. 1888. 6l8 ACCOUNT OF THE MSS. // (Bat. 27). Also in the Laurentian Library at Florence. It is a large folio on paper, the writing rounded but very upright and beautifully clear, not unlike Z, and therefore probably on the confines of centuries 14 and 15. One or two leaves occur in another and inferior hand. The character of the readings is, I think, not generally good, and it is scarcely a text of average value, and not in any respect a high authority. The value is, I think, very much exaggerated by Batines (q. v. for detailed description, vol. II. p. 22, &c.). The following are peculiar, and, it may be added, clearly inferior readings : Inf. vii. 108; xvi. 3, 14; Purg. xxi. 25 ; xxxi. 43 (Tuttavia perche vergogna porti). All of these and several other peculiarities it has in common with 79, to which it is evidently in close relation. It contains the epitaph ' Jura Monarchice" 1 &c. with the reading 1 pat is exterus ab oris" 1 in 1. 5. i (Bat. 2). In the Laurentian Library. One of the most beautifully written MSS. I have seen. It is on vellum, and dated 1347, and resembles, I think, very much the celebrated Trivulzian MS. T. The general average of the readings is not, I should judge, remarkably good, nor should I attach any special authority to the text. Witte's judgment however should not be over- looked: "e da annoverarsi trai codici piu corretti, benche alcune varianti a lui particolari non sembrino certamente provenire dall' autore " (Proleg. p. Ixx). Moreover it should be added that there are extremely few peculiar readings, which indicates at any rate a careful and conscientious scribe. For a fuller account of this MS. see Batines, II. p. 8. It is defective in several places. j (Bat. 224). In the Bibl. Comunale of Siena. It is on paper, and dated in the colophon (see Batines) 1453. It contains the Inferno only, and is de- fective at the beginning, as far as i. 73. After the colophon (given in Batines) the following barbarous lines occur: Dantis Elegerii florenti genitus urbe Conditor eloquii lumen decusque musarum Dominicis annis ter septem mille trecenis. Its readings are, I think, fairly good on the whole. Nothing peculiar was noticed except perhaps Purg. viii. 94. The following few notes on orthography may be added : eror (error), conobi, cintta,ffama, dentti, chupria* EXAMINED OR COLLATED. 619 k (Bat. 156). In the Bibl. Riccardiana at Florence (No. 1049 in Catalogue). An early i5th century MS.*bn paper (lof x 8J). It is interesting as exhibiting the 'Sienese' type of readings 1 in Inf. iii. in n places. The writing some- what resembles that of the Bodleian I, a MS. also having some (though less close) relationship with the Sienese family, but is much earlier and more Gothic in character, and has not the extremely long flourishes of the latter MS. It also has e for i regularly as the nom. plur. of the def. art., as T has. It is not nearly so closely related to the Sienese family, nor notably to the Bodleian F, as d is, but there is an evident connexion between them all. k is a volume rather miscellaneous in its contents. 1. Le laudi di frate Jacopo da Todi, in fine clear upright but some- what rounded Gothic. 2. Some pages of ' Ricordi ' of accounts and expenditure. 3. After several blank pages, the Divina Comedia (i.e. Bat. 156) occupying 104 leaves, which is the copy described above. After this two or three blank leaves, one containing the epitaph beginning Theologus Dantis &c. consisting of 14 lines. 4. Another copy of the Comedia, of the I4th century, the writing of which, at first rather good and fine, becomes coarser as it proceeds, but remains clear and upright. This is described here as /, and in Batines is numbered 133. There is a page lost in k from Inf. ix. 32 to x. 24. The following are noteworthy : Inf. xvi. 14 ; xix. 64 ; xx. 69 ; Purg. x. 14, 30 ; xviii. 83 ; xxvi. 41 ; Par. i. 135; xi. 26; xxii. 94; xxviii. 23. Par. vii. 4 'alia rota sua' is, I think, uncommon, but rota and nota would be easily confused. The general character of the readings is not, I think, above a fair average. / (Bat. 133). In the Biblioteca Riccardiana (No. 1049 bis). This is bound up in the same volume as the last MS. (q. v.}. It has four of the Sienese read- ings in Inf. iii., viz. at lines 68, 91, 102 and 127, and shews therefore a much slighter tinge of this influence than k. A note at the end of the Inferno gives the date 1392. The following are peculiar or unique : Inf. iii. 59 Guardai e vidi. (A very rare fals. lect. only in b and i besides.} 79 Poscia cogli occhi. (So i.) 1 See Witte Dante- Forsch. I. No. 16. 62O ACCOUNT OF THE MSS. Inf. iii. 123 Tucti ne vegnon. (Ne is over an erasure and is perhaps corrected from ci vegnon [found in m, e and 77] which probably arose from a misunderstanding of the abbrevi- ation cvegnon or cvegnon = convegnon.) xvii. 95 Ad altro periglio. (An audacious change found also in i and 77.) xix. 64 lo spirto storse allora i piedi. (So U, i and d marg. Comp. 77.) Add further: Inf. ix. 54; xii. 125; xxi. 112; Purg. ii. 93; v. 136; Par. vii. 114 ; xxii. 94. i and / (and often 77 also) seem to be in frequent agreement in Inf. and Purg., e.g., besides above, Inf. ii. 22 ; xviii. 42 ; Purg. xxi. 19, &c. ; but this does not seem to be the case in the Paradiso, nor has / broken lines such as Par. xi. 5, and others observed in ' i.' This also, so far as it goes, would confirm the evidence of the exemplars of the two MSS. being different in this Cantica. m (Bat. 125). In the Biblioteca Riccardiana at Florence (No. 1025). A folio MS. ^ x 9) on vellum, undoubtedly of I4th century, and I should judge with Batines (q.v. II. p. 71) soon after the middle of the century. The writing is something like the Malaspina MS. () but rather thicker, as if written with a broader pen. The first leaf seems finer and by a different hand, as if it were a sample written by a professional scribe. [Compare the Cambridge MS. ' Q.'] The text is, I think, rather below average (except perhaps in the Paradiso}, and I do not understand the high value set upon it by Tommaseo and Becchi (see Batines). Batines may also be referred to for a fuller description of the MS., its title, colophons, &c. In the epitaph 'Jura Monarchies,' &c. we have ' patriis exterus ab oris' in 1.5- The following may be noted : Inf. iii. 108 ciaschedun che Dio. (So K, a, 6.) 123 ci vengon qui. (So e and /. See note on last MS?) Purg. xxxi. 43 Tuttavia perche vergogna porti. Add further : Inf. ii. 22 ; xxvi. 25. n (Bat. 124). In the Biblioteca Riccardiana at Florence. A very beautifully written large folio (153X9!) on vellum, containing only Inf. and Purg., with a commentary in the same hand surrounding the text, as in the Bod- leian A. I should judge it to be of somewhat similar date, i.e. late in the I4th century or early in I5th (hardly so early as Batines supposes?) EXAMINED OR COLLATED. 621 and, like A, it is marked by strong Venetian dialect. The following samples will make this obvious : cossl, pellago, eror, fatore, magior, sucedete, into staxone, caxone (for stagione, &c.), rason (ragion\ esse (esce),possa (poscia) custui, dubiusi, rispitse, signur dil, dilla, digit (for del, &c.) carecar ( = carreggiar, Purg. iv. 72), saglita, alleluglando, Tro- glani (Troyani\ &c., &c. Inferno xxxiv. is lost except the first two lines. There is a curious note or colophon to the Purgatorio, " Or sum for del Borgatorio dco gratia." Batines is in error in stating that Ugolino is represented in a miniature to Canto xxxiii. as devouring one of his own sons. The Archbishop Ruggieri is clearly intended. The following may be noted: Inf. xvi. 14, 104; xxxiii. 72; Purg. xxi. 6 1 ; xxxi. 51. * (Bat. 253). This MS. in the Brera Library at Milan, containing the Par. only, is the continuation of the last described MS. in the Riccardiana at Florence containing the Inf. and Purg. (see Batines II. pp. 70, 135). It is also beautifully written (both text and commentary), and has very finely illu- minated small initial letters. For the title see Batines, and compare that of n (ib. p. 70). The colophon is "Maestro Galvano scrissit testo c la ghiosa, merce de quella itcrgene gloriosa." Overleaf are the lines (34 in number) beginning " Credo in una Seta Trinitatc." The reading in x. 112 seems quite peculiar, 'Tanto nell' alta mente Ci si' \inarg. "a/. entro "]. /t (Bat. 88). In the Magliabecchian Library at Florence. A folio MS. on paper (iox7f), containing the Inferno only. I should have thought certainly 1 5th century, the writing being thoroughly tondo (Batines says middle of I4th century, but??). It has a curious peculiarity in orthography of doubling the initial consonant passim, e.g. ssanza, ssuo, Jfalso, ttu, &c., &c. Nepetuno in xxviii. 83 is curious. See also xxvi. 25. It contains many very inferior readings and is, I think, a thoroughly poor text of little or no value. In iii. 9 it reads 1 O voi che entrate.' q (Bat. 101). In the Magliabecchian Library, known as the Codice Giraldi. It is on vellum (8^x5^), and in a print-like hand, very like d and u. It is interesting as exhibiting a very strong relationship to the Vatican family, the evidence of which will be found in the Table in App. n. ; and it is curious that v, which it resembles in its writing, also exhibits a similar t o was omitted from its liability to confusion in manuscript with o and o. 622 ACCOUNT OF THE MSS. type of text. The writing changes at the last line of Purg. xxxii. to a much thinner and more tremulous hand, but otherwise resembling the former, and probably an imitation of it. See Batines for title, colophon, &c. In orthography we may note: ke, cillia, vollia,sollion, &c. (for tiglia, &c.), also mi, tu, su occur for mto, &c., e.g. mi Virgilio, il tu talento, il su Polidoro, &c. And among readings : Inf. i. 28 ; ii. 22 ; Purg. xxi. 19 ; Par. xxii. 95. Also Purg. xxix. 14 'la donna a me tutta.' r (Bat. 174). In the Magliabecchian Library. A MS. on vellum (9 x 6J), in a sloping and rounded, but very clear hand. It has marginal notes occasionally by the same hand as the text, and consequently the variants given by them are important. For title and colophon see Batines. The text exhibits a singular combination of many very good (and sometimes rare) readings, with many very inferior ones. It also has curious and occasional points of contact with typical texts such as those of B and D, but these relation- ships are partial and capricious. The following passages may be worth noting : Inf. ix. 70 ; xvi. 104 ; xxviii. 10 ; Purg. xx. 90 ; xxvi. 41. S (Bat. 226). A 1 5th century MS. on paper in the Biblioteca Comunale of Siena. The writing is coarse and ugly, and the readings seem very poor, though none of a peculiar kind were noticed. It has a title in Italian : Incomincia il terzo et ultimo libro della comedia di dante nel quale tratta de la beat a gloria di Paradiso ; and a rather curious colophon in Latin : " Explicit tertia Cantica Comedie Dantis que tractat de superna gloria et de quibusdam gloriantibus in gloria Paradisi. Deo gratias. Amen." t (Bat. 223). In the Bibl. Comunale of Siena. A I5th century MS. on paper (ii^xSf), with two dates in the colophon (for which see Batines), viz. 1439, and afterwards 1435, m tne following words: "Ego luixi (sic] scrissi in Ancona ano 1435." ^ n spite of this (to say nothing of the evidence of the writing), it has an audacious title-page, " Dante Aldighieri, Manuscritto dell* 1st" Autore"\ It is written in very flowing round-hand, but is quite clear. Its orthography is curious, from the very frequent substitution of x for s, which produces some quaint results : e.g. qiiaxi, choxl (cos\\ dixio (desid], degnoxo, i senxi e polxi (Inf. xiii. 63) ; delxuo Pulidoro, Inf. xxx. 18 ; Yexu ( Jesu}. Inf. v. 37 reads thus : Intexi chacozi, &c. EXAMINED OR COLLATED. 623 At Inf. xii. 1 6 ; xxi. 93 ; xxviii. 83, Purg, xviii. 83, the readings are more or less peculiar. This MS. is interesting as being the typical MS. of Witte's so-called ' Sienese ' family, containing the whole list of the 20 characteristic readings collected by him. See^ Dante-Forsch. I. No. 16. u (Bat. 22l). In the Bibl. Comunale of Siena. A folio MS. on vellum (13^ x 9^), very beautifully written in clear Gothic hand, with a rather broad pen and singularly black ink. It has only one column, and the letters and lines are very close together, the margin being unusually broad ; and the MS. generally is in a beautiful state of preservation, except that it is unfortu- nately defective in one or two places: Inf. v. 91 to end, and viii. i 90. (It is Canto viii. as Ilari states, and not vii., though he is wrongly cor- rected in this respect by Batines.) In spite of the praise of Batines, the text is in no way, I think, remarkable. Inf. \. 42 and Purg. ii. 13 may perhaps be noted. V (Bat. 222). In the Bibl. Comunale at Siena. A MS. on paper, assigned by Batines to the I4th century. (See Batines, II. p. 115, for the title and further de- scription.) It is slightly defective, viz. Inf. xi. from 36 to end ; Purg. v. 77 to vi. 27 ; and Par. xxxiii. and part of xxxii. The text appears to have no special value, and has a considerable number of readings peculiar to itself, most of them having an obviously conjectural character. The following are some examples of this : Inf. \. 28; ii. 81 ; xviii. 42 ; Purg. vii. 51 ; xxiv. 61. The following are, as far as I know, unique : Purg. xxv. 131 S' intende Diana. xxxi. 43 perch non vergogna porti. Par. xi. in Che nel merito. w (Bat. 22$). In the Bibl. Comunale of Siena. A very beautifully written MS. on vellum, the letters being very clear and upright, with singularly long tails and heads : the ink is very black, and the margin broad. I should think certainly not later than c. I37O 1 , and I am surprised at the late date assigned to it by Batines, in opposition to the opinion of De Angelis and Ilari, cited in his note, and also that of Dr Witte (see below). This Codice has a special interest, since Dr Witte discovered that the frag- mentary Latin Commentary which it contains was (as far as it goes, 1 The writing somewhat resembles the Bodleian ' C.' 624 ACCOUNT OF THE MSS. which is not beyond Inf. xi.) that of Ser Graziuolo dei Bambagioli. This is a Commentary of extreme rarity. Its very existence was for some time disputed, until a copy of.it was discovered in the Bibl. Colombina at Seville, and this is, I believe, the only complete copy known to exist. I think, however, I have proved that it is identical (or practically so) with the anonymous Italian Commentary published by Lord Vernon (in 100 copies only) at Florence in 1848, and dated 1324 (var. lect. 1328). For a fuller discussion of the subject, a paper by myself in the Academy No. 492 (Oct. 8, 1881) may be referred to. It is well known that up to the time of his death Dr Witte was engaged in the publication of Ser Graziuolo's Commentary, which had been transcribed at the cost of the German Government from the Seville MS. This Sienese MS. was lent to him by the municipal authorities, and it now contains a fly-sheet with a highly interesting description of the MS. in Dr Witte's own handwriting, dated Halle, Nov. 10, 1882. In reference to the MS. itself he says : " Le lezioni sono per lo piu buone, benche non vi manchino esempi di alterazioni [giving as examples of such secondary readings, Inf. iii. 31, 36; Purg. ii. 26; vi. 1 1 1 (come si curd)\. In ogni modo il numero di nuove lezioni degne d' approvazione che potessero rilevarsi da questo codice sembra assai scarso. II carattere della scrittura e molto elegante e non sembra posteriore al mezzo del trecento? The MS. is very imperfect, beginning at Inf. i. 58, and breaking off after Par. i. 54. It seems to have a peculiarly large number of ' nuove lezioni] or at any rate many that are extremely rare, e.g. Inf. ii. 55, 81 ; iv. 141 ; xii. 16 ; xiii. 63 ; xxi. 112 ; xxiii. 83 ; xxvi. 25 ; xxviii. 83 ; Purg. x. 30; xxvi. 41. Note also : Inf. iii. 75 per lo fitto lume. (? unique?) 78 riva d' Acheronte. (So ' s.') 123 Tutti vengon qui. (So '7.') X. This MS. (not registered in Batines, being probably a later acquisition) is also in the Biblioteca Comunale at Siena (No. I. ix. 20 in the Catalogue). It is on paper (i i x 8f ) ; it is written in two columns, and the writing is flowing and has many abbreviations, making it somewhat difficult to read. The first page contains a beautiful illumination, and others, though less elaborate, occur at the beginning of Purg. and Par. It is dated 1477, as appears from the following colophon : Finito di copiare questo d\ viij di didembre MCCCCLXXVIJ per me aghostino de jac di lorenzo di canbino Rustichelghi (?) ritalgliatore. Laus Deo. Ghrazia Christo. It has four or five of the so-called Sienese readings in Inf. iii. indi- cating some slight tinge of this influence, viz. lines 62, 75, 105 (sementa), 108 (in the peculiar form ciaschedun che Iddio). EXAMINED OR COLLATED. 625 The following are a few samples of its orthography: illchorppo, fferro, portta, mezo. The number of its peculiar or rare readings is unusually large. The following are some of them : Inf. i. 28; ii. 22 ; 4 viii. 125; ix. 70; xvii. 95; xxiii. 83; xxvii. 21; Purg. xxi. 6 1 ; xxvi. 41 ; Par. xi. 26; xxxiii. 89. Nearly all these are quite peculiar and audaciously conjectural. I (Bat. 413). In the Bibliotheque Nationale at Paris. A very beautiful folio (13! x o/) on vellum. I should suppose of about the middle of the I4th century or not much later. At the same time the text is, I think, quite common- place, if not below average. It has a well marked relationship to the very much later British Museum MS. 'k,' illustrations of which are collected supra in the account of that MS., and they need not therefore be repeated here. 2 (Bat. 414). In the Bibliotheque Nationale at Paris and formerly at Padua ('de 1'abbaye de Sainte Justine')- A large folio (14 x loj) on vellum, written by Betinus de Pilis (comp. n and y) and dated 1351. It is however by different hands in different parts, and it has been very much erased and corrected. It contains only the Inferno and Paradiso. The earlier part is beautifully written, though there are careless mistakes and also many corrections and varr. lectt. The Par. begins in a huge hand, the small letters being \ inch high, and the Latin Commentary (by which the whole text is accompanied) proportionately large, its letters being about as large as those of the previous part of the text, where the Commentary was minute, though very clear. All seems to be by an entirely different hand, but still by a Venetian scribe. Yet both text and Commentary gradually become smaller, till they at last both resemble the original hand again. The orthography, though (as noted above) still Venetian, differs considerably in different parts. As the former hand reappears, the parasitic 11 (see below, and also the account of n and $) reappears everywhere, whereas it is comparatively rare in the earlier Cantos of the Paradiso. The orthography, especially in the earlier part, is (like that of n and s) of the most marked Venetian. Here are some specimens : Senecca, cortesse, parolla, vappor, Brutto (Brufo), remmi ( = remi), pasion, mezo, quela, vale, rabioso, disi, cognoser, &c., &c. The fondness for u for o appears everywhere, and the insertion of the parasitic u before o is also of extraordinary frequency, e.g. nuon, cuon, cuome, cuosi, cuosa, cuompagno,puossa (=posa), nuostre persuone, suonno D. I. 40 626 ACCOUNT OF THE MSS. (=sono), dttofe, paruole, alluor, Buoluogncse (!), guolla(gola), Puoliduoro, Truoyanni, cuompuorta, &c., &c., and also its substitution for o, as scuncia, mundo, umbra, lucco ( = loco). k appears for ch &c. commonly (this is archaic though not specially Venetian), e.g. ocki (occhi), malebranke, kiaro, inforka, fioko, poko, kon, skianti, &c., &c., ke and ka (for che), ciaskun, 'cuon luokio bieco' (Inf. xxiii. 85). Inf. ii. 26 reads thus Intisse cosse ke fuoron kaggione. In Inf. iii. 32 we have que for che. bien appears for ben in Inf. xiii. 20 and elsewhere. Other noticeable forms are quilt (quelli), ausello (augello), Maistro, Bia- trice. In the part written apparently by a different scribe the inserted u is comparatively rare, ke for che is continued but is not so frequent. The following words speak sufficiently however for the Venetian nationality of the writer : parolle, diffusso, piumme, carbonne, Polio (Par. xviii. 136, but Paullo in 1. 131), Paradisso. Add duona ( = donna), lioki (git occhi}, skieera (Par. xviii. 7$),pospuosi, rikiusa, poka. The following readings may be noted : Inf. iv. 142 (with marks of correction} reads Et vidi il geometra Tolomeo. xxxiii. 91 Puoi passamo oltre. To these we may add Inf. i. 28 ; xvi. 3. The description of this Codice with its colophons &c. in Colomb de Ratines (il. p. 227, c.) is so full that it is needless to repeat the details here. The following extract from the accompanying Commentary shews it to be a very ancient one. On Inf. xiii. 146 we read: quedam statuta (sic) Martis quae residet super caput pontis Ami, frustra fuisset edificata ipsa civitas super ceneribus, &c. 3 (Bat. 418). In the Bibliotheque Nationale at Paris. I paid special attention to this MS. on account of the extravagant praise of it by Marsand (ap. Batines II. p. 232). My own opinion of it is that it is a thoroughly commonplace text, or rather an exceptionally inferior one 1 . It is a large folio on vellum (15 x lof), and if of the I4th century quite late in it. The writing is very clear, but decidedly rounded Gothic, or mezzo-Gothic. See Batines /. c. for title and colophon. 1 Since writing the above, I find in Witte's Proleg. p. Ixiii that Marsand, meet- ing afterwards with this same MS. under another reference, failed to recognize it, and pronounced a very unfavourable opinion of it ! According to Witte this 'No. 418' is identical with ' No. 426' of Batines, which is characterized by Signor Ferrari as being "uno de' piu scorretti Codici che ci siacapitato alle mani," &c. Marsand in two other cases has described the same MS. twice, and in one of these instances with similar inconsistency ! (Witte /. c .) EXAMINED OR COLLATED. 627 In the matter of orthography, grolia and grolioso (which occur several times) would seem to suggest a Tuscan origin. The only other point I noticed was the not uncommon aversion to //, e.g. bacte, racto,factore, mocto and even ecterno and ectu (Inf. xxx. 115) occur. The following may^be noted : Inf. iv. 23 ; xxiii. 83 ; Purg. viii. 94 ; xviii. 83 ; xxvi. 72 ; Par. i. 141 ; xxvi. 96 ; xxxiii. 89. Several very inferior readings, though not specially uncommon, will be found noted in the collations. 4 (Bat. 423). In the Bibliotheque Nationale at Paris. A I4th century MS. on vellum (iif x 8|), the writing almost a facsimile of Bat. 413, and the date pro- bably about the same. For further description see Batines, II. p. 236. I noted in the matter of orthography, spollia, fillio, &c. (passim), si- nixtra, and ke for che regularly. The text seems to me remarkably good in the Inf., but decidedly inferior in the Purg. and Par. Many passages have been altered by a 8iop6u>Tr)s, who has generally substituted worse readings. Among the very few unusual readings may be noted Purg. ii. 13, 93 ; Par. xxvi. 134. 5 (Bat. 422). In the Bibliotheque Nationale at Paris. A MS. on vellum (11x7^) containing the Inferno only. The writing is very rough and entirely ' tondo] and I should judge certainly of I5th century (and not I4th as Batines says), and apparently somewhat late in it. It appears to have a thoroughly inferior text, and one of no value as an authority. We may note as more or less peculiar Inf. xii. 66 ; xiii. 63 ; xvii. 95 ; xx. 69. I am quite at a loss to understand the grounds of Ferrari's favourable judgment (see Batines, n. p. 235). 6 (Bat. 425). A MS. on paper in the same Library. Batines says of the I4th century, Marsand of the I5th. I am inclined to think, in spite of the rather poor writing, that it is late in the I4th century. It is imperfect, beginning at Inf. ii. 19. I should judge the text of the Inf. to be on the whole very good, and better than that of the Purg. and Par. The orthography is peculiar. Besides Latin and unassimilated forms of words (passim), the following may be noted : el to talento (Inf. ii. 81); vollia, filluol, iacere (Inf. xxxiv. 13), gacgione (cagione), Yo for to (passim'). 402 628 ACCOUNT OF THE MSS. Inf. ii. 22 was the only peculiar reading noticed. It contains the epitaph 'Jura Monarchic] &c., reading ' patriis exterus' in line 5. This is described as " Epitaphium Dantis factum a se ipso." The epitaph ' Inclita fama,' &c., is also given as " Epitaphium quod filius suus fecit." The verses of the colophon are rather incorrectly printed in Batines, and should read thus : Se pur disio ad alcun venisse Et vole sapere quello chi me scrisse Donno Antonio de Bellante dicto Se chiama chi lo libro ha scripto. 7 (Bat. 427). In the Bibliotheque Nationale at Paris. A MS. on vellum (14! x QJ), of very great interest, as containing in Inf. Canto xxxiii. the singular inter- polation of 1 8 lines, first discovered by Dr Gregorio Palmieri in the Bodleian MS. 'E.' It is undoubtedly, I should say, of the I4th century, though perhaps of the last quarter of it, i.e. at any rate 50 or 60 years earlier than the Oxford MS. in which the passage is found. [See further on this subject Appendix ill.] The dialect is not strongly marked, but is probably Venetian ('E' is unmistakably so). It has some remarkably good and some equally bad readings, and a very large proportion of peculiar ones ; on the whole therefore it is probably not a high authority. It has also been very much corrected by a Sioptfwn/s. In respect of orthography we may note lu calle, malvasio, piacti (piatti\ tracti, &c. ; but its dialect is by no means strongly marked, that of the Commentary being much more so ; e.g. zaschuna, verd, &c. occur when quoting the text in which the ordinary forms ciascuna, verra, &c., appear. The following readings are very unusual : Inf. iii. 123 Tutti vengon qui d' ogni paese. (So TV.) x. 96 Che qui inviluppata, &c. xiv. 48 Che mal tiri (sic). xxiii. 109 E quei ch' avea. (So D.} xxix. 46 Qual dolor sare' selgli spedali. xxxiv. 47 a tale uccello. To these we may add : Inf. i. 48; viii. 125 ; ix. 125 ; xii. 125 ; xiii. 21 ; xix. 23, 64 ; xxvi. 25; xxviii. 135 ; xxxii. 47 ; Par. xvii. 9. It is to be noticed that none of these peculiarities are found in E, and per contra none (I believe) of those in the list given for E (see p. 522), are found in this MS. It looks therefore as if the two MSS. had no special connexion, and this makes it probable that the singular interpolation in Inf. xxxiii. may again be found elsewhere. Its extreme rarity is shewn by EXAMINED OR COLLATED. 629 the fact that out of nearly 300 MSS. which have passed through my hands it is found in the text of two only. The Inferno has the colophon : Qui finisce il ninferno de Dante Alighieri di firenze e la sua irisione tucta per online secondo che scripto avemo di sopra a laude di dio et a utilita degli uditori. Amen. Amen. Amen. Then follow the lines 'Jamque domos stygias] &c., and afterwards the epitaph ' Jura monarchic] &c., with patriis expulsus in 1. 5. At the end of the Paradiso we read : Explicit liber dantis Allegherii de florentia. Amen. Deo Grazias. Amen. (This is written with upright letters and very long tails and nourishes, like the colophon to the Codice Landi at Piacenza.) This Codice must have belonged to Pio VI., since the binding has the tiara and the well-known Braschi arms. (See also under Y, p. 549.) 8 (Bat. 441). In the Bibliotheque Nationale at Paris. A paper MS. of the I5th century. See Batines (n. p. 248) for description, and especially Ferrari's note cited there. In respect of orthography, I noted possia (poscia], grisie (grtge), lu sdemo, giuvanni, anema, cactolicha, allocta, gractar, cipta; and caricare for carreggiare in Purg. iv. 72. A curious feature is the number of unfinished lines and blank spaces left. (On this comp. /, <, i, 95), e.g. Inf. xxxiv. 99 Cavea...suolo (sic). Purg. iii. 85 Cosi vidi mueaver la testa. iv. 26 Montasi et in cacume. xix. 55 ends at sospicion. - 73 pavimento is omitted. - 78 noi is omitted. 105 Che piu...sembran, &c. in all cases gaps being left. There is a very curious and much larger blank in Purg. xxv. and xxvi. Purg. xxv. ends at 1. 105 in the middle of a page, and Purg. xxvi. begins at 1. 19 (Ne sole, &c.) in the middle of the next page, the space being left, as though the scribe had intended to supply the omission after- wards. The following may be noted as peculiar: Inf. ii. 81 ; Purg. ii. 13 ; iv. 72 ; xxvii. 4, Si 1 ; xxx. 15 ; Par. vii. 114 ; xxv. 14 ; xxvi. 96, 104. According to Witte Proleg. p. Ixiv (note) this MS. is identical with that numbered 435 in Batines, who, following Marsand and Ferrari, has inad- vertently described the same MS. twice under different numbers. I am at 1 The reading attributed (supra, p. 425) by a misprint to '8' belongs to this MS. 630 ACCOUNT OF THE MSS. a loss to understand on what grounds Ferrari assigns so high a value to the text. To judge by the large number of secondary and peculiar read- ings, I should rather say that the text was an exceptionally bad one. 9 (Bat. 259). In the Bibl. Trivulziana at Milan. A late I4th cent. (c. 1380 ?) MS. on vellum containing only the Inf. and Purg., not however all in the same hand, much damaged in places by damp, and the ink occasionally so faded as to be illegible. It has no colophon, and a few peculiarities of orthography noted point to a Venetian origin, e.g. doi=due, c% = con, cusl (sometimes), possa=posa, rasgion, &c. The following readings are noteworthy : Inf. xii. 49 O cieca cupidigia e dira e folle. (Comp. in part 66.) xxvii. 85 Lo principe dei vivi Farisei. (Comp. the -well-known variants nuovi and vivi in Purg. xx. 90, which the corrector possibly had in his mind here.) Purg. xxxi. 43 Tuttavia per vergogna porti (sic). Add further : Inf. xii. 66 ; xvi. 14 ; xxi. 14 ; xxxiv. 99 ; Purg. xxv. 9 ; xxx. 73; xxxi. 51. All these are, I believe, peculiar and in most cases unique. Bossi (see Batines, II. p. 139) attached an entirely exaggerated value to this MS., which happened to belong to him ; and his statement that " mondizia solversi fa prova (so f and 51): and in xxxiv. 93 it has lasciato in error for passato (so A, C, G). Add further : Inf. xii. 125 ; xxvi. 25 ; xxx. 115 ; Purg. xxvi. 72 ; Par. ix. 129 ; xxvi. 134. 24 (Bat. 137). In the Bibl. Riccardiana at Florence. A small folio on paper of early 1 5th century or perhaps late I4th. It is very neatly transcribed, and the colophons are carefully written in red ink, and bear out the idea of a I4th century origin. They contain interesting peculiarities, such as the prayer in four lines at the end of the Inf. (see Batines, n. p. 77). Also the name of the writer, Bartolomeus filius Andree Mazzonis delucis (i.e. of Lucca). (Bat. No. 189 is written by the same scribe.) At the beginning of Purg., after a title of an ordinary type, he adds : Qui nel primo canto sono quelli che speran di venire, quando che sia, alle beate genti [thus quoting Inf. i. 120]. Another curious feature is that after Purg. vi. 87 (where there is certainly no natural break) there occurs an interpolation of six pages as follows : Questo ene (sic) lo credo di Dante in rima. Then follow the well-known lines l lo scrissi gia] &c. After this the versified forms of the Ten Commandments and the Paternoster. The text is then resumed at Purg. vi. 88, Che val percht, &c. The scribe commonly omits the final ' /,' as no, ma, colu, &c. The text is a curious one, since it abounds with the wildest variants and conjectural readings in the Inferno, while the text of the other Cantiche is quite sober and ordinary. I think the exemplar text was a good one. I fail however to find any trace (as suggested by Witte, Prol. p. xiv) of relationship with the Vatican text. The following passages have noticeable readings, these being in many cases peculiar, as far as I have noted, to this MS. : EXAMINED OR COLLATED. 637 Inf. i. 42 ; v. 126; viii. 125; ix. 54, 70 ; xii. 16, 66 ; xvi. 14 ; xvii. 95 ; xxi. 93, 112 ; xxvi. 25 ; xxviii. 83 ; Purg. xxi. 61 ; Par. xxvi. 96. Also Inf. xi. 37 Odio omicidio. //! xxii. 101 d' esse lor vendette. Par. xxv. 33 Quante fiate Jesu ai tre fe' piu chiarezza. xxvi. 136 e si convenne. 25 (Bat. 139). In the Bibl. Riccardiana at Florence. A folio MS. on vellum, the earlier partof which is very clearly and well written and looks like late I4th century. After about Canto xvii. the writing becomes coarse and clumsy, and seems to be by another and much later hand. It contains the Inferno only, which is followed by the two well-known ' Capitoli.' The text is poor, and quite uninteresting. Again (as in the case of ' 24') I failed to trace the Vatican connexion indicated by Witte (Prolcg. p. xiv). His judgment in both places was probably based upon resemblances to B in Inf. iii. only, a Canto in which I have not observed any special variations in the 'Vatican' family. The following seem peculiar : Inf. xiv. 126; xvi. 14, but none others were noticed. 26 (Bat. 126). Also in the Bibl. Riccardiana at Florence. A beautifully written MS. on vellum in large folio (15! x 1 1), of the i4th century. The writing is of a type often found (as in r, , T, &c.), very clear and upright, and with all the letters very carefully formed. This type of hand is said, on the authority of Boccaccio, to resemble the handwriting of Dante himself (see Barlow, Contributions, &c., p. 37). The text is of quite an ordinary and average type, and possesses no special interest. Par. x. 112, and xxxiii. 89 may perhaps be noticed. 27 (Bat. 127). In the Bibl. Riccardiana at Florence. A MS. on vellum in folio, late 1 4th century. The writing is beautifully clear and upright. The text is a decidedly good one, and is interesting both as being one of Witte's selected ist Class (see Proleg. p. Ixxvi), and also as exhibiting very strongly the peculiarities of the 'Vatican ' type (see Table in Appendix II.). I noted scarcely any readings peculiar to itself, but the following are noticeable as being rather uncommon : Inf. xxvii. 65 Non uscl mai alcun. Purg. vi. 109 1' oppressura 1 . (So B, p, s.) 1 It is curious that this very rare reading (as far as my experience goes) is registered by Witte in A v C, D ; whereas B, with which, as I have noticed, this MS. is so frequently in correspondence, has the ordinary (and probably correct) reading pressura. 638 ACCOUNT OF THE MSS. Par, xi. 105 Tornossi (rather uncommon}. Add further: Inf. ii. 22 ; Purg. i. 108; xxi. 19 ; xxvi. 72. 28 (Bat. 130). In the Bibl. Riccardiana at Florence, on vellum, and, like the last MS., one of Witte's 1st Class. The text is curious as exhibiting the 'Vatican' type strongly in Par., moderately in Purg., and not at all in Inf. In the Purg. several of the peculiar Vatican readings have been altered to the more ordinary type secunda manu. The readings of Inf. ix. 70 and Par. xxvi. 134 may be noticed. Also Purg. xviii. 57, E de prima appetibile, which seems to be a mixture of the two readings ordinarily found. 29 (Bat. 131). In the Bibl. Riccardiana at Florence. A beautiful MS. on vellum, containing a very interesting and, as I think, a remarkably pure and valuable text, in fact for the Inferno one of the best I have ever met with. There are a very few readings which I should judge inferior, such as x. 65, and xviii. 12. But as this MS. has the peculiarities which I have described as belonging to the 'Vatican' text very strongly marked, exception must also be made of one or two of the obviously inferior readings of that recension, e.g. Inf. xii. 16. Among members of that family it has a specially close relation to ' Z.' Note (inter alia) ' figlia o nuro ' in Par. xxvi. 93 (Z having figlitf o nuro) ; nuovi in Purg. xx. 90; El postia si chiamb in Par. xxvi. 136. See also the Tabular view of Vatican readings, where the correspondences between Z and 29 are most striking, there being only two or three divergencies as against upwards of 50 agreements. The relation of V 1 to both these MSS. is also very noticeable. Inter alia, V 1 and 29 have in common the curious reading i portai fuori ' in Inf. ix. 70*. Unfortunately some pages are lost, including parts of Inf. xxi. and xxii., and also from Purg. viii. 71 to xiii. 68 (incl.}. In the Purg. and Par. perhaps the text is not so exceptionally good as in the Inf., but it is quite above the average; and this is just such a case as is mentioned in the Proleg. (p. xxiii), in which we seem to have a very good text slightly tampered with, but in such a way that these defects or excrescences do not seriously affect its value. There are scarcely any peculiarities to notice ; but giustizia in Inf. xxiv. 119 is not very common. See also Purg. xxx. 15; Par. vii. 21. ( Piu vergogna' in Purg. xxxi. 43 is very rare, but is found in some MSS. of the Vatican family. 1 It should be observed that Z which now has 'porta fuori ' bears traces of alteration here. EXAMINED OR COLLATED. 639 30 (Bat. 388). A MS. on vellum, formerly in the monastery of S. Pantaleo, but now in the Bibl. Vittorio Emmanuele at Rome. Batines places it in the I5th century, but I should say it certainly belongs to the I4th, or at any rate the earlier part of the MS. does so. The Paradiso (as Batines notes) is by a later hand, and there seem to be also two or three hands in the Purg. It is defective at the commencement, beginning with Inf. i. 113 Che tu mi segui. The colophon to the Inf. is as follows : Explicit prima pars comedie Dantis in qua de inferis est tractatum (thus far in bright n red ink: then in paler red ink and in a larger hand) MCCCCXXVlllj Martis e (sic) xiij decembris. Then follow in the same hand and ink the two lines cited by Batines (. p. 208). For the title to the Purg. see Batines. The colophon is Explicit secunda pars comedie Dantis. In qua de Mis que sunt in purgatorio est tractatum. Then follows a very curious interpolation of two Cantos headed " Capitulum de Usurariis et nominatur (?) bonafidanza" Come le tre sorelle che un sol occhio Comune usa-van riguardando altrui, &c. (151 lines.) Then " Hie incipit De Gulosis capitulum? Non v* eran tutto la veduta sciolta Di not da Cerber per lo scender fatto (also 151 lines.) The colophon follows : Expliciunt duo capitula facta per alium quam per danteni*-. Then follows the Paradiso with title, Dicto de purgatorio est dicendum de gloria Paradisi : and with the colophon, the Capitolo of Busone, epitaph, &c., as given in Batines, II. p. 209. In the titles given to Dante however, Batines omits Rhetorici, which should be added after Astrologi. Then follows ' Canzone di dante de la gentilezza ' written con- tinuously like prose ' Poscia ch' amor del tutto m' ha lasciato' (Canzone iv. of Part n. of the Canzoniere, Ed. Giuliani). There are nine more of these Canzoni as follows : (2) Canz. viii. of Part 2 (3) ii- >i > J (4) i- ii ii x (5) ii vi - ii 2 (6) i. 3 (7) ? (8) i. 2 (9) ii xi - n 2 (10) v. 2. 1 These spurious ' Capitula ' have since been published and discussed by the learned Librarian of the Bibl. Vitt. Emm., Professor Ignazio Giorgi, " Aneddoto di un Codice Danlesco " (Livorno 1880). 640 ACCOUNT OF THE MSS. Next we find (i) Epistola missa ad Regem Romanorum per Dantem Allegherii Florentinum. The letter itself is in Italian. (2) The same in Latin. (3) Universis et singulis Italics regibus, &c. This is Epistle v., beginning u Ecce nunc tempus acceptabile? &c. Then there are a number of other fragments of Canzoni, &c. There is a blank page after Purg. ix., but nothing has been omitted. The readings are quite average on the whole, but there are several comparatively rare, both of the good and bad type. The following may be noted as peculiar : Inf. iii. 8 io in eterno duro. xxi. 1 13 Mille dugento e nonanta e sei. (This is very curious?) Add further : Inf. xviii. 12; xix. 64; xxi. 93; xxvii. 21 ; Purg. xxiv. 6 1 ; xxx. 73 ; xxxi. 51 ; Par. xviii. 75 ; xxix. 4; xxxi. 20. 31 (Bat. 365). In the Barberini Library. This curious little MS., on vellum, the smallest I have ever met with, except perhaps ' 109' (it being only 5! x 3f ), is most beautifully executed, and the writing, though extremely minute (for, small as the MS. is, it has two columns on a page), is perfectly clear. The text is also a very good one, as judged by the test-passages examined, and it is further interesting as exhibiting very strongly the peculiarities of the Vatican type of text (see Appendix n). As has been noticed in several other cases, some of the Vatican readings have been altered by a 8iop6, 46.) xxviii. 50 piu festine. (So ft, X and one or two others.} Other coincidences will be found in Par. xi. 82 ; xvii. 9; xxvi. 136 ; xxvii. 57 ; xxviii. 23, &c., and the following may be added from Inf.: iii. 30 quando turbo. (So ft and one or two more.} - 38 rubelli. (So ft, e.) 99 intorno gli occhi. (So ft, c.) 103 Bestemmiavano Dio. (So ft, , X, o and several others} On the other hand, in 1. 6, ft has the very strange reading c il sommo honore, and in 1. 3, fra la, in both which cases 46 has the lect. vulg. This however is only an instance of an individual peculiarity in ft, such as may be found more or less in almost all MSS. The above and other instances of agreement fully establish the close relationship of these MSS. 47 (Bat. 330). In the Vatican Library. A small folio MS. on paper, apparently about 1400, or a little later perhaps. It is imperfect, beginning at Inf.v. 139 (Mentre, &c.) and ending at Par. xxxii. 1 1 (doglid). The following is the title to the Purg. : Incipit liber secunde Cantice Dantis in quo sub nomine Purgatorii allegorice et more poetico tractatur de vita morali. Primus cantus ipsius secunde Cantice in quo tractatur prohemium, scientias ad hujus opere subfragium more poetico invocando, Catonein perfectissimum moralem pro clavi moralitatis allegorice ordinando. EXAMINED OR COLLATED. 647 To the Par. the title is : Incipit primum capitulum in quo pmittuntur dicenda, et invocatur ad deos more poetico. The readings are decidedly poor, and many are curiously conjectural, especially in the Paradiso. The following may be worth noting : Inf. xviii. 12 ; xix. 3 ; xxv. 8 ; Par. i. 141 ; ix. 129 ; xviii. 75. We may also add : Par. xi. 105 Rendesi. (So only F, ix. 34; xxi. 61. 52 (Bat. 332). In the Vatican Library. Batines describes it as early i5th century. I should have thought, both from the writing and the paper, thit it was of the 1 4th. One peculiarity of this MS. is that a blank spa;e is left between the terzine 1 . It is defective both at the beginning ard end, as noted by Batines (n. p. 174). It appears to me that its reading; are very bad and often obviously conjectural. The following are some specimens : Inf. xvii. 95 ; xviii. 12 ; xxvii. 2[ ; xxxiii. 72; Purg. ii. 13, 44. ; iv. 72; vii. 51 ; xix. 35. Purg. xx. 90, Entra Giudei ladroni is peculiarly audacious 53 (Bat. 333)- Also in the Vatican Library. A late isth cent. MS. (daed 1459) on paper, in very cursive writing. For the title, colophon, and otier contents of the MS. see Batines, II. p. 175. The readings are often nost extrava- gantly conjectural, and the text among the worst J have e\er met with. The following samples are sufficient to condemn it : Inf. i. 22, 81 ; iv. 141; xiii. 63; xvi. 3; xviii. 12; xx. 69; xxi. 112; xxiii. 83 Purg. xviii. 83; xxx. 15. 54 (Bat. 329). A small folio MS. on vellum (latish I4th century) in the Vatican Library, exhibiting, as will be seen, a curious and interestingtext, from its unmistakable relation to ' <' and to 106, and for its partial contact with the ' Vatican ' type of text. See under < (supra p. 585) for Inf. ii. 53; xii. 6; xxi. 112; Purg. ii. 99 ; xx. 114, &c. 1 So also in '80' infra. EXAMINED OR COLLATED. 649 Several other instances might be added from the collations. Perhaps the most remarkable evidence of connexion is the omission in all the three MSS., 0, 54 and 106, (without note) of Inf. x. 109 in. There are no broken or unfinished lines in this MS. as in <. The ' Vatican ' colouring of the text of this MS. is very strong in Purg., especially in the latter part, but it is wholly absent in Par., and there is scarcely a trace of it in Inf. The same remark applies to and 106. See on this App. II. This MS. often shews itself independent of <. In fact, however closely MSS. are related, I have never (except in the cases of ^ and o> ; 56 and 57 ; and perhaps 88 and 89) found instances of two MSS. being in any sense identical. Instances of divergence from $ will be found in Inf. ii. 8 1 aprire il tuo. ($ has il mio.) iii. 40 per non esser men che belli. ($ and 106 omit non.) 75 fioco. ($ has cieco.) xiv. 126 Tu a sinistra. (So 106 only.} To these we may add Inf. xxx. 18 ; xxxiii. 26; xxxiv. 13 ; Purg. ii. 93; xiii. 2 ; xxi. 61. The following passages are also to be noted, where is either defec- tive or has not been recorded : Inf. x. 1 1 3 Fovi (sic) a saper che '1 feci. xii. 28 giu per lo incarco. (106 has giu per lo scharche.} xxii. 6 et corre et giostra. (Comp. 106.) Par. xxix. 4 che '1 centre inlibra. (So 106 and also Tr and G.) The titles and colophons are different (see Batines, il. p. 173). There is a final colophon after the Capitolo of Bosone, as follows : ' Compiuto il chapitolo del figliuolo di dante Alighieri. Deo gratias. Qui scrisit scriba semper cum domino vibat (sic). It will be observed that this colophon assigns an erroneous authorship to the Capitolo, as does also the title, for which see Batines, /. c. In the partial correspondence which this MS. exhibits with the Vatican text in the Purg., its coincidence in nearly all cases with V and S is noticeable, as well as its almost complete agreement with and 106 in the Inf., though in the Purg. and Par., while 54 and 106 maintain their relationship most exactly throughout, < separates itself from them in Purg. (It does not contain Par.} See however further under ' 106.' Dr Barlow (Contributions, p. 103) mentions the singular reading porto for parte in Inf. iv. 36. This is not in $ or 106, which have parte as usual. 55 (Bat. 374). A large folio I4th century MS. on vellum in the Library of the Borghese Palace. It is beautifully written, and of the type found in the Malaspina MS. () and the Archinto MS. (o) and others, dating 1370 80 probably. It is unfortunately imperfect, several missing leaves having been replaced at a later date. It seems to have a good but quite ordinary 650 , ACCOUNT OF THE MSS. text, in which I did not notice any peculiar variants in the selected passages. As will be seen from the collations, it has been corrected or altered with very great frequency and often, I think, by means of a 'Vatican' text. Perhaps the chief interest of the MS. is derived from the curious note on the fly-sheet in a later hand, but early in the I5th century ; and probably, as has been suggested, added by the first, or an early possessor of the MS. This is given by Batines (n. p. 200), and also by Papanti (Dante secondo la Tradiziom, &c., p. 85), and elsewhere, with some variations, Papanti claiming that his copy is accurate. The note is too long to quote here, but its principal feature is the statement that Petrarch declared that while the De Monarchia might properly be styled the work of Dante, the Divina Commedia should be referred rather to the Holy Spirit than to Dante. The Borghese Library also contains a copy of the Latin hexameter version of the Divina Commedia by Matteo Ronto, of Monte Oliveto. As he died in 1343, this very early translation has a peculiar value and interest, and, as it is a line for line translation, often affords direct evi- dence as to the readings adopted in disputed passages. A full description of this translation with some extracts, will be found in Batines, I. p. 237, though the actual MS. there described is not this one, but one existing at Lucca. This particular MS. is not mentioned by Batines, unless it may be the one noticed p. 241 as once belonging to the Olivetans of S. Vettore dal Corpo at Milan, but no longer traceable. The present MS. is a very late copy, being of the i/th century apparently, and it is very like the late MS. in the Bodleian registered above as ' J.' It resembles that MS. in the character of the paper and the writing, and moreover in the punctua- tion which accompanies the Italian text. This seems to be copied from a printed edition, and (like J again) evidently from one of the Aldine texts. It is curious to note this, because though the Italian and Latin versions are on opposite pages, it is obvious time after time that the latter is translated from quite a different text. The following are a few examples of this : Italian. Latin. Inf. xii. 16 Lo savio mio Virgilio Dux meus intonuit dicens cla- mosus in ilium. 125 copria pur li piedi quo solum planta calebat (cocea). xxxiii. 43 G&eradesto Somnus abibat ab his (Gia eran desti). 74 E tre di Hos biduo postquam jam deces- sere vocavi (E due di, &c.). 78 Che furo all' ossa Qui forat et terebrat, &c. (Che foran 1' osso). Purg. v. 38 Di mezza notte, &c. nocte vapores Acre subflrtma (Di prima notte). - 88 lo fu'\...fui B. Jam de Monte fui Feltro, Buo- contus et exsto (io son Buon.). EXAMINED OR COLLATED. 651 56 (Bat. 377). A MS. in the Chigi Library at Rome. A small folio on vellum on the verge of the I4th and i5th centuries. It is beautifully written and much resembles Z in this respect, with whose text also it stands in very close and remarkable relationship. It is a very strong instance of the 'Vatican' type of text (see App. n.), but what is more singular is that it is practically identical with ' 57,' and they are both apparently written by the same hand. One MS. seems to be a replica of the other. This, strange to say, is a most rare and an all but unexampled phenomenon. Having examined nearly 300 MSS., I have met with it only once besides, viz. in the case of ^ and , but 88 and 89 are very nearly the same. Dr Witte somewhere states that his experience is similar to this. The title is the same in both MSS., see Batines (n. p. 203), by whom it is partially quoted. It continues as follows : nel quale /' autore mostra se smarrito in una valle e impedito da tre bestie e come Virgilio apparitogli segli offerse per duca ad trarlo di quel luogo mostrandogli per qual via. The colophon is also the same in both MSS. (see Batines /. c.\ including the following four Latin verses in each case : Finis adest longi dantis cum laude laboris Gloria sit summo regi matrique precamur Quos oro celsas prestent conscendere sedes Dum suprema dies veniet mortalibus egris. Pages 84 116 (of '56') contain a fragment of a commentary by Filippo Villani thus introduced : Expositiones seu comento Filippi Villani super comediam Dantis Allegherii. prefatio. Incipit feliciter. Though it covers 65 pages closely and beautifully written, it comprises the commentary on Canto i. of the Inferno only. (This Commentary is not added in ' 57.') The following are noticeable as the only instances of divergences observed out of nearly 250 passages collated: Inf. xx. 69 Sefesse quel (56). Se fosse quel (57). Par. xxvi. 136 Poscia si chiamo (56). El poscia si chiamo (57). In this last case there is probably a mere clerical error in 56. Also in iii. 31 and xxxiii. 26 there are slight differences, but as 56 has been evidently altered, it may be presumed that its original reading was the same as that of 57. As peculiar or noticeable readings we may add : Inf. xxi. II3 1 Mille dugent' uno. xxiv. 119 O giustizia di Dio. 1 These two MSS. as well as 90 and 119 (in which also this reading occurs) were examined in this passage after p. 333 (supra) was in type. 6$2 ACCOUNT OF THE MSS. Purg. xix. 34 lo volsi gli occhi e '1 buon Virgilio al men tre. xxx. 1 5 La rivestita carne. xxxi. 43 piu vergogna. (So only V 1 , Z, d, 29, 57, 68 and 90 out of about 1 80 MSS. examined?) Par. vii. 21 Fosse vengiata. In all these cases 57 has the same readings, and they are generally rare elsewhere. 57 (Bat. 380). This MS. (also in the Chigi Library) being almost a facsimile of the last, the description of '56' may be referred to. There are many marginal notes and varr. lectt. (see Batines, 1 1. p. 205), concerning which latter it is very curious to note that in almost every single case where a 'Vatican' reading occurs, the lect. vulg, is given in the margin, and also on the other hand, in the few cases where the Vatican reading does not occur in the text, that is given in the margin. 58 (Bat. 376). In the Chigi Library at Rome. A MS. on vellum of a somewhat unusual form (gf x 7), very beautifully written in large clear and upright roundish Gothic characters with rather long tails and flourishes. Though I do not remember any writing exactly like this, it seems probable that its date is about 1370 1380. The colophon, for which see Batines (n. p. 203), contains the writer's name but no date, and (besides what is given in Batines) ends with the familiar line "Finite libra refferamus (&ic) gras X^o." The orthography is apparently Venetian : e.g. mezo, eror, conubi, taccitti ( = taciti xxiii. i), temma ( = tema\ mobille, Pietolla, luntana, fo (=fu), ese (=esce),fillio, &c., &c. also x for s commonly, e.g. prexo (=presso), rixo, vixo, mixe (=rzso, &c.). The text is a very poor one, I think, and (inter alia) exhibits an especial tendency to 'assimilative' changes (e.g. Inf. xix. 23; xxv. 120; xxxiv. 13, 53). Some peculiarities of reading will be found noted at : Inf. i. 28 ; v. 126; Purg. vii. 51 ; xxi. 61 ; xxx. 73; Par. ii. 141. It has also the rather uncommon cortese e bella at Inj. ii. 53 ; and the very rare (and bad) reading Romani at Inf. xxviii. 10. 59 (Bat. 328). In the Vatican Library. A large folio on vellum, of the I4th century though late in it, but certainly not by the same hand throughout. It has been a good deal corrected, and almost always to introduce a worse reading of the secondary or tertiary type (e.g. Inf. xvi. 3 ; xxix. 46, &c.). It has otherwise quite an ordinary text throughout. We find the rather EXAMINED OR COLLATED. 653 uncommon reading vendetta in Par, xxvii. 57. The reading at Par. xxvi. 136 may also be noticed. The title, imperfectly quoted by Batines (ll. p. 173), concludes thus : et lo impedimenta di tre particolari -vizii et il soccorso di Virgilio Jigura di naturale ragione, et prohemiza a tucta r opera. 60 (Bat. 323). A MS. on vellum in the Vatican Library, having all the appearance of being of the i$th century, but it is distinctly dated, in an apparently genuine colophon, 1388 \ It has on the whole a very good text, and Witte (Proleg. p. Ixxvi) puts it in his first Class. It is however disfigured by several distinctly secondary readings, e.g. Inf. ix. 70 ; xvi. 3 ; xvii. 20 ; xviii. 12 ; xxi. 46 ; xxxi. 75. I did not notice any individual peculiarities of reading whatever in the text, which is a point in its favour, as has been noticed before. 6 1 (Bat. 334). In the Vatican Library. A MS. on paper dated 1461, and containing the Inferno only. For the colophon see Batines, II. p. I76 2 . The ortho- graphy seems to be Venetian (e.g. cusl, mezo, camin, pusato, dubiosi, cognnbi : also lascio occurs for lasso). It has the two epitaphs, Jura Monarchic &c., and Inclita fama &c., and in the fifth line of the former reads patruus (sic) exterus ab horis. The text exhibits a large number of curiously bold conjectural readings. See (inter alia) the collations of the following passages : Inf. iv. 23; v. 102; vii. 108; xii. 125; xvi. 14; xvii. 95; xviii. 12; xxxiv. 13. 62 (Bat. 381). In the Chigi Library at Rome. A MS. on vellum, containing the Inferno only, probably about the middle (Batines says 'beginning,' but?) of 1 5th century. The writing is in a sort of print-like hand which I do not think I have met with earlier than the middle of the I5th century (compare 'd' in Br. Mus. and 'u' in the Ashburnham collection). The form is rather an unusual one (7f X4|). It has no title or colophon, except "Finis liber primus" (sic). In orthography it is careless about double letters (e.g. doppo, sonno, vilan, &c.) and the text is a thoroughly bad and worthless one. The following may be noted : Inf. ii. 58 cortese e Mantoana. xi. 37 Onde omicidio. (So 51.) See further under Inf. i. 28 ; xiii. 20 ; xxi. 93. 1 So it appears in my notes. I observe that Batines says 1394. I cannot now positively say which is correct. 2 There are several obvious misprints : 9 is printed instead of an initial con- and a final -us ; and z is written several times for final m. 654 ACCOUNT OF THE MSS. 63 (Bat. 382). In the Chigi Library. A folio MS. on vellum described by Batines as early I5th century, but I should think certainly late I4th. The following passages may be noted : Purg. xxvi. 7 piu rolente (sic). (A curious confusion between the two readings dolente and rovente.) xxxi. 43 Tuttavia perche vergogna porti. See further under Inf. iv. 23 ; v. 38; x. 65 ; xxi. 112 ; xxx. 18 ; xxxiii. 26; Purg. xxiv. 6 1 ; Par. x. 112; xiii. 27 ; xvii. 9. There seems a slight tinge of 'Vatican' influence in the Inf. and Purg. occasionally (see App. n.). I seem to have examined only a few of the usual passages in the Par. in this MS., so that I cannot speak as to that part. 64 (Bat. 384). Also in the Chigi Library. A MS. on paper, I should say certainly of I4th century, to judge both from paper and writing, though registered by Batines as I5th. The text is, I think, a very poor one (though perhaps rather better in the Paradise). It has been a good deal altered, so that it is often hard to say what the original reading may have been. The following passages are noticeable : Purg. vii. 51 ; xxiv. 61 ; Par. iii. 15 ; xvii. 9 ; xxiii. 1 15. 65 (Bat. 370). In the Barberini Library at Rome. An early I5th century MS. on paper, containing the Inferno only. It appears to have a thoroughly bad text, full of secondary readings and conjectural variants. The fol- lowing may be noted : Inf. \. 42; ii. 55 ; v. 38; ix. 54, 125 ; xiii. 21 ; xviii. 12 ; xxi. 112 ; xxv. 120; xxvii. 21. We may add Inf. xxii. 6, e far lor giostra. 66 (Bat. 372). In the Barberini Library. A MS. on vellum, I suppose about 1460, the writing being rounded and sloping, with the letters run together and yet somewhat print-like. It has no title, and the colophon, if there were one, is lost, since the MS. ends at Par. xxxiii. 57 (pltraggid). The text stands in very close and remarkable relation to that of p. The following are some instances of their agreement : Inf. i. 28; ii. 22; x. 65; xvii. 20; xviii. 12; xxi. 93; xxiii. 83. In Inf. v. 1 20 both p and 66 have the unusual form cognocesse. At Inf. xix. 3 and xxvi. 25 noticeable readings occur in this MS. which are EXAMINED OR COLLATED. 655 not found in p. In /;// xii. 49 we have the curious reading ' O cieca gente cupidigia e dira e folle.' (I am not sure about p here, except that it has e ira e folle.} Compare '9,' supra. It is curious to note that Inf. xix. 53 (S tu gia costl fitto, Bonifazio) is omitted and the space left blank. Compare the treatment of Purg. xix. 106 in in '34.' See under 'p' for the confusion occurring in that MS. h. I. 67 (Bat. 364). A very beautiful late I4th century MS. on vellum, in the Barberini Library, but almost hopeless to collate, as the text is sprinkled about in most irregular fragments in the midst of a flood of commentary but as far as I could guess from the examination of a small number of passages, it seems to have a very good text. For general description see Batines, II. p. 195. The following are one or two peculiarities noted : Inf. i. 28 Quando posato un poco il corpo lasso. Par. xvii. 9 Segnata bene della materna stampa. 68 (Bat. 338). In the Vatican Library. A latish i5th century MS. in 4to, and in 3 vols. The title is '/,' Inferno Comedia di Dante Canto Primo? The first colophon, l Finit feliciter Infernus] with the verses as given in Batines, who however omits the second line (which occurs in the margin in Inf. and Par., but in its proper place in the colophon to Purg.) as follows : Vivat et hie fell 'x ceu mrtlto tempore senix (sic). At the end of the Par. the colophon is Finit feliciter totum opus. The text is generally an inferior one, and abounds in peculiar readings and very rash conjectures. The following may be taken as specimens : Inf. iv. 23; x. 65 ; xii. 16; xiv. 48; xvii. 95; xxiii. 83 ; xxvi. 25 ; xxvii. 21 ; xxx. 1 8 ; xxxii. 47 ; Purg. vii. 51 ; x. 30 ; xxiv. 61 ; xxvi. 41 ; Par. i. 141 ; vii. 21 ; x. 112 ; xxvii. 57 ; xxviii. 23. To these we may add : Purg. xxxi. 43 perche pi-it vergogna porti. Par. xxv. 33 a te fe' piu chiarezza. In Par. xvi. 38 we have the curious 'conflate' reading, 'E trenta tre fiate.' Inf. xxx. 18 is also a sort of ' conflate ' reading. At Purg. xviii. 91 seqq. the MS. reads thus: E qual Lungo di se di nocte Purche Theban, as if the copy had been defective. 656 ACCOUNT OF THE MSS. 69 (Bat. 335)- In the Vatican Library. A I5th century MS. (and probably about the middle) on paper. In respect of orthography, the double letters are dis- regarded, as is commonly the case in the Venetian dialect (e.g. collut, cossl (also cusl), ptage, &c.). The text is a very poor one and has many gross blunders. The MS. is imperfect, beginning at Inf. i. 58, and ending at Par. xxxii. 12. The colophon to the Inferno is : Qui si compie la prima parte della Comedia di Dante chiamata Inferno, nella quale li pea -atari si puniscono secondo loro peccata. Deo gracias. Amen. The following peculiarities are noticeable : Inf. ii. 37 che dissolve ci6. iii. 28 Facea un tumbolo che s' aggira. Purg. viii. 94* Com' io guardava. xxiv. 65 di s& fanno schiera. (So I, j and 94.) Par. ii. 42 2 e Dio servio. (So 72.) Add further : Inf. vii. 108 ; xvii. 95 ; xviii. 91 ; Purg. ii. 93; vii. 51 ; xxvi. 72 ; Par. xviii. 75; xix. 141 ; xxvi. 136; xxviii. 23. Many of these are probably merely unintelligent blunders, which are very frequent in this MS., and of which the following lines are striking but not unfair samples : Purg. iv. 19 Majore intrata non suole fare impuna. Par. i. 141 Come A (sic) terra che te in (!). (Similar erratic capitals occur elsewhere?) 70 (Bat. 326). In the Vatican Library. A late I4th century MS. on vellum, with singularly beautiful and large clear writing, like that of the sample page in the Cambridge MS. ' Q,' but I think larger. The text however is a thoroughly ordinary one, and indeed below average value. There is little to notice as peculiar, unless perhaps Inf. xii. 66, and Purg. ii. 13. 71 (Bat. 339). In the Vatican Library. This is quite the most magnificent MS. I have ever seen in respect of whiteness of vellum, splendid execution, illumina- tions and handsome binding. It is written in a large print-like hand, and is I suppose a late I5th century MS. For a full description see Batines, II. p. 179. The readings are, I think, decidedly good, and quite above average on the whole. The only peculiarity I have specially noted is 1 This is in error attributed to ' 62 ' supra p. 385. 2 The reference to '69 ' has been omitted supra, p. 442. EXAMINED OR COLLATED. 657 Par. xiii. 27, 'in una sustanzia] which I have found in only 11 MSS. altogether out of more than 200 examined. Par. xviii. 75 and xxviii. 23 may also be noticed. 72 (Bat. 347)- In the Corsini Library at Rome. A large folio MS. on vellum, cer- tainly, I think, of the I4th century (so also Batines, though Rossi registers it as i5th ; see Batines, n. p. 184 n.). It appears to have a thoroughly ordinary text, but it has been so much altered by a Stopflomyr that I have often had to omit recording its evidence as being too doubtful. There are several pages lost towards the end of Paradiso. It contains the epitaph 'Jura monarchic' &c. with patriis (or patruisT) exterus in 1. 5. The following may be noted : Inf. ii. 55 ; xxxiii. 26 ; Purg. xiii. 2 ; xxi. 61 ; xxiv. 61 ; Par. ii. 42 ; xvii. 9; xix. 141. In Par. iii. 15 we have 'piu forte' (znda manu}. After the end of the Purgatorio there is a curious repetition of Purg. xvii. 40 to 129 ' Come si f range' 1 &c. 73 (Bat. 348). Also in the Corsini Library at Rome. The MS. is on paper which seems to me unmistakably of the i4th century, and the writing though very minute is very clear and good, and probably dates about 1370 So 1 . It does not appear to be all in the same hand, the writing becoming inferior at the beginning of the Par., though afterwards improving again. It is of less consequence to determine the date, since in any case it has an extremely bad text, abounding in the most audacious conjectural changes. The following are some of its vagaries and other peculiarities : Inf. ii. 22 ; iv. 23 ; v. 38 ; ix. 54 ; xii. 16 ; xiii. 63 ; xvi. 14, 104 ; xviii. 12; xxi. 14, 93; xxxiv. 99; Purg. ii. 118; vii. 51 ; xvi. 145; xix. 34; xxiv. 61 ; xxx. 73; Par. x. 4; xxvi. 134, 136. Add further : Inf. iii. 30 Si come '1 mare quando turbo spira (!). Par. xix. 3 Faceva liete. 74 (Bat. 352). In the Corsini Library at Rome. A MS. on paper dated 1464. The writing is very fine and executed with an extremely sharp pen ; that of the commentary is still more minute. It has an ordinary text possessing no special interest, but at the same time a fairly good one. 1 I have since found that one at least of the water-marks is given as occurring c- 1375- D. I. 42 658 ACCOUNT OF THE MSS. We may note : Inf. iii. 31 la testa d' error cinta. Purg. xxxi. 43 Tuttavia perche vergogna porti. (So only five others out of about 1 80 examined?) Add Purg. xxiv. 61. 75 (Bat. 342). In the Casanatense Library at Rome. A MS. on paper, described by Batines as I4th century, but I should think both from the writing and paper, it cannot be before the I5th century. It is slightly imperfect, ending at Par. xxxii. 126 (venusto}. The writing is clear, uniform, and quite upright, but rather scratchy and with a tendency to flourishes. The text has many alterations and corrections. It is on the whole, I think, a very poor text, but with occasional good readings. The following are noteworthy : Inf. iii. 72 (Ch' io dissi Maestro or mi) ; ix. 54 ; xvi. 14 ; xxviii. 10 ; Purg. iv. 72; xiii. 2; Par. xix. 141 ; xxii. 94; xxv. 3; xxviii. 23. In Par. xxii. 95 over the word Piu is written Men (apparently prima manti). 76 (Bat. 343). Also in the Casanatense Library. A curious little thick volume on vellum (only 4^x3^) and with only 18 lines on a page. It is late I4th century (according to Batines), or perhaps more probably early I5th. In spite of the entry on the fly-sheet "Codice pregevolissimo de' primi lustri del secolo XV," the text is, I think, an extremely bad and worthless one. The following may be noted : Inf. xiii. 63 ; xviii. 12 ; xxiii. 83 ; xxx. 18 ; Purg. xxiv. 61 ; Par. xxvi. 136. 77 (Bat. 344). In the Casanatense Library^ at Rome. A small folio MS. on paper dated 1462. It is well and clearly written, and the text, though almost worthless as an authority, is curious, both on account of its monstrous and unintelligent blunders, and of its audacious variants, which amount (as in some cases quoted below) to a complete rewriting of the text. It is also curious on account of its marked relationship to i and to / in the Inferno (see below). There seems no trace of this in Purg. or Par. The text moreover of Par. seems quite a sober and ordinary one. For titles and colophons see Batines, n. p. 182. To the title of Purg., as there given, should be added the words, "et postea antequam mors raperel eos fecerunt (?) contritionem? There is no colophon to Purg. or title to EXAMINED OR COLLATED. 659 Par. The two common epitaphs are given. In the epitaph, "Jura Monarchie^ &c., line 5 reads patriis exterus, and in line 4 Auctorem is written Actorem. On the back of the first cover there is, as Batines observes, a graceful portrait of Dante ; and on the last cover there is a figure holding a skull in one hand, and in the other a scroll with the three lines 'Or superbite'' &c. from Purg. xii. 70 3. Underneath is written ' Dantes Poeta ' and above ' Tucto e Fume'' (sic). I cannot however trace any resemblance in this figure to Dante. The dialect appears to be Venetian, or Lombardo- Venetian : e.g. angossia, posscia, lassia, pertusio, nasstmento, riddi, nela, cole, ctta, &c. fo occurs forfu, and also malfagio occurs two or three times. Al occurs frequently for /, but this seems to be due merely to ' personal equation,' as perhaps is also malfagio. The following are some illustrations of the above remarks as to the text : Inf. iii. 83 bianco per lo vecchio pelo. 94 corciare. xi. 37 Odio omicidia (sic). (Comp. 24.) xxxiii. 33 S' aven missi, &c. 35 Leopardi e filgli. 75 Poscia potte il dolor piu che il digiuno. Note further : Inf. ii. 22, 81 ; xvi. 104; xviii. 12; xxix. 46; xxxiii. 72; Purg. iv. 72; v. 136; vii. 51; xxi. 61 ; xxiv. 61 ; xxx. 73; Par. ii. 141. Purg. iv. 124-5 is rewritten thus Dicendo oramai Me di' perch assiso Se' tu quinci? attendi gik esscorta?(!) and Purg. xxii. 1 14 appears thus E cole sore sue e vide ydamia. This and some other similar lines look as if they had been written from imperfectly understood dictation. Resemblances to i or / or both will be found in Inf. xvii. 95 ; xviii. 42 ; xix. 64 ; xxiii. 83 ; xxx. 1 8, &c., &c. Another curious feature about this MS. is the following interesting and ingenious, even if probably apocryphal, account of the occasion on which the so-called ' Credo' of Dante was written 1 (" lo scrissi gia d' amor piu volte rime," &c.). It states how that Dante " una mattina andando alia chiesa di San Francesco se scontro in uno frate valentehuomo e Maestro in Theologia quale era Inquisitore, et disse, ' Se' tu quel Dante fantas- tico che dici essere andato all' Inferno, Purgatorio et Paradiso?' et Dante con grande riverenza rispuse, ' lo son Dante.' Allora il frate dixe, 1 These lines are given in full by Fraticelli, // Canzoniere, p. 377, together with another account of their origin (which should be compared with that above) found in a MS. in the Bibl. Riccardiana at Florence, 'Batines 142'). 42 2 660 ACCOUNT OF THE MSS. ' Dante, Dante, tu faristi el meglio a fare uno libro in grammatica, et fondarti in la Sancta Chiesa di Dio, et non andare facendo rime n& sonecti n frasche.' Et Dante bene armato li volse rispondere, et lo inquisitore dixe, ' Responderai me di qui a doi di, perchfe ora e troppo tardo a desputare dela fede di Christo,' et allora Dante rispuse che '1 faria volontieri." The narrative goes on to say that at the time appointed Dante returned with the lines "lo scrissi gia," &c.; and the inquisitor had nothing more to say, but gave Dante his blessing, "et Dante non euro piii de' suoi facti." 78 (Bat. 349). In the Corsini Library at Rome. A small folio MS. on paper, con- sidered by Batines to be either late I4th or early I5th century. The paper would suggest the former, and the writing the latter date. It has a poor and quite uninteresting text. The colophon is : Expliciunt capilula para- disi. Deo gratias. Amen ( 10 times repeated). The following may perhaps be worth noting : Inf. xviii. 12 ; Purg. ii. 93 ; xxv. 9; xxxi. 51. 79 (Bat. 354). Also in the Corsini Library. A MS. on paper, apparently latish in the 1 5th century, the writing being very rounded and flowing. It has a very poor text, the only noticeable feature of which is its remarkable coinci- dences with h. We may note especially : Inf. iii. 31; iv. 141 ; vii. 108 ; xvi. 3, 14; xx. 69 ; xxxiii. 26; Purg. xxi. 25 ; Par. ii. 42. In Purg. xxxi. 43 we find (in this MS. and in five others only, viz.: t, h, m, 63, 74) 'Tuttavia perche vergogna porti.' It is a curious feature of this MS. that the Cantos are numbered straight on from I to 100 without reference to the divisions of Cantiche. In the colophon cited by Batines (ii. p. 188) ' inxxiv. 119; and 'piu vergogna' in Purg. xxxi. 43. In many of these passages however it will be found in agreement with some of the prominent members of the ' Vatican ' family. On a fly-sheet at the beginning occurs the note transcribed by Batines, II. p. 14, in which the false date ' Sept. 5 ' for Dante's death should be noted. This note is written in a coarse rounded hand, and different from that of the epitaph l jura Monarchie] &c. which follows. In the latter two points maybe observed. In 1. 2, fata is written flata; and in 1. 5 we have pair is (no doubt for pat 'r -its} es terus. 91 (Bat. 6). A MS. on vellum (dated 1396) in the Laurentian Library at Florence. The writing is rather like that of Z, but not nearly so neat and regular. It has many peculiar readings, and a large number of interlinear and marginal variants. The text is on the whole a fair one, and I think above the average. The following are peculiar or rare : Inf. i. 48 ; ii. 81 ; iv. 141 ; Purg. xxx. 15 ; Par. xxxiii. 126. It has also the very rare reading Rende in Inf. iii. 114. EXAMINED OR COLLATED. 665 92 (Bat. 1 8). Also in the Laurentian Library and on vellum. It is beautifully written in the same style as f, &c., but rather coarser, and therefore probably latish in I4th century. The readings are quite of an ordinary type. The only peculiarities noticed were at Inf. vi. 43 and Par. xxvi. 134. 93 (Bat. 14). A MS. on vellum in the same Library. The writing is a sort of rounded Gothic almost as regular as printing, something like that of the British Museum MS. 'b,' but rather smaller characters, and I should judge a little older, perhaps c. 1370. The text is very curious and interesting from the fact of its belonging to very different types in the several Cantiche. It is quite ordinary in the Inf., while it is strongly 'Vatican' in the Purg. and very fairly so in the Par. There is absolutely no trace whatever of any such affinity in the Inf. Moreover there are several peculiar and characteristic readings in the two later Cantiche, but none at all were noticed in the Inf. The following are some examples : Purg. vii. 51 ; xvi. 145 ; xxi. 19. Par. x. 4 per lume per loco. xi. 26 non a quel secondo (? an oral blunder for non nacque '1 sec.). So also x and 1 18. xiii. 27 Et in essa persona. 94 (Bat. 65). Also in the Laurentian Library, partly on vellum and partly on paper. The writing resembles that of the Sunderland MS. (now British Museum 31918) though not quite so good, and so no doubt of the i$th century. The text is rather a poor one, but there are occasional exceptions to this. The following specimens of orthography sufficiently indicate a Venetian origin : rato, conobi, vila, calc, cadi, &c., possia, lassare, repossato, cossl, rippa, rasom and rason ( = ragion\ disaso ( disagio xxxiv. 99). In ii. 55 we find hochi soi piii cha stella. The following are some peculiar readings : Inf. xxxiii. 74 due de li (sic}. Purg. vii. 51 1 o saria dunque perche non. (So three or four others.} xxiv. 65 di se fanno schiera. (So /,j, 69.) Add Inf. xx. 69 ; Purg. ii. 93 ; xxvii. 81. 1 The registration of this reading in '94' is omitted supra, p. 383. 666 ACCOUNT OF THE MSS. 95 (Bat. 75)- Also in the Laurentian Library. The characters are somewhat rounded, but very carefully written with almost the regularity of printing, and the MS. seems fairly early in the I5th century. A peculiarity is the number of defective lines, as has been noticed already in other cases (see description of v, , i, w, &c.). The following are examples of this : Inf. ix. 127 The line stops at ' Qui sono.' x. 9 is left blank altogether. xviii. in is blank after ' ove.' xxxiv. 52 Quindi...tutto si gelava. in traggon is omitted with blank space left. Par. vii. 143 La somma. ..(//& rest blank}. ix. 89 e Macra omitted and space left. 91 occaso do. do. 116 congiunta do. do. x. 128 left blank altogether. In most of these cases there is some unusual word or name which may have caused uncertainty to the scribe. The distinction of single and double letters is constantly neglected. 'Grorta' occurs in Par. ix. 124. The words disago and palago in Inf. xxxiv. 97, 99 exhibit a curious orthography in the case of g, which has been noticed elsewhere (see among others D, G and 98). In Purg. xx. 104 e I' altro e parricida seems to be a confusion or con- flation of the two readings generally found h. I. The following readings may also be noticed : Inf. xviii. 12; xxxiv. 99; Purg. ii. 13; Par. ix. 129; x. 112; xxvi. 136; xxxiii. 126. Add: Inf. iv. 146 Pero che s' io mi chaccio in. Purg. vii. 12 dicendo, ella e sie non e (sic). Par. xxx. 141 Che muor difama (an obvious error). It is curious that in Inf. ii. 12 and 53 we find the 'Vatican' readings l Anzi che,' and l cortese e bella,' while there is not another trace of any such readings throughout the Infer no , except perhaps 'tre di' in Inf. xxxiii. 74. The sheet however on which these lines occur (from ii. io to 83) is of paper and in a later hand, the rest of the MS. being vellum, so that the missing page 1 seems to have been supplied from a Vatican text. In the original text however of the Purg. and Par. there is a very considerable number of Vatican readings, with occasional gaps of several Cantos in which no such readings occur (see Table, App. II.), as if different MSS. 1 There is another such page including iii. 89 to iv. 21, but no test-passages occur there. EXAMINED OR COLLATED. 667 had been copied in different parts. (See further under '106.') The text generally, as judged by test-passages, is a thoroughly poor one having a very large number of certainly inferior readings. 96 (Bat. 25). A MS. on paper (rif x8|) in the Laurentian Library, containing the Purg. only. Both paper and writing indicate I4th century very decidedly, and the water-mark, which however is rather indistinct, seems to cor- respond (valeat quantum] with one dated 1377 by Midoux and Matton. The text is not generally of more than average merit, and among occasional curious readings may be noted : Purg. ii. 44; v. 38; x. 30; xxiv. 61 ; xxx. 73; xxxi. 51. 97 (Bat. 113). A MS. on paper in the Magliabecchian Library at Florence, the writing and paper being apparently of early I5th century. The writing is very good and clear but rather rounded and somewhat resembling that of Z. The text is of barely average merit, I think, but is interesting both from a considerable number of peculiar variants and also from its rather strong Vatican character in Purg. and Par. and in the latter half of the Inferno, there being no trace of it earlier than the end of Canto xii. Curious orthography is found in mezo, camin, drizo, roxtarsi ( = rostarsi}, which have a Venetian character, and also in grolia Par. i. i; ix. 124; and ringua (\ma manii) for lingua (var. for penna) in Inf. xxv. 144, which look Florentine. There are also some curious defects and omissions (to say nothing of accidental misplacements of pages), e.g. Inf. xxiv. 97 132 and Purg. xxxi. 4 45. In both these cases the omissions occur without any note or indication, and in the middle of a page. Another case is Inf. xxvi. 8 93, but here perhaps a page has been lost. Peculiar readings will be found registered in the collations at Inf. iii. 114; iv. 141; xxi. 112; xxvii. 21 ; xxix. 46; xxxiii. 78; Par. vii. 114. To these we may add : Inf. x. 1 1 1 ancor coi vivi congiunto. xvii. 96 mi giunse. (So 'a.') xxx. 86 Con tutto ch' ella volgesse miglia (sic). Purg. xx. 104 e P altro e parricida. (So '95.') 98 (Bat. 110). Also in the Magliabecchian Library. A ijth century MS. on paper dated 1462. The writing is something like that of the Bodleian %' but the letters have not nearly such exaggerated tails or flourishes. The orthography is curious in respect of the letter g, which is made soft before a, 0, u ; e.g. gorno, gunse^ gd, gu, ragon, re govannt, &c. A 668 ACCOUNT OF THE MSS. similar peculiarity occurs in'D' and 'G.' The readings are poorish on the whole. There is a considerable tinge of Vatican readings, though it is partial and irregular (see Tables, App. II.). At Par. vii. 21 Fosse punita is uncommon. 99 (Bat. 91). Also in the Magliabecchian Library. A i4th century MS. on paper, in size, writing, and other respects remarkably like the British Museum MS. 1 Egerton 2567', (n) which bears the date 1379. It has I think a very good text, especially in Purg. and Par., though it has been very freely altered by a 8iop6 and 54.) iv. 75 modo mi diparte. (< has gli and 54 has Ii.) v. 1 20 Che non conoscesti. (See under <.) ix. 54 Ma non veggiamo inteso ai 1' assalto. (Comp. $.) xxii. 6 e corir e giostra. (Comp. 54.) See also the following passages registered under

in the Purg. to that of the other two MSS. is very peculiar. Till about the end of Canto xiii. agrees with the Vatican family without an exception, while 54 and 106 never do so. From this point to the end of Canto xviii. the three MSS. (with one slight exception) agree in following the readings of that family. After Canto xviii. < suddenly loses almost all further trace of any connexion with the Vatican family, and with 54 and 106 also. I think it will be found that the passages recorded in the collations give similar evidence as to the change of character in the text of these three MSS. 1 107 (Bat. 445). In the Library of the Arsenal at Paris. A magnificently executed MS. on vellum (13 x 8|), in a large kind of Gothic character as regular as print- ing, something like (speaking from memory) the beautiful Br. Mus. MS. ' b.' The writing afterwards gets smaller and more like the sample page of the Cambridge MS. ' Q.' I should think it was latish in the Hth century, as judged by Marsand (whose expression, 'sec. xiii.,' so sharply criticised by Batines, is evidently a clerical error for ' sec. xiv.'), and not of the first half of the I4th century, as estimated by Batines. It is profusely illustrated in a very rude and inartistic manner. The initial letters are left blank for illumination, which has not been carried beyond Inf. Canto ii. There are no titles. The colophon to the Inf. is " Rigracio deo e santi tutli che d' inferno semo tsciuti (? uscitf). There is none to the Purg. That to the Par. is Explicit comedia Dantis alagherii Florentini. Deo gratias. Amen. The orthography seems to be distinctly Venetian, e.g. ebi, mezo, camin, dzse, rabia, tuto, leto (x. 65), qualle, possia, cossl, chosse, rappaci, Heccuba, sucedete, mudo (mondo\ abraciolo (Purg. vii. 15) ; fo (=fuza.ndfu) ; vesilla (Inf. xxxiv. i) ; cariqar (Purg. iv. 72). We have some cases of the parasitic u, which has been several times noted before, e.g. Inf. ix. 70, 'abate e puorta fori.' Comp. 'riende figura ' in xviii. 12. 1 It is not a little remarkable that several other MSS. in the Table exhibit a similar peculiarity in respect of the change of the character of their text about Canto xiii. See the record of V, S, 19, 28, 95, 97, 116; and also, to some extent, that of s and 63. These MSS. except ' s ' (and in a few Cantos ' 63 ') also display the same absence of Vatican affinities throughout the Inferno. EXAMINED OR COLLATED. 673 On a fly-sheet in later and different hands are found the two epitaphs Inclita fama, &c., and Jura Monarchic, &c. The fifth line in the latter reads patriis exieris ab horis (sic). The text has been much altered by a diopdaT^s, and there are also many marginal variants. It is, I think, quite an ordinary text, with a considerable number of secondary readings, but few peculiar to itself. I should regard it therefore as a moderate text carefully copied. The following may be noted : Inf. ii. 12 (with signs of alteration) Prima ch' al tuo passo. Purg. vii. 5 1 oel (?) seria che non. Also Inf. ii. 55 ; xx. 69 ; Purg. xxxi. 29; Par. xxvi. 136. 1 08 (Bat. 446). In the Library of the Arsenal at Paris. A MS. on paper (iix8) of about the middle of the I5th century. The writing is rounded, sloping, and cursive, but very clear. The orthography is marked by exceptionally strong Venetian dialectic peculiarities. Thus we have z for c passim, e.g. zinque, zerchio, ziglia (Par. xi. 88) ; and an utter disregard for double letters, e.g. mosa, dotore, stela, dona, sucedete, bela, segio, quela, &c., and on the other hand, paurosse, malle, golla, Pissa. Inf. xxvi. 25 reads thus : QuaF il vilan che apogio se ripossa. i and e again are frequently inter- changed. See also the orthography of the titles which follow. The title to the /;//. is Capitolo primo di dante aldeghieri chiamato inferno, dove prohemiza a tuta F opra del dito suo libra : and the colo- phon Explicit prima pars comedie Dantis aldeghierii de Florencia vocata Infernus. The Purg. has the title Commcia qui la seconda comedia e parte di Dante aldeghieri de fiorenza nela quale se punise e purgasi Ii pecati comesi deli qual /' onto e pztuto e confeso, &c. The colophon is similar mutatis mutandis to that of Inf., and the title to the Par. somewhat resembles that of Purg. The colophon is : Explicit liber par adisus Dantis. Deo gratias. Amen. It appears to me to have an exceptionally bad text, full of secondary readings, and abounding also in such as are peculiar and evidently con- jectural. Batines makes the strange statement that it generally agrees with the celebrated Estense and Vatican MSS. The former important MS. I have had no opportunity of examining. But I fail to find any trace whatever of resemblance to the Vatican MS. (B], and if the former de- serves its great reputation it would be difficult to believe that it had any affinity with this very inferior text. The following are some samples (besides a large number of the ordinary secondary readings to be found in the collations) : Inf. iv. 23; ix. 70; xxi. 93 ; xxiii. 83; xxviii. 10; xxix. 46 ; xxxiii. 74; Purg. iv. 72; vii. 51 ; xx. 90; xxi. 61 ; xxii. 6; xxx. 73 ; Par. ii. 141 ; xxvi. 134; xxix. 4. D. I. 43 6/4 ACCOUNT OF THE MSS. 109 (Bat. 447). In the Arsenal at Paris. A curious little MS. on vellum, something like '31' in the Barberini Library, and even a shade smaller, measuring only 5 x 3|. It is in two columns, the writing being beautifully clear and regular, though very minute and scratchy, as though written with a pin's point. The letters are rounded and look almost like printing. It is pro- bably late 1 5th century. The title-page has " Manoscritto dalP anno MCCCCX al MCCCCXX incirca." This seems to me to be too early a date. The Cantos run on from one to another without a break, but the initial letters are coloured, and the first line slightly set back. It exhibits the common Venetian disregard of double letters passim. Several lines have been omitted at first, and added later in different ink, e.g. Inf. vi. 20, 21, 53 (part); xxii. 12; xxxiii. 109; Purg. ii. 26, &c. It has a remarkable alliance with G, at any rate in the Inf., though I failed to trace this in the two later Cantiche. The following are some of its peculiarities, in most of which the resem- blance to G is striking : Inf. iii. 25 Diverse voci. (So G.) iv. 76 P antica nominanza. (So G.) vi. 86 Diverse pene. (So G, A, and 10.) vii. in Nude e graffiate. (So G.) ix. 125 Che son sepolte. (So G only.} x. 72 Supino cadde e piu non parve fora. (Compare G.) xvii. 95 Ad alti tosto forte che. (So only G andy.) xxix. 46 esce fuor. (So G and a few others.} xxx. 1 8 Del suo Polidoro. (So only cr, 54, and 106.) xxxi. 75 E vedi la che. (So G and a few others} Add Purg. ii. 26; Par. xxvi. 134. Other agreements in readings less rare but not very common will be found in several other passages in the collations. 1 10 (Bat. 437). In the Bibliotheque Nationale at Paris. A very beautifully written MS. on vellum, somewhat in the style of the Bodleian 'A.' It is early in the 1 5th century, but has the archaic spelling ke regularly. It has no title. The colophon is correctly given by Batines. It has a good many peculi- arities, but otherwise, I think, a fair text, especially in Purg. The following readings are more or less peculiar : Inf. vii. 108; xxxii. 47; xxxiii. 78; Purg. x. 30; xxi. 61 ; xxii. 6; xxiv. 6 1 ; Par. x. 4, 112; xxvi. 136; xxix. 4. Purg. Canto iii. ends abruptly at 1. 132 without any sign that there is an omission. This is probably due to theological considerations. See on this, Pro/, p. xviii. EXAMINED OR COLLATED. 675 in (Bat. 430). Also in the Bibliotheque Nationale. A MS. on vellum, certainly, I think, of the I4th century, though late in it. The writing is clear and upright, and not very unlike the style of 'f &c., but with exaggerated flourishes to the letters. It is imperfect, beginning at Inf. x. 82. The colophon to the Inf. is 'Compiuto I Loninferno (sic) di Dante Allaghieri Deo gras. Am. (Those to Purg. and Par. are somewhat similar.) At the end are the Capitoli of Jacopo and Busone. The text is an ordi- nary one, and of about average merit, and I did not observe any pecu- liarities worth recording. 112 (Bat. 438). In the Bibliotheque Nationale at Paris. A beautifully written MS. on vellum. If it is, as Batines judges, of the I5th century, it is very early in it, and may possibly be of the I4th. The writing is very regular and upright, with extremely long tails and flourishes to the letters, and gives the impression of a professional and somewhat artificial hand, which at any rate imitates archaic features if not itself ancient. In the top and bottom lines the heads or tails of such letters as h,f, l,p, &c. are often an inch, and sometimes fully two inches long. This is like (inter alia) the colophon of the celebrated Landi MS. (A), and this peculiarity is even more exaggerated in the two colophons of this MS., which are as follows : " Explicit tercia pars comedie dantis Aldigherii de florentia poete moderni (this epithet is curious, and I think unique) vocata paradisus. Deo gracias. Amen}' 1 " Qui scripsit scribat. Antonium de curtona (sic) Dominus benedicat." It has, I think, a very fair text, and the peculiarities noted were not very numerous. The following may be cited : Inf. ii. 22, 8 1 ; ix. 54; xix. 3; xxvii. 21 ; Purg. v. 136 ; Par. xxviii. 50. In Par. ii. 42, et io s 1 onib is perhaps an oral blunder. 113 (Bat. 433). Also in the Bibliotheque Nationale at Paris. A MS. on paper, bearing the date (see colophon in Batines) 1456. The title is " Comincia laprima parte della cantica overo comedia chiamata Inferno del chiarissimo poeta Dante alighieri di firenze. Et di quella prima parte il canto primo." The writing is very neat and clear but scratchy in appearance. It abounds with Latin and unassimilated forms (e.g. facto, scripse, lectere, &c.) and has many such dialectic forms as maze, canne, anderem, &c.; but these are not by any means general, as is often the case. The text has remarkably strong 'Vatican' affinities, and is, I think, an exceptionally good one, especially in the Inferno. Indeed I have seldom met with a 432 676 ACCOUNT OF THE MSS. better. The other two Cantiche are not, as I should judge, of so high a character (especially the Purg.}. Except certain of the 'Vatican' readings which are confessedly secondary, it has, I think, very few indeed in the Inf. which are inferior. I scarcely observed any 'peculiar' readings, which is itself a good sign, except perhaps nuovi in Purg. xx. 90 (occurring in 12 only out of about 220 MSS. examined) ; came in Purg. xxx. 15 ; and et latro patridda in Purg. xx. 104. (So V 1 and Z.) Par. vii. 21 and xxvi. 136 may also perhaps be noticed, and I believe that the reading 'dissi al Maestro' in Inf. iii. 72 is unusual, as is also giustizia in Inf. xxiv. 119. The well-known Braschi arms on the cover denote that this MS., like some others in this collection, formerly belonged to the Library of Pius VI. 1 14 (Bat. 424). Also in the Bibliotheque Nationale. A MS. on vellum of different dates, as noted by Batines, the earlier part being of the i4th century, and somewhat resembling f, &c. It is defective in parts (see Batines) and, like the last MS., came from the Library of Pius VI. The Venetian dialect is very marked throughout : inter alia the neglect of the double letters is very conspicuous, e.g. ecubba ( Hecuba \),fato=falt0, &c., &c. ; Medusa is written Medtigia, cun for con and so on. It has quite an ordinary text, except perhaps in Par., where the peculiarities are rather numerous. Among unusual readings I noted : Inf. i. 42 ; xi. 90; xviii. 12, 91 ; xxx. 18 ; Purg. x. 30; xxi. 61 ; Par. x. 112 ; xvii. 9 ; xxii. 17 ; xxvi. 134 ; xxix. 4 ; xxxiii. 89. Par. xxvi. 103 reads Indi sospirb. 115 (Bat. 434). In the Bibliotheque Nationale at Paris. A MS. on paper of very unusual form (15^ x 5^), like an account book. It is dated 'Jan. i, 1469' (see colophon &c., in Batines). The text appears to be an inferior one, and the readings generally of quite an ordinary type, except perhaps Par. x. 4 ; xxvi. 134; xxxiii. 89. 116 (Bat. 134). In the Biblioteca Riccardiana at Florence. A small folio on paper probably of late I4th century. This was very imperfectly examined, and chiefly with a view to test its 'Vatican' affinities, since I noticed one or two suggestive coincidences. It will be seen from a reference to the Tables in App. II. that the text is a curiously mixed one in this respect, ex- hibiting no trace whatever of any relation to this family in the Inferno, nor in about the first half of the Purgatorio, after which it adheres pretty constantly to this type, and so it does generally in the Paradise. EXAMINED OR COLLATED. 677 Its coincidence in this respect with several other MSS. (as noted supra under ' 106') is too remarkable to escape notice. The following are somewhat peculiar : Purg. xiii. 105 Fammiti chontro (perhaps in error for conto). xviii. 57 E de prima appetibile. xxv. 51 fe' io stare. 117 (Bat. 379). In the Chigi Library at Rome. A MS. in quarto (ii|x8) on vellum, decidedly, as I should say, of I5th century, though it is described by Batines as i4th. The letters are much rounded and irregular in size, and the lines also are not evenly written. It is very imperfect, since the whole of the Inferno and as far as Canto v. 76 of the Purgatorio are missing. See Batines, II. p. 204 for the colophon and other notes. [The words ' Laudetur Deus' should be added in the colophon after ' ' paradixi^. x for s is found occasionally (e.g. marchexe, dispoxata, &c.). The following readings are perhaps worth noticing, but the MS. was examined too late for their insertion among the collations. Purg. vii. 51 o seria dunque perche non potesse. xviii. 83 Picciola piu che nulla. xxi. 25 Perche che colei che (sic). 118 (Bat. 371). A folio MS. on paper in the Barberini Library at Rome, the date of which, 1465, is given in the colophon. (See Batines II. p. 198 for this and other details.) The MS. contains the Paradiso only. The text is, I think, rather poor, and is also remarkable (like five or six other MSS. that have been already noted) for containing a few broken or imperfect lines : e.g. Par. vii. 143 La somma. xxi. 14 de ardente. In Par. xxxiii. 89 we have the very peculiar reading ' Tanti construtti'; and in xi. 26 the apparently oral blunder (as in x) non a quel secondo. 119 (Bat. 411). This is a very beautiful and well preserved MS. on vellum in the Biblio- teca Nazionale at Palermo. It is late i4th century, I should say about 1480 90 (size 15 x io). There is a title-page to each Canttca, finely but rather rudely illuminated, and the initials to each Canto are also illumi- nated. The writing is beautifully clear, regular and upright until near the end of the Purgatorio, when it becomes larger and coarser in character. I noticed such orthography as cusl, cum (con), sepe (seppe], 6/8 ACCOUNT OF THE MSS. caregiar, pogiato, possa (posa), &c. There is a short colophon (in a later hand) giving the date of Dante's birth, '1265 sedente Urbano quarto^? There is no title except, in a very much later hand, ' expurgatum juxta indicem expurgatorium hispanum. panormi. ibiq (?) Fr Decio Carrega ord. pred. This refers to the complete erasure in ink of the three passages condemned by the Spanish Inquisition in 1614, viz. Inf. xi. 8, 9; xix. 106 117; Par. ix. 133 142 2 . In a comparatively modern hand the three passages have been reinserted above. The text seemed to me to be on the whole decidedly good. The following were some peculiar readings, noticed mostly too late for insertion among the collations, since the MS. was inspected after that part of the work was already in type. Inf. ii. 55 quanto la stella. (So U, and two others?) x. 65 gia di costoro si eletto. (So U only?) xviii. 12 dove '1 sol rende figura. xx. 69 se fosse in quel. xxi. H3 3 Mille dugent' uno. xxv. 105 E il trafitto. (So O, F, r.) xxviii. 24 Fesso dal mento. (So O, D, F, r.) Purg. ii. 13 E come qual sol presso. (So $ and two others only?) iv. 22 lo calla. (Sic?) xvi. 145 Cos! n' and6. (So 3> and two others only.} xix. 35 Voce che mi dicesse. (So * and v only?) Par. xiii. 27 in una sustanzia. xvi. 38 E ire* fiate. (So O, F, and V. Comp. ' 68.') xxviii. 23 Alo al cinger. 1 This is an error for Clement IV., since Urban IV. died Oct. 2, 1264. It occurs however in Boccaccio's Life, and in Benvenuto da Imola's Introduction to his Commentary, and I think I have met with it elsewhere also. 2 This expurgation occurs also, as noted supra-, in a MS. in the Barberini Library (p. 641), and I noticed it some years ago in two or three MSS. in the Royal Library at Madrid. 3 Since p. 332 was in type this rare reading has been found in the follow- ing MSS., 56, 57 and 90. See also the curious variant registered under '30' supra. 4 As I omitted to examine this passage in time to include it among the selected passages discussed supra, I may add a short note upon it here. The variants trenta and tre are interesting, but the latter has very slight MS. support, and is adopted, among the old Commentators, by Peter Dante alone. The question de- pends, as is pointed out in Scartazzini's interesting note, on the period assigned by Dante to the revolution of Mars. It has been assumed from a loose and inaccurate citation of Conv. ii. 15 that Dante took it at two years. But he expressly says 'un anno quasi' in reference to half a revolution of the planet, and the whole passage is one in which a rough and merely approximate description was all that was required. But assuming (thus needlessly and even wrongly) Dante to have committed himself to the period of two years, 580 revolutions would give the date of 1160 for the birth of Cacciaguida, whereas he died in the 2nd Crusade, which EXAMINED OR COLLATED. 6/9 There seems to \xprimafacie evidence of a relation between this MS. and U and *, but the two latter MSS. were not examined in all the selected passages, and were of course not at hand for comparison with this MS. O and F (which are closely related, see supra p. 539) have also some points of resemblance to this MS. began in 1147 (see Par. xv. 139). Hence the reading tre was suggested, which would give 1 106 as the date of his birth, at least a possible date. But as we know from numerous passages that Dante was acquainted with not only the principles but the details of the Ptolemaic planetary theory [see especially a recent work by Angeletti, Cronologia delle Opere Minori di Dante] it is far more probable that he refers here to the Ptolemaic period of Mars, viz. 687 days. This would give the date 109: for the birth of Cacciaguida, and would make him to be 56 years old when he joined the Crusade. The 'quasi' of Conv. ii. 15 would thus cover a difference of only about 2 1 days. There seems therefore no reason whatever to doubt the usual reading trenta. LIST OF MSS. COLLATED OR EXAMINED. Here designated Place or Owner. No. in Colomb de Batines. Present Press-mark taken from Colomb tie Batines, unless (as often) it was noted as having been since changed. A Oxford, Bodleian 491 Canon. Hal. 105, 6, 7 B 493 109 C 492 1 08 D n 487 97 E 489 103 F 495 112 G 485 95 H 486 96 I 488 98 J 494 1 10 K 496 in L 490 104 M 498 US N n 499 116 O Cambridge Univ. Library Mm. 2. 3 1 P Mm. 2. 3 Q > } Gg R Milan, Bibl. Ambrosiana 246 C. cxcvin S > 249 D. XLVII T Milan, Bibl. della Brera 251 AN. xv. 17 U 5> 252 AG. xii. 5 V 1 Lord Vernon's Library 184 2243 V 2 )> n 183 2242 V 3 ) 244 V 4 191 [No number] w British Museum 2 Egerton 2628 X Eton College Library 484 Y Taylor Institution, Oxford Z Principal of S. Edmund Hall, Oxford Gl Glasgow Univ., Hunterian Library 500 Q. 2. 2 7 Tr Treviso, Bibl. Comunale 283 1 O and P are bound in one volume. 2 Acquired by the British Museum, at the sale of the Wodhull Library, to which it belonged when I examined it. LIST OF MSS. COLLATED OR EXAMINED. 68 1 Here designated Place or Owner. No. in Colomb de Batines. Present Press-mark t Colomb de Batines, often) it was noted been since changed. iken from unless (as as having r Milan, Bibl. Trivulziana 257 No. II A Ambros. 248 D. DXXXIX e 247 A. XL A Piacenza 1 , Bibl. Comunale 237 n Ravenna, Bibl. Classense 402 2 403 $ Bologna, Bibl. Comunale Cod. 223 y Cod. 200 Q Cod. 700 a Ashburnham Collection 2 450 L. 828 L. 406 y 458 L. 836 8 460 L. 837 App. 185 C j> 4 6 4 3 App. 181 App. 182 6 L. 405 i 5J 453 L. 832 K J> 455 L. 831 X 457 L. 830 p ?> 459 L. 833 V 452 L. 827 J> >' 454 L. 829 451 L. 834 TT App. 184 P L. 404 a- 463(?) Barrois 23 T 1 88 App. 240 V App. 219 517 App. 242 X 456 L.835 T App. 183 to App. 241 s >-> 187 App. 239 F L. 408 9 461 L. 838 a British Museum 537 943 Egerton b 19587 c 479 3488 Harl. d 482 839 Lansd. e 536 10317 f 400 22780 1 The celebrated 'Codice Landi.' 2 All the MSS. of the Ashburnham Collection are now removed to the Laurentian Library at Florence. The Ashburnham press-marks are given, as preserving some clue to the previous history of the MSS. 'L' implies that they formed part of the Libri Collection, which was purchased by the late Lord Ashburnham about 1846. Those in the 'Appendix' were acquired later and from various sources. 3 Batines seems also to refer to this same MS. under No. 516. 682 LIST OF MSS. Here designated Place or Owner. No. in Colomb de Batines. Present Press-mark taken from Colomb de Batines, unless (as often) it was noted as having been since changed. g 1 British Museum _ 2II63 h 11 480 3513 Harl. i n 478 3460 Harl. j 2085 Egerton k 932 Egerton 1 477 3459 Harl. m 481 3581 Harl. n 55 2567 Egerton o 255 26836 P 26771 q , 31918 r Holkham, Earl of Leicester 510 513 s n 5 5" 5H t n i 5i3 515 u 11 > 512 516 V 11 i 5H 517 w 11 i 5i5 518 X Cheltenham,Phillips Library 508 247 y 11 i 507 9589 z 11 i 506 8881 a Florence, Bibl. Magi. 163 Palat. 313 b 11 11 98 Palch. I. 32 c Bibl. Laur. 7 Cod. Tempiano Maggiore d 11 22 Plut. XL. No. xxxii e 11 3 XL. No. xxii f 11 9 XL. No. xxxvi g 11 72 XL. No. iii h 11 27 Codd. Strozziani, No. CXLVIII i n 11 2 Plut. XC. Suf. No. cxxv j Siena, Bibl. Comunale 224 I. vi. 28 k Florence, Bibl. Rice. I 5 6 1049 I 11 11 133 1049 bis m 11 11 125 1025 n 11 11 124 1005 n 12 Milan, Brera 253 AG. xii. 2 P Florence, Bibl. Magi. 88 220 q 11 u 101 II. x. 29 r 11 11 174 128 s Siena Bibl. Comunale 226 I. vi. 32 t 223 I. vi. 30 11 221 I. vi. 29 V 222 I. vi. 27 IV 225 I. vi. 31 X I. ix. 20 I Paris, Bibl. Nationale 413 1 g 2 refers to a duplicate copy of the earlier Cantos of the Paradiso, forming part of the same MS. 2 n 1 was originally a part of the same MS. as n, but n 1 is now at Milan, while the rest is at Florence. [The letter o is omitted to avoid confusion with the Greek o.] COLLATED OR EXAMINED. 68 3 Here designated Place or Owner. No. in Colomb de Batines. Present Press-mark taken from Colomb de Batines, unless (as often) it was noted as having been since changed. 2 Paris, Bibl. Nationale 414 Ital. 538 3 jj 418 Ital. 69 4 jj 423 Ital. 543 5 jj 422 Ital. 534 6 jj jj 425 Ital. 1298 7 jj jj 427 Ital. 540 8 jj >j 441 Ital. 70 9 Milan, Bibl. Trivulziana 259 No. I 10 jj jj 2 5 8 XVII ii jj jj 260 in 12 jj jj 26l IV 13 jj jj 262 VII 14 jj jj 264 X 15 jj jj 271 XI 16 jj jj ? Codd. Dant. 49, MSS. Belg. 290 17 jj jj 272 No. xn 18 Florence, Bibl. Laur. 16 Plut. XL. No. xiii 19 jj jj 52 XL. No. xv. 20 jj jj 4 XL. No. ii. 21 jj jj 82 XC. Sup. No. cxxxii 22 Bibl. Magi. 89 Palch. I. 36 23 jj jj 112 j, i- 3* 24 Bibl. Rice. 137 1036 25 jj jj 139 1045 26 j j) 126 1010 27 jj jj 127 IOI2 28 jj jj 130 IO 33 29 jj jj I 3 I 1035 30 Rome, Bibl. Vitt. Emanuele 388 101 MSS. S. Pant. 8 31 Barberini 365 xliv. 5 32 jj u 366 xlvi. 34 33 367 xlvi. 54 34 jj jj 369 xlv. 69 35 Vatican 336 Codd. Capponi, 263 36 Chigi 383 L. viii. 292 37 Angelica 358 S. 2. 9 38 w jj 359 S. 2. 10 39 Vatican 322 Codd. Capponi, 266 40 jj jj 325 Codd. Urbtnati, 378 jj jj 324 367 42 320 366 43 Corsini 351 44- F. 33 44 jj jj 345 No. 1265 45 jj jj 346 44. F. 31 46 Vatican 327 No. 3200 47 jj jj 330 No. 4777 48 Barberini 363 xlvi. 59 49 jj 368 xlv. 118 5 ,, Vatican 321 No. 8376 51 jj jj 337 Codd. Ottoboniani, 2865 52 jj jj 332 2866 684 LIST OF MSS. Here designated Place or Owner. No. in Colomb de Batines. Present Press-mark taken from Colomb de Batines, unless (as often) it was noted as having been since changed. 53 Rome, Bibl. Vatican 333 Codd. Ottoboniani, 2864 54 11 11 329 11 11 2373 55 Borghese 374 56 11 Chigi 377 L. vii. 253 57 n n 380 L. vi. 213 58 11 11 376 L. v. 167 59 Vatican 328 Codd. Ottobontant, 2358 60 11 n 323 Codd. Palatini, 1728 61 334 Codd. Ottoboniani, 2863 62 11 Chigi L. iv. 109 63 11 11 382 L. viii. 294 64 n n 384 L. viii. 293 65 Barberini 370 xlv. 109 66 11 11 372 xlv. 68 67 n 11 364 xlvi. 58 68 ,, Vatican 338 Nos. 7566, 7, 8 69 n 11 335 No. 3201 70 11 n 326 No. 4776 11 11 339 Codd. Urbinati, 365 72 Corsini 347 44- G. 3 73 11 11 348 44- F. 3 74 11 11 352 44. F. 28 75 Casanatense 342 d. iv. i 76 11 11 343 A. v. 55 77 11 11 344 d. iv. 2 78 Corsini 349 44. F. 26 79 11 11 354 44- E. 33 80 11 11 353 44- B. 33 81 11 11 355 44- F. 37 82 11 11 35 44. G. 10 83 11 11 356 44- D- 7 84 Turin, Bibl. dell' Universitk N. vi. ii 85 11 11 318 N. iii. 12 86 11 11 317 CXXII. 1. v. 33 87 Milan, Brera 254 A. F. xi. 31 88 Florence, Bibl. Laur. 28 Codd. Strozziani, CXLix 89 11 29 11 11 CL 90 11 8 6 Tempi 11 6 Plut. xc. Slip. No. cxxxiii 22 11 18 XL. No. xvi 93 11 14 XL. No. xi 94 11 65 XL. No. i. 95 11 75 XL. No. viii 96 11 25 XC. Sup. No. cxxx 97 Bibl. Magi. Palch. I. 35 98 no i- 33 99 11 91 ,, I- 43 ,E. B. 5. i. 7 100 p JB. A. 2. p. 3. No. 10 \Palch. 1164. 320 101 11 11 97 Palch. I. 30 IO2 Paris, Bibl. Nationale 420 Ital. 531 COLLATED OR EXAMINED. 685 No. in Present Press-mark t aken from Here designated Place or Owner. Colomb de Colomb de Batines, often) it was noted unless (as as having Batines. been since changed. 103 Paris, Bibl. Nationale 432 Ital. 530 104 421 Ital. 533 105 439 Ital. 532 106 ?> 436 Ital. 71 107 de 1'Arsenal 445 No. 8530 1 08 )> 446 No. 8531 109 447 No. 8506 IIO Nationale 437 Ital. 72 in 43 Ital. 528 112 438 Ital. 529 H3 )> 433 Ital. 542 114 424 Ital. 79 US j) '434 Ital. 76 116 Florence, Bibl. Rice. 134 1024 117 Rome Bibl. Chigi 379 L. vii. 251 118 , Barberini 37i xlvi. 38 119 Palermo, Bibl. Nazionale 411 xiii. G. I The following MSS. I have not examined personally, but their readings are reported on the authority of printed Editions or Collations (more or less complete), or from information furnished by the kindness of friends, which is acknowledged under the description of the MSS. Designation of MS. Place or Owner. No. in Colomb de Batines. Authority. A Florence, Bibl. Laur. I \ B C Rome, Bibl. Vatican Berlin Royal Library 3'9 525 I Witte's 4to. Edition. D l Rome, Duke of Sermoneta 375 1 E Cortona, Bibl. dell' Acad. 215 Lorini Etrusca F Monte Cassino 409 Reproduced verbatim in the fol. Ed. of the Tipografia di Monte Cassino, 1865 G Padua, Bibl. del Seminario 279 \ H 11 280 ->Q T I Sicca, Rivista delle varie 7 11 11 11 11 2ol 282 1 Lezioni della Div. Com. K Udine, Bibl. Florio 37 ) L Venice, Bibl. di San Marco 293 M Modena, Bibl. Estense 227 N Copenhagen University 534 O 11 11 533 P Florence 165 Scartazzini (2 Naples 407 s Stuttgard, Royal Library 523 Mussafia u Udine, Bibl. Bartoliniana 306 Vivarini V Vienna, Imperial Library 522 Mussafia 1 In the case of D it is to be observed that the readings given frequently differ from those in the margin of Witte's text, since the latter have to be cor- rected by some long and complicated Tables of Errata which are printed at the end of his Prolegomeni- LIST OF LINES OMITTED, TRANSPOSED, OR REPEATED IN ERROR. [This list is given to facilitate the comparison of MSS., since errors of this kind are not unlikely to be repeated in related MSS.] Designation of MS. A Inf. v. 97 100; xiii. 71, 2; xxxii. 36, 7. E Inf. xxiii. 9, 10. F Inf. v. 100; xiv. 56; xviii. 105; xix. 52. G Inf. v. 646; viii. 45 7; xxix. 56, 7; Purg. iv. 1249. After Inf. xxxiv. 55, there are 36 lines repeated in error. H Inf. xix. 70 2 ; xxvii. 43 5. K Inf. ii. 112 4; viii. 49 51; x. 136. L Inf. xvii. 72 5; xx. 64 6; xxiii. 64 6, 127 9; xxv. 21, 2 ; Purg. Hi. 133 145. Also the whole of Inf. Canto xviii. is omitted. Q Inf. xxiv. 435. X Inf. xxvi. 136 8. Tr. Inf. vi. 468. 5 Inf. xxiv. 40 43. f Inf. vi. 6971 ; Par. x. 47. ; Inf. xxiii. 143. //] xxiv. 49 ; Purg. ix. 8, 9. X // ii. 97 9. p Inf. xix. 53. r // vii. 313. $ //! x. 109 ii ; xxxii. 66. a Par. xxv. 13 5. g Inf. v. 127 /0 vi. 39; xvi. 75. 1 Purg. xxviii. 131, and several other lines noticed in the de- scription of the MS. 1 Par. xi. 19 21. q Inf. iv. 142 4; xxxiii. 119, 20. s Purg. x. 67 9. t Par. xv. 67. x Inf. xxiv. 84, 5 ; Purg. xii. 52 4. LIST OF LINES OMITTED, ETC. 687 a Purg. iii. 133 to end; Par. i. 32; xxxi. 20. c Inf. v. 100 2. / Purg. xxxiii. 63. n Inf. ii. 557. 8 Purg. xxv. 105 to xxvi. 19. 24 Inf. ii. 557. 36 Par. xxx. 1 1 6 136. 54 Inf. x. 109 n. 66 Inf. xix. 53. 68 Par. xviii. 91. 8 1 //; xv. 39 42. 82 Par. xi. 71, 2. 95 /#/ x. 9; .Par. x. 128. 97 Inf. xxiv. 97 132; Purg. xxxi. 4 45. 1 02 Purg. xxvii. 79 81. 105 Par. v. 78, &c. (See supra p. 671.) 106 Inf. x. 10911. no Purg. iii. 133 to end. LIST OF PECULIAR READINGS. [This list of the principal 'peculiar passages' noticed in the MSS. examined is added to facilitate the comparison of MSS. in the following manner. If a peculiar reading be found in any MS. under examination, the present list will shew whether it, or any other unusual form, has been noted in the passage in question among the MSS. above described. In that case the comparison may be carried further by means of any other ' peculiar passages ' which are noted under the description of each MS. If a prima facie relationship with any of these MSS. be thus established, by these 'Instances of the Finger post,' if we might so call them, the collations of the selected passages (180 in number) will afford the materials for pursuing the comparison sufficiently far to determine with tolerable certainty the degree and extent of the relation- ship in question. It is to be observed that, to save space, references are not given in this list (unless for some special reason), either (i) to passages in the Inferno occurring in any of the 17 MSS. fully collated ; or (2) to the selected passages, the record of the collation of which is given in the discussions supra, pp. 257 504. These can of course be referred to directly if required.] Inf. Inf. i. 35 A*. iii. 20 V 4 . ii. 10 A. 23 6. 12 V'.io^ 25 0.109. 15 5. 28 69 28 u. 30 46.73- 37 5.69. 38 69. 53 <.x.54.58.io6. 40 ^.54.106. 58 62. - 42 ;. 8 1 q.8o. 47 6. iii. 1 3 V 2 .A*. 49 6. 6 . 59 e.d. 7 X.r. 63 q. 8 0.30. 64 r.46. 9 p. 68 .66.io6. 36 i?. 86 A. 10.109. 105 e. 32 d.e. 1 1 1 109. I 19 T). 124 T;. 78 IJ.V. 79 r,. 90 r. 54 ^.54.106. 68 A. 39 -S-. 72 106. 85 V.v. 1 1 .54. Inf. x. 96 7. - 101 q. - in 97- 113 54. - 136 V. xi. 8 j. 12 T). 34 d. 37 24.51.62.77. 91 d. - 95 V. 108 15. xii. 28 p-54. 49 9.66. - 92 V. - 134 h. xiii. 147 V. xiv. 36 E. 48 7- 52 0. Ill TJ. XV. I U. xvi. 93 T). 122 X. - 136 /s. xvii. 8 T). 96 Z-x.l8.: 99 S.t. xviii. 43 1. - 1 10 V. xix. 89 V. xx. 30 6. 32 17- 69 53. 75 -y. - 87 *. xxi. 46 b. 91 0.f. - JI 3 30.56.57.90.119. - 138 Q. xxii. 6 j. 54.65.106. 64 Tr. - 101 24. 109 7. D. I. 690 LIST OF PECULIAR READINGS. Inf. xxii. 142 xxiii. 63 82 - 98 xxiv. 12 62 - 119 xxv. 1 6 - 105 xxvi. 3 1 64 - 120 xxvii. 46 65 8 5 xxviii. 24 71 - 138 xxix. 46 xxx. 19 86 xxxii. 34 - 136 xxxiii. 33 35 75 78 82 91 - 98 xxxiv. 47 50 93 118 c.i 3. m. 1. V. V 4 .^.86. Q.f. d.q.29- 56.57. 90. 102. 105.113. E. 119. V. S. f. S. L.2 7 . 9- 119. .(>. 1 06. 7- r,.d. c.h.j.1.97.99. 7- / 77- 77- q.77.ioi. Tr. y- 2. 15- 7- K.S.U.V. 23- V. Purg. i. 3 f- 14 f. 33 I- 45 f- - 119 x. ii. 26 1. 99 $.54106. iii. 64 d. Purg. iii. 85 8. - 103 d. iv. 19 k. 1 .69. - 124 77- vi. 48 V. - 109 p.S.27. - Ill w. - 125 f.i 5 . vii. 12 95- 15 j- 51 117. viii. 64 k.i. x. 30 d.g6. 49 f. 73 s. 74 a. xi. 105 1 8. xii. 17 f. 66 f. xiii. 1 02 f. - 105 116. xiv. 79 f. - 133 G1.1./Z. xv. 73 0). xviii. 57 28.116. - 58 4>. 83 117. xix. 140 f. xx. 104 vl -95-97- I 5- II 3- - 114 ^.$.54.106. xxi. 21 n- 25 117. xxiii. 36 19. xxiv. 65 W.F.A.a.j. 5.39.69.94. - 153 f. xxv. 3 1 u. 51 51.116. - 131 V. xxvi. 7 63. xxvii. 8 1 1. xxviii. 92 V- xxix. 67 V. xxx. 74 E.&.. 93 a. in 19- LIST OF PECULIAR READINGS. 691 Purg. Par. xxx. 112 E. xi. 21 xxxi. 43 ' S.V 1 .A.i;.i>.i.k.tA;#.'r/. 22 1.9.11.29.56.57.63. 26 68.74.79.90. 91 xxxii. 9 f. 105 - 102 v. in - 159 f. xiii. 27 xxxiii. 49 f. xv. 114 107 U. xvi. 37 - 38 Par. xvii. 13 ii. 115 S. 76 iii. 15 82. xix. 3 47 G.io. xx. 14 iv. 121 M. xxiii. 67 - 140 E. xxiv. 118 vi. 102 d. 119 117 17. xxv. 33 129 a. xxvi. 93 vii. 4 Y./. - 103 ix. 132 s. - 136 x. 98 v. xxvii. 100 - 119 v. xxix. 14 - 136 v. xxx. 141 f.t. jr-93. 1 1 8. S.F.d.b.f.Le. Y.F.u.t.d. 1 8.27.46.47. z/. 93- 1 IQ. d. E. O..F.K68.II9. d. A 73- W.a. i. a. v. 24.68. 29. 114. 24. e.j.l.t.v.dl y- 95- 442 APPENDIX. APPENDIX I. DANTE'S REFERENCES TO CLASSICAL AUTHORS AS BEARING UPON TEXTUAL CRITICISM. I had hoped in the present Appendix to discuss the general subject of the extent of Dante's reading from the evidence to be derived from his various works in prose or verse ; I purposed also to have brought to- gether a number of instances in which direct quotation or evident reminiscence of Classical authors might be employed not only to de- termine between disputed readings, but also to throw light upon the interpretation of difficult passages (a study perhaps the most interesting of all) ; and to give other such references possessing some special feature of interest or significance of their own. But I soon found that the subject, however inadequately treated, would extend far beyond the limits here available. I am therefore compelled to restrict myself to giving a few examples bearing upon the first only of the points above mentioned, viz. references relating to passages where the reading is disputed or doubtful. After all, it must be confessed that this class of quotations is alone strictly germane to the general subject of this work, viz. textual criticism. I shall hope for another opportunity hereafter of dealing with the subject, which I venture to think is an extremely interesting one, from a more general point of view. The following then are some examples of the process by which, as I have suggested in the Prolegomena p. xxxix, Dante's quotations or imita- tions of his favourite authors may be employed in the solution of textual problems. (i) The Vulgate. It is not necessary to stop to prove Dante's very great familiarity with, and unbounded reverence for the Bible, i.e. of course the Vulgate Version. His quotations from it are most numerous. I select the fol- lowing as bearing on questions of reading. (i) In Purg. vi. 109, the reading lapressura is much better supported than /' oppressura, which latter probably arose from a misreading of '/a- pressura'; but, other considerations apart, it is made almost certain by the use of pressura exactly in this sense in S. Luke xxi. 25, 'pressura gentium,' ' distress of nations ' ; and in S. John xvi. 33, ' In mundo pressuram habebitis,' ' ye shall have tribulation.' 696 APPENDIX I. (ii) In Purg. xiv. 133, 'qualunque m' afiprende* doubtless represents 'omnis qui invenerit me,' in Gen. iv. 14, which is here being definitely quoted by Dante, and it is therefore much to be preferred to mi prende, which would convey quite a different meaning. (iii) In Conv. i. n, we have usually "Per6 e scritto che il cieco al cieco fara guida, e cosi cadranno" &c. and this is said by Giuliani to be the ordinary reading of MSS. I find that my MS. of the Convttd 1 (like one cited by Giuliani) reads ' che se '/ cieco] which prima facie seems more natural, since the passage is evidently quoted from S. Matt. xv. 14, 'caecus autem si caeco ducatum prasstet' &c. (The parallel passage in S. Luke vi. 39 differs considerably in form.) But the question is, how did the difficilior lectio arise ? We may note that in any case e cost would suit an assertion preceding better than a supposition, and would so far tend to support the lect. vulg. which omits se. It seems possible that this reading may have arisen from a confused reminiscence of the earlier part of the verse, one form of which at any rate (in the Vetus Italica and other ancient Latin versions) is 'caeci sunt duces caecorum' (in which form it is quoted by S. Augustine and S. Cyprian 2 ) and not 'caeci sunt et duces caecorum.' The former might certainly be taken to mean 'The blind are leaders of the blind,' and from this to ' il cieco al cieco fara guida,' the transition would be easy enough in a quotation from memory. If the text were familiar to Dante, as to S. Augustine, in this form, the ordinary reading in the Coimito 1. c. would be explained. (iv) See the discussion of Par. xxii. 94 (sup. p. 475), for the bearing both on the reading and interpretation of the passage of the obvious citation of Ps. cxiv. (cxiii.) 5. (v) There is a very curious citation apparently of S. Matt. x. 8 (' gratis accepistis gratis date'} in Conv. iv. 27, "A grado ricevo, se a grado e dato," of which I cannot at present offer any explanation, since I do not find any variation here in the text or in patristic citations of this passage. (2) Aristotle. (i) There is a disputed reading in Conv. iv. 17, where the question seems determined by an Aristotelian reference. Dante, professedly re- producing ' la divina sentenza d' Aristotele ' respecting the several virtues, describes 'fortezza 'as 'la quale e arme e freno a moderare 1' audacia e la timidita nostra nelle cose che sono correzione (lect. vulg.} della nostra vita.' The reading should clearly be corruzione, the words no doubt having reference to Ethics in. xii. 2, where Aristotle, comparing the degrees of compulsion in cowardice and intemperance, says of the former, r\ p,fv Awn; e^iVn^n KCU (f)6 eipei rf/v rov e^o/Tos (frvo-iv [' corrumpit naturam habentis' 1 Ant. Trans!.]. My MS. of the Convito reads che sono di correc- ' tione, &c. 1 dated 1463. 5 auct. Sabatier, h. I. APPENDIX I. 697 (ii) The very difficult passage on the development of the embryo and the origin of the soul in Purg. xxv. 37 seqq. reproduces in several places the teaching and the language of Aristotle. Mr Butler's interesting col- lection of illustrative passages from Aristotle in his note p. 319 should by all means be consulted. There is an interesting difference of reading in line 51 where the lect. vulg. 'constare' seems to be established as against ' gestare,' and ' gustare ' (characteristic of the ' Vatican ' family of MSS.) by the word o-uw'ori/o-i in De Gen. Anim. used over and over again in reference to the same circumstance. See especially I. xx. 19 ; n. iii. 16; iv. 29; &c., &c. (3) Virgil. The following passages occur to me in which Virgilian quotation may be thought to bear on the question of disputed readings. (i) The use that has been made by Foscolo and others of Virg. sn. iv. 175, as possibly bearing on the var. lect. 'moto' in Inf. ii. 60 has been referred to in the discussion of that passage (supra, p. 271). It should be noted that the same place in Virgil is definitely quoted by Dante in Conv. \. 3, so that he was clearly familiar with it. In that passage the quotation of Virgil also strongly supports the reading ' La Fama vige (lect. vulg. vive] per esser mobile.' It is hardly necessary to call in its assistance to enable us to reject the false reading nobile (as in my MS.). (ii) In Inf. ix. 70 the reading ' abbatte e portafuor? (much preferable on other grounds also) is supported by Georg. ii. 441, Quas animosi Euri assidue franguntque feruntque. (iii) In Inf. iii. 31 orror as against error is made probable (inter alia) by ^En. ii. 559, At me turn primum sasvus circumstetit horror. (See further the notes supra on Inf. ii. 81 ; iii. 114, &c.). (4) Ovid. The quotations in Dante from Ovid are extremely numerous. For his mythology Dante seems to have been indebted to Ovid from beginning to end. Without further illustrations of this we may select the following as throwing light on disputed readings. (i) There is a very curious instance of imitation of Ovid in Purg. xxxiii. 49. The original passage is Metam. vii. 759 seqq. The word Naiade, which appears in Dante's text is quite meaningless in such a con- text. But, as Scartazzini explains, Naiades was in Dante's time the only recognized reading in that passage of Ovid, the true reading Laiades (i.e. CEdipus) being a more recent, though certain correction. The old text read thus : Carmina Naiades non intellecta priorum Solvunt ingeniis : 698 APPENDIX I. where Burmann notes " Ita veteres omnes libri." The now corrected text reads Laiades...Solverat. If further proof were needed of the source of Dante's language here, it will be found by comparing 1. 51 just below Senza danno di pecore o di biade with Metam. vii. 764 Exitio multo pecorumque suoque Rurigenae pavere feram. Note also the mention of Themis in 1. 47 and also in Ovid 1. 762. This shews how Dante in his mythological allusions followed Ovid, as we may say, blindly, since no one ever thought of connecting the Naiades with the Sphinx, or with the solution of riddles. (ii) In Purg. i. 9 the ordinary reading surga in the invocation to Calliope, is protected against the gratuitous and tasteless conjecture titrga (to avoid the repetition of nearly the same word in the rhymes) by Ovid, Metam, v. 338, Surgit, et immissos hedera collecta capillos Calliope querulas prsetentat pollice chordas. It may be observed that the fable of the Piche (see 1. n) is the subject of the context in which these lines occur in Ovid (11. 294 340). (iii) In Purg. xxv. 130 the imitation of Ovid bears upon the reading which is disputed, and clearly supports si tenne against si corse. The inci- dent, like so many others of Dante's examples of Virtues and Vices in the Purg. has been apparently suggested to him by Ovid. See Metam. ii. 453, &c., and especially 1. 455, ' Nacta nemus gelidum' in connexion with 11. 464 5 , I procul hinc, dixit, nee sacros pollue fontes, Cynthia, deque suo jussit secedere coetu. From this it is clear that Helice was driven out of the wood and Diana remained there. (See again Scartazzinfs note h. I.) (iv) In Inf. xxiv. 1 10, we have the two readings, d' incenso lagrime ed amomo and e d' amomo. This is a point of distinction as to which the MSS. would not help us, but the former reading is certainly more probable from a comparison with Ovid Metam. xv. 394, Sed turis lacrimis, et succo vivit amomi, where lacrimis belongs to turis only, to which Scart. says (note h. /.) it could only properly apply. The whole passage shews other features of reminiscence of, and resemblance with Ovid. (v) I have often thought that Ovid Heroid. vi. 40 (Jason and Hypsipyle), Detegit ingenio vulnera facta tuo, APPENDIX I. 699 lends considerable support to the reading senno instead of segni (which occurs in a far larger number of MSS.) in Inf. xviii. 91, where the same incident is being treated of as in Ovid I.e. See further the discussion of Inf. xviii. 91, sup. p. 322. (vi) There is a remarkable passage in Conv. iv. 27, where a transla- tion by Dante of five consecutive lines from Ovid Metam. vii. 507 n conversely throws back light on several disputed readings in the text of Ovid, or at least indicates (compare Ex. (i) supra) the text as it was read by Dante. The ordinary modern text of Ovid reads thus : Nee dubie vires, quas haec (al. nunc) habet insula, vestras Ducite, et omnis eat rerum status iste mearum. Robora non desunt : superat mihi miles et hosti. Dante translates : "non dite a voi dubitose le forze che ha questa isola, e tutto questo \2\ stato delle mie cose : forze non ci menomano, anzi ne sono a noi di soperchio, e lo avversario 2 grande" &c. From this it looks as if Dante read (i) dubias. Some editors have dubice, and Burmann proposes with some MS. authority to read dubias, for the usual dubie ; (2) Dicite 1 ; which has more MS. authority, auct. Burmann, than Ducite, though he prefers the latter ; (3) probably erat, especially since some MSS. of the Convito (e.g. my own, dated 1463) read "e tutto questo e stato delle mie cose " which is probably correct ; and so also does Ed. Sessa 1531. Further erat was the older and commoner reading, and moreover that of most MSS. At any rate Dante's translation, though not very clear, seems to have more relation to erat than eat. (4) He seems certainly, from his rather paraphrastic rendering, to have read hostis and not hosti \ and this also appears to have been the reading of the majority of MSS. and earlier editions. (5) Lucan. A great number of Dante's historical allusions or illustrations relate to scenes described by Lucan, and seem to be suggested by his familiarity with the Pharsalia. Lucan is quoted by him about a dozen times by name in his prose works. The following however is the only passage which I recollect that happens to bear on a disputed reading. In Conv. iv. 1 1 the early editions and apparently several MSS. (including my own, dated 1463) read thus : " Cio testimonia Lucano quando dice a quelle \sc. le divizie] parlando : Senza contenzione periro le leggi : e voi ricchezze, vilissima parte, moveste delle cose battaglia." This should certainly be corrected, as it is by Fraticelli and others, to the following (which has also MS. support) : " e voi ricchezze, vilissima parte delle cose, moveste battaglia." This is clear from a reference to the original, which will be found at Phars. iii. 119 121: 1 Compare another case of the interchange of duco and dice s. v. 'Cicero' Ex. (ii). 700 APPENDIX I. Pereunt discrimine nullo Amissae leges : sed, pars vilissima rerum, Certamen movistis, opes. (6) Statius. Out of many interesting quotations or reminiscences of Statius (such as the honourable position assigned to him in the Purg. would lead us to expect) the following seems certainly available for the correction of a fals. lect. in the Convito, which is found, according to Fraticelli, in nearly all the MSS. 1 The passage is in Conv. iii. 8 "siccome dice Stazio poeta del tebano Edipo quando dice che ' con eterna notte (fals. lect. ' nota ') solvette lo suo dannato pudore.'" The language of Statius is : Merserat aeterna damnatum node pudorem. (Theb. i.47-) (7) Cicero. Dante's familiarity with a great variety of Cicero's works is proved by the extremely numerous quotations scattered throughout his prose works, (especially the Convito), and there are also some very interesting illustra- tions of passages in the Divina Commedia to be derived from Cicero, which however, as not bearing on questions of reading, I am compelled to omit here. The following two cases may be noticed. (i) I have already pointed out in the discussion of Purg. ix. 17 (supra p. 386) how the passage in De Sen. 80 strongly supports the reading 'Piu dalla carne,' which seems on other grounds also to be undoubtedly the true one. (ii) It may be worth while to refer to a long quotation of Cic. Paradoxa c. i, occurring in Conv. iv. 12, because it appears as though Dante had read dixi instead of duxi, a case exactly similar to one noticed in reference to Ovid, Ex. (vi). The portion bearing on this point is as follows : ne 1' allegrezze, delle quali massi- neque eas quibus maxime astricti mamente sono astretti, tra cose sunt voluptates in bonis rebus aut buone o desiderabili essere dissi. expetendis esse duxi. I conclude by apologizing for the very fragmentary character of this Appendix. The instances here given may however serve as characteristic examples of the process of which I promised in the Proleg. p. xxxix to give some definite illustrations. 1 My MS. of the Convito (1463) omits the word altogether, reading 'con eterna solvette.' APPENDIX II. 701 APPENDIX II. THE 'VATICAN' FAMILY OF MSS., AND OTHER GROUPS. I have already mentioned in the Prolegomena that one result of the collation and comparison of the MSS. which I have described has been to indicate the existence of at least one distinct family of MSS. constituting apparently about one-tenth or nearly so of all those which I have ex- amined 1 . I propose now to give a fuller account of this family, and to explain the process and evidence by which the result has been arrived at. I would first remind my readers of the so-named ' Sienese ' family traced out by Dr Witte from the evidence afforded by the collation of over 400 MSS. in Inf. Canto iii. His account of the matter will be found in his Proleg. p. Ixxv, and more fully in Dante-Forsch. I. p. 280, &c. In reference to the family which I am about to describe, the process by which I arrived at it was as follows. The recurrence of two or three remarkable readings of B (Vat. 3199) in one or two other MSS., and notably (by good fortune) in my own MS. (' Z '), which I was able to consult at lei- sure and in detail, suggested the suspicion that B might perhaps be one of the so-called ' patriarchal' MSS. of Dr Witte (Pro!, p. Ixxv). I then selected a large number of passages from the margin of Witte's 4to Edition, in which B is registered as the sole authority among his four MSS. for readings not admitted into the text, selecting those where the variation appeared to be characteristic. This long list was examined in several MSS. which seemed likely (on the strength of some resemblances noticed in the test-passages) to be related to B. The result was that in a considerable number of such passages B was found to stand alone, or to have the support of only one or perhaps two other MSS. The next step was to eliminate such passages, as being apparently more or less peculiar to B, most MSS. having many such passages, and B an exceptionally large number. This much diminished list was then examined in a number of MSS. (especially in the Bodleian col- lection) in order to ascertain whether the readings which seemed common to B and its usually related MSS. were in any sense characteristic, or were more or less frequently found elsewhere. More than half of the passages hitherto retained were next dismissed on the ground that the readings were too commonly found in MSS. of all sorts to be considered as tests. The final result, bearing a very small proportion to the whole number of passages examined for the purpose, was the list given in the Tables which follow. These seem however quite sufficiently numerous and character- istic to indicate clearly the existence of a 'family' of MSS. closely related to one another, which I have styled the 'Vatican' family, since the 1 This estimate is perhaps slightly excessive in reference to the whole body of MSS. , as latterly I was rather on the look out for MSS. exhibiting these peculiarities, selecting these by preference for fuller examination. 702 APPENDIX II. Vatican Codice 3199 (Witte's B) is its best known and most celebrated member. Thus the readings examined for this purpose fell into three classes, and this would probably be the case with any other of such ' patri- archal' MSS. (1) Those peculiar to the MS. itself, or nearly so. (2) Those belonging to the MS. itself with a limited number of other MSS. (constituting its so-called ' family ') but comparatively rare else- where. (3) Those in the possession of which the ' family ' generally agree, but which also occur with comparative frequency in other MSS. bearing no further obvious traces of relationship, and which are consequently too common to be employed as ' tests.' The passages ultimately selected and embodied in the following lists are those seeming to fall under the second of these Classes, though of course the line of demarcation between (2) and (3) cannot be clearly drawn l . It is unfortunate that a sifting process of this kind necessarily occurs after, and not before, the examination of a considerable number of MSS. dispersed through various countries. Consequently a number of passages now seen to be significant were not collated, while on the other hand a vast number of references have since had to be rejected as useless. When it is considered that Witte's Sienese family of 1 7 members repre- sents the result of comparing more than 400 MSS., whereas this more numerous Vatican family has been derived from the examination of about half of that number, though of course over a much wider area of text, it is probable that it is a much more extensive group or family than the former. On the other hand, it should be remembered that Witte's com- parisons were confined to one Canto ; and, as we have shewn in the Prolegomena (p. xxxiii), relationships may exist of which in one or two special Cantos no trace could be found, and moreover which sometimes are limited to one or to two of the Cantiche, or even to particular portions of them. The Tables here following afford ample illustration of this. The important question next arises, What is the probable value of MSS. constituting such a family as this ? Does it represent a deliberate depravation or unauthorized recension of the text ? Can it possibly be a revision by the author himself? Does it seem to approach more nearly than the ' textus receptus ' or the ' lezione volgata ' to the original as it left the author's hands ? Though I have observed more than once that I believe it will almost always be found that MSS. exhibiting the Vatican type of readings have 1 Some, I think, of Witte's 20 passages of the Sienese type in Canto iii. are not uncommon elsewhere, and so would fall rather under the 3rd than the 2nd of the above classes. But here, as in the case above, it is the cumulative evidence of the common possession of a considerable number of passages together that carries weight. APPENDIX II. 703 an exceptionally good text 1 , yet I am by no means prepared to maintain that all, or even the majority, of the typical or characteristic readings themselves share in that excellence. Some at any rate are pretty certainly inferior, having the air of emendations 2 , while in regard to the majority of them it would perhaps be impossible to pronounce a decided opinion on critical grounds. Even as to those which would be judged on these grounds to be secondary, there is always the possibility of some of them being the author's own emendations, he being prompted by the same considerations in favour of a lectio facilior^ as would be likely to occur to later emendators. In the existence of a class of emendations of this kind some distinguished scholars firmly believe. (See on this Proleg. pp. xxi, xxii). My impression as to the general excellence of the text of this family of MSS. is remarkably confirmed by the fact that about one-third of Dr Witte's 'ist Class MSS.' exhibit Vatican readings 3 . It should be re- membered that the processes by which Dr Witte and myself have arrived at the conclusion that any MS. has a text of special excellence are quite independent. His judgment is based on the evidence of a single Canto, Inf. iii. (a Canto, by the way, in which I have not observed any readings characteristic of the Vatican family) : mine is founded on the possession by a MS. of a large proportion of readings, judged to be primary, scattered throughout most of the Cantos of the three Cantiche. The resulting verdict is frequently, though certainly not always, in agreement. But Dr Witte has himself told me (see Proleg. p. xxxiii) that he was far from regarding that list as final or exhaustive ; and, as I have already pointed out {Proleg. z$.), the method of examining a single Canto only would often give misleading results. My general conclusion would be that we generally find in MSS. of this family an exceptionally good foundation-text, one not indeed by any means wholly untampered with, but one that, while admitting certain occasional ' refacimenti] has probably preserved the general character of the original text with more than average purity. There is another small group or combination of MSS. consisting of about six MSS. which may be mentioned here. Their connexion is well marked and unmistakable. My attention was directed to it by the accident of some of these striking coincidences occurring in 4 MSS. in a single collection of 27 lately belonging to Lord Ashburnham but now in the Laurentian Library at Florence. It is singular that the examination of more than 200 MSS. afterwards should have shewn not more than two 1 The only case in which I received a distinctly contrary impression was that of the Paris MS. '105.' 2 e.g. Inf. xii. 16, 125; xviii. 42; Purg. v. 88; x. 14; xxiv. 61; xxxii. 102; Par. v. 88; xxv. 14; xxxiii. 57; &c. 3 I have either examined, or obtained information about, 20 out of the 26 MSS. included in Witte's list, and find as many as 8 of those to belong to this family either wholly or partially. 704 APPENDIX II. other decided examples of a combination, which on the ordinary calcula- tion of chances promised to be much more numerous. There are many other instances to be found in the collations of agreement between these six MSS. ; but I think the most striking and conclusive are those included in the Table appended. The extraordinary blunder of Stazio for sitio (or sitiunf) in Purg. xxii. 6 is perhaps the most singular proof of connexion. It will be observed in the Table how the degree of relationship (as in the case of the Vatican family) varies in the different Cantiche, A and o having much less of this peculiar colouring in the Inf., and e and 33 much less in the Par. (See on this Proleg. p. xxxiii.) Finally I will add a list of still smaller groups or pairs of MSS., between which some relationship, and sometimes a very close one, seems to be established by the evidence of the collations above gathered. In several cases fuller details of the grounds on which the alleged con- nexion rests will be found under the account of the MSS. themselves. E, H, P and often A, E, H, P. T and X (very closely) and to some extent also W and Tr. and F (very closely). p and 66 (very closely). 1 and 86 (very closely). , 54 and 106. i and / and (partially) 77. C and u. G and 109. F and d, and to some extent also I, d and 86. S and t. r and f. h and 79. k and i. o and 37. It has been sometimes suggested that probably some common features might characterize MSS. written in the same province or district of Italy, so that the classification of MSS. by their dialectic varieties into those of the Venetian, Tuscan, Neapolitan and other provinces might correspond with some distinction of 'families' based upon the possession of a common type of text. Such a phenomenon, if shewn to exist, would be explained by their common parentage from some ancient or celebrated exemplar MS., which at an early date became thus localized. I must confess however that I have not myself been able to trace any such results. It may however be worth while to add what I should judge to be the geographical classification (whatever it may prove to be worth) of the 17 MSS. here collated in full throughout the Inferno : Venetian or Lombardo- Venetian : ACEKMOPQ. Tuscan: DFGIL. Uncertain, or without any marked dialectic peculiarities : BHZa. TABLE THE VATICAN FAMILY OF [ + signifies that the reading is found : - - that some other reading is found : o that the altered or corrected : J that the reading is peculiar, and that reference should be made either to to these Tables themselves.] Designation of the MS. Inferno ii 12 Anzi che ,, 53 cortese e bella ix 89 Giunse alia porta x 20 A te mio dir ,, in tra 1 vivi xi 90 giustizia xii 1 6 Lo savio mio Virgilio ,, 28 su per lo scarco ,, 1 25 copria pur xiii 25 Credo, (or) Cred? io che ,, 43 Cost di quella scheggia usciva .. xiv 52 i suoi fabbri xvii 50 Or coi piedi, or col ceffo, q. m. Da pulci son, da moschi, &c. 96 braccia w' aggiunse xviii 42 Gia di veder xxii 6 mtiover giostra xxv 144 la lingua abborra xxvi 41 T)e\ftioco xxvii 39 or ven lasciai ,, 65 Non ritornb alcun xxviii 71 E cui (io) vidi gia in xxx 87 E piil d' un mezzo xxxii 128 Cos! 1' un sovra 1" altro i denti.. xxxiii 43 Gia era desto 74 tre di xxxiv 93 Qual era il punto Designation of the MS. B B V 1 V e 1 18 18 Note. , and x are quoted here because in the other Cantiche they belong distinctly to the text of a MS. may differ widely in different parts of the poem. The same remarks would a] the Purg. and Par., since in the Inferno they seem to have no trace whatever of connexion with affinity with this family in the latter half of the Purgatorio only, and scarcely any such connexion Cantos of the Paradise, though not without some trace of connexion in the latter part of the Pur& in the case of 93, 95, 102, 106 and 116. The numerous marks of '?' in 99 are due to the fr To come between pp. 704, 705] I. MSS. IN THE INFERNO. is. is defective, or that the reading has not been registered : * that the MS. has been :he description of the MS., or to the collations supra, or else (in a few cases) to foot-notes 27 28 29 31 54 56 5' 63 90 93 95 97 98 99 IO2 105 1 06 "3 116 4- 4- 4- 4- 4, 4- 4- marg + 4- 4- 7 + 4- ^ ~ i 4- z 4- ~ 4- 4-? + 4- 4- 4- 4- o o 4- + 4 4- 4-? 4- 4- 4- 4- 4- o j 4- 4- 4- 4- + 4- 4 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- -*- 4- 4 4- 4- 4- 4- 4- j 4- 4- 4- 4- 4- 4- 4- 4-? 4- 4- 4- 4- 4- + 4- 4 4- 4- 4- 4-? 4- 4- 4- 4- 4- 4- 4- o 4- o 4- 4J 4- 4- 4- 4- 4- 4- 4- o 4- 4- 4- o 4- 4 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- + + + 4- o 4- 4- 4- 4- 4- + + 4- 4- 4- 4- 4- + 4- + -] + 1 o 4- + 4- 4- 4- 4- 4- 4- 4- o 4- o + . . 4- +\ 4- 4- 4-? 4- 4- 4- o 4- 4- H 4- J 4- 4- . 4- 4- 4- o 4- ? 4- 4- 4- 4- 4- 4- 4-? 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- o 4- 4- J 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- ; 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- 4- + 4- ! 4- 4- 4- 4- 4- 4- + 4- 4- 4- + + + 4- + + 4- 4- 4- 27 28 29 3i 54 56 57 6 3 9 93 95 97 98 99 IO2 105 1 06 "3 116 this family, and they thus serve as a good illustration of the fact that the character of ply to the case of S and V which are omitted above though included in the Tables of this type of text. The readings of '19' are recorded because this MS. exhibits a sudden elsewhere throughout the poem. '28' shews a similar peculiarity in reference to the later torio also. The unevenness of the record in the different Cantiche is also very noticeable :quency of erasures and corrections. TABLE THE VATICAN FAMILY OF MSS. IN T Designation of MS. B Z V 1 L S V 1 V * X s q 18 Pttrg. v ^8 mezzci notte 4- 4- 4- o _(- -If. 4- 4- , , 88 fui Buonconte 4- 4- + + o -f -f -1- 4- 4- 4- vi 49 Ed io, Buon Dttca + 4- + + o -J- -f + 4- 4- x 1 4 lo stremo 4- 4- 4- o -f + 4- 4- 4- xiii 1 05 Fammiti noto + + + 4- 4- ,, 121 ch' io levai 4- 4- 4- + -f 4- 4- 4- 4- 4- ,, 144 Di la in parte + + + + + -f + -f- + 4- o 4- 4- xvi 142 lo fiume 4- + + + + + + 4- o o 4- ,, 145 Cos! parlb + + + + 4- 4- 4- xviii 57 Ed e prima appetibile 4- + + ?J + 1 + + + 4- o 4- 4- ,, 58 Che sono 4- 4- + + + + + + + 4- o + + xx 104 V altro parricida 4- + t + + -f o 4- 4- 4- xxi 101 Giunse Virgilio 4- o + + o 4- 4- ,, 112 insieme assommi + 4- + + + + o 4- 4- 4- 4- xxiii 36 E quelli dunque (or dunqua) ... xxv 5 1 fe' gustare + + + + + + + + o + 4- 4- 4- 4- 4- 4- 4- 4- xxvi 7 piu dolente 4- + + + + + + + + 4- o 4- 4- xxviii 123 fiume ch' aspetta 4- + + + + + + + + 4- o o 4- xxix 14 la donna mia a me + 1 + + o o o | 4- ,, 71 solo il lunie + ' 4- -f + + + + o 4- + xxx Hi che le rote 4- 4- + +? . o + ? xxxii 1 02 Di quella tonna 4. + + + t 4- 4- Par. iv 121 s\ profonda + 4- -f o + v + 4- + v 88 Lo suo piacere 4- if + + + S 4- + 4- x H2 nell' alta luce + + f + 01 + o 4- o xxiii 103 che spiro 4- + + + + + rC + 4- 4- ,, in lo lume di Maria 4- c 4- 4- 4- xxv 14 Di quella schiera 4- + + + C5 + 1> + 4- xxvi 93 ciascuna cosa + + + + + O 4- 4- 4- ,, 96 in />' dico 4- 4- + + + 4- (J ^o + 4- xxviii 90 gli occhi sfavillaro 4- + + + + O v 4- 4- xxx 141 per fame + + + + + + + n 4- U5 4- 4- xxxiii 57 cede la materia + + + + O 4- 4- 4- ,, 80 Tutti conflati 4- 4- + + 4- 4- 4- Designation of MS. * B Z V 1 L ' S V e V X S 1 18 t 99 is introduced because of its partial coincidence with this family in the Inferno, though the fr such relationship. In Purg. xxix. 14 B has la donna n xviii. 57 ^ has E prima e There are a few other typical readings which frequently occur in MSS. of this family, but which are sometimes because they are not unfrequent in other MSS. outside the family. It is worth while perhaps to cal Inf. i. 28 iv. 141 v. 126 Poi posato ebbi un poco. Tullio Lino. Faro come. vii. 108 malvage piaggie. x. 65 xiii. 63 detto il nome. le vene e' polsi. Inf. xviii. 55 - 91 Pitrg. iv. 72 ,, xix. 34 Io son colui change in F Ivi con senno. La qual non. Io volsi gli 01 gilio e ment I suspect that in Inf. xviii. 43 gli occhi instead of /' piedi would be found to be a characteristic read To come between pp. 704, 705] II. HE PURGATORIO AND PARADISO. 19 27 28 29 3 1 54 56 57 63 90 93 95 97 99t i 02 105 106 113 116 4-* 19 28 29 3 1 54 56 57 63 90 93 95 97 99 105 1 06 113 | r 16 equency of corrections makes it doubtful whether even there the original text of the MS. had any lia a se mi torse. app. V has E de prima app. not included in the Tables, sometimes because they are not found in the archetypal MS. '/?,' and ' attention to some of them. ( Comp. the converse urg. v. 88.) :chi al (e '!) buon Vir- Purg. xxiv. 6 1 piu a gradire. xxx. 15 La rivestita came. ,, xxxi. 43 Perche piu vergogna porti. Par. vii. 2 1 Fosse vengiata, or Vengiata fosse. xviii. 123 muro di sangue. re (almen tre). ing of the Vatican family, but I did not notice it soon enough to collect sufficient information. APPENDIX II. TABLE III. THE ASHBURNHAM COMBINATION . xxviii 33 Al congiugner , , , , 50 tanto piu festine.... o , The following are fairly common elsewhere, but are perhaps worth adding as illustrating the relationship of these MSS. : Inf. i v 142 Et vidi il geometra + o + xx 30 compassion porta . + i . . xxii 101 Si ch' io non tema i , , 1 icxxiii 78 Che forar 1' osso , r . Purg. i 108 Prendere ii 93 tant' ora .. , , . , 1 xiv 126 vostra ragion . . t ^_ . Par. v 88 Lo suo piacere (1) 33 has che tu vedrai. (2) e h&sfusser. (3) X \\3.sfan...volge. (4) X has nessuno. (5) X has 'e '1 mio M. mentre.' (6) o has Ai vendetta. (7) 33 nas the rare reading giudizio. Note. The same capricious differences in the relationship of the MSS. in the different Cantiche that has often been noticed before will be observed in the above Table. D. I. 45 706 APPENDIX III. APPENDIX III. THE INTERPOLATED LINES IN INFERNO C. XXXIII. In 1874 the curious interpolation which is the subject of this note was discovered by Dr Gregorio Palmieri in the Canonici MS. 103 (denoted here by 'E') and afterwards communicated by him to the Athenceum, Aug. 21, 1875. These lines occur without any break in the text or note of any kind. In 1878 he came across them again, this time however not in the text, but on a fly-sheet of a MS. in the Chigi Library at Rome (L. viii. 292) denoted as '36' in these collations. This is described by Batines (No. 383) as being early in the i 5th century, though Dr Palmieri believes it to be somewhat older, as I certainly do myself. The only indication that the writer, either of the text or of the verses on the fly-sheet (whether he were the same or different), or any one else using the MS., knew that they were intended to come in at Inf. xxxiii. 90, is an asterisk in the margin at that place. It should be added that the same fly-sheet con- tains a fragment of 10 genuine terzine from the Paradiso (xxx. 115 145), which had apparently been accidentally omitted in the text : that place also having an asterisk in the margin. Finally, in the year 1883, I myself met with these lines again in one of the MSS. in the Bibliotheque Nationale at Paris, where they occur in the text as in the Oxford MS., and also without any indications of interpolation or other irregularity. The MS. is that here denoted as '7,' and it will be found described in its place. I subjoin the three texts at present known. The orthography of the several MSS. has been modernised, but no further changes have been made in the way of emendation 1 . 1 Perhaps it may be worth while to give a verbatim copy of the original as it stands in the Bodleian MS. Quando cussi parlato latraffita guarday dalaltro canto evidi un fricto loqual piangea tremado la corata Et io lidissi perche setu costi ficto io te cognosco ben che se lucesse qual fallo ti reco cussi conficto Et egli ame poy che tu say mie onfese perche pur mi molesti va ala toa via se torni may insu nel bo paese Io no mi partiro alui disio pria seno mi conti perche se qua dentro che no po esser senza gran follia APPENDIX III. 707 Bodleian MS. Quando cosi parlato latraffita[?] Guardai dall' altro canto e vidi un fritto Lo qual piangea tremando la corata. Ed io gli dissi : Perche se' tu costi fitto ? lo ti conosco ben che se' Lucchese, Qual fallo ti reco cosi confitto? Ed egli a me: Poiche tu sai mie ofifese, Perche pur mi molesti? Va alia tua via, Se torni mai in su nel buon paese. Io non mi partiro, a lui diss' io, pria Se non mi conti perche se' qua dentro, Che non puo esser senza gran follia. Poiche ti piace, dico fuor talento Che per 1' inganno ch' io ai grandi usai, Ch' al popolo i sommisi a tradimento, L' inferno mi riceve sempre mai. Vanne e non portar di me ambasciata Perche qua dentro tu trovato m' hai. Chigi MS. (on fly-sheet only). Quand' ebbe si parlato la ristata[?] Guardai dall' altro canto e vidi un fitto Che piangeva, e gli tremava la corata. Ed io gli dissi, Perche se' qui fitto? 10 ti conosco ben che se' Lucchese. Qual fallo ti reco costi confitto? Ed egli a me: Poiche tu sai mie offese, Perche pur mi molesti? Va a tua via, Se tu ritorni su nel buon paese. Io non mi partiro, diss' io, pria Se non mi conti perche se' qua dentro, Che non puo esser senza gran follia. Poiche ti piace, dico fuor talento Che per 1' inganno ch' io ai grandi usai, 11 popolo i' sommossi a tradimento, Perpetuo son qui dentro a questi lai. Vanne, e piu non mi far omai ambascia Poich' io t' ho detto li miei forti guai. Poy che ti piace dico for talento che per longano chio aigrandi usay chal populo ison mesi atradimento Lonferno mi receve sempre may vane eno portar dime ambasciata perche qua dentro tu trovato may 452 708 APPENDIX III. Paris MS. (No. 427 Batines). Quand' ebbi cosi parlato esta fiata, Guardai dall' altro canto, e vidi un fritto Piangendo orribilmente tutta fiata. Ed i' gli dissi, Se' tu costi dritto? I' riconosco ben che se' Lucchese ; Qual fallo ti reco tanto giu fitto? Ed egli a me, quand' a guardar mi prese, Perche pur mi molesti, e va tua via, Se tu ritorni su nel buon paese. I' non mi partirb, dissi lui, pria Se conto non mi fai perche qua entro Se' tanto basso tra la gente ria. Ed egli a me, I '1 dico fuor talento, Che per 1' inganno ch' io a' grandi usai, Ch' al popol' i sommisi a tradimento, Concito mi riceve sempre mai. Or ten va, non dir di me ambasciata Poiche tra' traditori trovato m' hai. 1. Since 1876 I have examined, for various purposes, more than 200 MSS. of the Divina Commedia, always (inter alia) looking for this passage. I have never met with any trace of it until I came upon it in this Paris MS. 2. It is very curious and instructive to note the wide differences of reading which occur even in a passage so rarely found. This is by no means an exaggerated illustration of the treatment to which the text of the genuine poem has been subjected in very many places, and may serve to indicate the great difficulty by which the restoration of the text is beset. 3. The three texts exhibit the same puzzling interchange of agree- ments and divergencies which are found in MSS. generally. This too may illustrate the extreme difficulty and intricacy of the problem of tracing families or groups of MSS. Thus, 11. 4, 6, 7, 11, 12, 13 shew the Chigi and Bodleian MSS. in general agreement, and the Paris text independent of them. Per contra, 1. 9 shews an agreement of the Chigi and Paris MSS. ; while in 11. 16 and 18 the Chigi text is independent, and the Bodleian and Paris MSS. are more in relationship, though not in close agreement. 4. The Paris MS. is, I think, certainly of the fourteenth century, though probably late in it. Consequently, the origin of these verses is thrown back at least fifty years earlier than was indicated by the Oxford MS. If Mr Butler's ingenious suggestion, in the letter referred to below, be correct, we may hope to find a still earlier copy. 5. One or two more points occur to me as to this fragment in the Paris text : a. Fiata repeated in the rhyme in 11. i and 3 is inadmissible. The APPENDIX III. 709 same remark applies to fitto in 11. 2 and 4 of the Chigi version. Dante admits the repetition of a word identical in spelling if it be a noun in one case and a verb in the other ; or even the same noun in a different sense ; or the same word, when formed from different verbs, as volse in Purg. viii. 64, 6 from volgere and volere. (See further on this subject what I have said in the note to Inf. viii. 78, among the passages discussed 1 .) b. In the Paris text there is a more direct imitation in 11. 4 and 6 of Inf. xix. 50 and 52 than in the other two versions. c. The orthography of this MS. is not very marked, but seems to have a Venetian tinge. (The Venetian character of the Oxford MS. is unmis- takable on every page. In the Commentary which accompanies the text, the Venetian o: lography is much more marked, so that the scribe was probably a Vent ^-;an, who naturally allowed himself more licence in dialectic forms in thekotes than in the text.) d. I cannot tra any relationship between the text of this MS. generally and that of tue Bodleian MS., so far as can be judged from some twenty passages which I noted as peculiar in the Paris MS. I unfortu- nately had not with me at the time in Paris a far larger number of peculiar or testing passages which I had noted when collating the Oxford MS. I can scarcely imagine that the spurious character of this interpolation can be seriously doubted, (i) The inevitable repetition within so short a space of the three rhymes in -ata at the beginning and end of the fragment would be (as Dr Witte has pointed out) unparalleled 2 . (2) An- other objection is the inordinate length to which the Canto would be drawn out. It is already the second longest in the poem, being only exceeded by Purg. xxxii. which has one terzina more. (3) Then again, the first line has no connexion with the context, there being nothing to explain parlato, though it almost looks as if the interpolator imagined the previous apostrophe to Pisa to be an actual speech of the poet. (4) Passing over minor difficulties of detail, the generally inferior character of the verses, the fragmentary bits of palpable imitation of other passages in the poem, the more than Dantesque vagueness which leaves no clue whatever as to the person intended (Dante does not thus 'draw his bow at a venture'), and other similar considerations, must surely compel us unhesitatingly to condemn these lines as spurious. Besides, if we could imagine these lines to be genuine, how should we 1 B. Latini seems to observe a similar rule, e.g. Tesoretto xxi. 69, 70. Amico or ben ti membra Se tu per belle membra. ib. xx. 23, 4 D' >dir infin la fine Amico caro e fine. 2 It is hardly necessary to observe that the word latraffita in the Bodleian MS. is a mere blunder, providing neither sense nor rhyme. 710 APPENDIX III. account for their disappearance from the vast majority of MSS. ? And ought not the few MSS. which have alone preserved them to exhibit an exceptionally pure text in other respects, which is certainly not the case? Unless indeed, and this is the only possible supposition consistent with their genuine authorship, Dante wrote them later and intended to fit them in here, but had not as yet obliterated the lines of junction, nor do we see how in respect of the rhymes he could well have done so. Such an hypothesis would still leave their extreme rareness unexplained, and would have to face the objection arising from their scarcely deniable inferiority, and others above stated. Several scholars have discussed these lines since they were first made public, but all, as far as I know, have decided unhesitatingly against their genuineness. The following may be specially mentioned: Prof. D'Ancona in the Rassegna Settimanale Jan. 19 and Feb. 9 1879; Dr Witte in the second series of his Dante-Forschungen p. 493 seqq. 1 ; Mr A. J. Butler in an interesting letter in the Athenaum of Aug. 24 1879, in which he ingeniously and with some probability endeavours to fix the actual date and purpose of this interpolation as follows : " Matteo Villani, v. c. 34, relates ' how they of Lucca thought to free themselves from the bondage of them of Pisa, but their traitors would not.' After telling how the Lucchesi with the aid of their Contadini were on the point of succeeding, he goes on: 'ma forse non compiuto ancora il termine de' lor peccati; e pero avvenne che certi popolani che erano meno male trattati dai Pisani che gli altri, e alquanti degP Interminelli... tradirono i loro cittadini.,.11 popolo vile...lievamente si lascio ingannare.' This happened in 1355, and it would appear likely that some personal enemy of the Interminelli conceived the idea of avenging the treachery by handing one of them down to eternal infamy, availing himself for this purpose of the Divine Comedy. Unluckily it would seem that he was hardly strong enough to wield Dante's weapons ; for no one who is familiar with the master's style could ever suppose these clumsy lines to have proceeded from his pen." Note. In regard to the strange word/rz'//0 in 1. 2 of the Bodleian and Paris MSS., it might perhaps stand imflitto or afflitto, though it may well be doubted whether Dante would have used either fiitto or fritto. I have found in the title of an unpublished MS. in the Bodleian Library at Oxford ("Add. A. 12") the words " di fricti (i.e. afflittt] che avvenghono nel corpo umano. The Vocab. Tram. s.v. ' Afflitto' 1 recognizes a collateral form affritto and also affrizione. See also s.v. ajfritti. In any case it cannot here be supposed to be the colloquialism, or vulgarism, from friggere = 1 Dr Witte collects here several other curious specimens of similar additions or interpolations. Prof. Giorgi published in 1880 two whole Cantos, which I also had come across independently in examining the MS. here marked '30' in the Bibl. Vittorio Emanuele at Rome. APPENDIX IV. 711 'perditus.' Compare Inf. xxvii. 10, where the MS. 'L' has delafrito for deW afflitto. Some further instances of the interchange of r and / are given in the note to Inf. xxx. 19, and many others will be found in the collations supra. It is, I believe, a Tuscan, or at any rate a Florentine peculiarity. APPENDIX IV. ON THE TEXT OF WITTE'S BERLIN EDITION. In the progress of this work I have been continually more impressed with the general excellence of Dr Witte's text, and of its quite exceptional position among all other editions, ancient or modern. Most of these are from a critical point of view very bad, being no doubt to a very consi- derable extent under the evil influence of the celebrated, but most uncri- tical, Aldine and della Crusca texts. Scartazzini's text should be to a considerable degree excepted from this censure, though it appears to me far inferior to that of Dr Witte. But for the existence of Dr Witte's Edition, I should have been led to propose either the construction of a new text, or a very considerable purifying of those now current. As it is, my results have generally tended to confirm Dr Witte's selections in the case of disputed readings ; but I am sure he would have been the last to maintain that a text based on the collation of four MSS. only could be regarded as final (see his Proleg. p. Ixxv. seqq.}. Such it will be recol- lected is his definite plan, and he admits having been sometimes com- pelled by it to reject a reading which on general critical principles he would have preferred to adopt, though in a few very rare instances he has allowed himself to make an independent selection. He mentions (Proleg. p. Ixxx, note) 15 cases of this. My own method has been to be guided mainly by the application of certain recognized critical tests, even though led thereby sometimes to run counter to the weight of the external evidence derived from MSS. I think it will be sufficient to append a list of the principal cases in which it appears to me desirable to modify Witte's excellent text 1 . I am veiy far however from claiming certainty for my own conclusions, and would leave my readers to form their own judgment. Witte's text. Proposed alteration. Inf. i. 1 1 6 Vedrai gli Di quegli. iii. 8 eterna eterno. 136 cui sonno che '1 sonno (?) 1 I have suggested an alteration of Witte's text in a note on Inf. xxiv. 143 (supra p. 177) which I should wish on further consideration to retract. I believe that Witte's reading Negri is correct, not only as being difficilior lectio, but also because I suspect that Dante purposely substituted Negri for the more usual Ncri, ' 5t' vfipiv.' 712 APPENDIX IV. WittS s text. Inf. v. 64, 5, 7 vidi (ter) 96 si tace 107 Caina vii. 86 Ella - 89 le fa - 108 maligne ix. 70 fiori xii. 130 altra piu xiii. 63 le vene e' polsi Piu la mano alia segni spose voi. compassion porta temetti non istra 1 xiv. 126 xv. 29 xviii. 91 xix. 3 xx. 30 xxi. 93 xxvii. 21 xxviii. 135 Giovanni xxx. 115 falso, tu xxxiv. 33 fatte parti Purg. ii. 13 sorpreso dal vii. 5 1 ovver saria ,, viii. 94 Com' to parlava x. 30 Che dritto di salita xxi. 63 volar xxii. 5 n' avean - 6 sitio e senz' altro xxvi. 75 morir xxvii. 8 1 lor di posa serve - in piu lontani xxviii. 135 perch' io xxx. 73 ben son, ben son Par. i. 141 Come in terra ii. 1 24 Riguarda bene omai xi. 26 Non surse il secondo xviii. 75 or hmga schiera - 123 di segni - 131 Pietro e Polo xxii. 94 volto e retrorso - 95 Piu f u il mar xxv. 29 1' allegrezza 33 Quanto xxvi. 104 Dante 134 L s' appellava xxxiii. 128 Pareva in tre 1 The 410 Edition has istra, Proposed alteration. vedi (ter). ci tace. Cain. Questa. (See collations.} la fa. ( Do. ) malvage (?) fuori. altr* a piu. i sonni (or il sonno) e' polsi. Pur. la mi a alia, senno (?) spose e voi. passion porta. temetti cK ei (or che). issa. giovane. falso, e tu. fatta parte. sul presso del o non sarria. Com' ei parlava. Che dritta di salita. voler. n' avea. sitiunt, senz' altro. viver (?) lor poggiato serve. men lontani. purch' io. ben sern, ben sent. Come a terra. Riguarda bene a me. Non nacque il secondo. or altra schiera. di sangue. Pietro e Paolo. volto retrorso. Piu fu, e il mar. la larghezza. Quante. Date. / (or y) s' appellava. Pareva in te. and the 8vo isia. APPENDIX V. Some of the above suggestions are obviously of little importance, and many must fairly remain open to question. I have been guided some- times merely by the overwhelming majority of the MSS. collated, when no critical principle seemed to be involved, since the concurrence of many MSS. having no special tendency otherwise to agree seemed to have some weight. In other cases my reasons will be found stated either in the notes to the collation of the Inferno, or in the discussions of the selected passages. APPENDIX V. ON THE METRE OF THE DIVINA COMMEDIA. [Mr Tozer has kindly contributed the following Essay on the principles of metre and scansion observed by Dante. These can only be determined by such a minute study of the poem itself from this special point of view, as Mr Tozer has carefully carried out. I regret that I had not given sufficient attention to the subject to enable me to employ as often as I might have done the arguments available from this source for the purposes of textual criticism. It is evident that when it is once clearly established what is admissible and what is not in respect of metre and scansion, we should be able sometimes to reject summarily a reading otherwise well supported, as not conforming itself to the rules and principles observed by the poet.] The subject of metre is necessarily a technical one, but, notwith- standing this, it is of importance, not only to the student, but also to the general reader, because of its intimate connection with the power of appreciating poetry. In order to discover a faulty line, if such occurs, and to distinguish a more rhythmical verse from a less rhythmical one and thus much it is desirable that an intelligent reader should be able to do it is necessary to understand something, at least, of the principles on which the metre is constructed. This is admitted on all hands with regard to the classical languages, but it is true also, though in a some- what less degree, of modern poetry. Without this knowledge, a person may read a poem with an irregular, vague impression of its rhythm ; but this will probably be something different from what the author intended, and occasionally will be extremely lame. Now, in order to be able to test the metre of the successive lines as we meet them, and to know why they should be read in this, and not in that way a process which a very little practice renders easy it is necessary to determine such questions as the following. To what extent is it allowable for the poet to depart, for the sake of variety or for other causes, from the exact metre and rhythm of the lines? How many vowels in the same word may be pronounced together, so as to form one syllable for metrical purposes? 714 APPENDIX V. Is it ever allowable to pronounce two vowels, with a consonant between them, as if they were one syllable? How many vowels at the end of a word can be elided before a vowel at the beginning of the next word ? Under what circumstances may a vowel be left open and unelided at the end of a word before a vowel at the beginning of the next? It is the object of the present paper to throw some light on these and similar points with reference to the Dimna Commedia; and they are the more worthy of consideration, because metrical difficulties are steadily ignored by the commentators on that poem. The study of this subject is im- portant also for the emendation of the text, because by this means in some cases we can test the accuracy of a reading, and in others we can discover the origin of a corruption. I. The rule of eleven syllables to the line. Dante's metre is the regular heroic verse of eleven syllables to the line, and the verses are arranged in stanzas of three lines each. To the rule of eleven syllables there are certain exceptions in the form of ten- syllable and twelve-syllable lines. Those with twelve syllables are rare, for I do not include such as end with words in -io, -ia, &c., as ufizio, rabbia, where the two last vowels, if they occurred in the middle of a verse, would usually be pronounced as one syllable. Eliminating these, we find twelve-syllable lines only in six passages ; viz. Inf. xv. i, Ora cen porta P un dd duri margini, with a corresponding rhyme in argini, but no third rhyme, because it is a doublet at the beginning of a Canto ; /;// xxiii. 32, where the endings are scendere, rendere, prendere ; Inf. xxiv. 62, malagevole,fievole, and disconvenevole; Inf. xxviii. 80, cattolica, Maiolica, Argolica; Par. xxvi. 125, inconsumable, razionabile, durabile ; Par. xxviii. 125, girano, rimirano, tirano. One of these instances, it will be seen, contains proper names, in which all poets allow themselves a certain license ; while the rest, with one exception, are forms, the last vowel of which might be omitted in the middle of a line. The exception is margini, argini ; but in these words the two last syllables belong to the inflexion, and are therefore weak. The ten-syllable lines are more numerous, occurring in fourteen passages, besides two, which we need hardly count, in the Provengal passage at the end of Purg. xxvi. Seven of these are accounted for by one of the rhymes in the terza rima being a proper name with the final syllable accented ; viz. Inf. iv. 56, D" Abel suofiglio, e quella di Not, with corresponding rhymes re, fe 1 ; xx. 74, pud, co, Po ; xxviii. 32, All, qui, cosl; xxxii. 26, Osteric, Tambernic, eric; xxxii. 62, Artu,piii,fu\ Purg. iv. 68, Sion, orizzon, Feton ; xii. 41, Gelboe, te, fe\ Three end with Hebrew or Latin words, viz. Purg. xxiii. 74, Che mend Cristo lie to a dire : Ell, with corresponding rhymes dl, qui ; xxxiii. 8, Modicum et non videbitis me, with// fosti a Tnfoz richie'sto 1 . But it can hardly be right, with Mr Butler ('The Purgatory of Dante,' p. 369), to correct E que'sti sdtte col primdio stuolo (Purg. xxix. 145), into ''come il primaio stuolo} on the ground that the termination aio elsewhere in Dante reckons as one syllable only. As to the origin of these unusual forms of synizesis, Prof. Caix thinks (Le origini della lingua poetica italiana, p. 133) that words like giota, noia were already abbreviated in the usual pronunciation, so as to be monosyllabic, and that the terminations aio, oio came to be abbreviated from the analogy of these. On the other hand, the practice, which is so common in English, of pronouncing two syllables, where a consonant intervenes between the 1 See several other variations in the form of this line recorded supra in the collations. /2O APPENDIX V. vowels, as one or, if we prefer to put it so, admitting a trisyllable in place of a disyllable into the metre is never found in Dante, because in the Romance languages the number of syllables is regularly counted in poetry. Thus, whereas English poets use such words as 'covenant,' 'spirited,' 'prisoner,' 'memory,' 'shadowy,' 'boundary,' 'pardoning,' 'in- finite ' as disyllables, in Dante the reading Di que*i signori ddll' altfssimo cdnto (Inf. iv. 95), for Di quei signor &c., must be condemned at once, though it is found in 55 MSS., out of 218 examined by Dr Moore (see above p. 280), because it involves the last three syllables of altissimo being pronounced as two. In the Divina Commedia there are very few passages which are questionable on this ground. In Par. xxii. 13 Witte reads Nel qudle, se intdso ave'ssi i prie'ghi suoi, where quale must be pronounced as a monosyllable. Other editions read qual, and that no doubt is right. The same thing has happened else- where in the case of qual, quale, for in Inf. xix. 28, where Witte reads qual, which suits the metre, some other editors have quale. Prof. Caix has shewn in his essay quoted above (p. 133), that it was the tendency of copyists to insert final vowels in poetry, even when they violated the scansion, where the prose form retained them 1 . Again, in Inf. iii. 64, where Witte reads Que*sti sciaurdti, che" mai non fur vivi, almost all his MS. authority favours sciagurati, a form of the word which occurs in Inf. xxii. 44, but as a quadrisyllable ; Che' tu sdppi chi e lo scidgurdto. But there is every reason to believe that in Dante's time the pro- nunciation of the word varied, and that in the former passage the g would not have been pronounced, even if it was written 2 . But the most interesting passage from this point of view is Par. v. 70, Onde pianse Ifigenia il suo bel volto ; where the question turns on the pronunciation of Ifigenia. In modern Italian it is Ifigenia; and if the word is to be thus accented in this line, the two first syllables must be regarded as forming one foot with pianse which precedes. But it may be considered almost certain that Dante pronounced it as the Greeks of his time pronounced it, and as it has often been pronounced since that period 3 , viz. as Ifigenia j and with this accentuation the metrical difficulty at once disappears. 1 An illustration of this habit will be found noted supra p. 534, sub fin. 8 Many instances have been given above of the evanescent pronunciation of g from the practice of some copyists in reference both to its insertion and its omission : e.g. pp. 5", 532, 542, &c. 3 For instance, by Goethe in his 'Iphigenie auf Tauris, ' and by the Germans APPENDIX V. 721 IV. Elision. We now turn to the question of Elision, or the cutting off of one or more vowels at the end of a word before a vowel at the commencement of the following word. With a single vowel this regularly takes place, except where it is strongly accented, as it is in Restate m' e"ra, non muto | aspe"tto ; on which subject more will be said below. Two vowels are also con- stantly elided, or form one syllable with the vowel following ; as Ch' e principzV> / cagion di tiitta gioia ; Ch' ei fu dell' alma Roma e di suo zmpeVo ; Tu cre'di ch qui sta z'l duca d' Atdne. And these two may be in different words ; as Non vorrestz a /nvitar molte par61e (Inf. xxx. 129). Even three vowels may be cut off, or pronounced together with a fourth ; as Perch' to zhdugiai al fine i buon sospfri (Purg. iv. 132). V. Hiatus. This brings us to the subject of Hiatus, or unelided vowels. When the final vowel is accented, it may either stand before the fol- lowing initial vowel without further change, as in non muto aspetto above ; or may destroy, or absorb, the initial vowel a process which generally takes place when the initial vowel is weak ; as Ulisse e Dfome'de, e cos/ znsie"me (Inf. xxvi. 56) ; or the two may be pronounced together, as a sort of diphthong ; as Da man sinistra m' apparz z/na ge"nte (Purg. iii. 58). When the final vowel is unaccented, it is elided, as we saw in the last section ; but in monosyllables more license is allowed than in other words, and many of them remain, or may remain, open, even when unaccented, if they end in a strong vowel, or bear a special stress of meaning ; as Ve"de qual loco d ; inferno 6 da | e"ssa ; Ma | e"sso die altra volta mf sovve*nne ; Pero pur va, | ed m anddndo ascolta ; Ho -| 10 il braccio a te"! mestie'r disciolto ; Dall' dmpio loco, ove tornar tu | ardi ; Che uscir dovda da lui, e il chf, | e il quale ; Vidi pr^sso di me | un vdglio solo ; Qudndo tre | ombre insidme si partiro. generally ; also by Uryden in his ' Cymon and Iphigenia,' and by English writers before his time. 46 722 APPENDIX V. But some words to which this rule applies, as fit, which is open in Dinanzi agli occhi mi si fu [ offdrto, can form one syllable, not only with a weak vowel following, as Ed 10 risposi come a me" iu z'mposto, but also with a strong one, as Per6 necessitate fu dgli Ebrdi. Chi is never elided ; but da is, in one passage, viz. Che inviscaVa la ripa da tigm parte (Inf. xxi. 18). Che, which is unelided as an interrogative, as Dunque che | e ? perche, perch ristai ? is usually elided as a relative ; but to this last rule there are several exceptions ; as Per torre il biasmo, in che* | eVa condotta (Inf. v. 57). Ignorance of this has produced various readings in several passages, viz. Che | e occulto, come in e"rba 1' angue (Inf. vii. 84), where there is a variant Che v 1 e occulto (Witte's A) 1 ; Chd | alcuna virtu nostra compre"nda (Purg. iv. 2), with a variant si comprenda (Cambridge MS. quoted by Mr Butler, ' The Purgatory of Dante,' p. 37) ; Che I e lassu, cosi c6rre ad amore (Purg. xv. 68), with variants (i) CK e lasstl, e cosl, &c. (Witte's A, C), and (2) Che lassu e, cos\ y &c. (Witte's B). The conjunction che does not seem to be unelided. In the one instance in which the conjunction se is unelided, viz. Ov' e la colpa siia, se | e"i non crdde (Par. xix. 78), the Aldine and Delia Crusca editions read sed ei. It remains to consider a number of passages, in which the hiatus is exceptional. In several lines final a is unelided, where naturally it would be cut off : Tal era | 10 in quella tiirba spessa (Purg. vi. 10) ; Tal era | io, con voglia acce"sa e spdnta (Purg. xxv. 13) ; Tale era | io, e tale e"ra sentito (Par. xvii. 4) ; Che d' acqua fre"dda | Indo o Etiopo (Purg. xxvi. 21) ; Ch' io fui de' monti la mtra | Urbino (Inf. xxvii. 29): Era dnorata | dssa, e suoi consorti (Par. xvi. 139) ; 1 See other variants in the collations supra. APPENDIX V. 723 (Here Witte reads ed essa, though only one of his four MSS. (A) reads ed, and that is omitted by the reviser of that MS. (A 2 ). Scartazzini, with- out comment, reads onorata, essa with a comma between the words.) D' dcqua e di frdnde, che* si chiama | fda (Inf. xiv. 98) ; (This is the ordinary reading of the editions, but Witte, in accordance with all his MSS., reads chtamd 1 .} O is unelided in Piu v61te il mondo | fn Ca6s converse (/;// xii. 43). / is unelided in D' infanti | i di fe"mmine 6 di viri (Inf. iv. 30) ; O tu, che onori | 6 scidnza ed arte (Inf. iv. 73) ; (Here the editions generally read ogni scienza, but all Witte's MSS. read e.) Tra' quai con6bbi | Ettore dd Ene*a (Inf. iv. 122) ; (Here Witte's B reads ed Ettore, which is the reading usually adopted in the editions.) Ahi Geneve's!, | u6minf divdrsi (Inf. xxxiii. 151); (Here the hiatus is probably modified by the semiconsonantal u of uomini, aided by the comma. We may notice in passing, that some- times a vowel is similarly unelided before the diphthong eu, which only occurs in proper names, and can in every case be traced to a Greek source ; as Ma ve"di | Eiinoe che la deriva (Purg. xxxiii. 127); Dinanzi ad e"sse | Euphrates e Tigri (Purg. xxxiii. 112) ; but before Europa a vowel is elided ; see Purg. viii. 123; Par. xii. 48 ; xxvii. 84.) Ponti | e Normandfa prese 6 Guascogna (Purg. xx. 66). (Here, as the first word stands for Ponthieu in Picardy, there can be no doubt that it should be accented on the last syllable, and pronounced Pontl.} E is unelided in O frate, | fssa ve"ggio, disse, il n6do (Purg. xxiv. 55) ; (Here the hiatus is somewhat modified by the stop.) Fe*de | ed innoce'nza s6n reperte (Par. xxvii. 127); GraVi, a morir gli parve | dsser tardo (Par. x. 135). (Here some of the editions adopt morire for morir, but Witte, following all his MSS. reads venirfor esser.} In conclusion, it is satisfactory to observe, generally, that though questions of metre have rarely been entertained in the correction of the text of the poem, yet, in proportion as the text has been improved by more careful collation of the MSS., an advance has also been made in respect of metrical accuracy. 1 See other variants in the collations supra. ambrtoge : PRINTED BY C. J. CLAY, M.A. AND SONS, AT THE UNIVERSITY PRESS. 02 A 000 672 333 2