A A = _ d AS — O ■ — . o o S -^— — < o m ^= 2D o ■ ^^S m 2 9 _ 52 ■ O o 3 — > i — I 5 m ^-^— r_ 1 = 65 ^^^ JO 9 S J> ■ 33 1 ' — — -< 7 S "" 3> 1 4 9 - O 1 ,^ r 1 ^ "H 1 ^ ^^™ ^^— ■< 1 9 LIBRARY jjniversity of Caltfornl IRVING THE LIBRARY OF THE UNIVERSITY OF CALIFORNIA IRVINE GIFT OF JOHN & DEBORAH STRONG 837 TIME-REFERENCES IN THE DIVINA COMMEDIA MOORE a ©ifori PRINTED BY HORACE HART, PRINTER TO THE UNIVERSITY [Only Two hundred and fifty copies printed '.] THE TIME-REFERENCES I N THE DP7INA COMMEDIA AND THEIR BEARING ON THE ASSUMED DATE AND DURATION OF THE VISION BY THE REV. EDWARD MOORE, D.D. PBINCIPAL OF S. EDMUND HALL, OXFORD AND BAELOW LECTUBEB ON DANTE IN UNIVEBSITT COLLEGE, LONDON LONDON DAVID NUTT, 270 STRAND, W.C. 1887 ' The central man of all the world as representing in perfect balance the imaginative, moral and intellectual faculties all at their highest is Dante.' — J. Ruskin. 'In all literary history there is no such figure as Dante.' — J. R. Lowell. INTRODUCTORY NOTE. The substance of this Essay was delivered in two Inaugural Lectures in connexion with the Barlow Lectureship on Dante, in University College, London, in Nov. 1886. I am not aware that the theory which I have here advocated has ever been applied consistently through- out the Poem before ; and certainly not, I think, in any case with identity of results. The central principle itself has been occasionally recognized, and it is hardly necessary to say that in the inter- pretation of individual passages my conclusions or opinions have very frequently been anticipated. If it were not so, my work would at once stand con- demned, considering the number and quality of pre- vious writers on this subject: — ov yap tovtovs evXoyov bLap.apTa.vetv rot? 0A01?, dAA' iv yi rt rj Kal ra TrAeicrra KaropOovv. E. M. S. Edmund Hall, Oxford, Bee. 15. 1886. AUTHORITIES. The principal authorities (as far as I am aware) on this subject are : — Giambullari. Del Sito, Forma, e Misure dello Inferno di Dante [Florence, 1544). Dionisi. Serie d* Aneddoti {Verona, 1785-90). Mazzoni. Difesa della Commedia di Dante [Cesena, 1688). Ponta. Nuovo Esperimento sulla Principale Allegoria della Divina Commedia [Novi, 1846). Capocci. Illustrazioni della Divina Commedia [Najrfes, 1856). Grion. Che 1' Anno della Visione di Dante e il MCCCI, &c. [Udine, 1865). Della Valle. II senso Geografico-Astronomico dei luoghi della Divina Commedia [Faenza, 1869). Antonelli. Studi Speciali [Florence, 187 1). Pasquini. La Principale Allegoria della Divina Commedia [Milan, 1875). Philalethes. Dante Alighieri's Gottliche Comodie [Leip- zig, 1877). Lubin. Commedia di Dante Alighieri preceduta da Studi, «&c. [Padua, 1881). Scartazzini. La Divina Commedia [Leipzig, 1874-82). THE TIME-REFERENCES IN THE DIVINA COMMEDIA AND THEIR BEARING ON THE ASSUMED DATE AND DURATION OP THE VISION. The references to details of time throughout the Inferno and Purgatorio are very numerous, and in many cases extremely obscure and difficult to inter- pret. At the same time, they are so pointed and definite in character that we are evidently intended to attach to them a precise meaning. Further, they clearly have relation to a comprehensive scheme or plan running throughout the whole poem ; yet we find that besides the obscurity of individual passages, it is by no means easy (though I hope to show it is not impossible) to give a consistent and connected interpretation of these passages in relation to one another. They involve references sometimes to the position of the sun, but more frequently to that of the moon. The difficulties chiefly attach to the latter class, partly because the lunar movements are so much more varied and complicated, and still more because they of course entirely depend on the determination "7^ B Time-References in of the day of the month and the year to which the commencement of the Vision is to be referred. As it has been disputed, at any rate by some recent critics, whether that year is 1300 or 1301, and as everything, in respect of references either to the moon's position or to days associated with the Easter Festival, will depend upon the year to which such references belong, we cannot altogether avoid some discussion upon this important initial point. I wish however to make it clear at once, lest my handling of this subject should be thought inadequate (since many of the passages bearing on it do not in- volve 'time-references' in our present sense), that my main purpose is to discuss the indications of the hours by which the different stages of Dante's poetic pilgrimage are marked, so as to fit them into a con- nected scheme ; and more especially to deal with those depending on lunar phenomena which involve some difficulty in their interpretation. It is clearly however impossible either to frame a connected scheme, or to deal with the difficulties of isolated passages, without a preliminary determination of certain fundamental data on which the interpretation of many of these passages in detail must depend. Before then we attempt to bring together into a general view (as I shall do later) the various time- references to be found throughout the poem, we must discuss, or at least endeavour to come to some under- standing upon, these fundamental points, viz. the the Divina Commedia. year, the month, and the day, to be assumed for the commencement of the Vision. One other brief explanation. Let it be remembered that I do not purpose to discuss astronomical refer- ences as such, but only when they convey data of time. Consequently many passages that are con- spicuous in the elaborate works of Delia Valle and others will find no place in the present Essay. First of all then we must say something as to the year. The date of 1300 has been all but universally accepted from the time of the earliest Commentators down to the present day. The best-known advocate of the year 1301 is Grion, who, with much ingenuity and learning, maintained this, and one or two other paradoxes, in an elaborate monograph published in 1865. It would be impossible for me (as I say) to under- take a full discussion of this point now, since Grion's theory depends on ingenious but questionable infer- ences drawn from a considerable number of passages, and chiefly such as involve prophecies (often very ob- scure and uncertain in their application) of events future to the assumed date of the Vision, e.g. notably Inf. vi. 67, x. 79 1 , and others. Grion cannot cite 1 Though I cannot discuss the question here, I may say that I believe the very difficult prophecy in Inf. x. 79 refers to the period of the departure of the Cardinal Niccol6 da Prato from Florence on June 4, 1304, when the failure of the embassy of peace-making between the rival parties, on which he was sent by Benedict XI, was finally recognised, and the Bianchi were abandoned to the unrestrained fury of their enemies. If B 2 Time-References in any of the old Commentators as having pronounced for any other date than that of 1300. The most he can do is to claim the author of the Ottimo Commento (date 1334) on his side, not indeed explicitly, but by a hazardous inference from some of his presumably, because habitually, inaccurate statements. Boccaccio also, it is true, shows in one place some hesitation. In his Commentary on i. 1 l (in that interesting pas- sage in which he relates Dante's account of his own age given to a friend on his death-bed) he declares distinctly for 1300; and so again in his explanation of vi. 69. But when commenting on iii. 60 (the cele- brated ' gran rifiuto ' passage) he excuses Dante for referring thus to Celestine on the ground that that Pope was not yet canonized ; since Dante (continues Boccaccio) entered on this journey, as will appear in the twenty- first Canto of this Book, in 1301. Boccaccio is referring here to the important passage which we shall discuss presently, Inf. xxi. 112, but unfortunately his Commentary breaks off abruptly at Inf. xvii. 17. We cannot say therefore whether or not he has made a slip in this passing reference, or which of the two views to which he has inconsistently so, this would imply the popular way of counting a lunation as equiva- lent to a calendar month (which is one of the disputed points in the interpretation of Dante's language in this passage ; see Pasquini Prino. Alleg., p. 239, &c), and it would support pro tanto the adoption of a popular and not strictly scientific interpretation of other astro- nomical allusions, such as those discussed in the text later on. 1 In Ed. Firenzi 1724, the references to the three passages are — pp. 19, 352, and 149 respectively. the Divina Commcdia. committed himself he would ultimately have main- tained, when brought to face the passage in question. It will be noticed that in regard to the ground alleged in iii. 60 (viz. the canonization of Celestine being subsequent), the difference between the two years is absolutely immaterial, since Celestine, who died May 19th, 1296, was not canonized till May 5th, Another advocate of the year 1301 is Vedovati (Exercitazioni Cronologiclie, &c, 1864), but the principal reason for his adopting this view seems to be, that he, being one of those who believe in a predominant political signification of the poem, finds that 1301 (especially in reference to Canto I.) suits this theory better than 1300 1 . Per contra, among the latest and most distinguished writers on Dante— Witte, Philalethes, Lubin, and Scartazzini — all regard the date 1300 as quite beyond question. The last-named says in one of his most recent works, ' That the date of the Vision is the year 1 300 is known to every one, nor should there arise on this point the very slightest doubt.' To quote only one other authority — the learned Dionisi, writing in 1788 (Anedd. iv. p. 45), says as to the year 1300, ' Niuno, ch' io sappia, ne dubita.' This might perhaps suffice for our present purpose. But I will briefly call attention to five or six passages which convey irresistible conviction to my own mind ; 1 See Pasquini, Princ. Alleg. delta D. C, p. 231, and Suppl. Note on Vedovati, inf. p. 131. Time- References in their plain and obvious meaning outweighing a host of ingenious inferences based upon doubtful and obscure prophecies, difficult astronomical allusions, or questionable various readings, (i) We have the very- first line of the poem taken in connexion with the well-known passage in Cotiv. iv. 23, where our human life is compared to an arch, whose highest point is at 35 years ; whence Christ willed to die in his 34th year, so that His Divinity might not ' stare in descensione.' (2) In Inf. x. in it is distinctly stated that Guido Cavalcanti was still living. As he died during the winter between 1300 and 1301, either late in 1300 or early in 1301, this statement would be certainly untrue in the Spring of 1301. (3) The language of Casella in Purg. ii. 98-99 seems quite conclusive. He states that the Indulgence connected with the Jubilee of Boniface began just three months before, and that during that time the spirits delayed outside the entrance of Purgatory had felt the benefit of it. This was actually proclaimed on Feb. 22nd, 1300, but its privileges were antedated in the Bull itself from the Christmas Day preceding. This clearly necessitates the Spring of J 300, not 1301. (4) The age of Can Grande della Scala is given in Par. xvii. 80 as ' pur nove anni,' ' only nine years.' Now as he was born on March 9th, 1291, he would have been just a little over ten years in the end of March or beginning of April 1301 1 . (5) In Purg. xxiii. 78 1 Scartazzini's interesting note on this passage, Par. xvii. 80, should the Divina Commedia. 7 Forese Donati says that five years have not yet passed since his death. He is generally said to have died in the end of 1295. It is true that Benvenuto da Imola (who however is distinctly admitted by Grion to have declared decisively for 1300 as the date of the Vision) gives 1296 as the date of Foresi's death. But it is a well-known source of confusion in these old dates (as I shall notice later on) that the year was sometimes held to commence on December 25th, sometimes on January 1st, and some- times on March 25th, so that an event before March 25th, 1296, would be sometimes described as occurring at the end of 1295, and sometimes as at the beginning of 1296. (The so-called 'Revolution of 1688' in English History is a familiar illustration, as it occurred in February, 1689). (6) There is another, and I think very conclusive passage, which I do not see that Grion has noticed, in Par. ix. 40, when Dante says that the fame of his rather questionable Saint, Bishop Fulk (or Folchetto) of Marseilles, shall endure until ' Questo centesim' anno ancor s' incinqua.' This expression more naturally applies to the initial be read. Grion (p. 16), on the authority of one writer, maintains that Can Grande was born in May, 1280, and that by nove anni Dante refers to the years of Mars (in which planet he then was), which are rather less than double the length of ours. But apart from the inap- plicability of the rest of the language of the passage if Can Grande were then about 21, nine revolutions of Mars will be insufficient, and of course more so for the date 1301 than for 1300. In another work (cited by Scart.) Grion proposes to read died for nove, but even this seems to be insufficient for his purpose. 8 Time-References in year of a century (as it is popularly considered) than to any other. (7) Finally, there is an argument not depending on the interpretation of isolated passages, but on a characteristic running like a thread throuo-h the whole texture of the Divina Commedia, to which I have not seen much weight given, viz. the well-known fact that Dante never forgets the assumed date of his Vision, and speaks of events which had then already happened as past, but of all that had not yet happened, even when they occurred very shortly afterwards, as future, under the guise of prophecy. It will scarcely be disputed (though it would take too long to illustrate it in detail here) that that important line of division between history and prophecy is drawn at 1300 and not at 1301 ; nay, I believe we may venture to say in April, or at least before May, 1300 \ Finally, there are many obvious reasons of a general kind which suggest themselves for the selection of this date. It was the central point of the ' arco ' of Dante's own life (see Conv. iv. 23, already quoted) : it was the year of his Priorate at Florence, which was the source of so much of his political troubles : it was the beginning of a new century (at any rate in the popular opinion and parlance) : and it was also the year of the first Jubilee of the Church 2 . We may now then I think confidently assume, as a 1 Some examples in illustration of this argument will be found col- lected in a supplementary note. 2 See further Dean Plumptre quoted inf. p. 114. the Divina Commedia. starting-point for the discussion of the month and daij of the commencement of the Vision, that its year at any rate was, as is commonly supposed. 1300. It will I think conduce to clearness, and also perhaps bespeak more interest in the discussion of the ques- tions which follow, if I first set down briefly the chief landmarks which are clear and undisputed, to which therefore any solution must conform. I hardly dare to use the term 'undisputed' of any subject, or any passage, connected with this controversy, but I think there can be no reasonable dispute about the signifi- cance of the passages here collected, so far as I am proposing to employ them at present. Anyhow, dis- puted or undisputed, these are the data with which we are required to deal. They are of course drawn mainly from the Inferno, as we are dealing now with the question of the date of commencement of the Vision. The central landmark, so to speak, is Inf. xsi. 112, from which it appears that it was then Easter Eve, it being universally agreed that the ruins, here and elsewhere referred to in the Inferno, resulted from the earthquake recorded at the moment of Christ's death. (This is in fact certain from Inf. xii. 34-45. See also Far. vii. 48 1 ). We shall find then that the night between Holy Thursday and Good Friday is supposed 1 An interesting article will be found in Fornacieri's Studi (p. 31, &c.) on a theory of Benassuti, that the ' ruina ' of Inf. v. 34 (a much disputed passage) is to be explained in reference to the same event. io Time- References in to have been passed by Dante in the Selva oscura : see i. 21 — ' La notte ch' i' passai con tanto pieta.' He emerges thence, and is encountered by the Tre Mere (whose significance has been so much disputed), on the morning of Good Friday (see i. 37), the season being that of Spring, and the sun among the same stars as when he and they were first created (lines 38-40); i.e., according to tradition, in the constel- lation Aries 1 . The whole day is spent in painful hesitation and alternate advance and retreat, from the dread of these three Beasts — (the long duration of the conflict being indicated in lines 31-36, 59, &c.) — and also in the interview with Virgil, who comes at last to Dante's aid (1. 61, &c.) 2 , so that it was nightfall on Good Friday before they two together approached the Entrance Gate of Hell (see ii. 1, &c, also lines 141-142, and iii. 1, &c). Observe in passing how significantly Dante enters the Inferno at nightfall, and both Purgatory and Para- dise 3 at daybreak, and moreover the Earthly Paradise 1 See further a supplementary note on this. 2 Vellutetto expresses this very clearly in his note on Purg. xviii. 76 : ' Consume- il poeta tutto quel di fin' alia sera in defendersi dalle here e nel parlamento ch' ebbe con Virgilio.' 3 See Purg. i. 19, &c. ; Far. i. 43 ; and Purg. xxvii. 133, &c, re- spectively. It should be added that there is some dispute as to the last point, Mr. Butler among others maintaining that Dante left the Earthly Paradise, and entered the Heavenly Paradise immediately after drinking the water of Eunoe, i. e. at noon on the Wednesday. I still hold to the commonly received view as in the text. But even the Divina Commedia. 1 1 as well. We find him leaving the fourth circle just after midnight (vii. 97-99), and passing from the sixth to the seventh circle between 3 and 5 a.m. on Easter Eve (see xi. 113, 114, compared with xv. 52 ier mattina &c). He is leaving the fourth Bolgia of Malebolge (i. e. in the eighth circle) about sunrise, or (as he pre- fers to describe it) at moonsetting on Easter Eve (see xx. 1 25). He is in the fifth Bolgia of the same circle at 7 a.m., as appears from the very definite statement in xxi. 112 already referred to. It would be 7 or 10 a.m., according as the death of Christ is supposed to have taken place at the sixth or ninth hour, but that seems settled for us, as far as Dante is concerned, in favour of the sixth hour by Com: iv. 23, as we shall see presently. We find him at the end of the ninth Bolgia of the eighth circle early in the afternoon of the same day, when the moon is directly under their feet (xxix. 10). He then traverses the ninth and last circle with its four divisions, and finally passes the centre of the earth to the other hemisphere, between 7 and 8 p.m. (see xxxiv. 68), which he suddenly finds now to be between 7 and 8 a.m. in that hemisphere ; as is clearly indicated by the words in xxxiv. lines 96 and 105. We see then that the whole time occupied in traversing the Inferno is not much more than twenty- four or twenty-five hours, in fact from nightfall on the evening of Good Friday till a little after sunset on if Mr. Butler be right, the symbolism to which I have culled attention would be no less appropriate. See further note on p. 54. 12 Time-References in Easter Eve. It should be noted however, that this 7-8 a.m. is not, as we might perhaps at first suppose, the morning of Easter Day, but apparently the morning of Easter Eve over again. This however is a disputed point, which I shall discuss later on. Just, then, as the whole of the daylight of Good Friday was passed between the Selva oscura and the entrance of Hell, so this intercalated space of twenty-four hours, or more precisely of about twenty-one hours, is spent in pass- ing from the centre of the earth to its surface at the Mountain of Purgatory. When the poets emerged 1 per un pertugio tondo,' it was ' riveder le stelle ' (last line of Inf. xxxiv.) ; i.e. the stars shining before daybreak on Easter morning, as I shall maintain, or on the morning of Easter Monday, as many others hold. It was in fact about 5 a. m. in the morning ; i. e. when Venus and the constellation of Pisces were on the horizon, and the ' sweet hues of orient sapphire' already in the sky, as we learn from the delicious passage in Purg. i. 13-21. After the interview with Cato, the sun is just rising (see ii. 1), and it is full and brilliant day after the landing of the spirits from the boat of the ' celestial nocchiero ' (ii. $$) : the beautiful description of which scene is enhanced by the contrast (as no doubt is intended in this and many other parallel scenes and incidents in the Inferno and Purga- torio) with Charon, the ' nocchier della livida palude ' and his cargo of the ' mal seme d' Adamo ' in Inf. iii. 98-117. the Divina Commedia. i \ o I will not, however, pursue this sketch of the poet's progress further into the Purgatorio at present. It will be seen that we have been able to follow his steps so far, almost from hour to hour by the help of passages which speak for themselves, if we can once determine our terminus a quo, i.e. the clay of the ecclesiastical or civil year on which he assumes his journey to commence. Now out of all these passages three points emerge clearly, and to these particular attention should be paid: — (i) It was at the time of the Spring Equinox (i. 37-40) \ (3) He entered the Inferno the evening of the day after the Full Moon (xx. 127). (3) The actual day was Good Friday (xxi. 112). Here we have apparently three very clear and precise conditions, and so I believe we have really, if we merely take them all three in their simple, popular, and as I may say, natural sense. But un- fortunately it is possible to understand every one of these apparently plain and precise data in two different senses. These may be roughly described as (1) the scientific or ideal sense, and (2) the popular 1 This also appears from the curious and obscure passage in Par. i. 38, &c, where quasi in 1. 44 seems intended to indicate (as Buti notes) that the Sun was not then exactly, but only approximately, at the en- trance of Aries. 14 Time-References in or natural sense. These I will now proceed to explain in each case. I. As to the Spring Equinox. This is, of course, generally and popularly under- stood to be March 21st. In this, as in other respects, the Calendar was regulated, chiefly with a view to the determination of Easter (since that is celebrated on the first Sunday after the first Full Moon after the Vernal Equinox) 1 , by ecclesiastical authority, as early as the Council of Nicaea ; and by some ' particular and national ' Churches at an even earlier date. As a starting-point, the Vernal Equinox is commonly believed to have been fixed (though this point is itself the subject of much controversy, into which we cannot enter here 2 ), by that Council to be on 1 It will be remembered that the Jewish months were strictly lunar, so that ' the fourteenth day of the first month,' the date prescribed for killing the passover (see Exodus xii. 2-6), would of course be the Full Moon of the first month. This explains the form of the present rule which took the place of the Quartodeciman practice. Easter was to be the first Sunday after the first Full Moon after the Vernal Equinox, at which time the year was thought to begin. Thus it was the best ap- proximation that could be made to ' the fourteenth day of the first month,' allowing for (1) the fact that the months were no longer lunar, and (2) the condition that the day must fall on a Sunday. Hence among varying views as to the date of the Equinox — (March 1 8th, according to Hippolytus, c. 220 A.D. ; March 19th, according to Anatolius, c. 270 A.D. ; or March 21st, according to the view adopted either at or soon after the Council of Nicaea) — it was always insisted on that Easter should under no circumstances ever be kept before the Equinox, for thus it would no longer fall within ' the first month.' (See also an old Anglo-Saxon Chronicle quoted later on in the note on Purg. ii. 1-9, inf. p. 71.) 2 This is not the place to discuss the thorny question as to the pre- the Divina Commcdia. 1 5 March 21st, though as matter of fact it occurred in the year 325, on the afternoon of March 20th, and though owing to the ' vibration ' of the Equinox, consequent on the occurrence of Leap-years, and the inexact length of the Julian year, it might then fall actually on March 20th, 21st or 22nd 1 . According to the previous chronological arrangements of Julius Caesar, it was on March 25th. Owing, however, to the large error in the Julian Calendar, which was afterwards corrected by the adoption of the New Style, in the year 1582, by the authority of Gregory XIII, the true Equinox in the time of Dante had fallen back as far as March 14th (or according to Dionisi, March 12), and this discrepancy between the true and the assumed Equinox was of course con- tinually increasing, until by the time of Gregory's correction of the Calendar the Equinox had fallen back to March nth, and consequently ten days had to be omitted from the Calendar to bring it back again to the 21st. This large and increasing error was eise time and manner of the general adoption of the present practice of the Western Church for determining the observance of Easter. It seems to have originated with the Church of Alexandria, and to have differed from the practice of the Church of Rome, but by what precise means the latter church was persuaded to accept the rule of the former is much disputed. Those who wish to pursue this curious subject further will hnd ample materials in Hefele's History of the Church Councils (pp. 316-3:9 of Clarke's Translation, second edition), or in the elabo- rate Appendix on the Paschal Controversy in Dr. Butcher's learned work on The Ecclesiastical Calendar. 1 See Butcher's Ecclesiastical Calendar, pp. 54-6. 1 6 Time- References in certainly known to educated people in Dante's time. In 1267 Roger Bacon had calculated the Equinox as falling on March 13th, and had invoked the aid of Clement IV to correct the anomaly. (See Lubin's note, Studi, p. 362.) Moreover, Dante was himself perfectly aware of the error, since he alludes to it in Pa?\ xxvii. 142, anticipating the time when January should pass entirely out of winter (i.e. when the Spring Equinox should fall back even beyond January into December) in consequence of the one-hundredth part of the day which is neglected on the earth. Per la centesma eh' fe laggiu. negletta, the Julian year being too long by about that amount *. This then is the first question raised : — Did Dante refer to the Equinox in its real and scientific, or in its ordinary and popular sense? I shall maintain pre- sently, undoubtedly the latter, but meanwhile let it be observed that this dispute exists as to the first of our data. II. As to the Paschal Full Moon. With a view to the determination of Easter, it was not only necessary to fix the Equinox, but also to adopt some ' mean ' and not ' real ' motions of the Moon, partly on account of the variability of the ' real ' motions of the moon, and partly on account of differences of meridian. For this purpose approxi- 1 See further supplementary note on this subject, p. 118. the Divina Commedia. 17 mate Lunar Cycles have been adopted, the most celebrated of which is the Metonic Cycle to which the so-called Golden Numbers refer. Though it does not appear that this Cycle was formally recognized at Nicaea, yet it is found in use not long afterwards, and was possibly employed in the Alexandrian Church even in the third century. Dr. Butcher, in his learned and exhaustive work on the Ecclesiastical Calendar (to which I am indebted for many of these details), states that the Calendar Full Moon may differ as much as two, or sometimes even three, days from the real Full Moon. It is very important for our purpose to note that as a matter of fact it did so differ, by two days in the year 1300, and by even three days in 1301 1 . It is to be observed then that the moon's position year by year was practically ascertained by the help of Cycles and rule-of- thumb calculations, such as those given in the Introduction to our Prayer Books, only of course much less accurate. There was no reference to, or verification by, independent astrono- mical observations. Now the actual Full Moon by astronomical calcu- lations for the year 1300 fell on Tuesday, April 5th, as is stated by Lubin, Scartazzini, Philalethes, and others. I may point to an obvious popular proof that this calculation is correct, from the fact that 1 See the Calendars of this portion of the two years at the end of the Book. C 1 8 Time-References in there is an Eclipse of the Sun recorded as having occurred on February 21st in that year ; and since Solem quis dicere falsum audeat? if we count the days from that date (remembering that 1300 was a leap year), the next Full Moon but one will be found to fall on April 5th. The same result follows from working out the simple rule-of- thumb calculations given by De Morgan in his Book of Almanacs, p. xiii, for ascertaining the real New and Full Moons in any given year. On the other hand, it is no less certain that the Full Moon by the Calendar fell upon Thursday April 7th, in the year 1300. This again may be found by working out as before De Morgan's formulae for the Calendar Moons from the Epact of the year, &c, or by referring to any of the standard works of Chronology in which the principal Epochs of past years are given ; e.g. in the well-known L 'Art de verifier les Dates, Table Chronique, the ' Terme Paschal' (as it is there called) for 1300 is placed on April 7th, and so it will be found in any other similar works. Here then is a second question, and one much more vigorously disputed even than the last. When Dante speaks of the Full Moon, does he refer to the Real (or Astronomical) or to the Calendar Moon ? And the curious point about this is that some of the most distinguished Commentators and Editors have (as we shall presently see) missed the significance of this discrepancy between the two Moons altogether, and the Divina Commedia. 1 9 have assumed, either that Dante has made an unac- countable blunder as to the date of the Full Moon, or that for some reason or other his lunar references appear to be in strange and unexplained confusion. III. As to the date of Good Friday and Easter. According to a prevalent mediaeval belief, the actual day of the Crucifixion was March 25th, that event having taken place on the thirty -fourth anni- versary of the Annunciation : Uodem die (says S. Cyril of Alexandria) conceplus est in utero Christus et mortuus in G'ruce 1 . Moreover that day of the month was also thought to be the day of the Creation of the first Adam 2 . The inconvenience however of having a variable day of the week for Good Friday and Easter (even if the day could be ascertained with certainty), was thought to overbalance the propriety of having a fixed day of the gear, and the varying of the day of the week was one serious objection to the Quarto- deciman practice. From this point of view March 25th may be regarded as a sort of ideal, as opposed to the conventional, Good Friday. Again, the question 1 This is quoted by Pasquini, Princip. Alleg., p. 257. Dionisi also claims the authority of Tertullian, Lactantius, Augustine, Chrysostom, and others for the belief that the death of Christ took place on March 25th. 2 Bede in his Chron. puts the creation of Adam (and also of Eve) and the Crucifixion, ' Eodem decimo Kal. Apr. (i. e. March 23rd). Decebat enim una eademque non solum hebdomadis sed et mensis die secundum Adam pro generis humani salute vivifica morte sopitum . . . qua videlicet die primum Adam . . . ipse creaverat, eique de latere costam tollens,' etc. C 2 20 Time- References in has been raised, and different answers have been given to it : Did Dante adopt an ideal Good Friday, viz. March 25, or did lie follow ordinary custom, and refer to Good Friday as generally observed, which would be in fact, April 8 in the year 1300? One very precisely worded passage, viz. Inf. xxi. 112, has been thought to favour the former view. It is very far however from being decisive of the question, and regard must be had also to many other considerations 1 . There is yet another suggestion made under this head, and strange and improbable as it is, it is made, and supported (though with some hesitation), by no less an authority than Philalethes. He thinks it 1 There is another mediaeval tradition assigning April 6th as the actual day of Christ's death, to which some interest attaches from its adoption by Petrarch. Moreover, though it does not seem to have made the basis of any theory making Dante's journey to commence on the night of April 5th in the selra oscura (it will be noticed that it is just one day wrong for Philalethes' suggestion, noticed afterwards in the text), yet its possible bearing on the question before us has not been wholly overlooked. (See supplementary note on Mazzoni's Difesa, &c.) Petrarch, with very questionable taste, dates the commencement of his affection for Laura from that solemn day. See Son. iii. — Era '1 giorno ch' al Sol si scoloraro Per la pieta del suo Fattore i rai Quand' i' fui preso, e non me ne guardai, Che i be' vostr' occhi, Donna, mi legaro, and again, Son. clvii. — Mille trecento ventisette appunto Su 1' ora prima, il dl sesto d' Aprile, Nel Labirinto intrai ; ne veggio ond' esca. He refers to it also in some other passages. I should observe that April 6th was not the actual date of Good Friday in 1327, as it fell on April 10th in that year. the Divina Commedia. 2 1 possible, from the difficulty (or supposed difficulty) attaching to the interpretation of one or two pas- sages, that Dante may have followed a third method, viz. that by which the Jewish Passover was computed. This would have been sacrificed on the 5th of April in 1300, and the utterance of the words in Inf. xx. 127, i.e. 'the night before last the moon was full' would be on the 6th. Then the 'real' Full Moon being by calculation at 3 a.m. on the 5th, the expres- sion iernotte as applied to it on the 6th in Inf. xx. 127 would be appropriately and naturally so applied : since though the Full Moon was astronomically on the 5th, yet occurring as it did at 3 a.m. it would be popularly described on the 6th as being on ' the night before last,' i.e. the night between the 4th and 5th. Thus our three simple data (as they seemed to be) open a vista of much confusion, and we find at once the following different conceptions possible, and not only possible, but actually contended for, of our three, so to speak, fixed points. I will recapitulate them for the sake of clearness. (1) The Vision commenced at the time of the Spring Equinox. If so, it may be held to have commenced either March 14th, or 21st, or 25th, or else at some time or other in the early spring, when the Sun was still in Aries, but not necessarily on the very day of his entering on that sign. Here are four different views, of which I shall maintain the last to be the correct one. 22 Time- References in (2) Dante entered the Inferno the day after the Full Moon at nightfall. If so, it was the day after the Heal, or the day after the Calendar, Full Moon ; i. e. either on April 6th, or on April 8th. Certainly, as I hope to prove, it was the latter, i.e. April 8th. (3) He entered it on the Evening of Good Friday. This again has been understood to signify either March 25th, April 5th, or April 8th. Again, as I think can be conclusively proved, it was on the last named, viz. April 8th; i.e. the actual Good Friday of the year 1300. Now I have said each of these days (and I must add several others besides) have found their advo- cates. Critics have pounced upon one or two passages which seemed readily (or as they may have thought, crucially) to satisfy some one or more of these required conditions. They have then con- veniently shut their eyes to other passages, and to the necessity of adopting a solution which should satisfy all the three conditions of the problem coin- cidently and simultaneously. Surely taken all together they form a threefold chain, which is not, or ought not to be, easily broken. I must next point out what are the principal views, amidst all this variety, that have been held or seriously maintained, though I may be excused from examining any but a few of the most important or best supported of them. In fact, I may allege the plea and borrow the language of Aristotle [Ethics, Bk. I. the Divina Commedia. 23 C. iv) : "Airaaas fxkv ovv k^ra^iv tcls ho^as fxaraioTepov icrajs ZcttIv, Ikclvov Se tcls /^aAicrra kirnroka^ovcras, r) 80- Kovaas tx €LV Tiva Xoyov. According to Dionisi (Anedd. iv. p. 45), Pierfrancesco Giambullari (who wrote a work on the Inferno, 1554) 1 was the first to ascertain by astronomical calculations that the Paschal Full Moon of 1300 was on Monday April the fourth (for so he puts it), and about fifteen hours after midday, i.e. what we should call 3 a.m. on Tuesday April the fifth ; and he was therefore the first to find any difficulty or discrepancy in the data given by Dante in his poem. To meet this he adopts a theory that Dante intentionally dis- regards the facts with a view to some mystical propriety in making the Moon Full while he was in the Selva Oscura, and so he describes it as occurring on Thursday instead of Monday in Holy Week. Giambullari's words are ' II Poeta nientedimeno, per servirsene forse al senso mistico, dice ch' ella fu tonda la notte che si ritrovo nella selva, laquale . . . f u la notte che e tra il Giovedi ed il Venerdi Santo.' This at any rate implies that Giambullari held to the view that the Vision commenced on Good Friday, April 8th. 1 Since writing the above, I have been able to meet with a copy of this now very rare work. It is entitled Del Sifo, Forma, et Misure dello Inferno di Dante. As this title indicates, it is mainly concerned with the topography of the Inferno and the measurements of its several parts ; following the lines of, and supplementing, the larger and better known work of Manetti with the same title. The passage in the text, though correctly cited by Dionisi, occupies a very subordinate position in the work. It occurs on p. 26. 24 Time- References in Dionisi adds that the discovery by Giarnbullari of this apparent discrepancy between Dante's language and the facts of the case attracted much attention, but that his solution was not generally accepted. Mazzoni 1 , Pelli, and others preferred to suppose that Dante's vision commenced on the evening of Monday in Holy Week 2 , in order that the day spent in Paradise should fall on Easter Sunday. It appears to me that this apparently appropriate feature in the scheme depends on the mistake made by the Commentators of not noticing that Giarnbullari, while correctly stating that the actual Full Moon occurred (note the unusual expression) fifteen hours after midday on Monday, departed from ordinary usage in speaking of that hour as part of Monday at all, whereas it was clearly 3 a.m. on Tuesday, April 5th. Since then it is at any rate clear that Dante did not enter the Inferno till the evening after the Full Moon, whenever it was, this would be the evening of Tuesday and not of Monday, and consequently the passage through Paradise would not fall on Easter Sunday at all, but on Easter Monday. Lombardi 3 appears to me to have fallen into a pre- cisely similar error, in making the commencement of 1 Mazzoni's treatment of this subject is quite ' a curiosity of litera- ture.' See Supplementary Notes. 2 Anedd. iv. p. 46. 3 See his note on Inf. xxi. 112, when he introduces the above view with the words ' viensi per le vie additateci dagli astronomi a rilevaro che.' &c, as above. the Divina Corn-media. 25 Dante's journey (i. e. the wandering in the Selva oscura) fall in the night between Monday and Tues- day in Holy Week. He observes that Dante did not therefore compute the Anniversary of the death of Jesus as falling on the Good Friday of that year, but on the Tuesday, April 5th, this being the da,y fol- lowing the Paschal Full Moon, and therefore (?) the day on which Christ was actually crucified (Come dal Vangelo si raccoglie .... nel giorno seguente al plenilunio antedetto). In other words, he seems to adopt a theory something like that suggested by Phi- lalethes, that the actual day of the Crucifixion is fixed by Dante according to the Jewish method. But he puts the Full Moon (misled apparently by the lan- guage of Giambullari) on the fourth instead of the fifth, yet by the compensating error of placing the Crucifixion on the day after the Full Moon, instead of on that of the Full Moon itself (Eccod. xii. 6, &c), his ultimate conclusion agrees with that of Philalethes. Dionisi himself speaks with some hesitation. At one time (p. 69, &c.) he appears to maintain strongly that March 25 was the day of Dante's entering the Inferno, pointing out that that day actually did fall upon a Friday in the year 1300. At the same time he throws out an alternative suggestion (p. 70) as tenable, though less probable, that Dante may have taken the true Equinox, which he puts at 4.40 p.m. on March 12th in the year 1300, and then have carried back to this corrected date the day both 26 Time-References in of the Annunciation and Crucifixion, and by conse- quence that of the commencement of his Vision. This is a series of the most extravagantly improb- able hypotheses, as he seems to be himself a little conscious. He finally defends the view which he prefers against what he calls ' la sola obbiezione che si possa fare ' by the consideration that the Vision though ' maravigliosa e quasi divina, e pure fittizia' (p. 75), and consequently that Dante may not have troubled himself about the actual days of the month, or of the week, or of Full Moon, or of Easter in 1300; but assumed everything to be then in a typical condi- tion : the year was the year of Jubilee ; the season that of the Spring Equinox ; the Moon Full l ; the Ecclesiastical Year culminating with the Festival of Easter ; and so on ; all this being irrespective of any idea of piecing these general ideas together into a consistent whole of actual occurrence at any given time. The answer to this would seem to be : — If so, why has he gratuitously puzzled and confused us by a series of minute and precise, yet altogether mean- ingless and misleading, data of time, such as those we have already noticed, besides many more, and even more minute, references in the Purgatorio 1 I pass over rapidly most of the theories to which 1 See p. 73 : ' Se non fu piena la Luna allora, lo fu nel dl della sua creazione ; che Dio certo mostrolla tutta illuminata dal Sole.' See further on, p. 31, note 2. the Divina Commedia. 27 these difficulties have given rise, such as that the Vision commenced (i. e. the wandering in the Selva oscura) on the night between April 2 and 3, i. e. on the eve of Palm Sunday (Gregoretti). So also Ponta, in his Appendix to the Princ. Alleg. della Divina Commedia, p. 227, where he starts from the strangely- inaccurate statement that the Paschal Full Moon in 1300 was oh Palm Sunday, the 3rd day of April ! Or again between April 3 and 4, that is on the evening of Palm Sunday itself (Torricelli) : or be- tween the 4th and 5th of April (Arrivabene and Philalethes, though the latter doubtfully) : or that it lasted from the 4th to the 16th of April (Lanci). Capocci again makes it begin on the night of Palm Sunday April 3rd in the Selva oscura, and extend to the end of Easter Sunday, April 10th. These and other theories or guesses are given by Pasquini in his Principale Allegoria della Divina Commedia (1875), pp. 229-30, and also by Lubin in his very elaborate and exhaustive Sfudi, p. 360. Many of these views, in the absence of reasons given (as I have not always been able to consult the original authorities), seem to me to have nothing whatever to recommend them. I only mention them to show how generally and how severely the difficulty that we are discussing has been felt. There seem, however, to be four views which either on their own merits, or else on account of the dis- tinguished names by which they have been advocated, 28 Time-References in deserve to be considered as within the range of prac- tical criticism. They are these : — I. That the Vision began on March 14th, or the day of the true Equinox ; which is the view most in favour with the early Commentators on the Divina Commedia. II. That entrance of the Inferno is to be placed on March 25th, i.e. the 'ideal' Good Friday, as is main- tained by the distinguished modern Commentators, Fraticelli and Scartazzini. III. That it is to be placed on April 5th, the date of the Jewish Passover. This is suggested, as we have seen, somewhat tentatively by Philalethes, who, how- ever, generally offers three alternative dates in his notes, regarding none of the views as free from difficulty. IV. That the entrance of the Inferno was on Good Friday, April 8th, but that all the lunar references are to the Astronomical or Real Moon on April 5th ; this is held, with candid admissions however of the unsolved difficulties which it involves, by Lubin. There are two of the passages to which I have referred, which must be kept continually in mind, viz. Inf. xxi. 112, which plainly states that it was then Easter Eve: and Inf. xx. 127, which as plainly declares that ' iernotte; that is on the night between Thursday and Friday, the Moon was Full. These are the two cardinal points which must never be lost the Divina Commedia. 29 sight of; the two main conditions which any theory- is bound to satisfy. I. Now of the four views just enunciated, the first was adopted among modern writers (I believe) by Giuliani, but chiefly (as I have said) by some of the early Commentators, who generally dashed at a con- clusion without looking beyond the passage in hand, just as the copyists of MSS. in their textual emend- ations, often limit their critical vision to the compass of a single line. Moreover, they wrote before the discrepancy between Dante's allusions and the actual position of the moon had been noticed or suspected, since this was first observed, as we have seen, in 1544. This theory may therefore be disposed of without waiting for the consideration of the catena of passages of which it took no account. The only reason for the assumption of the date of March 14th or 15th was that that was regarded as the date of the true Equinox at that time, and this seems to suit a very rigidly literal interpretation of a single passage, viz. i. 38-40, taken in conjunction with the recognized tradition as to the day of the beginning of Creation \ This implies that we are to bind down the reference in the passage in question to the actual day of the Sun entering Aries (and that moreover astronomically corrected), instead of re- ferring it merely to the season, viz. early spring, when the sun was still in that sign ; which sufli- 1 See further the supplementary note on this subject. Time-References in ciently answers to the expressions used by Dante. It is not worth spending time on the detailed ex- amination of this theory. It is enough to point out that the days in question {viz. March 14th and 15th) were Monday and Tuesday in 1300, Tuesday and Wednesday in 1301, and that the age and position of the moon would be quite unsuitable to all Dante's references in either case : the moon being then about at her Third Quarter in 1300, and just after New in 130 1 1 . At the same time this view seems to be favoured by several of the older Commentators who say that the Vision commenced circa mezzo Marzo. So Pieiro, the Ottimo, and Benvemito. But how worthless such opinions are, may be gathered from the glaringly false statements with which they are connected, e.g. ' f u allora la Pasqua fra Marzo' (Ott.); or 'circa la meta di Marzo, nel Venerdi Santo : la Pasqua caddi allora in Marzo ' (Benv.) 2 . We have seen that 1 Moreover there is a noteworthy passage in Par. xxvii. 87 where Dante, then himself in the sign Gemini, describes the position of the sun thus : — il Sol procedea Sotto mie piedi un segno e piu partito, i. e. the Sun was one sign and something more distant from Gemini. From this it seems clear that the assumed date of the Vision could not be the true Equinox, as the Sun would in that case still be near the beginning of Aries, and so very nearly two whole signs distant from even the nearest point of the constellation Gemini. 2 Pietro's comment on Inf. xxi. 112 may be added, ' Nota quod auctor ostendit in hoc capitulo Christum crucifixum fuisse in medio Martii in aetate annorum xxxiv, et hoc opus incepisse in medio dicti mensis mccc.' the Divina Commcdia. 31 Easter fell on April in both years. Did it never occur to these old writers to ' verify their references ' ? II. Next as to the theory of Scartazzini and Frati- celli, viz. that the Vision must be held to commence on March 25th 1 , as the actual traditional day of the Crucifixion, which happened also, as will be noticed, to fall on a Friday in the year 1300. Of course the obvious objection arises that the Moon was New just when Dante's references state that it was Full. Scar- tazzini in answer to this says, that this plenilunio was merely ' una finzione poetica alia quale fa piede la tradizione della creazione del mondo.' But, if I rightly understand this suggestion, it is that Dante has had regard to the ' ideal ' and ' traditional ' date, not only in respect of the Crucifixion, but also in respect of the Creation, and moreover in the case of the Creation of the Moon as well as that of Man ; the Moon being of course presumably created as a Full Moon 2 , ev\oy(t)T€pov yap rj aWrjv Ttva. Well, but if SO, 1 It may also be remarked that according to Florentine (and some other mediaeval) usage the year began on March 25th. 2 On this see Dionisi quoted sup. p. 26, note. B. Latini seems to have held the curious view that the Moon was created New. See Tes. ii. 48 : ' Et sappiate che '1 primo anno del secolo si fa el primo giorno de la Luna. La Luna hebbe el primo dl di Aprile 10 dl,' &c. In another part of the chapter however he seems conscious of some anomaly in supposing the moon to have been created in an invisible condition. The Tenth Century Anglo-Saxon Manual already quoted, sup. p. 14, says (p. 4), ' On the same (fourth) day He placed the moon Full in the evening in the East together with shining stars in the course of the autumnal Equinox, and fixed Easter-time by the beginning of the Moon.' 32 Time- References in the ideal plenilunium would occur on the fourth day of Creation, i.e. on Wednesday, March 23rd, and not on Thursday, March 24th, as would be required by the expression iemotte, spoken, as we have seen, on Saturday the 26th. And (as I argued before) grant- ing that Dante had full liberty to form or assume ideal dates to any extent, yet we cannot suppose that, having done so, he would bring everything into con- fusion by distinct and specific statements throughout the poem inconsistent with such original conceptions ; statements too quite gratuitous and arbitrary, which might just as easily have been made consistent, or indeed omitted altogether if they served no definite purpose. III. Next as to the hesitating' suggestion of Phila- lethes, that Dante assumes a different sort of ideal Good Friday, viz. April 5th (which was in fact a Tuesday in 1300), as being the day for the cele- bration of the Jewish Passover in the year 1300 in accordance with the date of the actual Full Moon on that clay. He clears the way for this by arguing that neither March 25th, nor the actual date of Good Friday in 1300, viz. April 8th, can be adopted con- sistently with the two crucial passages, to which I have directed special attention. As to March 25th, he says that the language of Inf. xx. 127 finally disposes of that, since gar keine vernilvftige Deutung ztrfasst (' it admits of absolutely no intelligible meaning ') : and similarly it forbids the acceptance of the date April the Divina Commedia. 33 8th, because 'jedoch trifft auch hier die Angabe des Vollmnnds nicht zu' ('the reference to the Full Moon does not agree with this either '). So again he denies that the Saturday on which these words are spoken can be supposed to be either March 26th or April 9th : — ' Nach der Angabe des 9 April hiitte sonach der Dichter sich um einige Tage geirrt ; bei der Annahme des 26 Marz ist aber die Sache noch irriger ' (i.e. on the adoption of April 9th the poet would have made an error of some days, while on the supposition of March 26th the matter is made worse). It is perhaps hardly worth while discussing the improbable suggestion of Philalethes further, since he does not profess to attach much value to it himself. He rather throws it out in despair, as his language just above indicates, of being able to rest in any other solution that has been pro- posed. IV. Finally we come to the view advocated by Lubin, which in itself I make no doubt whatever is the correct one, viz. that the commencement of the Vision is to be taken in the natural and obvious sense of Dante's words as occurring on Good Friday, April 8th, 1300. Lubin however is in constant difficulties, since he holds that Dante refers throughout to the real Full Moon which occurred on March v r ,th ; and there is no course open to him but candidly to admit that the poet has fallen into an error. He says of this D 34 Time- References in undoubtedly correct astronomical calculation, ' Cio non significa altro se non che Dante nella Corn- media siasi ingannato, ponendolo (il Plenilunio) due giorni dopo ' ; and adds ' Come cio avvenisse non e facile dire V Now it appears to me very singular that Lubin should not have seen the simple solution to be that Dante followed the Calendar and not the Ileal Moon. This is the more remarkable, because he does contend on the very same page [Studi, p. 363) that Dante ' che segwb le credenze popidari ' would naturally follow the Calendar of the Church in respect of the date of the Equinox, and also of the observance of Easter. (This is of course in refutation of the dates March 14th and 25th, to which we have already referred.) Now surely the same reasons precisely would lead us to infer, that he ' followed the Calendar of the Church,' and not independent astronomical calculations, also for the date of the Paschal Full Moon. In all the passages however where the question of the Moon's position is involved, Lubin gives that of the ' real ' Moon, and explains the hours intended by Dante as calculated on that hypothesis. Consequently, though his days are, as I believe, all right, his hours are, I venture to think, all wrong by nearly two hours, so far as they depend on indications of the Moons position. It is now high time to state distinctly and maintain 1 Studi, p. 362, note. the Divina Commcdia. 35 emphatically the central principle for which I am contending, the application of which I venture to think dispels at once (if it is not too bold to say so) these clouds of doubt and difficulty as to Dante's language and meaning in his various allusions to time. I could scarcely enunciate it better than by a slight adaptation of the language of Lubin just quoted. ' It is natural to suppose that ' (when re- ferring to the Moon) ' Dante should have followed the Calendar of the Church.' In other words, all his allusions are, I believe, to be connectedly and con- sistently explained as referring to the Calendar and not the Real Moon. Let me ask attention to these considerations. (1) It should be remembered that Dante is not describing the scene as an eye-witness at the time, but is relating it some years afterwards, and more- over he is not describing an actual scene at all, but a purely fictitious and imaginary one, to which how- ever he artistically imparts definiteness and reality by fixing very accurately the surrounding circum- stances of time and place in which it is supposed to occur. Surely then his obvious course would be to make his references square with the computed position of the Paschal Moon as taken from the Calendar, since this would be the only available source of informa- tion for any ordinary reader, who might wish to follow him closely, and realise for himself, by inter- preting the data so carefully and pointedly given to D 2 36 Time- References in him, the actual surroundings of each scene depicted. Consequently, if Dante, when writing, wished to indi- cate the hour of Moonrise, say, on April nth, he would adapt his reference to the supposition that the Moon was four days past the Full, as any one would find it recorded in the Calendar for the year. Remember too he is referring not to an ordinary Full Moon, but to the Paschal Moon of the year, information as to which would be universally and easily accessible ; the ' terme pascliaV being almost as conspicuous a landmark in the Calendar of any year as the date of Easter itself. (2) Moreover let us ask ourselves, why is Dante so careful to insert these various and frequent references to time, not less I think than forty in the first two Canttche % We may be sure that they are not mere poetic adornments, mere fixtures, so to speak, of Dante's poetic furniture. All know doubtless how one of the most characteristic and distinguishing features of Dante's poetry is its extraordinary mi- nuteness of detail in local description. Macaulay, in his Essay on Milton, has compared Dante's descrip- tions to the reminiscences of a traveller. Ruskin says they often resemble the notes of a land-surveyor. I believe that for vividness of effect he wished his readers not only to follow him step by step in the scenes which he depicted, but also hour by hour. Consequently, to have adopted any other than a popular computation of familiar celestial phenomena the Divina Covimcdia. 37 would not only be poetically superfluous, but posi- tively misleading 1 . (3) There could scarcely be a better proof of the inutility of adopting any other than the popular method of referring to astronomical phenomena, than the fact that it does not seem to have occurred to any one before Giambullari, i. e. for nearly two and a half centuries after Dante wrote, to go into the astro- nomical calculations. One is reminded of the familiar argument that the Bible uses the language not of science but of popular usage, because had it done otherwise it would have been unintelligible to those to whom it was addressed, and many generations must have passed before its true meaning could be ascertained. It seems that a similar fate would have befallen the work of Dante, had he corrected his astronomical references by independent scientific cal- culations. (4) The strongest argument however would be if this hypothesis alone should give a consistent ex- 1 A curious illustration of the difference between the ' Real ' and ' Calendar ' Moon, and of the necessity for any one writing for ordinary people, to keep clear of refinements of this kind, may be drawn from the phenomena of the year 1301, whicli some would assign as the year of the Vision. The Calendar Full Moon in that year was on Monday March 27th. Consequently Easter was kept on the following Sunday, viz. April 2nd. But the Ileal Full Moon was three days earlier, viz. on Friday March 24th (auct. Grion, Lubin, &c). Hence we have the curious result, that if regard had been paid to the Real Moon, Easter Sunday ought to ha\ e been kept on March 26th, and not, as it actually was kept, on April 2nd. 38 Time- References in planation of the various time-references in the poem, and this I must now endeavour to show. At any rate it is pretty generally admitted that other hypotheses do not do so, even by those who advocate them. On the assumption then that this would be a natural supposition, I have constructed for myself a specimen of the sort of working Calendar by which Dante is likely to have guided himself, so that his time-references might be approximately correct, and also (what would be quite as important) popularly intelligible to his readers. As the Calendar Full Moon fell on April 7th, and further, as we learn from Inf. xx. 127, during the night between April 7th and April 8th, when Dante would (poetically speaking) have so observed it during the night spent by him in the Selva oscura, we should not be far wrong in supposing that it set on the following morning (i.e. Good Friday), about Sunrise, or within at any rate ten or fifteen minutes of the time of Sunrise. We should then be able to calculate, in such a rough and popular way as would be sufficient for Dante's poetical purposes, its rising and setting for the next few days, by allowing a retardation be- hind the sun of twenty-five minutes for each twelve hours or fifty minutes for each complete day 1 . Further, 1 It is interesting to see Vellutello working out this kind of calcula- tion for himself, allowing a daily retardation of about an hour, in his note on Purg. xviii. 76. Moreover, since writing this, I have found a passage in which Dante's own Master, Brunetto Latini, gives precisely this very rule for ascertaining practically the Moon's position on any the Divina Commedia. 39 we should not be many minutes wrong if at a distance of about twenty days from the Calendar Equinox we assumed Sunrise to be about 5.15 a.m. and Sunset about 6. 45 p. m. We could thus calculate roughly the impression that would be likely to be conveyed to an ordinary contemporary reader by a reference to Moonrise or Moon-setting during any of the days mentioned in the poem ; and as the poet is surely likely to have used terms with a view not to the minute calculations of astronomers, but such as would be ' understanded of the people,' it seems to be most natural to suppose that he adopted some such rough and ready calculation as is here suggested. To go further than this would be (as Metastasio says of the too rigid application of rules like 'the Dramatic Unities ') ' confondere il vero col verisimile.' In fact Dante might well have reasoned with himself as to such scientific calculations in the language of ' il Filosofo,' whom he so often quotes : — Ae'yerai -nepl avrijs koX ev rots efcorepiKoi? Aoyoi? apKovvrcos evia, nal xpt]v irpoKzip\£vo}V. Here then is the sort of working Table he might have followed: — given day. See Tes. ii. 49, init. : ' E poi che 1' uomo sa cio (viz. the sun's position) e' pub leggermente sapere ov' e la luna, che ella si dilunga ciascun di dal Sole tredici gradi, poco vi falla.' Since 13 of space are equivalent to 52 minutes of time, this is just the way of finding the moon's position that we have been describing. 4v epyoiv 7rAetco ylyvr\rai. Again, though I am not (happily) very familiar with the aspect of the heavens ' about 3 a. m.,' I do not imagine that so early as that, 'the dawn is just beginning (as Mr. Butler says) to whiten in the East,' in the first half of the month of April ; nor is there, I think, any special propriety in saying (as he suggests) that Scorpio, which ' is just on the meridian ' at that hour, may be said to be on the forehead of the dawn. By pass/, in 1. 9, Mr. Butler understands 'signs,' but he candidly admits that ' there is, on any explanation, some confusion in lines 7-9.' 1 ought perhaps to say a few words as to the the Divina Commedia. 89 theory that the Solar Aurora is referred to, though I regard that idea as so utterly unsuited to the passage itself and all its surroundings as scarcely to deserve serious refutation. It will be found discussed and sufficiently refuted in Scad. p. 154. I will only again note, among many other objec- tions, that in this case the term concubina, as ap- plied to Aurora the wife of Tithonus, becomes both meaningless and offensive. (See note above on p. 84,) Still less worthy of consideration perhaps is the curious modification of this view, that Dante refers to the Solar Aurora in Italy \ in contrast with the nocturnal phenomena of lines 7-9 in Purgatory, i.e. ! nel luogo ov' eravamo.' It is ingeniously but fancifully argued that if it was about i\ hours after nightfall in Purgatory, it would be i\ hours after sunrise in Jerusalem, and consequently — Italy being, as has been already pointed out, according to Dante's geography about 45 W. Longitude, which would be equivalent to a retardation of 3 hours of time — about half-an-hour before Sunrise in Italy. No doubt Purg. iii. 25 and xv. 6, which we have already discussed, might be quoted as parallel cases of the contrast between the hours in Purgatory and in Italy, but it should be noticed that Dante is very careful there to indicate the contrast. Here it would surely be preposterous to suppose that all this brilliant description refers to an absent and invisible 90 Time- References in phenomenon. Further, what meaning are we to attach to lines 4-6 on this supposition ? Though I cannot here attempt an adequate ex- amination of other views that have been held, I must not pass over without notice the extraordinary in- terpretation which has been suggested in recent years, first I believe by Prof. Antonelli, and ex- pounded in his Studi Speciali (Florence 1871) and also in his Paper ' Sidle dottrine astronomiche della Divina Commedia? contained in the collection entitled ' Dante e Suo Secolo? This is adopted, and is defended, keenly and vigorously (as usual), by Scartazzini. It is this : — First of all, he reads Titan (i. e. the Sun) in- stead of Titone (or Tithonus) in 1. 1 *, adopting a variant which is found in the great Vatican MS., denoted as ' B ' by Witte, but has not (as far as I 1 In regard to the irregular form Titone, instead of Titono, from Tithonus, it may be remarked that old Italian abounds with every kind of anomalous interchange of the forms appropriate to different noun-declensions in Latin. A great part of Nannucci's elaborate Teorica del Nomi is taken up with examples of such irregularities in every declension, 'joer uniformita di cadenza? I select, out of many instances like the one before us, the following : — Radamante {Boccaccio) ; Berlinghiere ( = Berengarius), Sonnolente, Turbolente (all three in Pulci) ; Nile and Menale, in the Ditfamondo. Also in Dante himself, frodolente (Inf. xxv. 29) and pome (Par. xxvii. 45 and 115) Nannucci quotes one authority for Titono. Per contra, we very frequently meet with such forms as Tritono, Clemento, TStiopo, (e.g. Purg. xxvi. 21), Apollino, Cesar o, Atlanto, &c. The alleged irregu- larity then of the form Titone is no reason for preferring the more regularly-formed Titan. These anomalies would be facilitated by such truncated forms as Titon, padron, Nin (Purg. viii. 53), Nil (Par. vi. 66), &c. the Divina Commcdia. 91 know) much other authority. The main outlines of the interpretation are then as follows:— (1) that neither Solar nor Lunar Aurora are referred to, but that ' La concubina di Titan ' represents Tethys, the wife of Oceanus, and is in fact an equivalent to onda mar inn. The following lines are therefore tanta- mount to saying that the ocean waves towards the East were illuminated by light, probably from the rising Moon, but at any rate from some source other than the Sun (for thus Scartazzini strangely interprets 1. 3). (2) He thinks that the f red do animate is not the Scorpion, or indeed any other Constellation whatever, but only certain stars, among which might be some of those of Scorpio, ' disposte in forma di serpe,' the serpent being well known as ' frigidus anguis.' (3) He interprets passi in 1. 7 of the hours of the night, as we have done above, and consequently his final conclusion is precisely that which we have come to, viz. that the hour was towards 9 p. m. c McTaftaCvoov 8?) 6 Aoyos els tolvtov cufunTai.' Now I must say that this interpretation involves such a congeries of improbabilities or difficulties in every single line, that I think for once Scartazzini's usual judgment and common-sense seem to have strangely deserted him. I take his interpretation line by line. (1) In /. i, Scartazzini (as we have seen) censures Dante for describing the Lunar Aurora as the mistress 92 Time- References in of Tithonus, as thereby falsifying (his word is falsifi care, p. 152) Mythology. But what are we to say about his own theory that Dante here instals the Ocean Wave as the mistress of Titan or the Sun 1 Where is there any trace of this notion in Ancient Mythology 1 He quotes, it is true, a number of classical passages from which such an unpleasant inference might con- ceivably have been drawn. But there is not the least evidence that it ever was drawn, and it involves quite as much of an adaptation or falsification (which- ever you will) of traditional mythology as that of describing the Lunar Aurora as the mistress of Tithonus. In short all the rhetorical and prudish nonsense on p. 152 as to this latter idea involving 'lordura,' and its being 'sozza pittura ' from which we are compelled ' svolgere con nausea e con ribrezzo gli occhi' (!), all this I say applies with at least equal force to his own contribution to mythology as just expounded. In- deed I should maintain that these expressions apply with much greater force against his invention that Tethys is the mistress of the Sun, since she is un- doubtedly the lawful wife of Oceanus, whereas Dante's ' Concubina di Titone antico ' i. e. the Lunar Aurora, has at any rate no other known attachment. (2) Next in I. 2, the word balco (i. e. gallery or bal- cony) implies some elevation, and clearly indicates some phenomenon in the ski/, not on the ' suol marino,' as Dante calls it. It loses its significance then when the Divina Commedia. 93 applied to light on the waves, though it is appro- priate to the light of dawn on and above the horizon. Scarf, paraphrases lalco by lemho, which is clearly in- adequate 1 . (3) In I. 3, Scarf, takes fnor delle braccia apparently to mean that the wave, i. e. Tethys or the concubina of the Sun, was illuminated otherwise than by the Sun her ' dolce amico.' This is surely an impossible, or ex- tremely forced meaning for the words. Moreover, considerations of language apart, it can scarcely be doubted th&tfuor delle braccia represents the idea of efc \(X* Ti6ep.evov ovtco yap av . . . cocnrep nap avrols yiyvo\Atvos rols irpaT- ro'/xerot?, evpicTKOL to Trpiirov /cat iJKLcrTa av XavOavoi ra vTtevavria. [Poet. xvii. init\ The passages here discussed have been taken, as will no doubt have been observed, entirely from the Inferno and Purgatorio. It is hardly necessary to add that Dante gives us no such marks of time in the Paradiso 1 , since there he has passed from time to eternity, — All' eterno dal tempo era venuto. {Par. xxxi. 38.) Also there they have ' no need of the Sun, neither of the Moon, to shine in it,' for ' there is no night there.' 1 See Supplementary Note vi, p. 1 26. SUPPLEMENTAEY NOTES. [The following notes are printed here, as some of them seemed too long to appear as foot-notes on the pages to 'which they severally refer, and others relate to works that have only been met with since this Essay was written.] I. On the selection of 'the year 1300 for the Vision. [As an additional note on p. S.] Dean Plumptke in the valuable Introduction to his Translation recently published (pp. lxv-lxvii) makes an interesting, and I think very plausible, suggestion that the assumed date of the Vision had a very special significance for Dante himself as the turning-point in his own life (see especially Purg. xxx. 130-8). "We know from his own confessions in the Convito and elsewhere, how, after the death of Beatrice, he ' forsook his first faith/ and sought for consolation from the ' broken cisterns ' of heathen philosophy, and from le presenti cose Col falso lor piacer. {Purg. xxxi. 34.) Moreover, while asserting as a fact in his own history his recovery from these errors, and his return to Faith and to his ' first love,' he also distinctly connects that recovery, both in time and cause, with the contemplation of the scenes and subjects of this Vision. Dean Plumptre suggests that Dante may have been actually at Rome, and perhaps for the first time in his life, at the date assigned to his Supplementary Notes. 1 1 5 Vision, viz. Easter-tide of the year 1300 ; and that this was in fact the turning-point in his life from speculative, and more or less sceptical philosophy to Revealed Truth, as symbolized by Beatrice. If so, the assumed date was not an arbitrary poetic fiction, but corresponded to a fact of the deepest interest in his own history, 'the conversion crisis' of his life (p. Ixxvii). This would satisfactorily account for the pointed and repeated references by which the date assigned to the Vision is emphasized. It is fixed, as the Dean says, ' with a precision the only natural explanation of which is that it represents a fact.' At any rate this suggestion is a highly interesting one, though it cannot be regarded as more than an ingenious conjecture. It must not be forgotten however that in Inf. xviii. 28-33 we have very strong evidence that Dante was at Home sometime in the year 1300, and the esercito molto would naturally be at its height at the solemn season of Holy Week and Easter. II. On the assumed date J 300 never forgotten by Dante. [As a note on pp. 8, 136.] The following are, as far as I can recollect, the chief passages bearing on this. I will mention first those which are sometimes thought to be exceptions, or oversights, on the part of the poet. I. Inf. xviii. 28. Dante here refers to an incident occurring at Rome during the progress of the Jubilee in 1300. This was proclaimed by Boniface on Feb. 22 in that year. But (1) Dante may have been there in March or at Easter itself. (See last note.) Certain it is that early in 1300 two embassies went to Boniface from Florence, and as Dante took part in such an embassy shortly after, he may have accompanied either of these. (2) Whether I 2 1 1 6 Supplementary Notes. he was there or not, the incident he mentions may have occurred at or before Easter in that year. (3) Whether it did so occur or not, the comparison is introduced here not as having been spoken of by or to himself during his pilgrimage, but as an illustration given by the Poet when narrating that Vision afterwards. (I mention this at some length because Grion uses this as an argument for the assumed date of the Vision being 1301.) II. Precisely the same principle would apply in reference to the event referred to in Inf. xix. 19. Grion (p. 10) states that this occurred, according to Jacopo di Dante, on April 1st, 1 30 1. Again I would say, this is a statement evidently made by the Poet while writing his narrative, and it does not profess to have been referred to by himself or any of the characters introduced by him during the Vision itself. I might add that this would not help Grion or any one else to maintain 1301 as against 1300, since the expression ancor non e molt' anni is equally (from that point of view) inapplicable to either date. III. The ' ruina ' known as ' Slavino di Marco,' near Biva, which is supposed with much probability to be re- ferred to in Inf. xii. 5, &c, is stated to have occurred on June 20th, 1309. But granting both this date and the explanation of the allusion to be beyond doubt, this also is only referred to by the Poet as writing afterwards by way of an illustration, and we certainly need not suppose him to have been so pedantic as to refuse an apt comparison from a natural phenomenon then familiar to his readers, because it was non-existent at the assumed date of his poetic pilgrimage. IV. Another case deserving notice occurs in Purg. viii. 74, viz. the second marriage of Beatrice, the widow of Supplementary Notes. 117 Nino Visconti with Galeozzo Visconti, which was celebrated ' with extraordinary pomp ' (according to the chronicles) on June 24th, 13.00. If so (as Scartazzini observes) she was no doubt betrothed to him for some time previously, and had ' put off her widow's weeds,' which is all that Dante commits himself to, before the assumed date of the Vision. She had become a widow, it may be added, about four years previously, in 1296. V. Purg. xiii. 152 is supposed to refer to the purchase of Talamone by the Sienese, on Sept. 10th, 1303, for 8000 golden florins, during the time (as it is stated) when Dante was himself at Siena (see Aquarone, Dante in Siena, p. 70). No doubt if he were then present it would account for the impression which the incident made upon his mind, and it may have suggested an allusion to it. But his statement is far from being so definite as is assumed. Dante only says that the hearts and hopes of the foolish Sienese were set upon the place, and this may well have been the case for three or four years before they succeeded in securing it. Per contra, we have some remarkable instances of events occurring very soon after the assumed date which are spoken of as still future, and referred to under the guise of prophecy. The following are a few cases which occur to me : — Inf. vi. 64. The bloodshed of May 1st, 1300, at Florence (probably). (See Giov. Villani, Ghron. VIII. c. 38.) Inf. x. in. The death of Dante's friend Guido Cavalcanti in the winter of 1 300-1. Purg. xiv. n 8-9. The death of Mainardo Pagano, ( il Demonio,' Aug. 16, 1302. Purg. xviii. 12 /. The death of Alberto della Scala Sept. 10th, 1 30 1. 1 1 8 Stcpple77tentary Notes. III. On the date of the Vernal Equinox, and traditions of the Creation. [As a note on pp. 14-16.] Brunetto Latini seems to have placed the Vernal Equinox on March 18th and the Summer and Winter Solstices 011 June 17th {all' XV dl all' uscita del mese di giugno, Tes. ii. 43, part 2) and December 17th respectively, but, as far as I can see, gives no reason for this position of the Equinox beyond that 'dicono molti savi die 'Ifu XI I II dl all' uscita, del mese di Marzo' (Tes. I. c. 6), or, as it stands in the original, 'XIIII jors a Tissue dou mois de Mars! There are, as far as I know, two other passages besides that just cited when a similar statement occurs : — Et sappiate die 'I primo dl del secolo entro el Sole ne lo primo segno, cioe in Ariete. Et cid fit, XIIII dl all' uscita di Marzo, et altresi fa egli ancora (II. c. 42). And again in II. c. 43, part 2, the same statement is made in nearly identical words, and in the same chapter the Autumnal Equinox is similarly put a XV dl all' uscita di settembre, i.e. Sept. 18th. The words et altresl fa egli ancora seem to make it doubtful whether Brunetto was aware of the error in the Calendar or of the Precession of the Equinoxes, as they imply that no change had occurred in this date since the Creation. There is another passage, Tes. II. c. 48, where Brunetto says that all the heavenly bodies were created on March 21st ('AT" dl all' uscita di Marzo '), that being the fourth day of the week of Creation, which he supposed to have commenced, as we have seen, on March 18th. Hence (he proceeds) many say that the Equinox is on that day. When therefore he himself puts it on March 18th, it is probably not on astronomical grounds, but because, regarding that day as ' il primo dl del secolo,' he thought it appropriate that it should fall ' in Supplementaiy Notes. 1 1 9 quel buono e dritto punto,' when day and night were equal. This view is however much more ancient than Brunetto Latini. It is found in Bede, who says that the heavenly bodies were all created 'quarta die . . . quae nunc, quantum aequinoctii conjectura colligimus, XII Kal. Apr. vocatur.' Hence the 'prima saeculi dies' would be March 1 8th. So also in the curious ioth Century Anglo-Saxon Manual of Astronomy translated and edited by Wright (1841), p. 4, we read, 'The first day of this world we may find by the day of the Vernal Equinox, because the day of the Equinox is the fourth day of the Creation of this world We will now say briefly that the first day of this world is reckoned on the day which we call XV Kal. Apr.' i.e. March 18th. (He quotes Bede as his authority for this.) He says later that some have placed the Equinox on March 25th, but that it is certainly ' on S. Benedict's day,' i.e. March 21st. So also Hippolytus, a disciple of S. Irenaeus, and apparently the first author of a Paschal Cycle, placed the Equinox on March 18th (Hefele, Councils, p. 318). It is worth while adding a word as to the curious way of computing the day of the month by counting backwards, which is found in the above passages and elsewhere in the Tesoro, and in some other old writers. The month was divided in a way curiously resembling the Greek ^71/69 iaranivov and nrjvos fyBLvovTos (though there seems to have been no third division corresponding to the [xt]i>6s /xeaovvros), and, as with the Greeks generally, those of the former division were counted forwards and those of the latter backwards : e.g. Brunetto describes S. Barnabas' day as XI dl alT entrata di giugno (II. c. 20), and S. Thomas's day as XI dl alV uscita di decembre (II. c. 13). It is curious also to note that in the passage above cited from Tes. I. 6, the old editions, e.g. Ven. 1533, read ' che 'I fa XI III dl del 120 Stcpp lenient ary Notes. mese di Marzo! This is a mere Editor's correction, made no doubt by some one who knew that March 14th more nearly corresponded with the actual fact. It is however unsupported by the MSS. and clearly inconsistent with the author's repeated statements in B. II, as quoted above. As to the common belief that the Sun was created at the Vernal Equinox, which is of course implied in Inferno i. 38-40, Dante himself gives us a curious a priori reason in the very obscure passage, Par. i. 37, &c. and also in Conv. II. 4 (1. 50, &c), Avhich passages are both noticed by Dionisi, Anecld. IV. p. 51, &c. The same writer (p. 66) also quotes a curious fragment from the Acts of an early Council in Palaestine a. d. 196, which professed to de- termine ' quo modo in principio /actus fuerit mundus, id est die Domenico(I), Verno Tempore, in Equinoctio quod est Octavo Kalendarum Aprilium (i.e. March 25th), Luna Plena.' So that thus the Creation, Incarnation, and Cru- cifixion would all occur on March 25th. By Creation, when thus spoken of as the act of a single day, no doubt the Creation of man is generally intended, though I do not think those who indulged in these a priori specu- lations were always careful to attach a precise meaning to their words. It should be remembered also that at the time of the very ancient Council just cited the Equinox was held, according to the chronological arrangements of Julius Caesar, to fall on March 25th. This explains the date of 1 Octavo Kal. Ap.' in the above quotation. After the Equi- nox had been fixed (as some say by the Council of Nicaea) on March 21st, ecclesiastical ingenuity was equal to the occasion in discerning other a priori reasons for the fitness of things on this hypothesis. The first day of Creation was then found to fall most appropriately on March 20th, when it was pointed out God having created light, ' divided the Supplementary Notes. 121 light from the darkness,' i.e. made the light and darkness equal, which clearly points to the Equinox immediately following ! IV. Note on Mazzoni's argument in his 'Difesa di Dante.' [In reference to p. 24.] Mazzoni's discussion of Inf. xx. 127 in conjunction with Purg. ix. 1, &c. in his Difesa di Dante, I. c. 76, is very curious. He seems to regard it as axiomatic in regard to Purg. ix. 1, &c. (i)That the Lunar Aurora is referred to; (2) that lines 7, 8 describe 2-| hours of night or a little later : and (3) as regards Purg. xviii. 76, that the hour of moonrise is referred to. He quotes two passages of Pliny respecting the retardation in the rising of the Moon, and on the strength of these, constructs two elaborate Tables for finding the time of moonrise after sunset from Full to New. Next, describing the night referred to in Purg. ix. as the third, and that in xviii. 76 as the fourth day of the Moon's age, he finds Dante prima facie quite wrong in both cases, since in the former he implies a retardation of more than 2\ hours instead of ijV hours, as indicated by Pliny's Tables (p. 306) : and in the latter he implies at least 4 hours, whereas the correct time would be 2^\ hours (ib.). Mazzoni proceeds to defend him by explaining that Full Moon was on Monday, April 4th, in 1300. On Tuesday evening Dante entered the Inferno, and on the night of the 6th escaped from the perils of the Inferno 'per dar principio ad un altro viaggio.' (Here follows a very curious passage to prove that this day was specially chosen because on that day, viz. April 6th, were done by the Ancients many valorous actions, according to Aelian, whom he quotes to show that on that day were fought the Battles of Marathon, 122 Supplementary Notes. Plataea, and Mycale : also Darius was defeated by Alexander the Great; and moreover it was the birthday of Socrates! Hence Mazzoni thinks the day was significantly chosen by Dante for this part of his journey : and that in this he was wiser and more pious than Petrarch, who chose the same day for the commencement of his love : .) He thinks it reasonable that a clay was spent in the ascent from the centre of the Earth to its surface at the Mountain of Purgatory, since the same amount of time was spent in the descent of the Inferno. This he justifies in a quaint matter-of-fact manner thus : ' compensando la malagevo- lezza della salita, colla tardanza che s' era fatta nella scesa per ragionare con molte anime ! ' He is able thus to argue that the passage in Purg. ix. 41, &c. may relate to the fourth night there (though only the third night in our hemisphere), when, according to Pliny's computation, the moon would rise at 2f|- hours of night. He rather lamely continues that as on the next night, according to Pliny's rule, the moon would rise at 3f| hours of night, that might be described as about 4 hours, and therefore ' quasi a mezza notte (?).' He seems to feel, however, that bis laboured ' difesa ' of Dante has not been entirely successful, as he adds that Dante probably took account of the ' velocita del moto, ch' ella aveva in quel tempo, per partirsi dalla oppositione . . . seconda la quale non ci ha Plinio lasciata regola alcuna ! ' (P- 3°9-) V. On the position of the Earthly Paradise and Purgatory. [Note on p. 61.] In reference to the Earthly Paradise we may note that the curious notion of the Euphrates and Tigris having a 1 Mazzoni doubtless refers to the passages already cited, sup. p. 20. Supplementary Notes. 1 2 3 » common source (see Purg. xxxiii. 112) is frequently found in other writers, including at least one profane, as well as many Christian authors. In the case of the latter, and others acquainted with the Vulgate, it is no doubt derived from Gen. ii. 10 and 14, where a ' fluvius' rising in Paradise parts into four streams, two of which are Tigris (in Engl. Version ' Hiddekel ') and Euphrates. It is not so easy to account for its appearance in such an author as Sallust (Hist. Frag, cited by Isid. Orig. xiii. c. 21 § 10): ' Sal- lustius, auctor certissimus, ita asserit Tigrim et Euphratem uno fonte rnanare in Armenia.' Boethius (aptly quoted by Scart.) in Be Cons. Lib. v. Metr. 1, writes : Tigris et Euphrates uno se fonte resolvunt Et mox abjunctis dissociantur aquis. Dante may have derived the idea either from Boethius or Isidore or perhaps from his own master Brunetto Latini who writes thus in Tes. iii. c. 2 : ' Salustio dice che Tigris et Euphrates che passono per Armenia escono d' una mede- sima fontana.' Brunetto also says of the Euphrates ' corre per Armenia et movesi dal paradiso terreno.' S. Thomas Aquinas notes this geographical difficulty in Genesis, and states it as one of the arguments against Paradise being a ' locus corporeus.' He replies to it by adopting the solution of S. Augustine, who had previously observed the difficulty, and deals with it thus : — ' Ea flumina quorum fontes noti esse dicuntur alicubi esse sub terras, et post tractus prolixarum regionum locis aliis erupisse, ubi tan- quam in suis fontibus nota esse perhibent ' (De Gen. ad lit. Lib. viii. pp. 612-3, Ed. Basil. 1556). He proceeds further to identify Geon or Gihon with the Nile, and Phison with the Ganges. This identification of the four rivers however is not first found in S. Augustine. It occurs for instance in 124 Supplementary Notes. Josephus, Ant. Jud.I. iii. § 3 ; and many other writers are cited as adopting it by Corn, a Lap. in his Commentary on Genesis h. I. It is commonly repeated in later times, as by Isidore (Orig. xiii. 21); Sir John Maundeville (with much quaintness, see pp. 304-5, Ed. 1886) ; B. Latini (at any rate as regards the Nile, Tes. iii. c. 2) ; and it is also found in the Mediaeval Mappae Mundi, 2>^ssim. Eabanus Maurus, In Gen. Lib. I. c. xii. reproduces the explanation and almost the words of S. Augustine. So also does Gervase of Tilbury, Ot. Im]). i. 11, &c, &c. As regards the situation of Purgatory the various opinions, as well as the general consensus, on this subject may be gathered from the exhaustive discussion by Bellarmine, Be Purgatorio, Lib. II. cap. vi. headed l De loco Purgatorii! He there collects and criticises all the views known to him as having been held on this subject, to the number of eight, but there is no trace in them of any notion that Purgatory was in the Southern Hemisphere, or indeed on the earth's surface at all. He then gives the following opinion as ' communis Scholasticorum, Purgatorium esse intra viscera terrae, inferno ipsi vicinum. Constituunt enim scholastici communi consensu intra terrain quattuor sinus, sive unum in quattuor partes divisum, unum pro damnatis, alteram pro purgandis, tertium pro infantibus sine Baptismo obe- untibus, quartum pro justis qui moriebantur ante Christi passionem, qui nunc vacuus remanet.' (This last statement may be illustrated by Inf. iv. 52-63 and xii. 38-9.) So again the various legends collected by Wright in his 'Pur- gatory of St. Patrick ' generally agree in supposing it to be underground, various opinions prevailing as to the spot in the Earth's surface from which it might be approached. Sometimes, though rarely, it was believed to be in the air. Though it scarcely bears upon this, I cannot refrain from Supplementary Notes. 125 reproducing a quotation given by Wright from a pious Italian writer of the 17th century, as showing how very seriously literal was the belief in the topography of these nether worlds. After stating that it was most certain and beyond all doubt that Hell was situated in the centre of the earth, this writer gravely answers the objection that the space would be insufficient for the constantly increasing multitude of the lost, by stating that the souls of the damned could not expect to be allowed so much room as the blessed spirits in Paradise ! As Bartoli says, ' L' Inferno diventa nel concetto medievale un capitolo di geografia.' He mentions that the island of S. Brandan's vision was marked in maps, that it was once formally ceded by Portugal to Castile, and that even in 1721 a Spanish expedition was fitted out to discover and explore it ! It may be worth while to draw attention here to another point for which (as far as I know) Dante had no patristic or scholastic authority, viz. the introduction of a frozen region in Hell. The idea however is found in some of the current mediaeval visions which preceded Dante; e.g. in the vision of Tundalus (an Irish monk), which is placed in a.d. 1149, it occurs in a connexion something like that in which it is found in Dante, viz. he represents that a winged monster (not however described as Lucifer) is seated in ice devouring sinners who are also embedded in it. ' Sedebat autem haec bestia super stagnum glacie condensum et devorabat animas' {Be Raptu Animae Tundali, c. vii). In this case this punish- ment is reserved for licentious Monks and Ecclesiastics. So again in the very singular 1 ith century Visione di San Paolo (according to Tommaseo, of Anglo-Norman origin), ' gli dimoni si ardevano la metade, e 1' altra metade affreddavano.' (Scripture of course only describes S. Paul as having had a vision of Paradise — see 2 Cor. xii. 1-4 — but mediaeval 126 Supplementary Notes. fancy extended this to the Inferno also, and it is interesting to note that Dante appears distinctly to adopt this legend in Inf. ii. 28, &c). In the still more celebrated Visione di Frate Alberico (to which some have maintained that Dante was largely indebted) the lascivious are imbedded in ice to various depths in proportion to their guilt, like Dante's tyrants in the river of blood (Inf. xii. 121, &c). Cold is described as one of the torments of Purgatory (though not apparently of Hell) in the ' Visio Drycthelmi ' narrated by Bede (Hist. Eccl. Lib. V. c. xii), and it appears also in the Purgatory of S. Patrick. VI. On alleged notes of Time in the Paradise. The passage in Par. xxii. 151, &c. is sometimes quoted as a proof that 16 hours had elapsed to this point, i.e. starting from the very doubtful assumption that Paradise is entered at noon of Purgatory (or midnight of Jerusalem) on the Wednesday. (See on this point the notes on pp. 10 and 54.) If this be so, the indication is a very obscure and inferential one, and quite unlike those that we have met with in the other two Cantiche. I do not feel at all clear that it is so intended, and regard it as a purely ideal astronomical description, and so far as it has a literal interpretation, it is local rather than temporal. In any case the temporal details apply (if at all) only to this earth (aiuola), and not to the place where Dante was. I said the whole situation is a purely ideal one, because as soon as we attempt to draw definite inferences here (or in the somewhat similar passage in xxvii. 79, &c.) as we have done in regard to the precise indications given in the Inferno and Purgatorio, we begin to discover that we are in a sphere to which the ordinary con- Supplementary Notes. 127 ditions of time and space no longer apply: e.g. lines 151-3 imply that the whole inhabited hemisphere of the world was visible. Now remembering Dante's geographical theories already expounded, and having regard to the clear state- ments of Par. xxvii. 82-87, we observe (1) that such a vision as xxii. 15 1-3 would imply the condition that both the spectator himself and; the Sun were on the meridian of Jeru- salem : and (2) that such a condition was now impossible, since the spectator was in Gemini, and the Sun well advanced in Aries, ' un segno e piu partito ' (xxvii. 87). Hence as Delia Valle (who however struggles manfully with the difficulties of these two passages) says in reference to xxii. 150, etc., 'II luogo dove il poeta si trova e al tutto arbi- trario' (p. 119): and again (p. 120) he states that the position, of Gemini is ' non per legge astronomica ma solo per arbitrio e finzione del poeta, al pari di pianeti' (see 1. 144, &c). We might indeed justify this in Dante's own words — Che dove Dio senza mezzo governa La legge natural nulla rilieva. v (Par. xxx. 122-3.) I do not consider therefore that the discussion of these passages falls within the scope of our present subject. At the same time I admit that (as I have already said) Dante intends to give us generally to understand that though himself beyond the limits and conditions of time, still the time passing meanwhile on this earth was such that when he returned to it after his ecstatic vision of Paradise, it would be found to be the evening of Thursday, April 1 4th. (See su}). p. 59.) 128 Supplementary Notes. VII. Note on Bella ValWs ' Senso Geografico-Adronomico del luogld delta Divina Commedia.' I tried in vain to procure this work (which is out of print) before writing the above, and only succeeded in meeting with it after I had nearly finished, so that when Delia Valle is cited, it is generally on the authority of Scartazzini. His conclusions agree in a general way with my own, as may be expected from the fact that he states among his fundamental data that the Paschal Full Moon was on April 7th, and does not appear aware of (or at least makes no allusion in the body of his work as far as I can find to) any other view 1 : i.e. the disputed point as to the Calendar and Eeal Moon is not referred to. I find too that to a great extent we go over the same ground, but my work is mainly independent of his, except so far as sometimes Scartazzini's notes convey Delia Valle's results. In discussing Inf. xx. 126 and xxix. 10 (pp. 15 and 21) his conclusions at first are the same as mine as to the hours indicated, but these are corrected later on in a ' Ketractation ' (p. 68), (when he is discussing Purg. xviii. 76), by the addition of about one hour. This is to suit a theory then started, that the Moon was Full shortly after midnight on Wednesday April 6th, which would be, ' secondo il com- puto e la regola della Chiesa' a part of April 7 (p. 66). This theory is due to the supposed necessity of bringing moon-me (for so he interprets the passage) nearer to midnight on the evening referred to in Purg. xviii. 76. I observe too that he calls that evening Tuesday, but as I have argued (see p. 55, etc.), it must surely be taken to 1 The same seems to be the case with Pasquini, who is satisfied with referring to Hart de verifier les Dates, 'opera ... da fidarsene ad occh chiusi' (p. 253). Supplementary Notes. 129 be Monday. The result is that he gains (as compared with my calculation) about two days. In other words, his result is just the same as if for that passage he had adopted the Real instead of the Calendar Moon. In reference to what I have said about the difficulty resulting from this suppo- sition in Purg. ix. 1-9, this does not affect Delia Valle, since he explains that passage of the Solar Aurora of the Northern Hemisphere in 1. 1, in contrast with the evening hour 'nel loco ov' eravamo' in 1. 8. This is of course open to the obvious objection that the glowing and vivid description of verses 1—6 refer to an absent and invisible scene, besides others which I have noticed in the discussion of that passage. It should be added that near the end of the Supplemento to his work Delia Valle for the first time (I believe) refers to the difference between the Ecclesiastical (or Calendar) and Astronomical Moons. He makes the latter, however, fall by his calculations on the tenth of April in that year, and then proceeds to enunciate the principle for which I have contended, ' che Dante seguiva le opinion! correnti di allora, e sopratutto stava colla regola della Chiesa ' (p. 43). How far Commentators generally have been from recognizing this principle we have already seen ; and further how Delia Valle himself practically departs from it by putting a non-natural sense on Inf. xx. 126, and supposing the Full Moon to have been not ' iernotte ' as that term would be ordinarily under- stood, but on the previous night. This necessitates his adoption (inter alia) of the almost demonstrably false inter- pretation of Inf. xxi. 112 as being 10 a.m. instead of 7 a.m. This hour of 10 a.m. he simply assumes without (as far as I can see) a word of justification, and alludes to it more than once as if no doubt existed on the point. K 130 Supplementary Notes. VIII. Note on the Commentary of Talice di Ricaldone. Since these sheets were in the Press I have obtained a sight of the hitherto unpublished Commentary on Dante by Talice di Ricaldone, which has just been printed in a magnificent folio by order of the King of Italy. We have here the notes apparently of some Lectures written or de- livered in 1474 in burgo Liagniaci (i.e. probably Lagnasco, near Saluzzo. See Pref). He scarcely touches on any of the points discussed in this Essay, and passes over in complete silence the chief passages in which difficulties of date or time are involved. In his Note on Inf. i. 1 however he declares distinctly for the year 1300. I note the following brief extracts as having some bearing on points we have discussed : — On Inf. xxi. 112 he refers to the different ways of counting years a.d. in the words: ' Vel describit tempus more Tuscorum qui describunt annos ab incarnatione, et nos a nativitate.' He also declares, like the other early Com- mentators, for the hour of 7 and not 10 a.m., though by a clerical error he writes 5, thinking no doubt of the word cinque in the text. His words are : — ' Deus passus est hora sexta, et erat una hora diei ita quod 5 hora ' (sic). On Purg. ix. 1, &c. he comments thus: — 'La concubina, &c, id est aurora lune. . . . Et intelligunt autores de aurora solis. Sed Dante intelligit hie de aurora lune, et in hoc facit novam Actionem, ita quod aurora solis est uxor Tithoni et aurora lune est arnica Tithoni. . . . Del freddo animate, ab effectu. Illud signum effective est frigidum, et est tristius signum quod sit in caelo. Unde illi qui orti sunt sub scorpione semper faciunt vilia officia ' ! He takes Purg. xviii. 76 apparently as not referring to moonrise, since he comments thus : — ' Et describit tempus Supplementary Notes. 131 dicens quod luna clare lucebat ita quod offuscabat multas stellas splendore suo : et erat quasi media nox.' IX. Note on Vedovati s Esercitazioni Cronologiche, etc. [See p. 5.] This work also has only come into my hands since I began to print. I therefore add a few notes here. Vedovati main- tains that though the Florentine usage was to count years ab Incarnatione, Dante would certainly have counted them, as usually, a Nativitate (see sup. pp. 48, &c. and Vedovati, pp. 15, 23, &c). The reason for this assertion does not seem clear. Still less do I understand the statement made later by Vedovati (p. 27): 'L' anno 1300 ab Incarnatione Verbi Divini, in ultima, corresponde di fatto al 1301 a Nativitate Domini.' (Is not the reverse the case f) In regard to the prophecies of Inf. vi. 68 and x. 79 (referred to in a previous note), Vedovati strongly maintains that ' tre soli ' are three days, viz, Nov. 2, 3, and 4 in 1301, after the entrance of Charles of Valois into Florence ; and he advocates the strange notion that in the latter passage ' cinquanta volte ' means Quarters of the Moon (!), so that the period would be rather less than a year, and he becomes quite jubilant over this undoubtedly original suggestion (see p. 21). Further he states (I know not on what imaginable grounds) that the Paschal Full Moon of 1300 was on Palm Sunday, April 3 1 ! (p. 27), and he adds: 'E percib la Luna del Giovedi Santo di sera non poteva esser tonda, come piii volte afferma Dante.' By the way, he prudently omits to show how these statements of Dante can be made to square with his own date of 1301, which they certainly do not. (See the Calendar of 1301.) 1 So also, as we have seen, Ponta {sup. p. 27). K 2 132 Supplementary Notes. Though the point does not bear on our subject, I may perhaps mention another paradoxical theory of Vedovati, viz. that Benedict XI is the Veltro of Inf. i. 101. As Benedict died in 1304, probably before the passage was written, and certainly long before it was 'published,' it is inconceivable that Dante should have given a permanent and very prominent position to a prophecy which had so obviously failed of its accomplishment. X. On the unity and symmetry of the plan of the Purgatorio. [As a note in explanation of Table No. VII.] Lest any surprise should be felt at the minutely planned and connected scheme of the whole Purgatorio, implied by the series of time-references which I have traced out, it may be worth while to draw attention to the still more striking proof that through the whole Cantica ' one unceasing purpose runs,' which is afforded by the annexed Table (No. VI). Note here the following points of similarity throughout the description of the Seven Cornici : — 1. At the commencement of each, the Poets are greeted with examples of the Virtue which is opposed to the Vice being expiated in that Cornice, that Vice being in each case one of the Seven Deadly Sins recognized by the Church. 2. Towards the end of each, there are similar examples of each Vice held up to odium. 3. In each case the examples are taken alternately from Sacred and Profane history, except in the fifth Cornice, where the examples of the Vice alternate thus in groups. This systematic balancing of Sacred and Profane illustra- tions is found also in the Paradiso several times, e.g. iv. Stipplementary Notes. 133 83-4 (S. Laurence and Mucius) ; v. 66-70 (Jephthah and Agamemnon); viii. 130-2 (Esau and Jacob and Quirinus) 1 . 4. There is an obvious correspondence of the number of examples given of each Virtue and its corresponding Vice, though sometimes one or even two Cantos intervene between them. In Comici i. and v, three instances in the one case are balanced by three groups of instances in the other. In the former Cornice, the groups are marked off from one another by the repetition in twelve Successive Terzine, of the initial words, Vedeva, 0, and Mostrava, which are again repeated and gathered up into the three lines of a concluding Terzina (see Canto xii. lines 25-63) 2 . In the latter, the groups are marked off by an arrangement peculiar to that Cornice, viz. putting together two or more instances from Profane and Sacred History respectively instead of making the instances alternate. The only slight exception to this symmetry seems to be in Cornice vi, where the number of examples of the Virtue and of the Vice do not correspond. 5. Note especially that in every case some incident in the life of the Blessed Virgin is the first instance of the Virtue held up for admiration. 6. In every case also they are dismissed from the Cornice with the utterance by an Angel of a portion of one of the Beatitudes from the Sermon on the Mount. (In the 1st Cornice (xii. no) it is true the Angel is not expressly mentioned, and there is some doubt as to the reference of voci in 1. no. See Scartazzini's note, h.l.) 1 So in the JSp. to Can Grande, § I, an illustration is drawn from the Queen of the South and Pallas. 2 There is a very similar case of a Canto symmetrically constructed in Par. xix. 1 15-141. Compare also the six times repeated Ora conosce at the commencement of alternate terzine in Par. xx. 40-70. 1 34 Supplementary Notes. This last is a peculiarly interesting point, as it establishes I think beyond all possibility of doubt the reading Sitiunt instead of the commonly received Sitio in Purg. xxii. 4. This is the more interesting from the fact that the reading seems to have disappeared almost entirely from our existing MSS. In fact, it may be said to be practically devoid of MS. support. I have only met with it once, though I have searched for it in more than 180 MSS. This was in a MS. in the Corsini Library at Rome (No. 346 in Colomb de Batines). I find it registered, however, as occurring in two MSS. of some celebrity at Udine, known as the Godice Florio and the Codice Bartoliniano, which are stated by Colomb de Batines to have closely related texts. Other instances of this unity of design may be found in the way in which the first three Cornici are assigned to the first day and the last three to the second day, while the central Cornice (in which Accidia is purged) appro- priately occupies the intermediate night (see sup. p. 106). Also in the fact that a mystical dream occurs each day in the hour before sunrise (for this also see sup. p. 98). LIST OF TABLES. I. Calendar for March-April, 1300. II. „ „ „ 1301. III. Longitude of the Chief Places mentioned in Dante's Time-Eeferences. IV. Diagram of Simultaneous Hours at the above Places. V. Time-Eeferences in the Inferno. VI. „ „ „ Purgatorio. VII. A Table shewing the Unity and Symmetry of Plan in the Seven Cornici of Purgatory. Table I. Calendar, 1300. MARCH. 21 M 22 T 23 W 24 Th 25 F 26 S 27 s 28 M 29 T 30 W 31 Th API 1 ML. F 2 S 3 s 4 M 5 T 6 W r Th 8 F 9 S 10 s 1] M 12 T 13 W 14 Th SPRING EQUINOX. New Moon (Real). New Moon (Calendar. LADY DAY. Passion Sunday. PALM SUNDAY. Full Moon (Real, 3 a.m.) Kill Moon (Calendar). GOOD FRIDAY. Easter Eve. EASTER DAY. Table II. Calendar, 1301. MARCH. 24 F Full Moon (Keal). '25 S LADY DAY. '26 s PALM SUNDAY. ■27 M Full Moon (Calendar 28 T 29 W 30 Th 31 F GOOD FRIDAY. APRIL. l S Easter Eve. 2 s EASTER DAY. 3 M 4 T 5 W 6 Th h 2 Table III. « O H s Pn P Ah 180° W. i— ( Ph 90° Hi Hi «! P P3 t"9 45° 0° Longitude. 3 90° « o H « 1-1 1 — »> M P ,-,' -,\ ■ a fl §. § .& * -a 2 M rl > » 2 Ph B «1 <1 WPhO C 3 GO io '3 — b« "1 PQ <1 p— I p 1—1 o o <5 §5 :3 W 03 o s S5 2 « 5 -*5 CD I <3 •« K P9 PS O ->1 ^ pq Pm X ri esq 0J P. 2 § « o\ o g or M g § S :g £ W O x 1-4 oi co o ttf 5 pq pq . ° CO ^=5 CO « . S :S W ° X r- Ol " 03 £ ° 3 03 "tb < Cv ■a x pq w a Ph S- pq p 05 o 2 §• H =3 'S cq P3 o Pi c (4 D H O w > K 3 P - — - oq University of California SOUTHERN REGIONAL LIBRARY FACILITY 405 Hilgard Avenue, Los Angeles, CA 90024-1388 Return this material to the library from which it was borrowed. GAYLORD PRINTED IN U.S.A. UC SOUTHERN REGIONAL LIBRARY FACILITY AA 000 205 974 9