Morgan Callaway., Jr» Studies in tlie Syntax of tlie Linmsfarne Gospels * Supplementary Series: STUDIES IN ENGLISH PHILOLOGY Edited by James W. Bright ===== Number 5 = STUDIES IN THE SYNTAX OF THE LINDISFARNE GOSPELS With Appendices on Some Idioms in the Germanic Languages BY MORGAN CALLAWAY, Jr. Professor of English in The University of Texas baltimore The Johns Hopkins Press 1918 P£ UNIVERSITY OF CALIFORN1 1 . SANTA BARBARA PREFACE The present instalment * of my projected Studies in the Syntax of the " Lindisfame Gospels " is restricted to an in- vestigation of the Participle and of the Inlinitive. The main object of this instalment is to determine whether the syntax of these verbals in the Northumbrian dialect differs essentially from that in the West-Saxon dialect as set forth in the writer's monographs 2 on the Participle and the Infinitive in the latter dialect. The investigation is based upon a statistical reading of the four Lindisfame Gospels and of their Latin originals, as given in W. W. Skeat's The Holy Gospels: Anglo-Saxon and Northumbrian Versions, Cambridge, 1871-1887. In the study of each verbal, I have attempted to make my statistics com- plete, and have habitually given an account of the Latin corre- spondents of the Northumbrian gloss, and in the more doubtful constructions have cited the parallel passage in the Rushworth version of the Gospels, which latter has been read entire, although no account is taken thereof except in the way here indicated. Occasional omissions and misclassifications are in- evitable, but I hope that they will not prove so numerous or so serious as to invalidate the trustworthiness of this investigation. As the Lindisfame Gospels is merely an interlinear gloss, and in many respects a faulty one, a larger question at once presents itself, whether or not such a gloss can give any trust- worthy evidence as to the normal syntax of the dialect in which it is written. That very gross errors are made, is evidenced by such passages as the following, in which the author uses a nomi- native as the direct object of a verb and a dative as the subject of a finite verb: — John 18. 28 : geloeddon forSon se hcelend from 1 The next instalment will be devoted to the Subjunctive Mood. 2 The Absolute Participle in Anglo-Saxon, Baltimore, 1889; The Apposi- tive Participle in Anglo-Saxon, Baltimore, 1901; and The Infinitive in Anglo-Saxon, Washington, D. C, 1913. iii iv Preface caifa = Adducunt ergo iesum a caiapha (similarly in the Rush- worth version); Luke 16.5: mi^Sy weron geceigedo forSon syndrigum scyldgum hlaferdes his cuoaaS = conuocatis igitur singulis debitoribus domini sui dicebat (leaf lost in the Rush- worth version). But even these errors are not in reality so gross as they at first appear. The probability is that, in the former example, the glossator is merely naming the word, not the form thereof, to be used ; and that, in the second, he turns the ablative absolute of the participle in the native English way, by a finite verb, but, on reaching the ablative subject, has for- gotten about his rendering of the participle, and, naturally therefore, translates the ablative of the Latin noun by a dative. At any rate, such errors are exceptional ; and the proper answer to our question seems to be this. As to the normal order of words, this Northumbrian gloss, like most interlinear trans- lations, gives next to no evidence of value, since, as a rule, the glossator adheres strictly to the order of his Latin original. As to the normal idioms to be used in the combining of words into sentences, however, it gives invaluable evidence, especially in those locutions in which the Northumbrian gloss consistently diverges from the idiom of the Latin original. In a word, if in the syntax of any part of speech, as of the participle or of the infinitive, the glossator consistently shuns a Latin idiom, and consistently substitutes therefor another idiom, we are justi- fied in holding that the substitute idiom represents his native usage, — a principle that seems to me to hold perfectly in the syntax of the verbals. While, as stated above, the main purpose of the present in- vestigation is to discover what light this Northumbrian material may throw upon the syntax of the West-Saxon dialect, it is hoped that the studies may prove of interest from an absolute standpoint, the more so that they disclose several idioms not known in West-Saxon. Among the more noteworthy of these idioms may be mentioned the Absolute Participle with an Accusative Subject, possibly also with a Nominative Subject; the Infinitive as the Object of a Preposition ; the Imperative Preface v Use of the Infinitive; and that substitute for the Infinitive which I have ventured to denominate the Elliptical Accusative- with-Infinitive Construction. Moreover, certain constructions that are only slightly represented in West-Saxon, are somewhat frequent in the Lindisfarne Gospels, as the Inflected Infinitive with an Accusative Subject in Objective Clauses and the Unin- flected Infinitive with Accusative Subject in Subjective Clauses. So far as I have been able to learn it, I have given the history of opinion on all points discussed by me. Numerous works dealing with the phonology or the inflections of the Lindisfarne Gospels have been published, and are duly recorded in my Bibli- ography. Of these I need here mention only those that have been of most service to me : Professor Albert S. Cook's A Glos- sary of the Old Northumbrian Gospels (Lindisfarne Gospels, or Durham Booh), Halle, 1894; Dr. H. C. A. Carpenter's Die Destination in der Nordhumbrischen Evangelienuebersetzung der Lindisfarner Handschrift, Bonn, 1910; Dr. Theodor Kolbe's Die Konjugation der Lindisf aimer Evangelien, Bonn, 1912 ; and the several works by Professor Uno Lindelof listed in my Bibliography. So far as I know, my monograph is the second treatise devoted specifically to the syntax of the Lindis- farne Gospels, the first being Mr. C. E. Bale's The Syntax of the Genitive Case in the Lindisfarne Gospels, a University of Iowa Master's dissertation of 1907. But, in the nature of the case, some syntactical observations occur in the three works mentioned above, in the other treatises dealing with the inflec- tions of the Lindisfarne Gospels, and in the various editions of that text by Bouterwek, by Stevenson and Waring, and by Skeat, all recorded in my Bibliography. Wherever help has been found, it has been gratefully accepted and specifically acknowledged. Perhaps a word should be added as to the date of the Lindis- farne Gospels. And here I cannot do better than quote the most recent deliverance in reference thereto by the editor of our text, the late Professor Walter W. Skeat. In his English Dialects from the Eighth Century to the Present Day (Cam- vi Preface bridge, 1911), p. 22, Professor Skeat expressed himself as follows : " The Northumbrian glosses on the four Gospels are contained in two mss., both of remarkable interest and value. The former of these, sometimes known as the Lindisfarne MS., and sometimes as the Durham Book, is now MS. Cotton, Nero D 4 in the British Museum, and is one of the chief treasures in our national collection. It contains a beautifully executed Latin text of the four Gospels, written in the isle of Lindis- farne, by Eadfrith (bishop of Lindisfarne in 698-721), proba- bly before 700. The interlinear Northumbrian gloss is two and a half centuries later, and was made by Aldred, a priest, about 950, at a time when the ms. was kept at Chester-le-Street, near Durham, whither it had been removed for greater safety. Some- what later it was again removed to Durham, where it remained for several centuries." Since, as already incidentally stated, I have cited the Rush- ivorth Gospels in rare and difficult constructions, I quote, also, Professor Skeat's account of the Rushworth version of the Gospels, as given on p. 22 of his English Dialects'. " The second ms. is called the Rushworth MS., as it was presented to the Bodleian Library (Oxford) by John Rushworth, who was deputy-clerk to the House of Commons during the Long Parliament. The Latin text was written, probably in the eighth century, by a scribe named Macregol. The gloss, written in the latter half of the tenth century, is in two hands, those of Farman and Owun, whose names are given. Farman was a priest of Harewood, on the river Wharf e, in the West Riding of York- shire. He glossed the whole of St. Matthew's Gospel, and a very small portion of St. Mark. It is worthy of especial notice, that his gloss, throughout St. Matthew, is not in the Northum- brian dialect, but in a form of Mercian. But it is clear that when he had completed this first Gospel, he borrowed the Lindis- farne MS. as a guide to help him, and kept it before him when he began to gloss St. Mark. He at once began to copy the glosses in the older ms., with slight occasional variations in the grammar; but he soon tired of his task, and turned it over to Preface vii Owun, who continued it to the end. The result is that the Northumbrian glosses in this ns v throughout the three last Gospels, are of no great value, as they tell us little more than can be better learnt from the Durham book ; on the other hand, Farman's Mercian gloss to St. Matthew is of high value, but need not be considered at present. Hence it is best in this case to rely, for our knowledge of Old Northumbrian, on the Durham book alone.'''' The italics in the last sentence of this quotation are Professor Skeat's. As to the question raised in this sentence concerning the relative value of the Lindisfarne and the Rush- worth versions of the Gospels, the present writer prefers to re- serve judgment until the appearance of further instalments of his studies in the syntax of the Lindisfarne Gospels. As in my former syntactical studies, I have taken account of the verbals in the kindred Germanic languages. This fact will justify, I trust, the several Appendices strewn throughout the present monograph. In these appendices I have striven to bring up to date the chapter on the Absolute Participle in the Ger- manic Languages, published in 1889 ; that on the Appositive Participle in the Germanic Languages, published in 1901 ; and that on the Infinitive in the Germanic Languages, published in 1913. And I have devoted one Appendix (IX) exclusively to a consideration of some " Germanic Analogues to the Northum- brian Elliptical Accusative-with-Infinitive Construction." It is a pleasure to find that these later investigations strongly tend to confirm the theories formerly set forth by me as to the syntax of the verbals not only in Anglo-Saxon but also in the other Germanic Languages. With slight modifications here and there for individual languages, the theories derived from a minute study of the West-Saxon and the Northumbrian dialects seem to apply equally well to the other Germanic languages. Though fairly confident of the correctness of the view just expressed, I realize that my interpretation of the idioms of the participle and of the infinitive in the Germanic languages other than English rests upon statistics which, though carefully gathered bv others, are for several of the languages incomplete, and in viii Preface some instances take little account of the original Greek or Latin. I should be glad, therefore, if Germanic grammarians would test the theories in question by making as detailed a study of the syntax of the verbals in the other Germanic languages as I have attempted to give thereof in the West-Saxon and the Northumbrian dialects. Although my Bibliography lays no claim to exhaustiveness, I have striven to make it as nearly complete as was possible considering my remoteness from the larger libraries and the difficulties of communicating with Europe incident to the Inter- national War. Even partial success in this direction would have been impossible but for the kindness shown me by the librarians of several of our older universities. For such cour- tesies I wish to tender my cordial thanks to the librarians of the following universities: Chicago, Columbia, Harvard, and Johns Hopkins. And to the librarian of the University of Texas, Mr. John E. Goodwin, I am indebted for many kind- nesses. In the Bibliography I have given the titles not only of the chief accessible works on the Northumbrian Dialect of the Old English Period, but also of a few of the more noteworthy treatises on the other dialects of that epoch and on the Northern Dialect of the Middle English Period and of the Modern Eng- lish Period. And, as I am throughout comparing the Old Northumbrian syntax with that of West-Saxon and of the Ger- manic languages, I have added, in the Bibliography, the titles of the more noteworthy treatises dealing with the syntax of the participle and of the infinitive in English and in the other Ger- manic languages, in the later as well as in the earlier periods, that have appeared since the publication of my monographs on these verbals. A few dissertations that had appeared before the publication of those monographs, but that had escaped me or had been inaccessible, have likewise been added. In a word, as this study is supplemental to my former studies, so this Bibli- ography is supplemental to my former bibliographies. And only in the case of a few of the more important works have I repeated here titles given in my former monographs. Preface ix Despite the fact that this study of the verbals in Northum- brian is largely based upon the writer's former studies of the verbals in West-Saxon, I have tried to make the present mono- graph easily intelligible to the reader who may not have access to the earlier monographs. If I should not always have suc- ceeded in this, I hope the reader will generously grant me indul- gence for the endeavor's sake. And the further fact that the present study is a comparative one, will excuse, I trust, the fre- quent references in this monograph to my former studies. To the Editor-in-Chief of the Supplementary Series of Hesperia, Professor James W. Bright, I am indebted for his offering occasional suggestions for the improvement of my mono- graph, for his lending me rare books from his private library, and, most of all, for his volunteering to devote a number of Hesperia to this work. Professor J. L. Boysen, of the School of Germanic Languages of the University of Texas, has kindly read the proofs of the sections dealing with the Germanic lan- guages other than English, and has been of especial help in the subsections dealing with the Scandinavian Languages. My col- league and life-long friend, Professor Killis Campbell, has added another to his many former kindnesses to me. He has read the whole of the proof, has offered valuable criticisms upon the same, and has at all times given the undertaking his cordial sympathy. Where help is so cheerfully given as by these friends, indebtedness becomes, not a burden, but a perennial source of pleasure. Morgan Callaway, Jr. The University of Texas, Austin, Texas, March 31, 1918. CONTENTS CHAPTER I PAGE The Absolute Participle Introduction : Definition, Case 1 A. The Absolute Dative-Instrumental 2 General Statement 2 I. The Present Participle 3 Statistics 3 Notes 5 II. The Preterite Participle 7 Statistics 7 Notes 10 Origin of the Idiom 13 Appendix I. The Absolute Dative in the Other Germanic Languages 13 B. The Absolute Accusative 25 General Statement 25 I. The Present Participle 31 Statistics 31 Notes 33 II. The Preterite Participle : Statistics 33 Origin of the Idiom 34 Appendix II. The Absolute Accusative in the Other Ger- manic Languages 34 C. The Absolute Nominative 35 General Statement 35 I. The Present Participle: Statistics 36 II. The Preterite Participle: Statistics 37 Origin of the Idiom 38 Appendix III. The Absolute Nominative in the Other Ger- manic Languages 38 D. Synoptic Table of the Absolute Participle 39 E. The Uses of the Absolute Participle 39 General Statement 39 Synoptic Table 42 F. Origin of the Idioms with the Absolute Participle: Summary Statement 43 CHAPTER II The Appositive Participle Introduction : Definition, Regimen 45 A. The Present Participle 46 xi xii Contents PAGE I. With an Object 46 General Statement 46 Statistics 46 1. Of the Nominative 46 2. Of the Genitive 49 3. Of the Dative-Instrumental 50 4. Of the Accusative 50 Origin of the Idiom 51 Appendix IV. The Governing Power of the Apposi- tive Present Participle in the Other Germanic Languages 52 II. Without an Object 61 General Statement 61 Statistics 61 1. Of the Nominative 62 Notes 64 2. Of the Genitive 67 3. Of the Dative-Instrumental 68 4. Of the Accusative 69 Origin of the Idiom 70 B. The Preterite Participle 71 I. With an Object 71 General Statement 71 Statistics 71 Origin of the Idiom 72 II. Without an Object 72 General Statement 72 Statistics 72 1. Of the Nominative 72 2. Of the Genitive 74 3. Of the Dative-Instrumental 75 4. Of the Accusative 75 Origin of the Idiom 78 C. Synoptic Table of the Appositive Participle 78 D. Uses of the Appositive Participle 78 General Statement 78 I. The Adjectival Use 81 II. The Adverbial Use 82 III. The Co-ordinate Use 85 Notes 86 Synoptic Table 87 E. Origin of the Idioms with the Appositive Participle: Sum- mary Statement 89 CHAPTER III The Infinitive Introduction 90 Forms of the Infinitive 90 Contents x [[[ PAGE Functions of the Infinitive 9] Differentiation of Uninfected Infinitive and of Inflected Infinitive a Main Problem 91 The Origin of the Various Idioms with the Infinitive Another Main Problem 91 The Order of Words not Treated 92 The Voice of the Infinitive 9 2 The Subjective Infinitive 93 A. The Active Infinitive 93 1. With Active Finitive Verb 93 General Statement 93 Uninfected Only 95 Inflected Only 95 Uninflected and Inflected Each 9 g Statistics 9 g 2. With Passive Verbs 97 General Statement 9 7 Statistics 9 g 3. Differentiation of the Two Infinitives 98 4. Origin of the Idiom 99 Appendix V. The Subjective Infinitive in the Other Ger- manic Languages jqO B. The Passive Infinitive 101 C. Notes 102 The Objective Infinitive 102 A. The Active Infinitive 102 1. With Active Finite Verb 102 General Statement 102 Uninflected Only 104 Inflected Only 105 Uninflected and Inflected Each 105 Statistics 107 2. With Passive Verbs 109 General Statement 109 Statistics 1 10 3. Differentiation of the Two Infinitives no 4. Origin of the Idiom 1 1 1 B. The Passive Infinitive 1 13 C. Notes 113 III. Other Substantival Uses of the Infinitive 115 A. The Active Infinitive 115 As a Predicate Nominative 116 As an Appositive 116 As the Object of a Preposition 117 IV. The Predicative Infinitive with Auxiliary Verbs 118 A. The Active Infinitive 118 General Statement 1 IS II. v Contents PAGE Uninflected 118 Inflected 118 Differentiation of the Two Infinitives 119 Origin of the Idiom 119 B. The Passive Infinitive 119 C. Notes 119 V. The Predicative Infinitive with Verbs of Motion and of Rest Other Than Wutun 121 The Active Infinitive 121 General Statement 121 Uninflected Only 121 With Verbs of Motion 121 With Verbs of Rest 122 VI. The Predicative Infinitive with Wutun 122 A. The Active Infinitive 122 Origin of the Idiom 123 B. The Passive Infinitive 123 C. Notes 123 VII. The Predicative Infinitive with Bia{n) {Wosa) 123 The Active Infinitive 123 A. Denoting Necessity or Obligation 123 B. Denoting Futurity 124 C. Denoting Purpose 125 Origin of the Idiom 125 Notes 126 VIII. The Predicative Infinitive with Accusative Subject 126 General Statement 126 As Object 127 A. The Active Infinitive 127 I. Uninflected 127 General Statement 127 With Verbs of Commanding 128 With Verbs of Causing and Permitting 128 With Verbs of Sense Perception 128 With Verbs of Mental Perception 129 With Verbs of Declaring 129 With Other Verbs 129 Statistics 130 Origin of the Idiom 131 II. Inflected 134 General Statement 134 Statistics 134 Origin of the Idiom 136 Differentiation of the Two Infinitives: see Origin of the Idiom 136 B. The Passive Infinitive 137 As Subject 137 Contents PAGE A. The Active Infinitive 137 I. Uninflected 137 General Statement 137 Statistics 138 Origin of the Idiom 140 II. Inflected 141 Statistics 141 Origin of the Idiom 141 Differentiation of the Two Infinitives: see Origin of the Idiom 141 B. The Passive Infinitive 141 C. Notes 141 Appendix VI. The Predicative Infinitive with Accusa- tive Subject in the Other Germanic Languages 142 IX. The Predicative Infinitive with Dative Subject 143 The Active Infinitive 143 Appendix VII. The Predicative Infinitive with Dative Sub- ject in the Other Germanic Languages 145 X. The Final Infinitive 155 A. The Active Infinitive 155 1. With Active Finite Verb 155 General Statement 155 Uninflected Only 156 Uninflected and Inflected Each 157 Inflected Only 157 Statistics 158 2. With Passive Verbs 161 3. Differentiation of the Two Infinitives 161 4. Origin of the Idiom 161 B. The Passive Infinitive 162 C. Notes 162 XL The Infinitive with Adjectives 165 A. The Active Infinitive 165 General Statement 165 1. Uninflected 166 Statistics 166 2. Inflected 166 Statistics 166 3. Differentiation of the Two Infinitives lt>7 4. Origin of the Idiom 168 B. The Passive Infinitive 168 C. Notes 168 XII. Other Adverbial Uses of the Infinitive 160 The Causal Infinitive 1 *i!» The Consecutive Infinitive 169 The Instrumental Infinitive L69 Origin of the Idiom 169 Appendix VIII. The Consecutive Infinitive in Gothic 170 xvi Contents PAGE XIII. The Infinitive with Nouns 171 A. The Active Infinitive 171 General Statement 171 Uninflected 171 Statistics 171 Inflected 172 Statistics 172 Differentiation of the Two Infinitives 173 Origin of the Idiom 173 B. The Passive Infinitive 174 C. Notes 174 XIV. The Imperative Infinitive 175 The Active Infinitive 175 Uninflected 175 Statistics 175 Origin of the Idiom 176 XV. Some Substitutes for the Infinitive 176 I. The Predicate Nominative of the Present Participle for the Predicative Infinitive after Verbs of Motion. . 176 II. The Predicate Accusative of the Present Participle for the Predicative Infinitive with Accusative Subject 177 Origin of the Idiom 179 III. The Elliptical Accusative-with-Infinitive Construction 180 A. Made Up of an Accusative Plus a Present Par. ticiple 180 B. Made Up of an Accusative Plus a Preterite Par- ticiple 182 C. Origin of the Idiom 183 Notes 184 Appendix IX. Germanic Analogues to the North- umbrian Elliptical Accusative-with-Infinitive Construction 186 XVI. Notes 196 XVII. Synoptic Tables of the Uses of the Infinitive and of the Sub- stitutes Therefor 19* CHAPTER IV Results I. Concerning the Participle 199 A. The Absolute Participle 199 B. The Appositive Participle 201 II. Concerning the Infinitive 203 A. The Active Infinitive 203 B. The Passive Infinitive 213 C. Substitutes for the Infinitive 214 Bibliography 216 Addenda and Corrigenda 240 STUDIES IN THE SYNTAX OF THE LINDISFARNE GOSPELS CHAPTEK I THE ABSOLUTE PAKTICIPLE Introduction In this chapter, a participle is considered Absolute when its use conforms to the definition of the Absolute Clause given in The Absolute Participle in Anglo-Saxon, p. 1: "When to a substantive not the subject of a verb and dependent upon no other word in the sentence (noun, adjective, verb, or preposi- tion) a participle is joined as its predicate, a clause is formed that modifies the verbal predicate of the sentence and denotes an accompanying circumstance, as in Urbe expugnata, impera- tor rediit." From its apparent, but not real, grammatical in- dependence, this has been denominated an absolute clause. The case of the absolute participle varies in the Lindisfarne Gospels. As in West-Saxon, so in the Lindisfarne Gospels normally the absolute participle is in the dative-instrumental case. But in the Lindisfarne Gospels we find, too, the accusa- tive and, apparently, the nominative used absolutely, — both idioms that are unknown in West-Saxon, as I have tried to show ^in my aforementioned monograph. For each of the cases used absolutely the participle is at times uninfected; and for each case divergent participial forms are used occasionally ; hence at times, in absolute uses, the case of the clause has been deter- mined by the form rather of the substantival subject than of the participial predicate. At times, too, the form of substan- tive or of participle (occasionally of both) is indefinite; and we have what may be termed " crude " * forms of substantive 1 On the term " crude " here, see The Absolute Participle in Anglo- Saxon, p. 2. 1 2 The Syntax of the Lindisfarne Gospels or of participle (or of both) ; that is, a weathered form that cannot with certainty be assigned to any definite case, but that usually represents an unweathered dative-instrumental (occa- sionally an unweathered accusative or nominative) . That some of my assignments of case will not meet with acceptance, I do not doubt. Critics will be the more lenient, I believe, when they recall the fact that the monographs specifically dealing with the inflections of the Lindisfarne Gospels, as those by Dr. H. C. A. Carpenter and by Professor A. S. Cook, cited in my Biblio- graphy, have left many individual examples, even when specific- ally cited, unclassified as to case and gender. That the possi- ble differences as to case-assignment will not be so numerous as to invalidate the trustworthiness of my respective groups, is my hope and belief. A. The Absolute Dative-Instrumental In the Lindisfarne Gospels proper, exclusive of the " Intro- ductions " thereto, the Absolute Participle in the Dative- Instrumental is not quite so frequent as in the West-Saxon Gospels, there being 52 examples in the former to 66 in the latter. But, if we include the " Introductions," the number 2 of examples of the Absolute Dative-Instrumental is appreciably larger in the former (94) than in the latter (66). The following are typical examples of the Absolute Dative- Instrumental : — Mlc. 14. 22 : ettendum him onfeng se hselend hlaf = manducantibus ittis accepit iesns panem; Mk. 16.20 b > c : 3 2 The detailed statistics are as follows: Lindi sf arne : West- Saxon : Pres. Ptc. PastPtc. Pres. Ptc. PastPtc. In the Gospels 33 19 25 41 In the Introductions 20 13 00 00 Totals 62 32 25 41 Grand Totals 94 66 3 Throughout this study, superior letters distinguish different examples in the same verse. As a rule, I have expanded contractions. I have not differentiated p and S, but have uniformly used S. For the symbol I I The Absolute Participle 3 bodadon eghuser drihtne midwyrcende & <5set word trymende mi(5 fylgendum becenum uel tacenum = Praedicauerunt ubique domino cooperante et sermonem conformant e sequentibus sig- nis; Mk. 1.18: hreconlice mi(5(5y forleorton uel forletnum net- turn fylgendo weron him = protinus relictis retibus secuti sunt eum. The complete Statistics of the Dative-Instrumental Absolute are as follows : — I. THE PRESENT PARTICIPLE (62, including all in the Notes except Note 3) Dative-Instrumental, Singular or Plural (54 + 8 in the Notes) The Dative-Instrumental of the Absolute Present Participle ends normally in -nde in the singular (masculine and neuter) and in -ndum in the plural (masculine and neuter) : see Carpen- ter, I. c, §§ 527, 532, 537, and 542. In the singular (mascu- line) the participle ends once in -nd (L. 14.32) and once in -ndum (L. 21.5) ; in the plural (masculine), once in -ndu for -ndum (L. 9.43). Normally the vowel preceding the endings named above, is e, but occasionally it is a or u: see Carpenter, I. c.j § 529. When no ending is given below, -ndum is to be understood. astiga, descend (1):L. (1) : 9.37. bidda, ash, request (1) : L. (1) : I. 4 5.10. cliop(p)iga, cry out, exclaim (1) : Mat. (1) : I. 21.4. have written out the Latin equivalent, uel, and for the symbol ~] I have used &. Occasionally, for the sake of clearness, I have altered the punctu- ation of the original text; and I have habitually ignored the hyphenation of compounds in the Latin original and in the Northumbrian gloss. 4 1, here and elsewhere in the statistics, when used in connection with the name of a gospel, indicates that the example occurs in the " Introduction " to that gospel. The examples in the " Introductions " are cited by page and line; those in the Gospels, by chapter and verse. The segregation of the examples occurring in the " Introductions," here and throughout this monograph, from the examples occurring in the Gospels proper, facilitates a comparison of the respective idiom in the Lindisfarne Gospels and in the West-Saxon Gospels, in which latter no Introductions occur. 4 The Syntax of the Lindisfarne Gospels CWOeSa, say, declare (2) : L. (1) : I. 10.2V- J. (1) : I. 6.13. doa, do, make (2) : Mat. (1) : I. 19.1 (with the dative, a proper name, to be supplied from the Latin ablative). — L. (1) : 14.32 (-nd; Cook: pres. p. ; Carpenter, I. c, § 532 : dsm.). efneiorna, run together (1) : L. (1) : 11.29 a . efnespreca, talk, converse (1) : L. (1) : I. 11.12. eftcerra, return (1) : /. (1) : I. 8.3 b . eta (eatta), eat (1) : Mk. (1) : 14.22. fsera,#o (1) : Mat. (1) : 20.29. fraigna (fregna), enquire (1) : L. (1) : I. 6.19 b (-nde). fromgeonga, depart (1) : Mat. (1) : 11.7. gefeaga, rejoice (2) : L. (2) : I. 3.14 (-nde) ; I. 8.5 b (-nde). gefraigna (-fregna), enquire (1) : Mat. (1) : I. 20.3 (-nde: gefraigende ) . genera, ftear (1) : L. (1) : 19.11. gelefa, believe (3) : J. (3) : I. 5.6; I. 5.14; I. 6.9 b . geonga, go (2) : L. (2) : 9.34 b , 57. habba, have (1) : L. (1) : 7.42. hlioniga (hliniga), recline (1) : L. (1) : I. 5.13 (-nde). hrowa (rowa), row (1): L. (1): 8.23 (hrowundum: see Carpenter, Z. c, § 542, anmk.). innfara (-fsera), enter into (1) : L. (1) : 22.10 a . iorna, run (1) : J. (1) : I. 8.3 a . miSwyrca, cooperate (1) : MA;. (1) : 16.20 b (-nde). ofstiga, descend (2) : Mat. (1) : 17.9.— M. (1) : 9.9. ondswaeriga, answer, respond (1) : J. (1) : I. 5.15. onginna, begin (1): L. (1): 21.28 a . onsacca (onssecca), deny (l) : L. (1) : 8.45. singa, sm# (1) : Mat. (1) : I. 8.10 ( ?). slepa, sleep (2): Mat. (1): 28.13 b .— Mfc. (1): I. 3.5 b (-nde). smeaga, think, reflect (1) : L. (1): 3.15. SOeca, seek (5) : L. (2) : I. 6.19 a (-nde) ; I. 10.9 (with the dative, a proper name, to be supplied from the Latin). — J. (3) : I. 5.3 a ; I. 6.1; I. 6.14 a . Stiga, ascend, descend (2) : L. (2) : I. 4.1 a ; 2.42. The Absolute Participle 5 trymma, confirm (1) : Mh. (1): 16.20 c (-nde). undoa, unloose (1) : L. (1) : 19.33 (or Appositive?). ungelefa, disbelieve (1) : L. (1) : 24.41 a . WUndriga, wonder (3): Mh. (1): I. 4.12.— L. (2): 9.43 {-ndu: Cook: dp. ; Carpenter, Z. c, § 542: dpm.) ; 24.41 b . ymbsceawiga, look around (1) : L. (1) : 6.10. ymbsitta, sit around (1) : L. (1) : 22.55. ymbstonda, stand around (1) : L. (1) : 12.1 a . NOTES 1. J.n OZeZ English Absolute Dative-Instrumental Active Translates a Latin Ablative A bsolute Passive in the following : — Mat. I. 16.3 b : tungolcrseftiga stearra him hlatuu gesaegde ge- breingendum uel geafendum mi(5 fiingum togebedon — christum magi stella sibi duce nuntiante oblatis muneribus adorauerunt; 5 Mat. I. 20.20 c : Sser moder sunana sebedseis biddende swccendum sedlum cueS = matri filiorum zebedaei petenti negatis sedibus ait: — L. 6.20: he ahebbenduni (sic) egum on Segmim his cuoeS = ipse eleuatis oculis in discipulos suos dicebat (leaf lost in the Rushworth version) ; L. 15.13 : mi(5(5y gesomnandum allum . . . fserende wses = congregatis omnibus . . . profec- tus est (leaf lost in the Rushworth version). The West-Saxon Gospels has an active Appositive Participle (beseonde) in L. 6.20 and an active finite verb (gegaderude) in L. 15.13. — See the Note on the Voice of the Appositive Participle at the end of Chapter II, and of the Predicative Participle at the end of Chapter III. 2. The Dative-Instrumental Subject Is to Be Supplied in the following, in one of which the Latin has an appositive, not an absolute, participle: — Mk. 15.29 a : & Sa bifcerendum gee- bolsadon uel ebolsande hine cserrende heafda hiora & cuoeftende = Et praetereuntes blasphemabant eum mouentes capita sua et dicentes (Rushworth: ib.j W. S. : sfv.) ; — L. 24.47: sua? were * If miS is considered a preposition here, we have an appositive, not an absolute, phrase; but I consider mi?S an adverb here. See Note 3 below and Note 1 under " The Preterite Participle." 6 The Syntax of the Lindisfame Gospels rehtlic Saette crist ge(5rowade & eftarisa from deadum (5irdda daege & Saette were abodenn in noma his hreonise & eftfor- gefnise Sara synna in allum cynnum ongmnendum from hie- rnsalem = sic oportebat christum pati et resurgere a mortuis die tertio et praedicari in nomine eius paenitentiam et remis- sionem peccatorum in omnes gentes incipientibus ab hierosoly- ma; — J. I. 6.1: <5a aeldo mi(5 unsibsumnise gedroefedo uoeron soecendum he from Saem inlichtet cySig geworfrad U83S = pharisaeis anxietate turbatis atqne quaerentibus ipse ab inluminato cognitus adoratur (Cook: celdo, ap. ; soecendum, dp., but celdo must be np., I think). 3. A Prepositional Phrase Plus a Present Participle in the Dative-Instrumental Translates a Latin Ablative Absolute Passive in Mat. I. 3.1 a : we gehehtun Saet of dwm ana (5a Soht gesegon gecerde uel givixla gemendum Sa aefterra gewuna we ondetaS = imperauimus ut his tantum quae sensum uide- bantur mutare correctis reliqua manere pateremur. Strictly according to our definition, of course, we have not an absolute clause here ; nor does the clause seem to be appositive. Since it is such a botch, I have not counted the clause under either head. See Note 1 above and Note 1 below, p. 10. I consider the Northumbrian participle Attributive in such passages as the following, in which, again, a Lindisfame present participle in the dative corresponds to a Latin passive participle in the ablative: — /. 11.41: uutudlice mid hebbendum upp egum cuoe(5 — autem eleuatis sursum oculis dixit; ib. 17.1: mid un- derhebendum egum in heofnum cuoeS = subleuatis oculis in caelum dixit. In this use of the present participle in the dative after mid, as in the similar use of the preterite parti- ciple in the dative after mid, discussed in Note 1 below, we doubtless have an illustration of the Germanic striving for a method of translating the Latin ablative absolute that would not do violence to the English (Northumbrian) idiom. See the discussion of the West-Saxon be him liflgendum in The Absolute Participle in Anglo-Saxon, pp. 42-44, and of the The Absolute Participle 7 Scandinavian participial dative phrase introduced by a prepo- sition (usually at but occasionally med), below, in Appendix I. 4. An Old English Dative-Instrumental Absolute Is Made Up of a Noun and an Adjective in the Dative in Mat. I. 18.6 a : in eorSo Sara lioda halgum monnum diobles fara Serhgelefde uel sende in bergum = In terra genassenorum sanatis hominibus da?mones ire permittit in porcos. For the same idiom in West- Saxon, see The Absolute Participle in Anglo-Saxon, p. 44. 5. A Dative-Instrumental Participle ivith an Uninflected Pronoun Translates a Latin Ablative Absolute in L. 21.5 : Sara sum cuoedendum of temple . . . cuoe(5 = quibusdam dicenti- bus de templo . . . dixit. II. THE PRETERITE PARTICIPLE (33, including the examples in the Notes) The Dative-Instrumental of the Absolute Preterite Participle ends, in the singular, oftenest in -ed (masculine, feminine, and neuter), less frequently in -(e) ne (masculine and neuter), and occasionally in -eno (masculine, feminine, and neuter), in -de (neuter), in -edo (-ado) (feminine), in -t (neuter), in -en (masculine), in -na (feminine), and in -ni (masculine) ; in the plural, normally in -um (masculine, feminine, and neuter: of strong verbs in -num and of weak verbs in -dum), occasionally in -(e)no (masculine and neuter) and in -de (feminine). Con- cerning these endings, see Carpenter, I. c, §§ 549, 552, 553, and 556. As indicated below, some of the examples are ex- tremely doubtful. When no ending is given below, -um is to be understood. acuoeSa, say (1) : Mat. (1) : 26.30: efne acwoedni uteodon on mor = hymno dicto exierunt in montem. [Cook : efne, dsm. ; acwoedni, dsm. ; Carpenter, I. c, § 549 : dat.-inst. sg., masc] ahebba, lift up (1): L. (1) : 24.50. bega, bend (1) : Mh. (1) : 10.17: sum oSer cneuo beged fore bine baedd hine = quidam genu flexo ante eum rogabat eum. 8 The Syntax of the Lindisfarne Gospels [Professor Cook makes cneuo dsn., but does not assign any case to beged.'] bityna, close (1):J. (1) : 20.26. efnegeceiga, call together (2): Mh. (1): 8.1. — L. (1): 23.13. forblawa, blow (1): J. (1): 6.18: Se sse uutudlice winde miclum forblauene ofstod uel aras = mare autem uento magno flante exsurgebat. forcu(m)ma, become dry or hard (1) : L. (1) : 21.26. fordrifa, drive out, eject (2): Mat. (1): I. 18.11: ten dumbe tunga fodrifen uel gescyfen diubol — (blank) = muto linguam eiecto dcemone reddit. [Diubol may be nomina- tive here, as is claimed by Professor Cook; if so, fordrifen is predicate nominative or absolute nominative instead of unin- flected dative-instrumental.] Mh. (1) : 5.40. foresetta, propose (1): Mat. (1): I. 19.12 b : striones bisen uel meregrotta bigetna efennise foresetna gelic Son bisin of segna fiscum hine gehriordum ofgessede = Thesauri uel mar- garitse repertae comparatione proposita similiter parabolam de saginae piscibus se prandis (= separandis) exponit. [Cook: efennise, " ns. ? " foresetna, " apn. ? " But I do not see why the noun may not be considered dative-instrumental singular. The participle is perhaps apn., proposita being mistaken for the nominative plural neuter instead of the ablative singular femi- nine. Carpenter, I. c, § 551, makes the participle apn., also.] forleta, leave, dismiss (4): Mat. (1): 13.36. — Mh. (1) : 1.18. — L. (2) : 5.11: underlseded woeron to eor(5o scioppo for- letno allum gefyligde weron hine = subductis ad terram nauibus relictis omnibus secuti sunt ilium. [Professor Cook makes for- letno " npn. ? " here, but dpm. in the next passage that I quote. To me, the two passages hang together. Dr. Carpenter, I. c, § 550, considers forletno in L. 5.11 as napn. He does not cite L. 5.28.] L. 5.28: forleorte uel miS8y allum forletno aras fyl- gende waes him = relictis omnibus surgens secutus est eum. (ge)bega, bend (1) : Mat. (1) : 27.29 a : cnew gebeged bifora him bismeredon = genu flexu ante eum inludebant. [Cook: The Absolute Participle 9 cnew, ds. ; gebeged not classified. See Carpenter, I. c, p. 253, who seems to consider that the participle is used absolutely here although he assigns no particular case thereto.] (ge)binda, bind (1) : Mat. (1) : 22.13. (ge)bringa, bring (1): Mk. (1) : I. 3.10: sende twoelfe boderes miS bodum laereS gebroht gefea haelo uel halra = Mittens duodecim pradicaturos praeceptis instruit conlata gratia sanitatuni. [Cook : gefea, nsn. here though cited as ds. for other passages; gebroht, pp.; but see Carpenter, I. c., p. 253, who seems to consider that the participle is used absolutely here although he assigns no particular case thereto.] (ge)cuoeSa, say, speak (1) : /. (1): I. 1.13: gecuedne cri-ste = dicente christo. (ge)drysn(i)a, extinguish, end (1): L. (1): I. 3.4 a : biS said Ssette gode on god full & sune selenis gedrysned uel geendad gebed apostolum gewordne mi(5 hlod drihtnes gecorenes tal = datur ut deo in deum pleno et filio proditionis extincto (sic) oratione apostolis facta sorte domini electionis numerus. [Cook : gedrysned, pp. ; gebed, dsn. ; Carpenter : not cited.] (ge)hera, hear (1): Mat. (1): 15.12: wast forSon aeldor- menn geherde word uel gehered wses word geondspyrnede weron uel aron ? = scis quia pharisaei audito uerbo scandalizati sunt ? [ Geherde may be indicative preterite ; it is not cited by Cook for this passage. Carpenter does not cite geherde as a participle.] gesparriga, close, shut (1): Mat. (1): Q.d: & gesparrado dure Sin gebidd f aeder Sinne = et clauso ostio tuo ora patrem tuum. [Cook: gesparrado, " dsn. ? " dure, dsf. ; Carpenter, I. c, §553, anm. 3: "In gesparrado dure Sin, Mi. 6.6 (= clauso ostio tuo) sind kasus und genus schwer zu bestimmen."] (ge)Sreatiga, rebuke (1) : Mk. (1) : I. 2.16. (ge)waelta, bend (1) : Mat. (1) : 17.14: geneolecde to him monn cneum gewoelteno before hine cueS = accessit ad eum homo genibus prouolutis ante eum. [Cook: cneum, dpn. ; gewoelteno, pp. Carpenter, I. c, § 556, anm. 2, makes gewoelteno here napn.] ofgemearciga, designate (1) : L. (1): I. 6.14: ofgemcercade 10 The Syntax of the Lindisfarne Gospels asc oSrum tusem & unseofontigum deadum sile(5 bodo = Desig- natis et aliis Ixx duobus dat praecepta. [Ofgemcercade may be indicative preterite, 3 rd sg., as Professor Cook holds. Carpenter does not cite ofgemcercade as a participle.] onsetta, place upon (1) : Mk. (1) : 8.23. [Carpenter, I. c, § 556, amnk. 2, calls attention to the fact that in onsetnum we have an analogical strong dative plural instead of the expected weak form.] underlseda, put down (1) : Mat. (1) : I. 3.12 : miSSy Sonne <5a regnlas (5u redes (5a underbeged uel under(5iodid arun (5ses sceomaes uel telnisses dwala underlaeded = cum itaque canones legeris qui subiecti sunt confussionis (sic) errore sublato. [Cook: dwala, dsm. ; underlaeded, pp. Carpenter does not cite underlceded as a participle.] unforleta, not to leave (1): Mk. (1): 12.20: dead wses unforletne sed = mortuus est non relicto semen. [Rushw. : ib.; W. S. : na Icefedwm scede. — Cook: unforletne, adj. ptc. in dsn., but sed, asn. — Carpenter, I. c, § 549 : unforletne, disn. — Clearly the erroneous form of the Lindisfarne noun (sed) is due to the erroneous form of the Latin original (semen) .] untuna (untyna), open (1): Mat. (1): I. 4.4 b : miS (5y wutedlice untuned hoc swse o(5re bisen (5set uel Sa?t forecwide gewite (5u welle = Cum igitur aperto codice uerbi gratia illud siue illud capitulum scire uolueris. [Boc may be nominative, as is held by Professor Cook ; if so, untuned is predicate nomi- native or absolute nominative. Carpenter, I. c, § 455, however, gives hoc as one form of the dative, but not for this passage. He does not cite untuned as a participle.] WOrSa, become, be made (1) : Mat. (1) : I. 18.18. NOTES 1. A Prepositional Phrase Plus a Preterite Participle in the Dative-Instrumental Translates a Latin Ablative Absolute in Mat. I. 17.12 : micles beames to uel mid efennisse gewordeno lsera(5 = trabis comparatione facta docet. [Cook : efenisse, dsf. ; gewordeno, " nsf. (nfn. ?) ; " Carpenter, I. c, § 549 : " In Mt. The Absolute Participle . 1 1 Pr. 17.12 to uel mid efennisse gewordeno (c'omparatione facta) ist der Kasus schwer festzustellen." Personally I am inclined to consider to or wid as an organic part of the noun and to write toefennisse (or 'wid efennisse) ; in which case we should have a true dative-instrumental absolute. Similarly, in the following passage, I should read midceping as one word and consider that we have an absolute dative-instrumental: L. I. 5.2 : hia lycedon ... of fsestern forbodeno (5a?m brydgum aec woedes & wines niwes mid ceping becuoeS = murmurantes . . . de ieiunio athibita sponsi et uestimenti ac uini noui compara- tione redarguit. [Cook: forbodeno, ns. ; ceping, dsfn. ; Car- penter, I. c, § 549, anmk. 1 : "In L. Pr. 5.2, forbodeno ( = athibita) liegt vielleicht nsf. oder npn. vor. Cook halt die letzten 3 fur nsf."] In both of these examples, mid and to seem to me to be suggested by com in comparatione. I count these two examples. On the other hand, I consider that the Northum- brian participle is used Attributively in passages like the fol- lowing, in which, also, the phrase is introduced by a preposition (mid): — Mk. I. 4.15 c : haedno forebeadend gelicad uel gebis hersumnise mid bisene cedeawed = gentiles prohibens imi- tandos humilitatis exemplo monstrato ; ib. 14.26: mid sua cuoednum wordum (5ona foerdon on mor = liymno dicto ex- ierunt in montem ; ib. I. 3.1 a : <5egnas to bodanne miS word mid gebrohtum mcehtum sendeS = discipulos ad praedicandnm uerbo conlatis uirtutibus mittet (possibly mid and gebrohtum should be written as one word, in which case the participle would be used Absolutely) ; ib. 6.5 : untrymigo mid onsetnum hondum gegemde uel gehaelde = infirmos inpositis manibus curauit; — L. I. 7.5 a : Sone dumba diulas mid dy geboette gemeS dcet spree = Mutum daemonia cum restituto curat eloquio (possibly we should read middy as one word and consider it a conjunction?); ib. I. 6.2: mid gebroehtum moehtum sende(5 tuoelfum = Conlatis uirtutibus mittit duo- decim (see note on Mk. I. 3.1 a above); ib. I. 4.8: Srittig wintra mid dio gefuulivad drihtne Srinise on fulwiht assegdnise tosceaded is = triginta annorum baptizato domino trinitatis in 12 The Syntax of the Lindisfarae Gospels baptismo niysterium declarator; ib. 22.41: mid gesetnum cne- oum gebsed = positis genibus orabat ; ib. I. 9.6 a : e(5nise & gese- teno mid bisene from esne herende uel foedende lsereS = f acul- tatem positaque similitudine de seruo arante uel pascente docet ; ib. 10.30 a : mid wundum onsettenum fromfoerdon half cwic uel lifigiende forleten = plagis impositis abierunt semiuiuo re- licto; — /. 19.30: raid gebegdum heafde gesalde (5set gast = in- clinato capite tradidit spiritum. On the borderline between the Attributive and the Appositive use, but leaning more to the latter, is the participle in the following: — L. I. 4.13: eftaswgd so(51ice uel sec (5io widua of serapta uel (5a3s licSroueres neman syri raid clamsunge (5aette (5erh middum hiora . . . oferfoerde assegd is = relata etiam uidua de serapta uel leprosi neman siri mundatione quod per medium eorum qui eum de monte prsecipi- tare uolebant transierit indicatur; ib. I. 3.4 b : bi[(5 said (5gette gode on god full & sune selenis gedrysned uel geendad gebed apostolum gewordne mid Mod drihtnes gecorenes tal = datur ut deo in deum pleno et filio proditionis extincto oratione apostolis facta sorte domini electionis numerus etc. [Cook: gewordae, dsn. ; Mod, dsn ; Carpenter, I. c, § 540 : gewordne, dative-instru- mental, sn.] In L. I. 9.14 a , the phrase is introduced by the preposition of, and again the participle is more probably Ap- positive than Attributive: of gebed Sees aalaruas on temple & bsersynnig foregesettet laereS ne to worpanne (5a merdo ah to ondetende (sic) synno = Oratione pharisaei in templo et publi- cani proposita docet non iactanda merita sed confitenda peccata. — See Note 3 above, under " The Present Participle." 2. An Old English Dative-Instrumental Absolute Is Made Up of a Noun and an Adjective in the Dative, possibly, in L. I. 8.14: laSum (5a nedlicnm mi(5 Son & sauel his & ondfenge rode gefylgendo woero (5aet gelic getimbrendes torres & of gefeht cyninga tuoege gesceadeS = Oditnris necessitudines insuper et animam suam et assumta cruce secuturi similitudinem aedifi- candae turris et de bello regum duorum exponit (Cook: ond- fenge, not cited for this passage; rode, dsf.). The Ah solute Participle 13 ORIGIN OF THE IDIOM That the Dative-Instrumental Absolute here in the Northum- brian, as in the West-Saxon, is due to Latin influence is evident. (1) The dative-instrumental translates a Latin ablative absolute in all cases except two (Mb. 15.29 a and L. I. 5.13, in each of which the Latin has an appositive participle). (2) In the ma- jority of instances (157 out of a total of 249 or in the ratio of 1.58 to 1, for the Isindisfarne Gospels as a whole), the Latin absolute construction is rendered otherwise than by an absolute participle in the Northumbrian, usually 6 by a co-ordinated finite, verb (about 74 times) or by a subordinated finite verb (about 70 times), or in the ratio of 1.057 to 1, whereas in the West-Saxon Gospels the subordinated finite verb is somewhat more common than the co-ordinated, the proportion being 1.19 to 1. Despite this divergence, both glossator and translator are true to the native English idiom of finite verb instead of parti- ciple. (3) The idiom in West Saxon and probably in the Germanic languages as a whole was borrowed from the Latin. See The Absolute Participle in Anglo-Saxon, pp. 31 ff. APPENDIX I THE ABSOLUTE DATIVE IN THE OTHER GERMANIC LANGUAliKS Since the publication of my monograph on The Absolute Par- ticiple in Anglo Saxon in 1889, in which (pp. 31-36) I gave a brief discussion of the Absolute Participle in the Germanic Ian guages other than Anglo-Saxon, there have appeared a number of works dealing, usually incidentally, with the Absolute Con- struction in the Germanic languages exclusive of English. The more important of these treatises I mention in the following sec- tions as I take up each group of languages. * In the remaining 13 examples, the Latin absolute participle is rendered by a prepositional phrase, 6 times; by an appositive participle, 3 times; by a noun in an oblique case, twice; by an infinitive, once; and once it is omitted. 14 The Syntax of the Lindisfarne Gospels To begin with the Gothic, the noteworthy treatises dealing with the Absolute Dative that have been published since 1889 are as follows : Professor Heinrich Winkler's Germanische Casussyntax, i, Berlin, 1896, pp. 118-140; Professor Berthold Delbriiek's Vergleichende Syntax der Indogermanischen Sprachen, Vol. iv, Part 2, Strassburg, 1897, pp. 495-496 ; Pro- fessors Oskar Erdmann and Otto Mensing's 7 Grundzuge der Deutschen Syntax, Vol. n, Stuttgart, 1898, § 312 ; Dr. M. J. van der Meer's Gotische Casussyntaxis, Leiden, 1901, § 95 ; Dr. Ant. Beer's 8 Kleine Beitrdge zur Gotischen Syntax: I. Der Absolute Dativ (a reprint from the Sitzungsberichte der Konig- Uchen Bohmischen Gesellschaft der Wissenschaften for 1904),' Prag, 1904 ; Dr. H. Stolzenburg's " Die Uebersetzungstechnik des Wulfila Untersucht auf Grund der Bibelfragmente des Codex Argenteus," in the Zeitschrift fur Deutsche Philologie, xxxvn, 1905, pp. 178-179 ; Professor Wilhelm Streitberg's Gotisches Elementarbuch, Dritte und Vierte Verbesserte Aim., Heidelberg, 1910, § 260 ; Professor Joseph Wright's Grammar of the Gothic Language, Oxford, 1910, § 436 ; and Dr. J. M. N. Kapteijn's " Die Uebersetzungstechnik der Gotischen Bibel in den Paulinischen Briefen," in Indogermanische Forschungen, xxix, 1911, p. 330. As before 1889, so in these later discussions, two views as to the dative absolute construction in Gothic are advocated. The one school holds that the construction was independently devel- oped in the Germanic languages, and that the Greek influence was secondary only. Substantially 9 this view is held by Messrs. Winkler, Streitberg, van der Meer, and Beer, 10 and has 7 Though usually listed under both names, as here, the second volume is exclusively by Dr. Mensing. 8 In this article Dr. Beer announced that he was shortly to publish another article on the Absolute Participle in Gothic; if it has appeared, it has escaped me. 9 Slight differences are purposely ignored here. 10 Perhaps I should add here the name of Professor Hans Naumann, who, in his Kurze Historische Syntax der Deutschen Sprache, Strassburg, 1915, p. 14, holds that probably the Dative Absolute was independently devel- The Absolute Participle 15 been succinctly stated by Professor Streitberg, I. c, § 260 : " Der Dativus absolutus. Wir konnen die Entwickhmg der dativischen Partizipialkonstruktion, die man als die absolute bezeichnet, noch deutlich verfolgen. Den Ausgangspunkt bilden Satze wie qimandin pan in garda duatiddjedun imma pai blindans = £\66vti els ttjv oiKiav Trpoar)\dov avra> ol rvs = ela-eXdovri avrw TrpoarjXdev avTa> eKarovrapx ^ M. 8,5. Diese bequeme Konstruk- tion wendet nun der Ubersetzer mit Vorliebe an, urn den griech. Genitiv absol. wiederzugeben. Meist, doch nicht immer ist eine Beziehung des iibergeordneten Verbs zum Subjekt des Gen. abs. vorhanden ; sie fehlt ganz L. 3, 1 Rj 9, lk5, 3. Die Grund- lage dieser Dativkonstruktion mag germanisch sein, da auch das ISTordische sie kennt, vgl. Grimm, Gramm. 4, 1090 Neu- druck; ihre got. Ausgestaltung steht jedoch sichtlich unter griech. Einfluss. Vgl. Liicke S. 19 ff., Winkler 118 ff., v. d. Meer, 216 ff., A. Beer Kl. Beitrage zur got. Syntax (Bohm. Ges. d. W., phil. KL, 1904, Nr. xin)." — It is true that, as Pro- fessor Streitberg states, the dative absolute is found in Old Norse; but, as we shall see in our next section, it is now gen- erally considered an importation into Old Norse from the Latin. The other school, on the contrary, holds that foreign influence (in the Gothic the Greek and in the other Germanic languages the Latin) was paramount, and that the native influence of whatever sort was secondary only ; in a word, that the Germanic languages borrowed the dative absolute from the classical lan- oped in Gothic, but that it was borrowed from the Latin in Old High German. As the grounds of Professor Naumann's belief in the native origin of this construction in Gothic are substantially identical with those of Professor Streitberg, I content myself with quoting the fuller statement of the latter, as given above. 16 The Syntax of the Lindisfame Gospels guages (Greek or Latin). Substantially this view is held by Messrs. Delbriick, Stolzenburg, and Wright; 11 and it is hap- pily stated by Professor Delbriick, I. c, pp. 495 f. : " Der abso- lute Dativ ini Gotischen ist in etwas iiber zwanzig Fallen be- zeugt, z. B. nauhpan imma rodjandin gaggip sums manne = en avrov \a\ovvros epx^rai Tt? Luk. 8, 49., paruh nauhpan duatgaggandin imma gabrak ina sa unhulpa = ert 8e irpoo-- epxo^evov avrov epprj^ev avrov haip,6viov Luk. 9, 42 ; andanahtja pan vaurpanamma . . . oerun du imma = oi/rta? he ? ty 109 (vgl. 1, 299). An solche Vorbilder kann man sich bei dem Versuch, die abs. Genitive widerzugeben, soweit sie Personen bezeichnen, recht wohl angelehnt haben. Anders diirfte es sich mit den zahlreichen Zeitangaben verhalten. Es konnten darin Lokative von Zeitbegriffen vorliegen (vgl. 1, 225) zu denen ein Parti- zipium hinzugefiigt ist. — Eine andere Art, den griech. Gen. abs. wiederzugeben, ist at mit dem Partiz., z. B. jah sunsaiv nauh- panuh at imma rodjandin qam Judas = teal evdecos, ert avrov XaXovvTos, irapaytveraL Mark 14, 43 ; atiddjedun du pamma hlaiva at urrinnandin sunnin = epftovrai eVi to /xvwueiov avarei- \avros tov rjXiov Mark 16, 2. Dieselbe Ausdracksweise liegt nach Grimm 4, 906 vereinzelt in der Edda vor. Offenbar ist es un- richtig zu sagen, die Proposition sei der absoluten Partizipial- konstruktion vorgetreten. Vielmehr hat man in der Wendung mit at einen zweiten Versuch zur Wiedergabe der absoluten Konstruktion zu erblicken, und also wortgetreu (aber freilich nicht sinngetreu) zu iibersetzen: ' bei ihm, als er noch redete,' d. h. so viel als : ' wahrend er noch redete.' ' As in 1889, I believe that Dr. O. Liicke was right in claim- ing that the Absolute Dative in Gothic was borrowed from the Greek ; but I now think, also, that the secondary native influ- ence, as stated by Professor Delbriick, was unduly minimized both by Dr. Liicke and by me. And it has long seemed to me that, as declared by Professor Delbriick, the advocates of the native origin of this construction in Gothic allowed too little weight to the fact that in the Germanic languages as a whole the absolute construction is relatively rare, and that in one (Old Saxon) it is unknown. 2 18 The Syntax of the Lindisfame Gospels On the Dative Absolute in Old Norse, also, several treatises that have appeared since 1889 make noteworthy contributions. The more important are Professors H. Falk and A. Torp's Dansk-N orskens Syntax i Historisk Fremstilling, Kristiania, 1900, pp. 221-223, and Professor M. Nygaard's two works: " Den Lserde Stil i den Norr0ne Prosa " in the S pro glig -Hist o- rishe Studier Tilegnede Prof. C. R. linger, 1896 (= Nygaard 2 ), pp. 157-158, and his Norr0n Syntax, Kristiania, 1906 (= Ny- gaard 3 ), § 229. Two important works dealing with the abso- lute dative in Old Norse that appeared before 1889, but that were then unknown to me are G. F. V. Lund's Oldnordisk Ordfopiingslcere, K0benhavn, 1862, and Professor M. Ny- gaard's " Om Brugen af det Saakaldte Praesens Particip i Oldnorsk," in Aarbpger for Nordisk Oldkyndighed og His- torie for 1879 (= Nygaard 1 ), pp. 203-228. Lund's statement, I. c, § 54, does not make it clear to me whether he considered the absolute dative in Old Norse to be due to foreign influence or not : " Dobbelt hensynsform bruges for at betegne samme f or- hold som den latinske ablativus consequentiae (Madv. lat. sp. § 277) og den grseske dobbelt ejeform, ssedvanlig med tilfojelse af at (hvilket dog nseppe er den oprindelige udtryksmade), dog ogsa uden dette." Professor Nygaard, however, comes out squarely for the Latin influence on the origin of the dative abso- lute construction without a preposition, though he correctly observes that the native influence is shown in the phrase made up of a preposition 12 (usually at) plus the dative of a noun and of a participle. His statement concerning the latter locution is as follows (Nvgaard, 1 p. 207) : " Saerskilt mserkes, at part, stundom f0ies appositivt til et nomen, der styres af praep. at (isser forat udtrykke tid) saaledes, at part, med nominet smelter sammen til et begreb, og udtrykket bliver at over- saette enten ved til nominet i eieform at fole et verbal- substantiv eller ved en tidsssetning, hvori subjektet gjengiver nominet og praedikatet participiet. Dette udtryk, der oprin- u Concerning the analogous West-Saxon idiom of be him lifigendum etc., see The Absolute Participle in Anglo-Saxon, pp. 42-44. The Absolute Participle 19 delig vistnok er i overensstemmelse med sprogets egne analogier, bliver i latiniserende sprogbrug et bekvemt middel til at gj en- give den latinske absolute ablativ." He then cites as examples of this idiom: — Kgs. 174.33: at hanum lifanda (" i hans leve- tid," " medens han lever"); Harb. 58: at uppverandi solu (" medens sol er oppe "). Concerning the first idiom, the real dative absolute and, therefore, without a preposition, Professor Nygaard 1 writes (p. 207) : " Staerkest viser sig den latinske paavirkning, naar udtrykket gaar over til en absolut dativ uden prsep." Of this idiom he gives this illustration: — Eids. 22: f>oat hjun segi skilit millim sin, pa ma hvarki (?eira ser til for- rseSi leita badwm f>eim lifandum. Professor Nygaard's earliest statement concerning this idiom, just quoted, was reaffirmed in 1895, in his " Den Lserde Stil i den Non-one Prosa," p. 158, and in 1906, in his Norr0n Syntax, § 229, Anm. 2, which latter I quote: " Til et nomen, der er styret af en prsep. (isser at), kan prses. part. f0ies i app. for at udtrykke et samtidigt forhold eller en ledsagende omstsendighed. I F. S. 13 forekommer endel udtryk af denne art som tidsbetegnelse. . . . I L. S. 14 ogsaa for at betegne andre forholde. . . . Ved paavirkning af latinsk absolut ablativ bruges i L. S. paa samme maade ogsaa dativ uden prsep." To the same purport is the statement of Professors Falk and Torp, I. c, p. 221. After stating that the dative abso- lute in Gothic is borrowed from the classical languages (Greek), they add concerning Old Norse: "I samme retning gaar old- norskens vidnesbyrd, idet her den absolute dativ uden proposi- tion udelukkende tilh0rer den oversatte litteratur (som de fra oldfransk oversatte ' Strengleikar ') og senere retsdokumenter ; ogsaa her har da vistnok fremmede forbilleder vaeret be- btemmende." If I have dwelt long on the absolute dative in the Scandi- navian languages, it is partly because I gave so little about the construction in these languages in 1889, and partly because, as already stated incidentally in my discussion of the absolute con- 13 F. S.= Folkelig Stil, ' popular style.' " L. S. = Lard Stil, ' learned style.' 20 The Syntax of the Lindisfarne Gospels struction in Gothic, so distinguished a scholar as Professor Streitberg as recently as 1910 declared that the occurrence of the absolute dative in Old iSTorse was one ground for believing that the absolute construction is a native Germanic idiom. With all due respect to Professor Streitberg, it seems to me that, on the contrary, all that we know about the absolute construction in Old Norse tends to show that the dative absolute was not a native Germanic idiom, but was a foreign importation. This conviction will be strengthened, I believe, by a brief con- sideration of the chief treatises dealing with the Dative Abso- lute in High German that have been published since 1889. The more noteworthy of these treatises are V. E. Mourek's Weitere Beitrage zur Syntax des Althochdeutschen Tatian, Prag, 1894, pp. 35-37 ; Professor Heinrich Winkler's Germanische Casus- syntax, i, Berlin, 1896, pp. 118-140 (chiefly on this idiom in Gothic) ; Professor Berthold Delbriick's V ergleichende Syntax der Indogermanischen Sprachen, Vol. iv, Part 2, Strassburg, 1897, pp. 495-497; Professors Oskar Erdmann and O. Mens- ing's Grundzilge der Deutschen Syntax, Vol. n, Stuttgart, 1898, pp. 274-275; Dr. C. W. Eastman's Die Syntax des Dativs bei Notker, a Leipzig dissertation of 1898, pp. 41- 43 ; Dr. J. B. Crenshaw's The Present Participle in Old High German and Middle High German, a Johns Hopkins disserta- tion which, though dated 1893, was not published until 1901, pp. 8-13 ; Professor H. Wunderlich's Der Deutsche Satzbau, 2 nd ed., Vol. i, Stuttgart, 1901, pp. 392-393 ; Dr. W. Manthey's Syntaktische Beobachtungen an N others Uebersetzung des Mar- tianus Capella, a Berlin dissertation of 1903, p. 34 ; Dr. W. Gocking's Das Partizipium bei Notker, a Strassburg i. E. dis- sertation of 1905, pp. 27-33 ; Dr. K. Meyer's Zur Syntax des Participium Praesentis im Althochdeutschen, a Marburg dis- sertation of 1906, pp. 61-65 ; Professor W. Wilmanns's Deutsche Grammatik, Dritte Abteilung, 1. Halfte, Strassburg, 1906, p. 108 ; Professor W. Streitberg's Gotisches Elementarbuch, Dritte und Vierte Verbesserte Aufl., Heidelberg, 1910, pp. 174-175 The Absolute Participle 21 (chiefly on this construction in Gothic) ; and Professor H. Kall- mann's N others Boethius: Untersuchungen uber Quellen und Stil (= Quellen und Forschungen zur Sprcuch- und Cultur- geschichte der Germanischen Vblker, cxxi), Strassburg, 1913, p. 79. 15 As with reference to the Absolute Construction in Gothic, so here with reference to this construction in High German schol- ars fall into two groups. A few scholars of great distinction hold, or seem to hold, that the absolute dative is an idiom native to the Germanic languages as a whole (exclusive of Old Saxon), to Old High German as to Gothic, though in each instance for- eign influence was somewhat felt. This view is held, I take it, by Professor Winkler and by Professor Streitberg, though neither speaks specifically of the construction in Old High Ger- man. But what each says of the native development of the Dative Absolute in Gothic, is equally applicable, with only trifling modifications, to the Absolute Dative in Old High Ger- man ; hence I have put them down as favoring the native devel- opment of this construction in Old High German as well as in Gothic. The grounds for Professor Streitberg's belief have been quoted in full in my discussion of this idiom in Gothic, p. 15 above, and need not be repeated here ; and, as there stated, the view of Professor Winkler is in substantial accord with that of Professor Streitberg. I do not feel sure as to the position of Professor Wunderlich, but the words quoted below, I. c., p. 392, lead me to believe that he leans to the native-origin theory : " Wo das attributive Par- tizip mit Substantiven sich verbindet, die im freien Genetiv oder Dativ dem Satze sich angliedern, entsteht ungezwungen eine Fugung, die dem lat. Ablativ absolutus entspricht ; vgl. die " I regret to say that, on account of the European War, I have not been able to obtain Dr. J. Muhlau's Zur Frage nnch der Ootischen Psalmen- uebersetzung, a Kiel dissertation of 1904; Dr. P. Ebeling's Der Syntak- tische Gebrauch der Pwrticipia in der Kudrun, in a Halle a. S. Program of 1912; and, above all, Dr. J. von Guericke's Die Entwickelung des Alt- hochdeutschen Participiums unter dem Einfluss des Lateinischen, a Konigs- berg i. Pr. dissertation of 1915. 22 The Syntax of the Lindisfarne Gospels Beispiele aus der althochdeutschen Uebersetzerlitteratur in meinen Beitragen zum Notker'schen Boethius, S. 82. In un- serem wahrend des Krieges liegt noch der alte Genetiv wdhr- endes Krieges vor, dem Fiigungen wie in wdhrendem Kriege zur Seite stehen; vgl. Grimm S. 1085 ff. ; vgl. welche, wie ich deutlich verspure, wahrenden Redens bereits in tnir aufge- stiegen ist, Immermann, Miinchhausen 2, 288 (vgl. unten Teil II, Kapitel 1)." Nor do I feel certain as to the position of Professor Wilmanns (I. c, p. 108), although, from his state- ment given below, I suppose that he belongs to the second group, discussed in our next paragraph : " Sehr kraftig hat sich der Gebrauch satzartiger Partizipia im Griechischen und Latein- ischen entfaltet, am freiesten aussert er sich in den absoluten Partizipialkonstruktionen. In den germanischen Sprachen finden sich zu den absoluten Partizipien nur geringe Ansatze ; andere, die sich der Konstruktion des Satzes einfiigen (apposl- tive Partizipia), begegnen haufig zu jeder Zeit, gehoren aber doch mehr der Kunst- und Schriftsprache an und stehen oft sichtlich unter dem Einfluss fremder Originale und Muster." Dr. Mensing, I. c, pp. 273-274, as in the case of Gothic, seems to stand midway between the two schools. The second group of scholars hold that in High German the Absolute Dative was borrowed from the Latin, though, in the view of some of them, possibly slightly helped along by some tendencies within the language itself. To this group belong the remaining scholars mentioned at the beginning of this section. The late Professor Mourek, I. c, pp. 35-37, gives detailed statis- tics of the Absolute Dative in Tatian, and speaks of " der un- deutschen construction des absoluten dativs, die dem lat. abl. abs. sklavisch nachgeahmt, aber ungemein haufig belegt ist." Dr. Delbriick, I. c, p. 495, declares: " Im Altnordischen und Althochdeutschen gilt er ebenfalls als Eindringling." 16 Dr. C. W. Eastman, I. c, p. 41, thus speaks of the Absolute Dative in Notker: " Diese Verbindung des Dativs eines No- 18 The whole passage is quoted ahove, in my section on the Absolute Dative in Gothic. The Absolute Participle 23 mens mit einem Part. Praet. oder Part. Praes. kann keinesfalls als eine echt ahd. Construction angesehen werden,'' a judgment based on a minute comparison of the Old High German text of several of Notker's works with their Latin originals. Dr. J. B. Crenshaw, I. c, pp. 9 and 11, gives some illuminating statis- tics as to the Dative Absolute in Old High German. For the Present Participle in this construction the figures are as fol- lows: Otfrid, 1; Isidor, 2; Tatian, 45; Notker, 65; total 113; of which examples 95 translate a Latin Ablative Absolute. For the Preterite Participle the figures are : Otfrid, 2 ; Isidor, 2 ; Tatian, 71 ; Notker, 28 ; total, 103 ; of which examples only one is original in the German, the remaining 102 being due to Latin influence. Dr. Crenshaw sums up the matter on page 13 : " The results reached by Dr. Morgan Callaway in his thesis, The Absolute Participle in Anglo-Saxon, are accordingly abun- dantly substantiated for the Dative Absolute in German. The construction in question is directly borrowed from the Latin, and is the result of an attempt to graft on to German the Abla- tive Absolute of the Latin. The attempt failed because the trend of German syntax was not in that direction. The con- struction is, therefore, limited to the period characterized by translations from Latin and does not become an integral part of classical German syntax." Dr. Manthey, I. c, p. 34, speaks thus of the construction in Notker's Martianus Capella: " Der Dativus absolutus ist eine genaue Nachamung des lateinischen Ablativus absolutus und steht im M. Cap. a'uch nicht ein ein- ziges Mai selbstandig." Dr. Gocking, I. c, p. 27, gives statis- tics for Notker's Boethius as well as for his Martianus Capella, and likewise declares that the Absolute Dative is borrowed from the Latin. Dr. K. Meyer, I. c, p. 62, quotes approvingly this statement from Rannow : " Auch die Konstr. des dat. abs. mag von aussen eingedrungen und zu gewissen Zeiten, wie schon Ulphilas zeigt, nicht einmal selten in Anwendung gekommen sein, aber voiles Biirgerrecht hat sie in der deutschen Sprachen nie erhalten." Finally, Professor H. Naumann, I. c, p. 79, as recently as 1913 spoke thus of Notker's Latinized style, 24 The Syntax of the Lindisfame Gospels especially as exhibited in his use of the infinitive and of the participle, and in so doing differentiated the earlier stage of Old High German from the later stage thereof, represented by Notker : " In der alten Schule waren der Abl. absol. haufig ver- mieden oder aufgelost, das Partic. conj. gleichfalls oft aufge- lost, beide vielfach auch beibehalten; die Infmitivkonstruk- tienen sind bis auf wenige Ealle vermieden. In der Folgezeit (Exhortatio; Weissenburger Katechismus; — Trierer Kapitu- lare) ist vielleicht eine leise Steigerung der Bewahrungstendenz bemerkbar. JSTotker nun iibertrifft noch die altere Schule; der Ace. c. Inf. ist selten aufgelost; der Abl. abs. sehr oft nach- geahmt, doch sind die Falle, wo er aufgelost ist, immerhin zahl- reicher; das Partic. conj. ward sehr oft beibehalten, oft auch ohne Vorlage angewandt, doch oft auch aufgelost. Dazu kommt die vollendete Hypotaxe, auch ein Merkmal literarischer, vom Latein gelernter Syntax ; desgl. die fast vollige Vermeidung des verbalen und auch des substantivischen Asyndetons, das doch ein Charakteristikum der alteren deutschen volkstiimlichen Redeweise war. tibrigens weichen die Stellen aus Notkers eigener Feder sehr wesentlich von dieser lateinischen Diktion ab." And in his recently published K'urze Histarische Syntax der Deutschen Sprache (1915), p. 14, Professor Naumann is no less pronounced for the Latin origin of this construction in Old High German as a whole : " Der reine Dat. absol. im Ahd. (ohne Proposition) gilt als eine Nachahmung des lateinischen Abl. absol. : bin gote helphante thero arabeito zi ente; bislozanen thinen turin. Diese Konstruktion gehort zum gelehrten Stil der ahd. Autoren ; aber sie blieb ohne Einfluss auf die Zukunft, denn bereichern kann der Stil die Syntax nicht." I have given this rather detailed summary of opinion as to the origin of the Dative Absolute in Old High German bceause I desire to show how many different special investigators of this construction in Old High German have independently come to the same conclusion, that the idiom is borrowed from the Latin. To me the evidence given by these investigators seems con- clusive for the theory of the foreign (Latin) origin of this The Absolute Participle 25 construction in Old High German, although it is still possible that certain native tendencies in Old High German such as those indicated by Messrs. Winkler and Streitberg for Gothic, may have been of secondary, but not, as claimed by them, of primary, help. As to the Middle High German, Dr. Crenshaw, I. c, p. 10, declares that " In Middle High German no clear instance of the construction [the Dative Absolute] is found; only two where the use seems probable." The absolute Dative is not mentioned by Professor H. Paul in his Mittelhochdeutsche Grammatik, 9th ed., Halle, 1913, or by Professor V. Michels, in his Mittel- hochdeutsches Elementarbuch, 2nd ed., Heidelberg, 1912. Nor is the Dative Absolute found in New High German, but only the Accusative Absolute. Concerning the latter, see Blatz, I. c, pp. 354-357, and my section on the Accusative Absolute below. In Old Saxon, no instance is recorded of the Absolute Dative : see Dr. H. Pratje's " Syntax des Heliand, I. Das Verbum," in the Jahrbuch des Vereins fur Niederdeutsche Sprachforschung for 1885, xi, 1886, pp. 74-84, and Professor O. BehaghePs Die Syntax des Heliand, Wien, 1897, §§ 107-108. To sum up the matter : Gothic, Old Norse, Old High German, Middle High German, and Old Saxon, like Old Northumbrian, all point to the foreign origin of the Dative Absolute in the Germanic languages as a whole. B. The Absolute Accusative Several scholars have called attention to the fact that the Accusative is occasionally used Absolutely with a Participle in the Lindisfarne Gospels, as in Mat. 9.18: Sas hine sprecende to him uel Ssem, heonu aldormonn an geneolecde = Haec illo lo- quente ad eos ecce princeps unus accessit (Rushw. and W. S. : sfv.). In 1857, K. W. Bouterwek, in his Die Vier Evangelien in Altnordhumbrischer Sprache, Giitersloh, p. cv, cited four 26 The Syntax of the Lindisfarne Gospels examples 17 of the idiom; in 1866, George Waring, in the " Prolegomena " to The Lindisfarne and Rushworth Gospels, 4 vols., Durham, 1854-1866, independently cited one example (L. 8.49: illo loquente = hine sprecende), and commented: " Here the use of hine as a dative is to be noted " ; and, in 1899, Jacob Schipper, in his Konig Alfreds Uebersetzung von Bedas Kirchengeschichte, Leipzig, 1897-1899, p. xliii, quoted two other examples. 18 But no one of these three scholars seemed to be aware of the frequency of this idiom in Northumbrian ; nor has the fact of the occurrence of the idiom in this dialect been generally incorporated in the treatises dealing with Anglo- Saxon syntax. I regret to add that the fact was not known to me when I wrote my dissertation on The Absolute Participle in Anglo-Saxon (West-Saxon), Baltimore, 1889; indeed, I did not learn the fact until several years ago, when I began these studies in the syntax of the Northumbrian dialect. About twenty-one examples 19 of this idiom have been found in the Northumbrian, a construction unknown in the West- Saxon, 20 for, as I have tried to show in my dissertation, the cases hitherto cited as examples of the absolute accusative in West-Saxon are to be explained otherwise, — a conviction in no wise upset by my later discovery of the use of the absolute accusative in the Northumbrian. The number of examples found of this idiom precludes one's first impression that the use of the accusative in the gloss is due merely to carelessness. " Mat. 8.34: uiso eo — gesene hine; L. 20.1: docente illo = Icerende hine; ib. 22.47, 60 : eo loquente = hine sprecende. ™J. 4.51: eo descendente — hine stigende; ib. 8.30: haec illo loquente — Sas hine spreccende. 19 In all of these examples except one (L. 23.8"), the subject of the participle is a pronoun, and the form of the subject is indisputably accu- sative. In L. 23. 8" the noun subject (hcelend) is preceded by the accusa- tive of the article, tione. 20 In the third edition of his Historische Syntax, 1916, p. 59, Professor Eugen Einenkel expresses the belief that, in late Old English (Anglo- Saxon ) , a few genuine cases occur of the Absolute Accusative, but he does not cite any example of this idiom in Anglo-Saxon that seems certain to me. The Absolute Participle 27 That the accusative here is a genuine Northumbrian idiom seems indicated, also, by several other facts. (1) Once the dative and the accusative are given as alternatives in the abso- lute construction, in Mk. 14.43 : Sa get uel (5a geon him uel hine sprecende cuom iudas = athuc eo loquente uenit iudas (W. S. : dat. abs. ; Rushw. : dat. abs. (crude)). (2) The accu- sative is used in several other constructions in which normally (in West-Saxon, at least) the dative occurs, as (a) after the comparative degree of adjectives; (b) after the preposition mid (mid), as has been long known; (c) after the verb to be for the dative of interest; (d) after sella for the dative of the indirect object (or of interest?) ; (e) after gedcefniga for the dative of reference. Conversely, (/) after geceiga, ' call,' which in West- Saxon has the accusative only, we find occasionally the dative, though usually the accusative. Examples are : — (a) L. 3.16 : cymeS Sonne strongra mec (sic) = ueniet autem fortior me; — J. 14.28: forSon se faeder mara mec (sic) is = quia pater maior me est. [Or is the accusative here due to the Glossator's mistaking me, ablative, for me, accusative ?] (b) J. 8.29: seSe mec sende mec (sic) mid is = qui me misit mecum est. [Concerning the regimen of mid, see Erla Hittle, Zur Geschichte der Altenglischen Prapositionen "Mid" und " Wid" ( = Anglistische Forschungen, Heft II), Heidel- berg, 1901, pp. 6-7. Dr. Hittle refers to Miller's Bede, and declares that the accusative with mid is a chief characteristic of the Northumbrian (Mercian) dialect. (c) J. 21.22 : cueS him to se haelend Sus uel suae hine ic uillo geuuni uel Saette he gewuniga oS Saet ic cymo huaed is Se bi Sy uel is dec (sic) Sees? Su mec soec uel fylig Su me = dicit ei iesus sic eum uolo manere donee ueniam, quid ad te? tu me sequere. — For a striking parallel to this in Old High German, compare Tatian 45.2: waz ist thih thes inti mihf = quid mi hi et tibi est? and see the comment thereon by Arthur Kohler. " Ueber den Syntaktisehen Gebrauch des Dativ im Gotischen,'' in Germania, xi, 1866, p. 288. (d) Mat. 26.67: Sa speafton in onsione his & — [ = blank] 28 The Syntax of the Lindisfarne Gospels hine slogun o<5ro Sonne hondbreodo in onsione hine (sic) sal- don = Tunc expuerunt in faciem eius et colaphis eum caederunt alii autem palmas in faciem ei dederunt. (e) J. 3.30 hine uel him gedcefnad Ssette auexe mec uutud- lice (5set ic lytlege = ilium oportet crescere me autem minui (Rushw. : him gidcefnad etc. ; W. S. : hit geburaS (5set he weaxe etc.). The accusative is doubtless partly due to the presence of the accusative (ilium) in the Latin. (/) L. 19.15 : heht geceiga (5sem esnum uel (5a esnas Ssem gesalde <5aet feh = iussit uocari seruos quibus dedit pecuniam ; L. 14.13 : ah miSSy Su doest gebserscip geceig dorfendum un- lialum haltum blindum = sed cum facis conuiuium uoca pau- peres debiles clodos caecos; L. 15.6: cuom to hus uel to ham geceiged uel geceigde friondum & neheburum = ueniens domum conuocat amicos et uicinos. In twelve other citations for geceiga given by Professor Cook, it governs an accusative: Mat. 1.21, 23; 9.13; 10.25; 23.9 ;—L. 1.13, 31, 62; 14.9, 12;—/. I. 5.3; 9.18; 13.13; and in five others, it has no object: Mat. 8.29; 14.26 ;—MJc. 15.13, 14;— L. 1.42. In a word, there seems to have been an interchange of the accusative and the dative in Northumbrian that is unknown in West-Saxon ; of which interchange the accusative absolute offers the most striking illustration. After writing the foregoing paragraph concerning the inter- change of dative and accusative in the Lindisfatyie Gospels, I came upon this statement by the late Dr. Henry Sweet concern- ing the confusion of cases in the Northumbrian and Mercian dialects : " The peculiar feature of the Northumbrian and Mercian dialects is their combination of archaism and disinte- gration, which can only be compared with that of Southern English in the twelfth century. We find the same confusion of genders (dees moehtes, pi. burgas, etc.), of cases (heom ace. in Rush.), of strong and weak (dees lichomes) ." 21 Of the falling 21 The quotation is taken from Dr. Sweet's " Dialects and Prehistoric Forms of Old English," which originally appeared in the Transactions of the Philological Society, 1875-1876, pp. 543-569, and was subsequently re- The Absolute Participle 29 together of certain cases of nouns in Northumbrian, Professor Uno Lindelof tells us in his " Beitrage zur Kenntnis des Alt- Northumbrischen," Helsingfors, 1893. On p. 299 he speaks as follows of the Feminine Nouns: " Im Singular fallen der Nom- inativ und der Accusativ formell zusammen," probably out of analogy to masculine and neuter nouns, he explains. He then declares that, although the Dative ending is usually preserved in Feminine Nouns, " Es sind aber doch einzelne Spuren von Zusammenfall des Dativs mit dem Nominativ und Accusativ vorhanden, ein Vorgang, der sich in der Entwickelung des Nordenglischen jedenfalls bald vollzog." And it is well known, of course, that in the Modern Scotch dialects there is great con- fusion of cases in the pronouns. In his The Dialect of the Southern Counties of Scotland, London, 1873, p. 187, Dr. J. A. H. Murray speaks of the matter as follows : " The usage of the Personal Pronouns in the current Scottish dialects differs essen- tially from that of the Standard English, being in most respects identical with the French. There is a direct or proper Nomina- tive, and a direct Objective, as well as an indirect case, used like the French moi, toi, lui, eux, for both Nominative and Objective in certain positions. But while in French this indirect case or dative is in its history and derivation distinct from the direct accusative, the indirect case in Scotch is, viewed etymologically, really the objective of the English (the dative or accusative of the Anglo-Saxon), while the direct Objective is a contracted or mutilated form." And, on p. 189, he tells us that " The Indi- rect form is used for the Nominative (1) when the Verb is not expressed, as in answer to a question (so in French) ; (2) when the Nominative is separated from the Verb by a Relative or Relative Clause, a numeral or a substantive (so in French) ; (3) as the second Nominative (predicate) after the verb to be (so in French) ; (4) when the Nominative is repeated for the sake of emphasis, the added nominative being put in the indi- printed in the Collected Papers of Henry Sweet, Oxford, 1913, pp. 185-211. The passage given above I have quoted from the Collected Papers, p. 197. 30 The Syntax of the Lindisfame Gospels rect case (so in French) ; (5) when two or more nominatives form the subject of the same verb (so in French) ; (6) with a participle as the absolute case." This statement of Dr. Murray is confirmed by the later investigations in this general field, as by Dr. G. H. Cowling, in The Dialect of Hachness (North-East Yorkshire), Cambridge, 1915, p. 120, and by Sir James Wil- son, in Lowland Scotch as Spoken in the Lower Strathearn Dis- trict of Perthshire, Oxford, 1915, pp. 83-86. Whether or not Dr. Murray intends to attribute the interchange of cases in Scotch to French influence, is not clear to me. Dr. Otto Diehn, in his Die Pronomina im Fruhmittelenglischen, Heidelberg, 1901, p. 50, attributes the occasional interchange of dative and accusative forms of pronouns in Early Middle English partly to the analogy of nouns (in which dative and accusative had be- come identical) and partly to Scandinavian (especially Danish) influence. Concerning the latter factor he mentions an oral sug- gestion of Professor Sarrazin, and refers to Professor Jesper- sen's Progress m Language, London, 1894, pp. 182 ff., in which latter a general suggestion is made as to the possible influence of the Scandinavian languages on the English in the matter of interchange of case-forms. A very violent substitution in Mod- ern Scandinavian, well known but not mentioned by Professor Jespersen, is the use of the accusative (den) of the article for the nominative (der). In his later work, Growth and Structure of the English Language, 2 nd ed., Leipzig, 1912, §§ 80-81, Pro- fessor Jespersen returns to the topic, but adds nothing of impor- tance. Concerning the confusion of the nominative and of the dative of personal pronouns in Middle English and in Modern English, see, also, Professor F. B. Gummere's interesting arti- cle, " On the English Dative-Nominative of the Personal Pro- noun," in The American Journal of Philology, iv, 1883, pp. 283-290. The examples of the Absolute Accusative are in full as fol- lows (21) :— The Absolute Participle 31 I. THE PRESENT PARTICIPLE (18) The Accusative of the Absolute Present Participle ends nor- mally in -nde and once in -end, and is found in the singular, masculine, only. Compare Carpenter, I. c, §§ 527, 532, and 535. bodiga, preach (1) : L. (1) : 20. l b : kerende hine (5set folc in temple & bodande efnecuomon (5a alldormenn = docente illo populum in templo et euangelizante conuenerunt principes. [Rushw. : cfv. ; W. S. : sfv. — Cook: pres. p.; Carpenter: not cited.] fara, go away, depart (2) : Mh. (1) : 10.46: mi(5 (5y foerde he uel hine fcerende in 8a burug . . . blind gesaet = profi- ciscente eo hiericho . . . caecus sedebat iuxta uiam. [Rushw. : cfv.; W. S. : cfv. — Cook: pres. p.; Carpenter: not cited.] — L. (1) : 19.36 : fcerende Sonne hine underbraeddon uel legdon gege- relo hiora on woege = eunte autem illo substernebant uesti- menta sua in uia. [Rushw.: fcerende Sonne hine; W. S. : sfv. — Cook: pres. p.; Carpenter: not cited.] ofstiga, ascend (1) : Mat. (1) : 8.23: & ofstigende hine uel Sa he ofstag in lytlum scipe . . . gefylgdon hine uel him Segnas his = Et ascendente eo in nauicula secuti sunt eum dis- cipuli eius. [Rushw. : & (5a stag he ; W. S. : & he astah. — Cook: pres. p. ; Carpenter: not cited.] rsesta, recline (1): Mat. (1): 9.10: & geworden wses rcestende hine in hus . . . monigo . . . cuomun geraestun = Et factum est discumbente eo in domo . . . multi . . . uenien- tes discumbebant. [Rushw. : cfv. ; W. S. : sfv. — Cook : pres. p. ; Carpenter: not cited.] spreca, speak, declare (10) : Mat. (3) : 9.18: (5as hine sprec- ende to him uel Saem heonu aldormonn an geneolecde = Haec illo loquente ad eos ecce princeps unus accessit. [Rushw. : (5a he Sis sprsec; W. S. : sfv. — Cook: not cited; Carpenter: not cited.] Mat. 17.5 a : (5a gett uel geana hine sprecende uel forSor he waes sprecende heonu wolcen leht oferscyade hia = athuc eo loquente ecce nubis lucida obumbrauit eos. [Rushw.: sfv.; W. S. : dat. 32 The Syntax of the Lindisfarne Gospels abs. — Cook: not cited; Carpenter: not cited.] Mat. 12.46 a : geonae hine uel he spraec uel sprwcend to menigom heonu moder his & bro(5ero stondas uel gestodon = Athuc eo loquente ad turbas ecce mater eius et fratres stabant foris. [Rushw. : sfv. ; W. S. : sfv. — Cook: pres. p. ; Carpenter: not cited.] — Mk. (2) : 5.35 a : (5a geone hine uel he sprecende cnomon . . . aldermenn = adhuc eo loquente ueniunt etc. [Rushw. : (5a geona he sprec- ende comon etc. ; W. S. : dat. abs. — Cook : not cited ; Carpenter : not cited.] Mk. 14.43: Sa get uel (5a geon him uel hine sprec- ende cuom iudas = athuc eo loquente uenit iudas. [Rush.: dat. abs. (crude) ; W. S. : dat. abs. — Cook: not cited; Carpen- ter: not cited.] — L. (4): 8.49: Sageane hine spreccende — (= blank) from aldormonn somnunges cuoeS him = athuc illo loquente uenit a principe synagogae dicens. [Rushw. : geona hine sprecende com . . . cwae(5 ; W. S. : dat. abs. — Cook : not cited; Carpenter: not cited.] L. 9.34 a : Sas (5a hine sprecende aworden waes wolcen = haec autem illo loquente facta est nubis. [Rushw.: sfv. ( ?) ; W. S. : sfv. — Cook: not cited; Carpenter: not cited.] L. 22.47 : forSor (5aget hine spreccende heono (5a?t here uel (5a menigo = Athuc eo loquente ecce turba. [Rushw. : to him sprecende heono (5e here & seSe giceged waes iudas; W. S. : dat. abs. — Cook: not cited; Carpenter: not cited.] L. 22.60 : sona for(5or (5a get hine sprecende gesang se hona = con- tinue athuc illo loquente cantauit gallus. [Rushw. : sona for(5or Sagett hine sprecende Se hona gisang; W. S. : sfv. — Cook: not cited; Carpenter: not cited.] — J. (1): 8.30: (5as hine sprec- cende uel mi(5(5y he wses sprecende menigo gelefdon on him = haec illo loquente multi crediderunt in eum. [Rushw. : (5as hine sprecende monige gilef dun in hine ; W. S. : sfv. — Cook : pres. p. ; Carpenter: not cited.] Stiga, descend (1) : J. (1): 4.51 a : gee uutudlice hine sti- gende uel soSlice mi(5(5y Se geade . . . gwurnun him togasgnes = iam autem eo descendente serui occurrerunt ei. [Rushw. : gisceh wutudlice hine stigende esnas giurnon togsegnes him; W. S. : sfv. — Cook: pres. p.; Carpenter: not cited.] wyrca, work, do (1) : Mat. (1) : 6.3: (5u uel Beh uutedlice The Absolute Participle 33 wyrcende Sa aellmissa nyta winstra Sin huset wyrcas uel doas suiSra Sin = te autem faciente aelemosyna nesciat sinistra tua quid faciat dextera tua. [Rushw. : dat. abs. ; W. S. : sfv. — Cook: not cited; Carpenter: not cited.] NOTES 1. An Abortive Attempt at the Absolute Accusative occurs possibly in M at. 24.3 : wees sittende uutedlice he uel hine ofer mor oleuetes geneolecdon to him Segnas = Sedente autem eo super montem oliueti accesserunt ad eum discipuli (Rushw. : saet Sa he ; W. S. : Da he saet) . II. THE PRETERITE PARTICIPLE (3) The Accusative of the Absolute Preterite Participle ends in -ne, and is found in the singular, masculine, only. Compare Carpenter, I. c, § 549. (ge)sea, see (3) : Mat. (1) : 8.34: gesene hine uel Sa hine gesegon gebedon Sset ofereade = et uiso eo rogabant ut transiret. [Rushw. : & geseende lime bedun hine Saet he ferde ; W. S. : sfv. — Professor Cook considers gesene,, here and in the two pas- sages quoted below, an adjective (and inadvertently as ns.). It may be an adjective, but, if so, it is probably accusative singu- lar masculine, and we have an absolute phrase made up of an adjective plus a pronoun instead of a participle plus a pronoun. As, however, according to Professor Cook, at least three clear examples (Mat. I. 8.17, 25.29 ; L. 22.24) occur of geseen as the past participle in the nominative singular, I do not see why gesene may not here be considered a participle instead of an adjective. Dr. Carpenter, I. c, § 480, considers gesene an adjective, but assigns no case to it in this passage.] L. (2) : 10.31: gesene hine biwaerlde = uiso illo prseteriuit. [Rushw.: lacking; W. S. : sfv.] L. 23. 8 a : herodes Sonne gesene Sone licelend glsed wees suiSe = herodes autem uiso iesu gauisus est ualde. [Rushw. : herodes Sone gesene Sone licelend gla?d wses swiSe; W. S. : sfv.] 3 34 The Syntax of the Lindisfarne Gospels ORIGIN OF THE IDIOM That the Absolute Construction with the Accusative is due to the influence of the Latin original, seems likely from the fact that in each instance it occurs in translation of a Latin ablative absolute ; or, rather, the absolute construction is due to the Latin influence, the fact that the accusative is so used seems due to the Northumbrian interchange of dative and accusative dis- cussed above. Finally, it should be added that the glossator at times gives as an alternative rendering the native idiom of a subordinated finite verb (as in Mat. 8.23 : ofstigende hine uel &a he of stag = ascendente eo) or of a co-ordinated finite verb (as in Mat. 17.5 a : Sa gett uel geana hiiie sprecende uel forSor he ivo3s sprecende = athuc eo loquente), — a fact that further attests the ungenuineness of the absolute construction. APPENDIX II THE ABSOLUTE ACCUSATIVE IN THE OTHEE GERMANIC LANGUAGES In his Gotisches Elementarbuch, 3 rd and 4 th edition, Heidel- berg, 1910, §251, Professor Streitberg speaks as follows of a possible Absolute Accusative in Gothic: " Einen Akkusa- tivus absolutus nimmt man an auf Grund von M. 6.3 puk tau- jandan armaion, ni witi hleidumei peina hva taujip taihsiuo peina = crov 8e ttoiovvtos eXevfioa-vvvv und Mc. 6.22 atgag- gandein inn dauhtar Herodiadins jah plinsjandein jah galei- kandein Heroda jah paim mipandkumbjandam, gap piudans du pizai maujai = elae\6ovcrn 'HpcoSr) icai Tot? avvava/ceifAevois, elirev 6 fiacri- \ev? ra Kopaaiu>. Im zweiten Beispiel stellt man gewohnlich durch die Anderung *dauhtr einen Dativ absol. her ; im ersten lasst man den Akkusativ von witi abhangen, doch befriedigt diese Aushilfe nicht. Vgl. Riickert S. 416, Lucke S. 26 f." Despite this statement of Professor Streitberg, I am loth to con- sider these two passages genuine instances of the Absolute Accu- sative, not only because their genuineness has, as stated by Pro- fessor Streitberg, been questioned by several eminent Gothic The Absolute Participle 35 scholars, but also because the Absolute Accusative is most proba- bly not a native idiom in the other Germanic languages. See, further, H. Winkler, Germanische Casussyntax, p. 124, and M. J. van der Meer, Gotische Casussyntaxis, § 38. Concerning the Absolute Accusative in New High German, which, according to Blatz, is most probably not a survival of the dative absolute, but a new importation from the French, see F. Blatz, N euhochdeutsche Grammatih, 3 rd ed., Karlsruhe, 1895-1896, Vol. n, pp. 354-357 ; T. Matthias, " Zur Geschichte der Deutschen Mittelwortfiigungen," in the Zeitschrift fur den Deutschen Unterricht, xi, 1897, pp. 694-705; W. Wil- manns, Deutsche Grammatih, Dritte Abtheilung, 1. Halfte. Strassburg, 1906, p. 108 ; and the references given in The Abso- lute Participle in Anglo-Saxon, p. 33. For the Absolute Accusative in the Low Germanic languages, see J. Verdam, " Absolute Naamvallen in 't M. Ndl. en Ndl.," in Tijdschrift voor Nederlandsche Tool- en LetterJcunde, n, 1882, pp. 188-198; W. L. van Helten, " Bijdragen tot de Dietsche Grammatical viii. Nog iets over de Absolute Naam- vallen," in the same journal, v, 1885, pp. 207-220 ; and F. A. Stoett, Middelnederlandsche Spraahhunst: Syntaxis, 2 nd ed., 1909, pp. 127 ff. (in the 1889 ed. pp. 34 ff.). Dr. Stoett tells us that this construction in Middle Dutch is borrowed from the French. With this statement compare J. W. Bright, " The Objective Absolute in English," in Modern Language Notes, v, 1890, pp. 80-81 ; C. H. Ross, The Absolute Participle in Middle and Modern English, a Johns Hopkins dissertation of 1893 ; and Eugen Einenkel, Historische Syntax, Strassburg, 1916, pp. 58-60. C. The Absolute Nominative A few sporadic examples (about ten in all) occur of what appears to be an Absolute Construction in the Nominative Case, an idiom unknown in West-Saxon, as I have tried to show in my aforementioned dissertation. A typical example is found in Matthew 1.20 : Sas soSlice <5e he dencende uel tSohte Sa cuom 36 The Syntax of the Lindisjarne Gospels heno engel drihtnes in suoefnum uel in slepe setdeaude him = haec autem eo cogitante ecce angelus domini in somnis apparuit ei (W. S. : dat. abs. ; Rushw. : cfv.). In these Northumbrian examples it is difficult to decide whether the nominative is due merely to errors in case-relations (an error of a sort not infre- quent in these gospels), or whether the glossator thought the nominative grammatically allowable. Personally I am inclined to believe that the nominative is due to contamination, or, bet- ter, to the mixture of two constructions. In several of the ex- amples, as in the one just quoted, the glossator gives two alter- native constructions, (1) the finite verb and (2) the participle, the former calling for a nominative and the latter for an oblique case, but, in these sporadic instances of the nominative absolute, he gives us — in all but three examples — the nominative only. In these three exceptional instances we find the nominative in- terchanging with the accusative: in Mat. 6.3, 12.46, and Mk. 5.35, all quoted below and also under the Absolute Accusative. In nine of the examples, the Latin original has the ablative ab- solute; in one (Mk. I. 5.3) the Latin has an appositive dative of the participle, but the glossator has mistaken the dative of the noun (scriboe) for the nominative, plural. The examples of the Nominative Absolute are in full as follows : — I. THE PRESENT PARTICIPLE (6) The Nominative of the Absolute Present Participle ends nor- mally in -nde, occasionally in -nd. Compare Carpenter, I. c, §§ 527, 532, 534, and 538. fregna, enquire (1) : Mk. (1) : I. 5.3: da boecere of bod aes jregnende twufald lufes sedeawde = Scriba? de mandato legis interroganti geminum dilectionis osteudit. [Cook: jregnende. pres. p. ; boecere, npm. Carpenter, I. c, § 293, makes boecere nominative, plural, masculine, but does not cite jregnende. So far as its form is concerned, boecere could be a dative, singu- lar; and, of course, jregnende could be a dative-instrumental, singular, masculine; accordingly, at first I put this example The Absolute Participle 37 under the appositive use. But, as boecere is preceded by da, the noun was probably intended for the plural by the glossator, who mistook scribes for the nominative, plural, instead of the dative, singular ; I do not feel sure as to what case he intended fregnende to be in. geonga, go (1) : Mat. (1) : 9.27: & geongende uel $a geeode Sona (5e hoelend gefylgdon hine uel him tuoege bisene uel blinde — Et transeunte inde iesu secuti sunt eum duo caeci. [Rushw. : cfv. ; W. S. : sfv. — Professor Cook does not classify noun or participle here, nor does Dr. Carpenter.] spreca, speak (2): Mat. (1) : 12.46 a : geonse hine uel he spraecc uel sprcecend to menigom heonu moder his & broSero stondas uel gestodon = Athuc eo loquente ad turbas ecce mater eius et fratres stabant foris. [Rushw. : sfv. ; W. S. : sfv. — Cook: pres. p.; Carpenter: not cited.] Mk. (1): 5.35 b : 8a geone hine uel he sprecende cuomon . . . aldormenn = adhuc eo loquente ueniunt etc. [Rushw. : (5a geona he sprecende comon. — Cook: not cited ; Carpenter: not cited.] Senca, think (1): Mat. (1): 1.20: Sas soSlice Se he cfen- cende uel Sohte (5a cuom heno engel drihtnes in suoefnum uel in slepe aetdeaude him = haec autem eo cogitante ecce angelus domini in somnis apparuit ei. [Rushw. : cfv. ; W. S. : dat. abs. — Cook: pres. p.; Carpenter: not classified.] wyrca, work, do (1) : Mat. (1) : 6.3: flu uel Seh uutedlice wyrcende (5a aellmissa nyta winstra Sin huset wyrcas uel doas suiSra (5in = te autem faciente aelmosyna nesciat sinistra tua quid faciat dextera tua. [Rushw. : dat. abs. ; W. S. : sfv. — Cook: not classified; Carpenter: not classified.] II. THE PRETERITE PARTICIPLE (4) The Nominative of the Absolute Preterite Participle ends, in the singular, in -d or in -en; in the plural, in -do or in -de. eftasaecga, say, relate (1): L. (1): I. 5.4: soecende of niming hera in sabatum eftascegd dauid bisen forcuom = Quaerentes de uulsione spicarum in sabbatis relato dauid exem- 38 The Syntax of the Lindisfarne Gospels plo conuincit (or predicate nominative?). [Cook: eftascegd, pp. ; bisen, nsfn. ; Carpenter: not cited.] forleta, leave (1): L. (1): 10.30 b : fromfoerdon half cwic uel lifigiende forleten = abienmt semiuiuo relicto. [Rushw. : lacking; W. S. : cfv. — Cook: forleten, pp. ; half-cwic, " asm. ? " Carpenter: not cited.] (ge)ceiga, call (1) : Mk. (1) : 3.23: & efne geceigdo 6a ilco uel miSoy geceigd weron o"a ilco in bispellum cuoeS he = conuocatis eis in parabolis dicebat. [Rushw. : & efne gicegde da ilcv in bispellum cwEeo" ; W. S. : cfv. — Cook : efnegeceigdo, npm. ; ilco not cited for this passage; Carpenter, I. c, § 554: efnegeceigdo, npm.] (ge)claensiga, cleanse (1) : L. (1): I. 9.8: tea uel teno hreafo geclcensade an ... to (5ongunge gewoende = Decern leprosis mundatis unus . . . reuertitnr. [Geclcensade may be indicative preterite, 3 rd sg., as Professor Cook holds. Carpen- ter: not cited.] ORIGIN OF THE IDIOM As already incidentally indicated, the Absolute Nominative is in each instance suggested by the ablative absolute of the Latin original, and seems to arise from a mixture of constructions, the glossator wavering in his rendering of the absolute parti- ciple of the Latin between a finite verb, which requires a nomi- native as its subject, and an absolute participle, which requires an oblique case. APPENDIX III THE ABSOLUTE NOMINATIVE IN THE OTHER GERMANIC LANGUAGES By scholars of no slight distinction it has several times been claimed that we have two examples of the Absolute Nomi- native in Gothic, namely, in Mark 6.21: jah waur(?ans dags gatils = ical yevone'vvs r)fie'pa<; evicaipov and in John 11.44: urrann sa dau^a gabundans handuns jah fotuns f ask jam jah wlits is aural j a bibundans — /cal 97 oyfns avrov ire- pieSeSero. Although the absolute nominative is contended for here, among others, by so distinguished a scholar as Professor The Absolute Participle 39 Streitberg (in his Gotisches Elementarbuch, 3 rd and 4 th ed., Heidelberg, 1910, § 244), and as Professor Einenkel (in his Historische Syntax, 3 rd ed., Strasburg, 1916, p. 58), it seems to me that Dr. W. van Helten, in his " Gab es Einen Got. Nomi- nations Absolutus?" (in Paul and Braune's Beitrdge zur Ge- schichte der Deutschen Sprache und Literatur, xxxv, 1909, pp. 310-311), has shown that in each passage we probably have some sophistication of the text, and that Ulfilas probably used a finite verb in each. Nor does Dr. Ant. Beer's effort, in his " Gab es Einen Gotischen Nominativus Absolutus? " (in the same peri- odical, xxxvn, 1911, pp. 169-171), to overthrow the theory advocated but not originated by Dr. van Helten seem to me suc- cessful. For other discussions of these two passages, by Gabel- entz and Loebe, by Uppstrom, Bernhardt, Riickert, Gering, and Liicke, see the references given by Professor Streitberg, by Dr. van Helten, and Dr. Beer; also M. J. van der Meer, Gotische Casussyntaxis, § 5, b; and The Absolute Participle in Anglo- Saxon, pp. 31-36. Concerning alleged examples of the Nominative Absolute in New High German, see F. Blatz, Neuhochdeutsche Gram- rnatih, u, pp. 310-311, and T. Matthias, " Zur Geschichte der Deutschen Mittelwortfiigungen," pp. 705-708. On page 40 I give a synoptic table of the Absolute Participle in the Lindisfarne Gospels, showing the approximate number of examples in the Dative-Instrumental, the Accusative, and the Nominative. D. The Uses of the Absolute Participle As in West-Saxon, the chief use of the absolute participle in the Lindisfarne Gospels is to denote Time, as in Mat. 28.13 b : lorstelun uel stelende weron hine us slependum = furati sunt eum nobis dormientibus ; Mat. 9.18: Sas hine sprecende to him uel Ssem heonu aldormonn an geneolecde = Haec illo loquente ad eos ecce princeps unus accessit (Rushw. : Sa he Sis spraec). Almost a hundred examples of this use have been found. At times the absolute phrase seems to denote an action almost 40 The Syntax of the Lindisfarne Gospels (3 a Ph M o >— H | H a o x pq w I H fe O pq o r-l H Ph | "5 ■*-> o ■ V m ■- gPn irt im « HWIO (M r-l 10 05 tc tH r- «© rH lO CM 00 f rH Total Pret. Participle Tjt 00 «o rH OS t- (M ■«»< 00 «o 00 CM tH OlOlO •«»< CM rH Total Pres. Participle rH ■* C- +3 03 | a o rC s-. 0! r-t rH • «HN Tf CM CM EQ o u 0i Tf "* • (M rH r-l toian > cS w s 5 o <1 o> Ph rH i-H • (M CM • co eo • n 9 U Pi a 5 1— 1 Pi o Pi eo t- eo r-i oo ce cm * Oi U5 -«ti CO CM rH CO O CO CO CM rH a o> h Ph * rH ■* tr- i-l t«U5N CONH o • o r-l • r-l CM CO CR «5 CO CM 3 B PL, o a c a i o 4-1 c .1 1 a c tr c C M h CO 35 a o a c a i o M 3 .1 3 a c cr i ji o 1-5 U I- 3 "5 c a c C 3 o i <5 The Absolute Participle 41 Co-ordinate with that of the principal verb, a use of the abso- lute participle corresponding to what I have denominated the co-ordinate use of the appositive participle. 22 Possible exam- ples are seen in Mk. 16.20 b > c : bodadon eghussr drihtne mid- wyrcende & Saet word trymende miS fylgendum becenum uel tacenum = praedicauerunt ubique domino co-operante et ser- monem confirmante sequentibus signis. Other examples ob- served are: Mat. I. 21.4; I. 19.12 b ;— L. I. 8.5 b ; I. 10.2 b ; 6.20; 10.30b; 24.47;—/. I. 1.13; I. 6.13; I. 8.3 b ; I. 5.6; I. 6.9 b ; I. 8.3 a . Occasionally the phrase denotes Manner, as in Mat. 17.14: geneolecde to him monn cneum gewcelteno before hine cue(5 = accessit ad eum homo genibus prouolutis ante eum dicens. For other examples (all that I observed) see Mat. I. 20.20 c ; I. 17.12; 27.29 a ;— Mk. 10.17; 12.20 a ;— L. 24.50. At times the phrase denotes the Cause of the main action, as in L. I. 5.10: Srsel gemeS biddendum fore hine celdrum & megum = senium curat rogantibus pro eo senioribus et amicis. See, too, Mat. I. 18.18; 15.12;— L. 9.43;—/. I. 5.3 a : 6.18 (or means?). Rarely the phrase denotes Means, as in Mat. I. 8.10: set- eawas fewr ana uel noht mar a se reht godspellas to onfoanne & alle wiSerweardra gedwola deada swiSor slitenum Son ciricen- dum hlifienddum singendum = ostenditur quattuor tantum de- bere euangelia suscipi et omnes apocriforum nenias (sic for uenias?) mortuis magis hereticis quam ecclesiasticis uiuis ca- nendis. See, also, Mat. I. 16.3 b ; — L. I. 5.4. Rarely, too, the phrase denotes Concession, as in J. 20.26: cuom se hselend bityndum durum = uenit iesus ianuis clausis. See, too, Mat. I. 19.1 ( ?) ;— L. I. 9.8. The table below shows, approximately, the distribution of these several uses in the four Lindisfarne Gospels. As to the classification of individual examples, of course, there is no little room for difference of opinion. ■ See The Appositive Participle in Anglo-Saxon, pp. 285-289. 42 The Syntax of the Lindisfarne Gospels H O oo PQ < I O s p w W H O •< H O >— i H Ph O ■m o Eh CO 3 Ph iH rH OS 00 • rH Tf OJ -. Ph iH tH CO T»< CM iH CM t- CO «M rH rH tp t> • f CO CM O CM • rH i-{ to oo S 09 4-> o Eh CM CM • CM CM • MM ■ • to Ph CNJ CM • 1-H 1-H • ■ co Ph CO CM ■ rH rH • CO 03 i 9 s -t-» o Eh CO CO ■ i-i • rH CO Ph t4 • rH i-t CM CM • *# CM r-t NHMC0 rH rH LO CO OS • ■* CO CO ■* CM Ph 00 C- rH • t- CO • rH o oo co • rH U5 CM (4 Ph CO CO CO ■* i-H O lO CO CNJ rH lOOOt- • CO CM tO tp CM to DO (V O a o O -4-> o rH rH rH • rH l-t l-f • CO -•J Ph rH • • rH i-t i-i • CN Ph t-H iH i-< •a u o i o U o Eh IM CM eo eq • • US - tp 1ft id • CM rH CO H Ph co O - g cfl £ : ■ii +» ( o c« e EhQ-i * ■u cfl S r- 1 2: co cfl .£ 3 -• «j *j o S o cfl o o id h CC! - S a js s O cfl C CD M 3 i IB ) o - g> co- ca .C 3 , •*-> *J O C O Cfl CJ Z o >-i i so "cfl o The Absolute Participle 43 E. Origin of the Idioms with the Absolute Participle : Summary Statement For the sake of brevity and of convenience, the Origin of the Absolute Participle has been discussed as each case that is used absolutely has been treated. See the preceding sections, espe- cially pp. 2-13 for the Dative-Instrumental; pp. 25-34 for the Accusative; and pp. 35-38 for the Nominative. Here I will only add, by way of general summary, that, in the Lindisfarne Gospels, the absolute use of the participle, whether in the normal dative-instrumental case, or in the rarer accusative case, or in the very rare nominative case, corresponds almost without exception to an ablative absolute of the original Latin, and is patently due to Latin influence. The use of the accusative in absolute constructions seems to have arisen from an interchange of the accusative and the dative cases peculiar to the Northum- brian dialect. The use of the nominative in absolute construc- tions appears to spring from a confusion of constructions, the glossator halting between the use of a finite verb and of an abso- lute participle, as explained above, p. 36. But that the absolute use of the participle, whether in the dative-instrumental, in the accusative, or the nominative, is not a native Northumbrian idiom, but is due to Latin influence, is attested not only by the fact stated above, namely, that the absolute construction is found only in translation of a Latin absolute participle, but also by this further fact, that in the majority of instances the glossa- tor renders the Latin absolute participle by a finite verb, either co-ordinated or (somewhat less frequently) subordinated. In this preference for the finite verb over the participle, the Lindis- farne Gospels stands on the same plane as does the West-Saxon Gospels, nay more, as do West-Saxon as a whole and, in all probability, the Germanic languages as a whole. But, while in the West-Saxon Gospels the subordinated finite verb is slightly more frequently used than is the co-ordinated finite verb in translation of a Latin absolute participle, in the Lindisfarne Gospels the reverse is true. 44 The Syntax of the Lindisfarne Gospels But, while, as just stated, the Latin influence in the develop- ment of the absolute construction in Northumbrian and in West-Saxon was primary, there was at work a secondary influ- ence of native origin, which is manifested in the expressions made up of a preposition (in Northumbrian usually mid, in West-Saxon be) plus a substantive and a participle in the dative-instrumental each, possibly, also (as pointed out for Gothic by Professor Delbriick), of a substantive in the dative of the person interested plus a participle in the dative : see pp. 16-17 above. CHAPTER II THE APPOSITIVE PARTICIPLE Introduction A participle is used Appositively " when the connection be- tween the participle and its principal is so loose that the two seem to constitute two independent ideas ; or, to use the words of Sweet (§90): 'When the subordination of an assumptive (attributive) word to its head-word is so slight that the two are almost co-ordinate, the adjunct-word is said to be in apposition to its head-word.' " 1 Typical examples are the following: — Mat. 26. 7 a : cuom to him wif hcebbende staenna fulle smirinisse = accessit ad eum mulier habens alabastrum ungenti ; Mat. 26.10 : Sa wittende uel wiste uutedlice (5e haelend cue(5 = sciens autem iesus ait; Mat. 25.9 : geonduordon hogo cuoedendo = responder- unt prudentes dicentes. As the examples just given show, under the Appositive Par- ticiple I include not only the participle that is equivalent to a dependent adverbial clause, but also the participle that is equiva- lent to a dependent adjectival (relative) clause and the parti- ciple that is substantially equivalent to an independent clause. For a detailed discussion of this classification, see, below, the section (C) on" the Uses of the Appositive Participle." For the light that it throws upon Northumbrian and Ger- manic syntax, the appositive participle with an accusative 2 1 Quoted from The Appositive Participle in Anglo-Saaon, p. 143. 2 In a few instances, I have included under the heading " accusative object " objects that in form are not accusative but that, from the Modern English standpoint, seem to be direct objects, as in Mk. 1. 41: gerahte hond his & gehran uel hrmande him cue8 = extendit manum suam et tangens eum ait; and in L. I. 5, 6 a , where hrinendo governs him. In all the examples of hrina given by Cook except in these two, it governs the accusative of a noun. The use of him here but reenforces what was said above, p. 28, as to the interchange of accusative and dative forms of the pronouns. 45 46 The Syntax of the Lindi-sfarne Gospels object is everywhere separated from the appositive participle without an accusative object, — a distinction obviously of more importance for the present participle than for the preterite participle. A. The Present Participle I. WITH AN OBJECT The Present Appositive Participle with an Accusative Object is much more frequent in the Northumbrian Gloss than in the West-Saxon translation of the Gospels, 243 examples (299 in- cluding the " Introductions ") occurring in the former to 122 in the latter. The following are typical examples: — Mat. 10.12: groetas <5a ilco (5us cuefiende etc. = salutate earn dicentes etc. ; Mat. 26.27 a : & genimmende colic Soncunco dyde = Et accipiens calicem gratias egit ; Mat. 8.28 : geurnon him tuoege hcebbende uel hsefdon diobles = occurrerunt ei duo habentes daemonia. The statistics in full of the Present Appositive Participle having an Accusative Object are as follows (299 in all, 56 m the Introductions) : — 1. In the Nominative, Singular or Plural (239 in all, 26 in the " Introductions ") The Nominative of the Appositive Present Participle with an Accusative Object ends normally in -nde (in the singular, all genders, and in the plural, masculine and feminine) ; oc- casionally in -ndo (in the plural, masculine and, possibly, neuter 3 ) ; rarely in -nd (in the singular, masculine), in -nda? (in the singular, masculine), and in -nda (in the plural, mas- culine). See Carpenter, I. c, §§ 527, 532, 534, and 538. The vowel preceding the above endings is usually e, but is occa- sionally a or op or o: for details, see Carpenter, I. c, §§ 529 ff., 3 See L. 15. 9 (Sa icif . . . cuoeftendo) ; and compare LindeloP, I. c, p. 237, who tells us that, while wif is usually neuter, it is sometimes feminine. The Appositive Participle 47 and Kolbe, I. c, §§ 207 ff. When no ending is given in the fol- lowing statistics, -nde is to be understood. beada, order, command (1) : J . (1) : I. 8.11 b . bearna, bum (1) : J". (1) : 5.35 a . bebeada, order, command (1) : Mat. (1): 11.1 (or Pre- dicative?). behalda, behold (1) : Mh. (1) : 10.21 (-nd). bera, bear, carry (3) : L. (2) : 5.18 a : 22.10 b — J". (1) : 19.15. bida, await, expect (1) : L. (1) : 2.25 (-nd). bidda, pray (1) : Mk. (1) : 11.24. bita (bidta), bite (1): ifcffc. (1): 1.26 a (bidtende). brenga, bring (1) : ilffc. (1) : 2.3. caerra: see cerra. ceiga, call (2): ilfa*. (1): I. 1.8 (-nda?).—Mk. (1): 3.31 (-ndo). cerra (caerra), to, move (2): Mat. (1): 27.39 b . — Mk. (1): 15.29 c . cliop(p)iga, cry out, exclaim (4): Ifai. (2): 9.27 a ; 11.16 b — L. (2): 3.4; 4.41 a . COStiga, tempt (1) : ilfk (1) : 8.11 c (-ndo). cunniga, tempt (1) : Mat. (1) : 19. 3 a . cursiga, pZa£ (1) : Iffc. (1) : 15.17 (-ndo). CUOeSa, say, declare (155): Mat. (61): I. 5.1; I. 8.4; I. 16.14 b (-nd); 2.2; 8.27, 29, 31; 9.14 (-ndo), 27 b , 33; 10.7, 12; 12.38; 13.3 a , 31; 14.26 b , 27, 33; 15.1 (-ndo), 7, 21, 23, 25; 17.5 b , 10*; 18.1, 26, 28; 19.3 b ; 20.12, 21 (-ndo), 30 b (-ndo) ; 21.10, 20 (-ndo), 23 b , 25; 22.24; 24.3, 5; 25.9 (-ndo), 11 (-ndo), 44 (-ndo), 45 ; 26.8 b , 17 a , 27 b , 39 b , 42, 44 b , 48, 68; 27.11, 23, 29 b , 40, 41, 46, 54 b , 62; 28.13 a , 18 b .— Mk. (38) : I. 5.9*; 1.7 a , 15, 24 a 25, 27; 2.12; 3.11 (-ndo) ; 5.12 (-ndo), 23, 35 b ; 6.2, 35; 7.37; 8.15, 16, 26, 28, 33 ; 9.6, 11, 25, 38 a ; 10.26, 35, 49 b ; 11.9, 17, 31; 12.6, 18; 14.39 b , 44, 57 (-ndo), 60, 68; 15.29 d , 34.— L. (43): I. 4.5 b ; I. 7.9 a ; 1.66, 67; 3.10; 4.34 a , 41 b (^nda); 5.12, 21 b ; 7.20; 8.8 b , 24, 54; 9.18, 35; 13.25 c (-ndo), 31 (-ndo) ; 14.30 a (-ndo) ; 15.2, 9 (-ndo) ; 17.4, 13 (-ndo) ; 18.13; 19.16, 38; 20.2 (-ndo), 5, 21 (-ndo), 27 {-ndo) ; 48 The Syntax of the Lindisfarne Gospels 21.7 a (-ndo), 8 (-ndo) ; 22.19, 64 {-ndo), 66 (-ndo) ; 23.2 b (-ndo), 5 (-ndo), 18, 21 (-ndo), 35 b (-ndo), 37 (-ndo) ; 24.23 a (-ndo), 29 (-?m?o); 27.7.—/. (13): I. 4.5 (-ndo) ; I. 7.9 a ; 1.15; 4.31 (-ndo), 51 b (-ndo) ; 6, 52 a (-ndo) ; 11.3 (-ndo), 31 (-ndo); 12.21; 18.22, 40; 19.6 (-ndo), 12 (-rwZo). doa, do, make (2) : ». (1) : I. 1.2 a (-nd).-L. (1) : 17.16 (-nd). doema, judge (2) : Mat. (1) : 19.28.— L. (1) : 22.30. eftwaecca, resuscitate (1) : ilfatf. (1) : I. 18.9 C . forbeada, /orOtd, prohibit (1) : Mfc. (1) : I. 4.11 (-nd). forebeada, forbid, prohibit (1): Mk. (1) : I. 4.15 a (-nd). forleta, dismiss (3): Mk. (1): 26.44 a .— Mk. (2): 4.36; 14.50 (-ndo). fraigna: see fregna. fregna (fraigna), enquire, ask (2) : Mk. (2) : I. 4.8 a ; I. 5.4. fulwiga, baptize (1) : Mat. (1) : 28.19. gearwiga, prepare (2) : Mat. (1) : 27.28.— L. (1) : I. 8.13 b . gebidda, pray (1) : Mat. (1) : 26.39 a . gebloedsiga, bless (1) : L. (1) : 24.53 b . gelsera, teach (1) : Mk. (1) : 7.7. genim(m)a, take, accept (3) : Mat. (3) : 26.27 a , 37 a ; 27.6 genioma : see genim (m)a. gesea, see (2) : Mat. (2) : 26.8 a ; 27.54 a . geteara, tear (1) : ilffc. (1) : 9.26. habba, have (6) : Mat. (2) : 8.28; 26.7 a .— Mk. (3) : I. 2.5 b ; 1.22; 8.18.— J. (1): 5.5. halda, hold (1) : Mk. (1) : I. 3.11 (-nd). heriga, pra^e (1) : L. (1) : 2.20 b (-ndo). hrina, /onc/ t (2): Mb. (1): 1.41.— L. (1): I. 5.6 a (-ndo). hyhta, /tope (1) : L. (1) : 6.35 (-ndo). lsera, £eac/i (2) : Mat. (2) : 9.35 ; 28.20 a . lofiga, praise (1) : L. (1) : 24.53 a (-ndo; or Predicative?). merciga, seal (1) : Mat. (1) : 27.66. nim(m)a (nioma), take, hold (4): ilfa£. (3): I. 10.3; 26.57; 27.27.— J. (1): 2.6. Ilioma: see nim(m)a. The Appositive Participle 49 ondeta, confess (1) : Mk. (1) : 1.5. onginna, begin (1) : Mfc. (1) : I. 1.10. sella, #w>e up (1) : 3/a.^. (1) : 27.4. senda, send (2) : Jlfa<. (1) : 27.35.— L. (1) : I. 10.12. setta, set, place (1) : L. (1) : I. 8.3 b . sla, s«e (1) : Mat. (1) : 26.51. soeca, seek (7): Makka(5i henni mor- gum f>okkum Tcyssandi hana ok halsfadmandi. In his recent Norr0n Syntax (Kristiania, 1906), Professor Nygaard reaffirms his statement as to the original genitive regi- 56 The Syntax of the Lindisfarne Gospels men of the present participle in old Norse, and ascribes to Latin influence the acquisition by the present participle of the power to govern an accusative object. In § 242 he says: " Det oprin- delige forhold med hensyn til prses. part, synes altsaa at have vaeret: . . . prses. part, er i det vsesentlige at betragte som adjektiver; de har ingen fremtrsedende tidsbetydning og har nominal styrelse (genitiv). " Saaledes vedbliver ogsaa idethele sprogbrugen at vsere i F. S. 6 " I L. S. 6 udvides og modificeres brugen ved paavirkning af fremmede analogier, navnlig det latinske nutidsparticip og passive verbaladjektiv. " Det aktive part, dannes her ogsaa af transitive verber og faar verbal karakter (med tydelig tidsbetydning og verbal styr- else)." Moreover, in another section (233), Professor Nygaard tells us that, in the popular style, as a rule, the appositive present participle of the transitive verb is not used, but instead is used a subordinated finite verb or a co-ordinated finite verb. A more complete parallel to the situation with reference to the governing power of the participle in Anglo-Saxon could scarcely have been drawn by Professor Nygaard, even if he had had open before him my treatise on The Appositive Participle in Anglo-Saxon. His statements are the more remarkable when we learn that he had not seen that monograph. In his Altisldndisches Elementarbuch, Heidelberg, 1913 (a revision of the late Bernhard Kahle's work with the same title), Professor Andreas Heusler gives a treatment of the Present Participle in the Scandinavian languages, which, though brief and couched in general terms, seems to favor the thesis set forth by me (§422):— " Seine Verwendung in der volkstiimlichen Prosa ist be- schrankt: es wird fast nur von intransitiven Verba gebildet; abgesehen von festen Verbindungen (wie farande honor ' her- •F. S. = Folkelig Stil, 'popular style'; L. S. = Lserd Stil, 'learned style.' The Appositive Participle 57 umziehende Weiber,' dugande maf>r ' wackrer Mann,' sitiande hlutr ' sitzender Anteil = Zuschauerrolle ') erscheint es meist nur appositiv in Verbindungen wie: kom par ripande Hialte ' kam da H. geritten ' : hon gehh pegiande at Flosa ' sie trat schweigend zu Flose.' " Ein Hauptkennzeichen des ' gelehrten Stiles ' sind die den Bericht weiterfiihrenden und transitiven, auch die pradikativen Partt. wie: frcendr hans homo, eggiande hann at fa ser eignar- spuso ' seine Verwandten kamen, ihn antreibend, sich ein Ehe- gespons zu nehmen ' : konungs so7i var fyrstr ok fremstr allra sinna manna. . . . eige fyrr lettande, en hann er par komande, sem bardagenn var, pegar svno suerpe bregpande ok huern um annan hQgguande ok suma mep spiote leggiande ' der Konigs- sohn war der erste und vorderste aller seiner Mannen . . . nicht eber ablassend, als bis er dabin kommend ist, wo die Scblacbt war, alsbald sein Scbwert ziebend und einen nacb dem. andern bauend und etlicbe mit dem Spiese durcbbobrend.' " As to Old Higb German, justification for tbe statements made by me in 1901 concerning tbe regimen of tbe appositive present participle therein, is to be found in Dr. Konrad Meyer's Zur Syntax des Participium Praesentis im Althochdeutschen, a Marburg dissertation published in 1906. On page 21 Dr. Meyer tells us that, of the present appositive participles in their Latin originals, the Monsee-Wiener Fragmente avoids 5/7 ; Isidor, 2/3 ; and even Tatian, 1/5 ; that, when the present appo- sitive participle is found in these Old High German texts, it translates almost without exception a Latin appositive parti- ciple 7 ; and that, when the participle of the Latin original is not kept, in the majority of instances it is rendered in Old High German either by a coordinated finite verb (in Monsee-Wiener Fragmente and in Tatian chiefly) or by a subordinated finite verb 8 (in Isidor chiefly). On p. 40 he states that the appositive present participle is relatively rare in the more original Otfrid, ' See Meyer, I. c, pp. 22, 25, and 32. 8 See Meyer, I. c, pp. 23, 26, and 34. 58 The Syntax of the Lindisfame Gospels only about 44 examples occurring in the whole of his works. Of these 44 examples of the appositive present participle, I may add, not a few are modal; many (about 33) occur after verbs of motion; and in some 9 of these 33 examples, the participle seems to me predicative rather than appositive, and is inde- pendently so classed by Dr. Karl Rick in his Das Pradikative Participium Praesentis im Althochdeutschen, a Bonn disserta- tion of 1905. In a word, then, there are even fewer appo- sitive present participles in Otfrid than is stated by Dr. Meyer. But, what is more to my present purpose, Dr. Meyer segre- gates the present appositive participles having an accusative object from those not having such an object, and gives the fol- lowing totals therefor: 10 — Isidor: to 8 ; Otfrid: 12 to 44; Monsee-Wiener Fragments: 7 to 20; and Tatian: 131 to 466. Of 140 Latin present appositive participles rendered by a pres- ent appositive participle by Notker, 83 participles have an accusative object in the Old High German, according to Meyer, pp. 51-54. Meyer tells us, too, that, in about 212 instances, Notker uses an appositive present participle without having had a participle in his Latin original, but unfortunately he does not here segregate participles having an object from those not having an object, though he does say that in the majority of these 212 instances the participle has the adverbial form in -ndo; many of these, therefore, probably do not 1Qa have an object. In Otfrid the present appositive participle having an accusative object is less frequent than indicated by Dr. Meyer, who, as above stated, gives twelve examples, 11 for ten of these twelve "For example, in ii, 13.1 (fuar druhtin bredigonti) and in, 2.25 (unz er fuar ahtonti thes selben wortes nmhti ) , both of which are considered predi- cative by Dr. Rick, I. c, p. 30, though appositive by Dr. Meyer. 10 These statistics take account only of present participles modifying the subject nominative. 10a Since writing this statement, I have myself looked up eighty of these examples, taken at randon, and of the eighty examples only fourteen have an accusative object. As we should expect, most of the participles that do not have an object are used adverbially, and denote manner. "The twelve examples are found as follows: I, 4.20, 5.50 b , 13.18, 23.10; III, 2.25, 8.13 b , 10.13* ; IV 30.1; V, 4.15, 9.5, 9.7', 14.10. The Appositive Participle 59 participles are considered predicative by Dr. Rick, and in most instances he is justified in his classification, I think, the parti- ciples being predicate nominatives after intransitive verbs of motion (faran and gang an) ; though in one of these ten exam- ples (I, 4.20: Ingiang er tho skioro, goldo garo ziaro, mit zin- seru in henti thaz hus rouhenti) the participle is so far removed from the principal verb that it seems appositive to me rather than predicative. This leaves us at most only three present appositive participles with an accusative object in Otfrid (I, 4.20, just quoted: I, 5.50 b : Noh keisor untar manne, nimo geba hringe fuazfallonti int inan erenti; and I, 13.18: Thiu muater barg mit festi thiu uuort in iru brusti, in herzen mit githahti thiz ebono ahtonti). These three appositive participles may very easily have been suggested by any one of the numerous Latin sources 12 that have been suggested for Otfrid's Evange- lienbuch. To me the wonder is, not that Otfrid thrice uses the appositive participle with an accusative object, but that he uses it only thrice. As to Tatian and iSTotker, the frequency of the appositive present participle without an object is due, I think, largely to the influence of their Latin originals, but, also, in the case of JSTotker, partly to his fondness for the native Germanic idiom of the participle to denote manner, in which latter case, Notker more commonly uses, Meyer tells us (pp. 48-49), the adverbial form of the present participle in -ndo. With both Tatian and Notker, the frequency of the appositive present participle with an accusative object is due almost wholly to the influence of their Latin originals, I think. For the frequency of the appo- sitive present participle (whether with or without an object) in the added passages of Notker, Dr. Meyer, I. c, p. 50, offers another explanation, not, however, incompatible with my own: " Im Gegensatz zu Is., Mons., Tat. fiihrt also Notker sehr zahl- reiche deutsche Part. Praes. in seinen Text ein, denen im La- teinischen kein Part, entspricht. Das wiirde er schwerlich u For a summary statement concerning these sources, see A. L. Plumhoff's Beitrdge zu den Quellen Otfrids, Kiel Dis., 1898. 60 The Syntax of the Lindisfarne Gospels getan haben, wenn er, wie jene Uebersetzer, deutsche Satze in ununterbrochener Reihenfolge aneinander gefiigt hatte, ohne latein. Satze einzuschieben. Vielmehr sind die vielen Parti- cipiea, bedingt durch den eigentiimlichen abrupten Charakter von Notkers Mischprosa, die sich in ihrem Streben nach Prag- nanz der Sprechweise dem geniachlich fortschreitenden deutsch- en Satzbau, wie er in zusammenhangender Rede herrscht, nicht anzupassen vermag." See, too, Dr. Paul Hoffmann's Die Mischprosa N others des Deutschen (= Palaestra, No. lviii), Berlin, 1910; and Dr. Hans Naumann's Notkers Boethius: Untersuchungen ueber Quellen und Stil (= Quellen und Forschungen zur Sprach- und Culturgeschichte der Germa- nisclien V biker, No. cxxi), Strassburg, 1913. On p. 79, in a passage quoted by me on p. 24 above, Professor Naumann calls attention to the Latinized style of Notker, especially in the use of the participle and the infinitive. I have made little use of Dr. Wilhelm Gocking's Das Participium bei Notker, a Strass- burg dissertation of 1905, because for my present purpose Not- ker's use of the participle is better treated by Dr. Meyer, who, as already stated, segregates the participles having an object from those not having an object. Although Dr. Gocking does not do this, he does classify his participles according to the idea denoted into Temporal, Causal, etc. A combination of the plan of Dr. Meyer with that of Dr. Gocking would give the best results, it seems to me. As in the case of Old Norse, so in Old High German the situ- ation with reference to the appositive present participle with an accusative object very closely parallels that in Anglo-Saxon. Perhaps I should add that, although Dr. Meyer several times speaks of the Latin influence upon Old High German syntax, he does not specifically discuss the origin of the appositive use of the present participle (whether with or without an object) in Old High German; that apparently neither he nor Dr. Gocking had seen The Appositive Participle in Anglo-Saxon; and that, while gratefully availing myself, in the present study, The Appositive Participle 61 of the statistics given by these two scholars, the interpretation thereof is my own. As to Middle High German, valuable testimony is given by Dr. Theodor Matthias, in his " Zur Geschichte der Deutschen Mittelwortfiigungen," an article that appeared in the Zeit- schrift filr den Deutschen Vnterricht of 1897 (Vol. xi, pp. 681- 708), but that was not known to me in 1900, when I wrote The Appositive Participle in Anglo-Saxon. After telling us, on p. 689, that the appositive preterite participle is common in Mid- dle High German, Dr. Matthias declares, on p. 690, that the appositive present participle is rare, especially when having an object. Says he: " Viel seltener ist im Mhd. dieser Gebrauch des 1. Mittelwortes, und unzweifelhaft hat fremdes, nament- lich lateinisches Vorbild darauf gefuhrt. . . . Vor allem von der Beschwerung dieses 1. Mittelwortes mit Erganzungen und Umstanden lasst sich deutlich nachweisen, wie sie aus latei- nischen Vorlagen entsprang." II. WITHOUT AN OBJECT The Present Appositive Participle without an Accusative Object occurs about 108 times (168 including the " Introduc- tions") in the Northumbrian Gloss to 115 times in the West- Saxon Gospels. The following examples are typical: — Mat. 28.18 a : geneo- lecende Se hselend spreccend wses to him = accedens iesus locu- tus est eis; Mat. I. 19.16 a : geongende uel geeade ofer sae <5one peter Sruncnende ahefes = Ambulans supra mare petrum mer- gentem leuat; Mat. 20.30 a : heonu tuoege blindo sittende set weg geherdon = ecce duo caeci sedentes secus uiam audierunt. The statistics in full of the Appositive Present Participle not having an Accusative Object are as follows (167 in all, 59 in the " Introductions ") : — 62 The Syntax of the Lindisfarne Gospels 1. In the Nominative, Singular or Plural (88 in all, 14 in the " Introductions," including the examples in Notes 1 and 2) The Nominative of the Appositive Present Participle with- out an Accusative Object ends normally in -nde (in the singu- lar, all genders, and in the plural, masculine) ; occasionally in -nd (in the singular, masculine, and in the plural, masculine) ; rarely in -ndo (in the plural, masculine) and in -ing (in the singular, neuter ( ?)). Compare Carpenter, I. c, §§ 527, 532, 534, and 538. When no ending is given below, -nde is to be understood. adunestiga, descend (1): Mk. (1): 15.30. basniga, expect (1) : L. (1) : 23.35 a (or Predicative?). bodiga, preach (1):L. (1) 9.6 a . cerra, turn (1) : L. (1) : 8.37. cliop(p)iga, cry out, exclaim (2) : Mk. (2) : 5.7; 15.39. cuma (cyma), come (3): Mat. (1): 20.10 (cymende). — Mk. (2) : 7.1 (cymende) ; 16.1 (cymmende) . CUnniga, tempt (1) : Mat. (1) : 16.1 (-ndo). cuaciga, tremble (1) : L. (1) : 8.47. cyma: see cuma. doa, do, act (1) : L. (1) : 16.8 (-nd). ebolsiga, blaspheme (2) : Mk. (1) : 15.29 b .— L. (1) : 22.65. efsta, hasten (1) : L. (1) : 2.16. falla, fall (1):L. (1):4.7. fara, go, pass by (7): Mk. (4): 1.16* 35 (-nd) ; 13.1; 16.20 a .— L. (3) : I. 4.1 b ; 1.6 (-ndo) ; 4.30. foera, depart (1) : Mk. (1) : I. 3.7 a . fore(h)luta, bend down, stoop (1) : Mk. (1) : 1.7 b . foreliora, pass by (1): Mat. (1): 27.39 a (foreliornende (sic) in Skeat and in Cook; but is a scribal or typographical error for foreliorende, the form given in the Rushworth text ? Carpenter, I. c, § 527, has foreliornende without comment). fulwiga, baptize (1) : J. (1) : 1.31 a . gebidda, pray (1) : L. (1) : 21.36 a . The Appositive Participle 63 geecniga, conceive (1) : L. (1) : 1.31. gefeaga, rejoice (3) : L. (3) : 15.5; 19.6, 37 a . gehera, ftear (1) : ilfa*. (1) : 27.47 b . geneoleca, approach (1) : Ifai. (1) : 28.18 a . geonga, gro, walk (7) : Ma*. (3) : I. 19.16 a ; 26.49 ; 27.5 b .— Mk. (2) : 6.48b; I4.39 a — L. (1) : 24.17.— J". (1) : I. 4.16. gesea, see (2) : Mat (2) : 13.13, 14. habba, have (1): L. (1): 7.2 (-nd). hera, hear (1) : Mk. (1) : 4.12 (-nd). hlifiga: see lifiga. hrema, weep (2) : Ma*. (1) : 2.18b.— /. (i) . 20.11. hwispriga, murmur (1) : J. (1) : I. 4.20. ingeonga, enter (1) : L. (1) : 8.16. iorna, run (2) : Mat. (2) : 27.48 a ; 28.8 a . lsera, teach (3): ». (1): 12.35 (-nd).-L. (2): 5.17 a ; 13.22. lecniga, cure, heal (1) : L. (1) : 9.6 b . lifiga (hlifiga), live (1) : Ma£. (1) : 27.63. lixiga, shine (1) : J". (1) : 5.35 b . msena, mourn (1) : L. (1) : 2.48. ofcliop(p)iga, en/ (gecladed). (ge)doema, condemn (1) : J. (1) : I. 5.9 (-edo; Cook: asf., but the participle modifies dcet wif; Carpenter, I.e., §553: wk. as., but he assigns no gender). The Appositive Participle 77 (ge)friga, free, liberate (1) : £. (1) : I. 8.9 (gefreod). (ge)gearwiga, clothe (2): L. (2): 7.25 b (-ad); 23.11 (-ad). (ge)hera, hear (1) : Mat. (1) : I. 6.12 b (-ed). (ge)hiwiga, clothe in purple (1): L. (1): I. 9.2 {-due; Cook: asm.; Carpenter, I. c, § 553: asm.). gemenga, mix (1): Mat (1): 27.34 b (gemenced; Cook: gemenged, pp.). (ge)nemna, name (1): .Mai. (1): I. 2.12 {-edo; Cook: apm. ; Carpenter, I. c, § 555 : apm.). [But genemnedo modifies boec, and, according to Professor Cook, hoc is either f. or n. I take genemnedo to be feminine here. If genemnedo is mascu- line here, it is due to the too close following of the Latin, nuncupatos.~\ (ge)scrynca (gescrinca), shrink, dry up (1): L. (1): I. 8.4 (-an; Cook: pp. ; Carpenter: not cited). (ge)sea, see (1): Mat. (1): I. 6.12 c (gesege (sic); Cook: pp. ; Carpenter, I. c, § 547, Amk. 1 : " Der verlust des n in gesege, Mt. Pr. 6.12, wird schreib-fehler und nicht lautgesetz- lich sein "). (ge)temesiga, sift (?) (1): Mat. (1): 12.4 (-eda; Cook: apm. ; Carpenter, I. c, § 555 : apm.). (ge)Sersca, hill (1) : L. (1) : 20.10 (gedorscen). (ge)woeda, clothe (1) : Mat. (1) : 11.8 (-ed). (ge)wor(5a, become, be made (2) : MTe. (1) : I. 1.6 C (-en). — /. (1): 2.9 (-en). (ge)wyrca, make (1): Mat. (1): I. 5.5 (geworht). ofnioma, take, catch (1) : J. (1) : 8.3 (-en). senda, send (1) : Mat. (1) : I. 21.8 (sende; Cook: not cited; Carpenter: not cited). unwoeda, not to clothe (1) : Mat. (1) : 22.11 (-ed). ymbgearwiga, clothe (1): Mk. (1): 16.5 b (-ad). 78 The Syntax of the Lindisfarne Gospels ORIGIN OF THE IDIOM In most instances, 18 the Preterite Appositive Participle with- out an Accusative Object was probably a native idiom in the Lindisfarne Gospels. (1) Except in eight cases, it is true, the Lindisfarne Preterite Appositive Participle without an object corresponds to the same idiom in the Latin original. In the exceptional examples the Latin correspondents are as follows : a noun in the ablative (Mat. I. 6.10, /. 3.1), 2; a gerundive (one predicative: J. I. 8.12; one appositive: L. I. 5.9), 2; a noun in the genitive (Mat. 12.4, L. 6.4 a ), 2; a passive sub- junctive (Mat. 27.12), 1; a passive infinitive with accusative subject (L. 7.24 c ), 1. But in the majority of instances (in 81 out of 133 examples, or in the ratio of 1.64 to 1), the abso- lute participle of the Latin original is retained by the Northum- brian glossator. (2) In most of its uses, the idiom was native to West-Saxon and probably to the Germanic languages as a whole. See The Appositive Participle in Anglo-Saxon, pp. 297 ff., especially pp. 299-300. On page 79 I give a synoptic table showing the occurrences of the Appositive Participle, in the several cases, in the Lindis- farne Gospels. C. Uses of the Appositive Participle In the Lindisfarne Gospels, as in West-Saxon, the Apposi- tive Participle has three chief uses : — I. The Adjectival, in which the Appositive Participle is equivalent to a Dependent Adjectival (Relative) Clause, and denotes either an action or a state, as in Mat. 26. 7 a : cuom to him wif hcebbende stsenna fulle smirinisse = accessit ad eum mulier habens alabastrum ungenti ; L. 5.18 a : heono wseras berende on bed uel on ber mono = ecce uiri portantes in lecto "The details are given below in the section specifically dealing with " the Origin of the Appositive Participle." Total of Present and of Preterite Participles With- out Obj. O0U5M tO ^< CM NCOOl OS U5 CO cm t- ia MHrt CO I- OS lOlrtO) CM r-l With Obj. ■*HM rH © i-H iH i-H AAO t- ia cm 00 CO CM rH OSCM CO rH rH NlOt- ©TfUS CO CM Total of Preterite Participle With- out Obj. iH OS CM ■<*< rH cS Q With- out Obj. ■<*< CO iH M< CM CM OO Irt CO With Obj. rH rH • i-H rH • 3 With- out Obj. co • co •f rH CO t- rH CO With Obj. rH iH • rH rH • a 5 2 With- out Obj. © OS rH CM i-H ■* CO iH CO CO CO U5 -^< iH lO OS CO CO rH rH With Obj. 1 O M H K H co H a S < With- out Obj. OS U5 rt< t-MiP co c- co rH •*HM MCBt- CO rH rH With Obj. rH iH • i-t iH • 00 •*< •* co co co CM • CM t- OO OS C8 Q With- out Obj. 00 Tfl •*»< t- CO rH ia co cm i-H i-l CO ■ CO CO CO CO CO rH CM With Obj. •<* iH CO Tt* iH CO OS CD CO ia • ifl eq 00 ■* CM rH a 3 O With- out Obj. ■"*< CM CM CM rH 00 CO rH With Obj. «* oo c© OS 00 t»< •»!< © co ia iH © Tf CO CO U5 CM i-l OS CO CO CO rH CM CM CM H ca O Matthew: Total Gospel Introd'n Mark: Total Gospel Introd'n Luke: Total Gospel Introd'n John : Total Gospel Introd'n Totals Gospels Introd'ns 79 80 The Syntax of the Lindisfarne Gospels hominem ; Mat. 12.25 : hus todceled uel tosliten wi(5 him ne stondas = domus diuisa contra se non stabit. At times, how- ever, the Latin appositive participle that is equivalent to an adjectival (relative) clause, is awkwardly translated by a co- ordinate definite verb, as in Mat. 8.16: gebrohton him menigo diobles hcefdon = optulerunt ei multos dsemonia habentes (Rushworth: monige deofulseoke hwbbende; W. S. : manege deofolseoce). II. Adverbial, in which the Appositive Participle is equiva- lent to a Dependent Adverbial (Conjunctive) Clause, and de- notes time, manner, means, etc., as in Mat. 26.10: (5a ivittende uel wiste uutedlice Se hselend cueS = sciens autem iesus ait ; Mat. 27 .4 : ic synngade uel ic firinade sellende uel ic salde blod Sone so(5f sest = peccaui tradens sanguinem iustum ; Mat. 27.54 a : gesende eor(5 hroernise . . . ondreardon sui(5e = mso terrae motu . . . timuerunt ualde; Mat. 27.35: todseldon hrsegla his tan sendende = diuiserunt uestimenta eius sortem mittentes. III. Co-ordinate, in which the Appositive Partciple is sub- stantially equivalent to an Independent Clause, and either (1) denotes an accompanying circumstance (the " circumstantial " participle), or (2) repeats the idea of the principal verb (the "iterating" participle), as in Mat. 28.19: gaa(5 forSon laeraS alle cynno uel hsedno fulwuande hia in noma f adores = euntes ergo docete omnes gentes baptizantes eos in nomine patris; Mat. 8.29 : geceigdon (5us cue&ende — clamauerunt dicentes; Mat. 25.9 : geonduordon hogo cuoedendo = responderunt prudentes dicentes. The relative frequency of these three uses of the appositive participle is as follows: of the Adjectival, 249 examples occur, of which 168 are present participles; of the Adverbial, 78, of which 68 are present participles; and of the Co-ordinate, 231, of which 230 are present participles. In all, 558 appositive participles occur in the Lindisfarne Gospels, of which 466 are present and 92 are preterite. For the West-Saxon Gospels the figures are as follows: of the Adjectival Use, 61 examples occur, The Appositive Participle 81 of which 44 are present participles; of the Adverbial, 81, of which 69 are present participles; and of the Co-ordinate, 132, of which 124 are present participles; a total of 274 appositive participles, of which 237 are present participles. The larger number of appositive participles in the Lindisfame Gospels as compared with the West- Saxon Gospels seems due to two facts : the former has extended prefaces not found in the latter; and the Northumbrian glossator adhered more closely to his Latin original than did the West-Saxon translator. Of the 558 appo- sitive participles in the Lindisfame Gospels, 156 occur in the " Introductions," and 402 in the Gospels proper, as against 274 in the West-Saxon Gospels. The three chief groups of the Appositive Participle — Adjec- tival, Adverbial, and Co-ordinate — call for only brief indi- vidual comment. In the Lindisfame Gospels, as in the West-Saxon Gospels, the Adjectival Use of the Appositive Participle occurs more frequently with the present than with the preterite participle, there being 168 of the former to 81 of the latter in the Lindis- fame Gospels and 44 of the former to 17 of the latter in the West-Saxon Gospels. In West-Saxon as a whole, however, the preterite participle is much more frequent than the present in the Adjectival Use, there being 846 of the former to 377 of the latter. The Adjectival Appositive Present Participle is found of- tener without an object than with one in the Lindisfame Gos- pels, in the West-Saxon Gospels, and in West-Saxon as a whole, the figures being respectively 94 to 74, 24 to 20, and 270 to 107. Of the Appositive Preterite Participle in the Adjectival Use, 81 examples occur in the Lindisfame Gospels, 3 with an accu- sative object; 17, in the West-Saxon Gospels, none with an accusative object ; and 846 in West-Saxon as a whole, none with an accusative object. In all except thirteen instances, the Lindisfame Adjectival 6 82 The Syntax of the Lindisfame Gospels Appositive Participle translates the same idiom of the Latin original. Of these thirteen examples, two have a present par- ticiple with an accusative object translating a Latin substantiv- ized appositive participle with an accusative object: L. 3.4 and L. 6.32 ; five have a present participle without an accusative object, corresponding to a Latin appositive adjective {Mat. I. 20.12), a Latin gerundive {Mat. I. 20.16 b , L. I. 10.3), a Latin subjunctive {L. 16.18), and a Latin noun in the genitive (de- scriptive) (Mk. I. 1.4 a ) ; and seven have a preterite participle without an accusative object, corresponding to a Latin gerun- dive {J. I. 8.12, L. I. 5.9(?)), a Latin noun in the genitive (descriptive) {Mat. 12 A, L. 6.4 a ), a Latin noun in the ablative (descriptive) {Mat. I. 6.10, J. 3.1), and a Latin passive infini- tive with an accusative subject {L. 7.24 c ). These divergences from the Latin original are slight, and are such as are to be expected. It seems, therefore, that the evidence of the Lindis- fame Gospels tends to confirm the conclusion reached as to the origin of the Adjectival Appositive Participle in West-Saxon, as stated in my monograph on The Appositive Participle in Anglo-Saxon, pp. 298-300 ; namely, that the adjectival use of the present appositive participle except with a few slightly ver- bal participles like living, lying {licgende), etc., was in West- Saxon due chiefly to Latin influence, especially if the participle governed an accusative object; but that the adjectival use of the preterite appositive participle was a native West-Saxon idiom. Of the Adverbial Use of the Appositive Participle about 78 examples occur in the Lindisfame Gospels, of which 68 are present; against 81 in the West-Saxon Gospels, of which 69 are present. In the Lindisfame Gospels, an accusative object is found 28 times with the present participle to 27 in the West-Saxon Gos- pels. With the preterite participle an accusative object is not found in the former, and only once in the latter, in imitation of the Latin original. The Appositive Participle 83 In its adverbial use the appositive participle denotes subordi- nate relations of manner and means (combined here under the head of modal), of time, of cause, of purpose, of concession, and of condition. Some of the adverbial appositive participles admit of a twofold classification. According to my estimate, the approximate number of each use is: Modal, 23 ; Temporal, 32 ; Causal, 5 ; Final, 11 ; Concessive, 6 ; Conditional, 1. For the West-Saxon Gospels, the corresponding figures are: Modal, 15; Temporal, 32; Causal, 11;. Final, 11; Concessive, 10; and Conditional, 2. Of the Appositive participle denoting Time and Manner (in- cluding Means), examples have been given above in defining the groups. Here I need add only examples of the remaining adverbial uses. Of the appositive participle denoting Cause, I cite all the clearer examples observed (5): — Mat. 26. 8 a : gesegon uel ftcet (5a geseende . . . discipulas . . . wraSe weron = uidentes au- tem discipuli indignati sunt; Mat. 27.54 a : geseende eor<5 hroer- nise . . . ondreardon sui<5e = uiso terrae niotu . . . timu- erunt ualde; L. I. 8.13 b : sete(5 geddung of Ssem laSendum se(5e hine forcuoede gearuande uel geteldon = Ponit parabolam de inuitatis qui se excusare studentes (or Adjectival?); Mat. I. 8.13: gelaered sum o(5er alexandrinesca miS micile bigeong & ec hogahscipe enne uel an us fore feower godspellum offorleort = ammonius (for ammonitus?) quidam alexandrinus magno studio atque industria unum nobis pro quattuor euangeliis dere- liquit (or Adjectival ?) ; L. 7.13: se drihten mi(5 miltheort- nise gecerred ofer hia cuoeS = dominus misericordia motus super ea dixit (or Co-ordinate?). Of the appositive participle denoting Purpose, eleven exam- ples have been observed: — Mat. 16.1: togeneolecdon to him Sa aldo . . . cunnendo & bedon hine — accesserunt ad eum phari- saei et sadducaei temtantes et rogauerunt eum (or Co-ordi- nate?) ; Mat. 19. 3 a : geneolecdon to him — [== blank] cunnende hine = accesserunt ad eum pharisaei temtantes eum ; Mat. 19.28: sittes & gie ofer seatla tuelf doemende tuoelf strynda 84 The Syntax of the Lindisfarne Gospels israeles = sedebitis et uos super sedes duodecim iudicantes duo- decim tri bus israhel; M at. 12.43: miS Sy . . . gaast . . . gaaS uel geeade Serh stowa dryia gesohte uel soecende rest = cum autem . . . spiritus . . . ambulat per loca arida quaerens re- quiem (or Co-ordinate?); Mat. 12.46: moder his & broSero stondas uel gestodon bedon uel sohton uel soecende spreca him = mater eius et fratres stabant foris quaerentes loqui ei (or Co- ordinate?) ; Mat. 12.47; moder Sin & broSro Sin ute stondes sohton uel soecende Sec = mater tua et fratres tui foris stant quaerentes te (or Co-ordinate?) ; Mh. 8.11 b > c : ongunnon efne- gesoeca miS hine Saet hia sohton uel soecende from him becon of heofne costendo hine = coeperunt conquirere cum eo quae- rentes ab illo signum dae (sic for de f ) caelo temtantes eum ; L. 11.24: Serheode uel gaS Serh stowa wseterleasa soecende uel sohte raest = perambulat per loca inaquosa quaerens requiem ; /. 6.24. cuomon to Sa?r byrig sohton uel soecende Sone hselend = uenerunt capharnaum quaerentes iesum; J. 1.31 a : cuom ic in uaetre fulguande uel to fulguanne = ueni in aqua baptizans. Of the appositive participle denoting Concession, six exam- ples have been found: — Mh. 8.18 : ego habbaS gie uel hcebbende ne geseaS gie = oculos habentes non uidetis ;. J. I. 5.5 a : soecende hine to gegrioppanne ne maehtun = quaerentes eum adprehen- dere nequeunt; Mat. 13.13: forSon gesegende uel seende uel Sset geseas uel gesegon ne seaS & Sa geherdon ne heras hia = quia uidentes non uident et audientes non audient (or Co-ordi- nate?) ; Mat. 13.14: gesegende ge sciolon gesea uel ge geseas = uidentes uidebitis ( ?) ; Mh. 4.12 : Ssette gesegon geseaS & ne geseaS & Sa herend geheraS & ne oncnaweS = ut uidentes uide- ant et non uideant et audientes audiant et non intelligent; L. I. 3.6 : Sone long wiS priclom eftdrcegend drihten gecease = quern diu contra stimulos recalcitrantem dominus elegisset. Of the appositive participle denoting Condition, only the fol- lowing example has been found: — Mat. 18.8 b : god uel betra Se is to life ingae unhal uel halt Son tua honda uel tuoege foet hcebbende uel hsebbe sende in fyr ece = bonum tibi est ad uitam ingredi debilem uel clodum quam duas manus uel duos pedes habentem mitti [in] ignem aeternum. The Appositive Participle 85 In all cases except eight the Lindisfarne Adverbial Apposi- tive Participle corresponds to a Latin appositive participle of the original. In these eight examples, seven participles denote time, and one denotes cause. Six of the participles are present, have an accusative object, and correspond to the following Latin idioms: a gerundive in the accusative (Mat. I. 18. 9 C ), 1; and an ablative absolute in the passive (Mat 26.37 a , 44 a ; 2T.5 a , 6, 54), 5. Another of the participles is present, has no accusative object, and corresponds to a Latin subjunctive (Mk. 13.1). And one of the participles (in Mat. 27.12) is preterite, has no object, and corresponds to a Latin subjunctive passive. The evidence of the Lindisfarne Gospels, therefore, tends to con- firm the conclusions reached on the basis of the West-Saxon as to the origin in West-Saxon of the adverbial appositive par- ticiple, as stated in detail in The Appositive Participle in Anglo-Saxon, pp. 300-306 ; namely, that the appositive parti- ciple (present and past) denoting manner was probably a native West-Saxon idiom ; but that the appositive participle denoting other adverbial relations was in West-Saxon due mainly to Latin influence. The Co-ordinate Appositive Participle, in both its circum- stantial and its iterative uses, was illustrated above in defining the term co-ordinate. Here I should add that, as in the West- Saxon Gospels, so in the Lindisfarne Gospels, the co-ordinate use is very frequent, especially in its iterating species: about 155 examples occur of cuoefiende used co-ordinately. In all instances except one (L. 8.37), the Co-ordinate Appo- sitive Participle corresponds to the same idiom in the Latin original. In this one instance the Lindisfarne appositive par- ticiple translates a Latin finite verb that is in close proximity to a Latin appositive participle. I believe, therefore, that the Lindisfarne Gospels again tends to uphold the conclusion reached as to the origin of this idiom in West-Saxon ; namely, that it was imported into West-Saxon directly from the Latin, chiefly from the Vulgate New Testament. 86 The Syntax of the Lindisfarne Gospels NOTES 1. The Voice of the Appositive Participle. — As a rule, in the Lindisfarne Gospels the Present Participle is active in sense. Occasionally, however, the Present Participle is proba- bly passive in sense, as in the following passages, in which the Northumbrian present participle translates a Latin gerundive or preterite 19 participle: — Mat. I. 20.16 a ' b : foregeheht bren- gende him bloedsende lytla ne wiSsaca = Praecepit oblatos sibi benedicendos paruulos non repelli ; ib. I. 18.9 : geeade & eftwceccende dohter Sses aldormonnes (5set wif of herning blodes haeleS geweht & (5set maeden = Pergens et resuscitandam filiam principis mulierem a profluuio sanguinis sanat suscitans et puellam. At times, too, the Substantival Present Participle is passive in sense, as in L. I. 8.13 : sete<5 geddung of Saem ladendum se(5e hine forcuoede gearuande uel geteledon symbel ne woeron wyrSe = Ponit parabolam de inuitatis qui se excu- sare studentes cena non fuerint digni. — The Preterite Parti- ciple of intransitive verbs has an active sense, as in Mat. 7.6 : gewoendo uel gecerdo toslitas iuh = conuersi disrumpant uos. Of transitive verbs, the Preterite Participle is usually passive in sense, but occasionally it is active in sense, as is possibly true of geweht = suscitans in Mat. I. 18.10, quoted in the ear- lier part of this note under Mat. I. 18.9 ; and of offylgde = asse- cuto in L. 1.3, quoted on p. 71 above. But in J. 4.39 (ge- trymmedes = perhibentis), also quoted on p. 71, getrymmedes is probably a scribal error for getrymmendes. — Concerning the voice of the appositive participle in West-Saxon, see The Appo- sitive Participle in Anglo-Saxon, pp. 289-290. For the voice of the Absolute Participle in the Lindisfarne Gospels, see Note 1, on p. 5 above; and of the Predicative Participle, see the section on the Elliptical Accusative-with-Infinitive Construc- tion below. "Often, however, a Northumbrian appositive present participle, though translating a Latin passive participle (either absolute or appositive), re- mains active in sense by merely making the subject of the Latin participle the object in the translation, as in Mat. 26.44* : forletende hia eftersona eode = relictis Mis iterum abiit. The Appositive Participle 87 2. Some Possible Examples of the Appositive Participle are mentioned in Note 1, p. 5 ; Note 3, p. 6 ; and Note 1, p. 10. 3. An Alternative Choice between an Appositive Participle and a Finite Verb, as the Translation of a Latin Appositive Participle, is not infrequently given by the Lindisfarne glossa- tor, as in Mat. 26.49 : sona cwom uel geneolecde uel geongende to Sam hselende cue(5 = confestim accedens ad iesum dixit ; ib. 9.14: (5a geneolecdon uel comon to him Segnas iohannes cue- cfendo uel hia cuedon = accesserunt ad eum discipuli iohannis dicentes; etc., etc. 4. Supplementary Particles with the Appositive Partici- ple. — In his " Conjunction Plus Participle Group in English," a Master of Arts dissertation presented at the University of North Carolina and published in Studies in Philology, iv, 1910, pp. 1-29, Mr. O. P. Khyne discusses the earliest occur- rences of the Supplementary Particle with the Appositive Par- ticiple. On p. 8, he declares: " Einenkel . . . traces it [the construction in question] ... no further back than the mid- dle of the 18th century. Indeed, it seems to be the impression of all grammarians who have treated the subject that the con- struction first came into English about this time. But in real- ity, it goes much farther back. I have been able to trace it as far back as 1552" (in Roister Doister). The earliest occur- rence, however, antedates Mr. Rhyne's year (1552) by cen- turies. Although Mr. Khyne quotes from Matzner's Englische Grammatik, in, pp. 73-74, what is said concerning this con- struction, he seems to have overlooked page 90 of the same vol- ume, where one example is given of this construction in Anglo- Saxon, and The Appositive Participle in Anglo-Saxon, pp. 290- 291, where about a dozen examples are given from West-Saxon. I do not recall any example of the idiom in the Lindisfarne Gospels. I close this section with a Synoptic Table showing, approxi- mately, the distribution of the several Uses of the Appositive Participle in the four Lindisfarne Gospels. As to the classifica- tion of individual examples, of course, there is room for differ- ence of opinion. qg Ph GO H ►J Ph KH o H P3 -5 PM H CO O Pm Ph W H fa o DO w x P fa H H O fa PP H O E p "2-3 00 CO co at iH OS i-l CM CO CO £ ° 1ft ■<* i-t ■«»< CO lO i-l CO Oi 1-1 i-H i-t ■<»< o^ _, *j CS u I© r-t t- OS ■* 00 iH rH t- 9> +J O CP Tf Tf CO CO CXI M i-H CO 00 £ 2 S rH _ -t-j d ja u CO i-H OS • t- rH O iH OS CO *J *J 0) i-H t- • t- CO OS h£§ .-1 CO M CS O 'S" CO •* 00 00 co m tt rH ■<*< 00 "■3 ^O CO CO 1-H CO CO rH i-l o> c- CD J3 . O rH 00 • ■** iH CNI rH -«p 00 £S rH • CM t-\ t- tl • o o 1 o-ord. 1 Adverbi *0 t-( 1-H tH ■<>< T-i J3 . Is 1ft • rH • CO • ■*f • CO • 00 CO z§ tO tH (31 • i-H • CO • CO CO T-\ ja 00 • lO • OS ■ Ifl • t- u 53 t- • Irt • T»< . r-i ■ Oi rH 3 *3 o +> 13 js rH • y-i O ir * u £o a> o •=? CXI • t-( • rH • ■

'53 CO *o 0) J3 iH • rH • CO is a o O 5S "3 .3 fa m tH • l-i • CO J3 in • CM • tH • rH • OS > £o < "3 in 3 03 5S • i-i • i-t CO JS . O II CO ■ i-H • CO "3 o •ri' co m t> i-H "* iH CO • t> t~ o *S rH X3 TP • CO • i-H • 00 H £o o .^ t- • CO • 00 i-H . t- rH 3 '8 o *0 rH A . S Is CO • rH • rH • ia I o o .JA CO •**< 00 00 co in "* T-> rf 00 *s CO CO iH CO CO rH rH a> t~ XI < O tH 00 • ■f rH - 0>H| fa fc. tn t- (-> (h (h tH U a, fafa PmO, PmCh (U Ph tn J3 O ■M 5 o •-5 CS cU O -»-> ■a o Eh o 88 The Appositive Participle 89 ]). Origin of the Idioms with the Appositive Participle : Summary Statement The Origin of the Idioms with the Appositive Participle has, for the sake of brevity, been discussed in connection with the exposition of its several uses. See the preceding section, espe- cially p. 81, for the Adjectival Use; p. 85, for the Adverbial Use; and p. 85, for the Co-ordinate Use. The results there obtained may be summarized here as follows. In the Adjec- tival Use, the Appositive Present Participle, except with a few slightly verbal participles, is due to Latin influence, especially if the participle governs an accusative object ; while the Apposi- tive Preterite Participle is a native English idiom, except when governing an accusative object. In the Adverbial Use, the Appositive Participle (present and past) denoting manner is probably a native English idiom; but in all other adverbial uses the Appositive Participle is probably due to the influence of the Latin. And in the Co-ordinate Use, whether circum- stantial or iterative, the Appositive Participle is an importation from the Latin. That, in the uses above specified, the Latin influence is preponderating is evidenced by the fact that, in the alleged foreign uses, the appositive participle corresponds, al- most without exception, to the same idiom in the Latin original ; and that, in the majority of instances (in the ratio of 1.3 to 1 in the Lindisfarne Gospels as a whole) , the Northumbrian glos- sator renders the Latin appositive participle otherwise than by a participial phrase, usually by a co-ordinated 20 finite verb, less frequently by a' subordinated finite verb. By this prefer- ence for finite verb over participial phrase, the Lmdisfarne glossator again adopts the native English and the native Ger- manic idiom. 20 For the ratio of co-ordinated to subordinated finite verbs, see pp. 52 and 71 above. CHAPTER III THE INFINITIVE Introduction In considering the syntax of the Infinitive in the Lindisfarne Gospels, I follow the general plan of treatment given to the West-Saxon Infinitive in The Infinitive in Anglo-Saxon. As in West-Saxon, so in the Lindisfarne Gospels, from the standpoint of Form, there are two infinitives: (1) the unin- fected, or simple, infinitive in -a 1 or in -e (occasionally in -ob or in -i) , which corresponds to the West-Saxon uninflected infinitive in -an, and which in origin is the petrified and weathered nomi- native-accusative case of a neuter verbal noun ; and (2) the unin- flected, or gerundial, or prepositional, infinitive, made up of the preposition to plus the dative case of the uninflected infinitive ending in -anne 2 (-enne; occasionally -enna, -ennce, -enni, -enno; and, with simplification of the double consonant, occa- sionally -ane, -ano, -ene). Occasionally the to is followed by an infinitive in -an 3 and occasionally by an infinitive in -ende (-endo) (by confusion with the form of the present participle), 1 According to Dr. Theodor Kolbe, in his Die Konjugation der Lindis- fwrner Evangelien, Bonn, 1912, § 205, -a occurs 620 times, -e 175 times, ■ce 16 times, and -i twice. Once, too, we find an uninflected infinitive in -an (bian, 'be,' in Mat. 1.20) ; and once we have an uninflected infinitive without any ending ( do, ' do,' in Mat. I. 16.9 ) . See, too, Dr. H. A. Carpenter's Die Destination in der Nordhumbrischen Evangelieniibersetz- ung der Lindisfarner Handschrift, Bonn, 1910, §§ 43, 51, and 77. 2 For the relative frequency of these variant forms of the inflected infinitive, see Kolbe, 1. c, § 206, who states that -arme, occurs 250 times, -enne 49 times, and the others only occasionally. He does not, however, mention the -ene form of the infinitive, of which an example is found in L. I. 8.10. 8 Kolbe, I. c, § 206, cites examples of the -an infinitive preceded by to, but not of the -ende infinitive. But we have instances of the latter, I think, in L. I. 7.16" and I. 9.14 c , in which -ende (-endo) is preceded by to; and in Mat. I. 17.13", in which to is omitted before -ende. 90 The Infinitive 91 both of which forms are counted as inflected in this investiga- tion. Very rarely, too, we have the -anne {-enne) infinitive not preceded by to; 4 and once preceded by til. 5 While the dual nature of the Infinitive — the substantival and the verbal — is constantly borne in mind, the classification of the infinitive is based primarily on its Function in the sentence. According to function we have four chief uses of the infinitive, namely, the Substantival, the Predicative (or more Verbal), the Adverbial, and the Adjectival, each term here used as defined at length in the aforementioned monograph. And these chief uses are further subdivided here as there. Accordingly I give a section to each distinctive use of the Infinitive and to certain Substitutes for the Infinitive. It will be observed that two uses not found in West-Saxon occur in the Lindisfarne Gospels, namely, the Infinitive as the Object of a Preposition and the Imperative Use of the Infinitive ; and that some new and interesting modifications of the Predicative In- finitive with Accusative Subject have been discovered in the Northumbrian gloss. Again, although I have made the use rather than the form of the infinitive the determining factor in my section-division, I have everywhere sharply separated the inflected infinitive from the uninflected. And one of my chief problems has been to try to discover the grounds of differentia- tion between the two infinitives in the several uses. It is grati- fying to find that the general principles governing the differen- tiation in West-Saxon apply with only slight modification to our Northumbrian gloss. With each use, too, I have tried to estimate the bearing of the evidence derived from the Lindisfarne Gospels on the Origin of the constructions of the infinitive not only in the Northum- brian dialect but also in West-Saxon, — one of the two chief problems of my monograph referred to above. And, as an aid to such an estimate, with each use I have given the Latin cor- 4 Mat. I. 17.13 s and, possibly, 13 b (with -mde for -enne); I. 20.11;— L. I. 10.15; 21.36 b ;— J. I. 8.4. 6 In Mat. 26.17*. 92 The Syntax of the Lindisfarne Gospels respondents of the Lindisfarne infinitives. I have, moreover, made a minute study of the Rushworth Gospels, and,, in rare or difficult passages, have compared the Rushworth rendering with the Lindisfarne. For reasons stated in my " Preface," I have not treated the Order of Words. The Voice of the Infinitive in the Lindisfarne Gospels de- serves special consideration, and is discussed in each section below. Here I need make only a few general comments on the two infinitives. Although, as in West-Saxon, the uninflected infinitive that is active in form is normally active in sense also, deviations from this norm are more common in the Lindisfarne Gospels than in West-Saxon. In several of the uses (subject- ive, objective, and predicative (with auxiliary verbs and with accusative subject)), sporadic examples occur of an uninflected infinitive that, though active in form, translates a Latin passive infinitive, and seems itself to be passive in sense. In far the majority of instances, however, the Lindisfarne uninflected in- finitive that translates a Latin passive infinitive is itself active in sense. See on this topic the paragraph on the voice of the infinitive in each of the following sections, especially in Section II, where the matter is discussed at length with reference to the objective infinitive after verbs of commanding, etc. The inflected infinitive is active in sense as in form except in the predicative use with bia(n) (wosa) to denote necessity (and occasionally — once — to denote purpose), in which it is exclu- sively passive in sense; in the predicative use with accusative subject 6 (in object clauses), in which, though occasionally active in sense, it usually translates a Latin passive gerundive, and is passive in sense ; and in the final use, in which, though normally active in sense, it is possibly once or twice passive in sense. It is evident, therefore, that the use of the inflected infinitive in a passive sense is more extended in the Lindisfarne * In what, for lack of a better name, I have denominated the Elliptical Accusative-with-Infinitive Construction, the present participle, though ac- tive in form, is usually passive in sense. See Section XV below. The Infinitive 93 Gospels than in West-Saxon, a fact arising probably from the frequency of passive gerundives in the Latin original and from the aversion of the glossator to the use of the compound passive infinitive. Of the strictly passive infinitive (that is, of the infinitive made up of bia(n) (wosa) plus a perfect participle), I have found but two examples in the whole of the Lindisfarne Gos- pels, which are quoted in the section on " the Predicative Infini- tive with Auxiliary Verbs." The passive infinitive is rare, though not so rare, in West-Saxon, and in the latter, as in the two Lindisfarne examples, is due to Latin influence. I. THE SUBJECTIVE INFINITIVE A. The Active Infinitive The active infinitive as the Subject of a finite verb is some- what more frequent in the Lindisfarne Gospels than in the West-Saxon Gospels, there being 66 examples (74 including the "Introductions") in the former to 39 in the latter. As the subject of an active verb, the active infinitive occurs about 44 times (52 times including the " Introductions ") in the Lindis- farne Gospels to 31 times in the West-Saxon Gospels; as the subject of a passive verb, it occurs about 22 times in the former to 8 times in the latter. Under the subjective use I include sentences having hit as the grammatical subject and the infini- tive as the logical subject. 1. With Active Finite Verb I consider first the active infinitive as the subject of active verbs. As in the West-Saxon Gospels, so in the Lindisfarne Gospels the subjective infinitive is more frequently uninflected than inflected, the proportion in the latter being 26 to 18 (29 to 23 inclusive of the " Introductions ") to 17 to 14 in the former. In West-Saxon as a whole, however, the subjective infinitive is more frequently inflected than uninflected, there being 252 examples of the former to 104 of the latter. The 94 The Syntax of the Lindisfarne Gospels divergence from the West-Saxon norm in the translation and in the gloss of the Gospels is probably due to the fact that, in the Latin original and in the two renderings thereof, the infinitive is less frequently in proximity to a dative-governing adjective or verb, which dative-governing word in West-Saxon 7 attracted the uninflected infinitive into the inflected form. The subjective infinitive that is active in form, whether unin- flected or inflected, seems to me habitually active in sense. Occasionally, however, it translates a Latin passive infinitive, and seems itself passive in sense, as in the following: — Mat. 18.9 b : god uel betra (5e is an ege in lif ingeonga Son tuoe ego hasbbe gesende in tintergo f yres = bonum tibi est uno oculo in uitam intrare quam duos oculos habentem mitti in gehennam ignis; Mk. 9.45 b : god is <5e halt ingeonga in lif sece Son tuoege foet hsebbe sende in tintergo fyres = bonum est tibi claudum introire in uitam aeternam quam duos pedes habentem mitti in gehennam ignis; Mk. 9.47 b : god is Se anege ingeonga in ric godes Son tuoego ego hsebbe gesende on tintergo fyres = bonum est tibi luscum introire in regnum dei quam duos oculos haben- tem mitti in gehennam ignis. But in the majority of instances, even when translating a Latin passive infinitive, the Lindis- farne subjective infinitive that is active in form is active in sense, as in the following examples: — Mat. 22. 17: is rehtlic penningslseht gesella Ssem caseri uel no ? = licet censum dari caesari an non ? 8 Mk. 2.22 : ah Saet win niwe in byttum niwum senda is rehtlic = sed uinum nouum in utres nouos mitti debet; J. 3.14: sua? gefeage uel oftersuifia gedeafnad is sunu monnes = ita exaltari oportet filium hominis ; — Mat. I. 8.9 b : seteawas fewr ana uel noht mara se reht godspellas to onfoanne = ostenditur quattuor tantum debere euangelia suscipi; Mk. 13.10: serist gerises to bodanne uel to forescegcane uel Saette he sie boden godspell = primum oportet praedicari euangelium ; L. 13.16: Sios uutedlice dohter abrahames . . . ne were geris- T See The Infinitive in Anglo-Saxon, pp. 20-26. 8 Cf . L. 20.22 : is reht us to seaXlanne Saet geafel Stem caseri ? = licet nobis dare tributum caesari? The Infinitive 95 nelic uel reht to unbindanne uel to undoanne of bend Sissum daBge symbles ? = hanc autem filia (sic) abrahae . . . non oportuit solid a uinculo isto die sabbati ? Mh. 12.14: is gelefed to seallane geaf ol (5sem caseri ? = licet dari tributum caesari ? I. The Uninflected Infinitive only is found as the subject with the following verbs: — bia(n), be. bia(n), be, in predicative combination with Adjectives (oc- casionally with Nouns) : behoflic, necessary, expedient. gewuna, custom. gelicia, please. scina, ordinarily shine, but here be allowable, translating licuit. widliga, corrupt. A few examples will suffice for illustration: — Mat. 15.20 : unSuenum uutedlice hondum eatta ne widlas Sone monno = non lotis autem manibus manducare non coinci- nant (sic) hominem. L. 12.32 : for<5on gelicade woel feder iuer gesealla iuh (5set ric = quia complacuit patri uestro dare uobis regnum. Mlc. 15.6: Serb Sone dasge Sonne symbel forgeafa gewuna w&s him enne = Per diem autem festum dimittere solebat illis unum ex uinctis (or Predicate Nominative?). II. The Inflected Infinitive only is found as the subject with the following verbs: — behofia, behoove. bia(n), be, in predicative combination with Adjectives (oc- casionally with Nouns) : darflic, useful. longsum, long, tedious. ned(e), necessary. (5arf, necessity. Typical examples are: — 96 The Syntax of the Lindisfarne Gospels Mat. I. 5.7 : Sa to talanna loiigsum is = quos enumerare longissimum est. Mat. 14.16 a : cueS . . . ned is Saet hia gegae uel Barf is him to geonganne = dixit . . . necesse ire. III. The Uninflected Infinitive and the Inflected Infinitive are each found as the subject of the following verbs : — bia(n), be, in predicative combination with Adjectives: ge(h)risenlic (-hrisnelic), fitting. god, good. reht, right, proper. rehtlic, right, proper. gehrisa : see gerisa. gerisa, be fitting. The following are typical examples : — Mat. 15.26 a ' b : ne is god to onfoanne hlaf Sara suna & sende hundum = non est bonum sumere panem filiorum et mittere canibus. Mat. 23.23 a ' b : Sses is rehtlic uel rehtlic wcere to doanne & <5a ne forhycga = haec oportuit facere et ilia non omittere. I now give, in a single alphabetic list, the complete statistics of the active infinitive as the Subject of active verbs (U. : 9 29, of which 3 are in the " Introductions " ; I. : 9 23, of which 5 are in the " Introductions ") : — behofia, behoove: U.: 9 (0).— I. 9 (1) : L. (1) : 12.12. bia(n) (wosa), be: U. (2): Mat. (1): 20.23 a .— ». (1): 10.40' a .— I. (0). bia(n) (wosa), plus an adjective (or occasionally a noun) : — — behoflic, necessary, expedient: U. (1): Mat. (1): I. 14.15.— I. (0). — darflic, useful: U. (0).— I. (1) : L. (1) : I. 3.7. — ge(h)risenlic (-hrisnelic), fitting: U. (1): L. (1): 15.32 b .— I. (3): L. (3): 2.49; 13.16; 15.32 a . 8 In the statistics here and elsewhere, U. = Uninflected Infinitive; I. - Inflected Infinitive. The Infinitive 97 bia(n) phis an adjective, continued: — gewuna, custom: U. (1): Mh. (1): 15.6.— I. (0). — god, good: U. (10): Mat. (3): 15.26 b ; 18.9 a > b .— Mk. (7): 7.27 b ; 9.43 a > b ; 9.45 a > b ; 9.47 a > b — I. (5): Mat. (2): 15.26 a ; 17 A.— Mk. (2) : 7.27 a ; 9.5.— L. (1) : 9.33. — long, long, tedious: U. (0).— I. (1) : Mat. (1) : I. 5.7. — ned(e), necessary: U. (0). — I. (1) : Mat. (1) : I. 5.8 a . — rent, right, proper: U. (1) : Mk. (1) : 13.7.— I. (4) : Mat. (1) : I. 8.9 b — Mk. (1) : I. 1.8 b — L. (2) : 17.10; 20.22. — rehtlic, right, proper: U. (5): Mat. (4): 19.3 C ; 20.15; 22.17; 23.23 b — Mk. (1): 2.22.— I. (1): Mat. (1): 23.23 a . — Sarf, necessity: U. (0).— I. (1) : Mat. (1) : 14.16 a . forstonda, be expedient: U. (1) : Mat. (1) : 19.10.— 7. (0). forwosa, be of advantage (?) : U. (0). — I. (1): Mat. (1): I. 2.14(?). gehrisa : see gerisa. geliciga, please: U. (1) : L. (1) : 12.32.— I. (0). gerisa (-hrisa), be fitting: U. (4): L. (4): I. 3.8; 17.25; 18.1 a » b — I. (4): Mk. (1): 13.10.— L. (3): 11.42 a > b ; 13.14. scina, usually shine, but here be allowable : U. ( 1 ) : Mat. (1):I. 2.13(?).— I. (0). widliga, corrupt: U. (1) : Mat. (1) : 15.20.— I. (0). 2. With Passive Verbs The active infinitive is found as the subject of passive verbs about 22 times. The infinitive is inflected 10 times. I. The Uninflected Infinitive only is found as the subject of the passive of the following verb : alefa, allow. II. The Inflected Infinitive only is found as the subject of the passive of the following verbs: — (ge) sella, give. lefa, allow. III. The Uninflected Infinitive and the Inflected Infinitive 98 The Syntax of the Lindisfarne Gospels are each found as the subject of the passive of the following verbs : — gedsefna, be fitting. (ge) deafna: see {ge)dwfna. (ge)lefa, alloiv. A few examples will suffice for these three groups : — Mk. 3.4 a > b > c : is alefed rmestdagum wel wyrce uel yfle's 6a sawele hal gedoa uel losiga-f = licet sabbatis bene fa-cere an male ? animam saluam facere an perdere ? Mat. 13.11 : forSon iuh gesald is uel wses (5set ge witte uel to uutanne clseno hryno = quia nobis datum est nosse mysteria. L. 14.3 : is gelefed on symbeldsege gelecnia? = licet sabbato curare f Mat. 14.4 : ne is gelefed Se to habbanne <5a uel hia = non licet tibi habere earn. The complete statistics of the active infinitive as the subject of passive verbs are as follows (U. : 12, all in the Gospels proper ; I. : 10, all in the Gospels proper) : — alefa, allow: U. (6) : Mk. (3) : 3.4»M— L. (3) : 5.9 a M— I. (0). (ge)dsefna (deafna), be fitting: U. (2): J. (2): 3.14; 4.20.— I. (1):L. (1): 19.5. (ge) deafna: see {ge)dcefna. (ge)lefa, allow: U. (4): Mat. (2): 12.11, 12.— Mk. (1): 10.2.— L. (1) : 14.3.— I. (5) : Mat. (2) : 12.2; 14.4.— Mk. (2) : 6.8; 12.14.— L. (1): 6.4 b . (ge)sella, give: U. (0).— I. (2): Mat, (1): 13.11.— L. (1):8.10. lefa, allow: U. (0).— I. (1) : Mk. (1) : 2.26. DIFFERENTIATION OF THE TWO INFINITIVES Whether with active or with passive verbs, the Differentiation between the Uninflected Infinitive and the Inflected Infinitive as Subject in the Lindisfarne Gospels seems in the main to rest upon the same principle as in West-Saxon : " Verbs and verbal phrases that govern a dative (or occasionally a genitive) nor- The Infinitive 99 mally have the inflected infinitive as subject, especially if the infinitive is near its principal verb." 10 Accordingly, in the Lindisfarne Gospels, as in West-Saxon, we find that, as a rule, in a series of infinitives as subjects, although the first infinitive is inflected, the succeeding one is not, presumably because of its remoteness from the principal verb, as in Mat. 23.23 a > b : Sses is rehtlic uel rehtlic wsere to doanne & (5a ne forhycga = haec oportuit facere et ilia non omittere. 11 In L. 11.42 a > b , however, each of the two infinitives is inflected despite the sep- aration of the second from its finite verb (geras) by three words. Such a deviation from the general norm by the Lindisfarne glossator is due partly to analogy and partly to the lateness of the text, it is believed. ORIGIN OF THE IDIOM As to the Origin of the Subjective Infinitive, the evidence of the Lindisfarne Gospels tends to confirm the conclusion reached from the study of the idiom in West-Saxon, as stated in The Infinitive in Anglo-Saxon pp. 183-184, namely, that in West- Saxon the active infinitive, whether uninflected or inflected, as the subject of active verbs was a native idiom only slightly in- fluenced by the Latin, but that as the subject of passive verbs it was an idiom borrowed from the Latin. This conclusion was there based on the paucity of the latter construction in the more original West-Saxon literature and on the nature of the Latin correspondents (their diversity in the one construction and their substantial uniformity in the other). As is to be expected, the correspondence between Latin original and Old English rendering is closer in the Lindisfarne gloss than in the West- Saxon translation. The Latin correspondents of the active Sub- jective Infinitive in the Lindisfarne Gospels are as follows: — (1) with an active finite verb: an infinitive that is the subject of a finite verb (U. : 24; L: 16) ; an accusative and infinitive 10 See The Infinitive in Anglo-Saxon, p. 24. 11 A complete list of the subjective infinitive in a series is given in Note 1 at the end of this section. 100 The Syntax of the Lindisfame Gospels as the subject of a finite verb (U. : 5 ; I. : 5) ; a predicative in- finitive with an auxiliary verb (U. : 0; I.: 2); — (2) with a passive verb: an infinitive that is the subject of a finite verb (U. : 11; I.: 8) ; an accusative and infinitive as the subject of a finite verb (U. : 1 ; I. : 2). APPENDIX V THE SUBJECTIVE INFINITIVE IN THE OTHEE GERMANIC LANGUAGES In the Tijdschrift voor Nederlandsche Tool- en Letterhunde, xi, 1892, pp. 165-168, Professor W. L. van Helten gives an instructive discussion of the Infinitive as Subject in Dutch, in an article entitled " Over het Gebruik van den Infinitief als Subject, Object, of in Verbinding met Dan." I have been de- lighted to find that in this article, which I first saw during the summer of 1917, Professor van Helten's explanation of the dif- ferentiation between the uninflected infinitive and the inflected infinitive as subject in Dutch and in Old High German tallies in the large with that offered by me for West-Saxon and for Old High German in The Infinitive in Anglo-Saron, pp. 20-26, 232, and in the present section suggested for the Northumbrian as represented in the Lindisfame Gospels. Says Professor van Helten: " Gelijk vanzelf spreekt, is hier aan de onbep. wijs zonder te het praedicat van oorspronkelijkheid toe te kennen. Het gerundium kan niet anders zijn dan het product eener navolging, als wier hoofzakelijke factor zonder twijfel mag worden aangemerkt de bekende neiging der Germ, dialecten om in verloop van tijd steeds meer en meer den oorspronk. enkelen infin. door dien met te te vervangen. Hoofdzakelijke, zei ik; want in een bepaald geval is de wijziging der woordvoeging langs een anderen weg en reeds tamelijk vroeg tot stand ge- komen, t. w. in zinnen, welke een adject, met een begrip ' (niet) geschikt, nuttig, gemakkelijk ' als gezegde bevatten: bij ver- binding van zulk een bijv. nw. met een nomen als subject was naturlijk het gebruik van een gerundium als bepaling van 't The Infinitive 101 adject, de van rechtswege vereischte constructie, als b. v. in 't Olid. ' sin gisiuni ist uns zi sehanne urgilo swar,' O. 4, 24, 16 ; ' hwedar ist gazelira za quedanna ? ' Erg. th. 3, 14 ; ' wedar ist odira zi quedanne? ' Tat. 54, 6; ' dhazs izs widharzuomi . . . ist eomanne zi chilaubanne ,' Is. cap. 3, § 3 ; bij koppeling van 't bijv. nw. met een infin. als onderwerp was daarentegen oorspr. alleen de onbep. w. zonder praepositie op haar plaats, als in 't Ohd ' guot ist thir zi libe ingangen wanaheilan odo halgan,' Tat. 95, 4 ; ' mir ist guot ze gote haften,' 1ST. Ps. 72, 28 ; ' wieo guot ist sament puen,' ib. 132, 1 ; door verwarring der beide constructies, d. i. ten gevolge van den invloed, door de eerste op de laatste uitgeofend, had zich intusschen een woordvoeging ontwikkeld, gelijk we waarnemen in : ' guot ist uns hir zi we- sanne/ Tat. 91, 2 ; 'ist huot ze sagenne dina gnada,' N. Ps. 91, 3 ; ' unodi ist iz harto . . . thia kleini al zi sagenne/ O. 5, 14, 3 ; ' iu ist unnuzze fore tage uf ze stanne/ ~N. Ps. 126, 2 ; ' nist biderbi zi gihiwenne/ Tat. 100, 5 ; enz. ; vgl. ook. O. S. ' god is it her te ivesanne,' Hel. 3138." Concerning the Subjective Infinitive in the other Germanic languages, see The Infinitive in Anglo-Saxon, pp. 231-233. B. The Passive Infinitive Of the passive infinitive proper, that is, of the infinitive made up of bia(n) (wosa) plus a past participle, in the subjec- tive use, I have found no example in the Lindisfarne Gospels. At times, however, as stated above in the discussion of the voice of the subjective infinitives, an active infinitive seems to be passive in sense. Occasionally the Latin passive infinitive is turned by a passive finite verb, as in L. 17.25 : serist uutedlice gehrised him feolo geSrouia uel geSolia & dcette he se forcwnen from cneoreso Sasum = Primum autem oportet ilium multa pati et reprobari a generatione hac. But more frequently we find that, by some sort of periphrasis, the Latin passive infini- tive is turned by a Lindisfarne infinitive that is active in sense as well as in form: see above, pp. 94-95. 102 The Syntax of the Lindisfarne Gospels NOTES 1. The Subjective Infinitive in a Series. — In the follow- ing passages we have a series of infinitives in which the first is inflected, but the succeeding is not: Mat. 23.23 a > b , quoted on p. 96 above; Mat. 15.26' a > b ; Mh. 7.27 a > b ; and L. 15.32 a > b . In L. 11.42 a > b , each of the two infinitives is inflected. It seems unnecessary to give the series in which each infinitive is unin- fected. 2. The Subjective Infinitive Alternates with a Clause in Mat. 14.16 : cueS . . . ned is dcet hia gegae uel darf is him to geonganne = dixit . . . necesse ire; and in L. 8.10, 13.14. 3. The Choice between an Uninfected Infinitive and an Inflected Infinitive is occasionally given, as in Mat. I. 2.14: of Sami wutetlice ne in aldum gehrine uel see sefter unseofunti- gvm trahteras uel recceras eftniwige uel girihte hwset scean ne in niwe for senig wees sengum to boetanne uel to rihtanna? uel giboeta = quibus utique nee in ueteri instrumento post septua- ginta interpretes emendare quid licuit nee in nouo profuit emendasse. II. THE OBJECTIVE INFINITIVE A. The Active Infinitive 1. With Active Finite Verb In the Lindisfarne Gospels, as in the West-Saxon Gospels, the active infinitive as the Object of an active verb is more commonly uninflected than inflected, the proportion being 129 to 32 (or 157 to 37 including the " Introductions ") in the former and 94 to 20 in the latter. The objective infinitive that is active in form, whether unin- flected or inflected, is almost without exception active in sense, I believe. In the two following passages, the active druncniga translates a' Latin passive infinitive, and is probably passive in sense: — Mat. 14.30: mi(5<5y ongann uel ongunne druncnia uel gedrince = cum coepisset mergeri; L. 12.45 e : onginneS mi(5(5y slaa (5a ensehtas & Siuwas & aetta & drinca & druncgnia uel The Infinitive 103 Saette se druncenig = coeperit percutere pueros et ancillas el edero et bibere et inebriari. Usually, however, although trans- lating a Latin passive infinitive (or occasionally a gerundive that is passive in sense), the Lindisfarne objective infinitive that is active in form is also active in sense, as in Mat. 19.7 a ' b : hwset forSon — [= blank] behead sella boc freodomas & for- leta ? = quid ergo moses mandauit dart librum repudii et dimittere ? Mat. I. 20.1 : biddendum sealla of heofnum becen = petentibus dari de caelo signum; Mat. I. 17.1 13 : ne synngige gelcerde hates = non moechandum docens iubet; Mk. 10.49: geheht hine ceiga = praecepit ilium uocari; L. 12.39: ne walde leta d'erhlidelfa hus his = non sineret perfodi domum suam; — L. 23. 2 b : Siosne woe gemitton . . . forbeadende uel woerdende gsefelo to seallanne uel (Saette se gesald Ssem caseri = hunc inuenimus . . . prohibentem tributa dari caesari ; etc., etc. In passages like these, of which there are numerous exam- ples, some scholars may prefer to consider that the infinitive here is predicative, and that the accusative, instead of being, as I believe, the object of the infinitive, is its subject. They would, of course, consider the infinitive passive in sense, though active in form. Against this sort of interpretation in West- Saxon I have urged what seemed (and seem) to me cogent reasons in The Infinitive in Anglo-Saxon, pp. 29-30 ; and these reasons, on the whole, seem to me to apply equally well to this idiom in the Lindisfarne Gospels. The only noteworthy differ- ence is this: in the Lindisfarne Gospels the accusative usually precedes the infinitive, whereas in the West-Saxon translations it usually follows the infinitive. But, as the Lindisfarne Gos- pels is an interlinear gloss, we expect, as we find, that the posi- tion of the Old English accusative will be identical with that in the Latin original. Hence the position of the accusative in the gloss is not significant, although in the West-Saxon transla- tions it is, I think, highly significant, as I have tried to show in the discussion above referred to. Concerning the voice of the objective infinitive in the other Germanic languages, sec Note 8 below and The Infinitive in Anglo-Saxon , p. 333. 104 The Syntax of the Lindisfarne Gospels I. The Uninflected Infinitive only is found as object with the following verbs : — 12 beginna (1), begin (W. S.: U. and L). bidda (2), invite (W. S. : U. and L). bodia (1), announce (W. S. : I. only). cwoeSa (4), say (W. S. : lacking). eawa (1), show (W. S. : lacking). foregehata (1), order (W. S. : lacking). forgeafa (2), allow (W. S. : the same). gearwiga (1), attempt (W. S.: lacking). geblinna (1), cease (W. S. : lacking, but the simplex, blin- nan, has U. only). gehata (7), order (W. S. : same, but gehatan, 'promise,' takes I. only). gelsera (1), teach (W. S. : lacking, but the simplex, Iceran, has T. only). gelefa (4), allow (W. S. : lacking, but the simpler, liefan, has I. only). gewiga (1) (= giwiga), demand (W. S. : lacking). gewil(l)niga (2), wish (W. S. : IT. and I.; see wil(l)niga in III below). hata (7), order (W. S. : the same). inginna (1), begin (W. S. : lacking; but see aginna in II below and onginna in III below). leta (2), allow (W. S. : the same). oncnawa (2), know (W. S. : lacking). woena (1), ween, think (W. S. : IT. and I.). WUta (4), know (W. S. : I. only). Typical examples are : — Mat. I. 19. 6 b : becon biddendum Tones becon sella bodade uel ssegde = Pharisaeis signum petentibus ionse signum dari prwdicit. Mat. I. 19.4 b , 5 : cued ebalsung in halig gast ne forletta . . . 12 The figure in parenthesis with each verb indicates the number of its occurences in this construction. For the sake of comparison, I give in succeding parentheses the construction of each verb in West-Saxon. The Infinitive 105 A: rehtnisse of word idlum in dsege domes forgelda = dicens blasphemiam in sanctum spiritum non remitti . . . et rationem de uerbo otioso in diem iudicii reddi. Mat. 27.64: gehat forSon gehalda uel Saette sie gehalden byrgenn = tube ergo custodiri sepulchrum. — Mat. I. 16.9 a : hreunisse gedo uel gewyrce hates = penitentiam agere iubens. — Mat. 7.11 : wutas ge godo gesealla sinrnm iurum ? = nostis bona dare filiis uestris ? IT. The Inflected Infinitive only is found as the object of the following verbs : — 13 aginna (1), begin (W. S. : IT. and I. ; see beginna in I above and onginna in III below). forbeada (1), forbid (W. S. : TJ. and I.; see bebeada in III below). gesoeca (1), seek (W. S. : lacking, but the simplex, secan, has U. and I. ; see soeca in III). getella (1), compute (W. S. : lacking, but the simplex, tellan, has I. only). hyhta (2), hope (W. S. : lacking, but the compound, gehyh- tan, has I. only). sella (1), give (W. S.: U. and I.). (5afiga (1), allow (W. S. : lacking, but the compound, ge- dafian, has TJ. and I.). The following are typical examples: — L. 11.38: Sonne agann bituih him getalade to cooed anne = pharisaeus an tern coepit intra se reputans dicere. L. 6.34: gif huerf gie sellas Ssem from Ssem gie hyhtad to onfoane = si mutuum dederitis his a quibus speratis accipere. L. 22.5 : dafando woeron feh him to seallanne uel geseal- lanne = pacti sunt pecuniam illi dare. III. The Uninflected Infinitive and the Inflected Infinitive are each found as the object with the following verbs: — 14 13 See the footnote to page 104. "See the footnote to page 104. In this section the first figure stands for the uninflected infinitive; the second, for the inflected infinitive. 106 The Syntax of the Lindisfarne Gospels bebeada (6 and 1), order (W. S. : the same). eftongeata 15 (1), recognize (W. S. : lacking, but the sim- plex, ongietan, has I. only) . habba (3 and 1), have (W. S. : I. only). laera (2 and 1), teach (W. S. : I. only). ondreda (1 and 2), fear (W. S. : the same). onginna (84 and 3), begin (W. S. : the same). onsacca (2 and 1), refuse (W. S. : lacking). soeca (6 and 18), seek (W. S. : the same). wil(l)niga (5 and 2), wish (W. S. : the same). The following will serve as examples :- — Mat. 19.7 a > b : hiuet for-Son— [ = blank] behead sella boc freodomaes & forleta? = quid ergo moses mandauit dari librum repudii et dimittere? — L. I. 7.12 a > b ' c : (5cgnum behead & from dserstum esuicnise to behaldane & (5a (5a(5e lichoma acuellas ne scyniga ne Sset in oehtnise husetd (sic) hia wero sprecendo geMencce = Discipulis prcecepit et a fermento hypocrisis cauere et eos qui corpus occidunt non metuere nee in persecutione quid locuturi sint cogitare. J. 8.26 a ' b : feolo ic hafo of iuih spreca uel to sprecane & doema = multa habeo de nobis loqui et iudicare. — L. 14.14: eadig (5u hist forSon hia ne habhad eft to seallanne (5e = beatus eris quia non habent retribuere tibi. Mat. I. 17.4 a , 5 : nsefrse gesueriga laeres ne . . . suiga wifi- stonda = omnino iurare docet nee . . . mutnare uolenti re- sistere. — L. I. 6.11 : eSmodnise leered & ne to forbeadane maehto on his noma doende = humilitatem docet et non prohiberi nirtutes in ipsius nomine facientem. Mat. 2.22 : ondreard Sider fara uel to fceranne = timuit illnc ire. — Mat. 1.20: nelle Su <5e ondrede uel forht bian to onfoanne maria = noli timere accipere mariam. Mat. 4.17 a>b : ongann . . . bodage & cuoecfa = coepit . . . praedicare et dicere. — Mk. 13.4: Sonne Sas alle onginnad to 15 An alternative rendering with an inflected infinitive is given in the only passage in which this verb occurs in this use. The Infinitive 107 endanne uel Satte hia se geendado = quando haec omnia incip- ient consummari. — L. 19.37 b : ongunnon alle <5a menigo ofsti- gendra gefeadon uel gefeande to herganno god = coeperunt omnes tnrbae discendentiuin gaudentes laud-are deum. — L. 21.28 b : (5a?m & (sic for uel?) (5as uutedlice to wosanne ongin- nendum = his autem fieri incipientibus. Mk. I. 3.19 b : — [blank] becon uel tacon soecendimi sealla onsoc = Pharisaeis signnm quaerentibus dari negat. — Mk. I. 1.15: ne onsoce uel nalde onsacca soSSa to tellanne uel to clcensanne him = negaret denique amputasse sibi. L. 5.18 b,c : sohton hine gebrenge & gesette fora hine = quaere- bant eum inferre et ponere ante eum. — Mat. 21.46 : sohton hine to haldanne ondreardon (5a menigo = et quaerentes eum tenere timuerunt turbas. L. 23.20 : willnade forleta Sone hselend = uolens dimittere iesum. — L. 22.15: willum ic wilnade Sis eostro gebrucca uel to eattanne huh miS = desiderio desideraui hoc pascha manducare uobiscum. — L. 23. 8 b ivces forSon willnande of menigo tid to ge- seanne hine = erat enim cupiens ex multo tempore uidere eum. The complete statistics of the active infinitive as the Object of active verbs follow (IT. : 157, of which 28 are in the " Intro- ductions " ; I. : 37, of which 5 are in the " Introductions ") : — aginna, begin: U. (0).— I. (1) : L. (1) : 11.38. bebeada, command: U. (6): Mat. (2): 19.7 a > b — 1/X\ (1): 8.6(?).— L. (3): I. 6.18; I. 7.12 b , 13»— I. (1): L. (1): I. 7.12 a . beginna, begin: U. (1) : L. (1) : 3.8.— I. (0). bidda, request: U. (2) : Mat. (1) : I. 20.1.— L. (1) : I. 9.5. —I. (0). bodiga, announce, declare: U. (1): Mat. (1): I. 19.6 b . — I. (0). CWOeSa, say, declare: U. (4) : Mat. (3) : I. 19.3 b ; I. 19.4 b , 5.—Mk. (1): 5.43 a .— I. (0). eawa, show: U. (1) : L. (1) : 3.7.--I. (0). eftongeata, recognize: U. (1): Mat. (1): I. 15.4 a (inter- 108 The Syntax of the Lindisfarne Gospels changing here with an inflected infinitive). — I. (1) : Mat. (1) : I. 15.4 a . forbeada, forbid: U. (0).— I. (1) : L. (1) : 23.2 f . foregehata, order, command: U. (1) : Mat. (1) : I. 20.16 c . —I. (0). forgeafa, allow, permit: U. (2) : Mat. (1) : 19.8.— L. (1) : 9.59.— I. (0). gearwiga, attempt, seek: U. (1) : L. (1) : I. 8.13 a ( ?). Pro- fessor Cook makes forcuoede " opt. pret. 3rd sg.," as is possible ; but I take it to be a scribal error for forcuoede. — I. (0). geblinna, cease: U. (1) : L. (1) : 5.4.— I. (0). gehata, order, command: U. (7) : Mat. (5) : I. 20. 9 b ; 8.18 ; 14.9; 27.58, 64.— Mk. (2): 6.27; 10.49.— I. (0). gelaera, teach: U. (1) : Mat. (1) : I. 17.1 b .— I. (0). gelefa, allow, permit: U. (4): Mat. (1): 24.43.— L. (3): 4.41 c ; 8.32 b ; 9.61.— I. (0). gesoeca, seek: U. (0).— I. (1) : J. (1): 11.8. getella, compute, reckon: U. (0). — I. (1): L. (1) : 14.28 b (?). gewiga (for giwiga), demand, request: U. (1): J. (1) : I. 6.17 a .— I. (0). gewil(l)niga, wish, desire: U. (2): Mat. (2): 13.17 a > b — I. (0). habba, have: U. (3): /. (3) 8.26 a >b; 16.12.— I. (1): L. (1): 14.14. hata, order, command: U. (7) : Mat. (4) : I. 1.11 ; I. 16.9 a ; T. 17.2: T. 20.13.— Mk. (1): 8.7.— L. (2): 8.55 a ; 19.15.— I. (0). hyhta, hope: U. (0).— I. (2) : L. (2) : 6.34; 22.8 C . inginna, begin: U. (1) : L. (1) : 14.9.— I. (0). lsera, teach, instruct: U. (2): Mat. (2): I. 17.4 a , 5.— I. (1): L. (1): T. 6.11. leta, let, allow: U. (2) : Mk. (1) : 1.34 a — L. (1) : 12.39.— I. (0). oncnawa, know: U. (2) : Mk. (1) : I. 2.3( ?) ; I. 2.4( ?).— I. (0). The Infinitive 109 ondreda, dread, fear: U. (1): Mat. (1): 2.22.— I. (2): Mat. (1): 1.20.— L. (1): 9.45. onginna, begin: U. (85) : Mat. (16) : I. 7.10 ; 4.17 a > b ; 11.7 a , 20; 12.1 a > b ; 14.30; 16.21; 18.24 a ; 24.49; 26.22 b , 37 b > c , 74 a » b . — Mk. (36): I. 1.10 a ; 1.45 a > b ; 2.23; 4.1; 5.17, 18; 6.2, 7, 34, 55; 8.11 a , 25, 31 a , 32; 10.32, 41, 47 a > b ; 11.15; 12.1; 13.5; 14.19 a > b , 33 a > b , 65 a > b > c > d , 69, 71 a > b , 72; 15.8, 18.— L. (30): 4.21; 5.21 a ; 7.15, 24 a , 38, 49: 9.12; 11.29 b , 53 a > b ; 12.1 b , 4 5 b,c,d,e. I3.25 a > b ; 26; 14.18, 29, 30 b ; 15.14, 24, 28; 19.45; 20.9; 21.7 b ; 22.23; 23.2 a , 30.— J. (3) : 4.47; 13.5 a > b .— I. (3) : Mk. (1) : 13.4.— L. (2) : 19.37 b ; 21.28 b . onsacca, refuse, deny: U. (2) : Mk. (1) : I. 3.19 b — L. (1) : I. 8.8 a . Professor Cook considers gescilde here " ind. pret. 3rd sg.," as is possible, but I take it to be an infinitive. — I. (1) : Mk. (1): T. 1.15. sella, give: U. (0).— I. (1) : /. (1) : 5.26. SOeca, seek: U. (6) : L. (3) : 5.18 b > c ; 17.33.—/. (3) : 8.37; 10.39; 19.12 a — I. (18): Mat. (1): 21.46.— Mk. (1): 12.12. —L. (7): 6.19; 9.9; 11.54; 13.24 b ; 19.3, 47; 20.19.—/. (9) : I. 5.3 b ; I. 5.5 b ; 5.18 a ; 7.1, 19, 20, 25, 30; 8.40. Safiga, consent, agree: U. (0).— I. (1): L. (1): 22.5. wil(l)niga, wish, desire: U. (5): Mat. (1): 12.38. — L. (2): 17.22; 23.20.—/. (2): I. 6.14 b ; 7.44.— I. (2): L. (2): 22.15; 23.8 b . woena, ween, think: U. (1) : J. (1) : 5.39. — I. (0). WUta, know: U. (4) : Mat. (2) : 7.11 ; 16.3.— L. (2) : 11.13, 12.56.— I. (0). 2. With Passive Verbs The active infinitive as the retained object of passive verbs occurs about six times in the Lindisfarne Gospels. Of these six infinitives, two are inflected. The following examples are typical : — L. I. 7.10 a : mi(5 rclarua gehriordage gebeden wees = Apud pharisaeum prandere rogatus. J. I. 8.4 : gehaten is Segnum gesaeccanne = iubetur discipu- lis nuntiare. 110 The Syntax of the Lindisfarne Gospels The complete statistics of the active infinitive as the object of passive verbs are as follows (IT. : 4, all in the " Introduc- tions " ; I. : 2, of which 1 is in the " Introductions ") : — (ge)bidda, invite: U. (1) : L. (1) : 7.10 a — I. (0). (ge)hata, order, command: U. (0). — I. (1) : /. (1) : I. 8.4. (ge)lefa, allow, permit: U. (0).— I. (1): Mat. (1): 12.4 (see Xote 6 at the end of this section). (ge)neda, compel: U. (1) : Mat (1) : I. 7.3 a (or Consecu- tive?).— I. (0). (ge)Sreatiga, compel, force: U. (2): Mat. (2): I. 6.19; I. 7.2 (or Consecutive in each example?). — I. (0). DIFFERENTIATION OF THE TWO INFINITIVES The Differentiation between the Two Infinitives as Object in the Lindisfarne Gospels appears to rest upon the same general principle as in West-Saxon: the uninilected infinitive usually appears to the modern Englishman as a direct object, and is oftenest found with verbs whose noun object would be in the accusative case. On the other hand, the inflected infinitive, as a rule, appears to the modern mind as an " indirect object," under which phrase are included genitive (occasionally also instrumental) objects as well as dative objects, and is usually found after verbs whose noun objects would be in the genitive or the dative (occasionally in the instrumental). The double construction of uninflected infinitive and inflected infinitive is usually due to the double regimen of the governing verb, as I have sought to show is true in West-Saxon. 16 As my parallel columns above show, not a few apparent, if not real, divergences from the West-Saxon custom appear in the Lindisfarne Gospels. But these divergences do not seem to me in any appreciable degree to invalidate the general principle stated. The chief divergences consist in the fact that, in the Lindisfarne Gospels. some verbs {heginna, bidda, gewilniga) have the uninflected infinitive only as object, or some verbs {aginna, forbeada, (ge)- 16 See The Infinitive in Anglo-Saxon, pp. 60-71. The Infinitive 111 soeca, and Safiga) have the inflected infinitive only as object, whereas in each group in West-Saxon these verbs are found with both infinitives as object. But, as in almost every one of these instances the objective infinitive is found only once in the Lindisfarne Gospels, the divergence may arise, not from a dif- ference of idiom, but from paucity of texts in the Northumbrian dialect. Again, the objective infinitives (whether uninflected or inflected) that occur in the Lindisfarne Gospels but not in West-Saxon, seem to me in the large to conform to the general principle stated above. Moreover, in the Lindisfarne Gospels, with the more numerously recorded objective infinitives, of whichever variety, the usage is almost identical with that in Anglo-Saxon. I believe, therefore, that, despite the number of divergences between the objective infinitive in the Lindisfarne Gospels and in West-Saxon, some of which may be due to anal- ogy and some to the lateness of the text, no one will find in these divergences any ground for questioning the essential correctness of the general principle above laid down as explaining the use of the one infinitive or the other. On the contrary, I believe that the evidence of the Lindisfarne Gospels strengthens the theory that was originally based on West-Saxon texts only.- ORIGIN OF THE IDIOM Owing to the small number of verbs occurring more than once or twice with the Objective Infinitive, the Lindisfarne Gospel throws little light on the Origin of the idiom. So far as the evidence goes, however, it seems to me to tend to confirm the general results obtained for West-Saxon in The Infinitive in Anglo-Saxon, pp. 185-192. Of the verbs having only the unin- flected infinitive as object, listed on p. 104 above, only two (gehata, ' order,' and hata, ' order ') occur over a half dozen times ; and with these the usage in the Lindisfarne Gospels is identical with that in West-Saxon, in each of which the idiom seems native. The Latin correspondents for gehata are an objective infinitive active, 1 ; a passive infinitive with accusa- tive subject, 6; for hata: an objective infinitive, 3 (active, 2; 112 The Syntax of the Lindisfarne Gospels passive, 1) ; a passive infinitive with accusative subject, 4. Of the verbs having only the inflected infinitive as object, only one occurs over a single time, and that only twice (hyhta, ' hope ') — a basis too slender upon which to base a theory. Of the verbs having as object both the uninflected infinitive and the inflected infinitive, only the following occur over a half dozen times : bebeada, i order ' ; onginna, ' begin ' ; soeca, 1 seek'; and wil{l)niga, l wish.' The usage with each is iden- tical with that in West-Saxon, and the objective use with each seemed native in West-Saxon for reasons stated in the discussion referred to above. The Lindisfarne evidence seems to me con- firmatory of that conclusion. The Latin correspondents for these words respectively are: — for bebeada: U. : an objective infinitive, active, 5 ; a passive infinitive with accusative subject, 1 ; — I. : an objective infinitive, active, 1 ; — for onginna: U. : an objective infinitive (active, 82; passive, 2), 84; a final infini- tive, 1 ; — I. : an objective infinitive (active, 2 ; passive, 1), 3 ; — soeca: — U. : an objective infinitive, active, 6; — I.: an objective infinitive, active, 18; — wil(l)niga: — U. : an objective infini- tive, active, 5 ; — I. : an objective infinitive, active, 2. The active infinitive as the retained object is due to Latin influence, seldom occurring except when the Latin original has the same idiom. For the active objective infinitive as a whole, in the Lindisfarne Gospels, the Latin correspondents are as follows : — (1) with active verbs: — U. : an objective infinitive, 138 (active, 130; passive, 8) ; a passive infinitive with accusative subject, 18; a gerundive without esse and passive in sense, 1; — I.: an objective infinitive, 34 -(active, 33 ; passive, 1) ; a passive infini- tive with accusative subject, 2 ; a noun in the accusative, 1 ; — (2) with passive verbs: — TJ. : an objective infinitive, active, as retained object of a passive verb, 3 ; a subjunctive, active, 1 ; — I. : an objective infinitive, active, as the retained object of a passive verb, 1 ; a subjective infinitive, active, 1. For the Objective Infinitive in the other Germanic languages, see The Infinitive in Anglo-Saxon, pp. 233-236, and the articles on the Infinitive in Dutch by Professor W. L. van Helten, cited in my " Bibliography." The Infinitive 113 B. The Passive Infinitive Of the strictly passive infinitive (that is, the infinitive made up of bia(n) {wosa) plus a past participle) as the object of a verb, I have found no clear example in the Lindisfarne Gospels. Usually the Latin passive infinitive is rendered by an active infinitive in the Lindisfarne Gospels, as illustrated above, pp. 102-103. i NOTES 1. The Objective Infinitive in a Series. — In the following we have a series of two or more objective infinitives the first of which is inflected, but the succeeding ones are not: — L. I. 7.12 a > b , 13 a : (5egnum behead & from dserstum esuicnise to be- haldane & (5a (5a(5e lichoma acuellas ne scyniga ne (5aet in oeht- nise huaetd (sic) hia wero sprecendo gedenco? = Discipulis praecepit et a fermento hypocrisis cauere et eos qui corpus occi- dunt non metuere nee in persecutione quid locuturi sint cogi- tare. — I do not give here the series in which each infinitive is uninflected. 2. The Choice between an Uninfected Infinitive and an Inflected Infinitive is given at times, as in the following: — L. 22.15: willum ic wilnade Sis eostro gebrucca uel to eattanne iuih miS = desiderio desideraui hoc pascha manducare nobis- cum; Mat. I. 15.4: eftongeattas . . . lufu geworden Sing gesella uel to sellanne = recognoscant . . . fidem f actae rei tradere; Mat. 2.22: ondreard Sider fara uel to fceranne = timuit illuc ire; MJc. 1.34 a : nalde leta spreca hia uel ne lefde hia to spreccanne = non sinebat loqui ea; L. 21.7 b : mi(5(5y ge- worda uel Ssette sie uel to wosanne onginnad = cum fiieri incip- ient; J. 8.26 a : feola ic hafo of iuih spreca uel to sprecanne 17 & doema = multa habeo de uobis loqui et iudicare; J. 16.12 : gett uel forSor menigo uel feolo ic hafo iuh gecuoeda uel to scegc- ganne 17 = adhuc multa habeo uobis dicere; J. 8.37: ah gie " The infinitives may depend on the substantive rather than the verb : see The Infinitive in Anglo-Saxon, p. 43. 8 114 The Syntax of the Lindisfarne Gospels soecas mee gecearfa uel to accuellanne = sed quaeritis me inter- ficere; J. 10.39: ib. 3. An Interchange of Objective Infinitive and of Objective Clause occurs in the following: — Mat. 27.64: gehat forSon gehalda uel doette sie gehalden byrgenn = iube ergo custodiri sepulchrum ; Mk. 13.4, quoted above, p. 106; L. 9.59: forgef uel gelef me aerist geonga & 8a?t ic byrga feeder min = permitte mihi primum ire et sepelire patrem meum; L. 21. 7 b , quoted above, in ISTote 2, for another purpose; L. 12.56 : onsione earSes . . . wutad gie gecunnia uel (5sette see gecostad = faciem terrae . . . nostis probare; L. 12.45 e , quoted above, p. 102; L. 23. 2 b , quoted on p. 103 above. 4. .1 Lindisfarne Noun in -Ing (-Ung) Translates a Latin Objective Infinitive in L. 7.45: ne blann cossetunges foeta mine = non cessauit osculari pedes meos (Rushw. : lacking ; W. S. : ne geswac <5set heo mine fet ne cyste). See JSFote 3 to Chapter II, p. 65, for Professor Curme's contention that cossetunges is here a gerund, and that foeta is its accusative object. See, also, Note 4 to Section xiii of the present chapter, for an example of an inflected infinitive interchanging with a verbal noun in -nes(s). 5. Infinitive after a Finite Verb Passive in Form, but Active in Sense. — In the following passages we have an active infinitive that is the object of finite verbs that are passive in form, but active in sense, the passive forms being due to the glossator's mistaking the Latin deponent verbs for passives: — (ge)cunniga, ' attempt ' : Mat. I. 5.2 and I. 5.9: — forgetta 1 forget ' : Mk. S.14. — For the same phenomenon in West-Saxon, see The Infinitive in Anglo-Saxon, p. 72. 6. A Mixed Construction occurs in Mat. 12.4: hlafas . . . gebrec <5a neron gelef ed him to gebrucanne = panes . . . quos non licebat ei edere. We seem to have a cross between the sub- jective use of the infinitive (Sa n-ces gelef ed him to gebrucanne) and the normal use of the retained objective infinitive (Sa he nces gelef ed to gebrucanne). Although I have listed to ge- The Infinitive 115 brucanne in the passage under discussion as an imperfect exam- ple of the retained objective, the infinitive might be considered as denoting purpose. 7. An Inflected Objective Infinitive without " To " occurs in J. I. 8.4, quoted on p. 109 above. 8. The Voice of the Infinitive in Old High German. — A very thoughtful discussion of the Voice of the Infinitive in Old High German is given by Dr. Paulus Kaufmann, in his Leipzig dissertation, Ueber Genera Verbi im Althochdeutschen Be- sonders bei Isidor und Tatian, Erlangen, 1912, which appeared subsequent to the sending of The Infinitive m Anglo-Saxon to the printer, and was procured by me only recently. As its title indicates, this dissertation considers the voice of the infinitive not only in its Objective use, but in all the uses that are repre- sented in the texts studied by its author. As is evident from the following quotation from § 66 of this dissertation, Dr. Kauf- mann is more insistent than I am in his contention against the passive rendering of a Germanic infinitive that is active in form: " Passive Bedeutung hat der Infinitiv jedoch niemals, wie das fiir die Falle verschiedentlich angenommen wurde, wo er zur Wiedergabe von lat. passivem Infinitiv oder lat. Gerund- ivum dient. Es ist jedenfalls ein am Lateinischen ausgebildetes und an seinem Formensystem sich orientierendes Sprachgefuhl, das hier passives Genus Verbi zu empfinden glaubt. Die Sache liegt vielmehr so, dass in solchen Fallen der deutsche Infinitiv teils activisch und teils neutral ist. Danach sind die Belege aus Isidor und Tatian in zwei Klassen zu sondern." Many of Dr. Kaufmann's neutral infinitives, given in his § 68, seem to me passive in sense, as in Isidor 93.1: filius hominis tradendus est in manus hominum = mannes sun ist zi sellenne in hant manno. III. OTHER SUBSTANTIVAL USES OF THE INFINITIVE Of Substantival Uses of the Infinitive Other than as Subject and as Object only a very few examples occur in the Lindisfarne Gospels. 116 The Syntax of the Lindisfame Gospels A. As a Predicate Nominative Of the infinitive as a Predicate Nominative I have found but two examples, the first one of which is doubtful, the finite verb having to be supplied in both Latin and Northumbrian : — Mat. I. 1.4: arwyrfte were uel winn ah froecenlic foreonfeng uel dearfscip sengum to doemenne of oSrum = pius labor sed peri- culosa praesumptio iudicare de ceteris ; — J. 19.40 : sua Seau iuSeum is bebyrge uel bidelfa = sicut mox (sic for mos) iudeis est sepelire. In one of these examples, the infinitive is inflected, probably because of its proximity to a datival adjective or noun. The examples of the infinitive as a predicate nominative are too few in the Lindisfame Gospels to determine the origin of the idiom. But the probability is that the idiom is partly of native origin and partly of foreign origin, the reasons for which view are given in The Infinitive in Anglo-Saxon, p. 193. B. As an Appositive Somewhat more frequent is the use of the infinitive as an Appositive to a noun or pronoun. In this use the infinitive is usually uninflected, but occasionally it is inflected owing to its proximity to a dative-governing adjective. The examples in full are as follows : — 1. Uninflected: Mat. 9.5 a ' b : hucet is eabur cuoeda forgefon bioon (5e synna uel cueda aris ? = quid est f acilius dicere dimittuntur tibi peccata aut dicere surge ? MJc. 2.9 b : Hwcet is eaSur to coeSanne Ssem eorScryple for- gefen bioon oe synno uel cuoeda aris ? = quid est f acilius dicere paralitico dimittantur tibi peccata an dicere surge ? L. 5.23 a > b : huoeder is eaoor gecuoa?8a forgefen bioon <5e synna uel cuoeda aris = quid est facilius dicere dimittuntur peccata an dicere surge ? 2. Inflected: Mh. 2.9 a , already quoted above. The Infinitive 117 In Mk. 2.9 a ' b we have a series of two appositive infinitives the first of which, in proximity to a dative-governing adjective, eadur, is inflected, but the second is not. But, in Mat. 9.5 a and L. 5.23 a , the infinitive, though next to eadur, is not inflected. The situation in the West-Saxon is much the same both as to the infrequency of the appositive infinitive and as to the inflec- tion of the same. See The Infinitive in Anglo-Saxon, pp. 75-77. For reasons stated in the discussion of the origin of the appositive use of the infinitive in West-Saxon, it is probable that the idiom is partly native and partly foreign. See The Infinitive in Anglo-Saxon, p. 193. C. As the Object of a Preposition Aside from the inflected infinitive made up of the preposition to plus a dative of the verbal noun in -ne, which we regard as a unit and which, therefore, does not strictly fall under the above heading, I have found a few cases of the infinitive used as the Object of a Preposition. In all the examples except one, the infinitive is inflected. The examples in full are as follows: — 1 . Uninfl ected : L. I. 3.1 : bodade in monnes gecerrce in him <5erh sunu dyde — predicans in hominis redire in se per filium faceret. 2. Inflected: Mat. I. 21.10: gecunnedon of Seem gaefel uel gyld cseseres to seallanne ofer inwurittena uel licnessa miS frasung gefselde = Temtantes de tributo caesaris dando supra inscribtionis uel imaginis interrogatione destruxit. L. I. 9.16: gefraegn Sone weligo from to oyenna lif ece (5iu refterra onsuare unrod gecearf = Interrogans diues de possi- denda uita aeterna secunda responsione tristis abscidit. L. I. 10.8 : costende of gsefel cseseres to forgeldanne penninges onwriting forcuom = Temtantes de tributo caesaris reddendo denarii inscribtione conuincit. L. I. 9.12: to symble biddanne geddung from widua setteS wraco from wiSernarde his from unrehtwis doeme giwigende ■= 118 The Syntax of the Lindisfame Gospels Ad semper orandum parabolam de uidua ponit ultioneni de aduersario suo ab iniquo iudice postulante. Although we have the preposition to here plus a dative in -ne, the two words do not, I think, form a logical nnit as in the ordinary infinitive with to. Although some of the above examples are doubtful, some of the infinitives seem to me indisputably to be governed by a preposition. In all of the examples, the use of the preposition in the Lindisfame Gospels is in imitation of the idiom in the Latin original. Moreover, the differentiation between the un- inflected infinitive and the inflected infinitive rests upon the differences in the Latin original : in the former the Latin has a simple infinitive; in the latter, a gerund or gerundive, usually the latter. In West-Saxon no clear example was found of the infinitive as the object of a preposition : see The Infinitive in Anglo- Saxon, p. 78. In all three uses, the infinitive, whether uninflected or in- flected, is active in sense as well as in form. Concerning the Infinitive in Substantival Uses other than as subject or as object, in the other Germanic languages, see The Infinitive in Anglo-Saxon, pp. 236-237. IV. THE PREDICATIVE INFINITIVE WITH AUXILIARY VERBS A. The Active Infinitivi; As in West-Saxon, the uninflected infinitive as the Predica- tive Complement to Auxiliary Verbs is so frequent in the Lindisfame Gospels as not to call for collection. Except in very rare cases, cited in Note 1 below, this infinitive is active in sense as well as in form. Of the inflected infinitive as the predicative complement of auxiliary verbs I cite all the examples observed by me in the Lindisfame Gospels: — With agan, owe, ought: Mat. 18.24 b : gebroht waes him enne seSe ahte to geldanne tea The Infinitive 119 Susendo craeftas = oblatus est ei unus qui debebat decern milia taleuta. Mat. 18.28 a ' b : substantially as above. L. 7.41 : an ahte to geldanne penningas fit' hund = unus debebat denarios quingentos. L. 16.5: substantially as above. With dear(r), dare: Mat. I. 1.9 : godes wracco Se ich (sic) darr huelc hwoego in aldum bocum to eccanne uel ic toecade, geswiga, gegema = sacrilegum qui audeam aliquid in ueteribus libris addere, mutare, corrigere. With willa, will, desire : Mat. 5.40: Ssem seSe wil Sec miS to dome geflitta & cyrtel uel hra?gl Sin to niomanne forlet & hrsegl uel haecla uel bratt = illi qui uult tecum iudicio contendere et tunicam tuam tollere remitte et pallium. As is evident, the inflected infinitive occurs chiefly with the datival verb, agan, as is true, also, in West-Saxon. B. The Passive Infinitive Of the passive infinitive used predicatively with auxiliary verbs, I have found but two examples in the Lindisfarne Gos- pels, namely, Mat. 26.9' a > 1> : mcehte forSon Sis wosa uel were biboht in micil feh & wosa said Sarfum = potuit enim istud uenundari multo pretio et dari pauperibus. The passive infini- tive with auxiliary verbs in the Lindisfarne Gospels, as in West- Saxon, is clearly due to Latin influence. See, too, Note 1 below. For the Predicative Infinitive with Auxiliary Verbs in West- Saxon (where the situation is quite similar to that in the Lindisfarne Gospels), and in the kindred Germanic languages, see The Infinitive in Anglo-Saxon, pp. 79-88, 194, and 237-238. NOTES 1. Predicative Infinitives Active in Form, but Passive in Sense?— In the following passages we seem to have predicative infinitives that are active in form but passive in sense: — Mat. 120 The Syntax of the Lindisfame Gospels 5.14 a : ne mo?g burug uel ceastra gehyda uel gedeigla ofer nior geseted = non potest ciuitas abscondi supra monte posita (Rushw. : beon ahyded; W. S. : beon behyd) ; Mat. 2.18 c : Sset wif woeap suna hire & nalde froefra forSon ne sint = rachel plorans filios suos et noluit consular! (sic for consolari) quia non sunt (Rushw. : beon afroefred; W. S. : beon gefrefred) ; Mat. 10.34 a : Netted ge gedoema uel dcet ge se domad forSon ic cuom to sendenne sibbe = Nolite arbitrari quia uenerim mittere pacem (the glossator mistaking the deponent, arbitrari, for the passive?). — See, also, Notes 5 and 6 below. 2. Predicative Infinitive as Complement to Gesea, 'Seem.' — In Mat. I. 3.1 b we have a predicative infinitive as the comple- ment to gesea, ' seem ' : ana (5a (5oht gesegon gecerde uel gevixla = tantum quae sensum uidebantur mutare. [Professor Cook, in his Glossary, makes gecerde Ind. Pret. 3rd sing., but to me it seems a past participle, in the predicate nominative plural ; and it may be that the glossator intended that the user of his book should put gevixla into the same participial form.] 3. The Predicative Infinitive in a Series with Auxiliary Verbs. — In Mat. I. 1.9 a > b ' c , quoted on p. 119 above, we have a series of three infinitives the first of which is inflected, but the others are not. On the other hand, in Mat. 5.40, quoted on p, 119 above, we have a series of two infinitives the first of which is uninflected, but the second, contrary to the general rule, is inflected. 4. An Alternative Rendering of a Latin Deponent Infinitive by either an Active Infinitive or by a Passive Subjunctive is given in Mat. 10.34 a , quoted above in Note 1. 5. An Abortive Attempt at Rendering a Latin Passive In- finitive Complementary to " Debeo" is found in Mat. 3.14: cue(5 ic from (5e rehtra is gefulwia = dicens ego a te debeo baplizari (Rushw. : cweSende ic sceal from t5e beon uel wesa deped uel fullwihted; W. S. : cwaeS: Ic sceal fram Se beon geftdlod). The Infinitive 121 6. The Adverbial Use of the Auxiliary " Mwg " occurs in the phrase eade mwg {mwge, mwhte), as in L. 21.34: behaldaS Sonne iuih eade mcege uel Ssette ne sie ahefigad hearto iuero = Attendite autem uobis ne forte grauentur corda sua (Rushw. : eada mcege; W. S. : de Ices) ; Mat. 11.23 eada mwhte uel eade mcege were wungiende uel Ssette hia gewunadon wi(5 Sonne ondueard dseg uel Siosne ondueard dsege = forte mansissent usque in hunc diem (Rushw.: wen; W. S. : witodlice). Unin- telligible to me is the use of this phrase in Mat. 12.33 : gif ec soSlic from wssstm treo oncnaua eada mceg = si quidem ex fructu arbor agnoscitur (Rushw.: ford on de of western bid treow ongeten; W. S. : Witodlice be Sam wsestmme by<5 Sset treow oncnmven) . [In his marginal note to this passage, Skeat says of eada mceg: " intended as a further gloss to si quidem." I am inclined, however, to think that mceg is here a verb, not, as classified by Professors Cook and Skeat, an adverb ; and that oncnaua is the infinitive complement of mceg, and is here passive in sense, though active in form, because of the Latin passive, agnoscitur. ] 7. The Uninflected Infinitive after " Agan." — In his Syn- tax im Poema Morale, a Breslau dissertation of 1914, p. 33, Dr. Walter Preusler declares that the uninflected infinitive is not found in Anglo-Saxon after agan, ' owe,' ' ought.' Four examples of this construction, however, had been recorded by me in The Infinitive in Anglo-Saxon, p. 81 ; but that monograph had appe?red too recently to be accessible to Dr. Preusler. V. THE PREDICATIVE INFINITIVE WITH VERBS OF MOTION AND OF REST OTHER THAN " WUTUN " Of the Predicative Infinitive with Verbs of Motion, as in the Martyrology 26.10 (culfre com fieogan of heofonum ond gesset ofer his heafde, 'the dove came flying from heaven,' etc.), in which the infinitive, instead of denoting purpose, seems equiva- lent, in modern English, to a predicative present participle, no example has been found in the Lindisfarne Gospels. Nor was 122 The Syntax of the Lindisfarne Gospels an example of this construction found in the West-Saxon Gos- pels. The idiom is found frequently in Anglo-Saxon poetry and occasionally in the prose. Of a similar predicative use of the infinitive after Verbs of Rest, as in iElfric's Lives of Saints 512. 417 (Seer (5set scrpef wses tomiddes Se <5a seofon halgan lagon inne slapan, ' where the cave was in which the seven saints lay sleeping ' ) , no exam- ple occurs either in the Lindisfarne Gospels or in the West- Saxon Gospels. The idiom is very rare in Anglo-Saxon. Concerning each of these two idioms in West-Saxon and in the other Germanic languages, see The Infinitive in Anglo- Saxon, pp. 89-92, 194-199, and 238-239. Compare, too, Dr. J. H. Kern's De met het Participium Praeteriti Omschreven Werhwoordsvormen in 't Nederlands, Amsterdam, 1912, §§ 61 and 329, and the references there given to Stoett; also van Helten 4 , pp. 230-232. VI. THE PREDICATIVE INFINITIVE WITH " WUTUN " 18 A. Tiie Active Infinitive Of the active infinitive as Predicative Complement to Wutun, i let,' about seven examples occur in the Lindisfarne Gospels. The infinitive is always uninflected, and is active in sense as well as in form. The examples in full are as follows : — ga(geonga), #o (4): Mat. 26.46: wutum geonga = eamus. So: Mh. 1.38, 14.42; J. 14.31. gesea, see (1) : Mat. 27.49 a : wutun we gesea hwe(5er cyme = uideamus an ueniat. ofsla, slay, hill (2) : Mh. 12.7: cymes wutum we ofsla hine = uenite occidamus eum; — L. 20.14: wutun of sice hine = occidamus ilium. In each of these examples wutun plus the infinitive translates 18 The regular Lindisfarne form is either ivutum or icutun. The Infinitive 123 a Latin subjunctive. In one example, Mk. 12.7, both in the Latin and in the Northumbrian gloss, an adhortative (uenite = cymes) precedes ivutun. ORIGIN OF THE IDIOM The paucity of examples of the Predicative Infinitive with Wutun makes it unwise to dogmatize about the origin of the idiom in the Lindisfa/rne Gospels. But the evidence seems to support the theory that the idiom in West-Saxon was native, but was somewhat helped on by the occurrence of the sub- junctive in the Latin original. See The Infinitive in Anglo- Saxon, pp. 93-96, 199-200; and for this idiom in Old Saxon, p. 239 of the same. B. The Passive Infinitive Of the passive infinitive as the predicative complement of ivutun I have found no example in the Lindisfarne Gospels. Note. An Alternation of the Analytic " Wutun" Plus an Infinitive with the Synthetic Subjunctive is found twice: Mk. 1.38 : gee we uel wutum geonga = eamus; similarly Mk. 14.42. VII. THE PREDICATIVE INFINITIVE WITH " BIA(N) " (" WOSA ") A. The Infinitive Denotes Necessity or Obligation The Predicative Infinitive with Bia(n) (Wosa) Denoting Necessity or Obligation in the Lindisfarne Gospels seems to me always passive in sense, as in L. I. 11.1 : Saette neh eostro is to redenne = Quod propre pascha legendum est. In West-Saxon, however, this infinitive is occasionally active in sense. Without exception the infinitive is inflected. I cite all the examples found, arranged alphabetically (21, of which 17 are in the " Introductions ") : — ahoa, crucify (1) : J. (1) : 19.16. awrita, write (1) : J. (1) : 21.25. 124 The Syntax of the Lindisfarne Gospels behalda (bi-), hold, guard against (3) : Mat. (2) : I. I7.15 b ; I. 20.2.— L. (1): I. 2.10. geafa, give (1) : Mat. (1) : I. 1.12. gelefa, believe (1) : J. (1) : I. 4.2 b . reda, read (3): L. (1): I. 11.1.— J. (2): I. 4.19; I. 5.8 a . sella, give (1) : Mat. (1) : I. 17.13 b (in which sellende is for sellenne without £of). senda, send, put (1) : L. (1) : 5.38. setta, set, place (1) : Mb. (1) : I. 3.5 a . SOeca, see& (5) : Mat. (4) : I. 1.12; I. 2.11; I. 4.4 a ; I. 9.19. — Mh. (1) I. 4.8*>. swiga (suiga), keep silent (1) : Mat. (1) : I. 15.5. tOSOeca, seek, return (Latin referre) (1): L. (1): I. 9.9 a (woero tosceccenne; perhaps should be put under the simplex, soeca). WUndriga, wonder, admire (1) : /. (1) : 9.30. Of these twenty-one examples, fifteen translate Latin gerun- dives (of which one is preceded by ad), a correspondent which is quite common in West-Saxon, also, and which in both dialects often accounts for the use of the inflected infinitive of necessity. The other Latin correspondents are: an appositive participle, passive, 1 ; an objective infinitive, active, 1 ; a present passive infinitive, 1 ; a subjunctive passive, 1 ; a substantival participle, present, 1 ; an adjective in -bilis, 1. B. The Infinitive Denotes Futukity Of the Infinitive Denoting Futurity, I have found no clear example in the Lindisfarne Gospels, though it occurs several times in West-Saxon. Where the predicative infinitive with beon (wesan) denoting Futurity is found in the West-Saxon Gospels, the Lindisfarne Gospels use (1) a predicate participle (usually present, but occasionally past) or (2) a predicate adjective or (3) a predicative verbal noun: — (1) Mat. 11.3 : W. S., Eart (5u (5e to cumenne eartf = Lind., ar(5 (5u se(5e tocymende wees uel isf==ta es qui uenturus es? The Infinitive 125 Similarly: Mat. 16.27; L. 7.19, 20; L. 10.1; J. 1.15.— L. 22.23 : W. S., to donne woere = Lind., doend were = facturus esset. L. 9.31 : W. S., his gewitendnesse (5e he to gefyllende («— #e- fyllenne?) wees on hierusalem = Lind., tofser his <5one (stc) scealde gefylled wosa uel wses in h. = excessum eius quern com- pleturus erat in h. — Mat. 17.12: W. S., swa ys mannes sunu eac fram him to Srowigenne = Lind., sunu monnes ge&rowed bid from him = et filius hominis passurus est ab eis. [Skeat in a marginal note to gedrowed: " Ms. gefirowend altered to gedrowed."] (2) Mat. 11.14: W. S., he ys helias (5e to cumenne ys = Lind., Se is — [= blank] seSe toueard woes = ipse est helias qui uenturus est. (3) L. 24.21: W. S., we hopedon Saet he to alysenne wcere israhel = Lind., woe . . . gehyhton Ssette he were eftlesing israeles = nos autem sperabamus quia ipse esset redemturus israhel. C. The Infinitive Denotes Purpose Of the inflected infinitive with bia(n) (wosa) denoting Pur- pose, only two examples occur in the Lindisfarne Gospels, one of which is doubtful: — L. I. 7.9 b : Saet lehtfset cuoeSende ne were under mitte to gesettanne = Lucernam dicens non esse sub modio pani (or may denote Necessity) ; L. 5.17: msegen wees drihtnes to hcelenne hia = uirtus erat domini ad sanandum eos. The same idiom is found in the West-Saxon version of the second passage. In the first example, the infinitive, though active in form, is passive in sense, and translates a Latin pass- ive infinitive; in the second example, it is active in sense as well as in form, and translates a Latin gerund that is active in sense. ORIGIN OF THE IDIOM With the exceptions already noted, the situation with respect to the Inflected Infinitive of Necessity is substantially the same in West-Saxon as in the Lindisfarne Gospels. The probability 126 The Syntax of the Lindisfarne Gospels is that, when denoting necessity or purpose, the inflected infini- tive with bia(n) (wosa) was suggested by the Latin. See The Infinitive in Anglo-Saxon, PP- 97-106 and 200-203, for this idiom in West-Saxon, and pp. 239-240 for this construction in the other Germanic languages. NOTES 1. An Inflected Infinitive without " To " occurs in Mat. I. I7.13 b : halig hundum & bergum ne is sellende (for sellenne?) . . . foregemercade uel getachte = Sanctum canibus porcisque non dandum . . . praefigit. It is possible, of course, that the glossator purposely wrote the participle here (sellende) instead of the inflected infinitive (sellenne) without to. See A above ; also " the Elliptical Accusative-with-Infinitive Construction," below, in Section xv. 2. An Inflected Infinitive of Necessity Alternates with a Past Participle in /. 19.16: Ssette uoere ahoen uel to ahoanne = ut crucifigeretur. 3. An Inflected Infinitive of Necessity Alternates with an Adjective in J. 9.30 : in Sis forSon to uundranne uel uundorlice is = in hoc enim mirabile est. Professor Cook considers uundor- lice an adverb here, but it seems better to construe it as an adjective with an improper ending: compare Carpenter, I. c, §162. VIII. THE PREDICATIVE INFINITIVE WITH ACCUSATIVE SUBJECT The active infinitive as the Quasi-Predicate of an Accusative Subject is more commonly uninflected in the Lindisfarne Gos- pels as in the West-Saxon, but it is found inflected much more frequently in the former than in the latter. In the Lindisfarne Gospels we have about 38 (65 including the " Introductions ") examples of the uninflected predicative infinitive to about 4 (41 including the " Introductions ") examples of the inflected infinitive, in objective clauses ; while the West-Saxon Gospels has 44 and respectively. In subjective clauses, the predi- The Infinitive 127 cative infinitive (uninflected in all instances but 2) is likewise more common in the Lindisfarne Gospels than in the West- Saxon, about 38 examples occurring in the former to 7 in the latter. Of the strictly passive infinitive (that is, of the infinitive made up of bia(n) (wosa) plus the past participle) I have found no clear example in the Lindisfarne Gospels. At times, however, the active infinitive, whether uninflected or inflected (especially the latter), is passive in sense. But the voice of the predicative infinitive is too complicated for collective treatment, and must be discussed piece-meal in the several subsections of this division of my study. I. As Object A. The Active Infinitive I. Uninflected The uninflected active infinitive with accusative subject, in object clauses, seems to me almost exclusively active in sense as well as in form. Possibly the infinitive is passive in sense in the two following passages, in each of which it translates a Latin passive infinitive: — Mat. 23.7 lufaS . . . groetengo in spree & hia ceiga uel hia genemna from monnum laruas = amant . . . et salutationes in foro et uocari ab hominibus rabbi ; Mat. 27.21 : hue&er ne wallad gie iwh Sara twoege forletaf = quern uultis uobis de duobus dimittif (Rushw. : hwcecfer willaS ge eow Sara twegra forleten beon; W. S. ; Hwce&erne wylle ge <5set ic forgyfe eow of Sisum twam?). But in the two following examples, in which the Latin original has a predicative passive infinitive, the Lindisfarne infinitive is active in sense as well as in form: — L. I. 9.15 a : (5a lytlo from him hia adrifa forehead = Paruulos a se uetari prohibens; L. 18.40 : geheht hine tolceda hine = iussit ilium adduci se (Rushw. : heht hine Iceda to him ; W. S. : het hine loedan to him). Some hold, however, that, after verbs of commanding and of causing, we sometimes have a predicative infinitive that, though active in form, is passive in 128 The Syntax of the Lindisfarne Gospels sense. The grounds for the active interpretation have been given in Section II, on " the Objective Infinitive," pp. 102-103 above ; where I have stated that to me the infinitive in examples of the sort there cited seems, not predicative, but objective, and the accompanying accusative, not subjective, but objective. The groups of verbs with which we find the accusative-with- infinitive construction are much the same as in West-Saxon, but naturally the distribution within the groups differs appreciably. In the Lindisfarne Gospels, the predicative infinitive with accu- sative subject is found: — I. Occasionally with these Verbs 19 of Commanding (10): forbeada (2), forbid. gebidda (1), bid, request. gehata (4), order, command. hata (3), order, command. II. Oftenest with these Verbs of Causing and of Permit- ting (24) : doa (3), do, cause. forgeafa (1) : grant, allow. forleta (5) : let, allow. gedoa (3), do, cause. gelefa (2), allow. geSreaga (1), force, compel. gewyrca (1), make, cause. lefa (1), allow. leta (4), let, allow. ondeta (1), usually confess, but here allow. fterhgelefa (1), permit. wyrca (1), make, cause. III. Occasionally with these Verbs of Sense Percepti n (4): gesea (3), see. sea (1), see. 19 In these lists the figure in parenthesis indicates the number of occur- rences of each verb with this construction. The Infinitive 129 IV. Occasionally with these Verbs of Mental Perception (12): sedeawa (l), show. geeawa (1), show. gefsestniga (1), confirm. geondwearda (l), respond. getrymma (4), affirm, confirm (belongs partly under V.). lsera (2), teach. ongeatta (1), understand, know. woena (1), hope. V. Occasionally with these Verbs of Declaring (10) : CUOeSa (9), say, declare. onsacca (1), deny. VI. Karely with these Other Verbs (5) : lufiga (1), love. willa (4), wish, desire. As this idiom in the Lindisfarne Gospels is substantially iden- tical with the corresponding one in West-Saxon, I give here only a few examples: — Mat. 19.14: nallas ge hia forbeada to me cyme = nolite eos prohibere ad me uenire. — Mat. 14.28 : haat meh gecumce to (5e = iube me uenire ad te. Mat. 5.45 : sunna his arise doed = solem suum oriri facit. — Mat. 13.30 a : forletas egSer uel boege gewcexe = sinite utraque crescere. Mh. 13.29: mi(5(5y gie geseas das wosa = cum uideritis hcec fieri. — L. 24.39 : bano ne hsefeS suae mec gie seas habba = ossa non habet sicut me uidetis habere. Mat. I. 17.15 a : monige . . . inngae getrymes = multos . . . introire testatur. — Mat. 28.20 b : Icerende hia halda alle = do- centes eos seruare omnia. J. I. 6.8: latzarum . . . slepe cuoed = Lazarum . . . dor- mire dicit. — Mat. I. 6.17: (5a tSe onsaccas — (= blank) in lic- 9 130 The Syntax of the Lindisfarne Gospels homa gecomae = negant christian in carne uenisse (christum, being a proper noun, is not glossed). L. 19.27: (5a(5e ne waldon mec rixage = qui noluerunt me regnare. The following is a complete list of the verbs followed by the active uninflected infinitive used as the Predicate of an Accu- sative Subject, in objective clauses (65, of which 27 are in the " Introductions ") : — sedeawa, show (1) : Mat. (1) : I. 18.16. CWOeSa, say, declare (9): Mat. (3): I. 17.18; I. 18.15 a ; I. 19.18.— Mh. (1) : I. 3.5 C .— L. (3) : I. 7.5 b ; 11.18; 24.23 b .— J. (2): I. 4.18; I. 6.8. doa, do, cause (3) : Mat. (2) : 5.45; 21.7.— L. (1) : 12.37 b . forbeada, forbid, prohibit (2): Mat. (1): 19.14.— L. (1): I. 9.15 a . forgeafa, grant, allow (1) : Mh. (1) : 7.12. forleta, let, permit (5): Mat. (3): 8.21 a > b ; 13.30 a .— L. (1): 10.40.—/. (1): 18.8. gebidda, bid, request (1): L. (1): 5.3. gedoa, do, cause (3): Mat. (1): 5.33.— Mk. (1): 6.39.— L. (1): 24.28. geeawa, show, demonstrate (1) : L. (1) : I. 2.17. gefsestniga, confirm (1) : L. (1) : I. 7.19 b . gehata, order, command (4) : Mat. (2) : 14.22 a > b . — L. (1) : 18.40.—/. (1): I. 6.9 a . gelefa, allow, permit (2): L. (2): I. 5.20; 8.51. geondwearda, respond (1): L. (1): I. 10.14 a . gesea, see (3): Mk. (2): 7.2; 13.29.— L. (1): 13.28. getrymma, affirm, confirm (4): Mat (1) : I. 17.15 a . — L. (1): I. 6.13c.— J. (2): I. 4.2 C ; I. 7.9 b . geSreaga, force, compel (1) : Mk. (1) : 6.45. gewyrca, make, compel (1) : Mat. (1) : I. 1.1. hata, order, command (3): Mat. (1): 14.28. — Mk. (2): I. 5.9 c > d . lsera, teach, instruct (2): Mat. (2): I. 20.5 b ; 28.20 b . lefa, allow, permit (1): Mk. (1): T. 3.6. The Infinitive 131 leta, let, allow (4): Mat. (1): 23.13.— Mk. (1): 10.14.— L. (1): 18.16.—/. (1): 11.44 b . lufiga, love (1): Mat. (1): 23.7. ondeta, usually confess, but here aiZcw (1) : Mat. (1) : I. 3 . l c ( ondetad = pateremur) . ongeatta, understand, know (1): L. (1): I. 11. 8 C . onsacca, deny (1): Mat. (1): I. 6.17. sea, see (1): L. (1): 24.39. Serhgelefa, perwwtf (1): Ma^. (1): I. 18.6 b . willa, will, desire (4) : Mat. (1) : 27.21.— L. (1) : 19.27.— J. (2): 21.22, 23. woena, ween, think (1): L. (1): 8.18. wyrca, make, cause (1) : L. (1) : 5.34. ORIGIN OF THE IDIOM The Latin correspondents of the uninfected Predicative In- finitive with Accusative Subject are as follows: an accusative with an active infinitive, 58 ; with a deponent infinitive, 3 ; with a passive infinitive, 3 (2 present; 1 perfect) ; and with a future participle, 1. The completeness of the correspondence here between the Latin original and the Northumbrian gloss would lead one to suspect more or less of Latin influence, I believe. That the native trend in the Lindisfarne Gospels is decidedly against the accusative-with-infinitive construction is indicated by this fact : although in 64 instances the Latin accu- sative-with-infinitive construction is retained in the Northum- brian gloss, in 130 instances this Latin idiom is rendered other- wise than by a predicative infinitive (uninflected or inflected) with accusative subject. Possibly an inspection of the list of verbs after which this Latin idiom is avoided may help us to attain more specific results as to the origin of the accusative-with-infinitive construc- tion in the Lindisfarne Gospels. Accordingly I give in parallel columns an alphabetical list of the verbs after which in these Gospels the accusative-with-infinitive construction of the Latin original is retained or avoided : — 132 The Syntax of the Lindisfarne Gospels Retained : Avoided cedeawa, ' show ' 1 2 ceteawa, l show ' 3 awrita, ' write,' ' express ' 2 bebeada, ' command ' 2 ceiga, i call,' ' exclaim ' 2 cuoeda, ' say,' ' declare ' 9 49 doa, ' do,' ' cause ' 3 2 doema, ' judge ' 1 forbeada, ' forbid,' ' prohibit ' 2 1 forecuoeda, ' prophesy ' 3 foregehata, i promise ' 1 forescecga, i predict ' 1 forgeafa, ' grant,' ' allow ' 1 forleta, ' let,' ' permit ' 5 gebidda, ' bid,' ' request ' 1 gecyda, ' make known ' 3 gedoa,, i do,' ' cause ' 3 1 geeawa, ' show,' ' demonstrate ' 1 gefcestniga, ' confirm ' 1 3 gehata, ' order,' ' command ' 4 6 gehata, ' promise ' 2 gelcera, ' teach ' 2 gelecniga, ' take care ' 1 gelefa, ' allow ' 2 gemerciga, ' signify ' 1 genidriga, ' humiliate ' 1 geondswariga, ' respond ' 1 geondwearda, ' respond ' 1 4 gesea, ' see ' 3 5 gesweriga, ' swear ' 1 getrymma, ' affirm,' ; confirm ' 4 1 gedreaga, ' force,' ' compel ' 1 gewyrcw, ' make,' ' compel ' 1 l hata, ' order,' ' command ' 3 2 Icera, ' teach,' ' instruct ' 2 4 The Infinitive 133 Eetained : Avoided : lefa, ' allow/ ' permit ' 1 leta, ' let/ ' allow ' 4 1 ondeta, usually ' confess,' but here ' allow ' 1 ondswariga, ' answer ' 1 ongeatta, ' understand ' 1 onscecca, ' deny ' . ., 1 1 scecga, ' say/ ' declare/ 1 sea, ' see ' 1 soeca, ' seek ' 2 sodiga, ' attest/' ' prove ' 1 spreca, ' speak ' 2 todoema, ' judge ' 1 derhgelefa, l permit ' 1 willa (ne willa), 'will/ 'wish' (* will not/ 'wish not ') 4 1 wita, ' know ' 2 woena, ' think ' 1 3 wyrca, ' make/ ' cause ' 1 1 This list will not enable us to decide as to the origin of the accusative-with-infinitive construction with each word, but it makes highly probable, I think, several general conclusions. The idiom was probably native with a few verbs (1) of Com- manding (gehata, hata), (2) of Causing and Permitting (doa, gedoa, gewyrca(f), wyrca(f), forgeafa, forleta, gelefa, lefa, leta, derhgelefa), (3) of Sense Perception (gesea, sea), (4) of Mental Perception (getrymma) , and (5) of Inclination and of Will (ivilla( ?)). The idiom was probably of foreign (Latin) origin with all verbs of Declaring, for, although this construc- tion is found 9 times after cuoeda, it is avoided 49 times there- with and several times with other verbs of declaring (ceiga, forecuoeda, forescecga, gemerciga, geondsivariga, geondweardo , ondswariga, scecga, spreca). Moreover, the idiom is avoided, also, with this verb of Commanding (bebeada) , and with sev- 134 The Syntax of the Lindisfarne Gospels eral verbs of Mental Perception (ceteawa, awrita, doema, fore- gehata (' promise '), gecyBa, gefcestniga, gehata (' promise '), gelcera, gelecniga, geni&riga, gesweriga, soeca, sodiga, todoema, luita, woena). This situation with reference to the accusative- with-infinitive construction in the Lindisfarne Gospels is re- markably similar to that in West-Saxon, as may be seen by comparing my summary concerning the origin of this idiom in the latter, in The Infinitive m Anglo-Saxon, p. 208, which is too long for quotation here. However, the accusative-with- infinitive construction is more frequent in the Lindisfarne Gospels after verbs of declaring than in West-Saxon, nine ex- amples occurring in the former to only five in the whole of West-Saxon. II. INFLECTED About 41 examples (all but four in the " Introductions ") have been found of the inflected infinitive used as the Predicate of a Subject Accusative, in objective clauses, against less than a dozen clear examples in the whole of West-Saxon. As indicated below, more frequently than not the Lindisfarne inflected infinitive here is passive in sense, a fact explained by the further fact that in such instances it usually translates a Latin passive gerundive without esse, as in Mat. I. 20. 8 b : cwed cynn (5aet mi(5 gebed & mi(5 fsestern to fordrifanne = ait genus illud oratione ieiunioque pellendum. This idiom, both in the Lindisfarne Gospels and in the Latin original, may be consid- ered an elliptical form of the predicative infinitive of neces- sity with the verb bia(n) (wosa), the ellipsis (of esse) being patent in the Latin. Compare, in Section xv below, what is said of the " the Elliptical-Accusative-with-Infinitive Construc- tion," which in one of its phases is passive in sense. I give in full the examples of the inflected infinitive as Predi- cate of an Accusative Subject, in alphabetic sequence of the governing verb (41, all but four in the " Introductions ") : — The Infinitive 135 ACTIVE IN FORM AND IN SENSE (6, of which 3 are in the " Introductions ") CUOeSa, say, declare (1) : Mk. (1) : I. 4.13: cuoed nidriendo godra woruldra hundteantigsi(5a monigfallice mi(5 oehtnissum eft to onfoenne = ait contemtores bonorum saecularium cen- tuplum cum persecutionibus recepturus (for recepturos?) . foressecga, announce, declare (1) : J. (1) : I. 6.4: he uere cuoeS duru & hiorde scipa fore Ssem he his sauel Sset he ualde gesete & o(5ero scip to gelwdanne forescegde = seque dicit ostium et pastorem ouium pro quibus se suam animam positurum tt alias oues adducturum pronuntiat (with accusative subject to be supplied in the Old English). forleta, allow, let (1) : Mat. (1) : 8.22: fylge meh & forlet (5a deado to bebyrgenne (5a deado hiora = sequere me et dimitte mortuos sepelire mortuos suos. lefa, allow, permit (2): Mh. (2): I. 3.6: diowles fordraf gegeonga uel to geonganne lefde in bergum = demones eiciens ire concessit in porcos; 1.34 b : nalde leta spreca hia uel ne lefde hia to spreccanne = non sinebat loqui ea (or Objective?). leta, let, allow (1) : Mk. (1) : 5.37 : ne leort aenigne monno to fylgenne hine = non admisit quemquam sequi se. ACTIVE IN FORM, BUT PASSIVE IN SENSE (35, of which 34 are in the " Introductions ") beada, order, command (2) : L. (2) : I. 10.19 a , b : to wceccen- ne bead & to biddanne = uigilandum praecipit et otandwm (with the accusative subject implied in both Old English and Latin. Or the infinitive may be considered purely Objective ?). bebeada, order, command (3) : L. (3) : I. 5.8 a , b ; I. 7.15. CUOeSa, say, declare (3) : Mat. (2) : I. 20.8 b : cwed cynn Sset mi(5 gebed & mi (5 f western to fordrifenne = ait genus illnd oratione ieiunioque pellendum; 16.12. — L. (1) : I. 5.17 a . efnegefestniga, conclude, assure (1) : Mat. (1) : I. 21.7. forecuoeSa, predict, assert (1): L. (1): I. 10.15: ymbsel- 136 The Syntax of the Lindisfarne Gospels lenne from here hierusalem . . . forecuoed = circumdandam ab exercitu hierusalem . . . praedicit. foregehata, declare, promise (1) : Mat. (1) : I. 16.13 b . foresaecga, declare, urge (3) : Mat. (1) : I. 21.20 b : to wcec- cenne forescegde = Uigilandum praedicit (with accusative sub- jeer implied? or Objective purely?). — L. (2) : I. 9.11, 12. gebecniga, signify, indicate (1): J. (1): I. 7.16: <5a ilea & from feder to lufanne gebecnas = quos a patre amari signifi- cans. gehata, order, urge (6): Mat. (5): I. 16.15 b ; I. 17.1(?): I. 17.5 C ; I. 21.15; I. 21.20 a .— Mk. (1) : I. 3.20. getrymma, affirm (1) : Mat. (1) : I. 17. 5 a . hata, order, command (3) : Mat. (1) : I. 20.11 (forhyeganne without to).— L. (2) : I. 7.16, 17. lsera, teach, instruct (9): Mat. I. 17.6: (5a allmissa laeras deiglige to doanne = Aelemosynam docet iu abscondito /act- e/^A/m; I. 17.8 b , c ; I. I7.13' a .— Mk. (1): I. 5.8.— L. (3): I. 8.10; I. 9.14V.—,/. (1): I. 7.12. merciga, signify, indicate (1): L. (1): I. 9.1 5 b . ssecga, say, declare (1): J. (1): I. 8.2: saeged underni- omanne lichoma of byrgenne = nuntiat sublatum corpus de monumento. ORIGIN OF THE IDIOM In a few instances (a) (as in Mk. I. 3.6 and 1.34 b , quoted on p. 135 above), the inflection of the Predicative Infinitive with Accusative Subject seems to be due to the proximity of the infinitive to a verb of double regimen (lefa) in the Northum- brian. 20 But in most instances the inflection is due to the in- fluence of the Latin original, the Northumbrian inflected in- finitive corresponding to Latin locutions that in West-Saxon, also, usually cause inflection, though not in the predicative use: the accusative-with-infinitive construction in the Latin having (b ) a future active participle (2) or (c) a gerundive (32), 20 In each of these passages both the uninflected infinitive and the inflected infinitive are given as glosses. The Infinitive 137 the latter usually with the ellipsis of esse. The other Latin correspondents are: the accusative-with-infinitive construction having a present active infinitive (2), a deponent infinitive (2), a present passive infinitive (2), and a perfect passive in- finitive (1). The influence of the Latin gerundial and future participle constructions is stronger in the Northumbrian than in the West-Saxon, in which latter only a few examples occur of the inflected infinitive used predicatively with a subject ac- cusative (scarcely a dozen clear examples). Moreover, as stated above, p. 134, the voice of the inflected predicative in- finitive in the Lindisfarne Gospels is determined largely by the nature of its Latin original. For this idiom in West-Saxon and in the other Germanic languages, see The Infinitive in Anglo-Saxon, pp. 118-119, 213, and 247-248. B. The Passive Infinitive Of the serictly passive infinitive (that is, of the infinitive made up of bia(n) (wosa) plus the perfect participle), as stated incidentally at the beginning of this chapter, I have found no clear example. But, as shown in Section A, often an inflected infinitive that is active in form is passive in sense, and very rarely an uninflected infinitive seems to be passive in sense. II. As Subject A. The Active Infinitive I. UNINFLECTED In the subjective use, the uninflected Predicative Infinitive with Accusative Subject is much more frequent in the Lind- isfarne Gospels than in West-Saxon, there being 36 examples (of which six are in the " Introductions ") in the former to seven examples in the latter (all in the West-Saxon Gospels). In five of these 36 examples, the infinitive is the subject of a passive verb. In most instances here the infinitive is active in sense as in 138 The Syntax of the Lindisfarne Gospels form. But in a few examples, indicated below, the Lindisfarne infinitive, though active in form, translates a Latin passive in- finitive, and appears itself to be passive in sense. See above, p. 134. I give in full the examples of the uninflected Predicative Infinitive with Subject Accusative, in alphabetic sequence of the verb of which the infinitive phrase is the subject (36 ex- amples, of which six are in the " Introductions ") : — arisa, be fitting (4) : L. (4) : 9.22 a > b > c > d : cuoeS Ssette arised sunu monnes feolo uel micelo gedolia uel gefiroivia & forcunima from aldum uel from Saem seldestum & aldormonnum sacerda & uSwutum & ofslaa uel Saette se ofslaegen & Se Sirddan Sseg (sic) arisa = dicens quia oportet filium hominis multa pati et reprobari a senioribus et principibus sacerdotum et scribis et occidi et tertia die resurgere. Two of these infinitives, though active in form, translate Latin passive infinitives, and appear to be passive in sense. In these two instances, the W.S. Luke has a passive subjunctive (beo a-worpen) and a passive infini- tive (beon ofslegen). The Rushworth Luke has one appositive participle (forcumen) and three finite verbs. bia(ll) (wosa), be, plus an adjective (occasionally a noun) (20) : — — eaSor (eaSur), easier (3): Mat. (2): 19.24 a > b : ead'iir is camel Serh Syril nedles oferfcrra Son Se wlonca ingeonga in rice heofna = fa-cilius est camelum per foramen acus transire quam diuitem intrare in regnum caelorum. [Professor Cook makes wlonca nominative here, although he considers ingeonga an in- finitive. I think that wlonca is a weak accusative, and that 6e, though nominative in form, is accusative in function: see < Yir- penter, I. c, §§ 500 and 675.] Mk. (1) : 10.25. — gedoefenlic (gedaef enlic) , fitting (1) : /. (1) : I. !._": gedoefenlic is Ssette gewox hine hueSre lytlige = dicitur iesum . . . oportere crescere se autem minui. — god, good (2) : Mat (2) : 18.8 a - c : god uel betra de is to life ingae unhal uel halt Son tua honda uel tuoege foet haeb- The Infinitive 139 bende uel hsebbe sonde in fyr ece = bonum tibi est ad uitam ingredi debilem uel clodum quam duas manus uel duos pedes habentem mitti in iguem aeternum (or Subjective ?) Cf. the comments on ansa above. — rehtlic, right, proper (11) : Mat. (7) : 16.21' a > b > c > c > d > e . — sel, happy, fortunate (1) : Mat. (1) : 26.35: <5eh Se seel sie me suelta Sec wi(5 = etiamsi oportuerit me mori tecum. — were, work (2) : Mat. (1) : I. 1.5, 6( ?). gedsefniga, be fitting (3): L. (1): 4.43: cuoeS forSon & oSrum ceastrum gedcefned mec bodia = ait quia et aliis ciuita- tibus oportet me euangelizare. — J. (2) : 10.16; 20.9. gerisa (gehrisa), be fitting (3): L. (3): 13.33 a (or Sub- jective only?) ; 24.26 a ' b : ahne Sas geras ge&rouia crist & on (5a wisa ingeonga in wuldre his ? — nonne haec oportuit pati chi*istum et ita intrare in gloriam suam ? [ W. S. : hu ne ge- byrede criste (5as Sing Soligean & swa on his wuldor gan, in which we have the normal West-Saxon rendering of the Latin accusative- with-infinitive construction as subject by the subject- ive infinitive with a dative dependent on the main verb.] nioma (1), usually transitive and meaning take, but here probably intransitive and meaning be fitting or possible (as is probably true also of its Latin original, capit) : soS hueSre gehriseS uel gedsefneS me todaeg & tomerne & Ssem sefterfyl- gende geonga forSon ne nimed witge losia buta hierusalem = uerumtamen oportet me hodie et eras et sequenti ambulare quia non capit prophetam perire extra hierusalem (Rushw. : ibidem; W. S. : DeahhwseSere me bebyreS todseg & tomorhgen & <5y aefteran dsege gan, forSam Se ne gebyred <5set se witega for- wurde butan hierusalem. [Although I do not find in any of the glossaries, general or special, the definition here suggested 140 The Syntax of the Lindisfarne Gospels for nioma, for several reasons I feel confident that the definition is substantially correct. (1) The translation suggested is in harmony with the West-Saxon version. (2) It is based upon a well-authenticated use of capio, as a reference to the Latin dictionaries will show. And (3) it harmonizes in sense with the variant readings for non capit (inpossibile est and non est possibile) given in Wordsworth and White's Nouum Testa- mentum Latine, p. 409.] In the following five examples, the infinitive phrase is the subject of the passive of the verb indicated: — (ge)cuoeoa, say, declare (1) : J. (1) : I. 4.1 b : is gecuoeden done hcelend fulguge = dicitur iesum baptizare. gedsefniga, be fitting (2) : J. (2) : 4.24: gaast is god & &a forSon geuorSias hine in gast & soSfsestnisse is gedcefned uordia = spiritus est deus et eos quia adorant eum in spiritu et ueritate oportet adorare; 13.14. (ge)hata, promise (2): /. (2): I. 7.8 a > b : hine sib gesealla him & sib forleta gehaten bid uel is = seque pacem dare eis pacemque relimquere pollicetur. ORIGIN OF THE IDIOM In all of the aforementioned examples of the uninflected Pre- dicative Infinitive except one (J. 13.14), the accusative-with- infinitive phrase translates the same idiom of the Latin origi- nal ; in the exceptional instance, the Latin has a complementary infinitive. Although this idiom of the Latin is avoided about 29 times in the Lindisfarne Gospels, the influence of the Latin accusative-with-infinitive phrase as subject is much stronger in Xorthumbrian than in West-Saxon, in which latter the idiom is so generally avoided that by most Old English scholars the idiom has been declared non-existent in West-Saxon. As al- ready indicated, this is an error ; but, as was incidentally stated above, p. 139, under gerisa, the normal West-Saxon (and nor- mal Germanic) rendering of the Latin subjective accusative- with-infinitive construction is the subjective infinitive (either uninflected or inflected) with a dative dependent on the main The Infinitive 141 verb or verbal phrase. Concerning this idiom in West-Saxon and in the other Germanic languages, see The Infinitive in Anglo-Saxon, pp. 124-125, 214, and 245-247. II. INFLECTED Of the inflected Predicative Infinitive with Accusative Sub- ject, in the subjective use, I have found but two examples, one phrase being the subject of a passive verb: — bia(n) (wosa), be, plus rehtlic, proper (1): L. (1): I. 7.20 : onsione heofnes cunnande rehtlice were tid to oncnauanne eawunga & sec geSiodsumnise to Sseni fionde on woeg fore- saegeS = faciem caeli probantes debere tempus intellegere mani- festum atque consentiendum aduersario in uia pronuntiat. gedsefniga, be fitting (1) : L. (1) : 21.9: gedcefned is serist das to ivosanne = oportet primum haec fieri. ORIGIN OF THE IDIOM Here the inflection of the Predicative Infinitive is clearly due to the influence of the dative-governing adjective (rehtlic) and verb (gedcefniga). B. The Passive Infinitive Of the strictly passive infinitive (that is, of the infinitive made up of bia(n) (wosa) plus the perfect participle), as the Predicate of an Accusative Subject, in subjective clauses, I have found no clear example. But, as shown in my running comments on the active infinitive in section A above, pp. 127 if., several of the uninflected infinitives that are active in form translate Latin passive infinitives, and appear themselves to be passive in sense. NOTES 1. The Elliptical Accusative-with-Infinitive Construction. In sentences like the following we have what for lack of a better name I have denominated " the Elliptical Accusative- with-Infinitive Construction," in which the infinitive is to be 142 The Syntax of the Lindisfarne Gospels supplied both in the gloss and in the Latin original: — Mat. I. 20.4: gedrouende liine forescegde = Passurum se prcenuntians ; Mk. I. 4.14 a : ec forecueB hine slaende = Item praedicens se occidendum. But this idiom, of which several varieties occur, belongs rather in the section (xv) entitled " Some Substitutes for the Infinitive," where it is treated at length. 2. An Inflected Predicative Infinitive without " To " occurs in Mat, I. 20.11, in L. I. 10.15, and in J. I. 8.2 (quoted on pp. 135 and 136). APPENDIX VI THE PREDICATIVE INFINITIVE WITH ACCUSATIVE SUBJECT IN THE OTHER GERMANIC LANGUAGES Since the publication of The Infinitive in Anglo-Saxon, in 1913, there have come to my attention several articles that deal with the Predicative Infinitive with Accusative Subject in some one of the Germanic languages, and that call for brief notice here. In the first number of Vol. xxxix of Paul and Braune's Bei- triige zur Geschichte der Deutschen Sprache und Literatur, pp. 201-209, issued in September, 1913, about two months after the publication of The Infinitive in Anglo-Saxon, Dr. M. J. van der Meer published an instructive study of " Der Gotische Ace. c. Inf. in Subjectsatzen und nach Swaei und Swe." In this article Dr. van der Meer gives detailed consideration to each /nstance of the accusative-with-infinitive construction in subjec- tive clauses, in the Greek original (about 15 examples) and in the Gothic translation (about two examples), and re-affirms the view, earlier expressed by himself, by Professor Streitberg, and by a number of the elder Gothic scholars, 22 that the construction in Gothic is due to the influence of the Greek original. I was glad to find this fresh and independent confirmation of the view expressed in The Infinitive in Anglo-Saxon (which latter had appeared too recently to be seen by Dr. van der Meer) , that, not only in Gothic, but in the Germanic languages as a whole, the 21 These are mentioned in The Infinitive in Anglo-Saxon, p. 245. The Infinitive 143 aecu^ative-with-infinitive construction is a borrowed idiom in subjective clauses. Dr. van der Meer was in error, however, in declaring, as he did in the opening sentence of his article, that this construction is not found in any of the older Germanic languages except Gothic : " Weil diese constructionen in den anderen altgermanischen sprachen nicht vorkommen, bis jetzt wenigstens noch nicht nachgewiesen sind," etc. At least one example occurs in Tatian (187.9: gilimphit mih gangan = oportet me ambulare), according to Dr. A. Denecke, in his Der Gebrauch des Infinities bei den Althochdeutschen Uebersetzern des 8. und 9. Jahrhunderts, a Leipzig dissertation of 1880. And about fourteen examples occur in West-Sax^n. In both languages the construction is directly due to the influence of the Latin original: see The Infinitive in Anglo-Saxon, pp. 124- 125, 214, 245-246. Of the accusative-with-infinitive construction in Dutch, I gave no account in my section on this idiom in the Germanic languages in The Infinitive in Anglo-Saxon, pp. 241-248. But I wish now to call attention to the treatment of this subject by Dr. W. L. van Helten, in his " Over den Afhankelijken Infini- tief zonder Te," in the Tijdschrift voor Nederlandsche Taal- en Letterhunde, x, 1891, pp. 237-240 ; and by Dr. F. A. Stoett, in his Middelnederlandsche Spraakkunst: Syntaxis, edition of 1889, p. 34. Dr. Stoett declares that this construction in Dutch is due to Latin influences: " In 't mnl. vindt men na sommige werkwoorden meermalen eene constructie met den accusativus cum infinitivo, hoogst waarschijnlijk als eene navolging uit het Latijn. Die werkwoorden zijn sien, mercken, vernemen, fin- den, weten, horen, voelen, laten, seggen, doen, heeten." IX. THE PREDICATIVE INFINITIVE WITH DATIVE SUBJECT Several scholars of great distinction, as Jacob Grimm and Professor Wilhelm Streitberg, hold that, in the Gothic Mark 2.23 (jah warp pairhgaggan imma pairh atisk = ical iyeveTo TrapaTTopeveadai avrov . . . Sid rdv irvevfiarc ra> atcaQaprw i^e\0elv airb tov avdpdoirov) with 1 Timothy 6.13 (anabiudo . . . fastan puh po anabusn = TrapayyeXXco . . . Trjprjcrai gaswiltan pamma unledin." As al- M This example is taken from Professor E. C. Armstrong's Syntax of the French Verb, second edition, New York, 1915, p. 9. 31 This seems to me an infelicitous example, as I consider drigkan to denote purpose here: see The Infinitive in Anglo-Saxon, p. 148, Note 6. The Infinitive 155 ready intimated in preceding paragraphs, I hold that no con- vincing proof has ever been offered that a shifting of depend- ence from the finite verb to the infinitive took place in Gothic ; to me the last Gothic example of the dative-with-infinitive con- struction given by Professor Nauniann stands on substantially the same footing as the second Gothic example and the Old High German example thereof given by him, in each of which latter he admits the dependence of the dative upon the finite verb. Again I am pleased to see the phrase, " the Dative with Infinitive," extended to personal as well as. to impersonal verbs. So far as I can see, therefore, the studies that have appeared of the dative-with-infinitive construction since the publication of The Infinitive in Anglo-Saxon in 1913, tend to confirm the position then taken, and now reaffirmed with reference to the Northumbrian as represented in the Lindisfarne Gospels, that, in the dative-with-infinitive construction, whether after imper- sonal or personal verbs, the infinitive is substantival, not pre- dicative, in function, and that the dative is more intimately connected with the finite verb of its clause than with the in- finitive. X. THE FINAL INFINITIVE A. The Active Infinitive 1. With Active Finite Verb The active infinitive denoting Purpose, with active verbs, occurs about 124 times (136 including the "Introductions") in the Lindisfarne Gospels. Of these infinitives, 54 are unin- flected, and 82 are inflected. In the West-Saxon Gospels, there are 79 final infinitives, of which 42 are uninflected, and 37 are inflected. The final infinitive seems to me dominantly, if not exclu- sively, active in sense as well as in form. Possibly the inflected infinitive in Mat. 20.28 a , which translates a Latin passive in- finitive, is passive in sense, though active in form : sua sunu monnes ne cuom him to heranne ah he to embehtane oSrum & 156 The Syntax of the Lindisfarne Gospels sella sawel his eftlesing uel alesenis fore monigum = Sicut filius hominis non uenit ministrari sed ministrare et dare ani- mam suam redemtionem pro multis (Rushw. : ne cwom Sset him wcere dcegnad ah he Ssegnade ; W. S. : ne com (5set him man denode ac Saet he Senode). Some would hold that the inflected infinitives in Mat. 20.19 a ' b > c , which translate a Latin ad plus a gerund, are passive in sense : sellas hine ha?Snum to telenne uel to besuicanne & to suinganne & to hoanne = tradent eum genti- bus ad deludendum et flagellandum et crucifigendum. But to me the infinitive in such cases seems active in sense as in form, for reasons that are stated in detail in The Infinitive in Anglo- Saxon, p. 132, and that need not be repeated here. I. Only the uninflected Final Infinitive is found with these verbs : — 32 bringa (1), bring (W. S. : I. (6) only). foregeonga (1), precede (W. S. : lacking). gecuma (2), come (W. S. : I. (1) only). giwiga (1), request (W. S. : lacking). iorna (1), run (W. S.: 1.(1) only). ofstiga (1), descend (W. S. : lacking). winna (1), labor, strive (W. S. : I. (2) only). Typical examples are: — Mat. 28. 8 D : iornende beada uel scecga Segnum his = cur- rentes nuntiare discipulis eius. /. 4.33 : senigmon brohte him eatta? = aliquis attidit ei man- ducare ? J. 4.9 : huu . . . drinca from mec giues (5u ? = quomodo . . . bibere a me poscis? J. I. 3.15 a ' b : (5sette ne doema ah haele gecuome midgeard & Sset woere aedeauad cuse(5 uoerc = quod non iudicare sed saluare uenerit mundum et manifestari dicit opera. 32 The figure in parenthesis in these lists indicates the number of occur- rences of each verb in this construction. For the sake of comparison, I give also the construction of each verb in West-Saxon. The Infinitive 157 II. The Final Infinitive is found both uninflected and in- flected with these verbs: — 33 CUma (7 and 28), come (W. S. : 35 and 80). fara (2 and 4), go (W. S. : 5 and 15). foera (10 and 4), go (W. S. : 7 and 19). gegeonga (1 and 1), go (W. S. : lacking). geonga (2 and 7), go (W. S.: 40 and 20). gesea (1 and 2), see (W. S. : lacking). gesella (2 and 1), give (W. S. : 1 and 5). sella (13 and 7), give (W. S. : 257 and 76). senda (8 and 3), send (W. S. : 8 and 47). The following examples are typical: — Mat. 9.13: ne forSon cuom ic geceige so(5feaste ah synfullo = non enim ueni uocare iustos sed peccatores. — Mat. 2.2 : we cuomon to wordianne hine = uenimus adorare enm. L. 7.24 b : forhuon foerdon gie on woestern gesea? = quid ex- istis in desertum uidere? — Mk. 14.48: gie foerdon mi(5 suor- dum & stengum to gefoanne uel to Iceccanne mec = existis . . . comprehendere me. Mat. 27.48 b : salde him drinca = dabat ei bibere. — Mat. 14.16 b : seallas him ge uel iuh eatta = date illis uos manducare. — Mat. 20.19 a > b > c , with three inflected infinitives after sella translating Latin ad plus three gerundives in the accusative, is quoted on p. 156 above. MJc. 3.14: Saette sende hia bodiga godspell = ut mitteret eos praedicare euangelium. — Mk. I. 218: Segnas to bodanne miS word mi(5 gebrohtmn msehtum sended = discipulos ad praedi- candum uerbo conlatis uirtutibus mittet. III. The inflected Final Infinitive only is found with the following verbs : — arisa (1), arise (W. S.: IT., 2; I., 6). astiga (1), ascend (W. S. : the same). behalda (1), behold (W. S.: the same). 33 In this list the first figure indicates the uninflected infinitive; the sec- ond, the inflected infinitive. 158 The Syntax of the Lindisfame Gospels binda (1), bind (W. S.: the same). doa (1), do, make (\V. S. : the same). eftgecerra (1), return (W. S. : lacking). falla (1), fall (W. S.: lacking). forecuma (1), precede (W. S. : lacking). gearwiga (1), prepare (W. S. : the same). gecerra (1), return (W. S.: XL, 1; I., 1). gemoeta (1), find (W. S. : lacking). habba (2), have (W. S. : the same). infoera (1), enter (W. S. : lacking). lseda (1), lead (W. S. : the same). setta (1), set (W. S. : the same). SOeca (1), seek (W. S. : the same). stiga (1), ascend (W. S. : lacking). stonda (2), stand (W. S.: IT., 1; I., 3). tocuma (1), come to (W. S. : the same). togeeca (2), add (W. S. : lacking). ymbsceawiga (1). look about (W. S. : lacking). .V few examples will suffice : — Mat. 14.23 : astag in mor he ana to biddanne = ascendit in montem solus orare. Mat. 13.30 b : bindas (5a bunda uel byr(5eno uel sceafa to ber- nenne = alligate ea [in] fasciculas ad coniburendum. Mat. 4.9 : gif (5u f alias uel slsehtes to wordenne uel to wordi- anne mec = si cadens adoraueris me. Mat. 2.13 woen is . . . (5set hero(5es soecas Sone cnseht to fordoanne uel to forlosanne hine = futurum est enim ut herodes quaerat puerum ad perdendum eum. L. 13.15 : Icedes to wcetranne = ducit ad aquare. The statistics of the Final active infinitive with active verbs in full are as follows 34 (U. : 54, of which 8 are in the " Intro- ductions " ; I. : 82, of which 5 are in the " Introductions ") : — arisa, arise: U. (0).— I. (1) : L. (1) : 4.16. ** In this section, besides the finite verb (the initial word) I give the infinitive when uninflected, but not when inflected. The Infinitive 159 astiga, ascend: U. (0).— I. (1): Mat (1): 14.23. behalda, behold, look: U. (0).— I. (1) : L. (1): 1.26. binda, bind: U. (0).— I. (1): Mat. (1): 13.30 b . bringa, bring: U. (1):— eatta, eat (1): -/. (1): 4.33. — I. (0). cuma, come: U. ( 7 ) :— geceiga, call (1): Ma£. (1): 9.13. — — gedoa, make (1): L. (1): I. 9.20. gefriga (ge- friega), liberate (1) : i¥^. (1) : 27.49 b . gehalgiga, save, sanctify (1) : Mat. (1) : 18.11. losiga, destroy (1) : Mk. (1): 1.24 b . sella, give (1): i¥^. (1): 20.28 c . senda, send (1) : Mat. (1) : 10.34 c .— I. (28) : Mat. (12) : 2.2, 8; 5.16 a > b > c ; 8.29 ; 10.34 b , 35 ; 12.42 ; 20.28 a > b ; 28.1.— Mk. (3): ].24 c ; 2.17; 15.36.— L. (8): 1.59; 4.34 b ; 5.32; 11.31; 12.49, 51; 19.10 a > b .— /. (5): 1.31 b ; I. 6.18, 19; 4.7 ,a , 15. doa, do, prepare: U. (0).— I. (1) : Mat. (1) : 26.12. eftgecerra, return: U. (0).— I. (1): Mat. (1) : 24.18. falla, fall: U. (0).— I. (1): Mat. (1): 4.9. fara, go: U. (2) : — gearwiga, prepare (1) : L. (1) : 1.76. — — inlihta, illumine (1): L. (1) : 1.79*— I (4): Mk. (1) : 5.14.— L. (3): 1.77, 79 b ; 14.31. foera, go: U. (9): — efnegelseda, lead, conduct (1): Mat. (1): 20.1. gesea, see (7): Mat. (3): 11.7 b , 8, 9.—L. (4) : 7.24 b , 25 a , 26; 8.35 ,a . gewaecca, awake, arouse (1) : Mk. (1): I. 3.7 b — I. (4): Mk. (1): 14.48.— L. (3): 6.12; 19.12' a > b . forecuma, come in advance: U. (0). — I. (1): Mk. (1) : 14.18. foregeonga, precede, go in advance: U. (1) : — gegearwiga, prepare (1): L. (1): 1.17.— I. (0). gearwiga, prepare: U. (0).— I. (1) : Mat. (1) : 26.17 b . gecerra, return: U. (0).— I. (1): Mb. (1): 13.16. gecuma, come: U. (2): — doema, judge (1): J. (1): I. 3.15*. ^hsela, save (1) : /. (1) : I. 3.15 b .— I. (0). gegeonga, gro.-U. (1) : — gebidda, pra?/ (1) : Mk. (1) : 6.46. —I. (1): ¥^. (1): 25.10. gemoeta, find, invent: U. (0).— I. (1) : L. (1) : 6.7. geonga, go: U. (3):— fisciga, fish (1): •/. (1): 21.3.— 160 The Syntax of the Lindisfarne Gospels — gegearwiga, prepare (1): J. (1): 14.2. sede, sow (1): Mat. (1): 13.3 b .— I. (7): Mat (3): 5.24; 18.12; 26.55. — Mk. (2) : 3.21; 4.3.— L. (2) : 8.5; 14.19. gesea, see: U. (1): — geworpa, pluck out (1): Mat. (1): 7.5.— I. (2) : Mat. (2) : I. 1.7; 5.28. gesella, give: U. (2) : — drinca (dringa), drink (1) : Mat. (1): 25.35 b . eatta, eat (1): Mat. (1): 25.35 a .— I. (1): ./. (1): 6.52b. gesmiriga, anoint: U. (0). — I. (1): L. (1): 4.18 a . glwiga, request, demand: U. (1) : — drinca, drink (1): J. (1): 4.9.-1.(0). habba, have: U. (0).— I. (2) : L. (1) : 14.28°.—/. (1) : 4.32. innfoera, enter: U. (0).— I. (1) : L. (1) : 14.1. iorna, run: U. (1): — beada, announce (1): .Mai. (1) : 28.8 b .— I. (0). laeda, lead: U. (0).— I. (1): L. (1): 13.15. ofstiga, descend: U. (1) : — genioma, carry (1) : Mat. (1) : 24.17.— I. (0). sella, give: U. (13): — drinca, drink (5): i¥a£. (2): 25.42 b ; 27.48 b .— Mk. (1): 15.23.—./. (2): 4.7 b , 10.- — eatta, eat (7) : Mat. (2) : 14.16 b ; 25.42 a .— Mk. (2) : 5.43 b ; 6.37 a .— L. (2): 8.55 b ; 9.13.—/. (1): 6.31. geeatta, eat (1): L. (1): I. 6.1.— I. (7): Mat. (4): 20.19 a > b > c ; 27.34*.— Mk. (3): 6.37 b ; 13.22; 15.15. senda, send: U. (8) : — bodiga, preach (3) : Mk. (1) : 3.14. — L. (2): 4.19 a ; 9.2 a — - forleta, forgive (1): /. 4.1S C .— -fulwiga (fulguia), baptize (1) : /. (1) : 1.33 ,a . ge- hrioppa, reap (1) : /. (1) : 4.38. haela, heal (1) : L. (1) : 9.2 b — -togeceiga, call (1): i^a^. (1): 22.3.— I. (3) Ml. (1): I. 2.18.— L. (2): 4.18 b ; 14.17. setta, set, instruct (translating instituem) : U. (0). — I. (1) : Mk. (1): I. 1.13. SOeca, seek: U. (0).— I. (1) : Mat. (1) : 2.13. Stiga, go: U. (0).— I. (1): L. (1): 17.31. Stonda, stand: U. (0).— I. (2) : Mat. (1) : 6.5 b .— Mk. (1) : 11.25. The Infinitive 161 tocuma, come to: U. (0).— I. (1) : L. (1) : 23.48( ?). togeeca, add: U. (0).— I. (2) : L. (2) : 20.11, 12. winna, strive: U. (1) : — cweoba, say (1): Mk. (1): I. 1.11.— I. (0). ymbsceawiga, looh about: U. (0).— I. (1) : Mk. (1) : 5.32. 2. With Passive Verbs Of the Final active infinitive modifying passive verbs, I have found only one example, in Mk. 7.4 : oSero menigo sint (5a Se gesald aron Ssem uel him to haldanne = alia multa simt quae tradita sunt illis seruare. DIFFERENTIATION OF THE TWO INFINITIVES The Differentiation between the Uninflected Infinitive and the Inflected Infinitive, in the Final Use, in the Lindisfarne Gospels, rests upon the same general principles as in West-Saxon. Only the uninflected infinitive is found after a few verbs (1) of motion and (2) of giving, listed above under I, which list differs somewhat, but not fundamentally, from that of West-Saxon. But the uninflected infinitive and the inflected infinitive are each found with a larger number of verbs (1) of motion and (2) of giving (occasionally, also, with gesea, 'see'), listed above under II, which list largely coincides with the correspond- ing list in West-Saxon. Finally, as in West-Saxon, a large number of verbs of varied signification is found with the in- flected infinitive only. These verbs are given under III above. For the corresponding lists in West-Saxon, see The Infinitive in Anglo-Saxon, pp. 133-145. When we consider the lateness of the Lindisfarne Gospels and the fact that it is a gloss rather than a translation, the divergences from the West-Saxon appear negligible. The differentiation rests somewhat, also, on the Latin original, as is evident from what is said below. ORIGIN OF THE IDIOM The evidence of the Lindisfarne Gospels as to the Origin of the Final Infinitive, as far as it goes, is confirmatory of the 11 162 The Syntax of the Lindisfarne Gospels conclusions reached as to the origin of the idiom in West-Saxon, namelv : that the uninflected infinitive of purpose after verbs of motion was a native idiom in West-Saxon, the relative fre- quency of the idiom in the Lindisfarne Gospels, as in the West- Saxon Gospels, being due in part to the frequency of the final infinitive in the Latin original ; that the uninflected infinitive of purpose after verbs (1) of commanding and requesting and (2) of giving was probably due to Latin influence; and that the inflected infinitive of purpose, after verbs of whatever kind, was probably first suggested by the Latin (the constructions with gerund and gerundive, whether with or without the prepo- sition ad). The Latin correspondents of the final infinitive in the Lindisfarne Gospels are as follows: — (1) for the uninflected infinitive: a final infinitive, 51 ; a final participle, 1; a noun (potum), 1 ; — for the inflected infinitive: ad plus a gerundive, 9 ; ad plus a gerund, 10 ; ad plus a noun, 1 ; a final infinitive, 56 ; ad plus an infinitive, 1 ; an appositive participle, past, 1 ; an objective infinitive, 1 ; a final participle, present, active, 1 ; a subjunctive, 1. Concerning this construction in West-Saxon and in the other Germanic languages, see The Infinitive in Anglo-Saxon, pp. 132-148, 215-217, and 252-256 ; also van Helten, 4 pp. 228-230. B. The Passive Infinitive No clear example has been found of the strictly passive in- finitive used to denote purpose. NOTES 1. The Final Infinitive in a Series. — In the following pas- sages we have a series of two or more infinitives the first of which is inflected, but the succeeding is not: Mat. 20.28 b ' c , quoted on p. 155 above; L. 4.18 b ' c , 19 a : sende meh to bodianne ermingum uel gehaeftendum forgefnise & blindum gesihSo, for- leta <5a gebroceno on forgefnise, bodia ger drihtnes ondfenge & dcege (sic) eftselenise — misit me praedicare captiuis remis- sionem et csecis uisum, dimittere confractos in remissionem, The Infinitive 163 praedicare annum domini acceptum et diem retributionis. — On the other hand, as in the West-Saxon Luke, twice we have an uninflected infinitive succeeded by an inflected in- finitive, owing no doubt to the influence of the Latin original, in which an infinitive is succeeded by ad plus a gerundive: — L. 1.76, 77, 79 b ' c , quoted on p. 156 above; Mat. 20.28 a > b , quoted on pp. 155-156 above (here the two inflected infinitives are followed by an uninflected infinitive) ; L. 19.10' a > b : cuom forSon sunu monnes to soecanne & to hcelenne = Uenit enim filius hominis quaerere et saluare quod perierat ; L. 19.12' a ' b : foerde on lond unneh uel suiSe fearr to onfoanne him ric & eft he to cerranne = abiit in regionem longinquam accipere sibi regnum et reuerti; J. I. 6.18, 19: hine ne to doemenne ah to haelanne middangeard gecuome gecySed bi<5 = se non ad iudicandum sed ad saluandum mundum uenisse testa- tum — It seems unnecessary to give here the series in which each infinitive is uninflected. 2. An Alternation of the Two Infinitives occurs with a sin- gle word in the following: — Mat. 10.34 c : ne cuom ic fri(5 sende uel to sendenne = non ueni pacem mittere; Mk. 1.24 b : cuome Su losige uel to losane usig ? = uenisti perdere nos? L. 7.26: ah ymb huaed foerdon gie gesea uel to sceawnnef = sed quid existis uidere? L. 9.2 a : sende hia bodia uel to bodianne ric godes = misit illos praedicare regnum dei. 3. A Clause Alternates with an Infinitive in L. 1.17, in con- formity with the Latin original: he foregced fore (5sem in gast & of uel miS maeht helises Ssette gecerre hearta. ... In sunum 164 The Syntax of the Lindisfarne Gospels & Sa ungeleaffulo to hogascipe soSfaestra gegearuiga drihtne folc wisfaest = ipse praecedet ante ilium in spiritu et uirtute heliae ut conuertat corda patrum in filios et incredibiles ad pru- dentiam iustorum parare domino plebem perfectum. — An in- terchange of clause with infinitive for a single Latin word occurs in Mat. 12.42: forSon cuom from gemserum eorSo dcet hiu ge- herde net to heranne = quia uenit a finibus terrae audire. 4. "Til" with a Final Infinitive occurs in Mat. 26.17 b : huer wiltu Sset we gearuiga Se til eottanne eastro? = ubi uis paremus tibi comedere pascha ? 5. A Final Infinitive Alternates with a Present Participle in J. 1.31 b : cuom ic in usetre fulguande uel to fulguanne = ueni ego in aqua baptizans. 6. The Function of the Infinitive is ?wt Clear in the passage below, but to sceawanne probably denotes purpose: — L. 23.48: all Saet here Sara (5a<5e sedgesedre tocuomon uel toweron to scea- wanne (5aet & gesegon SaSe woeron aworden slsegendo woeron hiora breosto eftcerrdon = omnis turba eorum qui simul ade- rant ad spectaculum istud et uidebant qua? fiebant percutientes pectora sua reuertebantur (Rushw. : al Se here hiora SaSe estsce- owunga togedre comun to sceawunga Sset & gisegun etc., W. S. : eall wered Se cet disse wcefersynne wseron & gesawon etc. ) . See Note 3 to Chapter II, p. 64. 7. ' The Split Infinitive." — In his " Origin and Force of the Split Infinitive," Modern Language Notes, xxix, 1914, pp. 41-45, Professor George O. Curme declares that the oldest examples of the Split Infinitive known to him are found in Sir Gawayne and the Grene Knight (about 1320-1330 A. D.). Evi- dently he had not at that time seen my Note on this idiom in The Infinitive in Anglo-Saxon (1913),, p. 148 (at the end of the chapter on " The Final Infinitive "), in which I had cited an example of the split infinitive (with for to) from the Anglo- Saxon Chronicle, p. 256 of the Plummer edition, under the year 1127 of Manuscript E. The Infinitive 165 8. A Final Infinitive ivithout a Principal. — In the following passages we have a final infinitive without a specific word (verb) for it to modify: — iminflected: L. 1.72 b ; — inflected: L. 1.72 a ; 21.38. In the first two passages, the verb is implied in a noun, in Northumbrian and in Latin; in the third example, the Latin verb is not translated by the glossator. 9. The Text is Probably Corrupt in Mark I. 1.11 : ne wann accennisse lichomses Son in serrum awoendat were cweo&a = non laborauit natiuitatem carnis quam in prioribus uicerat dicere, in which we have an iminflected infinitive of purpose (cweoda) modifying the finite verb (ivann), corresponding re- spectively to the Latin dicere and laborauit. Although it occurs in several manuscripts, uicerat is probably a corruption for uiderat, which latter is the reading given in Wordsworth and White's Nouum Testamentum Latine, p. 172. My friend and colleague. Professor Edwin W. Fay, head of our School of Latin, has kindly called my attention to Professor W. M. Lind- say's Introduction to Latin Textual Emendation, p. 83, where we learn that, in Latin minuscule manuscripts, cl and d are often almost indistinguishable. The misreading of uiderat as uicerat probably accounts for the Northumbrian awoendat were. I cannot construe either uicerat or the other variant (uincerat) given for uiderat by Messrs. Wordsworth and White ; and con- sequently the Northumbrian rendering of uicerat, namely, awoendat were, is unintelligible to me here. XI. THE INFINITIVE WITH ADJECTIVES A. The Active Infinitive In the Lindisfarne Gospels, the active infinitive that modi- fies an Adjective is inflected 35 8 times, and is uninflected 35 V times. In West-Saxon, on the other hand, of 247 examples of the infinitive with adjectives, only 6 infinitives are unin- flected; and in the West-Saxon Gospels all examples (4) are inflected. 35 No examples occur in the " Introductions." 166 The Syntax of the Lindisfarne Gospels The infinitive, whether uninfected or inflected, is active in sense as in form. I. THE INFINITIVE UNINFLECTED The seven examples of the uninflected infinitive modifying an Adjective are as follows: — gearo, ready, prepared (1) : L. 22.33: Sec niiS gearo ic am in carcerne & aec in deaSe geonga = tecum paratus sum et in carcerem et in mortem ire. gidyrstig, audacious, daring ( 1 ) : Mat. 22 .46 : ne Se Son gidyrstig wses aenig . . . hine forSor gefregna = neque ausus fuit . . . eum amplius interrogare (or Consecutive?). msehtig, mighty, powerful (1) : Mh. 9.29 : cuoeS to him Sis cynn on nsenig mcehtig of geonga = dixit illis hoc genus in nullo potens ire. min, mine (1) : Mat. 20.23 b : sitta uutedlice to suiSra minra & winstra ne is min sella = sedere autem ad dexteram meam et sinistram non est meum dare. wyrSa, worthy (3) : Mat. 3.11 : his uel Sses nam ic wyrfie gesceoe beara = cuius non sum dignus calciamenta portare. Mh. l.Y c : Sees nam ic ivyrde forehlutende undoa Suongas scoe his = cuius non sum dignus procumbens soluere corrigiam calciamentorum eius. L. 21.36 c : Ssette gie se ivyr&o to habbanne gefleanne (5as alle (5a Se tocymendo sint & stonda fora sune monnes = ut digni habeamini fugere ista omnia quae futura sunt et stare ante filium hominis. II. THE INFINITIVE INFLECTED The eight examples of the inflected infinitive modifying an Adjective I likewise give in full: — eaSor, easier (1) : L. 10.12: ic cuoeSo inh Ssette sodomon on dsege Ssem forge- The Infinitive 167 fenra uel eador to forgeafanne bi(5 Son Seer ceastra = dico nobis quia sodamis in die ilia remissive erit qnam illi ciuitati. lset (hlaet), slow (1) : L. 24.25 : la unwiso & lilcetto ofer hearta to gelefanne in allum — o stulti et tardi corde ad credendum. mill, mine (1) : J/A\ 10.40 b : sitta uutedlice to swiSra minra uel to winstrai ne is min to sellanne = sedere autem . . . non est meum dare. [Cf. Mat. 20.23 b , above, where min is modified by an unin- flected infinitive.] nedSarf, necessary (1) : L. 23.17 : neddarf Sonne hsefeS uel hsefde to forgeafanne him . . . enne = necesse autem habebat dimittere eis . . . unum. [Or is neddarf a noun here, as is claimed by Professor Cook, in his Glossary ?] SCyldig, guilty (1): Mat. 5.22 : se(5e uutedlice cueSas broSre his <5u unuis uel idle, scyldig biS to hoetanne (sic) = qui autem dixerit fratri suo racha, reus erit concilio. [R'ushworth: he biS gemote scyl- dig; W. S. he by (5 gedeahte scyldig.] wyrSe, worthy (3) : L. 3.16: Sees ne am ic tvyrde to unbindanne Suongas sceoea his = cuius non sum dignus soluere corrigiam calciamentorum eius. L. 21.36 a ' b : quoted above, p. 166, under Uninflected. Notice there the erroneous translation of the Latin habeamini by to hahbanne. DIFFERENTIATION OF THE TWO INFINITIVES As to the Differentiation of the Two Infinitives, the relatively large number of uninflected infinitives modifying an Adjective is somewhat surprising; for in the Lindisfarne Gospels alone more examples (7) are found than in the whole of West-Saxon literature (6). But the lack of inflection is due, I believe, partly to the remoteness of most of these infinitives from the adjective modified ; partly to the frequency of the infinitive 168 The Syntax of the Lindisfarne Gospels instead of gerund or gerundive in the Latin original; and partly, in this as in other abnormal uses, to the idiosyncrasy of the glossator. As is implied in the preceding sentence, some- times (as in Mk. 9.29 and Mat. 20.23 b ) the infinitive is not inflected although in juxtaposition with the adjective. The Latin correspondents are as follows : — for the uninfected infini- tive: an adjective plus an infinitive, 6; an objective infinitive to ausus fuit, 1 ; — for the inflected infinitive : an adjective plus an infinitive, 4; an adjective with ad plus a gerund; 1; an adjective in the comparative degree, 1 ; an adjective plus a noun in the ablative, 1 ; and a subjunctive passive that is in prox- imity to an adjective and an infinitive, 1. ORIGIN OF THE IDIOM As I have tried to show in the discussion of this construction in West-Saxon, the use of the Infinitive with Adjectives, whether uninflected or inflected, appears to be a native English idiom. For the Infinitive with Adjectives in West-Saxon, see The Infinitive in Anglo-Saxon, pp. 149-159, 217-218; and in the other Germanic languages, pp. 256-257. B. The Passive Infinitive I have found no example in the Lindisfarne Gospels of the compound passive infinitive modifying an adjective. Only one example was found in all West-Saxon literature. NOTES 1. The Infinitive in a Series with Adjectives. — In L. 21.36 a » b ' c , quoted on p. 166 above, we have a series of three infinitives the first two of which are inflected, but the third is not. 2. The Inflected Infinitive without " To " occurs in L. 21.36 b , quoted on p. 166 above. The Infinitive 169 3. An Inflected Infinitive Erroneously Translates a Latin Subjunctive Passive that is in proximity to an adjective and an infinitive in L. 21.36 a , quoted on p. 166 above. XII. OTHER ADVERBIAL USES OF THE INFINITIVE Of Other Adverbial Uses of the infinitive than those already treated, namely, the infinitive of purpose and the infinitive of specification (with adjectives), but few examples occur in the Lindisfarne Gospels. Possibly we have an Infinitive of Cause in L. 16.3 : to giorn- anne uel to fceranne on selmessum ic sceomigo min = mendicare erubesco (Rushw. : lacking ; W. S. : me sceamad dcet ic wcedlige). Possibly, too, we have Infinitives of Result in the two follow- ing passages from Luhe: — 14.23: gened to ingeonganne Ssette sie gefylled hus min = compelle intrare ut impleatur domus mea (Rushw. : gined in to ingeonganne ; W. S. : nyd hig (5set hig gan in) ; 13. 24®". gedrincgas to ingeonganne Serh nearo gsett = contendite intrare per angustam portam (or Final?) (Rushw.: gedringas to onginnane; W. S. : efstad (5set ge gangen (5urh Saet nearwe get). Although I have put the infinitives in the fol- lowing passages under the objective use with passive verbs (= the retained objective), it is possible that the infinitives are con- secutive: — Mat. I. 7.3 a : were geneded (5set awritta = cogeretur ut scriberet; Mat. I. 6.19, 7.2 : gedreatad uel genedad is . . . godcunde haelendes hero uel heista awrita & . . . foreiorne uel bicymo = coactus est . . . [de] diuinitate saluatoris altius scribere et . . . prorumpere. Possibly, again, we have an infinitive denoting Means in Mat. I. 17.19: cwed . . . to geafanne miS efennisse getimbres hus ofer carr uel ofer sond = dicit . . . adhibita comparatione aedificantis domum super petram aut super herenam (sic). OKIGIN OF THE IDIOM The examples of the Causal, the Consecutive, and the Instru- mental use of the infinitive are too few to determine the origin 170 The Syntax of the Lindisfarne Gospels of the idioms here. In all probability, however, for reasons stated in my discussion of these idioms in West-Saxon, the causal used is more largely due to Latin influence, but the consecutive use chiefly to native influence. Of other adverbial uses than those specified, I have found no clear example in the Lindisfarne Gospels. In all the exam- ples given above, the infinitive is active in sense as in form. For the situation in West-Saxon, see The Infinitive in Anglo- Saxon, pp. 160-172, 218-220; and in the kindred Germanic languages, pp. 258-262. APPENDIX VIII THE CONSECUTIVE INFINITIVE IN GOTHIC In Dr. M. J. van der Meer's article on " Der Gotische Ace. c. Inf. in Subjectsatzen und nach Swaei und Swe," which ap- peared in the fall of 1913, and which was considered, in part, in Section vi, above, on " The Predicative Infinitive with Accusative Subject," he considers all the examples of the accu- sative-with-infinitive construction preceded by wcrre in Greek (21 examples in all) and preceded by swaei and swe in Gothic (5 examples for the two particles), and reaffirms the opinion that had been generally held theretofore, that these five examples of the consecutive infinitive in the Gothic are due to the presence of this construction in the Greek original. Of the correctness of this view, there seems to me no question. But Dr. van der Meer's statement, in his opening sentence, already quoted by me on page 143 above, that this construction is not found in any of the older Germanic languages except Gothic, needs modifica- tion. A somewhat similar use of the infinitive (not of the infinitive with an accusative subject) to denote result is found in Old Norse after saa and in West-Saxon after swa. Con- cerning the history of opinion as to the latter idiom and as to the former one, see The Infinitive in Anglo-Saxon, pp. 162-168, 218-219, and 259-261. The Infinitive 171 XIII. THE INFINITIVE WITH NOUNS A. The Active Infinitive In the Lindisfarne Gospels, as in West-Saxon, when a JSToun is modified by an infinitive, normally the infinitive is inflected. There are 17 36 examples of the inflected infinitive to 7 examples of the miinflected in the Lindisfarne Gospels. The proportion of inflected to uninflected infinitives in the West-Saxon Gospels is 26 to 1. The infinitive that modifies a noun seems to me active in sense as in form. I. THE INFINITIVE UNINFLECTED Of the uninflected infinitive modifying a Noun^ I give all the examples observed by me (7, of which one example occurs in the " Introductions ") : — mseht, might, power (6) : L. 5.24: mceht hafeS on eorSo forgeafa synna = potestatem habet in terra dimittere peccata. L. 12.5 : hsefeS mceht gesenda uel to gesendanne in tintergo = habet potestatem mittere in gehennam. J. 1.12: gesalde Ssem mceht suno godes wosa uel Sset hia uero = dedit illis potestatem filios dei fieri. J. 5.27: mceht salde him & dom gewyrce = potestatem dedit ei et iudicium facere. J. 19.10 a ' d : nastu Ssette mo?ht ic hafo gehoa (5e uel Sec to hoanne & mceht to forletanne Sec <5ec (sic) uel forleta Sec? = nescis quia potestatem habeo crucifigere te et potestatem demit- tere te? stefn, voice ( 1 ) : L. I. d.lQ^: miSSy acenned waes iohannes miS stefne gespreca Ssem dumbe gesald wass Ssem feder sona witgende = l^ato iohanne uox eloqui muto redditur patri ilico prophetanti. 38 Of these 17 inflected infinitives, 7 occur in the "Introductions": with one exception all of the uninflected infinitives occur in the Gospels proper. 172 The Syntax of the Lindisfarne Gospels II. THE INFINITIVE INFLECTED Of the inflected infinitive modifying a Noun, the examples in full are as follows (17, of which 7 are in the " Introduc- tions ") : — becon, sign (1) : J. I. 3.13 : huast becon gesalde to undoanne tempul & ym Srim dagum wseccennes clsen run uel godes degelnise sette(5 = quod signum daret soluendi templum et in triduo excitandi mis- terium ponit. bisen, example (2) : Mat. (1) : I. 20.19: bisen of Saem wyrcendum in wingeard tid ungelic uel unefne efnegebrohtum to brenganne an & gelic cue(5 hea onfeingon mearde = Parabolam (sic for parabola?) de operariis in uinea tempore diuerso conductis adhibita imam paremque dicit eos accepisse mercedem. — L. (1): I. 8.5' a : of gemnisse [= blanks] mi(5 bisseno oxes to ivcetranne aefsa?- geS = de cura sabbati murmurantes exemplo bouis adaquandi refellit. eare, ear (5) : ■ Mat. (1): 13.9: seSe haefes earo to herranne geheraS = qui habet awes audiendi audiat. — Mk. (2): 4.9; 7.16. — L. (2): I. 8.15; 14.35. (h)lattO, delay (1) : L. (1): 12.45 a : hlatto doaS hlaford min to cumanne = mo- ram facit dominus meus uenire. huothwoegU, something (1): L. (1): 7.40: ic hafo <5e huothuoego to cuoed anne = habeo tibi aliquid dicere. maeht, might, power (6) : Mat. (1) : 9.6: ha?fes mceht on eorSo forgefnisse uel to forge- fanne synna = habet potestatem in terra dimittendi peccata. — /. (5): I. 6.5; 10.1 8 a ' b ; 19.10V (quoted above under Unin- flected, p. 171). rod, rood, cross ( 1 ) : L. (1) : 23.26**: geseton him Saet rod to bearanne sefter <5one The Infinitive 173 hselend = imposuerimt illi crucem portare post iesum (or Final ?). [The W. S. has: (5a rode him onsetton fleet he hi bcere aefter Sam haelende.] DIFFERENTIATION OF THE TWO INFINITIVES As to the Differentiation of the Two Infinitives, when modi- fying a Noun, the lack of inflection seems to me due partly to the remoteness of the uninflected infinitive from the noun that it modifies, partly to the relatively large number of infinitives instead of gerunds or gerundives in the Latin original, and partly to the idiosyncrasy of the glossator. Sometimes, how- ever, as in L. I. 3.16 a , the infinitive is not inflected although in juxtaposition with the noun that it modifies. How puzzled the glossator must have been at times is shown in J. 19.10 a > b > c > d (quoted on p. 171 above) : with the same word, mceht, the glos- sator in the first instance writes the uninflected infinitive and gives the inflected as the alternative rendering, but in the second instance reverses the order! Again, in L. 12.5, also quoted on p. 171 above, after mceht the glossator writes first the unin- flected infinitive and, interchangeably, the inflected infinitive. That, as a rule, however, with nouns the infinitive was inflected in the Lindisfarne Gospels as in West-Saxon, is evident from the fact, already incidentally stated, that the inflected form predominates over the uninflected in the proportion of 17 to 7. The Latin correspondents to the infinitive with nouns is as follows: — for the uninflected infinitive: a noun with an infini- tive, 7 ; for the inflected infinitive: a noun with a gerund in the genitive, 10 ; a noun with a gerundive in the genitive, 1 ; a noun with an infinitive, 5 ; an absolute participle phrase, 1. ORIGIN OF THE IDIOM The evidence of the Lindisfarne Gospels as to the Origin of the Infinitive with Nouns in Anglo-Saxon seems to me in line with that derived from the West-Saxon. The idiom was native to West-Saxon, but the frequency of the construction therein 174 The Syntax of the Lindisfame Gospels was in part due to the frequency of the constructions with gerund or gerundive in the Latin originals. For this idiom in West-Saxon, see The Infinitive in Anglo- Saxon, pp. 173-182, 220; and in the other Germanic languages, pp. 262-264. B. The Passive Infinitive No example has been found of the compound passive infini- tive modifying a noun in the Lindisfame Gospels. Nor was any example found in West-Saxon. NOTES 1. The Two Infinitives Alternate in the following: L. 12.5 and J. 19.10 a ' c , quoted on p. 171 above. 2. An Uninfected Infinitive Alternates with a Clause in /. 1.12, quoted on p. 171 above. 3. The Inflected Infinitive Used as a Latin Gerundive. — In L. I. 8.5 a , quoted on p. 172 above, the Lindisfame inflected infinitive closely approximates a Latin gerundive. For this phenomenon in West-Saxon, see The Infinitive in Anglo-Saxon. p. 182. 4. An Inflected Infinitive Interchanges with a Verbal Noun in " -Nes(s)" that translates a Latin gerund in the genitive in the following: J. I. 3.13, wceccennes, quoted on p. 172 above; J . I. 6.5 : mceht hasfde he cuaeS setnese & eft to onfoanne sauel his — Potestatem habere se dicit ponendi ac resumendi animam suam. Cf. J. lO.lS 3 ^, in which we have two infinitives: mceht ic hafo to settanne hia ... & mceht ic hafo aeftersona to on- foanne hia = potestatem habeo ponendi earn et potestatem habeo iterum sumendi earn. — In the following an inflected in- finitive alternates with a verbal noun in -?ies that translates a Latin gerund in the genitive, for a single word: Mat. 9.6, for- gefnisse, quoted on p. 172 above; L. 14.35: se(5e hsefe(5 earo to heranne uel hernises gehereS = qui habet aures audiendi audiat. A noun in -nes translates a Latin gerund in the genitive in the following: — Mb. 4.23: gif hua hsefeft earo hemisses geheraS = siquis habet aures audiendi audiat; L. 8.8 C : seSe hsefe(5 earo The Infinitive 175 hemises geheraS = qui habet aures audiendi audiat. — See Note 4 to Section II of the present chapter, p. 114, for an example of the verbal noun in -ing translating- a Latin infinitive. XIV. THE IMPERATIVE INFINITIVE A use of the infinitive that is not found in West-Saxon, namely, the Imperative, occurs occasionally in the Lindisfarne Gospels. Of this use I have found the following seven exam- ples ; in each of which the infinitive is uninflected, and is active in sense as in form : — Mk. 11.23 a ' b : so<51ice ic cuoeS iuh Ssette suaha seSe cuoeSas Sissum more genioma & senda on sae = amen dico uobis quia quicumque dixerit huic monti tollere et mittere in mare. [W. S. : imperative subjunctive; Kiushw. : ginioma & sende in sae.] L. I. 6.13 13 : foxas holo habbaS getrymade esne fylgende for- geafa uel forleta Sa deado & <5one haldond sulh ne eftlocia on bsec — uulpes foueas habere testatur iuuenis (sic) sequent! dimittere mortuos et tenentem aratrum non respicere retro. [Or Objective? Cook: forgeafa, inf.; forleta, inf.; esne, ns. ; fylgende, not cited for this passage.] L. 17.6 a ' b : gie cuoede <5isum tree i heartbreer ofivyrtrumia & oferplontia uel gesette on saa & hersumia<5 iuh = diceritis huic arbori moro eradicare et transplantare in mare et obediret uobis. [W. S. : imperative subjunctive ; Rushw. : imperative subjunctive; Cook: ofivyrtrumia, inf.; oferplontia, inf.] L. 8.28 : ic biddo Sec ne mec drouiga uel Ssette <5u mec ne gegroeta uel ne pinia = obsecro te ne me torqueas. [W. S. : ic halsige (5e <5set &u ne Sreage me ; Rushw. : lacking. Possibly our infinitive is used Predicatively here? Cook: Browiga, in- finitive; gegroeta, optative present, 2nd singular.] •7. 8.5 : in 83 uutudlice moises bebead us Suslic gestwna Su ; f orSon hua3d cwaaSes ? = in lege autem moses mandauit nobis hujusmodi lapidare tu; ergo quid dicis? [W. S. : moyses us bebead on Ssere 03 Sset we sceoldon Sus gerade mid stanum oftor- fian; — Rushw.: bibeod us (5uslic nu gistcena (5u ; — Cook: ges- tcena, infinitive. ] 176 The Syntax of the Lindisfarne Gospels ORIGIxNT OF THE IDIOM Two of the above examples of the Imperative Infinitive (L. I. 6.13 b and L. 8.28), as there indicated, are doubtful; but the remaining five examples seem pretty certainly to be- long here. And clearly these examples are due to the close fol- lowing of the Latin original, in five of them (Mk. 11.23 ,a > b ; L. I. 6.13 b ( ?) ; L. 17.6 a > b ) to the too close following of the Latin, for we have the present infinitive translating a Latin passive of the imperative mood, which latter is identical in form with the Latin active infinitive. In J. 8.5, the Northumbrian impera- tive infinitive corresponds to the same idiom in Latin. In L. 8.28, the infinitive corresponds to a Latin jussive subjunctive. For alleged examples of the Imperative Infinitive in West- Saxon, see The Infinitive in Anglo-Saxon, p. 6. For this con- struction in the kindred Germanic languages, see the same mono- graph, p. 264. XV. SOME SUBSTITUTES FOR THE INFINITIVE I. THE PREDICATE NOMINATIVE OF THE PRESENT PARTICIPLE FOR THE PREDICATIVE INFINITIVE AFTER VERBS OF MOTION In the Lindisfarne Gospels I have not found any clear exam- ples of the Predicate Nominative of the Present Participle used for the Predicative Infinitive after Verbs of Motion, as in the West-Saxon He com fieogende from lie com fleogan. Two possible examples are the following: — Mat. 14.25: Sin feorSa uutedlice waccen naehtes cuom to him geongende net geeode ofer see = quarta autem uigilia noctis uenit ad eos ambulans supra mare (Rushw. and W. S. : ibidem); Mk. I. 3.14: om feorSa na3ht wacan cuom to Segnum geongende ofer uel buf a sa? = Quarta noctis uigilia uenit ad discipulos ambulans supra mare. But in each the participle may be used appositively instead of predicativcly ; and each has been put in the chapter on the Appositive Participle, p. 64 above. Concerning this idiom in West-Saxon and in the other Ger- manic languages, see The Infinitive in Anglo-Saxon, pp. 221-224. The Infinitive 177 II. THE PREDICATE ACCUSATIVE OF THE PRESENT PARTICIPLE FOR THE PREDICATIVE INFINITIVE WITH ACCUSATIVE SUB- JECT About 94 (96 including the " Introductions ") examples of the Predicate Accusative of the Present Participle instead of the Predicative Infinitive with Accusative Subject occur in the Lindisfarne Gospels to 80 in the West-Saxon Gospels. The following are typical examples: — Mk. 14.58 a : forSon ue geherdon hine cwoedne uel cwoedende = audiuimus eum dicentem. Mat. 26.40 : gemoete hia, slepende = inuenit eos dormientes. — Mat. 20.6: gemoette odero stondende = inuenit alios stantes. Mat. 26.64 a > b : gie gesead sunu monnes sittenda ... & cym- mende = uidebitis filium hominis sedentem . . . et uenientem. — Mat. 20.3 : gesceh odero standende in spree uel in Singstow = uidit alios stantes in foro. Mat. 25.44 b ' c : huoenne dec we segon hyncgerende uel fiyr- standa? = quando te uidimus esurientem aut sitientem? The statistics of the Predicate Accusative of the Present Par- ticiple are in full as follows (96 in all, 2 in the " Introduc- tions"): — behalda, 'see' (1): — geongende, ' going ' (1) : J. (1) : 1.36. forleta, ' dismiss' (2) : -fcestende, '' fasting ' (2); Mat. (1) : 15.32.— Mfc. (1) : 8.3. gehera, 'hear' (6) : — bifcerende, ' passing by ' (1) : L. (1) : 18.36. — cuoedende, ' saying ' (1) : Mk. (1) : 14.58 a . — forecwoedende, 'preaching' (1): Mat. (1): I. 6.9. — hucestrende, ' murmuring ' (1) : J. (1) : 7.32. — soecende, ' seeking ' (1) : Mk. (1) : 12.28. — spreccende, ' speaking ' (1):-/. (1) : 1.37. gemitta: see gemoeta. gemoeta (gemitta), ' find ' (20) : — cymmende, 'coming' (1): Mat. (1): 27.32. 12 178 The Syntax of the Lindisfarne Gospels — doende, ' doing ' (1) : L. (1) : 12.43. — forbeadende, ' forbidding ' (1) : L. (1) : 23.2 d . — jraegnende, 'enquiring' (1) : L. (1): 2.46 c . — hcebbende, 'having' (1): J. (1): 11.17. - herende, ' hearing ' (1) : L. (1) : 2.46 b . — Hccende (licgende), 'reclining' (1): Mk. (1): 7.30. — sittende, 'sitting' (3): L. (2): 2.46' a ; 8.35 b .— /. (1) : 2.14 a (sittendo). — slepende, 'sleeping' (6): Mat. (2): 26.40, 43. — Mk. (3): 13.36; 14.37, 40.— L. (1): 22.45. — stondende, 'standing' (2): Mat. (1): 20.6.— L. (1) : 19.32. — ■ under cerrende, ' subverting ' (1) : L. (1) : 23. 2 C . — wceccendo, 'watching' (1): L. (1): 12.37 a . gesea, 'see' (59) : L. (1): 9.49. Mat. (1) : 21.15 a . Mat. (1): 8.14 b . — aworpende, ' casting out ' ( 1 ) — clioppende, ' crying out ' ( 1 ) — cuacende, ' having fever ' ( 1 ) — cuedendo {sic for cuedendo), ' speaking ' (1) : Mat. (1) : 21.15 b . — cum(m)ende (cym(m)ende), 'coming' (11): Mat. (4): I. 20.6; 16.28; 24.30 ;*26.64 b — M. (3): 9.1; 13.26; 14.62 b . — L. (1): 21.27.— J". (3): 1.29, 47; 10.12. — cym(m)ende: see cum(m)ende. — fallende, 'falling' (1): L. (1): 10.18. — fylgendi (-do), 'following' (2): /. (2): 1.38 (-ndo) ; 21.20 {-ndi). — geongende, 'going,' 'walking' (6): Mat. (2): 14.26. a ; 15.31.— Mh. (3) : 6.3, 49 ; 8.24.— J. (1) : 6.19. — licgende (Uccende), 'lying down' (2): Mat. (1): 8.14 a (liccende). — /. (1): 5.6. — mcenende (mceniende), 'mourning' (2): Mat. (1): 9.23. — Mh (1): 5.38 b (mceniende). — ofdunestigende, ' descending ' (1) : J. (1) : 1.51 b . — ofstigende, ' descending ' (1) : Mk. (1) : 1.10 b . — sendende, 'casting' (2): Mk. (1): 1.16*.— L. (1): 21.2. The Infinitive 179 — sittende, 'sitting' (9): Mat. (2): 9.9; 26.64 a (-nda).— MJc. (4): 2.14; 5.15 a ; 14.62 a ; 16.5 a — L. (2): 5.27; 22.56.— ./. (1): 20.12 (sittendce). — standende : see stondende. — stigende, 'descending' (4): Mat. (1): 3.16 a .— J. (3): 1.32, 33 b ; 6.62. — stondende {standende), 'standing' (5): Mat. (1): 20.3. —Mk. (1): 13.14.— L. (1): 5.2 (-ndo).-J. (2): 19.26; 20.14. — up (p) stigende, ' ascending ' (2) : L. (1) : 12.54. — J". (1) : 1.51 a . — Sringende, ' pressing upon ' (1) : Mk. (1) : 5.31. — wcermigende, 'warming' (1): Mk. (1): 14.67. — woepende, 'weeping' (1): Mk. (1): 5.38 a . — wunigende, 'remaining' (2): Mk. (1): 1.10 c . — J. (1) : 1.33 c . — wynnende (winnende), 'laboring' (1): Mk. (1): 6.48 a . — wyrcende, ' doing ' (1) : J. (1) : 5.19. habba, 'have' (1) : — uuniande, ' abiding ' (1) : J. (1) : 5.38. infinda, find (l) : — restende, 'resting,' ' lying vacant' (1) : Mat. (1) : 12.44. onfinda, 'find' (1): — doende, 'doing' (1): Mat. (1): 24.46. sea, 'see' (5) : — kremende, 'weeping' (1): /. (1): 11.33. — hyncgerende : see hyngrende. — hyngrende (hyncgerende), 'hungering' (2): Mat. (2): 25.37 a , 44 b (hyncgerende) . — dyrstende, 'thirsting' (2): Mat. (2): 25.37 b , 44 c (dyr- stenda). ORIGIN OF THE IDIOM In all the 96 examples except two, the Northumbrian Pred- icate Accusative of the Present Participle translates the same idiom of the Latin; in the two exceptional cases, it translates a Latin adjective in the predicate accusative. The two exceptions 180 The Syntax of the Lindisfarne Gospels are as follows: — Mat. 15.32: forlette hia fwstende nwill (sic) ic = dimittere eos ieiunios nolo ; Mh. 8.3 : gif ic forleto hia fces- tende in hus = si dimisero eos ieiunios in domum. Moreover, the general dislike of the Northumbrian for the participle with verbal force is shown in the avoidance, by the Lindisfarne gloss- ator, of the predicative accusative of the participle even when the idiom occurs in the Latin original, as in his awkward rendi- tion, in Mat. 15.31 a ' b , of the participle by a finite verb: sua (5set <5reatas wundradun gesegon monigo gesprecon halto geeadon uel geongende blindo gesegon uel geseas = ita ut turbae mira- rentur uidentes multos (sic) loquentes clodos ambulantes caecos uidentes. The evidence, therefore, of the Lindisfarne Gospels tends to confirm the conclusion reached from a study of this idiom in West-Saxon, namely, that the predicative use of the accusative of the present participle was not a native idiom in English or in the Germanic languages as a whole, and that the substitution of the predicate accusative of the present par- ticiple for the predicative infinitive was due to Latin influence. See The Infinitive in Anglo-Saxon, pp. 225-230. III. THE ELLIPTICAL ACCUSATIVE-WITH-INFINITIVE CONSTRUC- TION A. Made up of an Accusative Plus a Present Participle Somewhat different from the preceding idiom is what for lack of a better name I should denominate the Elliptical Accu- sative-with-Infinitive Construction. This idiom is made up usually of an Accusative Plus a Present Participle, and trans- lates a Latin elliptical accusative-with-infinitive construction consisting either (a) of an accusative plus a future participle without esse or, more frequently, (b) of an accusative plus a gerundive without esse. In the idiom considered in the pre- ceding section, the principal verbs are usually verbs of sense perception and occasionally verbs of mental perception ; and the participle translates almost invariably a Latin predicative pre- sent participle. In the present idiom, the principal verbs are usually verbs of commanding or of declaring, never verbs of The Infinitive 181 sense perception; and the participle translates, not a Latin present participle, but a Latin future participle or, usually, a Latin gerundive. Of course, when translating the Latin gerundive, in (b), the Lindisfarne participle, though active in form, is passive in sense. I cite all the clearer examples ob- served of the Elliptical Accusative-with-Infinitive Construction of each of these two types : — (a) (8 examples, all in the " Introductions ") : — bebeada, command, urge (1) : L. (1) : I. 8.17: geddung . . . settee!) . . . behead (5set gefea from hreownisum hselo tocymende = parabolam . . . ponit . . . com.mendans gaudium de paeni- tentum salute futurum. forecuoeSa, announce, predict (1): L. (1): I. 10.16: hine tocymende on wolcen . . . forecuoed = seque uenturum in nube . . . praedicit. foressecga, announce, declare, prophesy (3) : Mat. (2) : I. 20.4 : gedrouende hine foresoegde = Passurum se prcenuntians ; I. 20.20 a .— L. (1) : I. 6.6. gehata, promise (1) : Mat. (1) : I. 18. l b : (5a iudeas of ric in Son cynna geheht tocymmende saegeS fordrifena = iudaeos de regno in quo gentes promittit uenturas asserit expellendos. geondwearda, respond, answer (1).: L. (1) : I. 9.10 a : wses frsegnend from tid rices godes geonduearde ne miS gehald tocy- mende = Interrogatus de tempore regni dei respondit non cum obseruatione uenturum (the subject accusative is to be supplied both in the Northumbrian gloss and in the Latin original). SOSssecga, declare (1) : J. (1) : T. 6.16 a : corn husetes dead monigf aid usestem sodsceges tobrengende embehtande = granum frumenti mortuum multum fructum pronuntiat allaturum min- istrantem. (h) (11 examples, all in the " Introductions ") : — beada, command, order (1): J. (1): I. 8.11: & foedendo Sa scipo gelic Sirda? beadande . . . tahte etc. = et pascendas ones a>que tertio commendans . . . significat ei quod etc. ( ?). CUOeSa, say, declare (2) : L. (2) : I. 5.17 b : & da (5set moder & broSro ceigendo cuoe& = et eos matrem fratresque uocandos 182 The Syntax of the Lindisfame Gospels ait; I. 9.10 b : niomande hia cuoed menn = occupandos eos dicit homines. forecuoeSa, predict (1) : Mk. (1): I. 4.14 a : ec forecued hine slaende = Item praedicens se occidendum. foresaecga, declare, predict (2): Mat. (1): I. 22.4: fore- sceged eastro sefter tusem dogrum & hine sellende = Praenuntiat pascha post biduum seque tradendum. — Mk. (1) : I. 4.14 a . geonduearda, respond (1): L. (1): I. 10.14 b : geonducerde yflo monigo forescyniga ne foresmeande huced gesald gescprecen weron = respondit mala plurima praecessura nee praecogitan- dum quid traditi loquantur. lsera, teach (1) : Mat. (1) : I. 17.10 (1) : ne oisignisse mettes & woedes hcebende ah ric godes allum fore — [= blank] Iceras = nee sollicitudinem escae et uestis habendam sed regnum dei omnibus prseferendum docet. Serhgebidda, persuade (3) : L. (3) : I. 7.4 a ' b > c : frionde Srio hlafas a?d middemasbt Sam biddende Sset getdung sette giwende, soecende, & cnylsende derhgebiddes = Amici tres panes media nocte petiti parabolam ponens petendum, quaerendum, pulsan- dumque persuadit. B. Made up of an Accusative Plus a Preterite Participle Again, we find a few examples (about 8 in all) of an Ellip- tical Accusative-with-Infinitive Construction made up of an Ac- cusative Plus a Preterite Participle and translating a Latin Elliptical Accusative-with-Infinitive Construction consisting (a) usually (six times) of an accusative plus a gerundive with- out esse and (b) twice of an accusative plus a past participle. In sense this locution is passive, of course. The examples in full are as follows : — (a) (6 examples, all in the "Introductions"): — bebeada, command, exhort (1): L. (1): I. 6.20 b : inlsedeS geddung Sees wundes Ssem gebroht from samaritanum mild- heortnise bebead gebisened = inducit parabolam uulnerati cui conlatam a samaritano misericordiam praecipit imitandam. The Infinitive 183 forebeada, forbid, prohibit (1) : Mk. (1) : I. 4.15 b : haedno forebeadend gelicad uel gebis hersumnise mi(5 bisene sedeawed = gentiles prohibens imitandos humilitatis exemplo monstrato. foresaecga, announce, declare (2) : L. (2) : I. 10. 2 a : fiondas rices his foresceged derhgelicade = inimicos regni sui pronun- tiat perimendos; I. 10.4 : gesaeh (5a ceastra gewsep ofer hia /ore- scegde ofacerred = Uidens ciuitatem fleuit super earn praenun- tians euertendam (in Northumbrian and in Latin the subject accusative is to be supplied from the context) . gelsera, teach (1): Mat. (1): I. 16.14°: gelaeres sodfeast- nisse ofersuided = docet iustitiam superandam. saecga, say, declare (1): Mat. (1): I. 18. l c : (5a iudeas of ric in Son cynna geheht tocymmende scegeS fordrifena = iudaeos de regno in quo gentes promittit uenturas asserit expel- lendos. (b) (2 examples) : ssecga, say, declare (2) : Mk. (2) : I. 1.6 b , 7 : in stefne engles soecgende gesended ne (5set ane word lichoma geworden ah lichoma drihtes (5erh word godcundes stefn gesaweled = in uoce angeli enuntiantis emissum non solum uerbum caro factum sed corpus domini per uerbum diuinse uocis animatum. ORIGIN OF THE IDIOM It seems clear that the Elliptical Accusative-with-Infinitive Construction, especially type A, made up of an Accusative Plus a Present Participle, arises from the very close translation of the Latin original. I am inclined to believe, too, that type B of this construction, made up of an Accusative Plus a Past Participle, is likewise due to the imitation of the Latin idiom, though some may prefer to consider it as merely a sub-variety of the native accusative plus a predicative past participle. Of neither of these two major types of the Elliptical Accusative- with-Infinitive Construction, have T found a clear example in West-Saxon. 184 The Syntax of the Lindisfarne Gospels NOTES 1. The Elliptical Accusative-With-Infinitive Construction Consists of an Accusative Plus a Past Participle Tlwi Is Active in Sense, and translates a Latin accusative plus a perfect active infinitive in J. 12.18 : forSa uel foreSon & ongaegn uel togsegnes cuom him Se here forSon uel Ssette geherdon hine geuorhto Sis becon = propterea et obuiam uenit ei turba quia audierunt eum fecisse hoc signum. 2. The Lindisfarne Present Participle Corresponds, not to a Latin Future Participle, but to the Perfect Participle of a Deponent Verb in Mh. I. 4.5 : iohannis — (= blank) weron Srowende he gelicra drowende cuoed = et iohannem heliae pas- suri ipse similia passum dicit. 3. An Abortive Attempt at the Elliptical Accusative-with- Infinitive Construction occurs in the passage below, in which the failure seems to arise from the glossator's misunderstanding of fore: L. I. 6.9 : efter sona cuoed hine drouande were toweard — Iterum dicit se passurum fore. 4. Apparent Examples of the Predicative Present Parti- ciple in a Passive Sense. — In his " Development of the Pro- gressive Form in Germanic " (in Publications of the Modern Language Association of America, xxvm, 1913, pp. 182-183), Professor G. O. Curme holds that, in the following passages of the Lindisfarne Gospels, the Predicative Present Participle is used in a Passive Sense: — Matthew 13.19, 20: eghuelc seSe heres word rices & ne oncneawu uel ne ongset, cuom Se Siowl uel Se yfle & genom uel gelahte Sset gesawen wses in hearta is, Ses is uel wses seSe neh strete uel woeg sawende woes; seSe soSlice ofer uel on staener sawende wees, Sis is uel wses seSe word geherde etc. = omnis qui audit uerbum regni et non intel- legit, uenit malus et rapit quod seminatum est in corde eius, hie est qui secus uiam seminatus est; qui autem supra petrosa seminatus est, hie est qui uerbum audit etc. ; Mat. 13.22 : seSe uutedlice ivces saivcende in Sornum, Ses uel Sis is seSe word heres etc. = qui autem est seminatus in spinis, hie est qui uer- bum audit etc.; Mat. 13.23: seSe uutedlice in eorSo godo sa- The Infinitive 185 wende wees, Sis is uel wses seSe heres word etc. = qui uero in terra bona seminatus est, hie est qui audit uerbum etc. ; Luke 7.12 : heono dead wees ferende sunu ancende moderes his = ecce defunctus efferebatur films unicus matris suae. The West- Saxon Gospels has the past participle in all of these examples, as has the Rushworth Gospels in all except in the example from Luke, which latter is lacking therein. Professor Curme may be right in contending that the present participle in the five examples of the predicative use just quoted is passive in sense; for, as we have seen, occasionally we find a present absolute or a present appositive participle of the Lindisfarne Gospels trans- lating a Latin passive participle: see pp. 5, 6, and 86 above. As there indicated, the present absolute and the present apposi- tive participles seem due to an oversight. I think it more probable, however, that in the first four examples just quoted the participle is active in sense. The glossator, seeing that the Latin verb has a personal subject, considers that the subject of sawende woes is literally the sower, not symbolically the sower standing for the seed; for, where the subject of the Latin verb is unmistakably impersonal, standing for the seed or the word, the Lindisfarne glossator uses the passive participle, as in Mat. 13.19 above (word . . . 8cet gesawen ivces = uerbum . . . quod seminatidm est). Compare, further, Mk. 4.15 a ' b (Sser bid gesauen word . . . word dcette gesawen ivces = ubi seminatur uerbum . . . uerbum quod seminatum est), in which we have a Lindisfarne impersonal subject and a passive participle trans- lating a Latin impersonal subject and a passive verb, with Mk. 4.16 (Sas sint gelic Bade ofer steenero saues uel sauad = hi sunt similiter qui super petrosa seminantur) and Mk. 4.18 (ox5ero sint dade on Sornum hia saued uel sanas = alii sunt qui in spinis seminantur), in which we have a Lindisfarne personal subject and an active finite verb translating a Latin personal subject and a passive finite verb. In Mk. 4.20 (Sa sint dade ofer eorSo god gesauen sint = hi sunt qui super terram bonani seminati sunt), however, we have a Lindisfarne personal subject and a passive participle translating a Latin personal subject and 186 The Syntax of the Lindisfarne Gospels a passive finite verb made up of the verb to be plus the passive participle : possibly the frequency of the Latin expression finally overrules the glossator's original interpretation? As to L. 7.12, quoted above, the present participle may be due in part to the proximity of the perfect participle (defunctus) of a deponent verb to efferebatur, since, in West-Saxon and in Northumbrian, Latin deponent verbs are often translated by the progressive tense. But, in the following example, which is not mentioned by Professor Curme, there is no such excuse for the error, real or apparent: — L. I. 9.9: ivces fregnend from tid rices godes geonduearde ne miS gehald tocymende = Interrogatus de tem- pore regni dei respondit non cum obseruatione uenturum. APPENDIX IX GERMANIC ANALOGUES TO THE NORTHUMBRIAN ELLIPTICAL ACCUSATIVE-WITH-INFINITIVE CONSTRUCTION In the Scandinavian Languages we find an instructive paral- lel to the Northumbrian Elliptical Accusative-with-Infinitive Construction, of which parallel, however, I did not become fully aware until after I had written the preceding part of this sec- tion, when I secured a copy of Professor Nygaard's " Om Brugen af det Saakaldte Prsesens Particip i Oldnorsk (1879)," to which article I have several times referred already in other parts of the present study. Concerning the idiom about to be discussed, I had read brief statements in the treatises on Scandinavian syntax by Lund (§154), by Falk and Torp (§138), and by Nygaard himself in his Norr0n Syntax (§§238- 242), but I did not realize the closeness of the parallel until I had read Professor Nygaard's aforementioned essay on the Present Participle in Old Norse. On pp. 219 ff. of this essay, Professor Nygaard tells us that, in later and learned Old Norse works, the Present Participle is at times used Predicatively 37 in a Passive Sense with vera to denote suitability, obligation, or necessity, as in Horn. 9.6: sja 37 Occasionally also attributively. The Infinitive 187 friftr er haldandi meS goSum (custodienda est) ; Horn. 28.18: f>at er vitanda mildlegt verk vist (sciendum est esse pietatis opus). He tells us, what is apparent from the examples just given, that the Norse idiom corresponds closely to the Latin passive periphrastic conjugation; and that it is, as a rule, found only in the writings that show Latin influence. The statement is of so great importance and the essay is so difficult to get hold of that I must quote Professor ISTygaard's own words (pp. 219 ff.) :— " I passiv betydning betegner prses. part. ' som er at gj0re ' d. e. ' som egner sig til at behandles eller til at blive gjenstand for en vis virksomhed.' Det udtrykker saaledes dels skikkethed eller vcerdighed (bekvem, skikket til at — , vcerdig, kvalificeret til at — ) dels mulighed (som lader sig behandle, kan behandles) dels tilb0rlighed og n0dvendighed (som fortjener at behandles, b0r, maa behandles). " Den passive betydning h0rer hjemme i transitive verber og findes baade i prcedicering og i attributiv stilling, men har over- hovedet ikke nogen udstrakt anvendelse. " Navnlig er det passive prsesensparticip meget sjeldent som attribut, og hyppigst bruges det i forbindelse med verbet vera forat betegne, at en handling bor eller maa ske (er tilb0rlig eller nddvendig). Denne forbindelse svarer til den latinske saa- kaldte omskrivende konjugation i passiv, og anvendes ogsaa fortrinsvis i saadanne skrifter, hvor latinsk paavirkning er kjendelig. " Hvor participiet er prsedikatsord, er subjektet, og hvor det er attribut, det substantiv, hvortil det er f0iet, en person eller ting, som i aktiv er gjenstand for verbets virksomhed. Det logiske subjekt saettes i dativ. " Sjelden bruges det passive part, upersonligt med vera: er fra hanum ekki lengra segjandi (Str. 35.38). " Af intransitive verber kan neutrum af participiet bruges i passiv betydning med verbet vera som et upersonligt udtryk, og 188 The Syntax of the Lindisfarne Gospels dertil kan foies den styrelse, hvornied verbet forbindes i aktiv : slikt hit sama er rennandi (currendum est) meS. sannri van til hugganar yfirlegrar mildi (Horn. 7. 2) ; allra hluta fyrst er manni leitandi (quaerendum est homini) hvat se sonn speki (Horn. 3, 6). . . " Ganske enkeltvis findes et lignende upersonligt udtryk af transitive verber med tilf0iet objekt i akkusativ: i hvern tima gefanda er bverja gjof (Kgs. 170.7) ; " etc. On p. 225, Professor Nygaard gives some interesting statis- tics as to the relative infrequency of the present participle (whether with or without an object and whether active or passive in sense) in the earlier and more original Old Norse works. He states that not a single example of the present participle with a passive sense is found in the Elder Edda. Still later, on p. 227, Professor Nygaard takes up the ques- tion as to the origin of the use of the present participle in a passive sense in Old Norse. Eecalling his statement that no example of this idiom is found in the Elder Edda, and that the construction appears chiefly in translations from the Latin or in imitations of Latin works, I expected him to say what all of his evidence seemed to make probable if not inevitable, namely, that the construction is due to Latin influence. But not so ; here is his to me surprizing answer : " Men hvorledes er de to forskjellige betydninger, den aktive og den passive, at stille til hinanden ? er den sidste yngre end den f0rste, og naar er den isaafald opstaaet? Derom er det vanskeligt at udtale sig med bestemthed. Vistnok forholder det sig, som paapeget, saa, at den passive brug ikke forekommer i den seldre edda, medens den f. ex. i ' Kongespeilet ' er temmelig almindelig. Deraf tor man dog ikke slutte, at den f0rst er udviklet i en senere tid og slet ikke fandtes paa det sprogtrin, der reprsesenteres af edda. Rimeligvis er den gammel, men bar i tidligere tider vseret endnu mere tilbagetra?dende i forhold til den aktive brug, end den senere blev. Men det synes at vsere sikkert, at de to betyd- ninger ikke er lige oprindelige i alle slags verber, at de tvserti- mod fra f0rst af har vseret knvttede til hver sin gruppe. Det The Infinitive 189 samme suffix liar, i'0iet til verbalstammen, af intransitive verber dannet et aktivt verbaladjektiv og af transitive verber et passivt. Saaledes er jo ogsaa tilfseldet med det andet verbaladjektiv, det saakaldte fortidsparticip, at de samme suffixer i intransitive verber f rembringer begrebet : ' indtraadt i en vis stilling eller tilstand ' (aktivt), og i transitive: ' bragt i en vis stilling eller tilstand, paavirket, behandlet ' (passivt). Det aktive particip har endvidere fra f0rst af havt sin anvendelse i apposition og som attribut, ikke i prsedicering, medens det passive fortrinsvis er benvttet i den sidste stilling. Senere er dette forbold for- rykket, idet ogsaa transitive verber danner part, med aktiv be- tydning (meget sjelden faar man derimod omvendt passive part, af intransitive verber), saaat altsaa begge betydninger kan findes ved siden af hinanden i samme ord. Disse aktive participier af trans, verber bruges f0rst som substantiver til at betegne den handlende person, siden ogsaa adjektivisk. Sam- tidig dermed er man ogsaa begyndt at anvende det aktive part, i prsedicering. Allersidst opstaar enkelte reflexiv-formede nutidsparticiper (sic) med reflexiv eller passiv betydning." Whether, in his more recent and brief treatment of the pres- ent participle with a passive meaning, given twenty-seven years later in his Norr0n Syntax (Kristiania, 1906), Professor Ny- gaard still holds to the same view as to the origin of this con- struction in the Scandinavian languages, it is difficult to deter- mine from his statements. To me, however, the later state- ments seem somewhat more favorable to the theory of Latin influence than do the earlier ones. Of the present participle with passive signification he says, in §239 of his Norr0n Syn- tax: " Som prsedikatsord f0ies det i F. S. 38 ikke ganske sjelden isser til vera. . . . Hyppigere bruges det saaledes i L. S. 38 Xavnlig bruges her meget ofte dette passive 39 part, i forb. med vera som en omskrivning for at betegne, at en handling M F. S. = Folkelig Stil, 'popular style'; and L. S. = Lserd Stil, 'learned style.' 39 He means the present participle used in a passive sense, both here and in my next quotation. 190 The Syntax of the Lindisfarne Gospels b0r, maa, kan ske (er tilb0rlig, n0dvendig, mulig) i lighed med den latinske omskrivende konjugation i passiv." And in his summary concerning the present participle, he says, among other things, in § 242 : " Det oprindelige forhold med hensyn til prses. part, synes altsaa at have vaeret : prses. part, af intran- sitive verber har aktiv, af transitive passiv betydning ; de f 0rste bruges i apposition og som attribut, sjelden i prsedicering ; de sidste, som idethele er lidet hyppige, saagodtsom udelukkende som prsedikatsord ; prses. part, er i det vsesentlige at betragte som adjektiver; de har ingen fremtraedende tidsbetydning og har nominal styrelse (genitiv). " Saaledes vedbliver ogsaa idethele sprogbrugen at vsere i F. S. " I L. S. udvides og modificeres brugen ved paavirkning af fremmede analogier, navnlig det latinske nutidsparticip og passive verbaladjektiv. " Det aktive part, dannes her ogsaa af transitive verber og faar verbal karakter (med tydelig tidsbetydning og verbal styr- else) . " Det passive part, bruges i forb. med vera hyppig som et omskrevet udtryk om pligt, n0dvendighed eller mulighed (i lighed med latinsk omskrivende konjugation i passiv)." Possibly Professor Nygaard's position is this : the passive use of the present participle was a native Germanic construction, but its use was extended by Latin influence. To me, however, it seems more probable that, in Old Norse as in Northumbrian, the idiom owes its origin to the Latin. This view gains in probability, I think, if we recall that in several other construc- tions, notably in the absolute use of the dative and in the apposi- tive use of the present participle governing an accusative object, Old Norse, according to Professor Nygaard himself, was a direct borrower from the Latin, as was Old English also. Professor Heusler calls the passive use of the present parti- ciple its gerundive use, and seems to consider that this idiom is not of native origin. See § 423 of his Altisldndisches Elemen- tarbuch (Heidelberg, 1913), which runs as follows: — The Infinitive 191 " Von dem Gesagten 40 sind auch in der heimischen Prosa ausgenommen: "1. Die substantivierten Partt., die namentlich in der Rechtssprache beliebt sind ; sie werden auch von Transitiva ge- bildet : veriande ' Verteidiger/ teliendr ' die Herzahler.' " Die Partt. mit gerundiver Bedeutung, meistens pradika- tive: er per pess ekke bipianda i dir ist nicht darum zu bitten (die Bitte stent dir nicht an) ' ; peim liQfpingia, er mer se eigande vinatta vip ' einem Hauptling, mit dem es sich mir verlohnte Freundschaft zu haben.' Selten attributiv: skogar- mapr feriande ' ein fuhrbarer (einschiffbarer) Waldmann.' ' On the other hand, Professor H. Logeman, in an instructive article entitled " Det Saakaldte Passive Nutidsparticip i Norsk og i Beslsegtede Sprog," published in the Arkiv for Nordisk Filologi, xxx, 1913-1914, pp. 17-42, takes a somewhat different view of the idiom that I have been discussing. He holds that, in the Scandinavian examples of the sort that I have quoted from Nygaard, we have in reality, not a present participle used in a passive sense, but a modified form of the infinitive, the d in -nde having been, he thinks, developed from a hypothetical inflected infinitive very much as in the High German Gerund, or Present Participle of Necessity; and that, in this supposed phonetic development in the Scandinavian languages, there is no sure ground for assuming outside influence either from the Latin or from the German. For this assumed native origin of the so-called present participle in a passive sense, in the Scan- dinavian languages, Professor Logeman offers interesting, but, to me, not convincing arguments. In particular, his assumption that the Scandinavian languages once had an inflected infinitive, seems to me not supported by sufficient evidence. But whether the Scandinavian -nde in a passive sense, as in the examples from Nygaard, was originally a participial form or a new for- mation from a hypothetical inflected infinitive in -enne, I must hold, contrary to Professor Logeman, that the Latin 40 " influ- 40 The reference here is to his § 422, which I have quoted in full on pp. 56-57 above. 192 The Syntax of the Lindisfarne Gospels ence was primary, and that the phonetic 4 ° a influence, if it- existed at all, was secondary. Otherwise I cannot account for the fact, vouched for by Professor ISTygaard, that in the Scan- dinavian languages the -nde form in a passive sense occurs first in, and is restricted almost solely to, translations from the Latin. This conviction as to the Latin origin of the Old Norse present participle in a passive sense is strengthened by a consideration of the kindred idiom of the Swedish language discussed in my next paragraph. • Another Scandinavian analogue to the Northumbrian Ellip- tical Accusative-with-Infinitive Construction is found in the Middle Swedish use of a present participle h* a passive sense, which Professor Adolf Noreen tells us was due to Latin influ- ence. Of this idiom Professor Noreen speaks as follows in his Geschichte der Nordischen Sprachen Besonders in Altnor- discher Zeit, 3d ed., Strassburg, 1913, § 274: u Ein Participium Futuri activi und passivi wird bisweilen im Altschwedischen seit um 1350 (bes. aber mschw.) zur Wiedergabe der latei- nischen Bildungen auf -urus und -ndus geschaffen und /.war durch Zusammensetzung des Infinitivs mit dem Part. Pras. skolande i werdend,' ' sollend,' z. B. homaskolande ' venturus,' dyrhaskoJande ' venerandus.' Wahrscheinlich sind diese For- men, die offenbar dem Lateinisehen nachgebildet sind und bald wieder schwinden, nie in die gesprochene Sprache einge- drungen." ■""Concerning the genesis of the -ende form preceded by to in Old English. Dr. Logeman makes a statement in the article under discussion, pp. 28-29, and in his " Introduction " to The Rule of St. Benet (London, 1889). SS 89- 92. He holds that this Old English form, by some called the Participle of Necessity, was due primarily to phonetic influence and only secondarily, if at all, to Latin influence. Accordingly he disapproves of the statement of Professor Eduard Sievers (in his Angelsachaiache Grammatik. 3d ed.. Halle, 1898, § 350, Anm.), who considers that the Old English development in question is due primarily to Latin influence. Neither of these scholars refers to what I have denominated the Elliptical Accusative-with-Infinitive Construction, but Professor Sievers does state that occasionally in the LvnMsfarne Gospels a Latin gerundive is rendered by a pure present parti- ciple instead of the participle of necessity, which latter is made up of to plus the form in -nde, as to donde = faoiendus. The Infinitive 193 In High German, also, we find a construction somewhat akin to the Northumbrian Elliptical Accusative-with-Infinitive Con- struction. In the New High German Gerund, or, as some prefer to call it, the Participle of Necessity, we have, as is well known, a present participle that is preceded by zu, that is used attributively only, and that is passive in sense, as in ein nicht zu ubersehender Umstand, i a circumstance not to be overlooked.' In his Beitrdge zur Geschichte des Gerundivs im Deutschen, a Heidelberg dissertation of 1909, Dr. Victor Eck- ert has summarized the several divergent views held by the grammarians as to the origin of the Gerund in German. De- spite Dr. Eckert's opinion to the contrary, it seems to me that the theory stated by the late Professor W. Wilmanns is most likely to prove the correct one. This theory is perhaps most succinctly stated in his Deutsche Grammatik ( Strassburg., 1896), ii, §385, which runs as follows: "Nicht selten ver- binden wir den Infinitiv mit zu mit dem Verbum Substanti- vum, 41 z. B. ' der Schritt war nicht zu vermeiden. Besserung ist kaum zu hoffen. Die Handlung ist zu loben.' Im Go- tischen begegnet diese Construction nicht, haufig im Ahd. Sie entspricht wesentlich dem lat. Gerundium und breitet sich wohl auch unter dem Einfluss des Gerundiums aus; z. B. ze hara- wenne sint praeparanda sunt ; ze kesezzenne ist constituenda est ; sindun zi chilaubanne credendi sunt. Seit dem 12. Jahr. fangt dieser Infinitiv an, die Endung -ende anzunehmen, zuerst im Md., dann auch im Alemannischen (Whd. § 372. 400) und in dieser mit dem Participium Priis. iibereinstimmenden Form ist im Nhd. der Infinitiv aus seiner pradicativen Stellung in attributiven Gebrauch iibergegangen, so dass er zu einem Par- ticipium necessitatis geworden ist, das jedoch auf den attribu- tiven Gebrauch beschra'nkt bleibt., z. B. ' die noch zu beset- zenden Stellen, ein nicht zu unterschatzender Gegner, eine kaum zu erwartende Wendung, hochzuverehrender Herr,' aber ' die 41 How the Infinitive with zu came to denote necessity is discussed by Wilmanns in the third volume of his Deutsche Grammatik (Strassburg, 1906), § 70, 6, and more at length by the present writer in The Infinitive in Anglo-Saxon, pp. 200-203, 239-240. 13 194 The Syntax of the Lindisfarne Gospels Stellen sind noch zu besetzen, der Gegner ist nicht zu unter- schatzen ' etc. Gescbmackvolle Darstellung geht diesen Formen noch gern aus dem Wege; oft stehen statt ihrer Verbaladjectiva auf -lich zu Gebote. Gr. 4, 66. 113. Heyse 1, 692 1" Wil- manns returns to the question in the third volume of his Deutsche Grammatik (Strassburg, 1906), § 31, 2, which sec- tion, though in part identical with what I have just quoted, deserves quotation here because of some additional information conveyed therein : " Das Part. Pras. fallt durch Assimilation des -nd und Apokope des auslautenden e {-ende > ene, > -en) vielfach mit dem Infinitiv zusammen 42 ). Auch diese Entar- tung begegnet zuerst im Ndd. und Md., (einmal schon im Arn- steiner Marienleich MSD. 38, 236), wurde dann aber in manehen Verbindungen, in denen man das Part, als selbstan- diges Satzglied nicht mehr fiihlte. von der Schriftsprache ange- nommen (§ 61). " Umgekehrt nahm der Infinitiv mit zu die Endung des Par- tizipinms an und ging als ein Part, necessitatis in attributiven Gebrauch iiber {ein nicht zu ubersehender Umstand u. dgl.). Die Beruhrung mit dem lat. Gerundium, das schon in der Bene- diktiner-Regel einmal durch diese Form wiedergegeben wird (Hattemer 98, 2 nuntianda = ze chundande), wird die Ent- wickehmg unterstiitzt haben. S. n, § 385. Erdmann § 137." This explanation of the Gerund offered by Wilmanns is sub- stantially concurred in by Oskar Erdmann, in his Grundziige der Deutschen Syntax (Stuttgart, 1886), i, § 137 ; by Friedrieh Blatz, in his Neuhochdeutsche Grammatik, 3rd ed. (Karlsruhe, 1896), ii, § 160, a. 2 ; and by Professor Otto Behaghel, 43 in his Geschichte der Deutschen Spra\che, 3rd ed. (in Paul's Grundriss der Germanischen Philologie, Strassburg, 1911), p. 184. But, whatever may be the final decision as to the precise origin of the High German Gerund, the Gerund certainly has ""Mit besonderem Fleiss, aber nicht iiberall mit richtijjer Beurteilunir hat Bech diese Erscheinung verfolgt. ZfdW. 1, 81." 43 Professor Behaghel, I. c, p. 184, gives substantially the same explana- tion of the confusion of the forms of the present participle and of the infinitive in German as does Wilmanns, but of the Gerund itself he does not speak specifically. The Infinitive 195 several points of striking similarity with the Northumbrian Elliptical Accusative-with-Infinitive Construction: a form of the present participle has a passive 4 * signification in High German and in Northumbrian; confusion of participle and of inflected infinitive forms (however brought about) occurs in High German and in Northumbrian ; and the inflected infinitive of necessity in Northumbrian and the prepositional infinitive of necessity in High German, from which latter probably was de- veloped the Participle of Necessity, or the Gerund, in High German, were each probably suggested by the presence of the gerund in the Latin originals. Of interest in this connection is Dr. J. A. H. Murray's state- ment concerning the passive of action in Scotch, in his The Dia- lect of the Southern Counties of Scotland, London, 1873, p. 225 : " To express the Passive of action, equal to the Latin cpclificatur, cedificabatur , cedificabitur , the Scotch uses the form the hoose is buildan.' This is not a contraction of the Old Eng. a-building, as the form is not the gerund but the participle, and represents the middle voice buildan' itseV , and thus bevn.g built. But as this form, being identical with the Active voice, would often cause ambiguity, it is usual in Scotch, as in French, to make such sentences active, with the indefinite Nominative tliay, pronounced (dh), Fr. on. Thus, 'many houses are at present being built here,' would be rendered ' The're buildan' monie hooses heir the-nuw.' " To consider the Modern English is doing in the sense of is being done would take us too far afield. An interesting dis- cussion of the divergent views concerning the origin of this idiom, however, is given by Dr. Alfred Akerlund, " A Word on the Passive Definite Tenses," in Englische Studien, xlvii, 1914, pp. 321-337. 44 To speak of the alleged instances of a Germanic present participle in a passive sense other than in the gerundial combination would carry us too far afield. On this point see Grimm, I. c, iv, p. 68; Wilmanns, I. c, in, §58; Erdmann, Grundziige der Deutschen Syntax, I, § 132; and Blatz, V enhochdeutsche Grammatik, II, § 161, a. 4. 196 The Syntax of the Lindisfarne Gospels PASSAGES IN WHICH AN INFINITIVE IS USED UNALLOWABLY IN THE " LINDISFARNE GOSPELS " I gather together here several passages in which an Infinitive Is Used Unallowably in the Lindisfarne Gospels: — 1. The Uninflected Infinitive Mk. 3.2 : (5a3tte hia geteldon uel nidria hine = nt accusarent ilium (Rushw. : (5a3t hise teldun uel nifiradun hine ; W. S. : Sset hi hine gewregdun). [Here, clearly, the glossator merely names by the infinitive form a word synonymous with geteldon without going further and putting the word in the proper form of the finite verb.] J. I. 1.3 : (5one of hsendum uel of brydlopum (5a3t lust uifes gemana eftgeceigde god = quern de nuptis uoluntem (sic for uolentemf) nub ere reuocauit deus. [The glossator seems to misunderstand uoluntem for uoluntatem instead of uolentem, and leaves gemana without a governing verb.] 2. The Inflected Infinitive Mat. I. 20. 5 a : alle willo to fylgenna hine lsereS onsaca hine seolfne = omnem uoluntatem (sic for uolentem?) sequi se do- cet abnegare se ipsum. [The error arose as in the preceding example.] MJc. I. 2.2 : for(5on alexandriniscse biscob was Sses uel his (5erh ana uel syndrigo woerc wiste uel to uutanne godspelles in him gecuoedna tosceada & (5one (5eodscip in him ass (5sette oncnea- wa uel were oncnauen (5set godeund = nam alexandriae episco- pus fuit cuius per singula opus scire euangelii in se dicta dis- ponere et disciplinam in se legis agnosceret. [Or is the infini- tive here, in Northumbrian and in Latin, a predicate nomina- tive ?] I close this chapter with Synoptic Tables showing the Uses of the Infinitive in the Lindisfarne Gospels and Some Substi- tutes for the Infinitive therein. SYNOPTIC TABLE OF THE USES OF THE INFINITIVE IN THE "LINDISFARNE GOSPELS" The Active Infinitive with Active Verbs Use: Totals U. I. Matthew U. I. Mark U. I. Luke U. I. John U. I. Subjective: Total. Gosp. Int. 29 23 2G 18 3 5 12 8 10 4 2 4 11 4 11 4 6 11 5 10 1 1 .. Objective : Total. Gosp. Int. 157 37 129 32 28 5 45 2 29 2 16 45 3 41 2 4 1 54 21 48 19 6 2 13 11 11 9 2 2 Other Substantival Uses : (1) As a Pred.-Nom. — Total. Gosp. Int. (2) As an Appositive — Total. Gosp. Int. (3) As the Obj. of a Prep. — Total. Gosp. Int. 1 1 1 1 5 1 5 1 1 4 i i 1 '.'. i 2 2 1 1 1 1 1 1 2 2 1 3 i 3 1 1 Predicative with Aux. Verb : Total. Gosp. Int. Not counted 7 6 1 5 4 1 2 2 Predic. with Wutun: Total. Gosp. Int. 7 7 2 3 3 1 1 1 1 Predicative with Bia(n) ( Wosa) : (1) Of Necessity — Total. Gosp. Int. (2) Of Purpose — Total. Gosp. Int. 21 4 17 2 1 1 9 9 2 2 4 1 3 2 1 1 6 3 3 Predicative with Acc. Subj. : (1) As Object— Total. Gosp. Int. (2) As Subject — Total. Gosp. Int. 65 41 38 4 27 37 31 1 28 3 1 24 17 14 2 10 15 14 12 2 10 6 6 2 4 4 5 5 22 14 14 S 14 y 1 9 1 9 4 4 5 4 3 2 1 Final : Total. Gosp. Int. 54 82 46 77 8 5 20 30 20 29 1 5 16 6 14 2 2 15 29 13 28 2 1 11 7 9 6 2 1 Infin. with Adject. : Total. Gosp. Int. 7 8 7 8 3 1 3 1 2 1 2 1 2 6 2 6 Other Adverbial Uses: (1) Of Cause — Total. Gosp. Int. (2) Of Result— Total. Gosp. Int. (3) Of Means — Total. Gosp. Int. 1 1 2 2 i 1 1 i ; 1 1 ! 2 2 Infin. with Nouns: Total. Gosp. Int. 7 17 6 10 1 7 3 1 2 2 2 ?, 6 2 3 1 3 4 6 4 4 2 Imperative: Total. Gosp. Int. 7 6 1 2 2 4 3 1 1 1 197 SYNOPTIC TABLE OF THE USES OF THE INFINITIVE IN THE " LINHISFARNE GOSPELS" (Continued) The Active Infinitive with Passive Verbs Use: Totals U. I. Matthew U. I. Mark U. I. Luke U. I. JOHN U. I. Subjective : Total. Gosp. . Int. 12 10 12 10 2 3 2 3 4 3 4 3 4 3 4 3 2 1 2 1 Objective: Total. Gosp. Int. 4 2 1 4 1 2 1 1 2 1 1 1 1 .'. 1 1 Predicative with Acc. Subj. : (1) As Object— Total. Gosp. Int. (2) As Subject— Total. Gosp. Int. 5 1 2 1 S i l 5 .'. 2 3 '.'. Fhial : Total. Gosp. Int. 1 1 .. 1 1 . . :: :: The Passive Infinitive with Active Verbs Predicative with Aux. Verbs: Total. Gosp. Int. 2 2 2 2 . . " The Passive Infinitive with Passive Vfbbs No example has been found. SOME SUBSTITUTES FOR THE INFINITIVE Predicate Nominate i i Present Ptc. : Total. Gosp. Int. Predicate Accusative op Present Ptc. Total. 96 27 26 19 34 Gosp. 94 25 26 19 24 Int. 2 2 •• Elliptical Accusative- witm-Infinitive Construction : (a) Acc anJE THE LIBRARY -3 / UNIVERSITY OF CALIFORNIA t> C3 Santa Barbara STACK COLLECTION THIS BOOK IS DUE ON THE LAST DATE STAMPED BELOW.