Early Syriac Translation Technique and the Textual Criticism of the Greek Gospels 9781463235833

Citation preview

Early Syriac Translation Technique and the Textual Criticism of the Greek Gospels

Texts and Studies

2 Series Editor Hugh Houghton

Editorial Board Jeff W. Childers Christina M. Kreinecker Alison G. Salvesen Peter John Williams

Text and Studies is a series of monographs devoted to the study of Biblical and Patristic texts. Maintaining the highest scholarly standards, the series includes critical editions, studies of primary sources, and analyses of textual traditions.

Early Syriac Translation Technique and the Textual Criticism of the Greek Gospels

P. J. Williams

9

34 2004

CONTENTS

Contents ................................ ................................................................ ......................................... ................................ ......... vii Preface................................ ................................................................ ............................................. Preface ................................ ................................ ............. xi Abbreviations and Conventions ................................ ..................................... ..... xv 1

.............................................................. Introduction ................................ .............................. 1 1.1 The Syriac Texts Considered ................................8 1.2 The Use of Syriac Witnesses since 1894..............14

2

Proper Nouns, Common Nouns, and Pronouns ..... 23 2.1 Addition and Omission of Personal Proper Nouns ..................................................................23 2.1.1 The Name ‘Jesus’.......................................23 2.1.2 Other Personal Names ..............................38 2.1.3 Conclusions................................................46 2.2 Addition and Omission of Verbal Complements ......................................................47 2.2.1 Direct Objects ............................................47 2.2.1.1 Hear / Say + ôá™ôá................................ 48 2.2.1.2 The Greek Sequence Verb–Object–Verb, etc. ............................ 52 2.2.1.3 Textual Implications ............................ 57 2.2.1.4 Miscellaneous Objects .......................... 59

2.2.2 Indirect Objects .........................................60 vii

viii

CONTENTS

2.2.2.1 Forgiving People Their Sins ................ 60 2.2.2.2 Giving................................................... 63 2.2.2.3 Preparing ............................................. 66

2.3 Possessive Pronouns ............................................67 2.3.1 Inherent Possession ...................................69 2.3.1.1 Hands ................................................... 73 2.3.1.2 Other Inherent Possessions.................. 81

2.3.2 Relational Terms .......................................87 2.3.2.1 Father ................................................... 89 2.3.2.2 Other Relational Terms ....................... 95 2.3.2.3 Generic Paired Relationships ............. 101 2.3.2.4 Conclusion.......................................... 102

2.3.3 ïj ìáèçôáß vs. ïj ìáèçôár ášôï™ ...................103 2.3.3.1 Related Phrases .................................. 115

2.3.4 Other Variations in the Use of Possessives................................................119 2.4 Independent Pronouns .....................................122 2.4.1 The Representation of ášôüò ...................122 ............................126 2.4.2 The Use of ðOò and 2.4.2.1 The Addition of ‘All’ .......................... 126 2.4.2.2 The Omission of ‘All’ ......................... 129 2.4.2.3 Other Features ................................... 131

3

Articles and Particles ................................ ............................................. ............. 133 3.1 The Syriac Indefinite Article ........................133 3.2 The Greek Definite Article................................141 3.3 Prepositions .......................................................143 .........................................................143 3.3.1 3.3.2 Greek Case Distinctions in Syriac............146 3.3.3 Repeated Prepositions.............................147 3.3.4 Semiprepositions .....................................147

CONTENTS

ix

3.4 The Differences between êáß and waw ..............149 3.4.1 Verbal Asyndeton.....................................150 3.4.2 ‘And’ Necessary during Translation of a Greek Participle ................................154 3.4.3 êáß as ‘Also’...............................................156 3.4.4 Lists ..........................................................157 3.4.5 Miscellaneous...........................................158 3.5 Adverbs ..............................................................160 3.5.1 dêåsèåí ......................................................161 3.5.2 hîù............................................................162 3.5.3 hôé .............................................................163 3.5.4 }äç ............................................................165 3.5.5 ìüíïí.........................................................168 3.5.6 í™í.............................................................169 3.5.7 ôüôå ...........................................................170 4

Transgressing Formal Boundaries ........................ 175 4.1 Alteration in Number ........................................175 4.1.1 Number in Nouns ....................................176 4.1.2 Number in Verbs .....................................178 4.1.2.1 3 sg. and 3 pl...................................... 178 4.1.2.2 2 sg. and 2 pl...................................... 181 4.1.2.3 1 sg. and 1 pl...................................... 182

4.2 Interchange of Active and Passive.....................183 4.2.1 Greek Passive with ›ðü of the Agent ........184 4.2.2 The Passive of öÝñù and Tãù....................189 4.2.3 Greek Active Translated by Syriac Passive ......................................................191 4.3 Alteration in Grammatical Person ....................193 4.3.1 Third Person and Second Person............193 4.3.2 Third Person and First Person ................195

x

CONTENTS

4.3.3 First Person and Second Person..............196 4.4 Divergence in Use of Tenses.............................198 5

Word Order ................................ ........................................................... ........................... 203 5.1 The Reversal of Paired Items............................204 5.2 Name Preceding Function ................................235 5.3 Other Questions of Word Order.......................245

6

Words .................................................. Words for Speech ................................ .................. 249 6.1 Semantic Despecification of Verbs of Speech ...249 6.2 ‘Asking’ ..............................................................252 6.3 ‘Answering and Saying’ .....................................255

7

Miscellanea ................................ ............................................................ ............................ 263 7.1 The Translation of PêïëïõèÝù...........................263 7.2 Mistakes, Ancient and Modern .........................272 7.3 Other Variants ...................................................275

8

Conclusions ................................ ........................................................... ........................... 285

Appendix 1: Brief Brief Rules for the Use of Syriac in NT Textual Criticism ................................ ............................................... ............... 293 Appendix 2: Suggested Emendations to the Apparatus of NA27................................ NA27................................ ....................................................... ....................... 297 Appendix 3: Agreements Agreements between Syriac Witnesses and Codex Codex Bezae for Which a NonNon-genetic Explanation Is Possible ................................................ ................................................ 307 Bibliography ................................................................ .................................. 311 Bibliography................................ bliography ................................ ................................ Indices ................................ ................................................................ .......................................... ................................ .......... 321

PREFACE It may be of interest to readers to have a description of the process by which I arrived at the basic argument of this book, a work that questions common citation of Syriac witnesses for textual variants in the Gospels, and that also formulates new rules for their use. It all began in the Faculty of Oriental Studies, Cambridge University, with my doctoral work on Syriac syntax applied to the Old Testament Peshitta, specifically to the book of 1 Kings. At a fairly advanced stage in my research for this I was invited to present a paper to the University’s Old Testament Seminar. The seminar was comprised of senior Old Testament scholars and a number of postgraduate students, and, although some knew Syriac, knowledge of Syriac could not be assumed. I had therefore to devise a subject related to my research, yet accessible to those who did not know the language I was researching. The most natural way of doing this was to look at the Syriac Old Testament by way of what is probably the public face of Syriac to those who do not know the language, namely the text-critical apparatus of the Biblia Hebraica Stuttgartensia (BHS). The eventual title of the seminar was thus ‘The Peshitta as a Text-critical Witness to 1 Kings’, read on 13 November 1996. The key surprise both to me and then to my hearers was just how unreliable and misleading many of the notes in BHS were. I later came to realise that this fact was widely appreciated amongst xi

xii textual scholars, even if they did not come to quite such negative conclusions as I did. In particular it seemed that those who had realised the unreliability of the textual apparatus of BHS had contented themselves with observing that fact in general, but had not gone on to point out that there were actually systematic errors due to failure to observe syntactic tendencies. The first account of such a systematic error I published in 1999.1 In the same year I was invited to give another paper, this time to the Seminar for the Study of Judaism in Late Antiquity, Durham University. Because of the constitution of this seminar I thought it most appropriate to extend my examination of critical apparatuses to include those of the Greek New Testament as well as of the Hebrew Old Testament.2 Preparation for this paper was what brought me to the realisation that the problem I was aware of in the representation of Syriac in BHS also extended to editions of the Greek New Testament. Thereafter I began groundwork for a thorough investigation of the systematic divergencies between the Greek and Syriac texts of the Gospels due to syntactic causes. With time it became possible to see wider textual implications of this course of research, in particular for identifying possible pseudoagreements between textual witnesses. There are many whom I have to thank for help with this monograph. Throughout the time of research I have held the position of Research Fellow in Hebrew and 1

‘“According to All” in MT and the Peshitta’, Zeitschrift für Althebraistik 12 (1999) 107–109. 2 The paper was ‘Retroversion from Ancient Versions in Critical Editions of the Bible’, read on 18 November, 1999.

PREFACE

xiii

Aramaic at Tyndale House, Cambridge, a position generously supported by the Kirby Laing Foundation. During 2002 the Academic Committee of Tyndale House kindly granted me a sabbatical from the teaching and editorial responsibilities of the post and it was during that time that much of this book was written. For four months from the beginning of April 2002 I was hosted by the Institut für Neutestamentliche Textforschung in Münster and granted access to the wonderful facilities there both in literature and microfilms. For this I am most grateful to the Institut’s Director, Prof. Lic. Dr. Barbara Aland. In addition to the delights of daily conversation with the Mitarbeiter there over coffee and lunch I also benefited from the specialist scholarly input of many. I am grateful to Luc Herren for help with fonts, Dr Franz-Jürgen Schmitz for help with a matter of Coptic, Dr Michael Welte for supplying me with books I requested from the store, Dr Klaus Wachtel for interaction on many issues, and to the librarian Christine Willruth for her patience with my questions. In particular I would like to thank my host Dr Andreas Juckel for practical help and especially for our many conversations about the Syriac issues raised by this book. He showed a remarkable dedication to Wissenschaft and his dictum ‘keine Kritik ist negativ’ will, I am sure, mark the spirit in which he receives the contribution of this monograph no less than that in which it is offered. Dr Juckel’s own insights into the Syriac manuscript tradition have helped shape much of this work. Besides help from those at the Institut I have benefited from critical comments on parts of this manuscript by a number of individuals. Dr Peter Head

xiv (Cambridge) gave comments on chapters 1, 5, and 6, and Dr Ulrich Schmid (Gütersloh), the Revd David Lane (Kirkcudbrightshire, Scotland), and Dr Terry Falla (Melbourne) gave comments on chapter 5. Professor William Petersen (Pennsylvania State University) selflessly devoted over three days to ironing out problems in my argument in chapters 1, 5, and 8, and though some disagreements will doubtless remain these chapters have benefited from his comments throughout. I thank the series editors Professor David Parker and Dr David Taylor for accepting this work in Texts and Studies and for their editorial input and encouragement. I am also grateful to Dr George Kiraz and Gorgias Press for overseeing production. The Syriac text is set using the MELTHO fonts from Beth Mardutho: The Syriac Institute [www.BethMardutho.org]. My wife Kathryn and my daughter Magdalena accompanied me on the sabbatical in Münster. The trip was one of great pleasure, but also of domestic upheaval of which they bore the brunt. To them I am profoundly grateful. I am privileged to have a wife and mother who have been prepared to proofread versions of the following pages, and who can spell better than I. It is appropriate therefore that I dedicate this book to my parents, Anthony and Diana Williams, who, besides the interest they have shown in this work, can be said to have provided its initial impetus by thoughtfully teaching me about the Bible and by providing me with my first Greek New Testament while still at school. Eis gratia sed Soli Deo gloria.

ABBREVIATIONS AND CONVENTIONS A word may be said about some of the conventions used. Many of the abbreviations I use are those of the NestleAland Novum Testamentum Graece. I use the manuscript sigla and the signs from its apparatus without explanation since those who find this book useful will need none. I use txt simply to refer to the main text of the Nestle-Aland edition which is referred to as NA27 (and its predecessor as NA26, etc.). Apart from in Appendix 2, I have preferred to use S, C and P for NA27’s sys, syc and syp, respectively. Parentheses appear round these signs reflecting equivalent parentheses in the NA27 apparatus, etc. OS is the Old Syriac. Occasional other abbreviations are explained in the text. References like Mark 14:3b do not necessarily refer to the second half of a verse, but rather to the second occurrence in that verse of whichever phenomenon is presently receiving comment. In quoting OS readings occasionally sey me have been supplied where one of the two witnesses has them and it was irrelevant to the discussion to draw attention to their absence in the other witness.

xv

1I

NTRODUCTION

This work has two main aims: positively, to formulate new guidelines about the use of Syriac witnesses to attest Greek variants, and negatively, to show that early Syriac witnesses of the Gospels do not support many of the Greek variants they have been claimed to support. These aims are intimately connected. In its negative aim the monograph is almost certainly less original than in its positive one. Careful scholars have, after all, long been cautious about any piece of versional information given in a critical apparatus; it is simply a suggestion to be checked. However, the small community of scholars able to make a truly independent assessment of the validity of a Syriac entry in an apparatus has expended relatively little effort in developing and refining precise rules for the use of Syriac versions. It is hoped that the following pages can make a positive contribution in this regard. It will be argued in what follows that, whereas critical editions of the Greek New Testament confidently cite Syriac witnesses on behalf of this or that Greek variant, the Syriac evidence is often more ambiguous. Consequently, some Syriac witnesses need henceforth to disappear from the discussion of a significant number of disputed texts. The particular witnesses with regard to which this claim is made are the Old Syriac (OS) manuscripts Sinaiticus (S) 1

2

INTRODUCTION

and Curetonianus (C), and to a lesser extent the Peshitta (P). Since Syriac Bible translations become progressively more literal through the centuries,1 the uncertainties that accompany studies on the Vorlage of the Old Syriac texts diminish as one approaches the Peshitta, and are drastically reduced on reaching the Harclean text of the Gospels.2 The Harclean text is therefore left entirely out of consideration in the pages that follow. In contemplating the Vorlage of the text of any ancient Bible translation there are basically three major types of hypothesis that present themselves. The first is the hypothesis that the translation is essentially a literal representation of its Vorlage, the second, that formal alterations were made in the process of translation, and the third, that alterations were made in the process of transmission of the translated text to us. We may call these the Vorlage Hypothesis, Translation Hypothesis, and Transmission Hypothesis, respectively. In no sense are these three hypotheses mutually exclusive, and one often has recourse to more than one of them simultaneously. What is at question is the relative frequency with which one should invoke each hypothesis. It is contended here that the Vorlage Hypothesis has been used too much and the Translation Hypothesis explored too little. The Transmission Hypothesis—the possibility of change during 1 S.P. Brock, ‘Toward a History of Syriac Translation

Technique’, 1–14. 2 To such an extent that Aland and Juckel are able with relative certainty to attempt a retroversion of the Harclean into Greek. See B. Aland and A. Juckel, Das Neue Testament in syrischer Überlieferung (1986–).

INTRODUCTION

3

transmission—needs to be considered carefully, but is not frequently invoked here since all the Syriac texts studied are extant in exemplars dating from the sixth century or earlier. The relatively short period between composition and our extant manuscripts is moderately favourable to an optimistic view of our access to the original translation. However, the relationship between translation and transmission needs to be clearly defined. S and C give the impression that they have a complex textual prehistory, which, unfortunately, cannot be reconstructed with certainty. It will, however, be argued in line with the opinio communis that both C and to a lesser degree S contain texts revised by comparison with Greek manuscripts at a point or points subsequent to the original composition (which could have been a process) of the OS. Since such historically secondary readings are hard to distinguish from ‘original’ OS readings, for most purposes translation is defined as the last stage of revision according to Greek manuscripts before S and C were copied. Transmission, then, for S and C refers hereafter only to any changes made subsequent to the last comparison with Greek texts. To claim that the Vorlage Hypothesis has been used too much is in no way to denigrate its use generally. It is and should remain the most significant hypothesis. When we look at large textual units, the reason that the Syriac Gospels correspond basically in the order of their contents to the Greek ones is not related to oral transmission nor to translation technique, but simply to the fact that the Syriac Gospels were translated or revised from Greek manuscripts with this order. At the other end of the scale of size the Vorlage Hypothesis provides a convincing explanation for

4

INTRODUCTION

small textual features such as the text of P in Mark 1:4. Given the choice between dãÝíåôï EÉùÜííçò ¿ âáðôßæùí dí ô† dñÞìv and the same words without the article ¿ it seems eminently reasonable to suppose that P’s ‘John was in the desert baptising’ goes back to a Greek text without ¿. What supports this view is the presence of Greek manuscripts both containing and lacking the ¿. The Vorlage Hypothesis thus may be used to attest a single Greek letter. The Vorlage Hypothesis’ chief attraction is that if the readings of S, C, and P are translated back into Greek so many of the resultant translations actually exist within the Greek manuscript tradition. But, while it is quite probable that Syriac translations reflect Greek variants no longer extant in our manuscripts, one should avoid postulating many such cases unnecessarily. In particular one should avoid using a translation to postulate multiple lost Greek variants when a consistent relationship can be established between the translation and the extant Greek texts. This is why scholars do not posit a Greek text with êáß every time a Syriac translation has the conjunction waw. It is a consistent pattern for Syriac translations to add waw, and positing literally thousands of lost Greek readings is unnecessary. The Translation Hypothesis, on the other hand, though the one championed in this monograph, provides no plausible explanation for many cases convincingly dealt with by the Vorlage Hypothesis. Nevertheless, it will be seen that its explanatory powers, though acknowledged, have still been widely underestimated. This is because mere correspondence between a Syriac and a Greek variant does

INTRODUCTION

5

not prove that the Syriac was made from that Greek text. It might also arise from an independent concurrence of the Syriac and Greek, since not all agreement is due to genetic relationship. Agreement, especially agreement between manuscripts in different languages, can be independent. The sorts of errors to which scribes were prone are by their very nature likely to be repeated. Agreement between Syriac and Greek texts can also be independent if an adequate reason internal to Syriac or the method of a translator can be given. Such an explanation must not be given in an ad hoc way, but must be shown repeatedly to be able to explain the features of the translation. Ad hoc explanations should not, moreover, repeatedly be introduced to explain multiple agreements of different kinds. Where such agreements exist we must conclude that the majority are genetic, even if we cannot be sure of the genetic nature of individual agreements. This leads us to a prediction that can be used to test the Translation Hypothesis. If particular features of a Syriac translation are caused by the conscious or subconscious preferences of the translator or by the internal constraints of the target language, and if such features occur often enough, we should expect them to occur sometimes when no Greek texts formally correspond to the Syriac. The Translation Hypothesis becomes ever more plausible an explanation as the number of Syriac cases showing a feature unparalleled in Greek texts mounts. If Greek variants always parallel features of the Syriac translation then the Vorlage Hypothesis gains support. Contrariwise, if a feature of the translation

6

INTRODUCTION

regularly occurs without formal Greek support, the Translation Hypothesis is more probable. This test, which will be used regularly hereafter, needs some further refinement. There is a sufficient mass of Greek manuscripts that a large number of features of Gospel translations might be paralleled somewhere in the Greek tradition. But the question is not whether a conjunction in Syriac can be paralleled in some obscure Greek manuscript, but whether there is any genetic link between the two. It must at least be plausible, therefore, that the Greek variant existed at the time the Syriac translation was made. For this reason, as well as because of the practical difficulties of any other procedure, the variants in the Nestle-Aland Novum Testamentum Graece (27th edn; hereafter NA27) and the United Bible Societies’ Greek New Testament (4th edn; hereafter UBS4), along with the works of Swanson,3 von Soden,4 Legg (Matthew and Mark only),5 and the International Greek New Testament Project (Luke only),6 have been taken to contain all the significant and plausibly early variants in the Greek Gospels. Where these record no variants it has been assumed that the Greek manuscript tradition is without variant (such minor variants as exist are unlikely to affect 3 Reuben Swanson, New Testament Greek Manuscripts: Variant

Readings Arranged in Horizontal Lines against Codex Vaticanus: Matthew, Mark, Luke, John (4 vols). 4 H.F. von Soden, Die Schriften des Neuen Testaments (4 vols). 5 S.C.E. Legg, Evangelium secundum Matthaeum; idem, Evangelium secundum Marcum. 6 International Greek New Testament Project, The Gospel according to St. Luke, Part One: Chapters 1–12, Part Two: Chapters 13–24.

INTRODUCTION

7

the argument). Establishing locations without Greek variants is usually important at the outset of consideration of a particular translation feature. At these points the unity within the Greek tradition has been highlighted and any deviations within it noted. When the united Greek witness has been established, the more frequent and consistent a Syriac deviation from it is, the more plausible the Translation Hypothesis has been held to be. Though the test of repeated deviation from the united Greek witness is usually enough to establish that variation is due to translation,7 investigation has not always stopped at merely establishing that a variation is translational. Usually a reason for the variation has been sought. This has often involved leaving the safety of observable data and seeking plausible conjectures to explain the data. These conjectures vary greatly in certainty, but the rejection of the explanation given here for a translational divergence should not necessarily lead to a rejection of the appropriateness of the Translation Hypothesis itself as the right model for the origin of a set of variants. The Translation Hypothesis is, of course, a model which works best with a set of variants. That is not in any way to conclude that all variants in that set must have a translational origin. It is rather to say that wide application

7 Repeated deviation from extant Greek witnesses does not

rule out the existence of a rogue Greek manuscript, no longer extant, into which a pattern of repeated variation has been deliberately entered. Such a manuscript, besides being itself improbable, is unlikely to provide grounds for the peculiar combinations of repeated deviations Syriac texts show from Greek ones.

8

INTRODUCTION

of the Vorlage Hypothesis for the set of variants has become implausible. Explaining a single text from a particular Greek Vorlage has not usually been excluded, only made unnecessary.

1.1 THE SYRIAC TEXTS CONSIDERED S, C, and P are clearly genetically related texts, but they are also independent to a considerable extent. S and C show a large amount of common agreement against P, and hence are classified under the common title OS. However, S and C also differ radically in many places. For instance, S lacks Mark 16:9–20 and C seems to have contained these verses.8 S and C show some striking agreements with ‘Western’ witnesses, such as Codex Bezae (D) and Old Latin witnesses,9 and this forms a central puzzle in New Testament textual studies. Dates vary for the translation of the Old Syriac from the mid-second century, shortly before Tatian’s Diatessaron, to the fourth century.10 Common agreements, however, between S and the Latin manuscript k, and three-way agreements between OS, the Old Latin Gospels, and early patristic sources are sometimes thought to mark the antiquity of significant aspects of the OS text to at least the second century. In general, C shows greater conformity to Greek texts and is therefore thought to represent a later text than S. At any given point neither or both texts may retain the original stratum of their textual history. 8 They survive in C from Mark 16:17 to 16:20. 9 F.H. Chase, The Syro-Latin Text of the Gospels. 10 The Introduction to NA27 gives their date as third or

fourth century (p. 65*).

THE SYRIAC TEXTS CONSIDERED

9

In analysing the significance of an OS text one is often simultaneously making several comparisons. There is natural comparison with various Greek or Latin witnesses. There is also comparison between the two OS witnesses when both are extant. Yet there is also internal comparison within each witness. Both S and C show significant tendencies within their texts to assimilate similar but not identical phrases. Hence, in Matthew 4:16, where Greek witnesses read ‘The people that sat in darkness have seen a great light, and to those seated in the land and shadow of death a light has dawned’, the OS texts assimilate ‘great ‘a light’ in both light’ and ‘light’. S reads simply ‘a great light’ in both places. places and C This indicates the force of internal assimilation in the two texts and the readings of the manuscripts are most readily explained by the Transmission Hypothesis. Similarly, in John 3:15–16 S has twice ‘that whoever believes in him should not perish but has/should have life for ever’ and C twice has merely ‘that whoever believes in him should have life for ever’.11 Whatever their relationship to Greek Vorlagen, both OS texts are consistent within themselves, and the divergence between the two OS witnesses is most readily explained if the assimilation has taken place subsequent to translation. Assimilation is therefore an important factor to take into account when using Syriac texts for the purposes of discovering the Greek texts from which they were made, and it is considered in this

11 Another probable example is Matthew 5:23–24 where S

reads ‘on the altar’ twice and C ‘before the altar’ twice, but Greek witnesses ‘on’ in v. 23 and ‘before’ in v. 24.

10

INTRODUCTION

monograph in the places where it is most obviously relevant.12 On the whole, P seems to be a revision of OS on the basis of Greek witnesses. The textual transmission of P is more unified than that in our two witnesses of OS but it has been suggested that the standard edition of the Peshitta Gospels, that of Pusey and Gwilliam,13 may give too unified an impression of the state of the text.14 The fortytwo manuscripts on which they based their edition were not fully collated,15 and Pusey and Gwilliam showed a bias towards West Syriac manuscripts. At the moment the extent of survival of OS readings within P manuscripts is still an open question. There is also the related question of the survival of OS readings outside the Gospels, since it is widely thought that the OS covered other parts of the New Testament including Acts and the Pauline Corpus.16 In order, therefore, to expand the textual base of this study, at certain points the variants of two key manuscripts of P have also been taken into account in considering the P 12 To my knowledge, as yet no thorough study of internal

assimilation within these texts has been undertaken. Since assimilation is not taken into account in the present work, its questionings of reconstructions of a Greek Vorlage based on OS evidence will, if anything, be less radical than they should. Where assimilation impinges on positive arguments put forward it is of course fully considered. 13 P.E. Pusey and G.H. Gwilliam, Tetraeuangelium Sanctum. 14 Werner Strothmann, Das Wolfenbütteler Tetraevangelium Syriacum: Lesarten und Lesungen, 1–12. 15 Andreas Juckel, ‘A Re-examination of Codex Phillipps 1388’, par. 5. 16 A. Vööbus, Studies in the History of the Gospel Text in Syriac, II, 181–83; J. Kerschensteiner, Der altsyrische Paulustext.

THE SYRIAC TEXTS CONSIDERED

11

tradition. Though both were cited by Pusey and Gwilliam, neither was fully collated. The manuscripts are codex 3. 1. 300 Aug. fol. of the Herzog August Bibliothek, Wolfenbüttel (Pusey and Gwilliam’s ms 39, hereafter Pwolf), which comes from the end of the sixth century, and codex Phillipps 1388 of the Staatsbibliothek in Berlin (Pusey and Gwilliam’s 41, hereafter Pphil), from around AD 500.17 Manuscript variants mentioned by Vööbus are also included when relevant.18 In general Tatian’s Diatessaron has been left out of discussion of textual variants. This procedure, though not without its disadvantages, has several justifications.19 First, to explain the text of S and C by recourse to the Diatessaron is frequently to explain notum per ignotum, to explain an extant text by means of a reconstructed one. The possibility of influence from the wording of the Peshitta on Ciasca’s Arabic text renders the Arabic problematic for a number of the variants we consider here. Even where Ephrem’s indisputable citations from the Diatessaron agree with the text of OS witnesses, the interpretation of the agreement is not transparent. To take 17 The collations of Strothmann, Das Wolfenbütteler Tetra-

evangelium Syriacum, and Juckel, ‘A Re-examination of Codex Phillipps 1388’, have been followed for Pwolf and Pphil, respectively, with additional reference to the microfilms of these manuscripts at the Institut für Neutestamentliche Textforschung. 18 Vööbus, Studies in the History of the Gospel Text in Syriac, II. 19 For an authoritative account of the complexities of Diatessaron scholarship the reader is referred to William L. Petersen’s Tatian’s Diatessaron: Its Creation, Dissemination, Significance, and History in Scholarship.

12

INTRODUCTION

the example of Matthew 1:25 C, which reads ‘he was dwelling purely with her’ as Ephrem repeatedly reads in his commentary on the Diatessaron (II, 5, 9–11), just on the assumption that the Diatessaron preceded the OS, at least two possible explanations present themselves. (1) C might preserve the original OS reading, and (2) C might preserve a secondary assimilation of OS to the Diatessaron. The question as to when C’s reading arose in OS distracts from the more important question as to how the Syriac reading was produced in the first place. Is it a literal translation of its Vorlage, a theologically motivated alteration, or just a rendering of a usual Greek reading according to social or linguistic preferences operative among Syriac speakers? Secondly, the type of explanation for variants most frequently invoked here is one based on the difference between Greek and Syriac language and idiom. Whether such differences from the Greek arose in the Syriac edition of the Diatessaron and were then adopted into the OS Gospels or arose independently in the translation of the OS Gospels makes comparatively little difference. Either way the rendering of Greek into Syriac has resulted in particular adaptations, and in neither case is there reason to postulate a Greek Vorlage. Thirdly, the Diatessaron is only relevant to discussion of translation method in the OS if it preceded the OS. Though much evidence has been produced to suggest that the Diatessaron did indeed come first,20 a whole string of earlier scholars thought that OS was prior to Tatian’s 20 Petersen, Tatian’s Diatessaron, 130–33.

THE SYRIAC TEXTS CONSIDERED

13

Gospel harmony,21 and this has occasionally been maintained in recent years.22 To have included here the presupposition of the priority of the Diatessaron into discussion, would have restricted this commendation of the Translation Hypothesis to a particular literary model, when in fact its strength is its independence from such models. Of course where information from the Diatessaron is indispensable for the proper evaluation of variants, it has been readily considered (as in chapter 5). On the other hand, the sorts of variants for which this is the case have not generally formed the focus of this study. Moreover, because this is broadly a study of cases where there are multiple examples of the same type of variation it is in the main the results of this study that have implications for isolated minutiae in Diatessaron scholarship, not isolated features of the Diatessaron which have significant implications for this study.

21 A. Vööbus, Studies in the History of the Gospel Text in Syriac, I,

167–68, lists T. Zahn, A. Hjelt, Agnes Smith Lewis, H. Gressmann, C. Brockelmann, A. Mingana, H.C. Hoskier, C.C. Torrey, and M. Black. C.R. Gregory was also open to this conclusion: ‘Ich meine, dass die alt-syrische Übersetzung nicht weit von der Zeit Tatians, vielleicht sogar vor seiner Zeit, entstanden ist’ (Textkritik des Neuen Testamentes, 493). 22 See J.P. Lyon, Syriac Gospel Translations: A Comparison of the Language and Translation Method Used in the Old Syriac, the Diatessaron, and the Peshitto, 203–206. H.S. Pelser is keen to argue against the position that the Diatessaron must have preceded the OS. See Pelser, ‘The Origin of the Ancient Syriac New Testament Texts—A Historical Study’.

14

INTRODUCTION

1.2 THE USE OF SYRIAC WITNESSES SINCE 1894 The debate about the level of certainty we can reach about the Vorlage of Syriac versions is not new. Often in history new textual discoveries are greeted by an optimistic assessment of their textual import, and this was certainly the case for the most recently discovered text considered here, namely S, first published in 1894.23 Its discoverer, Agnes Smith Lewis, favoured explaining S’s deviations from other witnesses as the touches of Jesus’ original eyewitnesses.24 Very shortly after the original publication of S, Adalbert Merx began his massive work, whose title indicates his optimistic assessment of S’s textual significance: Die Vier Kanonischen Evangelien nach ihrem ältesten bekannten Texte: Übersetzung und Erläuterung der Syrischen im Sinaikloster gefundenen Palimpsesthandschrift. The first volume consisted of a translation of S, and there followed three volumes containing detailed commentary on the text of the Gospels. These learned volumes are still an invaluable resource—by far the most detailed discussion of the OS text to date. But despite their erudition, these volumes are fundamentally flawed by the presupposition, never once established, that S represents its Vorlage literally.25 Merx hardly considers the Translation Hypothesis as able to

23 R.L. Bensly, J. Rendel Harris, and F. Crawford Burkitt, The

Four Gospels in Syriac Transcribed from the Sinaitic Palimpsest. 24 A. Smith Lewis, Light on the Four Gospels from the Sinai Palimpsest, 5–11. 25 In Lyon’s words, ‘… he has too often assumed that L [= S] is a verbatim translation’ (Syriac Gospel Translations, 13–14).

THE USE OF SYRIAC WITNESSES SINCE 1894

15

explain variants. If S differs from all Greek manuscripts then he is happy to assume that it is usually the latter that have been altered, even as late as the early third century.26 Merx is thus a maximalist in his assessment of our ability to reconstruct the OS Vorlage. Merx’s detailed treatment of individual passages is discussed very little in the following pages. This is firstly because, not having justified his presupposition of literal translation, there is nothing that someone taking a different view needs to refute. Secondly, it is because the types of variants dealt with here and in his work differ. He discusses more exegetically significant variants where the Vorlage Hypothesis is more appropriate, while the discussion here affects interpretation less but certainty of wording more. Thirdly, his method of argumentation is much more historically orientated, and the method here more linguistically orientated. Frequently Merx argues in an a priori manner from known custom to what the text must have read.27 However, alongside Merx, there was also a strand of scholarship that was more careful about the reconstruction of the Vorlage of the Syriac. Back in 1885, before the dis26 A. Merx, Matthaeus, XVII and 118. Alterations could of

course take place this late in transmission, but one would not expect a uniform Greek testimony to result. 27 Thus to settle the textual issue of the number of cups in Luke’s account of the Last Supper, Merx first wants to know what the custom was (Merx, Markus und Lukas, VII). Those who merely refer to this work on isolated readings may not note the distinctively anti-Jewish framework in which Merx’s text-critical work to uncover the ‘real Jesus’ takes place. See Merx, Markus and Lukas, 1–2.

16

INTRODUCTION

covery of S, Friedrich Baethgen, who was enough of a retroversional optimist to produce an edition of the Greek underlying C, expressed his caution as regards the work of others thus: Als allgemeine Regel gilt hierbei, dass die äusserste Vorsicht und das äusserste Misstrauen geboten ist, wenn die scheinbar von Sc [C] gebotene griechische Lesart von keinem andern Zeugen unterstützt wird. ... Jedenfalls sind in dem Apparat bei Tregelles und Tischendorf eine Menge angeblicher Varianten des Sc zu streichen.28

According to Baethgen the singular Greek readings attested by C alone begin with Matthew 3:7,29 considerably later than in the apparatus of NA27. Somewhat later, with the discovery of S still fresh, F.C. Burkitt showed remarkable judgement in his evaluation of the OS, pointing out that a number of divergences from Greek witnesses seemed to be translational.30 On the whole, though, the early twentieth century was not much occupied with the study of translation method, and of systematically investigating which variants were translational and which represented a variant Vorlage. Some steps were made in this direction in the latter decades of the twentieth century. In 1968 Clemons offered some cautions about the citation of Syriac witnesses in critical apparatuses.31 In 1977 Metzger produced a classic 28 F. Baethgen, Evangelienfragmente: Der griechische Text des

cureton’schen Syrers, 32. 29 Baethgen, Evangelienfragmente, 32. 30 Burkitt, Evangelion Da-Mepharreshe: Volume II, 38–99. 31 J.T. Clemons, ‘Some Questions on the Syriac Support for Variant Greek Readings’, 26–30.

THE USE OF SYRIAC WITNESSES SINCE 1894

17

volume on the early versions of the New Testament, including contributions by specialists on the limitations of each language in representing Greek. The relatively brief contribution of Brock laid down a number of solid principles restricting the use of Syriac to support Greek readings.32 A similar but less structured work edited five years earlier by Kurt Aland33 had treated all the major early versions and contained an extensive examination of the linguistic constraints involved in translation from Greek into Coptic,34 though, sadly, no counterpart treatment of Syriac. Nor has one been made since then. Stemming from a similar date to these two volumes were the beginnings of Falla’s extensive project of producing an analytical key to the vocabulary of the Peshitta Gospels, considering what the Greek equivalent to each Syriac word was, including an analysis of minor particles, which happen generally to be found in the parts of the alphabet his work already covers.35 Falla’s policy was that ‘only extant variant Greek readings are cited for corresponding terms. Presumed retroversions of Peshitta renderings such as we find in the critical apparatus of Hermann von Soden’s Die Schriften des Neuen Testaments are 32 S.P. Brock, ‘Limitations of Syriac in Representing Greek’. 33 K. Aland (ed.), Die alten Übersetzungen des Neuen Testaments,

die Kirchenväterzitate und Lektionare. 34 Gerd Mink, ‘Die koptischen Versionen des Neuen Testaments: die sprachlichen Probleme bei ihrer Bewertung für griechische Textgeschichte’ in Aland (ed.), Die alten Übersetzungen, 188–273. 35 Terry C. Falla, A Key to the Peshitta Gospels, Volume One: ’ laph–D lath; Volume Two: H –Y dh. Further volumes are still awaited.

18

INTRODUCTION

not included.’36 His discoveries about Syriac–Greek equivalents led him to note the following: The present author has come upon literally hundreds of citations of the Peshitta in critical editions of the Greek New Testament that can be shown to be invalid when the evidence is evaluated in the light of all such relevant data [data concerning correspondences between Syriac and Greek].37

His words are striking in that they apply to the Peshitta, which is widely acknowledged to follow its Greek Vorlage more closely than does the OS. Vorlage conclusions from the OS are likely therefore to be even more precarious. Most recently reservations about current use of Syriac to support Greek readings have been expressed by Lyon. With particular reference to the use of the OS in apparatuses he says, ‘The rule of thumb is less.’38 He formulates rules for the use of S as a witness to Greek. These rules are more ‘stringent’ than Brock’s rules.39 On the other hand, Lyon is quite positive about the witness of P to particles and possessives.40 In characterising OS he says: This translator inserts ‘Lo,’ abbreviates ‘anwered [sic] and said,’ removes redundancies of any sort, and adds the occasional helping word. He leads us to false assumptions, for he follows the Greek closely much of the time and then suddenly departs.41

36 Falla, Key, Volume One, XXXII. 37 Falla, Key, Volume One, XXXVI. 38 Lyon, Syriac Gospel Translations, 39 Lyon, Syriac Gospel Translations, 40 Lyon, Syriac Gospel Translations, 41 Lyon, Syriac Gospel Translations,

193. 194, n. 12. 196. 193.

THE USE OF SYRIAC WITNESSES SINCE 1894

19

For Lyon there seems to be a somewhat random element to the OS as a translation. He talks of the translator’s ‘mood’ as being a significant factor in shaping the product.42 It will be argued in the following chapters that many of these ‘departures’ are not random, but rather are motivated by structural aspects of the languages. Back around the same time that Merx began his work, Eberhard Nestle was also preparing his first edition of the Greek New Testament, which appeared in 1898. Based on the preceding editions of Tischendorf, Westcott and Hort, and Weymouth (from the third edition replaced by Weiss), it initially did not contain references to the Syriac versions. This is hardly surprising since at first the apparatus only listed a few variants without indicating the primary support for them. The edition gradually grew. By the ninth edition (1912) there were only five references to Syriac in the Gospels.43 This number of references to Syriac was matched already by Matthew 1:16 in the seventeenth edition (1941) and the apparatus changed little with regard to Syriac between then and the twenty-fifth edition, by now Nestle-Aland, i.e. NA25 (1963). With NA26 (1979) and NA27 (1993) the apparatus experienced further growth and substantial change. Many references to the Harclean Syriac were added, as well as to earlier Syriac authorities. Some references were also dropped44 or provided with parentheses or a question mark, qualifying

42 Lyon, Syriac Gospel Translations, 194. 43 Matthew 1:16 S, 8:5 S, Mark 10:11–12 S, John 4:52 C,

14:22 C. 44 E.g. Matthew 1:14, 1:25, 2:2, 2:6, 2:9, 3:3.

20

INTRODUCTION

the Syriac support.45 This shows a growing caution towards the use of versional evidence which characterises more recent text-critical work. In fact, for the first time in the history of the edition, both NA26 and NA27 in their Introductions express caution about the use of versions. To quote NA26 (p. 54*): In collating the versions it is necessary to be aware constantly of how they each differ from Greek in their linguistic structures. Correspondingly the witness of a version should be accepted in a critical apparatus only when it offers irrefutable evidence for a difference in its underlying Greek text. Recognizing such instances and eliminating all variants that are merely intraversional requires a degree of specialization and long familiarity with the versions that is rare among New Testament scholars.

NA27 introduces a new principle (pp. 63*–64*): The versions are cited only where their underlying Greek text can be determined with confidence. They are generally cited only where their readings are also attested by some other Greek or independent versional evidence. Only in rare instances do they appear as the sole support for a Greek reading …

However, despite these principles, and though NA27 is a considerable way from the maximalist position of Merx, the sheer growth in the apparatus of the edition means that it makes many positive claims about Vorlage that Merx never made. In most ways the growth in the apparatus has been positive. It is precisely its comprehensive inclusion of information that has made the Nestle-Aland Novum Testamentum Graece completely without rival as a manual edition. Yet its unique position of influence also means 45 E.g. Matthew 1:22, 3:17, 4:1.

THE USE OF SYRIAC WITNESSES SINCE 1894

21

that it, more than anything else, represents the Syriac witnesses to the wider scholarly community. There are far more non-specialists who refer to the Syriac here than there are specialists who work on the original texts. It is for this reason that NA27 and, to some extent, its partner edition the fourth edition of the United Bible Societies’ The Greek New Testament (UBS4) have been chosen as the chief interlocutors for this work.46 Since NA27 is more cautious than earlier critical editions which have included OS readings (Tischendorf,47 von Soden,48 Vogels,49 Merk50) then a fortiori questions raised about current procedures will apply to these earlier editions. NA27 contrasts very favourably with these editions in displaying consciousness of the problems involved in the use of versional material. Though attitudes towards the use of OS and P as witnesses to a Greek Vorlage are certainly much more cautious than a century ago, there is still a considerable range of opinion. This is represented inter alios by NA27 on the more maximalist side and by Lyon and Falla on the more minimalist side. This work will not only argue that in 46 The variants in K. Aland’s Synopsis Quattuor Evangeliorum

have not been excluded for any reason of principle, but in order not to augment the size of the task. 47 Constantinus Tischendorf, Novum Testamentum Graece: Editio Octava Critica Maior. Of course, this edition had no access to S. 48 Von Soden, Die Schriften des Neuen Testaments. 49 H.J. Vogels, Novum Testamentum Graece et Latine (3rd edn). Vogels outdoes other editors in his zeal to retrovert Syriac readings into Greek. 50 A. Merk, Novum Testamentum: Graece et Latine (8th edn).

22

INTRODUCTION

general we have not been minimalist enough, but also seek to formalise the boundaries for the use of the earliest Syriac versions and to some extent to fill the lacuna of works systematically considering the Vorlage of the earliest Syriac translations of the Gospels.

2 PP

ROPER NOUNS, COMMON NOUNS, AND RONOUNS

2.1 ADDITION AND OMISSION OF PERSONAL PROPER NOUNS OUNS 2.1.1 The Name ‘Jesus’ The OS and P show notable features vis-à-vis Greek texts in the representation of titles of Jesus. First of all there is the tendency for Syriac witnesses to add a suffix in trans‘my Lord’ lating Greek êýñéïò. Whether it is translated ‘our Lord’ depends, particularly in OS, on the or number of people concerned in the address to or of Jesus. A second peculiarity is the presence, particularly in S, where Greek texts have arthrous or anarthrous of EÉçóï™ò. This is not ubiquitous, but happens in distinct blocks (e.g. Matthew 8:3–11:7, but not 8:26; John 1:29– 6:5),1 giving the impression that the prehistory of S’s text is complex. 1 These definitions are minimal, extending from the first to

the last reference to . The two references to ‘Jesus’ in John 1:45 and 5:15 are necessary because of the content. If the same argument were applied to the occurrence of ‘Jesus’ in John 1:17, then the entire beginning of John’s Gospel could be designated a m ran-section. The end of the section may be 6:10 because there a reference to ‘Jesus’ is omitted, and omissions of the name

23

24

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

Though there remains much to be discovered in and in the investigating the use of the terms Syriac Gospels, these titles have already received some scholarly attention,2 and so call for little comment here. Editors of modern critical editions are aware that they cannot reconstruct a possessive pronoun in the Vorlage of a . Syriac text based on the fact that it has A less obvious peculiarity, and therefore one more likely to lead editors astray, is the use of the name ‘Jesus’.3 It is clear from even a superficial analysis of the Gospels that Syriac texts, particularly P, introduced the where there was no EÉçóï™ò (arthrous or word anarthrous) in their Vorlage. This appears first statistically in the following table showing the number of times that the name ‘Jesus’ appears in P, in NA27’s txt, and in K. Aland’s Vollständige Konkordanz. Matthew Mark Luke John Total

Peshitta 187 116 175 268 [+ 4] 746

NA27 150 [+ 2] 81 [+ 1] 88 240 [+ 4] 559

Konkordanz 174 [+ 2] 94 [+ 1] 100 252 [+ 5] 620

‘Jesus’ are characteristic of the m ran-sections. The lacunose nature of S in John chapters 1, 2, 4, and 5 precludes certainty about the boundaries of this particular m ran-section. 2 A.-G. Martin, ‘La traduction de Êýñéïò en Syriaque’. Lyon, Syriac Gospel Translations, 156, says, ‘In L [S] is found 93 times, 28 times in C, 49 times in the Peshitto Gospels, and 50 times in the Madrid Polyglot citations. is freely substituted for êýñéïò (even without the possessive ½ì§í as in Mt 25:11, 37, 44) or for EÉçóï™ò …’ 3 For another analysis of the Greek behind in P see Falla, Key, Volume Two, 143–46.

PERSONAL PROPER NOUNS

25

The numbers in brackets are occurrences of the name ‘Jesus’ in the name ‘Jesus Barabbas’ in Matthew, in the ‘shorter ending’ of Mark, and in the Pericope Adulterae in John, respectively. These are not relevant to discussion of the translation profile of P. From the table it is clear that P has about 1.335 times more occurrences of the name ‘Jesus’ than txt. That most of this deviation from txt is not to be explained on the basis of Vorlage is evident for several reasons. (1) The occurrences of the name in P considerably exceed all the occurrences in the Vollständige Konkordanz together. Since this contains all the occurrences of the name in a number of editions, including the more Byzantine Textus Receptus, and yet none of the editions shows all the occurrences listed in the concordance, it is clear that P is distinguished from a wide variety of Greek texts. (2) The striking difference between P and txt in Luke, with txt having only about half of the number of occurrences of the name that P has, is scarcely attributable to Vorlage. (3) Broadly speaking the number of occurrences of the name in each gospel in P is a function of two factors: the number of occurrences in Greek witnesses, and the length of the gospel. For example, in Luke, Greek witnesses have a small number of occurrences of the name ‘Jesus’ relative to the length of the gospel and there are many additional examples of the name in P. On the other hand, in John, the concentration of the name is relatively high in Greek witnesses and there are relatively fewer additional occurrences in P.

26

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

(4) The addition of the name ‘Jesus’ in P has a ready explanation in the desire to make the translation unambiguous. Frequently the Greek Gospels use ‘he’ throughout a narrative, or even at a narrative beginning (e.g. Matthew 5:1), without explicitly identifying the ‘he’ as Jesus. This may have been felt to be less natural in Syriac. Particular Syriac additions of the name ‘Jesus’ may be explained as attempts to remove ambiguity: Matthew 4:11 SC (so that angels are not ministering to the Tempter) and Mark 5:24 P (to indicate that Jesus went with Jairus, not the reverse). (5) The contrast between P and Greek witnesses is yet stronger than the table suggests because there is not even a strict correlation between the presence of the name ‘Jesus’ in the Greek and that name in P. P may also omit occurrences of the name in contrast to the Greek. (6) As is shown in more detail below, Syriac witnesses often add or omit the name ‘Jesus’ in contrast to all the Greek manuscripts represented in our control editions. Thus, while some of the additional occurrences of the name ‘Jesus’ in P might be accounted for by the presence of Greek variants, such variants are not able to provide the wide-reaching explanation needed. (7) That the Syriac did not literally follow its Vorlage is also shown by places where the name ‘Jesus’ has been ‘moved’ in Syriac relative to the equivalent Greek clause showing the name. in contrast to all the In 82 places S, C, or P have Greek readings contained in our control editions. 62 cases occur in P, and 35 in one or both OS witnesses, revealing a partial but not substantial intersection between the oc-

PERSONAL PROPER NOUNS

27

currences in P and in OS. Sometimes the use of the proper name appears to correlate to ášôüò in the Greek,4 but there is often no feature in the Greek to which the name can be said to correspond. The 82 instances are: Matthew 4:11 SC, 4:21 C, 8:24 P, 8:26 SP, 8:28 P, 9:32 P, 13:54 C, 14:18 P, 15:3 P, 16:23 C, 21:7 CP, 21:12b C, 21:23 P, 26:7 SP, 27:3 P, Mark 1:45 P, 2:23 P, 3:1 P, 4:11 P, 4:33 P, 4:38 P, 5:24 P, 6:56 S, 7:6 S, 7:24 P, 8:29 P, 8:34 P, 9:28 P, 9:35 P, 10:46 P, 11:7b P, 13:3 P, 14:3b P, Luke 2:44 S, 4:17 P, 4:23 P, 5:12a P, 5:13 P, 5:14 S, 6:1 P, 6:17 P,5 7:20 P, 7:50 P, 8:19 C, 8:42 CP, 9:14 P, 9:16 P, 11:5 SC, 11:17 P, 11:28 SC, 12:1 P, 13:18 P, 13:20 P, 13:24 SCP, 15:3 P,6 15:11 P,7 16:1 C, 17:1 CP, 17:11 P, 18:39a SCP, 19:4 SCP,8 19:28 P, 21:5 P, 22:38 S,9 22:70 P, 23:9 SCP, 23:49 SCP, 24:25 SCP, 24:38 SCP,10 John 2:23 P, 4:43 CP, 7:3 SP, 7:5 P, 9:35b S, 11:45 S,11 12:9a P, 12:42 S, 13:26b SP, 18:24 P, 21:16 S,12 21:17 S (2×). 4 Examples are Matthew 8:24 P, Mark 4:38 P, 8:29 P, Luke

5:14 S, 11:17 P, 11:28 SC, 23:9 SCP. On ášôüò see pp. 122–25. 5 Ms 1071 adds the name ‘Jesus’, but this occurs in the following clause. 6 Èc has ¿ êýñéïò. 7 Ms 118 has ¿ êýñéïò here. 8 The addition in P is placed quite differently from that in SC. 9 Immediately after the addition of ‘Jesus’ S omits the vocative address, which is êýñéå in the Greek and in CP. 10 Ms 28 has ¿ êýñéïò. 11 S has ‘Jesus’ twice whereas Greek witnesses have the name once, or not at all. However, the position of neither occurrence in S corresponds to the position of the name in Greek. 12 D has ¿ êýñéïò.

28

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

These can be analysed as 15, 18, 36, and 13 additions respectively in Matthew, Mark, Luke, and John, or giving the figures just for P, 10, 16, 29, and 7. Broadly speaking these rates of addition in contrast to our control editions correlate to the general rates of addition relative to NA27’s txt, and therefore are in an inverse proportion to the frequency of occurrences of the name ‘Jesus’ in each of the four Gospels. Such a distribution fits well with the view that the additions are largely translational. In addition to these occurrences there are several places where Pwolf contains the name ‘Jesus’ in contrast to the Greek texts of our control editions: Matthew 9:28a, Mark 11:1, Luke 6:6, John 1:51. In each case other than Luke 6:6 it is joined by other manuscripts of P. Whether these represent variants stemming from the translation or transmission of P is hard to say. Whichever is the case, these Syriac variants only emphasise the distance between extant P manuscripts and Greek witnesses in their use of the name ‘Jesus’. As well as additions where our control editions show no variants there are also additions where published Greek support for a variant or variants with the name ‘Jesus’ is weak. Often witnesses cited together below have different variants from each other, further suggesting that these occurrences of the name are independent features of the transmission process: Matthew 24:3 P [0138*vid],13 Mark 2:25 P [124 700], 7:14 P [036 64 229* 475], 7:17 P [028], 13:1 SP [713], 14:10 P [l49], Luke 8:1 P [0211], 8:23 P 13 This is how the evidence is interpreted by Legg. It seems

better, however, to see the reading êáèçìÝíïõ ôï™ EÉçóï™ as that of 0138c.

PERSONAL PROPER NOUNS

29

[579], 8:29 P [l524], 8:37b P [1215], 8:51 P [579 655 1604], 10:18 C [l854c l1231], 11:2 P [1080], 12:14 P [027 0211 1071], 13:32 P [903 2542 l1599 l1663], 17:20 P [063 544 713 1192c], 18:29 SCP [0211 230 1071], 22:14 P [157 1093], 22:34 CP [033 213 443 1071], John 7:50 P [713], 11:11 P [157]. The presence or absence of the name ‘Jesus’ is a matter in which Greek manuscripts show some degree of fluctuation. This centres around whether the name occurs in formulae introducing speech and at turning points in the narrative. Many of the Greek variants may have arisen as tradents found the writing style of the Gospels, particularly of Luke, to use the pronoun too much and the explicit identifier of the person too little.14 This means that a considerable number of the Syriac differences from txt coincide with Greek variants containing the name ‘Jesus’. This might be thought to support the Vorlage Hypothesis as a broad explanation. However, though it is impossible to be certain with regard to individual readings, it is likely that many of these agreements are coincidental. Though Jesus is the main protagonist of the Gospels, the number of places where his name could be added is still limited by subject matter. For instance, the name does not appear during unbroken discourse of Jesus, such as the Sermon on the Mount. The number of places where a Syriac translator and a Greek copyist might add the name in contrast to

14 By the term ‘pronoun’, here and hereafter, I denote

pronominal elements in verbs and suffixes. In some cases, such as the 3 m.sg. perfect of the verb, it is best to consider the pronominal marker as zero.

30

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

their Vorlage is restricted, and therefore additions in the two languages may overlap coincidentally. It is naturally not possible to attain certainty about the presence of the name ‘Jesus’ in P’s Vorlage in any given instance. However, some estimate of the probability of its presence can be made when the occurrences are viewed as a whole. If P were made from a Vorlage with the same number of occurrences as txt and we turn to a randomly in the Gospels, the probability chosen example of that it rests on a Greek [¿] EÉçóï™ò is about 0.75 (559 ÷ 746). Taking the artificially optimistic view that P was translated from a Vorlage with all the occurrences in the Vollständige Konkordanz the probability would be about 0.83 (620 ÷ 746). However, a randomly chosen example from Luke’s Gospel would only have a probability of about 0.5 and 0.57 under these two conditions respectively. At this point there seems little value in citing P in a critical apparatus for the presence of the name. Nevertheless, all these probabilities are too optimistic in the actual practice of textual criticism since examples are not selected at random. Since many of the occurrences of the name in P overlap with its presence in substantial Greek witnesses, it must be supposed that the presence of EÉçóï™ò in the Vorlage regularly led to the presence of in P in the translation. Therefore occurrences of where substantial Greek support for the presence of the name is lacking have a much slimmer probability of resting upon a Greek Vorlage. In many cases where Greek witnesses support the presence or absence of the name

PERSONAL PROPER NOUNS

31

equally,15 a very approximate estimation—for that is all we can make—suggests that if P has the name, the probability that it rests on witnesses with the name and the probability that it rests on witnesses without the name are too near parity for P to be used in a critical apparatus. However, a slight qualification to all these calculations is that the Syriac treatment of the name ‘Jesus’ is not unidirectional. There is a smaller number of instances where Syriac witnesses omit the name ‘Jesus’ in contrast to Greek readings. This is more marked in OS than in P, and most marked in S. Almost all the cases of omission in contrast to Greek readings are in the OS: Matthew 18:22 SC, 20:34 C, 26:1 S, Mark 9:4 S, Luke 10:30 S, John 4:10 S, 6:10 S, 6:43 S, 8:39 S, 8:58 S, 9:41 S, 11:40 S, 11:41 S,16 13:23a P, 13:36 S, 20:17 S. Of these 16, 11 are found in John’s Gospel—the gospel which otherwise shows the highest concentration of the name. The translation method of addition or omission of the name in this way evens out the distribution of names relative to the Vorlage. It might be suggested that lectio brevior potior supports the originality of the reading without the proper name ‘Jesus’, and hence the likelihood that OS represents a Greek Vorlage. Who, after all, would delete that name? This is probably to seek too theological an explanation for

15 Of course one is not weighing the Greek witnesses with

regard to the likelihood that they contain the original reading, but with regard to the likelihood that their reading was the one from which P was revised (or which first entered the OS and then was not revised). Such an estimation defies mathematics. 16 Though ms 579 lacks ‘Jesus’, its reading is too far removed from that of S to be relevant.

32

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

a matter that is largely linguistic. There are linguistic pressures to avoid both under- and overuse of proper nouns. When a translator has not consciously set out to represent every proper noun in his Vorlage by a proper noun in translation, substitution of pronoun for proper noun and vice versa is likely to occur. In addition to the cases above where no Greek support could be found for the omission there are also cases where published Greek support for the omission is weak: Matthew 17:19 S [655 1071 1093 1424], 21:27 CP [1574; but CP with ) then add the name later in the verse], 22:41 C [l184], 26:57 S [40 122 157], 26:64 S [1424], Mark 11:33a S [59 73 1071], 11:33b P [4 273 659], Luke 10:29 P [659], John 1:37 S [033*], 4:34 S [71], 6:24 S [013], 9:3 S [1093], 13:26b S [69 213], 20:15 S [28sup]. The data suggest, though not strongly, that P might also omit the name. The explanation that a proper name could be dropped to avoid excessive use is supported by the instances in Mark 11:33a and b (different omissions by S and P), John 1:37 (before 1:38), and John 13:23a, where nearly all Greek witnesses have two occurrences of the name ‘Jesus’ in very close succession. Greek witnesses to these omissions may also result from the same pressure to avoid repetition. That the OS translators were not striving to represent their Vorlage literally is also suggested by the fact that they did not refrain from ‘moving’ the proper name ‘Jesus’ to an immediately adjoining clause (e.g. Matthew 26:50b S). The above-established translation tendencies with respect to the name ‘Jesus’ have consequences for many places in the critical apparatus of NA27. It is noteworthy

PERSONAL PROPER NOUNS

33

that, taken as a whole, the apparatus has many more references to Syriac witnesses having the name ‘Jesus’ than it does to them lacking it. References to P or to OS and P together omitting the name are noticeable by their absence. This imbalance itself might suggest that variants from translation are being treated as Vorlage variants in the apparatus. There are eight cases where P alone of the Syriac texts being considered is cited for the presence of arthrous or anarthrous EÉçóï™ò: Matthew 9:36, 15:30, 17:2, 17:20, 22:37, 26:25, John 19:38b, 19:39. The likelihood that the citation is correct varies from case to case. External evidence indicates that the citations in Matthew 15:30 and 22:37 are more probable than that in 9:36. However, where the external evidence is strong the citation of P is scarcely going to reach the levels of probability needed to meet NA27’s stated requirement for citations. The citation in Matthew 17:2, where D and some versions are the only other evidence for the presence of ‘Jesus’, should no longer be regarded as evidence for a genetic link between the reading of P and that of these texts, since P could agree with them independently. There are five more cases where S and C, when extant, are cited together with P as supporting the presence of [¿] EÉçóï™ò: Matthew 8:3 SCP,17 Mark 12:41 SP (also in UBS4), 16:19 CP (also in UBS4), John 8:21 SP, 13:3 SP. None of these citations are improbable, even though none are compelling. The division of Syriac witnesses in close association with the name supports the 17 S has

.

34

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

citations in some cases. In Mark 12:41, S uses the word ‘stand’, and P has ‘sit’, both with significant Greek support, suggesting that they were independent translations. The same, with significant implications, can be said for Mark and P . 16:19 where C has P would be unlikely to have revised a Syriac text like C away from the Greek order êýñéïò EÉçóï™ò, and it is likely therefore that at this point, and therefore presumably for the ‘longer ending’ of Mark, C and P are not directly related. P is certainly not a revision of C. The Syriac witnesses in John 13:3 are also divided with P formally corresponding to Greek witnesses with äÝ and S to those without. The likelihood of the addition of the name ‘Jesus’ needs therefore to be considered in conjunction with the representation of Greek particles in Syriac. In addition to these five cases there are instances where the OS witness is split in some way. In Matthew 8:7 NA27 cites CP for the addition of the name and S supporting txt in its omission. For this see pp. 36–37. In Matthew 21:27 NA27 cites CP along with ) (0293) pc (it) for the addition of the name ‘Jesus’ in contrast to txt.18 S probably lacks the addition because, like Greek texts, it has already used the name earlier in the verse. The Syriac readings could have been based on several possible Vorlagen and without the support of CP the variant in the apparatus looks rather weak. In John 4:16 (S)CP are cited for the addition of ‘Jesus’, the parentheses being used . because S has

18 Ms 0293 has êáéïiwåé[ðå]í.

35

PERSONAL PROPER NOUNS

On a couple of occasions txt rell are cited against a small number of witnesses omitting ‘Jesus’ (John 13:10, 19:30). It should be remembered that SP and P, respectively, are only questionable members of rell. Whereas P has a major tendency to add the name ‘Jesus’, and only a minor tendency to omit it, we have seen that OS has significant tendencies both to add and to omit the name. Hence it is no surprise that many citations in NA27 represent P as supporting, but OS witnesses as opposing, the presence of the name. This could be taken as a sign that OS generally preserves a more authentic text than P, which represents a step in a more secondary, Byzantine direction, with greater representation of the name ‘Jesus’. But even if this is so, the translation tendencies set out above cannot be gainsaid, though one may decide that the Syriac witnesses should not be completely abandoned with regard to establishing the presence or absence of the name in the Greek text. NA27’s citations where OS and P are opposed are as follows: Reference Matthew 8:29 Matthew 9:12 Matthew 13:36 Matthew 14:14 Matthew 17:20 Mark 14:22 Luke 10:21 John 4:46 John 6:14 John 18:5 John 20:21

With ‘Jesus’ P P (P) P P P P P P P txt P

Without ‘Jesus’ txt S txt S txt SC txt SC txt SC txt S txt SC txt C txt SC txt S S

UBS4

as NA27 as NA27 as NA27 as NA27

The citation in Matthew 8:29 is almost certainly correct. The vocative ‘Jesus’ in the middle of the demoniacs’ cry is

36

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

of a quite different sort from the other variants treated here. The citation of (P) in Matthew 13:36 is much more problematic. The parentheses are used because the proper name is actually in a different clause relative to Greek witnesses. This, combined with P’s known tendency to add the name, suggests that the citation should be abandoned. In Matthew 14:14, SC are not cited with txt, but as agreeing uniquely with some Old Latin witnesses in lacking both the name ‘Jesus’ and the participle dîåëèþí. How closely the reading of SC should be compared with that of the Old Latin witnesses is debatable. Unlike them SC lack reference to a boat in Matthew 14:13 and therefore must omit the reference to Jesus ‘going out’. That SC lacked ‘Jesus’ in their Vorlage seems probable, even if textual complexities need further consideration. The striking feature of the list is also the close relationship between txt and SC. Only in John 20:21 does txt depart from S, and in that location NA27 prints the name ‘Jesus’ in brackets to mark its uncertainty. While it is difficult to single out any of txt’s readings as improbable, we must be aware of the possibility that txt itself may have been formed by giving excessive weight to the readings of SC, which are unreliable witnesses with regard to the presence or absence of the name in their Vorlage. If the maxim lectio brevior potior was not held with regard to these titles, and SC were also given relatively little weight in such decisions, the weight of evidence might be tipped against txt in some cases. In several cases NA27 cites OS witnesses against the presence of the name: Matthew 8:7 S (see p. 34), 8:22 S, 9:22 S, 14:27 C, Mark 9:39 S, 10:52b S with txt, Luke 9:43

PERSONAL PROPER NOUNS

37

SC with txt, John 21:5 S, 21:17b S. Matthew 8:7, 8:22, and 9:22 are in the middle of a ‘m ran-section’ within S, and the possibility must be considered that the name ‘Jesus’ , and then subsequently revised out was translated by due to its lack of conformity to the Greek (the same may have happened in John 6:10 S). The recurrence of certain witnesses to the omissions, such as ) D W, could support a special relationship between these manuscripts and OS, though the treatment of the name ‘Jesus’ in these manuscripts should also be made a subject of separate study. The omission in John 21:5 is cited with relatively weak support (A*vid W a) and the agreement of these witnesses with S may be coincidental. The ‘omission’ in John 21:17 needs to be considered in the light of the fact that S has added the name ‘Jesus’ already twice in the verse in contrast to all the variants in our control editions. To sum up, it seems that neither OS nor P tried to imitate the use of the proper name ‘Jesus’ in their Vorlagen. P in particular shows a tendency to add the name, especially where its Vorlage had a low distribution of the name. The explanation for the distribution is largely a matter of linguistics combined with the more subjective tastes of any translators, revisers, or tradents, who wittingly or unwittingly established the ratio of nominal to pronominal identification of the main figure in the Gospels. There has to remain some doubt over the use of any Syriac witness to testify to the presence or absence of the name, and in many cases the doubt is so great that SCP need to be abandoned as witnesses to Greek variants.

38

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

2.1.2 Other Personal Names If the preceding explanation for the distribution of the name ‘Jesus’ is correct in broad outline then we would expect similar features to occur in SCP with regard to other personal proper names. The proportions of additions to omissions might vary, but one could hardly expect translators to fail to represent literally the name ‘Jesus’ and pronouns substituting for it, and yet to treat references to all other individuals in an exact way. Here we will consider the possibility that other proper names substitute for pronouns or vice versa.19 The pattern we find is that OS and P both show additions and omissions of names where our control editions show no Greek variants. Additions outweigh omissions, and OS’s departures from the Greek outnumber P’s departures. The additions in contrast to our control editions are listed below. Matthew 1:19 SC (Mary), 2:14 SCP (Joseph), 2:21 P (Joseph), 2:22 SC (Joseph), 13:55a SC (Joseph), 14:5 C (John), 14:12 C (John), 17:25 P (Cephas), 27:18 P (Pilate), Mark 14:12 P (the Jews), 15:10 SP (Pilate), Luke 1:11 SP (Zechariah), 1:22 P (Zechariah), 1:42 SP (Mary), 1:56 SP (Elizabeth), 1:61 S (John), 2:28 S (Simeon), 2:39 S (Joseph and Mary), 3:19b S (Herod), 11:15 SC (the Pharisees), 22:33 P (Simon), John 1:45b S (Philip), 1:45b S (Nathaniel), 5:46b P (Moses), 8:25 P (the Jews), 11:6 S (Lazarus), 11:13 S (Lazarus), 11:17 S (Lazarus), 11:29 SP 19 In a slightly different context, the broad principle that the

OS translator might add or omit proper names is already mentioned in A. Smith Lewis, Zu H.J. Vogels Schrift, Die altsyrischen Evangelien in ihrem Verhältnis zu Tatians Diatessaron, 9.

PERSONAL PROPER NOUNS

39

(Mary), 11:49b S (Caiaphas), 12:11 S (Lazarus), 18:26b S (Simon), 21:17b S (Simon), 21:19 S (2× Simon).20 S shows the most additions proportionate to its extant text. A few comments are necessary about individual cases. Matthew 2:21 and 2:22. P adds ‘Joseph’ in the former text and SC add ‘Joseph’ a verse later. The versions are thus independent, but achieve approximately the same end, namely adequate specification of the figure involved. This supports the view that the additions are linguistically motivated. Matthew 13:55. The name ‘Joseph’ is probably substituted for ‘the carpenter’ in Greek in order to make the text easier to understand. Matthew 17:25. P’s addition of ‘Cephas’ is to be compared with C’s addition of ‘Simon’ in the same verse, but in a different clause (matched by Greek ms 713 with ¿ ÐÝôñïò). The two Syriac versions thus show themselves to be independently reflecting the same tendency to use a proper noun.21 Luke 1:11. ‘Zechariah’ is added in SP to avoid the understanding that the angel appeared to the crowd. Luke 3:19b. ‘Herod’ is added in S to make it clear that Herod had not married John the Baptist’s sister-in-law. It may be that after making this explanatory ‘addition’ S omits a further reference to Herod later in the verse in 20 The addition of ‘Joseph’ in Matthew 1:16 S should also be

in this list, but because of its controversial nature it is treated separately on pp. 240–44. 21 The complexities of the use of names for Peter in the Syriac gospels are discussed at length by Burkitt, Evangelion DaMepharreshe, II, 92–96; see also Brock, ‘Limitations of Syriac’, 88.

40

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

agreement with ms 1194. On that analysis S then adds a further reference to Herod at the beginning of 3:20 in contrast to Greek witnesses. A simpler analysis is that S analysed the sentence division in 3:19 and 3:20 so that ðïíçñ§í ended a sentence and the name ‘Herod’ began the next and governed ðñïóÝèçêåí. Whichever analysis is taken two Greek uses of the name ‘Herod’ correspond to three in S. John 1:45. The two additions in S are to make clear who was speaking and who was addressed. In addition to the occurrences above there are also places where a Syriac addition is paralleled by a small number of Greek witnesses. In some cases it is possible to demonstrate that any link between the Syriac and Greek is highly unlikely. Matthew 1:20a C (Joseph). C’s addition might seem to be paralleled in È, which reads åöáíç êáô ïíáñ ôù úùóçö ëåãùí ûúïó äáõåéä for txt’s êáôE –íáñ döÜíç ášô² ëÝãùí EÉùóxö õj’ò Äáõßä. However, È’s addition of ‘Joseph’ in the speech formula is connected with its omission of ‘Joseph’ in the address. C, however, has both. C’s reading is also paralleled in 280 892. ‘and he Matthew 1:24 C (Mary). C’s took Mary [as wife]’ might seem to find some parallel in the corrector of ) who inserted ìáñéáì to read ðáñåëáâåí ìáñéáì ôçí ãõíáéêá áõôïõ. The insertion was subsequently erased. However, as C lacks ‘his wife’ the variants are probably independent. Despite the theological import of the context, one should avoid assuming a theological motive for this variant, as if C is avoiding the term ‘wife’ (despite the fact that such avoidance may occur in 1:20 C).

PERSONAL PROPER NOUNS

41

It should be remembered that C also adds names in 1:19 and 1:20 (see pp. 38 and 40), and in neither of these cases is a theological motive likely. In 1:24 both theological and linguistic motives are possible, and the matter should not be prejudged. Matthew 17:25a C (Simon). The addition is paralleled in 713. Mark 6:27a P (John). P’s first addition of ‘John’ in this verse has some parallel in the reading of 485 1396. Mark 6:27b P (John). P has already added ‘John’ once in the verse. The second addition has some parallel in È 470 999. Luke 3:19 SP (John). The addition is paralleled in the different readings of F and 903. Luke 16:31 CP (Abraham). The addition is attested in the different readings of 1071 and l1663. Luke 19:3 CP (Zacchaeus). Though the addition is paralleled in 0139vid, one only has to read C’s text to realise that without explicit mention of Zacchaeus the text would be in danger of being read as saying that Jesus, not the tax collector, was vertically challenged. C reads ‘and he wanted to see Jesus, but was not able because he, Zacchaeus, was small in stature’. The addition of the name ‘Zacchaeus’ may be a secondary development after ‘because of the crowd’ ‘was able’. This was dropped out after restored in P, but the additional proper name was left unrevised. John 19:5 P (Pilate). Although P’s addition is paralleled in Nc 1321, the addition is absolutely obligatory

!

" #!

$

%

42

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

in the Syriac because without it the words ‘Jesus went out … and he said to them’ could only be read as indicating that Jesus, not Pilate, addressed the crowds. Although the number of instances where Syriac added the name ‘Jesus’ is much greater than the number of additions of other names, this can be explained on the basis of the relatively minor roles that other figures play in the Gospels. When other figures are prominent, such as Mary and Joseph in the birth narratives, Lazarus in John 11–12, and Pilate in the trial narratives, their names are added several times in contexts which do not suggest theological or harmonistic motive. It is best therefore to conclude that substitution of a proper name for a pronoun or common noun was an accepted translation method, even though it cannot be ruled out that some of the proper names first arose in transmission rather than translation. The comments here, of course, only really apply to proper names in a narrative context where there is an issue of how often a protagonist is to be portrayed nominally and how often pronominally. In genealogies, direct address, and citation of prophetic writings the inclusion or exclusion of particular names is affected by different factors. The number of Syriac omissions of names against Greek witnesses is relatively small: Luke 1:12 S (Zechariah), 4:27 S (Naaman), John 2:20 S (the Jews), 11:55 S (the Jews). Luke 1:12. The omission of ‘Zechariah’ in S is probably causally connected to the addition of the same name in 1:11 S. This could then be viewed as a case of

PERSONAL PROPER NOUNS

43

movement of a proper name. Such movement seems to be attested with the name ‘Herod’ in Luke 23:7 C. Luke 4:27 S is almost certainly an inner-Syriac corruption. Naaman’s name was removed in the phrase "& '( when the second word became understood as ‘Jeremiah’ rather than correctly as its homograph ‘the Aramaean’. The omissions of ‘the Jews’ in John’s Gospel need to be considered alongside the omissions of the same word in John 10:31 S, 19:21a P (both with a few Greek witnesses), and alongside John 7:13 S where (without Greek support) ‘the Jews’ become ‘the people’ (cf. 7:12). With these should be included Mark 14:12 P and John 8:25 P where ‘the Jews’ is added. There is thus a fair degree of variation in the use of the term ‘the Jews’, and the translators’ readiness not to imitate their Vorlage in the use of proper names is certainly an explanation to consider for this phenomenon.22 The view that Syriac witnesses might add or omit proper names in contrast to their Vorlagen calls for a reconsideration of how one compiles a textual apparatus. The replacement of a pronoun by a proper noun or vice versa seems to many modern scholars to transgress a boundary of acceptable translation and therefore to call for special attention. Occurrences of proper names in Syriac are therefore given a prominent place in textual apparatuses, whether in the editions of Legg, of the International Greek New Testament Project of Luke, or of von Soden. 22 Though see the comments on John 9:22 on p. 230, n. 45.

44

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

In the apparatus of NA27 the number of references to the inclusion or exclusion of a proper name is not very large, though there are many notes which have to do with how exactly an individual was designated. Precisely how someone was designated depends mainly on the designation in the Vorlage and the conventions of designation in the target community. That someone should be designated by a proper noun or pronoun is much more a matter to do with the texture and clarity of a narrative. A few comments on some of NA27’s notes for the inclusion or exclusion of a proper name follow. Matthew 13:35. As NA27 and UBS4 note, SCP do not add ‘Isaiah’. Since the name is not in narrative there is no reason to question this citation. ' ), which NA27 and Matthew 17:26. CP have UBS4 take as supporting the presence of ¿ ÐÝôñïò against the two readings of txt and S. The division of Syriac witnesses supports their analysis, though the addition in CP of proper name designations for the same disciple only a verse earlier despite Greek witnesses shows just how uncertain any citation of this sort will be. The citation of S, being an omission, is more plausible. Mark 1:16. As NA27 notes, SP omit ‘Simon’. Though there is good Greek support for the omission, it follows almost immediately after another mention of Simon. This increases the likelihood that Syriac witnesses could have avoided the proper name in contrast to their Vorlage. Mark 15:46. P adds ‘Joseph’. Despite good Greek support for the addition NA27’s citation of P is questionable because there is a clear reason for P to add a proper noun, namely to avoid the understanding that the

PERSONAL PROPER NOUNS

45

subject of the preceding verb (Pilate) was also the subject of ‘bought a cloth’. Luke 2:33. NA27 and UBS4’s citation of P for the reading EÉùóÞö for txt’s ¿ ðáôxñ ášôï™ (a reading S seems to support) is reasonable because in P it occurs in the combination ‘Joseph and his mother’. The use of a common noun alongside the proper noun makes implausible any explanation that the latter represents a common noun in its Vorlage. Luke 22:58. S, along with D and 69vid, lacks ‘Peter’ in ‘but Peter said’. That the citation is for an omission speaks in its favour. John 7:32. NA27 cites S as only having the last word in the sequence ïj Pñ÷éåñåsò êár ïj Öáñéóásïé ›ðçñÝôáò. ,(+ * However, earlier in the verse, S reads ‘the chief priests and Pharisees’ where Greek witnesses merely have ïj Öáñéóásïé. It seems, then, that S has merely merged the two references to Jesus’ opponents, showing a similar freedom to that shown in 7:45. At any rate S attests the presence rather than absence of ‘the chief priests’. John 13:6. This exemplifies the need to consider context before concluding the presence of a proper name in the Vorlage. SP read ‘And when he [Jesus] came to Simon ' ) & ]’. NA27 Cephas, Simon said to him [ notes that SP add Simon, here using Latin in the apparatus. Since SP are the only witnesses for this reading one may already suspect that the issue is merely translational. Further support is given to this when the Syriac sentence as a whole is considered. Take the wording in S: ' ) & -. ' ) / 0 . If the second Simon were removed the sentence would clearly

46

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

mean ‘When Jesus reached Simon, Jesus said to him, etc.’ If various Syriac alternatives are considered for translating the sentence unambiguously it will soon be realised that the addition of the explicit proper name is by far the most obvious, particularly if it were motivated by an dêåsíïò in the original. However, there is no particular reason why this sort of translation feature should appear in the textual apparatus.23 From the additions in John 13:6 and 19:5 dealt with above (pp. 41–42, 45–46) we can establish the principle that since the Syriac sequence Verb–Subjectival Proper Name– Verb Introducing Speech is used to signal that the speaker is the proper name (provided it agrees with the preceding verb), it is not an appropriate representation of the Greek sequence Verb–Subjectival Proper Name–Verb Introducing Speech, where the subject of the latter verb may be different from that of the former. Vorlage inferences from the Syriac addition of a proper name after the second verb are therefore hazardous. 2.1.3 Conclusions Under most conditions some uncertainty is involved in reconstructing a proper name in the Vorlage from a proper 23 Merx, Matthaeus, 160–71, in his discussion of the names

given to Peter in the Gospels, is very ready to posit otherwise unattested Vorlagen reading Óõìåþí: ‘Das Ergebnis ist, die Urtexte der Evangelisten hatten—vermutlich ausschliesslich— Óõìåþí, das im Laufe mannigfacher Redaktionen verdrängt und durch ÐÝôñïò ersetzt ist’ (Merx, Matthaeus, 168). On Merx’s reasoning Burkitt, Evangelion Da-Mepharreshe, II, 96, comments: ‘A similar argument might be used to prove that the Evangelists wrote Ìåóóßáò where our Greek MSS have ×ñéóôüò.’

PERSONAL PROPER NOUNS

47

name in S, C, or P. This can bring critical editions citing such reconstructions into conflict with their stated criteria for the use of versions. Moreover, in many cases the likelihood of such a reconstruction being correct is almost equal to the likelihood that it is not. At this point citation in the apparatus becomes pointless. This brief study has only considered personal proper nouns, but if the model presented here that versions avoid excessive nominal or pronominal reference is correct, we should expect similar patterns to occur with common nouns as occur with proper nouns. Preliminary survey suggests that SCP do indeed sometimes substitute common nouns for pronouns and vice versa. A full study is needed for this to be confirmed, and its frequency and conditions established. Needless to say, the text-critical implications would be considerable.

2.2 ADDITION AND OMISSION OF VERBAL COMPLEMENTS 2.2.1 Direct Objects While it is reasonable to expect that in general an object in the Syriac text will correspond to one in its Vorlage there are also reasons to suppose that this will not always be the case. The nearest semantic equivalent to an intransitive verb in one language may be a transitive verb in the other. Transitive verbs could be further subcategorized according to how likely they are to have an object. If a translation equivalent has a different level of preference for the presence of an explicit object this may, again, cause formal contrast with the Vorlage. A further issue to contend with is the use of shared objects. Both Greek and Syriac can share

48

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

an object between two verbs, but the object may occur more closely connected to a different verb in each language (see pp. 58–59 on John 19:6b P). Alternatively, a shared object in Greek might be expressed twice in Syriac (see pp. 56 and 57 on Mark 15:46 P). Or again, a verb in one language may indicate the object within itself (see pp. 59–60 on Mark 1:16). To date, users of Syriac for textcritical purposes have had too atomistic an approach to the presence or absence of objects. One needs to consider the contexts of the local clauses as well as the preferred constructions of the verbs used before one can draw conclusions about Vorlage. 2.2.1.1 Hear / Say + ôá™ôá A peculiar situation is that NA27 has four variants for which OS witnesses are invoked as supporting the absence of ôá™ôá, used absolutely as an object, in reference to what is said or heard. The similarity of these four texts, quoted below, invites attention. Luke 18:11 txt ¿ Öáñéóásïò óôáèårò ðñ’ò eáõô’í ôá™ôá ðñïóçý÷åôï; according to NA27 and UBS4 S omits ôá™ôá. Luke 18:22 txt Pêïýóáò äc ¿ EÉçóï™ò åqðåí ášô²; according to NA27 witnesses including A W È ˜ add ôá™ôá after äÝ, but SC lack it. John 7:32 txt }êïõóáí ïj Öáñéóásïé ôï™ –÷ëïõ ãïããýæïíôïò ðåñr ášôï™ ôá™ôá; according to NA27 C, along with D L* f 1 et al., omits ôá™ôá.24 24 S also lacks an equivalent of ôá™ôá, but this is not cited since

S also lacks an equivalent of ðåñr ášôï™ (ZLOds in C), and the variant in C is cited in connection with the whole phrase ðåñr ášôï™ ôá™ôá.

VERBAL COMPLEMENTS

49

John 9:40 txt }êïõóáí dê ô§í Öáñéóáßùí ôá™ôá ïj ìåôE ášôï™ –íôåò êár åqðïí ášô²; according to NA27 S, along with )*.2 D et al., omits ôá™ôá. That the word Öáñéóásïò occurs in three of the texts is only partly coincidental (as we will see on p. 51). The Vorlage Hypothesis seems initially plausible in at least the last three cases. It is less likely in Luke 18:11 because there is no Greek support for the omission. However, the Vorlage Hypothesis is significantly undermined by a small number of additional texts where similar conditions occur and absolute ôá™ôá used as an object of ‘hear’ or ‘say’ is omitted, despite being attested throughout the Greek tradition: Matthew 9:18 txt has ôá™ôá ášôï™ ëáëï™íôïò ášôïsò but S # 1 0 ‘But when he merely was speaking with them’. Luke 4:28 txt has êár dðëÞóèçóáí ðÜíôåò èõìï™ dí ô† ' ) 0 óõíáãùㆠPêïýïíôåò ôá™ôá but S reads ‘And when those who were $ 2 1 /& $) 3 in the synagogue heard they were filled with anger’. does not Noteworthy in this reading is that correspond to ôá™ôá, but is used to represent the personal subject of the sentence. It is possible to find some rationale for S’s rendering. The 0 -clause is naturally fronted. This leaves the dilemma that the ‘word for word’ translation 1 ' ) 0 ) would in fact (* $) 3 and 1 too close to avoid analysis as ‘all leave ( ' ) 0 these [people]’. P has ‘and when those 1 $ 2 1 /& $) 3 who were in the synagogue heard these things, they were all filled with anger’. This involves what is for Syriac an

50

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

unusually large distance between the agreeing words ( and 1 , thus showing the difficulty of a literal rendering of txt in Syriac. But, whatever the motivation for the change in S, the text shows how easily an equivalent of ôá™ôá could be dropped. Luke 7:9 txt has Pêïýóáò äc ôá™ôá ¿ EÉçóï™ò dèáýìáóåí 4 ) 0 ášôüí, which is translated in S simply /& ‘and when Jesus heard, he marvelled at 25 him’. Luke 24:36. SC lack an equivalent of ôá™ôá as object of , but it is analysable as the subject the verb ‘say’. P has of the verb ‘say’ and therefore as personal. John 8:26. txt’s êPãù S }êïõóá ðáñE ášôï™ ôá™ôá ëáë§ åkò 3 $' ) 1& ô’í êüóìïí is simplified in S to 1' 311 ‘and those things that I have heard from him I say in the world’. These passages, where ôá™ôá is present throughout the Greek tradition (except where stated) call into question the likelihood of an otherwise unattested Greek text in Luke 18:11 (as reconstructed by UBS4). This, of course, makes a total of six passages where OS witnesses diverge significantly from Greek ones. These in turn call into question the citations in Luke 18:22, John 7:32, and 9:40. In addition an equivalent of ôá™ôá is omitted after 4 ) in Luke 14:15 S (contrast CP) with support only from )* 71 827 1093 1458 2096 2643 l211 l1056, and in 20:16 is added after 4 ) despite the lack of ôá™ôá in SC the Greek. This fluctuation may find some explanation in the hypothesis that ‘these things’ did not appear to the 25 ôá™ôá is omitted in mss 700 713 1038 1247.

VERBAL COMPLEMENTS

51

translator to be a prominent element either in the Vorlage or translation, and therefore not an element with respect to which consistent correspondence between the two was sought. As intimated earlier an explanation for the occurrence of the word Öáñéóásïò in the plural in two of the variants in NA27 can readily be found. The problem with the in the Syriac in both cases would be that inclusion of it would be read as agreeing with ‘Pharisees’. Thus, for were included it would be instance, in John 7:32 if read ‘and these chief priests and Pharisees heard’. The same explanation applies, but to a lesser extent, in John is a 9:40. The key point motivating the omission of plural personal subject with which it could be read as agreeing. The explanation is thus basically of the same kind as that applied to Luke 4:28, and certainly the right explanation for the phenomena in Luke 24:36. The same sort of explanation may apply to the omission of the whole phrase ôï™ôï or ôá™ôá + åkðþí in Luke 23:46 SC and åqðåí äc ôï™ôï in John 12:6 S. Assuming that the singular ôï™ôï was in the Greek Vorlage this can be represented by the feminine 5 , but also by the masculine ( when it occurs in the combination ( (e.g. Matthew 6:25 CP, 12:27 CP). The use of the masculine in this expression may be because a masculine 6$- is understood, or stood at the root word such as of the expression. This might mean that ôï™ôï, when referring to a thing said, rather than to a thing done, might notionally be represented by ( . However, this would not generally occur because in combination with the & in the singular it would be read as its subject. verb

52

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

certainly may be used absolutely to represent ôï™ôï (e.g. Matthew 26:9 S, 26:26 SP, 26:28 SP). The possibility of the omission of ôá™ôá near a verb of speech allows the reading of SP to have been derived from D’s variant involving ôá™ôá before ëáë§ in John 8:38. For further doubt about NA27’s note here see pp. 90–91.

(

2.2.1.2 The Greek Sequence Verb–Object–Verb, etc. When Greek has two coordinated verbs with the same understood or expressed object, Syriac has a distinct preference for the sequences Verb–Object–Verb–Object (VOVO) and Verb–Verb–Object (VVO) as opposed to various Greek sequences such as VOV or OVV. This means that in general the object is expressed later in the Syriac translation, or is repeated so as to occur both earlier and later. The data are not simple because there are so many Greek variants in relation to the presence of pronouns, but even allowing for various Greek Vorlagen consistent deviations in the Syriac can be observed.26 Object, for this purpose, will refer not only to so-called direct objects, but also to patients connected with the verb by means of a preposition. VOV VOVO Matthew 5:11 txt variant S C

“íåéäßóùóéí ›ìOò êár äéþîùóéí äéþîùóéí ›ìOò êár “íåéäßóùóéí

-"

(& 3 08

37 (& 3 +

( -"

26 The record of Greek variants here omits minor variations

which do not affect the argument.

53

VERBAL COMPLEMENTS

P

-"

08

Matthew 18:8 txt variant P

hêêïøïí ášô’í êár âÜëå hêêïøïí ášôN êár âÜëå

Matthew 18:9 txt S P

hîåëå ášô’í êár âÜëå

Luke 20:15 txt

dêâáëüíôåò ášô’í hîù ôï™ Pìðåë§íïò PðÝêôåéíáí

SCP V[O]V VOVO Luke 10:31 txt variant SC

0) 0) 0)

1 9

;

:2 :2

%-&

kä¦í ášô’í Píôéðáñyëèåí kä¦í Píôéðáñyëèåí

Luke 10:32 txt variant SC

kä¦í Píôéðáñyëèåí kä¦í ášô’í Píôéðáñyëèåí

Luke 10:33 txt variant SCP

kä¦í dóðëáã÷íßóèç kä¦í ášô’í dóðëáã÷íßóèç

VOV VVO Matthew 5:30 txt CP

9 8-

1#

;#

2

;#

2

2"/&

2

hêêïøïí ášô’í êár âÜëå

0) < 8-

54

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

Matthew 18:8 txt variant SC

hêêïøïí ášô’í êár âÜëå hêêïøïí ášôN êár âÜëå

Matthew 18:9 txt C

hîåëå ášô’í êár âÜëå

V[O]V VVO Mark 11:2 txt variant S P VV VVO Matthew 21:2 txt C OVV VOVO Mark 14:1 txt S OVV VVO Mark 14:1 txt P

0) < 80) :2 ëýóáôå ášô’í êár öÝñåôå ëýóáíôåò PãÜãåôå

$& &$ &

) )

(& $ &

)

ëýóáíôåò PãÜãåôÝ ìïé

ð§ò ášô’í Pðïêôåßíùóéí

( 1 %( ð§ò ášô’í Pðïêôåßíùóéí

( 1 %(

dí

äüëv

( 02.( dí

äüëv

02.(

êñáôÞóáíôåò

1 3

3 &

êñáôÞóáíôåò

1 3

3 &

The general Syriac tendency to put the object later and the particular preference for VOVO and VVO is evident. So far it has not been possible to find a way to explain the distribution of VOVO and VVO constructions relative to

55

VERBAL COMPLEMENTS

each other, except to note that VVO is preferred when the two verbs fulfil the conditions for asyndeton (on which see pp. 150–54). Besides these transformations where the boundaries of the structure in the Vorlage are fixed, there are also some transformations where variants are significant for our purposes, but the objects bracketed below were not sufficiently attested to be recorded in NA27. VOV[O] VOVO Matthew 26:57 SP, Mark 6:13 S, 14:44 SP, Luke 9:21 SCP, 11:28 SCP, 13:12 SCP, 23:16 SCP VOV [VVO] VOVO Luke 23:22 SCP The same tendency to add objects and to position objects later in the sentence can be observed when a series of three verbs occurs. Thus the patterns VOVOVO and VVOVO emerge as favoured in Syriac. These may be regarded simply as the VVO or VOVO patterns mentioned above with a VO extension on the end. In John 19:15 P (below) we have VOVOVOVO. The reading of txt comes first and attested variant orders are placed in brackets. In Matthew 27:2 and Mark 12:3 the VVV orders are only inferred from the apparatus of Legg. Matthew 21:39

VOVV [VOVOV]27 VOVOVO

SP VVOVO, C

27 The reading of È is ignored here as irrelevant.

56

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

Matthew 27:2 Mark 12:3 Mark 15:1 Mark 15:46 Luke 14:4

VOVV [VVV, VOVOV, VOVVO] SP VOVOVO VOVV [VVV] S VOVOVO, P VOVO VOVV [VOVVO] SP VOVOVO VOVVO [VVOVO, VOVOVO] S VVOVO (corruption), P VOVOVO VVOV [VOVV, VOVOV] SCP VOVOVO

In John 19:15, txt has Wñïí Wñïí óôáýñùóïí ášôüí for which NA27 records no variants. The only variant from the VVVO structure in our control editions is a minor variant VVV. The contrast between this and P’s VOVOVOVO structure, %) %) ‘take him away, take him away, crucify him, crucify him’, is quite striking, but well illustrates the Syriac tendency to add the object with short transitive imperatives. Including some cases discussed above, five potential transformations should be dealt with together, with their better attested variants in brackets: Matthew 21:2 C VV VOVO, 27:2 SP VOVV[O] VVO, 26:57 SP VOV[O]28 VOVO, Luke 19:30 VOVOVO, Mark 14:44 SP VOV[O]29 VVO. VOV [VVO]30 Although in the cases of Mark 14:44 and Luke 19:30 it is not unlikely that the Syriac translations were made from Vorlagen with the object expressed, this is much less likely for Matthew 21:2 and 26:57. We can only deal with probabilities, but taken as a whole these texts suggest that Syriac is quite likely to add an object under particular 28 Ms 700 adds the object. 29 D È 157 565 add the object. 30 This is attested by A K È Ð.

VERBAL COMPLEMENTS

57

circumstances. The circumstances which are common to the five texts listed above are that Tãù or PðÜãù is the second verb, and follows another verb or participle with the same subject and understood or expressed object. In each case NA27 reads the text without an object for Tãù or PðÜãù and Syriac texts have an object with their equivalent of that verb. That the object has always been at least contextually expressed facilitates its formal expression in Syriac. 2.2.1.3 Textual Implications The implications of the Syriac tendency to use the constructions VVO and VOVO (and VVOVO and VOVOVO) are obvious, affecting the critical apparatus of NA27 in a number of examples, most of which are from John’s Gospel. Luke 23:53. Where txt has êár êáèåë¦í díåôýëéîåí ášôü NA27 cites SCP in favour of an additional ášôü after êáèåëþí (as in A and ˜), thus to support VOVO rather than $2& ‘and he took it [Jesus’ VVO. SCP read body] down and wrapped it …’ Yet in the parallel text in Mark 15:46 P reads the same where txt has êáß … êáèåë¦í ášô’í díåßëçóåí, with no explicit object after the second verb. John 7:34. Where txt has æçôÞóåôÝ ìå êár ïš÷ å›ñÞóåôÝ [ìå] NA27 cites SCP with the VOVO structure in its support against the variant VOV without the second ìå. The support of SCP has to be dropped, but without this support the balance of external evidence shifts yet further away from txt, which, on the principles on which the

58

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

edition was compiled, had only a marginal lead over the variant. John 7:36. Not only are txt and the variant with NA27’s citation of ‘sy’ exactly the same as in John 7:34, but the evidence for both is remarkably similar. The citation of ‘sy’ is slightly misleading since S goes its own direction and does not have a straightforward correspondence with txt or the variant. CP, however, have the expected VOVO structure. Again, without their support, txt loses any slender lead it had over the variant. John 12:47. SP’s VOVO structure does not support the addition of ášôÜ for which NA27 cites it. John 18:12–13. NA27 cites SP for PðÞãáãïí ášôüí instead of }ãáãïí in txt’s êár häçóáí ášô’í êár }ãáãïí ðñ’ò $ &, and are evidently being GÁííáí ðñ§ôïí. SP have cited by NA27 because they express the object, not because Tãù and PðÜãù would necessarily be distinguished in Syriac.31 The citation should be dropped not only because this could be a simple VOV to VOVO transformation, but also because Tãù or PðÜãù are involved, and some tendency for these verbs to be represented with objects in Syriac has been noted above. John 19:6. After txt’s reading óôáýñùóïí óôáýñùóïí, NA27 cites P as adding ášôüí. P in fact reads ‘crucify him, crucify him’, i.e. VOVO. This suggests that the presence of the object may be a feature of the language. Support, though not of a compelling nature, is given to this notion in several ways. First, the active32 31 The reading }ãáãïí ášôüí is also attested by the minuscules

69 124 788. 32 Matthew 27:22–23 has the passive in both Greek and SP.

VERBAL COMPLEMENTS

59

imperatives calling for Christ to be crucified always have the pronominal object expressed (Mark 15:13 S [2×], 15:13 P, 15:14 SP, Luke 23:21 SCP [2×]). The significance of this is lessened by the presence of objects in Greek witnesses. Secondly, later in John 19:6 Pilate taunts the Jews saying ëÜâåôå ášô’í ›ìåsò êár óôáõñþóáôå. P’s =9 ! $(& ‘you take and crucify equivalent him’, transfers the object to ‘crucify’, which would fit with the hypothesis that this is a preferred idiom. 2.2.1.4 Miscellaneous Objects Mark 1:16. According to txt Andrew and Simon were PìöéâÜëëïíôáò dí ô† èáëÜóów; D f 13 et al. read PìöéâÜëëïíôáò ôN äßêôõá; A W ˜ read PìöéâÜëëïíôáò Pìößâëçóôñïí. NA27 cites P in support of the reading of D f 13. The problem with this is that PìöéâÜëëù has within itself not only the sense of ‘throw’, but also of what is being thrown, i.e. a net. Syriac has no equivalent verb and therefore needs the /0 : + " object. P says that Simon and Andrew were ‘throwing nets in the sea’. It is probable that merely " ‘throwing in the sea’ would not make sense.33 /0 : " ‘throwing their nets in the S has sea’, for which we certainly have no authority to reconstruct a form with a possessive. S’s reading does, however, fit with the hypothesis that Syriac needed an object. In this case the hypothesis is hard to prove since PìöéâÜëëù is a hapax legomenon in the NT.

33 Matthew 4:18 S does not contradict this since there m\g

governs the object jNL‡K[{g, which was placed after the preceding verb m\l}ˆg, ‘preparing’.

60

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

Support might be sought for NA27’s citation in two facts: (1) that P has revised S by the omission of the possessive—this might be explained by P seeking to follow its Vorlage closely—, and (2) that P has the plural, hence it had ôN äßêôõá, not Pìößâëçóôñïí. However, the second argument, about number, does not hold since P also uses /0 : + " in Matthew 4:18 where the Greek text âÜëëïíôáò Pìößâëçóôñïí åkò ôxí èÜëáóóáí is certain. The first argument, that P has omitted the possessive, is a more powerful argument against P having read merely PìöéâÜëëïíôáò in Mark 1:16. However, this still would not allow us to decide whether P read PìöéâÜëëïíôáò ôN äßêôõá or PìöéâÜëëïíôáò Pìößâëçóôñïí. 2.2.2 Indirect Objects 2.2.2.1 Forgiving People Their Sins In the Greek Gospels when there is talk of someone’s sins (or debts, or transgressions) being forgiven, there are two ways that this can be expressed. The fuller form, represented in Töåò ½ìsí ôN “öåéëÞìáôá ½ì§í (Matthew 6:12 txt), contains both a dative of the person forgiven and a genitive marking their possession of the sins or debts. In txt this form also appears in Matthew 6:14, Mark 11:25, Luke 5:20, 5:23, and 11:4. The shorter form appears in Matthew 9:2 txt: Pößåíôáß óïõ áj Qìáñôßáé. Here the genitive possessive is retained, but the dative is absent. This appears in txt also at Matthew 9:5, Mark 2:5, 2:9, Luke 7:47, and 7:48. To these two constructions appearing in txt, we should also add a third, which sometimes appears in

VERBAL COMPLEMENTS

61

manuscripts, namely the one with the dative but not the genitive, e.g. PöÝïíôáß óïé áj Qìáñôßáé (Matthew 9:5 f 13). Although there is considerable variation in the Greek witnesses for many of these texts there is a striking uniformity of expression in the Syriac witnesses, a uniformity that seems independent of the Vorlage. In each of the above-cited texts the extant Syriac witnesses have both the dative introduced by l madh, and the possessive. The order is always >;), l madh + noun or pronominal suffix, sins/debts/transgressions + pronominal suffix.34 Thus, although in txt some instances have the dative and 2+ ? % ;) occurs in Matthew 9:2 some do not, ? SP, 9:5 SP, Mark 2:5 P, 2:9 P, Luke 5:20 SP, 5:23 SP.35 A particularly clear example showing that a Syriac text can add a dative contrary to the Greek is found in John 20:23, not dealt with above because of its unusual construction. There txt has Tí ôéíùí Pöyôå ôNò Qìáñôßáò PöÝùíôáé $(& ášôïsò Tí ôéíùí êñáôyôå êåêñÜôçíôáé. S has

$(& 0 2&

1#

% $ ( 2 %;)/ 02&/ ‘Of whom you forgive to him

his sins they will be forgiven to him, and [of] whom you retain [his sin] upon him, it is retained.’ Thus although txt in both the prohas a dative in the apodosis, S contains + 2 %;)/ & tasis and apodosis. P reads @(. 0 2& @(& 02&/ & % $ ( ‘If you forgive sins to anyone they will be forgiven to him and if 34 A structure which also occurs in the colophon to S, where

the scribe prays that his own sins may be forgiven. 35 The use of sey me is not uniform. For further comments on this uniformity of construction vis-à-vis the Greek see the comments on tense on pp. 199–200.

62

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

you retain [them] of anyone, they are retained.’ P here has a dative in the protasis where Greek texts have a genitive. At this point NA27 cites SP in support of ôéíïò instead of ôéíùí for both of the latter’s occurrences in txt. The Greek support for ôéíïò is ms B. This citation is somewhat shaky for the following reasons. Though both S and P have a singular in the verse they contrast with B in using in the first apodosis where B has plural ášôïsò. singular Thus while the Syriac texts have the person consistently in the singular B does not. Secondly, S also makes ‘sin’ in the 0 2& ‘it is second apodosis singular by using retained’. It does not therefore seem to be following the number of the Greek carefully at this point. Thirdly, the Greek plural, ôéíùí, though formally plural is not semantically distant from a singular. It does not require that more than one person’s sins should be forgiven for Jesus’ saying to work. That is why a translation like the NEB uses a singular in this verse. Given these matters and the fact that both S and P have dative constructions in contrast to the Greek, it is difficult to argue that their translation is literal enough for us to know that they had ôéíïò. To consider less complicated variants, we have to conclude that, given the observations above, Syriac witnesses cannot be cited to decide whether a pronominal dative of the person forgiven should be present in the Greek, or whether a genitive should be present. This involves removing NA27’s citations of Matthew 9:2 SP and Luke 5:23 SP. In Matthew 6:15 the second reference to P should be removed, as perhaps also the second reference to C. In this latter case C is cited alongside Latin ms c and no other

VERBAL COMPLEMENTS

63

witnesses to support the reading PöÞóåé ›ìsí. This has in its favour the lack of a 2 pl. suffix with ‘father’ immediately before. Without the support of C it is unlikely that the Latin ms alone can confidently be used to reconstruct this Greek text. A similar reconstruction, which the Syriac cannot support, is in Matthew 12:31. There NA27 cites SC along with some Latin and Coptic witnesses in favour of the addition of ášô² after txt’s ½ äc ôï™ ðíåýìáôïò âëáóöçìßá ïšê PöåèÞóåôáé. The problem here is that SC have # [ ] personalised the whole clause to read > $ ( C0D( [ ) 9 ] 2 " ‘Everyone who blasphemes against the [Holy] Spirit, it will not be forgiven him.’36 The construction thus cannot be compared with that in txt, and certainly one part of it cannot be extracted and used to support a variant from txt. 2.2.2.2 Giving Frequently äßäùìé and its compounds occur with a dative of the person given to. This is not always the case since these words have a wide semantic range, and moreover there may be no specific beneficiary in view (e.g. Luke 6:38a). As might be expected Syriac texts do not generally introduce beneficiaries where there are no specific ones at least implied in the context of the Greek. Nevertheless, there are still a number of places where the Syriac witnesses use suffixed l madh where no or little support for the presence of a Greek dative is found in our control editions. This addition is not confined in Syriac to a single lexical item

36 The brackets in the Syriac text mark elements only in S and

C respectively.

64

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

(e.g. the verb E /F$(), but seems to characterise the general category we might call ‘beneficiaries’. Matthew 18:30 fùò Pðïä² ô’ “öåéëüìåíïí ‘until he # =( 0# should give what was owed’. C reads G 2 H0 ‘until he should repay him what he owed’ and P G2 F$( 0# ‘until he should reads 37 give him what he owed him’. Matthew 19:7 ôß ï¤í Ìùûóyò díåôåßëáôï äï™íáé [SC F$( ] âéâëßïí Pðïóôáóßïõ; ‘Why did Moses command to give [SC give to her] a certificate of divorce?’ Luke 12:42 ôßò Tñá dóôrí ¿ ðéóô’ò ïkêïíüìïò ¿ öñüíéìïò •í êáôáóôÞóåé ¿ êýñéïò dðr ôyò èåñáðåßáò ášôï™ ôï™ äéäüíáé [SC F$( ] dí êáéñ² ô’ óéôïìÝôñéïí; ‘Who is the faithful and wise steward whom the master will set over his servants to give [SC give to them] the apportioned food in its time?’38 E ] ô’í ëüãïí ôyò ïkêïíïìßáò Luke 16:2 Pðüäïò [P óïõ ‘give [P give me] the account of your stewardship’.39 In addition to these texts in Matthew 14:7 a single Greek dative might have been analysed as going with more than one verb: ©ìïëüãçóåí ášô† äï™íáé • dNí ákôÞóçôáé ‘he

37 P here shows the additional suffixed l madh following the

word ‘owe’. This is also present in other texts in the context, namely Matthew 18:28 SCP, 18:34 P. In the latter case there is considerable Greek support for the addition, in the former only 892c. 38 The dative ášôïsò is supported by 027 13 28 69 124 346 543 788 826 903 983 1346, italicised witnesses being assigned to the entity f 13. 39 The dative ìïé is attested in 157 1654.

VERBAL COMPLEMENTS

65

declared to her to give whatever she might ask’.40 SCP with their equivalents of both ‘declare’ and ‘give’. have The polyvalent dative in óõíÝèåíôï ášô² Pñãýñéïí äï™íáé ‘they agreed with him to give money’ (Luke 22:5) is ‘understood’—if this does not suggest too naïve a view of equivalents—by SCP as only going with ‘give’. This may reflect a preference for the suffixed l madh with ‘give’. To these cases where Syriac idiom seems to have preferred suffixed l madh to occur with ‘give’, we may add two more cases where the addition of suffixed l madh seems more closely bound to the particular understanding of the Syriac translators. In Matthew 20:23 ô’ äc êáèßóáé dê äåîé§í ìïõ êár dî åšùíýìùí ïšê hóôéí dì’í ôï™ôï äï™íáé ‘but to sit at my right and at my left is not mine to give’ // ] [C one] on my becomes ‘that you may sit [SC right hand and [C one] on my left hand, this is not mine to F/& ]’. In Luke 12:58 for ä’ò dñãáóßáí give you [SC E ‘give him his work [= deserts?]’, and S has 0;# E ‘give him his benefit’. C has 3 ( Enough examples have been gathered to raise doubts over whether the presence of suffixed l madh in a Syriac text can reasonably be used to infer a dative in its Vorlage. NA27 cites Syriac support for a Greek dative in Matthew 14:9 SCP, 18:28 SCP, 27:26 S. The dative in 14:9 is poorly attested in the Greek, rendering the citation less likely to 40 K W à Рf

1

along with a significant number of minuscules have the single dative after äï™íáé ‘give’. If their reading was the Vorlage of SCP then the Syriac texts still add a suffixed l madh in contrast to the Greek. Moreover, given the substantial differences between SC and P, the addition seems to have occurred at two different historical junctures.

66

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

be correct, though even in 18:28 with more support the Syriac texts are only doing what we would expect. Both these citations should therefore be dropped. The citation in 27:26, though well supported by Greek texts, has the problem that SP have suffixed l madh in a similarly worded parallel passage in Mark 15:15, but where Greek support for the dative is rather weak. NA27’s citation of S for the omission of the dative in Matthew 18:26 cannot be faulted on linguistic grounds, given the general tendency of the Syriac to add a reference to the beneficiary, though there is the marginal possibility of a harmonisation towards the reading of S in Matthew 18:29. One is pleased not to see SP cited in support of the one Greek manuscript which adds the dative in Matthew 27:58. 2.2.2.3 Preparing In Luke 17:8 NA27 records that SCP along with ) and some versional witnesses add ìïé after eôïßìáóïí. However, in two other places in the same gospel (12:47 SCP, 22:12 S41) Syriac witnesses can be observed to add a beneficiary in translating eôïéìÜæù. In this connection we may note the OS rendering of the Isaian quotation eôïéìÜóáôå ôxí ¿ä’í êõñßïõ ‘prepare the way of the Lord’ (Matthew 3:3 SC = 2" & ; I ‘prepare the way for Luke 3:4 S) by 2" & 3-, which the Lord’. In Luke 3:4, C has seems like a secondary assimilation to the Old Testament Peshitta (Isaiah 40:3), though it is not impossible that the OS readings in Matthew and Luke originally used different verbs and that S’s reading in Luke is an assimilation to its

41 The dative ½ìsí is added in l950.

VERBAL COMPLEMENTS

67

reading in Matthew. Whichever way, the difference of verb from the Old Testament Peshitta shows that at least once, independent of the Old Testament Peshitta, a Syriac translator used a suffixed l madh to express what was represented in his Vorlage by a Greek genitive construction.

2.3 POSSESSIVE PRONOUNS In dem von mir reconstruirten Text ist auch hier ein entsprechendes griechisches Pronomen nur dann aufgenommen, wenn griechische Handschriften es bieten. Aber auch in diesem Fall darf man, wie bereits bemerkt, auf die Bezeugung durch Sc [C] kein grosses Gewicht legen.42

Baethgen, in his edition of the Greek text behind C, was of course describing how C could not be relied on to provide evidence of the addition of a possessive unless there were Greek support for that possessive. However, if Baethgen’s method is seen as anything other than a crude way of arriving at a decision it is problematic since, if a translation is capable of adding a possessive when no Greek text has one, it is also capable of adding a possessive when some Greek texts have one. If the process of addition can take place in translation, then it is possible that the Syriac has come to agree with Greek witnesses independently. Baethgen, however, had a problem that the compiler of a critical apparatus does not have. Since he was publishing a reconstruction of the entire Vorlage of C, he had to make regular decisions one way or the other with regard to the presence or absence of a possessive. The compiler of 42 Baethgen, Evangelienfragmente, 21.

68

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

a critical apparatus faces far fewer such decisions, and is able simply not to register the presence of a witness if its testimony is ambiguous. Nevertheless, it is important to formulate more precise criteria for deciding what can be known about the use of possessives in the Vorlage of the Syriac. There are of course many types of nouns in Syriac with which possessives occur. These need to be considered separately and their tendency to carry a possessive likewise independently evaluated. To help in this evaluation we should note that the level of uncertainty in the use of a Syriac witness to attest the presence or absence of a possessive increases as the proportion of cases in that witness showing the same reading without Greek support increases. For example, if Syriac is prone to add possessives with particular types of nouns then we expect possessives to occur with those nouns where all Greek witnesses concur in attesting no possessive. The observation that Syriac NT translations can add possessives is not new. Back in 1850 Wichelhaus gave a number of examples of where P adds possessives with terms of address.43 Baethgen, considering C, suggests situations in which it added suffixes.44 Lyon believes that the OS should generally ‘not be cited for … the presence or absence of Greek possessives, articles, and demonstratives’.45 43 Ioannes Wichelhaus, De Novi Testamenti Versione Syriaca

Antiqua quam Peschitho Vocant, 252–53. 44 Baethgen, Evangelienfragmente, 21: ‘Bekannt ist, wie beliebt bei P die Ergänzung des Possessivpronomens ist; bei Sc [C] ist diese Erscheinung noch weit häufiger…’ 45 Lyon, Syriac Gospel Translations, 194.

POSSESSIVE PRONOUNS

69

However, Lyon is more positive about using P as a witness to Greek possessives: When P corrects toward the Greek it represents a fairly reliable and independent witness to its Greek text, often even in the use of particles, conjunctions, possessives, etc.46

Further observations have been made about particular situations, some of which are cited in the more detailed discussion below. However, it is fair to say that in general there has been little investigation of specific situations in which there is a tendency for Syriac to add or omit possessives. The variation between Greek and Syriac witnesses on the presence or absence of possessives is considerable, and the discussion below represents an attempt to bring order to this area. This treatment is, however, highly provisional, and there remains much work to be done, particularly on the theoretical framework in which tendencies to use or omit possessives should be considered. 2.3.1 Inherent Possession Previous researchers have especially noted variation in the presence or absence of possessives with body parts. Brock says, ‘Syriac idiom in fact virtually demands the use of the suffix with, for example, words denoting parts of the body…’47 He illustrates this with reference to ÷åßñ in Matthew’s Gospel. Lyon concurs:

46 Lyon, Syriac Gospel Translations, 196; the original quotation

is in italics. 47 Brock, ‘Limitations of Syriac’, 95.

70

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

…in Syriac, unlike Greek, it is more common to say ‘He laid his hand’, or ‘he washed his face,’ while in Greek, the article can alone suffice to show possession. L [S] always uses a possessive with hand in the above two idioms, whether the Greek does or not.48

It is important, however, not to let the focus of discussion be too narrow. The reason for difference in the treatment of body parts is most probably related to a wider issue. In recent years linguists have increasingly been interested in a category of possession sometimes known as ‘inherent’ or ‘inalienable’.49 According to Waltke and O’Connor, ‘A genitive of inalienable possession refers to something intrinsically proper to its possessor, chiefly body parts.’50 The essential thing with this type of possession is that possession of the possessed is a normal part of the possessor’s being. Such things include body parts (if the possessor is corporeal), but also facilities (‘voice’, ‘power’) and existential attributes (‘life’, ‘name’). Languages express such possession in different ways. French needs no explicit possessive marker in Je me brosse les dents, it being clear whose teeth are referred to. Likewise Greek with many constructions with body parts has no possessive, particularly when the possessor– possessed relationship is direct and unambiguous. Syriac, on the other hand, generally uses an explicit possessive in such cases. It is possible to view the addition of possessives in contrast to its Vorlage as an oddity of Syriac. However, 48 Lyon, Syriac Gospel Translations, 81. 49 For an application of this to Semitic see W. Diem,

‘Alienable und inalienable Possession im Semitischen’, 227–91. 50 B.K. Waltke and M. O’Connor, An Introduction to Biblical Hebrew Syntax, 145.

POSSESSIVE PRONOUNS

71

an alternative way of viewing things is to regard the omission in Greek texts of an explicit possessive as a less straightforward realisation in the surface structure of a language of the deeper semantic relationships expressed. The term ‘inalienable’ in relationship to possessions inherent in the possessor’s present existence is not always the one chosen by linguists. There is in fact a cluster of types of possession associated with terms such as ‘inalienable’, ‘inherent’, and ‘intrinsic’ that in some languages are marked differently from other forms of possession.51 It is also accepted that alienability and inalienability are on a sliding scale and are not absolute categories.52 I have chosen the term ‘inherent’ as seeming most suited to the Syriac data. A second related feature that seems to have a role in deciding whether a suffix should be used is the prominence or level of individuation of the possessor in the text. In some cases the possessor has receded so far into the background that although one knows that the possessed belongs to someone the relationship of possession is not clearly in view. In this case, even if the noun is a body part, Syriac texts tend not to give it a possessive suffix. This will be illustrated below in the discussion of possessives with the word ‘hand’.

51 This may be seen in some cases as a distinction between

‘active’ and ‘passive’ possession. In Hawaiian, for instance, there is a strong syntactic opposition between ‘inherited’ possession and ‘acquired’ possession (Anatole V. Lyovin, An Introduction to the Languages of the World, 262). 52 See, e.g., Diem, ‘Alienable und inalienable’, 282.

72

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

It is the recognition of the importance of the semantic relationship in focus that will help our consideration to embrace further examples. Body parts are the most frequent entities in this category to which the Syriac adds a possessive, but the same addition may also happen with a noun like ‘voice’, which is an inherent possession. Thus in 19 Luke 17:13 Greek öùíÞ ‘voice’ is translated by ‘their voice’ in SCP with no evidence of a possessive in Greek witnesses, and the generally anarthrous form of the Greek making it even less likely that there ever was a Greek text with a possessive. Now it is important not to extrapolate from this addition of a possessive to suppose that a possessive will generally be added with this or any similar noun. The addition of a possessive is very contextually dependent. Generally ‘voice’ in the Gospels does not receive the addition of a possessive. Usually when someone cries out ‘in a loud voice’ no possessive is added in SC,53 because the reference is to the quality of sound produced (and the quality is not inherently possessed) rather than to the individual’s voice (which is inherently possessed). The interplay of these considerations with other variations can be seen in Luke 11:27 where (ignoring variations in word order) we have dðÜñáóÜ ôéò öùíxí ãõíx dê ôï™ –÷ëïõ åqðåí ášô² which is rendered by SC

/ &

"

1%

3

6

502

/$(&

‘a certain woman from the midst of the crowd said to him in a loud voice’. The addition of the Syriac word ‘loud’ changes the voice from an inherent possession to a quality, 53 Matthew 27:46 S, 27:50 S, Mark 1:26 S, 15:34 S, 15:37 S,

Luke 4:33 S, 8:28 SC, 17:15 SC, 19:37 SC, 23:23 SC, 23:46 SC, John 11:43 S.

POSSESSIVE PRONOUNS

73

and so no possessive is present in SC. The Peshitta, however, renders the construction more literally without the word ‘loud’ and therefore adds a possessive: $ "& / & 3 19 502 /$(& ‘a certain woman from the crowd raised her voice and said to him’. This example is to give a caveat to Lyon’s already quoted statement that When P corrects toward the Greek it represents a fairly reliable and independent witness to its Greek text, often even in the use of particles, conjunctions, possessives, etc.54

In Luke 11:27, P is in general more literal than OS, but in its use of the possessive it is less close to the Greek, which, ironically, is caused by its generally more literal rendering. Now it is not correcting towards the Greek text precisely as Lyon envisaged (he is referring specifically to correction with regard to the possessive), but the principle should not be ignored that the semantic context needs to be considered before making any judgement about likely Vorlage. This departure of P from the Greek with regard to a possessive occurs where there are no variants in the Greek witnesses. Nevertheless, it is the sort of pattern that could occur alongside independent variation in Greek witnesses and be mistaken for an imitation of the Greek by P. 2.3.1.1 Hands In general a specific, literal hand with a clear possessor will receive a suffix in OS and P even if there is none in the Greek. This occurs whether ‘hand’ is in the singular or plural. The tendency is sufficiently marked that it is 54 Lyon, Syriac Gospel Translations, 196.

74

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

possible, making only occasional allowance for textual peculiarities, to pick up a Greek concordance and search on ÷åßñ and, ignoring whether a possessive is present in the Greek, to predict successfully which occurrences will have a possessive in SCP. This applies not only to the word ‘hand’; rather, the expectation of a possessive marker can be applied to any word that is inherently possessed. The ability to predict with a high degree of accuracy the Syriac readings from a Greek concordance shows that we are here dealing with a matter where the presence or absence of a possessive in the Vorlage only makes a difference if its presence or absence alters the semantic relationship between possessor and possessed. In the following examples Syriac witnesses add possessives in contrast to Greek witnesses in our control editions, which have ÷åßñ without a possessive (occasional witnesses with a possessive are marked in brackets). ‘Hand’ in singular. + 3 m.sg. Matthew 8:3 SCP, 12:10 SCP, 14:31 SCP, 26:23 SP, 26:51 SP, Mark 3:1 SP, 3:3 SP, 7:32 S,55 Luke 5:13 SP, 6:8 P. ‘Hand’ in plural. + 3 m.sg. Matthew 15:20 SCP, 22:13 SCP, 27:24 SP [with 1093], Mark 6:5 S, 10:16 S [with l36], Luke 13:13 SC, 24:40 P [with 1200], John 11:44 SP, 13:3 SP; + 3 m.pl. Matthew 4:6 P, 26:50 P (for S see p. 78),56 Mark 7:2 SP, 7:3 SP, 7:5 SP, 16:18a C, 16:18b CP, Luke 4:11 S.

55 Some texts read a plural here, but still lack a possessive. 56 Though ms 28 has the possessive here, its reading is in

other ways different from that of P, and therefore unlikely to have been its Vorlage.

POSSESSIVE PRONOUNS

75

In addition to these straightforward cases there are also some cases where the possessives follow the same principles, but other factors make the equivalents initially less intuitive. In Matthew 4:6 SC and Luke 4:11 P the angels are #* # ‘on their arms’ said to be ready to carry rather than dðr ÷åéñ§í ‘on [their] hands’, an alteration which still leaves the semantic relationship of inherent possession intact; hence the Syriac suffix. The most striking transformation is that from the plural of ÷åßñ without any possessive to the singular of 50 & ‘hand’ with a 3 m.sg. possessive. This happens in eight passages (Matthew 19:13 SCP, 19:15 SCP, Mark 6:5 P, 8:23b SP, 8:25 SP, 10:16 P, Luke 4:40 SP, 13:13 P),57 7, and the Greek on every occasion the Syriac verb is verb is dðéôßèçìé or simple ôßèçìé. The 3 m.sg. suffix is simply a feature of the relationship of possession. The change from Greek plural to Syriac singular is already well attested in OS, though in several of the cases listed above (Mark 6:5 S, 10:16 S, Luke 13:13 SC) OS witnesses have a plural in contrast to P’s singular. In this instance, since it always contrasts with the Greek, P is generally less literal than OS, though more internally consistent. The singular may well reflect later ecclesiastical expression, a hypothesis supported by the plural to singular transformations attested in P in Acts 8:18, 1 Timothy 4:14, 2 Timothy 1:6, and Hebrews 6:2. Moreover, P’s use of the singular 50 &

57 There are minor Greek variants in Matthew 19:15, Mark

8:23b, 8:25, and Luke 4:40 where manuscripts have a possessive with the plural of ÷åßñ. Cf. also Mark 5:23 SP.

76

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

more consistently than OS may represent ecclesiastical practice as well as expression. On the hypothesis that the use of the plural is the earlier construction the occurrence of the plural in CP in the longer ending of Mark (16:18) needs to be explained. Most probably it should not be considered alongside the rest since it is the only case when reference is to a plurality of persons laying on hands. ‘Hands’ naturally occurs in the plural when they belong to a plurality of people. From Mark 10:13 S and Luke 18:15 S the singular ‘hand’ seems to be the preferred Syriac construction, since in these cases there is no Greek equivalent. A minority of occasions shows the plural in Syriac (always with a possessive): Mark 6:5 S, 10:16 S, 16:18 CP, Luke 13:13 SC. In some cases Syriac witnesses lack a possessive with ‘hand’. For instance, in Matthew 18:8b SCP and its parallel Mark 9:43b SP Jesus says that it is better to attain life than /*/ ‘two hands’ and perish. The lack of to have 0 + & possessive parallels the Greek, but it may also be observed that no close semantic relationship is in view. General hands and their notional possessors are mentioned rather than specific hands and specific possessors. Thus a possessive is not needed. It is the lack of specificity about ‘hands’ and their possessors which seems to cause Syriac witnesses to drop possessives found in the Greek. If we simply suppose that the word ‘hand’ tends to get a possessive added to it in Syriac, whether or not it has one in Greek, it will be hard to explain the converse, where a possessive is present in Greek but not in Syriac.

POSSESSIVE PRONOUNS

77

Such is the case in Luke 21:12 where dðéâáëï™óéí döE ( ›ìOò ôNò ÷åsñáò ášô§í is rendered by S, C, and P by +0 & 1# ‘they will cast hands upon you’, without any possessive. The hands here receive no possessive because they are so general and unspecified (or unindividuated).58 This is also the explanation for the lack of a possessive with ‘hands’ in Mark 14:46 SP. There NA27 cites SP in support of txt’s omission of ášô§í. The lack of a possessive is a regular feature of the Syriac construction of " with 0 + & as seen in Luke 20:19 SCP, John 7:30 SCP, 7:44 SCP. This is not merely the feature of a particular expression. As will be seen below, the expression 58 Entities in the text are more individuated or salient

according to a number of different features, for instance, as they are specific rather than general, represented by a proper noun rather than a common one, or as they are represented by a full noun as opposed to a pronoun. For a fuller discussion of this term see G.A. Khan, ‘Object Markers and Agreement Pronouns in Semitic Languages’, 469–73, and the abbreviation of Khan’s treatment in P.J. Williams, Studies in the Syntax of the Peshitta of 1 Kings, 52–53. Burkitt, Evangelion Da-Mepharreshe, II, 47 n. 2, notes the general absence of the possessive with the phrase ‘lay hands upon’ in OS. Though Burkitt explains this construction as a case of ‘fixity of Syriac idiom’ it is more likely that the presence or absence of the possessive is related to the level of individuation involved in the reference to hands. Luke 21:12 is a highly unindividuated phrase with a general subject ‘they’, who occupy no specific time or location. The possessive is present in a similar phrase in the Peshitta of Luke 22:53 when the nominal individuation of Jesus is used and literal hands actually on a person are more clearly in view. Thus the less individuated text of S ‘they cast hands upon him and seized Jesus’ is to be contrasted with P’s more individuated phrase ‘they cast their hands on Jesus and seized him’. Cf. p. 78 on Matthew 26:50.

78

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

can occur with a possessive. Rather this is a feature which shows up in similar expressions because they, likewise, refer to unspecified ‘hands’. Thus, in Luke 22:53, SC have 0 + & 1# $ ) ‘you were not $ ) & stretching out hands against me’ and P +0 & 1# ‘you did not stretch out hands against me’. In Matthew 26:50 there is a division of Syriac witnesses: txt dðÝâáëïí ôNò ÷åsñáò dðr ô’í EÉçóï™í êár dêñÜôçóáí ášôüí 02& 0& 1# "& S ‘they cast hands upon him and seized Jesus’ 02& # 0+ & "& P ‘they cast their hands upon Jesus and seized him’ Here in contrast to examples elsewhere P, though not S, has a possessive suffix on ‘hands’. However, in this case alone of all the examples the hands receive specific focus. The person on whom the hands are thrown is highly individuated, being represented by a proper name within the clause itself. Whereas in S a general casting on of hands is mentioned, and the person affected is only specifically nominalised in the following clause, P portrays more individuated hands being laid on a specified patient. That P agrees superficially with a minor aspect of a rather different reading in ms 28 is likely to be coincidental. In the light of the preceding discussion showing ways in which Syriac texts formally contrast with extant Greek ones in the use of possessives, it seems unjustifiable to use the Syriac to attest the presence or absence of a possessive in the Greek where there is a division among Greek witnesses, especially when the reading for which the Syriac is attested merely conforms to its usual patterns of usage.

POSSESSIVE PRONOUNS

79

However, such citations occur on several occasions in NA27: Matthew 15:2: SCP attest ášô§í after ôNò ÷åsñáò. Mark 1:31: P attests ášôyò after ôyò ÷åéñüò. Mark 3:5: SP attest óïõ after ôxí ÷åsñá. Luke 9:62: SCP attest ášôï™ after ôxí ÷åsñá.59 In each case the hand or hands are specific and the possessor is in view. It is to be recommended that the Syriac witnesses are no longer cited for these variants in future editions. There are also the following variants which need more discussion. Matthew 12:1. NA27 cites C (with Latin manuscript c in parentheses) in support of the reconstructed text êár ôásò ÷åñórí ášô§í øþ÷åéí which NA27 notes comes from a parallel passage (presumably Luke 6:1). However, in Luke 6:1 the evidence for the possessive ášô§í is slight. If we are therefore to suppose that C’s assimilation goes back to a Greek Vorlage—and there is no strong reason to suppose that it did—it is more probable that it went back to one without the possessive of NA27’s retroversion. The possessive in Syriac is only what one would expect (see Luke 6:1 P). Mark 5:23. Where txt has ëÝãùí … líá dëè¦í dðéè†ò ôNò ÷åsñáò ášô†, NA27 cites (SP) in support of D’s reading åëèå 1# 7 / áøáé áõôçò åê ôùí ÷åéñùí óïõ. S reads 7 / J0 & ‘come put upon her your hand’ and P 1# J0 & with the same meaning. SP contrast with txt 59 Although the actual variant for which SCP are cited is much

longer, the presence of the possessive is the only difference between that variant and txt. In UBS4, SCP are also cited as supporting the possessive.

80

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

and D in having ‘hand’ in the singular, but as we have seen this is quite normal in Syriac for the imposition of hands. SP agree formally with D over against txt in having a possessive, but according to principles established above this would be expected irrespective of the presence of a possessive in their Vorlagen. SP appear to agree with D in having an imperative, but as txt’s líá dëè¦í dðéè†ò is an imperatival construction its most natural translation in Syriac would be by two imperatives.60 EÅëèþí is an imperatival participle and cannot therefore be rendered by a Syriac participle. Thus the Syriac use of singular, a possessive and the imperatives do not allow us to decide which Greek Vorlage SP had. Ironically, however, the 7 followed by # is vocabulary choice of SP with somewhat closer to the vocabulary of txt than it is to that of D, as a concordance search on Rðôïìáé and dðéôßèçìé and their Syriac equivalents will quickly establish. The Syriac texts thus may give support to the opposite reading to that for which they are cited, though it is probably wisest to drop reference to them altogether from the apparatus. John 20:20. txt reads häåéîåí ôNò ÷åsñáò êár ôxí ðëåõñNí 7 0+ & (& 2 ‘he ášôïsò, and SP read showed them his hands and his side’. NA27 cites SP for two variants: first the insertion of ášôïsò after häåéîåí, and secondly, the substitution of ášôï™ for txt’s ášôïsò at the end. These variants are not distinct. Since the ‘hands’ and ‘side’ are specific and individuated we expect SP to have a possessive on both. Moreover, given that Syriac rules for 60 A.-G. Martin, ‘Le palimpseste syriaque du Sinaï et le codex

de Bèze’, in D.C. Parker and C.-B. Amphoux (eds), Codex Bezae: Studies from the Lunel Colloquium, June 1994, 250.

POSSESSIVE PRONOUNS

81

word order would mean that in SP’s (& 2 ‘he (& ‘them’ has to appear showed them’ the pronoun 61 SP cannot be cited for immediately after the verb, placing ášôïsò earlier rather than later in the sentence. SP’s reading could derive from the variants NA27 suggests, but is also a predictable rendering of txt. 2.3.1.2 Other Inherent Possessions What has been worked out in detail for the word ‘hand’ is briefly here applied to other similar nouns. Most, but not all, of these are body parts. We may include also parts of animals, e.g. ‘wings’, or of plants, e.g. ‘leaves’, ‘roots’. Below is a list of suffixes attached to words in contrast to Greek manuscripts in our control editions (exceptions in brackets). Words are listed according to English headings in order to stress that their meaning is the significant feature attracting explicit presence of a possessive. In general the Greek and Syriac equivalents are regular, and it can be assumed unless stated that Syriac retains the number of the Greek word. Unless the textual situation has required citation, additions of possessives have only been given in abbreviated form. The presence or absence of the Greek article has been ignored since in several cases it is impossible to be certain about its presence in the Vorlage of the Syriac. Belly. + 3 m.sg. Mark 7:19 P [È].

61 Object suffixes exist in Syriac for all persons except 3 m.pl.

and 3 f.pl. These functions are fulfilled by (& and (& respectively, which, like suffixes, must follow the verb immediately.

82

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

Cheek. + 2 m.sg. Luke 6:29 SP. The tendency for a possessive to be used is also attested by ‘his cheek’ in John 18:22 SP and ‘his cheeks’ in John 19:3 P, which have no direct Greek equivalent.62 Ear. + 3 m.pl. Matthew 13:15b SCP [)c 157]; ‘ear’ ‘ears’ + 2 m.pl. Matthew 10:27 SP. Eyes. + 3 m.pl. Matthew 13:15b SCP [251], John 12:40b SP. Face. + 3 m.sg. Luke 5:12 SP, 17:16 SCP [28 213 1424]; ‘face(s)’ + 3 f.pl. Luke 24:5 P [C*]. Feet. + 3 m.sg. Luke 24:40 P [1200], John 11:44 SP, 13:10 SP, 20:12 P. Head. + 1 sg. John 13:9 SP [157]; + 3 m.sg. Luke 9:58 SCP [l1127], John 19:30 P. Heart. + 1 sg. Matthew 11:29 SCP; + 3 m.sg. Mark 12:33 S; + 3 m.pl. Matthew 5:8 SCP, 13:15 SCP, John 12:40b P. In Mark 7:19, P has a 3 m.sg. possessive in agreement with D Ä 265. This Syro-Bezan agreement need not be genetic. Knees. + 3 m.sg. Luke 22:41 SCP [1424 l1056 l1642]. Leaves. + 3 f.sg. Matthew 24:32 SP. Left (i.e. left side or left hand). + 3 m.sg. Matthew 25:33 SP [)], 25:41 SP, 27:38 SP [l48], Luke 23:33 SCP. Lips. + 3 m.sg. Matthew 15:8 SCP, Mark 7:6 SP.

62 In John 19:3 the plural (dual) of the Syriac ‘cheeks’

corresponds to the plural ¼áðßóìáôá, whereas in John 18:22 the Syriac singular ‘cheek’ corresponds to the single ¼Üðéóìá. Evidently a single blow could not affect both cheeks, and multiple blows would be likely to be applied to more than one cheek.

POSSESSIVE PRONOUNS

83

Loins. In John 13:4 SP, 13:5 SP, 21:7 SP, :2+ has a 3 m.sg. suffix, and in 21:18 SP (2×) a 2 m.sg. suffix. The Greek has no specific reference to loins. in S or Name. In John 10:3 êáôE –íïìá becomes + in P. Power. + 3 m.sg. Mark 12:33 S. Right (i.e. right side or right hand). + 3 m.sg. Matthew 27:38 SP, Mark 15:27 SP, Luke 23:33 SCP. %# in Mark 11:20 SP. Root(s). dê ¼éæ§í becomes Soul. In Mark 12:33, S has the unusual order ‘heart … soul … power’, each word being suffixed. S’s addition of a 3 m.sg suffix to ‘soul’ formally agrees with D’s øõ÷çò áõôïõ, but this is not necessarily related since D has the order ‘heart … power … soul’. The addition of a 3 m.sg. suffix is 2 ( ‘that one natural in Syriac after the 3 m.sg. should love him’ for the Greek’s ô’ PãáðOí ášôüí. Spirit. + 3 m.sg. Matthew 27:50 SP, John 11:33 SP, 13:21 SP, 19:30 P; + 3 m.pl. Matthew 5:3 SC so that Jesus blesses those ‘poor in their spirit’ not simply ïj ðôù÷ïr ô² ðíåýìáôé. "& Voice. In Luke 17:13 ƒñáí öùíÞí is translated 19 ‘they raised their voice’ in SCP. See the discussion above on pp. 73–74. Wings. In Luke 13:34 ›ð’ ôNò ðôÝñõãáò [213 472 1009 =3 $ 2/ in S and 1079 1424 l48 have ášôyò] becomes + =6 $ 2/ in CP, both meaning ‘under her wings’. Just as possessives are added with body parts and inherent possessions when they are linked with specific individuals, so possessives are absent in Syriac when specific individual possessors are absent. For instance,

84

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

‘mouth’ Matthew 15:11 SCP (2×), 15:17 SCP, 15:18 SCP, or ‘belly’ Matthew 15:17 SCP. In the light of the argument above we have to question some more of NA27’s citations. In particular the following inherent possessions have possessives claimed for them: Matthew 13:15: SCP attest ášô§í after ôïsò ¨óßí. Matthew 14:12: SCP attest ášôï™ after ðô§ìá. Matthew 20:21: SCP attest óïõ after äåîé§í. Matthew 20:23: SCP attest ìïõ after åšùíýìùí. Luke 2:28: SP attest ášôï™ after ôNò PãêÜëáò. Luke 6:45: SP attest ášôï™ after ôyò êáñäßáò. Luke 10:11: SCP attest ½ì§í after ðüäáò. John 6:52: SCP attest ášôï™ after ôxí óÜñêá.63 In Mark 15:45, txt reads ô’ ðô§ìá, many mss read ô’ ó§ìá, but D reads ô’ ðô§ìá ášôï™ for which NA27 cites S with 01) ‘his corpse’ as support. P has D- ‘his body’, with the possessive in contrast to the Greek witnesses with ó§ìá, which suggests that the possessive is idiomatic to Syriac. If the same applied to S’s reading then its agreement with D might be non-genetic, and NA27’s note would be misleading. At any rate, both ‘body’ and ‘corpse’ are inherent possessions in the context, and the possessor is prominent in the narrative. For the same reason we may question NA27’s citation of P for ›ì§í after ô² óþìáôé in Luke 12:22.64 63 This citation is also in UBS4. The representation in SCP of

Greek óÜñî glossed as ‘flesh’ by Syriac 5 D- glossed as ‘body’ is quite normal (e.g. John 1:14 C). 64 SC lack the possessive on ‘body’, and S lacks it earlier in the verse with ‘soul’. SC’s construction with d lath allows focus to be

POSSESSIVE PRONOUNS

85

John 20:12 provides a clear example of the forces influencing the absence or presence of a suffix. Mary sees two angels, one ðñ’ò ô† êåöáë† and one ðñ’ò ôïsò ðïóßí. These two terms of location are taken by P to be related to the body of Jesus and both terms are accordingly trans+ 07& ‘at the location of his lated with a suffix, i.e. 16* ‘at his feet’, respectively. The head’ and complicating factor is, of course, that the body is not there. This may explain the choice of the word 507& as ". S, on the other hand, translates the two opposed to + ‘at the location terms without use of possessives 507& ‘at the location of feet’. The of the head’ and 16*& absence of the possessive on these terms is necessitated by the fact that in S they are related not to a person or body 02 (i.e. an inherent possessor), but to a place. S reads

02 16*&

"

$

+ 507&

‘one at the head of that place where Jesus had lain, and one at the feet’. Moreover, the prosthetic laph ‘place of feet’, clearly shows that S is using the word ‘foot’.65 S, therefore, in that it lacks the possessive not suffixes is closer to the Greek than P, but this agreement with the Greek has nothing to do with Vorlage. Ironically, the closeness to the Greek in this respect is caused by S’s divergence from the Greek in relating the terms ‘head’ on the qualities ‘of the soul’ and ‘of the body’ rather than the inherent possessions ‘soul’ and ‘body’, and the absence of the possessive is therefore consistent with the character of the language. On the other hand, the more abstract expression in S might be taken as reflecting the lack of a possessive in Greek (which the translator would be likely to take as abstract). 65 The identification of the lexeme is confirmed by a citation from Bar Ali in R. Payne Smith, Thesaurus Syriacus, column 3811.

86

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

and ‘foot’ to their location rather than to the (absent) body. Clearly the constructions in both S and P are to be understood primarily on grounds internal to Syriac, rather than by reference to the Greek. John 11:41 illustrates a further problem. Whereas P is cited by NA27 merely as adding ášôï™ after ôï˜ò “öèáëìïýò, an addition that is necessitated by the inherent possession ‘eyes’, S is cited as adding ášôï™ åkò ô’í ïšñáíüí along with K and some other witnesses. This variant has two parts, the addition of ‘into heaven’ and the addition of the possessive. However, to save space NA27 has assimilated the reading of K, which lacks the possessive, to the text that S is believed to support. From a consultation of Swanson ad loc. it is not clear that there are any Greek texts that have both the possessive and the phrase ‘into heaven’. Despite all these caveats about the use of Syriac as a witness to the presence or absence of a possessive, we should not conclude that Syriac witnesses can never be used for this purpose. In fact the general observation of addition of possessives with inherent possessions can give added certainty to citations in the apparatus of NA27. Mark 9:41 contains the variant where SP are cited in support of txt’s dí “íüìáôé as opposed to dí “íüìáôß ìïõ. Someone’s name is inherently possessed, so the absence of a possessive in the Syriac can only be explained on the basis of a Vorlage without a possessive, because ‘name’ in a Vorlage without a possessive would not be an inherent possession. This example contrasts with many Greek variants to do with body parts where the presence or absence of a possessive does not distinguish meaning.

POSSESSIVE PRONOUNS

87

One case where, though it may be slightly misleading, the presence of a citation in a manual edition can be justified on grounds of economy of space is Mark 14:64 where, for txt’s zêïýóáôå ôyò âëáóöçìßáò, P is cited as supporting zêïýóáôå ôxí âëáóöçìßáí ôï™ óôüìáôïò ášôï™. The wider difference between the phrases supports NA27 in citing P, and even though D f 1 2542s pc have the longer text without ášôï™, the point of NA27’s citation of P is clearly to support the longer text with less concern about the presence of the possessive. To conclude, inherent possession is a semantic relationship and as such it is not intrinsic to particular nouns, though it frequently characterises the relationship between particular nouns and specific possessors. In judging the witness of a Syriac text to the presence or absence of a possessive in its Vorlage it is necessary first to understand the semantic relationships in both the variant Greek texts and in the extant Syriac one. Because Greek habitually omits an explicit marker of the possessive when possession is inherent and Syriac includes one it will often be found that the semantic equivalent of the Greek’s omission of a possessive will be the Syriac’s inclusion of one. If it is found that different semantic relationships are marked by the inclusion or omission of a possessive in the Greek, one is in a much better position to use a Syriac translation as a witness to its Vorlage. 2.3.2 Relational Terms Another set of terms that need to be considered separately are relational terms, most of which are also kinship terms. In fact the type of possession involved in these terms is not

88

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

radically different from the inherent possession dealt with above. Possession of family members is not an intrinsic part of existence. However, possession of certain relations is implicit in the use of kinship terms. ‘Father’ implies the existence, whether alive or not, of ‘child’, ‘son’ implies ‘father’ and ‘mother’, ‘sister’ implies ‘sibling’, and ‘wife’ implies ‘husband’. Beyond the family, ‘servant’ implies some sort of ‘master’ or ‘mistress’, and ‘friend’ tends to imply reciprocally ‘friend’ (though not necessarily a peer). Consequently, while aside from ‘parents’ no relationships are necessary for a particular human being, when that individual is denoted by a relational term they possess inherently the implied relation (this term being used for persons in relationships, familial or not). There is a clear tendency for relational terms to bear a possessive when they are not generic and the person to whom they are related is present implicitly or explicitly in the narrative. This tendency is much stronger with family terms than with a term like ‘servant’, where the relational aspect of the term is usually not in the foreground, but its functional aspects are. However, even kinship terms like ‘father’, ‘son’, and ‘wife’ can be used in a way that is primarily functional rather than relational. The narrative prominence of functional and relational aspects of nouns is significant in explaining the distribution of suffixes. We will first consider the term ‘father’ in detail, before reviewing the other terms more summarily.

POSSESSIVE PRONOUNS

89

2.3.2.1 Father66 This term can of course refer both to God, the Divine Father, and to human parents. In many cases the same rules apply to both, though some features of reference to God as Father need to be considered at a level beyond the syntactic. As a relational term ‘father’ very frequently receives a possessive. How characteristic it is for Syriac to use the form & to represent ðÜôåñ can be judged from the fact that the hyper-literal Harclean version can use this equivalent (e.g. Luke 23:34), to be distinguished from ‘my & (e.g. John 20:17). Even the father’ which is 1 &, which in Syriac lacks the possessive, needs to form be handled with caution. )b@f)a (NT áââá) was of course the Jewish Palestinian Aramaic form for ‘my father’,67 and given the prominence of this term in early Christian & discourse it is possible that, for instance, the use of to represent ðÜôåñ (e.g. Matthew 11:25 SC, John 11:41 SP) is not to be taken as telling us something about Syriac grammatical preferences, but rather is a Palestinian Aramaic element reflected in an early stratum of Syriac Christian discourse. & and In considering the relationship between & there is a mass of data and a level of fluctuation in SCP that has not yet received an explanation. This means that the use of the Syriac evidence is more open to debate than is the case with entities that are inherently possessed. NA27 cites Syriac witnesses as supporting the presence of 66 For another discussion of possessives with ‘father’ see

Burkitt, Evangelion Da-Mepharreshe, II, 47. 67 E.g. Genesis 44:32 in Targum Onkelos.

90

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

the 1 sg. possessive ìïõ with ‘father’ in a number of cases.68 Of these the most convincing cases are those where other Syriac witnesses have no possessive, e.g. John &. It is possible that P represents 6:65, where SC have a correction according to its Greek text, but the text of SC seems best explained by a Greek Vorlage without a possessive. Likewise in John 16:16 the variant is the addition of ”ôé ›ðÜãù ðñ’ò ô’í ðáôÝñá supported by SP, to which S is said to add ìïõ. In these cases the contrast between the witnesses seems to give a reason to think that the Syriac can be relied on to supply information about possessives in its Vorlage. However, in John 16:17, the verse immediately following the second example, SP have an identical phrase to that in 16:16 with the possessive on ‘father’ despite its absence from Greek witnesses (except ms 472). This then brings the citation of S for the possessive in 16:16 into question. Furthermore, distinction in witnesses can be present even if there is none in their Vorlage, e.g. John 14:9a, where S has a possessive in contrast with P, and distinction between occurrences in a verse can occur in a version without distinction in the Greek. Thus there is no correspondence between the occurrences of ‘father’ in the Greek of John 14:9–10 and the presence or absence of a possessive in the translation of S. John 8:18 is also instructive. A possessive occurs in CP despite the fact that if a possessive were in the Greek it would be ungrammatical. On several occasions NA27 cites Syriac witnesses for the presence of ìïõ with Johannine occurrences of arthrous ðáôÞñ (John 6:65 P [contra SC], 8:38 SP, 16:10 SP, 68 After arthrous ðáôÞñ: John 6:57 S, 6:65 P, 8:38a SP (also

cited in UBS4), 16:10 SP, 16:16 S, 18:11 SP, 20:17 SP.

91

POSSESSIVE PRONOUNS

16:16 S, 18:11 SP, 20:17a SP). Though these citations are all possible, it is not certain that when Jesus is the speaker (and therefore a 1 sg. possessive is appropriate) Syriac witnesses do not show a slight preference for & over &, whatever their Vorlage. Below are the examples from John 14 of equivalents of arthrous Greek ðáôÞñ in Jesus’ speech without a possessive in txt. verse

& & verse

& &

verse

& & verse

& &

6 SP

9a S P

9b SP

10a P S

10b SP

10c CP

11a CP

11b CP

12 SPmss69 P

13

24

SP

16 SCP

26 SCP70

28a SCP71

31a SP

31b SP72

P

Even P, which has more examples of & than the other &. In several witnesses, still shows a preference for places & occurs in a Syriac witness though there is no Greek evidence for the presence of a possessive (verses 6, 9a, 10abc, 11ab, 16, 31a), and in other cases the Greek evidence for a possessive is weak (verses 28, 31b). This suggests that the presence of the possessive is translational. 69 A significant number of Greek witnesses have the pos-

sessive. Pwolf along with Syriac mss 4 9 23 36 37 read 70 D È Ðc 118 have the possessive. 71 G f 13 157 have the possessive. 72 473 has the possessive.

&.

92

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

The agreement between SCP and D in John 14:26 as to the presence of a possessive therefore need not be genetic. Some of the occurrences without a possessive in SP can be explained on Syriac-internal grounds rather than with primary reference to the Greek. John 14:9b, though quoting Jesus’ speech is actually quoting him quoting Philip. The 1 sg. possessive is therefore inappropriate. If the information of the above table is brought to bear on the variant in 14:28b where NA27 cites P as supporting the presence of a possessive, it is seen how problematic this citation is. Why should P support a possessive in 14:28b when it attests the possessive in contrast to the Greek in eight other places within the same chapter? There remains research to be done on the Syriac witnesses, in particular to see to what extent and why Syriac witnesses diverge, and to investigate differences between vocative forms of ‘my father’ and other forms. However, it is not only the first person suffix which is frequently added to the word ‘father’. It seems that the main candidates for a Syriac addition of a possessive are those cases where a specific relationship is implied. As with items inherently possessed we could argue that the oddity in linguistic coding is not the addition of possessives in Syriac, but rather the Greek’s omission of possessives or use of possessives which do duty for more than one noun. Thus it hardly seems correct for NA27 to cite SCP as twice supporting the addition of óïõ in ôßìá ô’í ðáôÝñá êár ôxí ìçôÝñá (Matthew 15:4), nor to cite SCP for the addition of óïõ after ôßìá ô’í ðáôÝñá óïõ êár ôxí ìçôÝñá (Luke 18:20). Though it is entirely possible that the Vorlagen of the Syriac witnesses had possessives in one or more cases, this

POSSESSIVE PRONOUNS

93

is hardly certain. The Syriac can no more omit the possessives than could the original commandment run: *M)'wF b)f db@'k@a ‘honour father and mother’. Semitic languages are also much less inclined than Greek to have double duty possessives, a fact no doubt related to the tendency of possessives to be suffixed rather than independent.73 In addition, Syriac witnesses contain possessives with ‘father’ and ‘mother’ in other statements of the fifth commandment and where possessives in the Vorlage are unlikely, e.g. Matthew 19:19 SCP. Greek [¿] ðáôÞñ without a possessive receives a 3 m.sg. possessive in Syriac in John 1:18 CP, 5:19 SC, and 13:1 SP without Greek variants, and with only minor Greek support for the possessive in Luke 15:12 S [1675] and 15:22 SCP [131 1200 l1016]. This suggests that in Luke 9:26 where CP ‘agree’ with D 544 903 1574 1689 l10 in showing the possessive the Syro-Bezan agreement need not be genetic. The same addition of a 3 m.sg. possessive happens in Matthew 10:37 SC, 15:4b SCP, 15:5 SC [with à 1071], Mark 7:10 S, 7:11 SP [with D 50 68 90 483* 484] alongside the probable addition of a 3 m.sg. suffix to [½] ìÞôçñ, and Luke 12:53 SCP [with D]. Mark 7:11 could be another case of non-genetic agreement between D and the Syriac, as is almost certainly the case in Luke 12:53 where D also adds a verb, making it highly unlikely that its reading was the Vorlage of SCP. In John 7:22 dê ô§í +& ‘from ðáôÝñùí ‘from the fathers’ becomes your fathers’ in SC. NA27’s citation of SCP for the inclusion of ášôï™ after ‘father’ in Luke 15:29 may 73 See further Lyon, Syriac Gospel Translations, 100.

94

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

accordingly be questioned, as may perhaps a similar citation of CP in John 4:53. NA27 and UBS4 suggest that Syriac witnesses had a second person plural possessive in their Vorlage in John 6:58 and 8:38b. For the former case SC are said to add ›ì§í after ïj ðáôÝñåò, and P is said to add ›ì§í ô’ ìÜííá. This interpretation of the evidence is possible, particularly since if there was a process of addition of a possessive which took place in Syriac independent of the Greek the Syriac might as well have added a 1 pl. suffix as added a 2 m.pl. suffix. Likewise the supposition that the Vorlagen of SP contained a 2 m.pl. possessive after the second occurrence of ‘father’ in John 8:38 seems initially reasonable. However, given that in 8:38b in NA27 txt, whether or not the reading of txt is original, the definite article on ‘father’ also serves to mark possession, and the possessor is clearly contextually 2 m.pl., the text of the Syriac witnesses is also the semantic equivalent of txt, and thus could conceivably have been produced from it.74 Although the general pattern is to add a possessive with the term ‘father’, there are some cases where a possessive in the Greek is omitted in translation. & in It seems that ¿ ðáôÞñ ìïõ is translated in C by Matthew 10:32, John 6:32, and 14:21, where each time SP have &.75 In this last case C uses the active construction & 2 ( ‘the Father will love him’ where the 74 For a preliminary caution about citing Syriac witnesses to

attest possessives with, for example, ‘father’ see Lyon, Syriac Gospel Translations, 142. 75 See also John 15:8, where SP lack the possessive in agreement with ms 579. Cf. John 8:54 S.

POSSESSIVE PRONOUNS

95

Greek, followed by S, has a passive construction: PãáðçèÞóåôáé ›ð’ ôï™ ðáôñüò ìïõ. The substitution of an active for a passive is a characteristic of OS (see below pp. 184–89), and this suggests that C here, and by analogy in Matthew 10:32 and John 6:32, is preserving the original OS reading from which S presents a revision. & If the presence or absence of a suffix on depends considerably on the narrative prominence of the relationship, it may be possible to posit the omission of the possessive in the Syriac in contrast to its Vorlage. For instance, in John 8:54 txt’s hóôéí ¿ ðáôÞñ ìïõ ¿ äïîÜæùí ìå by virtue of being a statement of existence has less focus on the father–son relationship than many other cases of ¿ & $& ðáôÞñ ìïõ. It may therefore be that S’s ‘there is a father who glorifies me’ comes from ; txt rather than from the reading of W, the only Greek witness to lack ìïõ. The omission of the first person possessive in Matthew 6:4 S may be an assimilation to Matthew 6:18 (where OS is represented by C; cf. 6:15 C). 2.3.2.2 Other Relational Terms Similar principles apply to other relational terms. The Greek frequently omits explicit expression of the implied relationship using merely the definite article. In Syriac, relationships tend more often to be explicitly marked by a possessive suffix. If a term means ‘his brother’, ‘my mother’, etc., then the possessive is generally marked. Moreover, whereas Greek allows a single possessive in a list of kinship terms to cover them all, Syriac needs a possessive on each item. This is illustrated in Luke 14:26,

96

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

where all six family members listed have a possessive in SCP. By contrast almost all Greek witnesses have a possessive only with the first noun (and D with the first two). Other examples of the relationship between Syriac and Greek are listed below, with abbreviated reference to Syriac possessives added in contrast to Greek. Brother. + 3 m.sg. Matthew 10:21 SP, Mark 13:12 SP; ‘brothers’ + 2 m.pl. Luke 21:16 SCP. NA27’s notes that P attests ášôï™ after ïj Päåëöïß in John 2:12 and ášô§í after ôï™ Päåëöï™ in John 11:19 are hardly convincing. The former note agrees with UBS4 ad loc. Child. See on ‘Son’, pp. 99–100. Daughter. + 3 m.sg. Matthew 10:37 SC; èõãÜôçñ (used vocatively) or èýãáôåñ + 1 sg. in Matthew 9:22 SP, Mark 5:34 P, Luke 8:48 SCP. If ðáéäßïí can be appropriately translated by ‘daughter’ (with possessive already explained), then NA27’s citation of P in Mark 7:30 loses authority.76 Enemy. There is no concrete evidence that Syriac ; 01' , translations add possessives with the term though it is certainly used relationally. The English word ‘enemy’ may be a more consistently relational term than ; 01' , and the possibility that this latter term may denote a person possessed of some sort of relationshipindependent characteristic needs to be investigated. However, if ; 01' did imply a relationship UBS4’s

76 In this connection it should be noted that the discussion

under ‘Son/Child’ (pp. 99–100) shows the tendency to use gender-specific terms in translating Greek gender-neutral ones. Absolute vocative ôÝêíïí always becomes in SCP.

POSSESSIVE PRONOUNS

97

citation in Luke 1:74 that SP support ô§í d÷èñ§í ½ì§í rather than just d÷èñ§í might be called into question. Friend. + 1 sg. Luke 11:5 SCP (this might possibly motivate the omission of a 1 sg. possessive in translating ‘my friend’ in Luke 11:6 SP because ‘friend’ refers to another person; cf. also 11:7 P), 14:10 SCP; ‘friends’ + 1 sg. John 15:15 SP [with 954]; + 3 m.sg. Luke 7:6 SP; + 2 m.pl. Luke 21:16 SCP; ‘female friends’ + 3 f.sg. Luke 15:9 SCP. ‘my lord’ and ‘our lord’ are Master. Both frequently attested. In general the singular suffix is used is when a master is being addressed by an individual. used when a master is being addressed by a plurality of people, and also, particularly in S, corresponding to where Greek texts have [¿] EÉçóï™ò. The plural suffix may be used when a master is addressed by an individual who is part of a group to whom the address is also pertinent, e.g. John 14:8 SP. These criteria do not allow us to predict the Syriac form used in every case, though they do trace broad patterns in the distribution. Naturally the same text may be rendered differently by OS and P (e.g. John 13:36, and P , contrasting with John 14:5, where S has and P has ). Even when not 14:22 where SC have referring to Jesus, the term ‘master’ still may have possessives added to it: 2 m.pl. in John 13:14 SP; 3 m.pl. in Luke 12:37 SCP [with 903 l184]. This brings into question NA27’s citation in Matthew 24:45 of SP in favour of ášôï™ after ¿ êýñéïò. Mother. + 3 m.sg. John 19:26a P [with È Ù]; + 3 f.sg. Mark 5:40 P, Luke 8:51 P [with È 1047]; + 3 m.pl. Matthew 19:12 SCP [with 892 954 1293 1424 1574]. The

98

PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

presence of possessives with ‘mother’ in Matthew 15:4 SCP has already been discussed in the discussion of ‘father’ above. NA27 also suggests that P’s Vorlage had a 3 m.sg. possessive in John 19:26b, i.e. ášôï™ after txt’s ëÝãåé ô† ìçôñß. However, as it is likely that kä¦í ôxí ìçôÝñá earlier in . 2 the verse has already been translated by P as ‘he saw his mother’, P may well have added the possessive independently. In fact, given observed tendencies we could go further and insist that P had no option but to use a possessive suffix in both cases. Similarly, NA27 is misguided in suggesting that SP and SCP respectively attest óïõ after ôxí ìçôÝñá in Mark 10:19, Luke 18:20 (see p. 92 on Matthew 15:4). The variant NA27 records in Luke 2:33 is dealt with at greater length in UBS4. That cites S for ¿ ðáôxñ ášôï™ êár ½ ìÞôçñ ášôï™, and P for EÉùóxö êár ½ ìÞôçñ ášôï™ with txt reading ¿ ðáôxñ ášôï™ êár ½ ìÞôçñ. It is possible, however, that S was derived from txt since the possessive on ‘mother’ needed to be expressed in Syriac. Similarly it is just possible that P was derived from a text like N and 579 which read ¿ EÉùóxö êár ½ ìÞôçñ. The Translation Hypothesis as an explanation for the presence of many of these possessive suffixes on ‘mother’ gains further support from the presence of the suffix on references to someone’s mother, which are not paralleled in the Greek, e.g. John 9:1 SP, 9:32 S, 11:37 S. Mother-in-law. See pp. 101–102 on Luke 12:53. Neighbour. ‘The neighbours’ + 3 m.sg. Luke 15:6 SCP [with N]; ‘female neighbours’ + 3 f.sg. Luke 15:9 SCP. Given the relational status of the term ‘neighbour’, one may debate NA27’s citation of SP in support of ášôï™ after

POSSESSIVE PRONOUNS

99

ïj ï¤í ãåßôïíåò in John 9:8, and of óïõ after [ôï˜ò] ãåßôïíáò in Luke 14:12 SCP. Parents. + 3 m.sg. Luke 2:27 SP; + 3 m.pl. Matthew 10:21 SP, Mark 13:12 P; + 2 m.pl. Luke 21:16 P (SC read + ‘your people’ [cf. English ‘your folk’]). (& Relatives. + 2 m.pl. Luke 21:16 SCP. Servant. + 3 m.pl. Matthew 13:28 SCP. In Luke 20:10 and 20:11 the master sends äï™ëïí ‘a servant’. Since the possessor is active and prominent in the narrative this is 0;# ‘his readily represented both times in SCP by servant’. The same correspondence between äï™ëïí and 0;# occurs in the parallel passage in Mark 12:2 SP. In the light of the likely relational status of the Syriac term, NA27’s notes that Luke 14:23 SCP77 and John 4:51 CP attest ášôï™ after äï™ëïò are at best questionable. Siblings. According to NA27, SP attest ìïõ after Päåëöïß in Mark 3:33 and after PäåëöÞ in 3:35. However, these possessives are obligatory in Syriac. Son/Child. Frequently an implied 3 m.sg. possession in the Greek is made explicit in Syriac witnesses.78 Given these additions it is also possible that, contra NA27, SCP do not attest ášôï™ in their Vorlage after ô’í õjüí in John 3:16 and 3:17, and that, contra NA27 and UBS4, the 2 m.sg. possessive present in SP of John 17:1b does not reflect a possessive in the Vorlage. In John 8:35 ¿ õj’ò ìÝíåé åkò ô’í ák§íá ‘the son remains [in the house] indefinitely’ clearly 77 The possessive here is another apparent Syro-Bezan

agreement. 78 Matthew 10:21 SP, 10:37 SC, 21:38 SC [with 1071], Luke 12:53 SCP, 15:13 SCP, John 5:20 SCP, 5:22 SC, 14:13 SP. Cf. Luke 15:12 P.

100 PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

refers to the status of a son as opposed to a slave. No relationship is in view and therefore the extant Syriac witnesses, SP, do not have a possessive. It is likely that NA27’s citation of SCP for retaining ášôï™ after ¿ ðásò in Matthew 8:13 is legitimate. This is because ¿ ðásò could take on the significance of ‘the servant’ or ‘the lad’, i.e. there is a semantic difference in Greek between the presence and absence of a possessive. SCP most probably support the presence of a possessive. Vocative ôÝêíïí ‘my son’: Matthew 9:2 SP, 21:28 SCP, always becomes Mark 2:5 P [)* has ôÝêíïí ìïõ], Luke 2:48 SP, 15:31 SCP, 16:25 CP (S’s H is probably a secondary corruption from ). This is analogous to the way ðÜôåñ so frequently becomes &. The vocative plurals ôÝêíá or ôåêíßá become 3 + ‘my sons’ in Mark 10:24 SP and John 13:33 SP.79 Nonvocative ôÝêíïí receives a 3 m.sg. possessive in Mark 13:12 SP, and ôN ôÝêíá receives a 3 m.sg. possessive in Matthew 18:25 SCP [with 1396 1515 1579 1675]. Teacher. The terms Rabbi and Rabban show how readily the term ‘teacher’ was taken in Aramaic to be a term with implied relationships.80 Normally the Syriac versions use " when the teacher’s relationship with an " when the individual is contextually relevant, and teacher’s relationship with a plurality of people is contextually relevant (e.g. Matthew 26:18 SP with " vs. "). Naturally it is not only first person 8:19 SP with possessives that may be added: John 13:14 SP attests " with a 2 m.pl. suffix. Although in Matthew 10:24 NA27 cites SP in support of ášôï™ after ïšê hóôéí ìáèçôxò 79 In this latter case the possessive occurs in ms 579. 80 These remarks concern only the term ", not 3=1 .

POSSESSIVE PRONOUNS

101

›ðcñ ô’í äéäÜóêáëïí ‘a disciple is not above [his] teacher’, this could be an example of Syriac addition of a possessive in contrast to its Vorlage. The fact that the reciprocally related term ‘disciple’ is so near ‘master’ means that these form a pair, and we thus expect a suffix to occur with one of the pair in Syriac (see pp. 103–115). Wife. + 3 m.sg. Mark 10:2 SP. NA27 takes SCP to support the addition of ášôï™ after ôxí ãõíásêá in Matthew 18:25, even though the same witnesses also add a possessive with ‘children’ later in the verse. Similarly, NA27 takes P to attest ášôï™ after ôxí ãõíásêá in Mark 12:19 even though the possessive is in SCP in the parallel Luke 20:28, where there is little support for the presence of ášôï™ [only 565c 700 1194 1215 1424 2487]. 2.3.2.3 Generic Paired Relationships When generic paired relationships are spoken of in Syriac, one of the terms in the pair lacks the possessive and the other has it. Luke 12:53 is most instructive. Txt reads: They will be divided, father against son and son against father, mother against daughter and daughter against mother, mother-in-law against her daughter-in-law and daughter-in-law against £ mother-in-law.

The insertion sign marks the possessive under dispute. NA27 cites SCP in support of ášôyò with the final term ôxí ðåíèåñÜí. A consideration of the whole context indicates that this is a misuse of the witness of SCP. P reads: For a father will be divided against his son, and a son against his father, a mother against her daughter and a daughter against her mother, a mother-in-law against her daughter-in-law and a daughter-in-law against her motherin-law.

102 PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

Thus the possessive is always present on the second of the pair of relations and not on the first. This pattern is also that of S and C, though with slightly different texts. The consistent Syriac pattern is different from that in all Greek witnesses. It is therefore impossible to conclude anything about the presence or absence of the possessive in the Vorlage of SCP for the variant under dispute. The Syriac texts have surely adopted the most natural use of possessives for Syriac, and show here a pattern of being able to omit one of the possessives in a clear reciprocal relationship. It would be worthwhile considering whether such a pattern can at all explain the distribution of suffixes in the Father–Son passages in the Syriac texts of John’s Gospel. 2.3.2.4 Conclusion A noticeable tendency when looking at the variants related to the use of possessives with kinship terms in NA27 is that the vast majority of the time Syriac witnesses are being quoted for the presence of a possessive, but only relatively infrequently for the absence of a possessive. This imbalance should give us pause about the approach being taken. Of course it is possible that many of the notes in NA27 accurately describe the Vorlage of the Syriac texts. Nevertheless, it would be a most improbable coincidence if texts in a language that can be shown to prefer the presence of a possessive with kinship terms just happened to be translated from Vorlagen that also had an unusually high incidence of possessives with kinship terms. It seems wiser to conclude that frequently Syriac added an explicit possessive, and to refrain from using the Syriac trans-

POSSESSIVE PRONOUNS

103

lations as a witness to the presence of a possessive except in unusual circumstances. 2.3.3 ïj ìáèçôáß vs. ïj ìáèçôár ášôï™ Merx long ago noted that S generally referred to Jesus’ disciples as ‘his disciples’ while Greek witnesses vary between ïj ìáèçôáß and ïj ìáèçôár ášôï™. He saw that the construction with the possessive was more Semitic, and concluded that S’s consistency in using the possessive was a sign that it represented the earliest gospel text from which Greek witnesses had been altered so as to omit ášôï™.81 Thus Aramaic idiom was translated literally into Greek Gospels which were then translated literally into Syriac. After the Syriac translation was made Greek texts were then somewhat graecized by the removal of the possessive. However, if Merx’s presupposition of literal translation is dropped, it is obviously simpler to conclude that S simply represents Aramaic idiom, and tells us little about its Vorlage. Agnes Smith Lewis had come to just this conclusion in her own criticism of a work of Vogels: Dr. Vogels hat keine Rücksicht auf eine wohlbekannte grammatische Eigenschaft der syrischen Sprache genommen, nämlich, die Einschleppung von Fürwörtern und fürwörtlichen Suffixen an jeder möglichen Stelle. Zum Beispiel, müssen wir immer, mit einer einzigen Ausnahme, wo das Wort ,Jünger‘ in den Evangelien steht, mit Anhängung des besitzanzeigenden Fürworts, auf syrisch 81 Merx, Matthaeus, 142–45. Merx’s consideration of the texts

here is limited to an incomplete survey of twelve texts in Matthew, and thus does not give an indication of the full extent to which his theory, if carried through consistently, would have to posit otherwise unattested Greek variants.

104 PROPER NOUNS, COMMON NOUNS, AND PRONOUNS ,seine Jünger‘ lesen yhwdymlt. Ohne diesen Überreichtum an Fürwörtern wäre keine Evangelienübersetzung den Aramäischen Völkern annehmbar gewesen, weil sie durchaus nicht für syrisch angesehen wäre.82

In the following discussion the explanation for the presence of the possessive will be broadly the same as in the previous section. ‘Disciple’ is usually a term implying a teacher/master–disciple relationship. Since the teacher or master is prominently present throughout the narrative the possessive is easily added. Furthermore, Greek has a widely used definite article, which fulfils the function of marking entities as identifiable (i.e. as old or accessible information). Syriac has no such article, but marks identifiability in other ways. At a nontechnical level we could say that it uses suffixes more than Greek to compensate for its lack of a definite article. This seeming ‘compensation’ is an extremely frequent phenomenon with wide explanatory powers, which are considered here merely in relation to expressions for ‘disciples’. The data related to the relationship between the Greek and Syriac texts are complicated and are represented by the table below. It only includes references to ‘disciples’ that are plural in Syriac, to some degree semantically definite, refer to disciples of Jesus, and do not occur alongside the phrase ‘the twelve’ (or ‘the eleven’). Occurrences of ‘disciples’ as the first part of a genitive construction or with a first or second person possessive, or a third plural possessive have also been excluded. In this

82 Smith Lewis, Zu H.J. Vogels Schrift, 3–4.

105

POSSESSIVE PRONOUNS

table ‘variant’ refers to one or more variants recorded in the apparatus of NA27, as opposed to txt. Reference

Texts with possessive

Matthew 5:1 8:21 8:23 8:25 9:10 9:11 9:19 9:37 12:1 12:49 13:10 13:36 13:51 14:15 14:19a 14:19b 14:22 14:26 15:12 15:23 15:32 15:33 15:36a 15:36b 16:5 16:13 16:14 16:20 16:21 16:24 17:6 17:10 17:13 17:19 18:1

SCP SCP SP SP SP SP SP SP SCP SCP SCP SCP C SCP SCP SC SCP CP SCP SCP SCP SCP SCP SC SCP SCP C CP CP CP C CP SC SC S

Texts without possessive

txt txt txt variant txt txt txt txt txt txt

variant variant

txt

txt variant variant variant txt txt variant variant variant txt variant txt txt variant

P

P

txt txt txt txt txt txt variant txt txt txt txt txt

P P P P

txt txt txt txt txt

106 PROPER NOUNS, COMMON NOUNS, AND PRONOUNS Reference

Texts with possessive

19:10 19:13 19:23 19:25 21:1 21:6 21:20 23:1 24:1 24:3 26:1 26:8 26:17 26:19 26:26 26:35 26:36 26:40 26:45 26:56 27:64 28:7 28:8 28:13

SCP SCP SCP C CP C C SCP SP SP SP SP

txt variant

SP SP

variant

SP SP SP S86 SP SP P P

variant variant variant variant txt txt txt txt

P

txt

Mark 2:15

Texts without possessive variant txt

txt txt variant

txt txt

SP P SP

P SP85

P

txt txt83 txt txt txt variant84 txt txt txt txt txt txt txt txt txt variant

83 txt here has äýï ìáèçôÜò, but the variant äýï ô§í ìáèçô§í

ášôï™ has significant support: È f 13 et al. 84 The variant for the omission of the possessive is only supported by D, which also lacks the following word ïnäáôå. 85 SP read 50 + / 1 ‘all the disciples’. The presence of may have motivated the lack of the possessive. But see n. 86 on Matthew 26:56. 86 S has 1 0 / ‘his disciples, all of them’, with the possessive in contrast to 26:35. It is interesting to note, however, the different positions of relative to the noun in these verses.

107

POSSESSIVE PRONOUNS

Reference

Texts with possessive

2:16 2:23 3:7 3:9 4:10 4:34 5:31 6:1 6:29 6:35 6:41 6:45 7:2 7:17 8:1 8:4 8:6 8:10 8:27a 8:27b 8:33 8:34 9:8 9:14 9:28 9:31 10:10 10:13 10:23 10:24 10:46 11:1 11:14 12:43 13:1

P SP SP SP S P P P SP SP SP SP SP SP SP SP SP SP SP SP SP SP S S SP SP SP SP SP S SP SP SP SP SP

txt txt txt txt variant txt87 txt txt txt txt txt txt txt txt variant txt txt txt txt txt txt txt txt txt txt txt txt txt txt txt

Texts without possessive

variant variant

txt

P P

txt txt txt

P

txt

87 Here txt has ôïsò käßïéò ìáèçôásò, but the use of näéïò is

naturally to be ranked alongside the use of ášôï™ for the purposes of comparison with Syriac translations.

108 PROPER NOUNS, COMMON NOUNS, AND PRONOUNS Reference

Texts with possessive

14:4 14:12 14:13 14:16 14:32 14:50 16:7 16:19

SP SP SP SP SP SP C

Luke 5:30 6:1 6:13 6:17 6:20 7:11 8:9 8:22 9:14 9:15 9:16 9:18 9:43 9:54 10:1 10:22 10:23 11:1a 12:1 12:15 12:22 16:1 17:1 17:22 18:15 19:29

P P SP SP SP SP SCP SCP SC SCP SCP SCP SCP SP P SCP SCP SCP P SCP SCP SCP SCP SCP SCP

variant txt txt variant txt variant txt

txt txt txt txt txt txt txt txt txt

txt variant

txt txt txt variant txt

Texts without possessive P

variant txt

P88

txt txt txt variant txt

variant txt txt txt txt

88 The addition in P of the word ‘disciples’ is naturally related

to its omission of the same term in the previous verse.

109

POSSESSIVE PRONOUNS

Reference

Texts with possessive

Texts without possessive

19:37 20:45 22:39 22:45 22:49 24:9

S89 SCP SCP SCP S

P

John 2:2 2:11 2:12 2:17 2:22 3:22 4:2 4:8 4:27 4:31 4:33 6:3 6:8 6:11 6:12 6:16 6:22a 6:22b 6:24 6:60 6:61 6:66 9:2 11:7 11:8 11:12 11:54 12:4 12:16a

txt

txt variant txt txt

SC

P P P SP SP SCP SCP SCP SCP SCP

txt txt txt txt txt txt txt txt txt

SCP SCP S SCP SCP S CP CP SCP SCP SCP SP SP SP P SP P SP

txt txt variant txt txt txt txt txt txt txt txt txt variant variant variant txt txt

89 C omits reference to ‘disciples’.

variant

P

txt txt variant

P

S

variant txt txt txt txt variant

110 PROPER NOUNS, COMMON NOUNS, AND PRONOUNS Reference

Texts with possessive

12:16b 13:5 13:22 13:23 13:28 16:17 16:29 18:1a 18:1b 18:2 18:15 18:19 18:25 20:10 20:18 20:19 20:20 20:25 20:26 20:30 21:1 21:2 21:4 21:8 21:12 21:14 21:23

P SP S SP S SP SP SP SP SP SP SP

variant txt txt txt txt txt txt

S S SP

P

SP90

txt txt variant

SP SP

Texts without possessive

txt txt variant txt

variant

SP SP SP SP P SP SP P SP P S

txt txt

variant

txt txt txt txt txt variant txt txt txt txt txt

The data in the table clearly reveal a strong preference in both OS and P for the expression with the possessive. Whereas txt has 67 references without the possessive there are only 32 such references in P, 14 in S, and one in C. Of the Syriac references without suffixes a large number are found in the last two chapters of John’s Gospel. While this 90 For SP’s

singular.

50 + /

02 ‘one of the disciples’, txt has a

POSSESSIVE PRONOUNS

111

might be seen as evidence for a different hand in the translation of OS, it should also be remembered that many of these texts refer to the activity of the disciples independent of their master. It is in just such a postresurrection context that the only reference without a possessive in C is found (Luke 24:9). The disciples here are not disciples in a relational context, but are disciples absolutely. Functional aspects of the designation therefore come to the fore. Moreover, in some of these cases specific factors militate against the use of the usual expression ‘his disciples’. For instance, in John 20:10 SP only two disciples functioning separately from the wider group are referred to, and in John 21:23 S a much wider group of Christians than the original disciples is clearly meant. Furthermore, if a suffix had been used in this text ‘his disciples’ might be misunderstood as referring to the apostle John’s disciples. In Luke 24:9 the reference is to a group of disciples beyond the ‘eleven’. A further exception to the tendency for OS to add a possessive seems to be John 12:4 where a possessive is omitted in S in contrast to most Greek witnesses. Jesus is not, however, the protagonist in the action immediately preceding, and this may bring some relational distance between him and Judas who is described as ‘one of the disciples’ rather than ‘one of his disciples’.91 In Matthew 19:25, SP lack the possessive but C has it. Since in the context S and C differ in vocabulary and C is slightly more distant from the Greek (using E / 91 Of course, since many of the OS occurrences of ‘disciples’

without a possessive are in John, it is relevant to note that Lyon, Syriac Gospel Translations, 188, has suggested that this gospel may have been translated by a different hand from the other gospels.

112 PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

‘again’ where there is no Greek equivalent) it may be that S’s text represents a revision of OS, which is preserved in C. Thus the overwhelming tendency is for OS to add a possessive, a process that seems also to be preserved or to occur in P, though to a lesser extent. Various explanations have been proposed for the exceptions, which, at any rate, do not threaten the general conclusion. Whereas the table shows a regular pattern of OS, and to a lesser extent P, adding possessives relative to txt, in only a small number of cases are there no Greek witnesses in our control editions with a possessive: Matthew 14:19b, 21:6, 26:19. This is to be attributed to the wide degree of fluctuation in Greek witnesses between the expressions ‘the disciples’ and ‘his disciples’. However, the fact that in almost all cases some Greek witness has a possessive should not be taken to indicate that the Syriac texts were translated from such witnesses. The Greek witnesses with possessives in these cases are quite diverse and often of little proven import. It is implausible to maintain that Syriac witnesses consistently represent readings that are now coincidentally only preserved in diverse minor witnesses. Further consideration reveals that this overwhelming pattern is likely to be caused in part by a preference in the Syriac witnesses for the suffixed form. First of all, to consider the evidence of OS. Frequently when the word ‘disciples’ is added in OS in contrast to

POSSESSIVE PRONOUNS

113

Greek witnesses it has a 3 m.sg. possessive suffix,92 as it does in paraphrases of other phrases.93 The tendency to preserve a possessive added in OS is also witnessed in P. P on eighteen occasions uses + 0 / where NA27 records simple ïj ìáèçôáß (in any case) in its text without variants.94 Moreover, P may use a possessive on occasions where it adds an explicit nominal reference to the disciples,95 and in other cases too.96 In contrast to this there are only a few occurrences where it is particularly likely that there was a possessive in the original which has not been represented in translation.97 This means that P is a moderately reliable guide for the omission of a possessive, but an unreliable guide for the presence of one.

92 Matthew 8:18 C, 13:51 C, 16:14 C, Mark 9:8 S (contrast P),

16:19 C. 93 E.g. Luke 10:1 S, John 4:1 C. In John 11:16 ôïsò óõììáèçôásò, perhaps followed by ášôï™, ‘to the [his] fellow disciples’ becomes in S 50 / ; ‘to his fellows the disciples’ with the possessive expressed on another word. Similarly in John 13:28 ‘those who were dining’ becomes ‘his disciples’ in S. 94 Matthew 13:10, 14:19a, 19:13, 24:3, 26:19, Mark 10:10, 10:13, Luke 9:16, 9:18, 10:23, 17:22, 18:15, 19:29, 22:39, 22:45, John 4:31, 11:8, 13:5. 95 Luke 12:15. 96 For instance translating ïj äþäåêá in Luke 9:12. See also Luke 10:1. 97 John 6:22a, 20:26, 21:2. John 21:2 has a partitive construction. In P, partitive constructions have a general tendency not to have a possessive. John 9:27 is a different construction altogether. Unsuffixed 50 + / is added in Mark 9:8 and Luke 9:15.

114 PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

Matthew 14:19 is most instructive in indicating differences between SC and P. In the Feeding of the Five Thousand txt reads êár êëÜóáò häùêåí ôïsò ìáèçôásò ôï˜ò + 0 / for both Tñôïõò, ïj äc ìáèçôár ôïsò –÷ëïéò. SC read references to the disciples. P does so for the first occurrence, but for the second it reads 50 + /, which has ( ‘and they’ prefixed to it as in the text of C. This marks the second reference to ‘disciples’ in P as denoting the same group as the first. This is consistent with the hypothesis that the suffixes in the OS and to a lesser extent in P are being used to mark identifiability, and that if identifiability is established it is more likely that the suffix will be disposed of. Similarly in John 4:1 there is a division + 0 / and SP have of Syriac witnesses: C reads 50 + /. However, SP each have an indefinite context (‘he made many disciples’) whereas C has the definite ‘his disciples were many’. The tendency for ‘his disciples’ to be used to denote Jesus’ disciples means that when the phrase denotes John’s disciples an explicit reference to John may be added (Matthew 14:12 C [contrast S]).98 On a number of occasions NA27 cites a variant involving Syriac as a witness to the presence or absence of a possessive with a word denoting the disciples of Jesus. There is a noteworthy sameness about many of the variants. NA27 cites Syriac as supporting ášôï™ after reference to ïj ìáèçôáß (in any case) in Matthew 8:21 SCP (with UBS4), 14:15 SCP, 14:22 SCP, 15:12 SCP, 15:33 SCP, 98 A couple of times ‘the disciples … to Jesus’ in Greek

witnesses corresponds to ‘the disciples of Jesus’ in Syriac ones (Matthew 18:1 C, 26:17 S).

POSSESSIVE PRONOUNS

115

15:36a SCP, 16:5 SCP, 16:20 CP, 17:10 CP, 19:10 SCP (with UBS4), 26:8 SP, 26:36 SP, 26:40 SP, 26:45 SP, 26:56 S, 27:64 SP, Mark 6:41 SP (with UBS4), 8:1 SP, 14:16 SP, Luke 9:54 SCP, 12:22 SCP (with UBS4), 16:1 SCP, 20:45 SCP (with UBS4), John 2:12 P (with UBS4), 11:7 SP, 11:54 SP, 13:22 S, 20:30 SP (with UBS4), 21:1 SP, 21:14 SP. Likewise in Matthew 8:25 NA27 (with UBS4) quotes SP for the addition of ïj ìáèçôár ášôï™ to txt, where some Greek witnesses also support the addition of just ïj ìáèçôáß. Since OS tends to use the suffixed form ‘his disciples’ throughout, when NA27 quotes S as supporting variants adding possessives to txt in Matthew 26:56 and John 13:22 it is overstating the evidence, whereas if it were to have given the evidence for txt it would have been able to include P with some confidence since it lacks a possessive in these cases. It is implied in the apparatus of NA27 that P may add a suffix since P is listed without parentheses in support of the addition in Luke 10:22 êár óôñáöårò ðñ’ò ôï˜ò ìáèçôNò åqðåí, even though it has the possessive with ‘disciples’. Agreements between D and SP in using a possessive in John 11:8 and 13:5 can no longer be used as evidence that the Syriac was translated from a Vorlage like D. 2.3.3.1 Related Phrases It is not merely with the word ‘disciple’ that Syriac tends to add a possessive. What applies to this term also applies to similar expressions. ‘The twelve’ in Greek often,99 though

99 Mark 6:7 P, 9:35 S, 10:32 SP, 14:17 SP, Luke 8:1 SCP, 9:1

SCP, 9:12 S, 18:31 SCP, John 6:67 SCP.

116 PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

not always,100 become ‘his twelve’ in Syriac. When no subject is expressed in the Greek of Luke 9:6, SC add 1)+ ‘his apostles’, and the occurrence of ‘apostles’ in Luke 22:14 C also has a possessive. In Matthew 20:17 where Greek variants attest ôï˜ò äþäåêá () D L È), ôï˜ò äþäåêá ìáèçôÜò (txt), and ôï˜ò äþäåêá ìáèçôNò ášôï™ (13 892c 1424 et al.), SC read /,8#"/ ‘his twelve’ and P reads + 0 / 8#"/ ‘his twelve disciples’. Despite NA27’s citation of P in support of the reading of ms 13 et al. it is at least as likely that P read the much better attested txt as its Vorlage as that it read the text of ms 13 and relatively few other witnesses for a possessive in this case. Matthew 26:20 is very similar. NA27 cites (S) with / 8#"/ in support of ô§í äþäåêá, and P with 8#"/ + 0 / in support of ô§í äþäåêá ìáèçô§í ášôï™. The citation of P considerably adds to the attestation of what is otherwise a very marginally supported reading. P, however, may well have been produced from the very widely attested text ô§í äþäåêá ìáèçô§í, which is read by ) A L W Ä È and many others.101 100 In particular there is no possessive with the frequent

phrase ‘one of the twelve’: Matthew 26:14 SP, 26:47 SP, Mark 14:10 SP, 14:20 SP, 14:43 SP, Luke 22:47 SCP, John 6:71 SCP. There is also no suffix in Mark 11:11 SP and Luke 22:3 SCP. John 6:70 P is not really a parallel construction since it is vocative. 101 There is, moreover, an inconsistency in the use of parentheses to mark the witness of S in Matthew 26:20, but not of SC in 20:17. For complete consistency SCP should also occur in parentheses in Luke 9:1 where they have ‘his twelve’ in contrast to Greek texts. The key thing is that parentheses should be used consistently. In my opinion it would be perfectly legitimate to omit the parentheses in cases where a possessive is

POSSESSIVE PRONOUNS

117

When these patterns are brought to bear on the variants in Luke 22:14 it is seen that a more radical amendment of NA27’s apparatus may be necessary. Where 1)+ ‘his apostles’. NA27 txt has ïj Pðüóôïëïé, C has cites C in parentheses since it adds a possessive which was almost certainly not in its Vorlage.102 However, NA27 also + 0 /, in support of the reading ïj cites S, which reads ìáèçôár ášôï™. The problem here is that ïj ìáèçôár ášôï™ is not attested in Greek manuscripts. At first sight, the reading of S could have been produced from a Vorlage with ìáèçôáß or with ìáèçôár ášôï™, but neither is particularly more probable than the other. However, further consideration casts doubt on any retroversion involving the word ìáèçôÞò. Syriac witnesses, especially S, often seem to use ‘his disciples’ as a paraphrase of their Vorlage. In Matthew 10:2, S also gives the reading ‘these are the names of his twelve disciples’, where other witnesses read ‘the twelve apostles’; in 10:5, S has ‘these twelve his disciples’ where other witnesses have ‘these twelve’; in 26:51, S has ‘one of the disciples of Jesus’ for ‘one of those with Jesus’; in Luke 22:49, S has ‘his disciples’ for ‘those around him’; in John 13:28, S has ‘his disciples’ for ‘those who were dining’. CP likewise have ‘his disciples’ in Luke 9:12 for ïj äþäåêá, where S has ‘his twelve’.103 All Syriac versions, especially S, thus seem to paraphrase de-

present in line with a translation tendency and there is minimal probability that the versional reading was derived from a Vorlage other than the one for which it is cited. 102 Note the addition of the possessive to the word ‘apostles’ in S in Luke 17:5. 103 Ms 267 has ‘his twelve disciples’.

118 PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

signations of the disciples towards the most common ‘his disciples’.104 So NA27’s retroversion of S in Luke 22:14 is 104 This has to be a more natural explanation than to

suppose that variants using widely different expressions for ‘his disciples’ arose secondarily. Reference should be made to the theory of Merx, Uebersetzung, 242–44, that S is generally original even when it alone avoids Pðüóôïëïé and has ‘[his] disciples’. There are several problems with this theory, which is part of Merx’s general preference for readings of S. First, while Merx seeks to explain the term ‘apostle’ as a later ecclesiastical term, he cannot explain the variety of designations for the disciples that grow up in other witnesses. It is not easy to envisage a transmissional mechanism that will replace ‘his disciples’ by diverse, contextually appropriate phrases ranging from ‘those who were dining’ to ‘those around him’. On the other hand, it is simple to envisage varied phrases with the same referent being made more monolithic in translation. Secondly, the use of possessives in S is best explained as a Syriac-internal issue, which undermines Merx’s consistent assumption that S is literal. Thirdly, Merx’s method results in sweeping aside all other witnesses based on his preference for S on internal grounds. While we cannot rule out that a single witness preserves the correct reading where all others err, it stretches credulity to suppose that, despite the level of independence in other manuscript, versional and citational evidence, the true wording is consistently found in the same witness—and that a translation—where all the others have united to go astray. Fourthly, Merx wrongly bases his preference for S on the view that its text is older than all other witnesses. Merx’s view can best be represented in his own words: ‘Alles das führt zu der Einsicht, daß der Wortgebrauch im Syrsin [S] nicht zufällig ist, sondern innerlich begründet, so daß aus Gründen der inneren Kritik seine Lesarten die echten sind, wogegen bei den Griechen und allen anderen Zeugen Verderbniß vorliegt, die aus mangelhafter Aufmerksamkeit auf die Tragweite des Wortgebrauchs und daraus hervorgehender Ungenauigkeit entstanden ist. ... Nach alledem wird man sich ernstlich fragen müssen, ob nicht eine Reihe von innerlich so consistenten

POSSESSIVE PRONOUNS

119

entirely possible, but perhaps should be recorded in Latin to give due weight to the uncertainty that it reflects a Greek variant. Against this the analogy of Mark 4:10 might be argued. There S seems to agree with D W È Old Latin et al. in support of ïj ìáèçôár ášôï™ for txt’s ïj ðåñr ášô’í ó˜í ôïsò äþäåêá. S’s reading is manifestly a lectio facilior, and the possibility must at least be considered that S agrees with the witnesses independently.105 In terms of transmissional probability, moreover, the simplified expression of the Syriac ‘his disciples’ is best seen as generally secondary to the more varied expressions to which it corresponds. A final text that may suggest that the underlying motive for the addition of possessives with expressions for disciples is related to identifiability is the expression ‘the seventy[-two]’ in Luke 10:17, which in CP becomes ‘those seventy whom he sent’, and in S ‘those seventy-two whom he sent’. Here to use the form ‘his seventy’ was not an option, and so to mark identifiability a paraphrase is produced. 2.3.4 Other Variations in the Use of Possessives There are other cases of additions of possessives, for many of which no patterns emerge. Given the number of examples of possessives in the Greek and Syriac Gospels, it was bound to be the case that some cases of non-

Lesarten, wie die eben vorgelegten, obwohl nur Syrsin sie bietet, original sind, so daß diese eine Handschrift mit ihrem Zeugniß alle übrigen zusammen aufwiegt—und aus dem Felde schlägt.’ (Merx, op. cit., 244). 105 A similar agreement between S with ‘his twelve disciples’ and D with ôï˜ò äþäåêá ìáèçôÜò occurs in Mark 6:7.

120 PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

correspondence would elude classification or would call for ad hoc explanations. Nevertheless, the following cases illustrate wider principles. Angels. The addition of ášôï™ after ôï˜ò PããÝëïõò in Mark 13:27 for which NA27 cites SP is paralleled by the addition of a 3 m.sg. suffix to ‘angels’ in Matthew 25:31 P,106 Mark 8:38 P (contra S) [with 122 124 346 483* 1346], Luke 9:26 CP (contra S). In all four situations there is a 3 m.sg. element referring to the possessor positioned in the sentence very close to the phrase mentioning ‘angels’ leading to explicit mention of possession in Syriac witnesses. It may even be that ‘angel’ in Syriac can imply a relationship with a sender or director, and thus obtains a possessive in certain contexts as a relational term. Gospel. In Matthew 26:13 SP, Mark 8:35 SP [with 1 sg. in 1082], 10:29 SP [with 1 sg. in 482], 13:10 P, 14:9 S, 16:15 P / ;7 ‘my gospel’ appears where Greek witnesses have arthrous åšáããÝëéïí or ô’ åšáããÝëéïí ôï™ôï. In Mark / ;7 ‘his gospel’ likewise corresponds to 1:15 S, arthrous åšáããÝëéïí. The possessive marks definiteness—it is not just any good news that will be preached, but a particular good news. In revising OS in Mark 14:9, P adds 5 to represent ôï™ôï, but leaves the 1 sg. possessive on / ;7. To use a possessive on the term ‘good news’ is clearly the most normal Syriac translation. Nets. In Matthew 4:20 and Mark 1:18 NA27 suggests that SCP and SP respectively had ášô§í after ôN äßêôõá. 106 Ms 692 is said by Legg to have ášôï™ but to lack the

following ìåôE ášôï™ (whereas P has # ‘with him’). Its reading generally would not give rise to P, and the variant presumably arose through the accidental omission of ìåôÜ.

POSSESSIVE PRONOUNS

121

Actually ôN äßêôõá ‘the nets’ appears elsewhere in txt without a possessive or a variant with a possessive in NA27 apparatus but where Syriac witnesses read a 3 m.pl. possessive.107 The singular ô’ äßêôõïí has a possessive added where no variants are recorded in our control editions,108 as does the similar noun PìöéâëÞóôñïí in Matthew 4:18 SC (contrast P). A possessive is also used in a similar phrase in Mark 1:16 S (contrast P) where some quite different Greek readings related to a ‘net’ are well attested, but all lack the possessive. There is nothing therefore implausible in allowing that the renderings with possessives in Matthew 4:20 and Mark 1:18 may have been produced from Vorlagen without possessives, especially in the case of the OS witnesses. Sins. In John 16:8, S reads: ‘And when he comes he will rebuke the world for its sins and about his righteousness and about judgement.’ Clearly the translator was uncomfortable with leaving the possessors of the ‘sins’ and ‘righteousness’ unspecified. This seems best taken as an alteration for theological reasons, though it also leaves open the uncomfortable possibilities of ‘his sins’ and ‘its righteousness’. When Jesus says to the paralytic Pößåíôáß óïõ áj Qìáñôßáé (Matthew 9:2 txt) or PöÝùíôáß óïé áj Qìáñôßáé óïõ (Luke 5:23 txt), NA27 records SP as supporting Pößåíôáß/PöÝùíôáß óïé áj Qìáñôßáé óïõ along with some Greek witnesses in both cases. However, wherever extant, OS and

107 Mark 1:19 SP, Luke 5:2 SP (contrast 5:5 SP). In both

places a small group of Greek texts has ášô§í. 108 John 21:6 SP with 2 m.pl. suffix; contrast 21:11 (2×).

122 PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

P have ? 2 ? % ;) for this phrase,109 even though Greek witnesses for both the dative and possessive are relatively minor in Matthew 9:5 and Mark 2:9. It seems that the dative is simply added for idiom.110 Given the tendency to add a dative it is possible to derive SP in Matthew 9:2 from txt, and in Luke 5:23 from the variant PöÝùíôáß óïõ áj Qìáñôßáé. It is even possible that the possessive rather than the dative has been added in either case and that the minority readings Pößåíôáß/PöÝùíôáß óïé áj Qìáñôßáé might have been the Vorlage of SP. At any rate it is hard to be confident about the Vorlage of the Syriac in any of these texts.

2.4 INDEPENDENT PRONOUNS 2.4.1 The Representation of ášôüò In Greek ášôüò in the nominative may serve a variety of purposes, including to identify a subject, to introduce a change of subject, and to begin an apodosis. It seems clear from the lack of a Syriac equivalent in several places where the control editions uniformly attest ášôüò that it was not always represented in translation: Matthew 14:2 SC,111 Luke 8:1 SC, 9:51 SCP, 16:24 SCP, 22:23 SCP,112 John

109 Matthew 9:2 SP, 9:5 SP, Mark 2:5 P, 2:9 P, Luke 5:20

SP, 5:23 SP. 110 The same addition in contrast to the overwhelming weight of Greek authorities happens in the analogous construction of Luke 7:47 SCP, 7:48 SCP. 111 C f 1 and the sixteenth century ms 61 have ï£ôïò. 112 827 2766 lack ášôïß. The reading of W, êáé çñîáôï óõíæçôåéí ðñïò áõôïõò, is sufficiently different from the Syriac

INDEPENDENT PRONOUNS

123

12:24 SP, 17:8 S.113 The contrasting structures in Luke 8:1 illustrate the dynamic. txt êár dãÝíåôï dí ô² êáèåîyò êár ášô’ò äéþäåõåí êáôN ðüëéí êár êþìçí ‘and it happened afterwards that he was travelling by city and village’ K ¾Ø˜ÍùÁ K ¿†… ûÜÿâ çÙ߅ ˜ÿÁ ç↠SC ¿ÿæØÊãÁ† ‘and after these things he was going round among villages and cities’ SC, as is typical, represent êár dãÝíåôï by just †. Thus, whereas in the Greek ášôüò serves partly to mark the change of subject between dãÝíåôï and äéþäåõåí, there is no verb preceding Syriac ûÜÿâ, and no other subject with which its subject may be contrasted. SC therefore have less reason to represent ášôüò than the Greek has to have it. In Luke 24:15, NA27 cites SC, with only versional support, for the omission of êár ášôüò in txt’s êár dãÝíåôï dí ô² ¿ìéëåsí ášôï˜ò êár óõæçôåsí êár ášô’ò EÉçóï™ò dããßóáò óõíåðïñåýåôï ášôïsò. The lack of êáß is explained in the comments on this verse on p. 157. As noted above ášôüò is often omitted where redundant (which it certainly is before “ÍýØ). In one of the two other occurrences in this Gospel where ášôüò appears strengthening a proper name, it is

versions that, though it lacks ášôïß, it is unlikely to have been their Vorlage. 113 W may read ášôü for ášôïß. If so this would be a mistake since it would have no referent in the context.

124 PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

not represented in OS (i.e. S to Luke 3:23; contrast SC to Luke 20:42). This raises the possibility that SC might have omitted the phrase êár ášôüò as a whole in translation. Moreover, ášôüò can function in the first, second, or third person. Hence it is not surprising to find that ášôïß is sometimes rendered by the first person plural pronouns 32 and 3 (&. For instance, in Luke 22:71 txt’s ášôïr ãNñ 3 (& 6 ‘for zêïýóáìåí is rendered ' ) 6 32 ‘for we behold we have heard’ in SC and ' ) 114 have heard’ in P. This means that in Luke 11:4, where NA27 cites (SC) with 3 (& C& and 32 C& respectively in support of D’s reading ©ò êár ½ìåsò instead of txt’s êár ãNñ ášôïß one may at least wonder why SC might not equally support )*’s reading ©ò êár ášôïß. Greek ášôïß in this context means ‘we’, and therefore is naturally represented by Syriac ‘we’. This is made abundantly clear by P which has 3 (& C& 6, clearly by its use of 6 a correction to txt, and yet retaining 3 (& as equivalent of ášôïß. In John 4:42, NA27 cites C in support of ášôï™ for txt’s ášôïß in ášôïr ãNñ Pêçêüáìåí. This makes C agree with the reading of D. However, in C’s reading ' ) 6 3 (& 3 ‘for we have heard from him’, 3 (& could stand for ášôïß (as 32 in P seems to). Consequently C is far closer to the reading of ) f 1 f 13 565, namely ášôïr ãNñ Pêçêüáìåí ðáñE ášôï™ than it is to the reading of D. Luke 11:14 represents a quite different situation to do with ášôüò. NA27 reads êár ƒí dêâÜëëùí äáéìüíéïí êár ášô’ ƒí êùöüí dãÝíåôï äc ôï™ äáéìïíßïõ dîåëèüíôïò dëÜëçóåí ¿ 114 It is less likely that the translations of SC and P are to be

derived from D’s reading zêïýóáìåí ãÜñ.

INDEPENDENT PRONOUNS

125

êùö’ò êár dèáýìáóáí ïj –÷ëïé. SC are cited in company with 45.75 et al. for the omission of êár ášô’ ƒí, which is placed within single brackets in the text. SC read: 0 >=( 0 )2 5 .) >= ... ‘and it happened as he was casting out a demon from a deaf man, might and when it came out …’. The opening correspond to the dãÝíåôï äÝ later in the Greek text. We , this time forming part of the then have another equivalent of the Greek compound tense in ƒí dêâÜëëùí. The question is, is it reasonable to expect that the in representing êár translator might then use a third ášô’ ƒí? There is something rather clumsy about a >= 0 hypothetical sentence like 5 .) . Moreover, such a sentence would not )2 be clear. Whereas Greek ášôü, being neuter, obviously , whether the personal or refers to äáéìüíéïí, Syriac demonstrative pronoun, might refer to the subject of the sentence, namely Jesus. Syriac could avoid this difficulty by 5 .) translating äáéìüíéïí êár ášô’ ƒí êùöüí by in the preceding ) 2, but, given the double use of context the redundant relative clause could be avoided ) 2 5 .). SC have not, with the simple formula 5 .). This however, chosen this, but rather ) 2 expression is probably motivated by a desire to avoid the dual attribution of the term ‘dumb’ as found in Greek texts. These use êùöüò both of the spirit and of the person from whom the spirit is cast out. In sum, whatever their Vorlage, SC paraphrase, but there are plausible reasons internal to Syriac why they might not represent êár ášô’ ƒí if it were in their Vorlage.

126 PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

2.4.2 The Use of ðOò and The uses of Syriac ‘all’ and of the Greek words ðOò and ”ëïò do not entirely overlap.115 For instance, it has been established that Syriac and Greek follow different principles in the position of the word ‘all’ in relation to a head noun.116 As with any other word, genuine textual variants occur with the word ‘all’, but it is necessary to examine overall patterns of its use in Syriac before drawing conclusions about Vorlage. 2.4.2.1 The Addition of ‘All’ Baethgen, writing with reference to C, has helpfully observed one recurring pattern for us already. Ich mache endlich darauf aufmerksam, dass d (Nm) lk = quisquis oft zur Auflösung des Participiums dient, und dass aus lk nicht auf ein griechisches ðáò etc. geschlossen werden darf.117

This is the case for the second and third participles in can be introduced with a Luke 11:10 SC. Similarly, neuter participle. In Luke 10:8 SC ôN ðáñáôéèÝìåíá ›ìsí 09 7$ H0 F . The reason for becomes in these examples is their generalising the use of also to occur in renderings nature: it is very natural for 115 For differences between Hebrew and Syriac in use of the

word ‘all’, see Williams, Studies, 39–40; idem, ‘“According to All” in MT and the Peshitta’, 107–109. This article demonstrates the avoidance of ? & followed by in many situations. This avoidance is also found in P’s rendering of êáôÜ when meaning ‘according to’ + ðOò in Acts 3:22. 116 P.J. Williams, ‘Some Problems in Determining the Vorlage of Early Syriac Versions of the NT’, 539–40. 117 Baethgen, Evangelienfragmente, 21.

INDEPENDENT PRONOUNS

127

of Greek phrases meaning ‘whatever’ or ‘whoever’, regardless of whether ðOò occurs in the Greek text itself.118 occurs several times in reIn John’s Gospel was presentations of ôïóï™ôïò. The presence of genuinely motivated by ôïóï™ôïò as is shown by its consistent use despite differences between S and P (12:37, 14:9, and 21:11). For instance, in 14:9 S and P represent 5 6 ( and ( ôïóï™ôïò ÷ñüíïò respectively by 1 ‘all these’ 3 ! 1 . In John 6:9 SCP have + ) for the plural of ôïóï™ôïò. (S is added with the term ‘flock’ In John 10:16 SP when the ‘togetherness’ of the flock is a focus. Likewise, is added in John 21:2 S before the word ‘together’. Since marks ‘totality’ and ‘totality’ and ‘unity’ are related concepts, this translation development is not surprising. In a few places ‘all’ is added when there is another ‘all’ nearby in the context, e.g. Luke 11:17b SC (from 11:17a— arguably the Greek ðOóá, though not in grammatical agreement with ïqêïò, covers it as well as âáóéëåßá); 21:24b SC (from 21:24a); John 11:50a S (from 11:50b). The simplest cause of the addition of ‘all’ is assimilation.119 More speculatively one might suppose that the translator was encouraged to use universal or general language by the

118 Added to which, in Luke 10:8 the additional

F in SC is not a careless appendage in translation, but is the only element in the Syriac that expresses the plurality explicit in the Greek. For further additions of see also John 6:47 C and 16:23 SP. The latter text is relevant despite the uncertainty over its Vorlage. 119 In the case of John 11:50 the assimilation of ‘people’ to ‘all the people’ must have taken place within the Syriac since the Greek has two different words for ‘people’.

128 PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

subconscious influence of other generalising elements in the context. Various loosely bound proposals have been made in Syriac. In above seeking to explain the addition of some cases, however, it is hard to find a reason for the addition. A couple of such ‘alls’ occur in Luke 15, perhaps heightening rhetorical effect. There is joy before ‘all the angels’ (15:10 SC), and the prodigal’s father divides up ‘all his livelihood’ (15:12 SC). The likelihood of assimilation to Matthew 25:31 and Luke 8:43, respectively, is remote. Luke 15:10 has nothing in common with Matthew 25:31 except for the mention of angels, and the difference in gender on the suffix would make ‘his livelihood’ in Luke 15:12 unlikely to be assimilated to ‘all her livelihood’ in Luke 8:43. which There are isolated cases of the addition of do not allow any certainty of analysis. Without recurring patterns of occurrence there is a real possibility that these represent genuine textual variants. Yet even in cases where no parallel exists there may be indications that a Vorlage variant is still not the most obvious conclusion. One such case is Matthew 20:12, the parable of the hired workers. The workers complain: 9 1%) S ‘We have carried the burden of the day in heat.’ 2 1 9 1%) C ‘We have carried the burden of the whole day and the heat.’ 2 9 1%) P ‘We have carried the burden of the day and its heat.’

INDEPENDENT PRONOUNS

129

Only C has , yet all three Syriac texts contain paraphrase in one way or another. Obviously, simply to write 2 9 ‘the burden of the day and heat’ did not seem to bring out the force the translators wanted. Critical editions have a few variants related to a particular addition of ‘all’. In Matthew 24:6 NA27 and UBS4 cite P in favour of the addition ðÜíôá (neuter plural) and S for including ôá™ôá before ãåíÝóèáé in the phrase äås ãNñ ãåíÝóèáé. The Vorlage ascribed to P is highly plausible, but the one ascribed to S is more debatable being much less well attested within the Greek manuscript tradition. 2.4.2.2 The Omission of ‘All’ An equivalent of ðOò is sometimes omitted without support from our control editions (e.g. Matthew 6:32 C, Mark 14:23 S, Luke 3:5b S, 4:28 S, 13:5 SC [contrast 13:3], 21:4 SC). In some cases specific motives can be posited. is used in the representation In Matthew 5:11 S no of Greek ðOí ðïíçñüí. In the Greek Jesus is pronouncing a benediction on those who suffer a large quantity or intensity of bad things said against them. However, if ðOí the ðïíçñüí had been translated by Syriac @ benediction would have been rather on those who had ‘any bad’ (whether of small or great quantity and intensity) said against them. This may be a reason for S’s omission of ‘all’ H0 F ‘everything in contrast to Greek texts. C’s @ that is bad’ does not formally correspond to the Greek, though semantically it does. In Luke 9:23 txt’s hëåãåí äc ðñ’ò ðÜíôáò ån ôéò èÝëåé êôë & ‘And he is simplified in SC to said “Everyone who wants, etc.”’ SC have no ‘all’ in the

130 PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

speech formula, but use ‘all’ in rendering ån ôéò. This simplification is in line with the tendency, discussed elsewhere, for Syriac speech formulae to show less variety than the Greek.120 Luke 19:37 presents a problematic reference in NA27. Most probably in the interests of economy of space its apparatus suggests that SC have omitted both ðáó§í and äõíÜìåùí from txt’s ðåñr ðáó§í ®í åqäïí äõíÜìåùí. The problem here is that along with some potential support from the Old Latin, NA27 gives the impression that a Greek text with ðåñr ®í åqäïí once existed. In fact SC read 2 H0 # ‘concerning everything they had seen’. In other words the word ‘all’ is in S and C. This ‘all’ could have been added in translating ðåñr ®í åqäïí, or it is just possible that S read a text more like that of D (ðåñé ðáíôùí ùí åéäïí ãåéíïìåíùí), merely failing to represent ãéíïìÝíùí. A related phenomenon may be the omission of ‘all’ preceding ‘things happening’ in Luke 21:36 SCP. Unnecessary fullness of expression is avoided in Syriac. In Luke 1:65 NA27 says that SP omit ðÜíôá in txt’s äéåëáëåsôï ðÜíôá ôN ¼Þìáôá ôá™ôá. Yet SP in fact have 11 $ ‘these things were said’. This translation not only fails to represent ðÜíôá, but also has no equivalent of ôN ¼Þìáôá. Of course a literal equivalent of ¼Þìáôá meaning ‘things’ may be omitted (Luke 2:19 S), but could not a translation confessedly free enough to omit an equivalent of ¼Þìáôá also omit an equivalent of ðÜíôá?121 It must be 120 Chapter 6, pp. 249–52. 121 There may be a good reason for the omission of an

equivalent of ¼Þìáôá. ¼Þìáôá would most naturally be translated by 1 + , which would be tautological after the verb 1 . Where a

INDEPENDENT PRONOUNS

131

remembered that the occurrence of ðÜíôá in Luke 1:65 follows two occurrences of ‘all’ within that verse, and precedes a further occurrence in 1:66. The Greek does not sound repetitive having both ðOò and ”ëïò to represent ‘all’. However, Syriac, only having one equivalent item of vocabulary, may have preferred not to use it as often. NA27’s citation is yet more problematic for S, which has no equivalent of ðÜíôåò ïj Pêïýóáíôåò at the beginning of 1:66. Thus S’s failure to represent ðÜíôá in 1:65 may be part of a wider omission. 2.4.2.3 Other Features It seems that ‘all’ may sometimes be transferred during translation from one phrase to an adjacent related phrase. This has probably happened in Mark 1:39 where it is said that Jesus travelled preaching åkò ôNò óõíáãùãNò ášô§í åkò ”ëçí ôxí Ãáëéëáßáí. S, on the other hand, records that Jesus was preaching ‘in all the synagogues of Galilee’. Since ‘the synagogues of all Galilee’ and ‘all the synagogues of Galilee’ refer to the same synagogues the transfer is easily made, subconsciously, by a translator. P at this point reads that Jesus was preaching ‘in all their synagogues in all twice. P’s reading was most probably Galilee’, with produced as a correction of S’s omission of ‘all’ with Galilee, the corrector failing to remove the ‘all’ with ‘synagogues’. Hyperbolic Greek ‘all’ referring to a large group of people may be rendered by ‘many’ in Syriac. In Mark 1:37, Simon’s words ðÜíôåò æçôï™óßí óå become in S ‘many people different verb of speech is used Greek ‘all’ (Luke 24:9 SC).

1 + may even correspond to

132 PROPER NOUNS, COMMON NOUNS, AND PRONOUNS

are seeking you’, corrected in P to ‘all people are seeking you’. In John 3:26 a similar statement by John’s disciples ðÜíôåò hñ÷ïíôáé ðñ’ò ášôüí ‘everyone is coming to him [Jesus]’ becomes a sober ‘many are coming to him’ in SP.

3A

RTICLES RTI CLES AND PARTICLES

3.1 THE SYRIAC INDEFINITE ARTICLE In Syriac the status emphaticus no longer functions to mark definiteness, although such was its function in earlier Aramaic. Yet this change in the role of the status emphaticus does not mean that definiteness and indefiniteness are unimportant for Syriac. Rather, Syriac finds other ways of marking definiteness and indefiniteness, or what may more broadly be seen as information which the textproducer believes the target audience will be able to identify as known or old, and that which they will find new. To mark definiteness Syriac can use demonstratives and several different types of suffix.1 To mark indefiniteness there is the status absolutus and also the indefinite article , positioned after the noun, a device frequently or used in the Syriac Gospels.2 The following table shows thirty-seven instances in the Gospels where some Syriac or , but the Greek testimony against witnesses have 1 For the use of suffixes to mark definiteness see Williams,

Studies, chapters 2–4; for the indefinite article in P see Falla, Key, Volume Two, 69–71. See also J. Joosten, The Syriac Language of the Peshitta and Old Syriac Versions of Matthew: Syntactic Structure, Inner-Syriac Developments and Translation Technique, 59–61. 2 When placed before the noun, usually functions as a numeral.

133

134

ARTICLES AND PARTICLES

reading a form of ôéò or åpò is uniform. Where a Syriac witness is not mentioned it is either not extant for the or .3 relevant text, or does not contain Syriac Witnesses and Text Matt. 2:23 8:2

C C

8:5 9:9 9:32 12:22

CP4 S S SCP

15:22 21:2 26:7 27:16 27:57 27:60

C C S S S6 S

Mark 1:40 3:1 5:2 7:24 7:32

S P S P S

(P

)

)

Greek

(P

ðüëéí ëåðñüò eêáôüíôáñ÷ïò Tíèñùðïí Tíèñùðïí êùöüí5 äáéìïíéæüìåíïò ãõíÞ –íïí ãõíÞ äÝóìéïí Tíèñùðïò ëßèïí

ëåðñüò Tíèñùðïò Tíèñùðïò ïkêßáí7 êùöüí

3 The marker of indefiniteness always follows the noun, but

whereas words sometimes intervene between the nouns and the marker of indefiniteness these are omitted in the table. Irrelevant Greek variants, e.g. eêáôüíôáñ÷ïò vs. eêáôïíôÜñ÷çò, are ignored. 4 S here reads without the article. 5 ) B et al. omit Tíèñùðïí. 6 Vööbus, Studies in the History of the Gospel Text in Syriac, II, 62, 83, mentions two manuscript of P with . See also the same addition in manuscripts of P to Matthew 17:14 (Vööbus, op. cit., 119, 123). 7 D W È 565 even read ôxí ïkêßáí.

THE INDEFINITE ARTICLE

9:36 14:3 14:13 14:15

Syriac Witnesses and Text S S S S

Greek ðáéäßïí ãõíÞ Tíèñùðïò PíÜãáéïí

Luke 2:25 5:12 5:18 8:32 9:38 9:47 9:52 13:11 19:2 22:12 23:33 23:50

SP S8 SP C SCP C C SC SCP SCP9 P SCP

Tíèñùðïò PíÞñ Tíèñùðïí PãÝëç PíÞñ ðáéäßïí êþìçí ãõíÞ PíÞñ PíÜãáéïí ô’í ôüðïí PíÞñ

John 3:25 4:5

P SC

5:2

CP

9:1

! )

S

(S

135

EÉïõäáßïõ ðüëéí êïëõìâÞèñá

" #

Tíèñùðïí ôõöëüí

There are also some variants in manuscripts of P: . Matthew 26:7, ms 14 reads after . Mark 3:1, mss 21 and 36 omit c . Luke 5:12, mss 4* 6 7 8 12 17 36 37 read phil after . Luke 22:12, P lacks Aside from the last case the variants all bring P closer to OS. 8 For the addition of

in a manuscript of P see Vööbus, Studies in the History of the Gospel Text in Syriac, II, 63. 9 Pphil lacks .

136

ARTICLES AND PARTICLES

From this table several things emerge. First, it seems that the addition of the indefinite article was much more common in the OS than in P. This is even more so with C than with S since C, when extant, attests all the additions found in S, and only lacks two of the additions found in P (Luke 23:33, John 3:25). These two additions in P are somewhat different from the others. Luke 23:33 P is striking because of the correspondence of the Syriac indefinite article to the Greek definite article. John 3:25 P in order to prevent the seems to have the addition of ! as plural in line with many Greek interpretation of texts. C, having the plural here, naturally could not use the indefinite article. C’s propensity for the article is sufficiently strong that on several occasions it attests its presence against S and P together (Matthew 2:23, 15:22, Luke 8:32, 9:47, 9:52). Secondly, the vast majority of the cases of addition of a Syriac indefinite article occur when a figure marked by a generic title is being introduced into a narrative. This seems to reflect either a conscious or a subconscious desire to mark the figure as new to the narrative. Thirdly, it is not possible to equate the presence of the article unequivocally with an early or late stratum in the Syriac translations. Although a simple binary comparison between S and P reveals more articles in the former, there are also cases where P attests articles not present in S. It seems unlikely that we are generally to connect the presence of the indefinite article with the presence of ôéò or åpò in the Greek. This is suggested not only by the absence of Greek witnesses in a large number of cases, but also by the fact that the Syriac tendency with the article is

THE INDEFINITE ARTICLE

137

largely unidirectional. There are many more cases where it adds an article against all Greek witnesses than there are where it lacks one against all Greek witnesses (though see p. 138 for the latter). It is improbable that the texts of S, C, and P all just happened to be translated or revised from Greek manuscripts that fortuitously showed the same grammatical peculiarity of greater use of the indefinite article. Further to the list of places where the Greek text is firm, there are a few other instances of the addition of the indefinite article in Syriac when the Greek witness for its presence is still very weak. If a tendency to add an indefinite article, particularly introducing new narrative characters, is established, then it becomes quite likely that here the Syriac does not witness to the Greek variant. Matt. 12:10 Mark 7:25 Luke 8:41

Syriac SCP P SCP

Greek Tíèñùðïò ãõíÞ PíÞñ

Mss with ôéò 1293 700 139610

It seems that the addition of the indefinite article is a matter basically internal to Syriac. Its special presence in C, which often otherwise presents a later stage of OS than S, could be explained if the addition of the article is seen as a process of naturalising the Syriac. Alternatively, this could be an aspect in which C retains a more primitive characteristic of OS. P represents, with some exceptions, a slight step back towards the Greek, but precise imitation of 10 The peculiar reading of ms 1093, with Tíèñùðüò ôéò for txt’s

PíÞñ, has a sufficient number of contextual differences to be ignored here.

138

ARTICLES AND PARTICLES

the Greek indefinite was not yet seen as important enough for one to one correspondence between Syriac and Greek articles to be firmly established. Alongside the tendency to add an indefinite article are a few occasions where the Syriac omits the number ‘one’ or the indefinite article, when it is present throughout the Greek witnesses (exceptions noted). Matt. 12:1111 Luke 7:4112 Luke 10:38 Luke 11:113 Luke 21:214

Syriac SC S P S C

$ %

Greek ðñüâáôïí fí äáíéóô† ôéíé ãõíÞ ôéò ôüðv ôéíß ôéíá ÷Þñáí

From this small set of data one potential pattern emerges. The omissions in Luke 7:41, 10:38, and 11:1 occur ad. This is sufjacent to another example of the word ficiently close in 7:41 that the possibility of haplography in may therefore be to S has to be considered. Omission of avoid repetition. This illustrates how tendencies in the Syriac, that to add an indefinite article, and that to avoid repetition,15 may conflict, producing irregularities. renders its The inner-Syriac tendency to use citation in critical apparatus problematic in the following cases:

11 Ms 1071 omits fí. 12 The immediately afterwards clearly goes with the fol-

lowing phrase. 13 Ms 1200 omits ôéíß. 14 Ms 063 omits ôéíá. 15 Cf. pp. 263–64.

THE INDEFINITE ARTICLE

139

Matthew 21:28. NA27 cites SCP with in 16 favour of ôéò after Tíèñùðïò. in Mark 5:25. NA27 cites SP with support of ôéò after ãõíÞ. Mark 12:1 and Luke 20:9. For the parable of the vineyard Greek variants occur in both Mark 12:1 and Luke 20:9 between Tíèñùðïò and Tíèñùðüò ôéò. For the two texts NA27 cites P and SCP respectively in support of Tíèñùðüò ôéò. These citations are problematic since the reading is only what one might expect, given the tendency observed above. Nevertheless the citation of P in in Mark is stronger because it contrasts with simple S—and P is more likely to have revised towards the literal. However, as seen in the table above, the third synoptic in all three parallel in Matthew 21:33 attests Syriac texts. Yet the Greek syntax of Matthew precludes any possibility of the presence of ôéò. Given the uncertainty, it seems best, then, to delete the reference to Syriac witnesses in Mark and Luke. Luke 11:37. NA27 cites P for the addition of ôéò after & is exactly what Öáñéóásïò. However, P’s phrase is in SC, and therefore need not represent an attempt by P to follow its Vorlage literally. The agreement between SC and D’s ôéò Öáñéóásïò could be fortuitous but this is less probable given the similar agreement in Matthew 18:2 16 The expression

was popular enough for the reverse addition to occur, with added to to render åpò or ôéò in Matthew 19:16 SC, Luke 9:49 C, 9:57 C, 12:13 SC, 13:6 SC, and 13:23 S. See also Luke 18:35 SC. In all these places the Greek testimony is united against the inclusion of PíÞñ or Tíèñùðïò.

140

ARTICLES AND PARTICLES

where NA27 cites SC for the addition of fí before ðáéäßïí as read in D. Naturally SC place the article after the noun. Chase saw this latter agreement as one of a number of genetic agreements between the OS text and that of the ‘Western’ witnesses.17 However, the same forces that make a Syriac text add ‘one’ might also have been operative in other Aramaic texts which might have influenced D. But though the agreement with respect to ‘one’ could be linguistically motivated rather than genetic, the agreement between SC and D in lacking dí ô² ëáëyóáé still suggests common genetic origin. Luke 20:9. See p. 139 on Mark 12:1 above. Without the support of SCP cited in NA27, the balance of evidence for the inclusion of ôéò shifts decidedly away from txt. in John 4:7. NA27 cites SC with support of ôéò before ãõíÞ, which otherwise has only slight Greek support: ) pc. for the John 6:9. NA27 cites SP with addition of fí after ðáéäÜñéïí. Although this citation is not strong, the absence of an indefinite from C, with just , is sufficiently unexpected that one wonders whether C might not be a better witness to the omission of the indefinite article than SP are to its inclusion. The context in S is rather paraphrastic, while C is rather literal for the whole verse. We may in this case, therefore, be dealing with a layer in C that is more imitative of the Greek. On the other hand, the citation in Luke 12:25 by NA27 of SCP in support of fíá after ðy÷õí is not

17 Chase, The Syro-Latin Text of the Gospels, 8.

141

THE INDEFINITE ARTICLE

problematic. Here in is functioning as a numeral, not as a marker of indefiniteness. In general the problem with all these citations in NA27 is that they are unidirectional. If they generally represented real aspects of the Vorlage we might expect that Syriac support for the inclusion or omission of the indefinite article would be cited on both sides, probably with approximately equal frequency. That the citations go one way suggests that we are dealing with a feature of translation not of Vorlage. There are a couple more citations related to, but different from, the type dealt with above. For convenience they are dealt with here. Mark 7:1 and 7:2. In the former of these verses txt’s ïj Öáñéóásïé êáß ôéíåò ô§í ãñáììáôÝùí is rendered by SP ‘Pharisees and scribes’. A similar construction is ignored by S in the next verse when the Pharisees and scribes instead of seeing ôéíNò ô§í ìáèçô§í ášôï™ see simply ‘his disciples’. This case joins a whole range where references to disciples are simplified and unified (see pp. 103–119). ) ) ( Luke 9:8. NA27 cites SC’s ' to support åpò instead of ôéò in txt’s ðñïöÞôçò ôéò ô§í Pñ÷áßùí can correspond to either åpò or ôéò.18 PíÝóôç. However,

3.2 THE GREEK DEFINITE ARTICLE Early Syriac Bible translations are not often used as significant witnesses to the presence or absence of a definite article in Greek. Of course Syriac may attest the presence 18 Syriac occasionally uses the construction

John 18:15 where S represents Tëëïò ìáèçôÞò by and P by + *) ( .

, e.g. in

( *

(

142

ARTICLES AND PARTICLES

or absence of an article, when such a variant has the potential to change clause structure, as in Mark 1:4 with txt’s dãÝíåôï EÉùÜííçò ¿ âáðôßæùí dí ô† dñÞìv compared with the variant dãÝíåôï EÉùÜííçò âáðôßæùí dí ô† dñÞìv. This variant is legitimately represented in NA27 and UBS4 (see p. 4). Nevertheless, due to constraints of space, NA27 is often forced to make decisions about the presence or absence of the article for its apparatus. Readers need to be aware of this because the citation of Syriac witnesses for one particular reading can produce the misleading impression of a degree of support for a reading which it does not really deserve. The following examples are selective. Mark 1:1. P does not give greater support to txt’s õjï™ èåï™ than to the variant õjï™ ôï™ èåï™. Neither do the Latin and Coptic witnesses cited in favour of the former. Of course, they are cited primarily to attest the reading ‘son of God’, in contrast to those witnesses that omit this phrase. Mark 1:2. NA27 and UBS4 hold P to witness for txt’s ô² EÇóáÀu ô² ðñïöÞôw, while some Greek witnesses support merely EÇóáÀu ô² ðñïöÞôw, a reading which P could equally well support. John 6:47. After ¿ ðéóôåýùí NA27 cites SC as adding åkò èåüí and UBS4 cites SC as adding åkò ô’í èåüí. Neither suggestion has any Greek support, so there is no way of knowing. John 8:54. NA27 and UBS4 cite SP particularly to show that they support reading ½ì§í as opposed to ›ì§í. There is simply no way of knowing whether they really support txt’s èå’ò ½ì§í rather than the variant ¿ èå’ò ½ì§í.

143

PREPOSITIONS

3.3 PREPOSITIONS 3.3.1

ÞØ~

It has already been shown that the preposition ÞØ~ can behave quite differently from its partial Semitic cognate, Hebrew k.19 A fortiori the possibility needs to be considered that ÞØ~ may behave in ways different from Greek ©ò and ©óåß, with which, despite differences, it overlaps in areas of meaning and function. In the Gospels there are a number of variants relating to the presence or absence of an approximator (©ò or ©óåß) before a numeral. The data for ©ò as a numerical approximator suggest that it might occasionally be left untranslated. Reference

txt

Mk. 5:13

©ò

Mk. 8:9

©ò

Lk. 1:56

©ò

Lk. 8:42

©ò

Jn. 1:39

Greek variants ©óåß

S

C

P

ÞØ~

not extant not extant not extant

ÞØ~

ÞØ~

ÞØ~

ÞØ~

ÞØ~

©ò

©óåß; ) pc omit ©óåß; D pc omit ©óåß; D 579 omit G omits

ÞØ~

ÞØ~

ÞØ~

Jn. 4:6

©ò

©óåß

nothing

ÞØ~

nothing

Jn. 6:10

©ò

©óåß

nothing20

nothing

nothing

ÞØ~

ÞØ~ ÞØ~

19 Williams, ‘“According to All” in MT and the Peshitta’, 107–

109.

20 S’s account of the Feeding of the Five Thousand is peculiar

in that it has no mention of the number of men who ate in John

144

ARTICLES AND PARTICLES

Reference

txt

Jn. 6:19

©ò

Jn. 11:18

©ò

Jn. 19:14

©ò

Jn. 19:39

©ò

Greek variants ©óåß; ¹; 28 omits D W* al. omit 66 ©óåß; * omits ©óåß

Jn. 21:8

©ò

no variant

S

C

P

,

nothing

,

nothing

not extant not extant not extant not extant

,

not extant not extant omits phrase

, , ,

In the six Gospel accounts of the feeding of a multitude the Greek narrative ends by saying that there were five or four thousand men (Matthew 14:21, 15:38, Mark 6:44, 8:9, Luke 9:14, John 6:10). In several of these verses txt contains an approximator before the numeral (©óåß in Matthew 14:21, Luke 9:14; ©ò in Mark 8:9, John 6:10). In one case, Matthew 15:38, NA27 cites SCP as supporting txt in not having ©ò or ©óåß before the number, and in another, Matthew 14:21, NA27 cites SCP as supporting the omission of txt’s ©óåß (variant, ©ò). In both cases, Syriac witnesses are cited in support of the lack of an approximator. The validity of this assessment needs to be weighed against other locations where Syriac witnesses likewise lack an approximator. The lack of an equivalent of in P) or of txt’s txt’s ©óåß in Luke 9:14a SC (contrast , ©ò in John 6:10 SCP (on S see n. 20) but the inclusion of , for ©ò in Mark 8:9 SP make it difficult to establish clear tendencies.

6:10. Instead one must wait till 6:13 where the number is given, without approximator.

PREPOSITIONS

145

The inclusion or lack of an approximator in Syriac texts is not generally an issue of syntax, but rather can be related to the felt importance of marking approximation. Several approximators which are significant for concord in the historical account of the four Gospels are retained in Syriac witnesses: that Jesus was approximately 30 (Luke 3:23 SP), that the Transfiguration took place approximately 8 days after the previous events (Luke 9:28 SCP), that the space between Peter’s second and third denials was approximately one ‘hour’ (Luke 22:59 S; contrast CP), and that at the crucifixion it was approximately the sixth hour when darkness came over the land (Luke 23:44 SCP). The variation in the witnesses to Luke 22:59 can be explained on the grounds that, of the instances given, the approximator there is the least important for evangelic concord. If we accept that the likelihood of an approximator being represented in the Syriac witnesses is related to its narrative importance we may conjecture that even if ©óåß in txt’s [©óår] PíN ðåíôÞêïíôá in Luke 9:14b had been in the Vorlagen of the Syriac versions they probably would not have represented it. NA27’s note ad loc. mentioning P as a witness for the omission of ©óåß can have no great certainty attached to it. can Besides its function marking approximation , also introduce a simile. However, in Luke 11:44 txt’s simile for the Pharisees (dóôc ©ò ôN ìíçìåsá) corresponds to a -. ‘you are tombs’, blunt metaphor in SC: + which NA27 cites as agreeing with D’s dóôc ìíçìåsá. The citation is proper, though the analogy of other cases shows

146

ARTICLES AND PARTICLES

that it is at least possible for the Syriac side of this agreement to be explained as translational. In Luke 24:6 where txt has ìíÞóèçôå ©ò dëÜëçóåí ›ìsí hôé Dí dí ô† Ãáëéëáßu, NA27 cites SC along with D in support of ”óá for ©ò. The Greek words have just about enough resemblance to seek to explain D’s reading as an accidental inner-Greek alteration. Yet, another explanation presents itself: ©ò following on from ìéìíÞóêïìáé does not fulfil a is acceptable Greek, but Syriac , similar function after the Syriac word to ‘remember’ (here: ! ). It is almost inevitable that some Syriac expression with d lath would be used in representing txt. SC read: / 0 ( ! ‘remember that which he said’, which ( ! , which can be derived from txt. P has merely / is no doubt a correction to txt, since it can legitimately be translated ‘remember how he spoke’. However, it is a correction with a disadvantage, since it can also be translated ‘remember that he spoke’, and there is no formal way in P’s text of disallowing this reading. The text of SC is therefore, in fact, a very good way of rendering txt, since it unambiguously puts the emphasis on the content of something Jesus said while with the disciples in Galilee but precludes the interpretation that the angel is merely wanting to draw attention to Jesus’ action in speaking with 0 in them in Galilee. Furthermore, support for / the OS as a translation of ©ò + ëáëÝù in Luke 24:6 can be found in John 7:46 (see UBS4 apparatus ad loc.). 3.3.2 Greek Case Distinctions in Syriac In Luke 19:44, NA27 cites SCP for dí óïr ëßèïí dðr ëßèv as opposed to txt’s ëßèïí dðr ëßèïí dí óïß. Of course such a

PREPOSITIONS

147

difference between dðß with the accusative and with the dative cannot be detected in Syriac translation, and the note in the apparatus is a combination of two variants to save space. It is similarly unlikely that we can know whether Syriac witnesses support ðáñÜ with dative or genitive in John 8:38 (ad fin.). NA27’s note that SP has ô² ðáôñr ›ì§í ðïéåsôå where txt has ôï™ ðáôñ’ò ðïéåsôå is another note resulting from the combination of two separate variants.21 3.3.3 Repeated Prepositions In Matthew’s account of the triumphal entry the significance of the two beasts (21:5) as opposed to the one in parallel accounts is often discussed. One variant related to this is whether we read dðr –íïí êár dðr ð§ëïí (txt), or simply dðr –íïí êár ð§ëïí. In support of the second dðß NA27 cites CP (S not being extant). However, Syriac seems to avoid double duty prepositions, and therefore NA27’s citation does not seem certain. CP could have added the second es due to preferences internal to Syriac. 3.3.4 Semiprepositions In Matthew 2:15, where txt has líá ðëçñùè† ô’ ¼çècí ›ð’ êõñßïõ äéN ôï™ ðñïöÞôïõ ëÝãïíôïò, NA27 suggests that S attests the addition of ôï™ óôüìáôïò EÇóáÀïõ after äéÜ. Since S is the only witness for this, it is a bold retroversion. It is a retroversion that is no doubt included mainly not for the sake of the prepositional phrase äéN ôï™ óôüìáôïò, but rather for the sake of the name ‘Isaiah’. As the following 21 For discussion of the variant with which it is combined and

of UBS4’s joint note see p. 94.

148

ARTICLES AND PARTICLES

quotation comes from Jeremiah, ‘Isaiah’ could be seen as the lectio difficilior, a reading whose origin is hard to explain, but which would readily be expunged by scribes seeking to rid the text of inaccuracies. A similar reading is found in Matthew 1:22 where S(C) support D, pc, the Harclean, Old Latin, and one Sahidic manuscript in reading äéN EÇóáÀïõ ôï™ ðñïöÞôïõ. Although the reading NA27 reconstructs for S is alluring, its charms should be resisted. The word ‘mouth’ seems to be able to be added, not reflecting óôüìá in the original but merely as a feature of translation. NA27 represents C within parentheses in " 0 1 . However, C Matthew 1:22 because it reads has the same reading in 8:17 and 12:17 where txt has äéN EÇóáÀïõ. S is extant in Matthew 1:22, 8:17, 12:17 but reads " . Despite the differences between S and C simply it seems reasonable, given the lack of the word óôüìá in Greek witnesses in any of the four locations where 0 1 occurs in OS witnesses in Matthew, to suggest that " the Syriac is paraphrasing its Vorlage when it uses the term ‘mouth’. The likelihood of paraphrase in the OS is borne out by consideration of other examples of Matthew’s fulfilment formulae. In 2:17, SC substitute the active ‘Jeremiah said’, and in 3:3, SC have ‘concerning whom it was written in Isaiah the prophet’, despite the fact that there is no reference to ‘writing’ in the Greek. On several occasions C ‘by the hand of’ for introduces the semipreposition is a Greek äéÜ + genitive (2:15, 4:14, 21:4). Although more common semipreposition than 0 1 , there is in theory no less reason for allowing a Syriac translation to represent a Greek preposition by a semipreposition

PREPOSITIONS

149

containing the element ‘mouth’ than by one containing the element ‘hand’. In neither case can we reconstruct the is P’s regular representation body part for the Vorlage. of äéÜ or ›ðü + genitive introducing prophetic word (Matthew 1:22, 2:17, 3:3, 4:14, 8:17, 12:17, 13:35, 21:4, 27:9). To conclude, if it were judged that there was a significant possibility that S’s reading in Matthew 2:15 is not just an assimilation to the reading ‘by the mouth of Isaiah the prophet’ found elsewhere in the OS (C of Matthew 1:22, 8:17, 12:17), and that the reading therefore merits inclusion in an apparatus, it seems best to include Isaiah’s name in the variant, but not ôï™ óôüìáôïò.

3.4 THE DIFFERENCES BETWEEN ÊÁÉ ÊÁÉ AND WAW … the presence of waw (especially in S and C) by no means implies that êáß featured in their Greek Vorlage.22

The differences in use of ‘and’ between a Hebrew and a Syriac biblical text have been dealt with elsewhere.23 Here there is some degree of overlap with that treatment, particularly as regards constant features of the Syriac language. However, those who translated or revised the Gospels from a Greek Vorlage faced some challenges that those who translated the OT from one Semitic language into another did not confront.24 This meant that in some instances tendencies in the target language motivated 22 Brock, ‘Limitations of Syriac’, 84. 23 Williams, Studies, 84–99. 24 For further discussion of waw see Baethgen, Evangelien-

fragmente, 24–25; Falla, Key, Volume Two, 42–53; Lyon, Syriac Gospel Translations, 94, 121, 191.

150

ARTICLES AND PARTICLES

them not to represent êáß by waw. The most important of these are summarised below. 3.4.1 Verbal Asyndeton Asyndeton is the joining of two elements without an explicit conjunction such as waw. The following factors dramatically reduce the likelihood that waw will occur between two coordinated Syriac verbs.25 (1) The verbs have the same tense (2) The verbs are adjacent (3) The first verb is one of motion or movement, especially 0 or 2$ (4) The verbs are imperatives The first condition is important.26 An example of the lack of asyndeton because the two verbs have different tenses is 3 & & ‘she got up and Matthew 8:15 SCP was serving’. If the verbs are not adjacent waw is generally used. The occurrence of a verb of motion or movement as the first of the verbs is very high, but Joosten has already noted that it is quite regular in Matthew’s Gospel that other verbs occur in this position.27 This seems to be the 4 ‘she has case elsewhere (e.g. Mark 14:8 S "5 3 3 aromatised beforehand’, John 13:26 S !* ! 6 7*, ‘it is he with respect to whom I dip 8# ‘take, keep’). The and give bread to him’, 17:11 S omission of waw seems to occur particularly when the action of the first verb is closely connected with that of the 25 For further discussion see Joosten, Syriac Language, 133–

38; Williams, Studies, 96–99. 26 Exceptions are noted in Joosten, Syriac Language, 136. 27 Joosten, Syriac Language, 133–34.

ÊÁÉ AND WAW

151

second, either by being part of the same action, or by being preparatory to it. This unity of action is probably the underlying principle behind the phenomena in conditions and (1)–(3). It also explains the high incidence of 0 2$ as the first component, since these verbs often function only in an ancillary capacity to a following verb. Occasionally asyndeton may occur with a sequence of 3 /1 ‘he came [and] three verbs: !* 9# fell down [and] worshipped’ (Matthew 9:18 S). The first verb is a verb of motion, moving to the position from which the actions in the other verbs take place. What is meant by the verbs /1 and 9# is signalled by the two verbs together, which explains why they are found in asyndetic combination elsewhere (e.g. Matthew 2:11 CP, 4:9 SP). Other examples of asyndeton for a sequence of three verbs may be similarly explained as making up a single occurrence (Matthew 14:12 P, Mark 14:40 P). If the occurrence is not viewed as a unity then it is and 2$ to occur in first possible even for verbs like 0 position without asyndeton. Thus in Matthew 8:26 SP we ' ‘he arose and rebuked have 6 the wind and the sea’. Since the action denoted by ' is given full semantic weight (Jesus was lying down asleep and now gets up) and is not a mere prelude to the next action, waw is used between the two verbs (similarly Matthew 8:7 SCP, Mark 14:37 S, contrast P). All four above-mentioned conditions apply in Mark *$ cannot be used, as they 6:38. Therefore SP with : are by NA27, to support txt’s ›ðÜãåôå näåôå without êáß in between.

152

ARTICLES AND PARTICLES

Rather than seeing in (SP) a confirmation of D’s asyndetic construction åëèå áøáé in Mark 5:23, we should shows the expected conobserve that SP’s ' # struction without waw given its fulfilment of conditions (1)– (3) above. D’s reading would be expected if it rendered literally an Aramaic Vorlage. At any rate the Syriac side of the agreement is explained. For an explanation of the other parts of this agreement between S and D see pp. 79– 80. The supposed agreement between S and D earlier in the verse (NA27 cites (S) for D’s ðáñáêáë§í … êáß) is based on S’s use of waw to join the verbs " ‘sought’ and ‘said’. As the text lacked the conditions for asyndeton S’s reading could have been produced from txt. The other side of the fact that certain conditions make asyndeton more likely, is that when these conditions are not fulfilled waw is used, and its presence in Syriac witnesses has no bearing on the presence or absence of êáß in the Greek Vorlage. Applying this to Matthew 27:40 could shift the balance of evidence against the reading given in txt. NA27 cites (S)P alongside only )* A D pc and the Old Latin in favour of txt’s êáß in ó§óïí óåáõôüí, åk õj’ò åq ôï™ èåï™, [êár] êáôÜâçèé Pð’ ôï™ óôáõñï™. However, the waw in S and P is liable to occur since the two imperatives are not adjacent, represent two distinct actions, and the former is not preparatory to the latter. If the other reading is adopted a significant exegetical change results. Those who pass by the cross make three independent taunts: (1) This is [vocatively: you are] the one who said he could destroy the temple and rebuild it in three days; (2) If you are God’s son then rescue yourself; (3) Come down from the cross.

153

ÊÁÉ AND WAW

Taking the three taunts as independent also might explain the apparent lack of logical sequence between the phrases. In Mark 13:8, P has - ! twice and S has it but once. The waw is obligatory in all cases, criteria for asyndeton being absent. NA27’s citations of SP and P, respectively, as witnesses for êáß twice in this verse need to be deleted accordingly. The apparatus for UBS4 also needs some amendment. Given that waw is likely to be used when coordinated verbs refer to distinct actions SCP cannot be used to support the weak Greek testimony for the presence of êáß before dëèþí in Luke 22:45, despite the fact that the preceding verb is 0 , since this verb is here given full semantic weight. Other cases where the presence of a predictable waw before a verb is wrongly used by NA27 as a witness for a êáß are Luke 9:28 SCP (before PíÝâç) and 15:24 SCP (before Pðïëùë¦ò ƒí).28 In Luke 18:11 it seems that the rules for asyndeton can be used to explain more clearly the reasons for particular wording in Syriac versions: txt ¿ Öáñéóásïò óôáèårò ðñ’ò eáõô’í ôá™ôá ðñïóçý÷åôï variant ¿ Öáñéóásïò óôáèårò ôá™ôá ðñ’ò eáõô’í ðñïóçý÷åôï 6*; !&1 * ‘blessed and divided’ is gave’ or in P’s ? expected, and would occur even if the first Syriac verb corresponded to a participle in the Vorlage. There seems therefore no justification for NA27’s citation of C(P) in support of åš÷áñßóôçóåí êár häùêåí against txt’s åš÷áñéóôÞóáò äéÝäùêåí. John 13:12. SP are cited by NA27 in support of txt’s êár PíÝðåóåí against the variants Píáðåóþí or êár Píáðåóþí. # in SP could be a normal But the main verb , replacement of a Greek participle, and the waw preceding it is obligatory in the absence of conditions for asyndeton. Consequently SP’s reading could be derived from txt or either of the variants recorded in the apparatus. An extension of the fact that Syriac uses subordinate participles less than Greek is found in translation of ëÝãùí, etc., at the introduction of speech. When ëÝãùí is dependent on a preceding verb, the Syriac equivalent has . The waw of SCP at to have waw before the participle

156

ARTICLES AND PARTICLES

the beginning of Matthew 3:2 should therefore not be used by NA27 to support Greek êáß. 3.4.3 êáß as ‘Also’ When the word êáß means ‘also’ it may of course be represented by Syriac @ or @ . However, an equivalent of a êáß meaning ‘also’ and attested throughout our control editions is sometimes not present in Syriac translations: Matthew 5:39 SC, 7:12 C, 18:33a S, 18:33b SCP, 25:41 S, Luke 3:14a P, John 14:12 S; to these one must add Matthew 21:24b SC where ms 1574 omits êPãþ as a whole, and John 11:37 S where ms 33 omits êáß, but this is connected with a word order change in the context. Since ‘also’ is not always represented, there is at least some question as to whether P can legitimately be used as a witness against êáß before ôïsò ìåôE ášôï™ in Luke 6:4 (NA27), and before ôï™ óáââáôï™ in Luke 6:5 (NA27 and UBS4). In favour of the citations is that they are using P, and most of the omissions of ‘also’ are in OS. As a result slightly more doubt exists over NA27’s citation of S’s lack of an equivalent of êáß before ¿ ðáôxñ ›ì§í in Luke 6:36, though êáß is represented by @ in P, or over the citation that SP lack êáß in Matthew 24:37. The key thing to observe is the preponderance of occasions when the Syriac witnesses are invoked for the absence of ‘also’ as opposed to its presence. That it was sometimes omitted in translation seems the most natural explanation of this preponderance. In Mark 15:41, while NA27 is right that it is unlikely that the wording of SP could have been produced from a text reading êáß for txt’s ál, the outside possibility that S or

ÊÁÉ AND WAW

157

P were produced from a text reading ám êáß as D È f 1.13 ˜ should not be overlooked, when full consideration is made of the tendency of Syriac translations to overlook êáß in the meaning of ‘also’. In Luke 24:15, NA27 cites SC for the omission of êár ášôüò in txt’s êár ášô’ò EÉçóï™ò dããßóáò óõíåðïñåýåôï ášôïsò. The êáß in this instance if taken as ‘also’ might be omitted in translation. The êáß could also be understood as marking the boundary between protasis and apodosis, in which case the omission could be explained on the grounds that Syriac texts typically lack waw at the beginning of an apodosis.30 The omission of ášôüò in this verse is explained on pp. 123–24. This means that it is possible that Syriac witnesses omitted the phrase êár ášôüò as a whole. 3.4.4 Lists Syriac with considerable regularity uses waw between different items in a list, especially in OS.31 For example, a whole list of eight vices is recorded in txt in Matthew 15:19. Texts sometimes vary about the order or content of the vices, but in all Greek texts referred to in our control editions the vices are listed without êáß joining them. SC, however, have waw between every one. When Jesus lists six commandments in Matthew 19:18–19, throughout the Greek tradition the first five are not joined by êáß. On the 30 Williams, Studies, 93–94, noting especially the parallels be-

tween the example given there in 1 Kings 22:32 and the example here. Th. Nöldeke, Grammar, § 339, states the absence of waw in apodoses too absolutely. See E. Beck, ‘Grammatisch-syntaktische Studien zur Sprache Ephräms des Syrers’, 16–26. 31 Cf. Williams, Studies, 90–91.

158

ARTICLES AND PARTICLES

other hand, P has waw between every commandment, C lacks it only before the fifth listed (‘honour your father and your mother’), which contains a shift of focus from the previous commandments, and S lacks waw before the second and fifth commandments, but lacks the third commandment entirely. In the light of this regular addition of waw in representing lists, SP’s use of waw to join names in the list of disciples in Luke 6:14–16 cannot be used to support the presence of êáß in txt. In fact, all names in New Testament disciple lists in P, and OS when extant, are joined by waw (Matthew 10:2–4; Mark 3:16–19; Acts 1:13). This means that references to SP need to be deleted from four notes in NA27 to Luke 6:14–16. The use of waw in lists naturally means that in Matthew 15:31 the individual types of people with disabilities are joined by waw. P cannot therefore be used as in NA27 and UBS4 as a witness to êáß before êõëëï˜ò ›ãéåsò (for a similar list with waw see Matthew 11:5 SCP). 3.4.5 Miscellaneous Whereas Greek frequently uses êáß … êáß and ôå (…) êáß with the meaning ‘both … and’, Syriac does not use waw … waw in the same way. An equivalent of the first êáß is often missing in Syriac translation, e.g. Matthew 22:10 SCP, John 7:28 SC ( 66* omits êáß), 9:37 SP. It is therefore questionable whether SCP can be held to witness against the reading of B in Luke 24:39, even though in terms of probability SCP are unlikely to rest on that reading. Despite NA27’s suggestion that P attests êáß after líá in líá ¿ óðåßñùí ¿ìï™ ÷áßñw êár ¿ èåñßæùí (John 4:36), the Syriac

ÊÁÉ AND WAW

159

waw (present in S and C as well as P) functions quite differently from the Greek êáß, which means ‘both’. The Syriac waw, on the other hand, joins clauses, and preceding a verb in the imperfect tense is rather the equivalent of Greek líá with the subjunctive. This equivalence explains the lack of a formal equivalent of líá such as d lath in the Syriac. Just as Syriac does not accept two coordinated verbs without intermediate waw, except under special circumstances, so waw is also needed to join two coordinate + participle clauses. P therefore should not be used in NA27 to attest êáß before ©ò äéÞíïéãåí in Luke 24:32, nor can SC be used to attest its omission since they have no opportunity to use waw, having omitted the whole phrase ©ò dëáëÞóåí ½ìsí (or perhaps its meaning is subsumed under the paraphrastic equivalent of äéÞíïéãåí, namely A&1 ‘explain’). Narrative ‘and it happened’, i.e. êár dãÝíåôï or dãÝíåôï äÝ, even when attested throughout the Greek tradition, is not always fully represented in Syriac translation, being sometimes merely translated by waw:32 e.g. Matthew 7:28 C, 13:53 SC, Luke 1:23 P, 2:15 S, 3:21 S, 5:1 S, 8:1 SC, 9:18 SCP, 9:37 SC, 9:51 S, 11:1 SC, 11:27 SCP, 14:1 SC, 17:11 SC, 19:15 SCP, 24:4 SC, 24:15 SCP, 24:30 SC, 24:51 S; see also Luke 17:14 SCP, where Ë* alone omits dãÝíåôï. ãßíïìáé may also be unrepresented when it is a prelude to an action in the protasis of a conditional sentence, as in Matthew 18:13 SCP. 32 For this with reference to C, see Baethgen, Evangelien-

fragmente, 23. See also Lyon, Syriac Gospel Translations, 129–30; cf. Williams, Studies, 108–109.

160

ARTICLES AND PARTICLES

In Luke 9:28 NA27 cites SC (contra P) in support of txt’s êáß before ðáñáëáâþí following on from dãÝíåôï äÝ. 3 (= dãÝíåôï äÝ) the sentence However, in OS after may be continued by a verb preceded by waw without any reason to suppose that êáß was in the Vorlage (e.g. Luke 16:22 S). The same applies to John 10:22 where SP do not particularly support êáß before txt’s ÷åéì¦í ƒí. There are other texts where NA27 cites Syriac witnesses for the omission of narrative dãÝíåôï: Mark 4:4 SP (producing an agreement with D), Luke 8:40 SCP, 9:57 SCP, 10:38 SC. In this last case NA27 cites P for the inclusion of dãÝíåôï. Such a citation for the inclusion of dãÝíåôï is fully justified.

3.5 ADVERBS In this section we deal with a number of Greek adverbs and their representation in Syriac. In broad outline the data below suggest that the OS tradition in particular did not feel constrained always to represent these elements when they were present in the Greek. It should be remembered that the OS translation is not one that seeks formal correspondence with its Vorlage, and that in many cases no equivalent Syriac adverb was readily at hand to match the Greek one. There is no reason to believe that there was a systematic decision not to represent adverbs. Rather, whether consciously or subconsciously, these adverbs were sometimes felt not to be important enough to be translated. Below is a preliminary investigation of some adverbs, but these are not necessarily all the cases where there is no direct Greek–Syriac correspondence. Sometimes there were too few examples to establish any patterns worthy of

ADVERBS

161

mention. In general text-critical practice it seems that OS more reliably witnesses to the presence of an adverb than to its absence. 3.5.1 dêåsèåí EÅêåsèåí ‘from there’ (24× in txt of the Gospels) has a direct Syriac equivalent in ( ( . This is a regular equivalent in P, and a frequent equivalent in OS. However, dêåsèåí can have different degrees of focus placed on it in the narrative. In a sentence like ïš ìx dîÝëèwò dêåsèåí fùò Uí Pðïä²ò ô’í hó÷áôïí êïäñÜíôçí ‘you shall not get out of there until you pay the last farthing’ (Matthew 5:26), dêåsèåí has considerable focus as the ‘there’ where the unsuccessful defendant will have to spend his days. On the other hand, in a sentence like êár ðñïâNò dêåsèåí åqäåí Tëëïõò äýï Päåëöïýò ‘and going on from there he saw another two brothers’ (Matthew 4:21), there is comparatively little focus on dêåsèåí. It marks the location that is going out of view as the new one comes in. In fact, little would be lost semantically if dêåsèåí were dropped from the sentence since ðñïâÜò already has within itself the sense of movement from one location to another. The level of focus on dêåsèåí seems to be a consideration in whether it is represented in S. On three occasions despite the presence of dêåsèåí throughout the Greek tradition S lacks any equivalent. One is Matthew 4:21, discussed above,33 and the others are Matthew 9:27 and 19:15. In Matthew 9:27 the situation is similar to that in 4:21. Jesus is moving on to a new location, and the focus 33 Here S does add

the function of dêåsèåí.

=

‘further’, which may fulfil some of

162

ARTICLES AND PARTICLES

has shifted from the old one. In 19:15 the focus is also more on Jesus’ movement than on the place he is leaving. The validity of this explanation is also confirmed by the reading of S in Matthew 9:9 where we seem to have a ( ( B partial doublet. S reads ‘and he removed from there and when our Lord passed ( B can be explained as a along’. The phrase ( is the attested OS corrective addition, since reading in Matthew 9:27 where S lacks an ‘equivalent’ of ( B , on the other hand, is more literal dêåsèåí. ( and therefore likely to have come later, becoming in P the regular phrase for such narrative markers of movement with dêåsèåí. Its presence in OS at, for example, Matthew 11:1, 12:9, 12:15, 13:15, shows how already our extant OS witnesses may present a revised text, or possibly that the reviser of Matthew 9:9 S modelled his revision on equivalencies he saw elsewhere in OS. Either way, it seems that at the earliest traceable stage of OS it was quite regular not to have a formal equivalent of dêåsèåí when it represented a horizon that was disappearing from view in the narrative. But it is in exactly this sort of context that NA27 twice cites S for the omission of dêåsèåí (Mark 7:24 and John 11:54). 3.5.2 hîù Greek hîù is very regularly rendered by * in all three Syriac texts in a wide variety of situations. There is then prima facie strong support for the few notes in NA27 using Syriac witnesses to attest the absence of hîù in Greek: SP in Luke 24:50, SC in John 6:37, and S in John 18:29. All three passages use Syriac A1 in the peal or aphel, which

ADVERBS

163

contains inherently the sense of ‘out’. It should be considered therefore whether the omission of a formal equivalent of hîù in these locations could be because it would be semantically redundant. The same omission can be found in Mark 11:4 S,34 and Luke 22:62 C [1242 l48 l150 l1642 omit hîù].35 However, someone seeking to explain the omission as a matter internal to Syriac would have to explain why of only a few examples of nonrepresentation of hîù, so many occur precisely where Greek witnesses divide over its presence.36 This agreement might occur by coincidence as in many cases the omission of hîù does not detract from the sense, allowing scribes to omit this element accidentally with comparative ease. 3.5.3 hôé Greek hôé is a word which receives no fixed equivalent in Syriac because it fulfils so many different functions. Glossed as ‘yet’, it can be applied to time and to other quantities. When, for instance, a number of Greek witnesses concur about the omission of hôé and it is also not represented in Syriac witnesses the Vorlage Hypothesis is naturally invoked. However, it is possible that factors causing variation in Greek manuscripts and in Syriac texts could be partly independent. Some variation in Greek 34 Though % ‘courtyard’ indicates that the foal is outside. 35 The latter case has A1 . We should also consider John 9:35

S, where D W omit hîù, and John 20:11 SP, though there hîù is also omitted by )* and A, and there are Greek witnesses suggesting Mary was in the tomb. 36 Cf. also the striking agreement between S, È, some Old Latin witnesses and the Sahidic reading ‘will be thrown down’ in John 12:31 where txt has dêâëçèÞóåôáé hîù.

164

ARTICLES AND PARTICLES

manuscripts could be explained by the small size of the particle and therefore its natural oversight in transmission. Variation in Syriac witnesses could be ascribed to the lack of a ready Syriac equivalent in many contexts in which it occurs. Of course, variation in Syriac manuscripts cannot be entirely independent of variation in Greek ones. Widespread variation in Greek manuscripts inevitably will also affect the Greek manuscripts from which Syriac translations are made. Nevertheless, it is as well to consider the cause of Syriac variation as being potentially twofold. A Syriac tendency to omit a representation of hôé is suggested by the three places in the Gospels (out of 37 occurrences in txt) where Syriac witnesses lack any equivalent despite the presence of hôé throughout the Greek tradition. Reference Lk. 8:49

txt hôé ášôï™ ëáëï™íôïò

Syriac S/

;C

/

Lk. 9:42

hôé äc ðñïóåñ÷ïìÝíïõ ášôï™

SCP !*

Jn. 16:12

hôé ðïëëN h÷ù

S B*

= B9#

Three out of 37 may not seem like a very high rate of omission, but alongside these we may place two other cases where Greek manuscript support for the omission is weak: Matthew 19:20 P [with 2145], Luke 22:71 S [with 205 579 l524]. These five cases take the edge off our confidence in the correctness of NA27’s notes saying that C and SP witness against hôé in John 4:35 and 11:30, respectively. There is no particular reason to question NA27’s notes relating to hôé in Mark 8:17, 12:6, or Luke 22:37.

ADVERBS

165

3.5.4 }äç Of the 41 occurrences of }äç ‘already’ in txt of the Gospels the majority are not represented in one or more Syriac witnesses. The best Syriac semantic, if not functional, ( (e.g. Luke 12:49 equivalent of }äç is probably SCP), which is not nearly so common as the Greek word.37 Syriac also has 3 (e.g. Matthew 17:12 P), which also 7 (e.g. Matthew 5:28 P), which functions as ‘behold’, 3 (Luke 24:29 P), also functions as ‘immediately’, and which usually means ‘now’. Since the best equivalent is not common and the other possible expressions carry other meanings, the most common single ploy in the Syriac translations is not to represent the word. In eleven cases Syriac witnesses fail to represent }äç formally while the entire Greek tradition supports its presence: Matthew 5:28 SC, 14:15 CP,38 Mark 4:37 P, 6:35a SP, 15:42 P,39 Luke 7:6 SP,40 11:7 SC, John 3:18 SC, 4:51 CP, 9:22 SP, and 9:27 SP. In a further fifteen cases Syriac witnesses fail to represent }äç and Greek support for the omission is not strong: Matthew 14:24 SCP, 17:12 SC, 24:32 S, Mark 6:35b SP, 11:11 SP, 13:28 SP, Luke 19:37 SCP, 21:30a SCP, 21:30b SC, John 6:17 SCP, 7:14 SCP, 13:2 SP, 19:28 P, 21:4 SP, 21:14 SP. In these latter cases, though it is possible that the Vorlagen of the Syriac 37 It occurs in the P Gospels as an equivalent of }äç seven

times: Mark 15:44, Luke 12:49, John 3:18, 4:35, 11:39, 15:3, 19:33. 38 S lacks the whole phrase in which }äç is found. 39 Again S lacks the whole phrase in which }äç is found. 40 S adds 3 ‘behold’ later in the sentence, which may fulfil some of the function of }äç in the Greek.

166

ARTICLES AND PARTICLES

texts lacked }äç in isolated instances this does not provide an adequate explanation in general. In each case it is different Greek witnesses that show an omission of }äç. It is hard to believe that successive Syriac translations could coincidentally have used Vorlagen showing variants at exactly the points where }äç appears in txt, when no extant witness in the Greek tradition shows such a consistent pattern of variants. In Luke 23:44 SCP, 24:29 SC, and John 5:6 SCP, Syriac witnesses also lack an equivalent of }äç when a more substantial body of Greek witnesses attests its absence. Further confirmation that the lack of an equivalent of }äç is generally an inner-Syriac issue and not one of Vorlage is found in the paraphrases present in some texts, reflecting the translator’s need to restructure the sentence because of the lack of a similar expression in Syriac. In John 11:17 instead of saying that Jesus found that Lazarus ‘had already (}äç) been in the tomb for four days’, S says that Jesus found that ‘they had buried Lazarus four days ( )’. In Mark 4:37 for txt’s ªóôå }äç before him (/ C ( ( " from that of SC: >/ D!- >/ 8 #5 ‘and there will come days when your enemies will surround you and will oppress you from every place’. SC have only two verbs to express the enemies’ actions against three in D and txt. 16 On its own dêÝëåõóåí ášô’í P÷èyíáé could mean ‘he

commanded him to be led away’.

282

MISCELLANEA

Nevertheless D and SC might have a closer relationship. SC might be derived from D’s reading if it were supposed that OS abbreviated three similar verbs or verbal phrases to two. Since the first of these phrases (D’s êáé âáëïõóéí åðé óå ïé å÷èñïé óïõ ÷áñáêá) is the one least obviously represented semantically in SC this method of abbreviation would very naturally lead to ignoring D’s dðr óÝ.17 The consequence of this is that it does not seem a straightforward interpretation of the evidence to say that D and SC omit dðr óÝ. D has it, and SC may well fail to represent clearly the whole phrase in which it stands. John 1:6. In txt John the Baptist is introduced thus: dãÝíåôï Tíèñùðïò PðåóôáëìÝíïò ðáñN èåï™ –íïìá ášô² EÉùÜííçò. NA27 cites C along with )* D* Ws Irlat as sup7 porting ƒí before –íïìá. In C the verse reads: , ( 2 ! 2 ‘There was a man who was sent from God. His name was John.’ The ‘was’ after 2 ‘his name’ would be position of expected in Syriac, whatever its Vorlage. There are few close parallels to the structure of this verse, though John 3:1 is the closest in the NT: ƒí äc Tíèñùðïò dê ô§í Öáñéóáßùí Íéêüäçìïò –íïìá ášô². C does not survive here, ( ( but S translates the introduction thus: ‘Now < 2 (7= 6783 there was a man there from the Pharisees. His name was representing what is a clause Nicodemus.’ S thus has that is verbless throughout the Greek tradition. Moreover, Luke 2:25 S has the same structure (an existential clause 2 followed by the asyndetically followed by 17 Though evidently dðr óÝ might still be felt to have some

correspondence to the suffix on >/

C.

OTHER VARIANTS

283

name) as the OS readings in John 1:6 and 3:1, despite a different Greek structure. Luke 1:5 SP should likewise be compared, though it contains a relative clause. The best deduction, then, seems to be that the structure in John 1:6 C was the usual OS structure, and cannot be used to attest Greek ƒí. John 4:51. NA27 and UBS4 cite CP in favour of õjüò óïõ instead of txt’s ðásò ášôï™. However it is possible that C or P, with )7 ‘your son’, read the intermediate reading ðásò óïõ with È f 1 565 ˜ et al. Although ðásò is more often #@ ‘boy’ than by 67 ‘son’ this is not rendered by #@ only rarely universal (see Luke 1:54 S). Moreover, occurs with a suffix since it has less relational focus than 67 .18 Since the lad, in contrast to the figure in the Matthean and Lucan parallels, is explicitly stated to be a son, and the nobleman is called the father, it would hardly be surprising if in this context ðásò were to be rendered by 67 . Thus, though CP may properly be used to support the second person rather than the third person suffix, it is less certain whether they read ðásò or õjüò. John 10:38. For txt’s líá ãí§ôå êár ãéíþóêçôå, NA27 cites S along with D 1424 (it) for the omission of êár ãéíþóêçôå. However, as Syriac lacks the possibility of ready expression of two different aspects of the same verb, and as SCP have been shown to avoid redundant repetition,19 one would almost expect the reading of S, with a single equivalent of ãéíþóêù, to result if its Vorlage was txt. C is not 18 Of the four suffixed occurrences in P, three are in a single

passage and another in its synoptic parallel: Matthew 8:6, 8, 13, Luke 7:7. 19 See, e.g., pp. 263–64.

284

MISCELLANEA

extant, and P follows a variant involving ðéóôåýù. The agreement between S and D would result not only from S being translated from a text such as D, but would be likely to result if D were influenced by any Aramaic text, since all Aramaic dialects would have the same restriction on expression of verbal aspect as Syriac. John 13:2. txt reads åkò ôxí êáñäßáí líá ðáñáäïs ášô’í EÉïýäáò Óßìùíïò EÉóêáñéþôïõ and NA27 records that SP along with many Greek texts place the líá clause after EÉóêáñéþôïõ and thus bring the possessed êáñäßá and the possessor, Judas, into immediate proximity. One problem with NA27’s citation is that the Syriac construction for possession could not tolerate disruption by a subordinate clause even if such a disruption were in the Vorlage. John 14:26. S attests " /! after ‘send’, which is, reasonably enough, taken in NA27 to attest ›ìsí after ðÝìøåé. However, could the presence of " /! here be in some way related to its absence at the end of the verse where another Greek ›ìsí occurs?

8C

ONCLUSIONS

In the preceding pages we have briefly surveyed some features in which the Syriac translations OS and P systematically diverge from extant Greek texts. It has been argued that it is plausible and in many cases probable that these divergences were produced during the process of translation. The problems involved in explaining some of these Syriac divergences as resulting from a variant Vorlage are immense. Though the initial step of positing that a lost or poorly attested Greek reading has been preserved in Syriac is not in itself unreasonable, many of the variants dealt with here are not just isolated readings: to pursue the Vorlage Hypothesis one would have to suppose that features (whole groups of similar readings) were lost in the Greek transmission of all four Gospels. Moreover, not even that would be an adequate explanation. If one wants to maintain the Vorlage Hypothesis here, one has to posit multiple lost features, that is, distinct groups of readings that were independently lost in extant Greek manuscripts. To put this in more concrete terms: in what has preceded it has been argued that in certain situations both the addition of the or and the use of an active for a passive indefinite were features of the OS. To avoid the Translation Hypothesis one must suppose that Greek manuscripts have 285

286

CONCLUSIONS

not only lost one group of readings, but that they have quite independently lost another group. Moreover, no rationale can be given for this to occur within Greek. In both cases, however, a consistent rationale can be given for why the Syriac is as it is. This makes the Translation Hypothesis the preferred explanation for a significant number of features. Yet the features dealt with in this book are only a selection of those in which Syriac and Greek texts systematically diverge, albeit a selection of what the author considers the most important divergences. Coverage of further topics would almost certainly reveal further areas in which the Translation Hypothesis should be applied. The conclusion that the Translation Hypothesis has been underused, since adequate criteria for its use had not been articulated, may be seen as enhancing some existing developments in modern textual scholarship. In comparison with the work of previous centuries, NA27 puts relatively little emphasis on the place of the early versions of the Bible. These were given pride of place in works such as the great polyglot Bibles of Paris (1645) and London (1657), and occupied relatively large amounts of space in the critical apparatuses of later Greek New Testaments.1 In NA27, however, versional citations are normally restricted to Coptic, Latin, and Syriac. The papyri discovered during the twentieth century have greatly increased the Greek documentation for the New Testament, and have removed the necessity of relying 1 Such as Tischendorf’s Editio Octava Critica Maior and von

Soden’s Die Schriften des Neuen Testaments (especially disproportionate in the third apparatus).

CONCLUSIONS

287

exclusively on translations and patristic citations for access to the second century text. The arguments in this book can therefore be seen as simply granting further grounds for decreasing stress on some versional witnesses for establishing certain specific types of minutiae of the wording of the original text. The versions remain as vital as ever for many larger variants, and naturally where they give unambiguous testimony about more minor variants they should be used as they always have been. However, their testimony is ambiguous more often than is widely supposed by users of critical apparatuses. This is not a call to ignore the versions. Rather, a practical consequence of the preceding discussion is that more work than ever needs to be done on the Syriac versions in order to establish fully their proper use. There is still a vast amount of research to be undertaken. Many of the errors in use of the versions have occurred because the versions have been consulted rather than analysed. This is a longstanding and deeply rooted scholarly practice of which to date virtually all critical editions of the Greek NT have been guilty to varying degrees. Though it is particularly notes of NA27 that have been queried here, it should be remembered that this edition is among the best in its treatment of Syriac. Moreover, the notes have been produced by the diligent following of established scholarly procedures. The way the procedures have been followed in NA27 is not to be criticised; it is the procedures themselves—which have developed over time and with wide consensus—not the contents of a particular edition, that need to change. Versional consultation rather than analysis

288

CONCLUSIONS

is a habit built on convenience: it is easy simply to turn to a version and see whether it has a proper name or not. It is quite another matter to establish whether a proper name in a version is likely to reflect one in its Vorlage. To use versions properly a study needs to be made of each feature of the translation for which it is consulted. Too often critical notes have been drawn up on the basis of a prejudicial equivalence—the equation of a word in the translation with one in the Vorlage without examination of whether the words were treated as equivalent by the translator. For instance, a modern scholar may have learned Greek Pðïêñßíïìáé as ‘answer’ (or a similar word in his or as her native tongue) and also have learned Syriac and ‘answer’. The wrong step is then to treat Pðïêñßíïìáé as a translational equation. In fact, though these two words were equated in later Syriac translation, in as much as it is by the OS Pðïêñßíïìáé is rendered by . Neither nor is more normal an equivalent than the other.2 The only way for the compiler of a textual apparatus to avoid basing judgement on prejudicial equivalence is to study each lexeme independently— evidently a vast undertaking. Thus far the conclusions about the text-critical use of OS and P. The preceding study, however, has also uncovered some important features about these texts themselves. First, there is the profile of these texts as translations. Broadly speaking P is more literal than OS, i.e. it has a greater level of formal correspondence with its Vorlage. 2 See pp. 255–62.

CONCLUSIONS

289

Nevertheless, this is only a generalisation, and it is important to note the occasional aspects in which P is less literal.3 Secondly, OS is a more predictable translation than has been previously thought. Many of its formal departures from the Greek are consistent and have identifiable motives. In so far as its formal departures from the Greek are constrained by motives it cannot be described as a ‘free’ translation. Thirdly, from a textual point of view OS is more ‘normal’ relative to other witnesses than has been previously thought. What is meant by this is that this study has narrowed the gap between the readings of OS and those found within extant Greek witnesses. There are fewer unexplained singular readings in OS, and some of its notable alleged singular readings (e.g. Matthew 1:16 S) have been explained on the basis of extant Greek texts.4 The likelihood that OS witnesses preserve important readings not preserved in Greek witnesses has therefore been reduced. Related to this it has also been shown that some of its formal agreements with isolated Greek witnesses can be explained on a basis other than that it was translated from a Vorlage with their reading. Nowhere is this more striking than in the case of agreements between OS (or P) and D (see Appendix 3). The chief contention here is that at a number of points the Syriac translations show agreements with D which are also part of the translation profile of the Syriac. Consequently it is possible to explain the Syriac text 3 See p. 73. 4 See pp. 240–44.

290

CONCLUSIONS

as having arisen from a Greek Vorlage containing a reading other than that of D. What this does not explain is the reading of D itself. That lies outside this study, though it is only right here to point out that from a linguistic point of view (and therefore dropping historical considerations for a moment) there is more than one type of explanation. The readings of D and the Syriac could converge independently, or be genetically linked. Genetic linking includes the Vorlage Hypothesis—that the Syriac was translated from a Vorlage with readings like that of D—and the reverse possibility of some kind of Syriac influence on the Greek textual tradition of D.5 Independent convergence could be coincidental (in the case of single readings), or due to coincidentally similar tendencies. Independent convergence could theoretically also be motivated rather than coincidental. Motivated convergence could include hypothesised convergence between Syriac texts and D due to the influence of some Aramaic linguistic features upon the latter. Provided the features were common to Syriac and to the relevant hypothesised Aramaic source of influence convergence would be expected. Which of these several explanations is relevant, in which texts, and to what extent, I leave to others. The chief point should not be lost: for a number of striking ‘agreements’ between Syriac witnesses and D, the Syriac rendering would also be expected if the Syriac were translated from a Vorlage quite different from D. This has the effect of limiting the demonstrated Greek circulation 5 This possibility was famously championed by F.H. Chase in

The Old Syriac Element in the Text of Codex Bezae, 1893, and in The Syro-Latin Text of the Gospels, 1895.

CONCLUSIONS

291

of some of D’s readings. Moreover, since the Syro-Bezan ‘agreements’ whose Syriac side fits the translation profile of Syriac were only discovered incidentally rather than by systematic research, and since only a limited number of features of the Syriac translation profile are discussed here, it is likely that the number of Syro-Bezan agreements that could be thus explained is considerably larger than Appendix 3 suggests. Alongside re-evaluation of the relationship between D and Syriac witnesses must be a re-evaluation of the textual affinity of OS. This must not be done on the basis of any readings whose Vorlage is uncertain, but only on the basis of a much more limited number of readings where clear allegiance is shown.

APPENDIX 1: BRIEF RULES FOR THE USE OF SYRIAC IN NT TEXTUAL CRITICISM This appendix aims to state concisely a number of rules of thumb that can be used in handling the evidence of OS and P. These principles have been stated starkly and without nuance since too much qualification would risk losing sight of a useful generalisation. For a more detailed and qualified discussion the main text of the book should be consulted. These rules do not exhaust the cautions given within this book and are not intended to be a substitute for carrying out a full search on a word or construction and its usage. (1) Do not use OS or P to attest the presence of [¿] EÉçóï™ò or any personal name within narrative if that person is already contextually implied. (2) Do not use OS to attest the absence of ôá™ôá with a verb of ‘saying’ or ‘hearing’. (3) Do not use the Syriac sequence Verb–Object–Verb–Object or Verb–Verb–Object to attest an explicit object after the second verb. (4) Do not use Syriac witnesses to attest a dative pronoun after the verbs ‘forgive’ and ‘give’. (5) Do not use Syriac witnesses to attest a possessive with something that is inherently possessed (such as body parts, 293

294

APPENDIX 1

personal attributes) when a specific possessor is implicit in the narrative. (6) Do not use Syriac witnesses to attest a possessive with a relational term when the relative denoted by the possessive is implicit in the narrative. (7) Do not use OS witnesses to attest ášôï™ after the arthrous plural of ìáèçôÞò referring to Jesus’ disciples. Do not use P for this purpose if it agrees with OS. (8) Do not use Syriac witnesses to attest the absence of ášôüò when it is a subject pronoun. (9) Do not use Syriac witnesses to attest ôéò or åpò when used as indefinite articles. (10) Do not use Syriac witnesses to reconstruct unattested semiprepositions. (11) Do not use Syriac witnesses with waw to attest the presence or absence of êáß between two verbs. (12) Do not use Syriac witnesses with waw to attest the presence of êáß joining items in a list. (13) Be careful using OS witnesses to attest the omission of the adverbs dêåsèåí, hîù, hôé, ìüíïí, í™í, and ôüôå. Never use OS to attest the absence of }äç. (14) Do not use Syriac to attest the singular of Tñôïò, or the plural of êáñðüò. (15) Do not use Syriac to attest a plural verb when the subject is compound. (16) Do not use OS to attest an active as opposed to a passive, especially when ›ðü + genitive accompanies a passive verb in Greek witnesses, or the verb is a form of öÝñù or Tãù. (17) Do not use Syriac to attest that ðÜíôá follows ôá™ôá.

BRIEF RULES

295

(18) Be careful using Syriac witnesses to attest the order of a pair of items since Syriac can sometimes reverse the order of a pair. (19) Do not use Syriac witnesses to attest the order Name– Function as opposed to Function–Name. (20) Do not use Syriac witnesses to attest a form of ëÝãù as opposed to a form of PããÝëëù, dñùôÜù, or Pðïêñßíïìáé. (21) Do not use Syriac to reconstruct a dative object after PêïëïõèÝù, nor to deny the presence of another verb of motion in the immediate context.

APPENDIX 2: SUGGESTED EMENDATIONS TO THE APPARATUS OF NA27 Below are listed those places discussed in this work where sufficient doubt has been cast upon a note in NA27 that in my judgement the note clearly needs to be amended. The list does not include all notes which were questioned, but only a list of the notes which most clearly lacked justification. NA27’s stated editorial policy that ‘versions are cited only where their underlying Greek text can be determined with confidence’ (p. 63*) obviously could allow some flexibility since confidence can exist in varying degrees. A very strict application, understanding ‘confidence’ as ‘certainty’ would obviously need to include a more wide-ranging modification of the apparatus. Changes listed here consist entirely of reductions in the number of witnesses cited, and for this purpose the symbols for Syriac witnesses used in NA27 have been used. Since the list is only for the purpose of emendations by deletion of evidence a recommendation such as ‘for syc.p read syp’ should be understood as advising the removal of C from the apparatus rather than commending the note on P (about which nothing is said here). Sometimes, when only Syriac support is given for a variant, removal of that support will also mean removal of a whole note, or variant within a note. 297

298

APPENDIX 2

Ref.

Recommendation

Page(s) of Discussion

Matt. 1:16 2:15 3:2 3:15 4:20 5:11 6:15 6:28 6:32 7:16 8:21 8:34 9:2 9:21 9:36 10:24 12:1 12:4 12:22 12:31 13:15 13:31 13:36 14:9 14:12 14:15 14:18 14:19

om. sys and syc om. sys or cite in Latin for sy read syh om. sys.(c) om. sys.c.p om. (syc) for syp.h read syh om. (syc) om. syc for sy read syh for sy read syh om. sys for sy read syh om. sys for syp.h** read syh** for sy read syh om. syc or cite in Latin om. syp for sy(s.c.p) read syh om. sys.c om. sys.c.p om. (sys.c) for sy(p).h read syh om. sys.c.p om. 2nd sys.c.p for 1st sy read syh; om. syc.p for syp.h read syh for sys.c.hmg read syhmg

240–44 147–49 155–56 275–76 120–21 224–25 62–63 179–80 204 176 114 276–78 61–62, 121–22 168–69 33 100–101 79 175 190–91 63 84 252 36 65–66 84 114, 167 245 177–78

299

EMENDATIONS TO NA27

Ref.

Recommendation

14:22 14:24 15:1 15:2 15:4 15:12 15:31 15:33 15:36 16:5 16:11 16:12 16:20 17:10 18:25 18:28 19:10 20:13 20:17 20:21 20:23 21:5 21:11 21:28 23:3 24:8 24:33 24:34 24:45

for sy read syh om. 1st syc.p for sys.c.h read syh for sy read syh om. sys.c.p (2×); for sy read syh for sy read syh for syp.h read syh for sy read syh for sy read syh for sy read syh for syp.h read syh for syp.h read syh for sy read syh for sy read syh for sy read syh for sy read syh for sy read syh for sy read syh om. 2nd syp for sy read syh for sy read syh for syc.p.h** read syh** for sy read syh for 1st sy read syh om. syc om. sys.p om. syp om. sys.p for sy read syh

Page(s) of Discussion 114 167 229–30 79 252, 92, 98 114 158 114 115 115 175 175 115 115 101 65–66 115 262 116 84 84 147 239 139 226–27 204 204 204 97

300

APPENDIX 2

Ref.

Recommendation

26:8 26:20 26:36 26:40 26:45 26:56 27:26 27:40 27:64

for sy read syh om. syp for sy read syh om. sys.p om. sys.p om. sys om. sys om. sy(s).p for sy read syh

Page(s) of Discussion 115 116 115 115 115 115 65–66 152–53 115

Mark 1:16 1:18 1:31 2:5 2:9 3:5 3:31 3:33 3:35 4:4 5:23 5:25 6:38 6:41 6:45 7:6 7:19 7:24

om. syp for sy read syh for syp.h read syh for 2nd sy read syh for 1st sy read syh for sy read syh om. sys.p for sy read syh for sy read syh om. sys.p om. (sys) and (sys.p) for sy read syh om. sys.p for sy read syh om. 2nd sys om. 1st sys.p om. sys om. 1st sys

44 120–21 79 199 199 79 179 99 99 160 79–80, 152 139 151 115 200 258 154–55 162

301

EMENDATIONS TO NA27

Ref.

Recommendation

8:1 10:19 10:30 11:7 11:15 11:33 12:1 12:19 12:38 13:8 13:27 13:30 14:16 14:46 15:44 15:45 15:46

om. sys.p om. sys.p om. sys for sy read syh om. (syp) for syp.h read syh om. syp for 2nd syp.h read syh om. (sys) for sy read syh; for syp.h read syh for 1st sy read syh om. sys.p for sy read syh for sy read syh om. sys om. sys for syp.h read syh

Page(s) of Discussion 115 98 169–70 199 274–75 262 139 101 279–80 153 120 204 115 77 167 84 44–45

Luke 2:28 2:33 3:7 3:9 4:17 5:23 6:3 6:14 6:15 6:16

for sy read syh for sy read syp.h om. sys.p and syc om. sys.c.p for sy read syh for sy read syh om. syp om. sys.p om. 1st and 2nd sys.p om. sys.p

84 98 187–88 176 239–40 62, 121–22 262 158 158 158

302

APPENDIX 2

Ref.

Recommendation

6:43 6:45 8:40 8:50 9:8 9:28 9:54 9:57 9:62 10:11 10:38 11:14 11:24 11:26 11:37 11:43 12:1 12:22 12:53 12:56 14:5 14:12 15:24 15:29 16:1 17:8 18:11 18:20 18:32

om. sys.p for sy read syh om. sys.c.p for sys.h read syh and om. syc for sys.c.h read syh om. sys.c and 2nd sys.c.p for sy read syh om. sys.c.p for sy read syh for sy read syh om. sys.c om. sys.c om. sys.c.p om. syc for syp.h read syh om. sys.c.p for sy read syh for sy read syh; om. syp for sy read syh om. sys.c om. sys.c.p for sy read syh for sy read syh om. sys.c.p for sy read syh for sys.c.p.h** read syh** om. sys om. sys.c.p om. syp

Page(s) of Discussion 176 84 160 259 141 160, 153 115 160 79 84 160 124–25 171 171 139–40 280 247 115, 84 101–102 230 258 99 153 93–94 115 66 48–50, 153–54 92, 98 234

303

EMENDATIONS TO NA27

Ref.

Recommendation

18:40 19:29 19:43 20:9 20:10 20:19 20:45 21:11 22:14 22:45 22:64 23:5 23:8 23:53 24:6 24:9 24:15 24:29 24:32

om. sys.c om. sys om. sys.c for sy read syh for (syp.h) read (syh) om. sys.c.p for sy read syh for sy read syc.p.h; om. sy(s).c om. sys om. sys.c.p om. syp for sy read syh om. syp for sy read syh om. sys.c for syp.h read syh om. sys.c om. sys.c; for sys.c.h read syh for syp.h read syh; om. 2nd sys.c

Page(s) of Discussion 280–81 244 281–82 139, 140 246 231 115 246, 231 117–19 153 255 155 178 57 146 204 123–24, 157 166–67, 200f. 159

John 1:6 1:25 1:49 2:12 4:7 4:36 4:42 4:51

om. syc om. syc for sy read syh for sy read syh om. sys.c for syp.h read syh om. 1st syc for syc.p.hmg read syhmg

282–83 255 259 96, 115 140 158–59 124 283

304

APPENDIX 2

Ref.

Recommendation

4:53 5:7 5:15 6:9 6:11 6:42 6:46 6:52 6:57 6:65 7:20 7:32 7:34 7:36 8:38 9:8 9:36 9:40 10:14 10:22 10:38 11:7 11:19 11:41

for sy read syh om. sys.c om. sys.c.p for sys.p.h read syh om. syc.(p) om. 2nd sys.c om. sys or use Latin for sy read syh om. sys for syp.h read syh for syp.(h) read sy(h) om. 1st syc and om. sys for sy read syh for sy read syh for 1st sy read syh om. sys.p for sy(p).h read syh om. sys om. (sys) for sy read syh om. sys for sy read syh for syp.h read syh for syp.h read syh; for sys.hmg read syhmg om. 1st sys and for sy read syh om. sys.p om. sys.p om. 2nd sys.p

11:54 12:16 12:47 13:12

Page(s) of Discussion 94 258 251 140 155 170 236 84 89–90 90–91 259 48–50, 45 57–58 58 52, 90–91 98–99 259 48–50 188–89 160 283–84 115 96 86 162, 115 171 58 155

305

EMENDATIONS TO NA27

Ref.

Recommendation

13:22 13:26 13:27 13:36 14:28 16:10 16:16 18:11 18:13 19:6 19:26 19:28 20:17 20:20 20:30 21:1 21:14

om. sys om. syp om. sys om. sys for syp.h read syh for sy read syh om. (sys) om. sys.p for sy read syh for sy read syh for sy read syh om. syp for sy read syh for sy (2×) read syh (2×) for sys.p.h** read syh** om. sys.p for sy read syh

Page(s) of Discussion 115 259 171 267 92 90–91 90–91 90–91 58 58–59 98 167 90–91 80–81 115 115 115

APPENDIX 3: AGREEMENTS BETWEEN SYRIAC WITNESSES AND CODEX BEZAE FOR WHICH A NON-GENETIC EXPLANATION IS POSSIBLE Below is a list of texts where an agreement between Syriac and Codex Bezae can readily be explained by forces of independent convergence rather than by a genetic link. These are agreements where the Syriac reading is explicable on the basis of an extant Greek Vorlage other than the reading of D. No attempt is made to explain D’s side of the agreement with the Syriac witnesses. The exact nature of the ‘agreement’ is either specified within the discussion or in the apparatus of NA27. The examples listed below have come to light incidentally in the process of investigation. Systematic analysis of Syro-Bezan ‘agreements’ would therefore be likely to add to their number considerably, as would the study of constructions and idioms which are not touched on in this book. Reference Matthew 3:17 5:11

Witness Involved in Apparent Agreement

Page(s) of Discussion

SC C

195 224–25 307

308

Reference

APPENDIX 3

6:15 13:23 14:19 14:24 16:7 17:2 18:2 23:37 26:40 27:28

Witness Involved in Apparent Agreement P S SC CP S P SC S SP S

Page(s) of Discussion 62 172 177–78 167 172 33 139–40 194–95 114 227

Mark 1:16 3:7–8 4:4 5:23 7:11 7:19 10:30 12:33 12:38 15:45

P S SP S and SP SP P S S S S

59–60 268–72 160 79–80, 152 93 82 169–70 83 279–80 84

Luke 3:9 3:16 6:43 7:22 9:26

SCP 2× C SP SP CP

176 258 176 251 93

SYRO-BEZAN AGREEMENTS

Reference

309

11:26 11:37 11:43 12:53 14:23 14:26 18:32 18:40 19:43 22:64 24:6 24:29

Witness Involved in Apparent Agreement C SC SCP SCP SCP SCP P SC SC SCP SC SC

Page(s) of Discussion 171 139–40 280 93 99 95–96 234 280–81 281–82 255 146 200–201

John 1:6 4:42 5:7 6:54 6:68 7:32 8:59 9:27 9:40 10:38 11:8 13:5 13:36 14:1 14:26

C C SC S SC C S SP S S SP SP SP S SCP

282–83 124 258 195 257 48–50 172 257 49–50 283–84 115 115 257 172–73 91

BIBLIOGRAPHY CSCO = Corpus Scriptorum Christianorum Orientalium MPIL = Monographs of the Peshitta Institute, Leiden Aland, B., et al. (eds), The Greek New Testament (4th edn; Stuttgart: Deutsche Bibelgesellschaft and United Bible Societies, 1983). ——, Nestle-Aland: Novum Testamentum Graece (27th edn; Stuttgart: Deutsche Bibelgesellschaft, 1993). Aland, B. and A. Juckel, Das Neue Testament in syrischer Überlieferung, I. Die grossen katholischen Briefe, II. Die Paulinischen Briefe, 1. Römer- und 1. Korintherbrief, 2. 2. Korintherbrief, Galaterbrief, Epheserbrief, Philipperbrief und Kolosserbrief, 3. 1./2. Thessalonicherbrief, 1./2. Timotheusbrief, Titusbrief, Philemonbrief und Hebräerbrief (4 vols.; Berlin, New York: De Gruyter, 1986, 1991, 1995, 2002). Aland, K. (ed.), Die alten Übersetzungen des Neuen Testaments, die Kirchenväterzitate und Lektionare (Arbeiten zur Neutestamentlichen Textforschung 5; Berlin and New York: De Gruyter, 1972). ——, Computer-Konkordanz zum Novum Testamentum Graece (Berlin, New York: De Gruyter, 1980). ——, Vollständige Konkordanz zum Griechischen Neuen Testament (Berlin, New York: De Gruyter, 1975–1983). 311

312

BIBLIOGRAPHY

——, Synopsis Quattuor Evangeliorum (13th edn; Stuttgart: Deutsche Bibelgesellschaft, 1985). Aland, K. and B. Aland (eds), Text und Textwert der griechischen Handschriften des Neuen Testaments, IV. Die synoptischen Evangelien, 1. Das Markusevangelium, Band 1,2: Resultate der Kollation und Hauptliste (Berlin, New York: De Gruyter, 1998). Baarda, Tjitze, ‘“The Flying Jesus”: Luke 4:29–30 in the Syriac Diatessaron’, Vigiliae Christianae 40 (1986) 313– 41. ——, ‘The Syriac Versions of the New Testament’, in Bart D. Ehrman and Michael W. Holmes (eds), The Text of the New Testament in Contemporary Research: Essays on the Status Quaestionis (Studies and Documents 46; Grand Rapids, MI: Eerdmans, 1995) 97–112. Baethgen, Friedrich, Evangelienfragmente: Der griechische Text des cureton’schen Syrers (Leipzig: J.C. Hinrichs, 1885). Beck, E., ‘Grammatisch-syntaktische Studien zur Sprache Ephräms des Syrers’, Oriens Christianus 68 (1984) 1–26. Beermann, Gustav and Caspar René Gregory, Die Koridethi Evangelien (Leipzig: J.C. Hinrichs, 1913). Bensly, R.L., J. Rendel Harris, and F. Crawford Burkitt, The Four Gospels in Syriac Transcribed from the Sinaitic Palimpsest (Cambridge: Cambridge University Press, 1894). Black, M., ‘The Syriac Versional Tradition’ in K. Aland (ed.), Die alten Übersetzungen des Neuen Testaments, die Kirchenväterzitate und Lektionare (Arbeiten zur Neutestamentlichen Textforschung 5; Berlin and New York: De Gruyter, 1972) 120–59.

BIBLIOGRAPHY

313

Brock, S.P., ‘Limitations of Syriac in Representing Greek’, in B.M. Metzger, The Early Versions of the New Testament: Their Origin, Transmission, and Limitations (Oxford: Clarendon, 1977) 83–98. ——, ‘Toward a History of Syriac Translation Technique’, in R. Lavenant (ed.), III Symposium Syriacum, 1980 (Orientalia Christiana Analecta 221; Rome: Pontificio Istituto Orientale, 1983) 1–14. Burkitt, F. Crawford, Evangelion Da-Mepharreshe: The Curetonian Version of the Four Gospels, with the Readings of the Sinai Palimpsest and the Early Syriac Patristic Evidence: Volume I, Text, Volume II, Introduction and Notes (Cambridge: Cambridge University Press, 1904). Chase, Frederic Henry, The Old Syriac Element in the Text of Codex Bezae (London: Macmillan, 1893). ——, The Syro-Latin Text of the Gospels (London: Macmillan, 1895). Ciasca, P. Augustinus, Tatiani Evangeliorum Harmoniae Arabice (Rome, 1888). Clemons, J.T., ‘Some Questions on the Syriac Support for Variant Greek Readings’, Novum Testamentum 10 (1968) 26–30. Diem, W., ‘Alienable und inalienable Possession im Semitischen’, ZDMG 136 (1986) 227–91. Falla, Terry C., A Key to the Peshitta Gospels, Volume One: ’ laph–D lath (New Testament Tools and Studies 14; Leiden, New York, København, Köln: E.J. Brill, 1991), Volume Two: H –Y dh (New Testament Tools and Studies 29; Leiden, Boston, Köln: Brill, 2000). Gelston, A., The Peshitta of the Twelve Prophets (Oxford: Clarendon, 1987).

314

BIBLIOGRAPHY

Greenberg, Gillian, Translation Technique in the Peshitta to Jeremiah (MPIL 13; Leiden, Boston, Köln: Brill, 2002). Gregory, Caspar René, Textkritik des Neuen Testamentes (Leipzig: J.C. Hinrichs, 1909). Hill, J. Hamlyn, The Earliest Life of Christ Ever Compiled from the Four Gospels (Edinburgh: T & T Clark, 1894). Horner, G., The Coptic Version of the New Testament in the Northern Dialect, Otherwise Called Memphitic and Bohairic, Vol. 1: The Gospels of S. Matthew and S. Mark, Vol. 2: The Gospels of S. Luke and S. John (Oxford: Clarendon, 1898). International Greek New Testament Project, The Gospel according to St. Luke, Part One: Chapters 1–12 (Oxford: Clarendon, 1984), Part Two: Chapters 13–24 (Oxford: Clarendon, 1987). Joosten, Jan, The Syriac Language of the Peshitta and Old Syriac Versions of Matthew: Syntactic Structure, InnerSyriac Developments and Translation Technique (Studies in Semitic Languages and Linguistics 22; Leiden, New York, Köln: E.J. Brill, 1996). ——, ‘“Le Père envoie le Fils”: La provenance occidentale d’une locution syriaque’, Revue de l’Histoire des religions 214 (1997) 299–309. Juckel, Andreas, ‘A Re-examination of Codex Phillipps 1388’, Hugoye: Journal of Syriac Studies 6, 1 (2003) [http://syrcom.cua.edu/Hugoye]. Jülicher, Adolf, Walter Matzkow, and Kurt Aland, Itala: Das Neue Testament in altlateinischer Überlieferung: I Matthäus-Evangelium (2nd edn; Berlin and New York: De Gruyter, 1972), II Marcus-Evangelium (2nd edn; Berlin: De Gruyter, 1970), III Lucas-Evangelium (2nd

BIBLIOGRAPHY

315

edn; Berlin and New York: De Gruyter, 1976), IV Johannes-Evangelium (Berlin: De Gruyter, 1963). Kerschensteiner, Josef, Der altsyrische Paulustext (CSCO 315, Subsidia 37, Louvain: Secrétariat du CorpusSCO, 1970). Khan, G.A., ‘Object Markers and Agreement Pronouns in Semitic Languages’, Bulletin of the School of Oriental and African Studies 47 (1984) 468–500. Kiraz, G.A., A Computer-Generated Concordance to the Syriac New Testament (6 vols.; Leiden, New York, Köln: E.J. Brill, 1993). ——, Comparative Edition of the Syriac Gospels: Aligning the Sinaiticus, Curetonianus, Peshî tâ & arklean Versions (4 vols.; New Testament Tools and Studies 21; Leiden, New York, Köln: E.J. Brill, 1996). Koster, M.D., The Peshitta of Exodus: The Development of Its Text in the Course of Fifteen Centuries (Assen and Amsterdam: Van Gorcum, 1977). Künzle, Beda O., Das altarmenische Evangelium (2 vols.; Bern, New York: Peter Lang, 1984). Legg, S.C.E., Evangelium secundum Marcum (Oxford: Clarendon, 1935). ——, Evangelium secundum Matthaeum (Oxford: Clarendon, 1940). Leloir, Louis, Saint Éphrem, Commentaire de l’Évangile Concordant: Texte Syriaque (Manuscrit Chester Beatty 709) (Chester Beatty Monographs 8; Dublin: Hodges Figgis, 1963), Saint Éphrem, Commentaire de l’Évangile Concordant: Texte Syriaque (Manuscrit Chester Beatty 709) Folios Additionnels (Chester Beatty Monographs 8; Leuven: Peeters, 1990).

316

BIBLIOGRAPHY

——, Saint Éphrem, Commentaire de l’Évangile Concordant: Version Arménienne (CSCO 137; Louvain: L. Durbecq, 1953). Lyon, Jeffrey Paul, Syriac Gospel Translations: A Comparison of the Language and Translation Method Used in the Old Syriac, the Diatessaron, and the Peshitto (Leuven: Peeters, 1994). Lyovin, Anatole V., An Introduction to the Languages of the World (New York and Oxford: Oxford University Press, 1997). Martin, Alain G., ‘Le palimpseste syriaque du Sinaï et le codex de Bèze’, in D.C. Parker and C.-B. Amphoux (eds), Codex Bezae: Studies from the Lunel Colloquium, June 1994. ——, ‘La traduction de Êýñéïò en Syriaque’, Filología Neotestamentaria 12 (1999), 25–54. McConaughy, Daniel L., ‘A Recently Discovered Folio of the Old Syriac (Syc) Text of Luke 16,13–17,1’, Biblica 68 (1987) 85–88. McHardy, W.D., ‘Disputed Readings in the Syriac Sinaitic Palimpsest’, JTS 45 (1944) 170–74. Merk, A., Novum Testamentum: Graece et Latine (8th edn; Rome: Pontifical Biblical Institute, 1957). Merx, Adalbert, Die vier kanonischen Evangelien nach ihrem ältesten bekannten Texte: Uebersetzung der syrischen im Sinaikloster gefundenen Palimpsesthandschrift (Berlin: Georg Reimer, 1897), Das Evangelium Matthaeus nach der syrischen im Sinaikloster gefundenen Palimpsesthandschrift (Berlin: Georg Reimer, 1902), Die Evangelien des Markus und Lukas nach der syrischen im Sinaikloster gefundenen Palimpsesthandschrift (Berlin:

BIBLIOGRAPHY

317

Georg Reimer, 1905), Das Evangelium des Johannes nach der syrischen im Sinaikloster gefundenen Palimpsesthandschrift (Berlin: Georg Reimer, 1911). Messina, Giuseppe, Diatessaron Persiano (Roma: Pontifical Biblical Institute, 1951). Metzger, Bruce M., ‘On the Citation of Variant Readings of Matt I 16’, JBL 77 (1958) 361–63. ——, The Early Versions of the New Testament: Their Origin, Transmission, and Limitations (Oxford: Clarendon, 1977). Mink, Gerd, ‘Die koptischen Versionen des Neuen Testaments: die sprachlichen Probleme bei ihrer Bewertung für die griechische Textgeschichte’ in K. Aland (ed.), Die alten Übersetzungen des Neuen Testaments, die Kirchenväterzitate und Lektionare, 160– 299. Müller-Kessler Christa and Michael Sokoloff, A Corpus of Christian Palestinian Aramaic, Volume IIA: The Christian Palestinian Aramaic New Testament Version from the Early Period: Gospels (Groningen: Styx Publications, 1998). Nöldeke, Th., Compendious Syriac Grammar (ET James A. Crichton; London: Williams & Norgate, 1904). Parker, D.C., Codex Bezae: An Early Christian Manuscript and Its Text (Cambridge: Cambridge University Press, 1992). Parker, D.C. and C.-B. Amphoux (eds), Codex Bezae: Studies from the Lunel Colloquium, June 1994 (Leiden, New York, Köln: E.J. Brill, 1996). Payne Smith, R., Thesaurus Syriacus (Oxford: Clarendon, 1879–1901).

318

BIBLIOGRAPHY

Pelser, H.S., ‘The Origin of the Ancient Syriac New Testament Texts—A History Study’, in I.H. Eybers et al. (eds), De Fructu Oris Sui: Essays in Honour of Adrianus van Selms (Pretoria Oriental Series 9; Leiden: E.J. Brill, 1971) 152–63. Petersen, William L., The Diatessaron and Ephrem Syrus as Sources of Romanos the Melodist (CSCO 475; Louvain: Peeters, 1985). ——, Tatian’s Diatessaron: Its Creation, Dissemination, Significance, and History in Scholarship (Supplements to Vigiliae Christianae 25; Leiden, New York, Köln: E.J. Brill, 1994). Pusey, P.E. and G.H. Gwilliam, Tetraeuangelium Sanctum (Oxford: Clarendon, 1901). Smith Lewis, Agnes, Light on the Four Gospels from the Sinai Palimpsest (London: Williams & Norgate, 1913). ——, Zu H.J. Vogels Schrift, Die altsyrischen Evangelien in ihrem Verhältnis zu Tatians Diatessaron (Leipzig: J.C. Hinrichs, 1913). von Soden, H.F., Die Schriften des Neuen Testaments (4 vols., Göttingen: Vandenhoeck & Ruprecht, 1911–13). Strothmann, Werner, Das Wolfenbütteler Tetraevangelium Syriacum: Lesarten und Lesungen (Göttinger Orientforschungen: 1. Reihe: Syriaca, Band 2; Wiesbaden: Harrassowitz, 1971). Swanson, Reuben, New Testament Greek Manuscripts: Variant Readings Arranged in Horizontal Lines against Codex Vaticanus: Matthew, Mark, Luke, John (4 vols.; Sheffield: Sheffield Academic Press, 1995). Taylor, Richard A., The Peshitta of Daniel (MPIL 7; Leiden, New York, Köln: E.J. Brill, 1994).

BIBLIOGRAPHY

319

Throckmorton, Burton H., Jr., ‘A Reply to Professor Metzger’, JBL 78 (1959) 162–63. Tischendorf, Constantinus, Novum Testamentum Graece: Editio Octava Critica Maior (Leipzig: Giesecke & Devrient, 1869–72). Vogels, H.J., Novum Testamentum Graece et Latine, Pars Prima: Evangelia et Actus Apostolorum (3rd edn; Freiburg im Breisgau: Herder, 1949). Vööbus, A., Studies in the History of the Gospel Text in Syriac (CSCO 128, Subsidia 3, Louvain: L. Durbecq, 1951). ——, Studies in the History of the Gospel Text in Syriac: New Contributions to the Sources Elucidating the History of the Traditions, II (CSCO 496, Subsidia 79, Louvain: Peeters, 1987). Waltke B.K. and M. O’Connor, An Introduction to Biblical Hebrew Syntax (Winona Lake, IN: Eisenbrauns, 1990). Walton, B., Biblia Sacra Polyglotta (London: Thomas Roycroft, 1657 [actually 1658]). Wichelhaus, Ioannes, De Novi Testamenti Versione Syriaca Antiqua quam Peschitho Vocant, Libri Quattuor (Halle: Orphanotropheum, 1850). Williams, P.J., “‘According to All’ in MT and the Peshitta”, Zeitschrift für Althebraistik 12 (1999) 107–109. ——, Studies in the Syntax of the Peshitta of 1 Kings (MPIL 12; Leiden, Boston, Köln: Brill, 2001). ——, ‘Some Problems in Determining the Vorlage of Early Syriac Versions of the NT’, New Testament Studies 47 (2001) 537–43. ——, ‘Bread and the Peshitta in Matthew 16:11–12 and 12:4’, Novum Testamentum 48 (2001) 331–33.

320

BIBLIOGRAPHY

Yoder, James D., Concordance to the Distinctive Greek Text of Codex Bezae (New Testament Tools and Studies 2; Leiden: E.J. Brill, 1961). Zohrabean, Y., AstowacašownL‘ Matean Hin ew Nor Ktakaranac‘ (Venice: St. Lazar, 1805). Zuurmond, Rochus, Novum Testamentum Aethiopice, The Synoptic Gospels: General Introduction, Edition of the Gospel of Mark (Äthiopistische Forschungen 27; Stuttgart: F. Steiner Verlag, 1989).

INDICES MODERN AUTHORS, EDITORS, AND EDITIONS OF THE BIBLE Greenberg, G., 205–206, 226,

Aland, B., 2, 219

227

Aland, K., 17, 21, 24, 219 Amphoux, C.-B., 80

Gregory, C.R., 13

Baarda, T., 273

Gressmann, H., 13

Baethgen, F., 16, 67, 68, 126,

Gwilliam, G.H., 10–11, 208, 221, 274

149, 159, 203, 255 Beck, E., 157

Harris, J.R., 14

Bensly, R.L., 14

Hill, J.H., 214

Black, M., 13

Hjelt, A., 13

Blass, F.W., 242

Horner, G., 220

Brock, S.P., 2, 17, 18, 39, 69,

Hort, F.J.A., 19

149, 203–204

Hoskier, H.C., 13 Institut für Neutestament-

Brockelmann, C., 13

liche Textforschung, 11,

Burkitt, F.C., 14, 16, 39, 46,

219

77, 89, 242–43, 244, 249,

International Greek New

255, 259, 278 Chase, F.H., 8, 140, 290

Testament Project, 6, 43,

Ciasca, P., 11, 214–15

185, 187, 207, 208, 273

Clemons, J.T., 16

Joosten, J., 133, 150

Diem, W., 70, 71

Juckel, A., 2, 10, 11

Falla, T.C., 17–18, 21, 24,

Kerschensteiner, J., 10, 233 Khan, G.A., 77

133, 149, 170, 210 Gelston, A., 204–205

Koster, M.D., 205 Künzle, B.O., 219

321

322

INDICES

Lane, D.J., 210

81, 84, 86–87, 89–90, 92–

Legg, S.C.E., 6, 28, 43, 55,

94, 96–102, 105, 113–18,

120, 190, 207

120–21, 123–24, 129–31,

Leloir, L., 215

139–42, 144–48, 151–59,

Lewis, A.S., 13, 14, 38, 103–

162, 164, 166–72, 175–76,

104, 179, 203

177–81, 183–84, 187–88,

London Polyglot, 286

190–91, 194–95, 198, 199–

Lyon, J.P., 13, 14, 18–19, 21,

200, 204, 207, 224–28,

24, 68–70, 73, 93, 94, 111,

230–34, 236, 238–47, 251–

149, 159, 255

52, 254–55, 258–59, 262,

Lyovin, A.V., 71

267–68, 274–76, 278–84,

Madrid Polyglot, 24

286–87

Martin, A.-G., 24, 80

Nestle, Eberhard, 19, 240

Merk, A., 21

Nestle-Aland, 6, 19–20

Merk 8th edn, 21

New English Bible, 62

Merx, A., 14–15, 19, 20, 46,

Nöldeke, Th., 157

103, 118, 168, 224, 273

O’Connor, M., 70

Messina, G., 215

Paris Polyglot, 286

Metzger, B.M., 16

Parker, D.C., 80

Mingana, A., 13

Payne Smith, R., 85

Mink, G., 17

Pelser, H.S., 13

Müller-Kessler, C., 220

Petersen, W.L., 11, 12, 213–

Nestle 1st edn, 19 Nestle 3rd edn, 19 Nestle 9th edn, 19 Nestle 17th edn, 19, 241–42 Nestle-Aland 25th edn, 19 Nestle-Aland 26th edn, 19, 231, 241 Nestle-Aland 27th edn, 6, 8,

16, 233 Pusey, P.E., 10–11, 208, 221, 274 von Soden, H.F., 6, 17, 21, 43, 184–85, 191, 207, 213– 15, 244, 273, 286 Sokoloff, M., 220 Strothmann, W., 10–11

16, 19–21, 24, 28, 32–36,

Swanson, R., 6, 207

44–45, 48–49, 51–52, 55–

Synopsis Quattuor Evan-

60, 62–63, 65–66, 77, 79–

geliorum, 21

323

INDICES

Taylor, R.A., 205–206, 210,

Vogels, H.J., 21, 38, 103–104,

237

179, 203

Tischendorf, C., 16, 19, 21,

Vööbus, A., 10, 11, 13, 134,

286

135, 208, 222–23, 233

Tischendorf 8th edn, 21

Waltke, B.K., 70

Torrey, C.C., 13

Weiss, B., 19

Tregelles, S.P., 16

Westcott, B.F., 19, 242

United Bible Societies, 6, 21

Weymouth, R.F., 19

United Bible Societies 4th

Wichelhaus, J., 68, 206

edn, 6, 21, 33, 35, 44–45,

Williams, P.J., 77, 126, 133,

48, 50, 79, 84, 90, 94, 96,

143, 149–50, 157, 159,

98–99, 114–15, 129, 142,

175–76, 204, 205, 246,

146, 147, 153–54, 156,

249

158, 167, 171, 175, 180,

Yoder, J.D., 172

183, 245, 283

Zahn, Th., 13

Vogels, 3rd edn, 21

Zohrab, Y., 219 Zuurmond, R., 220

INDEX OF BIBLICAL CITATIONS Matthew

2:4, 250

2:21, 38–39

2:6, 19

2:22, 38–39, 204,

1:14, 19

2:9, 19

1:16, 19, 39, 240–

2:11, 151, 212,

44, 289

233

1:18, 235

2:13, 204

1:19, 38, 41

2:14, 38

1:20, 40–41

2:15, 147–49,

1:22, 20, 148–49 1:24, 40–41

204, 236 2:16, 184, 186,

238–39 2:23, 134, 136 3:2, 155–56 3:3, 19, 66–67, 148–49 3:4, 278 3:6, 186–87 3:7, 16

1:25, 12, 19

276–77

2:1, 204, 236

2:17, 148–49

3:11, 207, 211,

2:2, 19

2:18, 204

3:13, 186

2:3, 236

2:19, 236

3:14, 186–87

217

324

INDICES

6:7, 250

8:21, 105, 114

6:8, 192

8:22, 36–37

3:16, 186

6:12, 60

8:23, 105, 265

3:17, 20, 195

6:14, 60

8:24, 27

4:1, 20

6:15, 62, 95

8:25, 105, 115

4:6, 74–75

6:17, 207, 213,

8:26, 23, 27, 151,

3:15, 258, 260, 275–76

4:9, 151

215, 216

223

4:11, 26, 27

6:18, 95

8:27, 223

4:13, 276–77

6:19, 209, 273

8:28, 27

4:14, 148–49

6:20, 209, 273

8:29, 35–36

4:16, 9

6:25, 51

8:34, 276–78

4:18, 59–60, 121

6:28, 180, 225

9:2, 60–61, 62,

4:20, 120–21

6:32, 129, 204

100, 121–22,

4:21, 27, 161

7:12, 156

199

4:22, 224, 263

7:16, 176

4:24, 234, 273

7:23, 250

4:25, 263

7:28, 159

9:9, 134, 162

5:1, 26, 105

8:1, 263

9:10, 105

5:2, 250

8:2, 134

9:11, 105, 171–72

5:3, 83

8:3, 33, 74

9:12, 35

5:8, 82

8:3–11:7, 23

9:18, 49, 151

5:11, 52–53, 129,

8:5, 19, 134

9:19, 105, 263

8:6, 283

9:21, 168–69

8:7, 34, 36–37,

9:22, 36–37, 96

224–25 5:13, 184, 186 5:23–24, 9

151

9:5, 60–61, 122, 199

9:27, 161–62, 263

5:26, 161

8:8, 168, 283

9:28, 28

5:28, 165

8:10, 263

9:32, 27, 134

5:30, 53

8:13, 100, 171,

9:36, 33

5:36, 223

283

9:37, 105, 171

5:39, 156

8:15, 150

10:2, 117, 244

5:45, 207, 213,

8:17, 148–49, 199

10:2–4, 158

8:18, 113

10:5, 117

215, 218 6:4, 95

8:19, 100, 263

325

INDICES

10:18, 191, 207,

12:38, 258

14:19, 105, 112,

12:39, 195, 260

113, 114, 177–

10:21, 96, 99

12:44, 171

78

10:22, 184, 186

12:49, 105

10:24, 100–101

13:10, 105, 113

10:27, 82

13:11, 260

14:22, 105, 114

10:28, 207, 211,

13:14, 240

14:24, 165, 167,

211, 215, 218

215, 216, 218

13:15, 82, 84, 162

14:21, 144, 225– 26

245

10:32, 94–95

13:23, 172

14:26, 105

10:37, 93, 96, 99

13:26, 171

14:27, 36

10:42, 168–69

13:28, 99

14:28, 258

11:1, 162

13:31, 252

14:31, 74

11:4, 223, 260

13:35, 44, 149

15:1, 225, 229–

11:5, 158

13:36, 35–36, 105

11:18, 194

13:37, 260

15:2, 79, 177

11:19, 194

13:51, 105, 113

15:3, 27, 177, 260

11:21, 280

13:53, 159

15:4, 92–93, 98,

11:25, 89

13:54, 27

11:28, 194

13:55, 38–39

15:5, 93

11:29, 82

14:2, 122

15:6, 177

12:1, 79, 105

14:5, 38

15:7, 240

12:4, 175

14:7, 64–65

15:8, 82

12:9, 162

14:9, 65–66

15:11, 84

12:10, 74, 137

14:11, 189

15:12, 105, 114

12:11, 138

14:12, 38, 84,

15:13, 260–61

12:15, 162

114, 151

30, 232

252

15:17, 84

12:17, 148–49

14:13, 36

15:18, 84, 225

12:19, 207, 212

14:14, 35–36

15:19, 157

12:22, 134, 190–

14:15, 105, 114,

15:20, 74

91

165, 167

12:25, 207

14:16, 192

12:27, 51

14:18, 27, 245

12:31, 63, 177

15:21, 276 15:22, 134, 136, 276–78 15:23, 105

326 15:24, 260–61 15:26, 258 15:28, 171, 194, 258 15:30, 33, 272

INDICES

17:17, 207, 212– 13

19:15, 75, 161–62 19:16, 139

17:19, 32, 105

19:18–19, 157–58

17:20, 33, 35

19:19, 93

17:21, 207, 217,

19:20, 164

218, 221–22

19:21, 264

15:32, 105

17:25, 38–39, 41

19:23, 106

15:33, 105, 114

17:26, 44

19:25, 106, 111–

15:36, 105, 115

18:1, 105, 114

15:38, 144, 225–

18:2, 139–40

19:27, 260

18:7, 280

20:12, 128–29,

15:31, 158, 272

26 15:39, 276, 278 16:2, 258, 260

18:8, 53–54, 76, 272

12

214–15, 245 20:13, 262 20:17, 116, 278–

16:5, 105, 115

18:9, 53–54

16:7, 172, 196

18:13, 159

79

16:11–12, 175

18:15, 168

20:21, 84

16:13, 105, 276

18:17, 207, 213

20:23, 65, 84

16:14, 105, 113

18:22, 31

20:34, 31

16:17, 258, 260

18:24, 189, 191

21:1, 106, 171

16:20, 105, 115

18:25, 100, 101

21:2, 54, 56, 134

16:21, 105

18:26, 66

21:3, 192

16:23, 27

18:28, 64, 65–66

21:4, 148–49

16:24, 105

18:29, 66

21:5, 147

17:2, 33

18:30, 64

21:6, 106, 112

17:3, 179

18:33, 156

21:7, 27

17:4, 258

18:34, 64

21:9, 263–65

17:6, 105, 212

19:1, 276

21:11, 239

17:9, 250

19:4, 260

21:12, 27, 207,

17:10, 105, 115

19:7, 64

17:11, 258

19:10, 106, 115

21:13, 209

17:12, 165

19:12, 97

21:20, 106

17:13, 105

19:13, 75, 106,

21:21, 260

17:14, 134

113, 190, 191

213, 215, 218

21:23, 27

327

INDICES

21:24, 156, 260

24:43, 209

26:47, 116

21:27, 32, 34, 258

24:45, 97

26:50, 32, 74, 77–

21:28, 100, 139

25:9, 258

21:29, 258

25:11, 24

26:51, 74, 117

21:33, 139

25:12, 260–61

26:55, 209

21:35, 234

25:26, 260

26:56, 106, 115

21:38, 99

25:31, 120, 128

26:57, 32, 55–56

21:39, 55

25:33, 82

26:64, 32

22:10, 158

25:37, 24, 258

26:65, 169, 192

22:13, 74, 211

25:41, 82, 156

27:2, 55–56

22:21, 171

25:44, 24

27:3, 27

22:29, 260

26:1, 31, 106

27:9, 149

22:35, 229

26:3, 235, 244

27:12, 184–85

22:37, 33

26:7, 27, 134–35

27:16, 134, 171

22:41, 32

26:8, 106, 115

27:18, 38

23:1, 106

26:9, 52

27:22–23, 58

23:3, 226–27

26:13, 120

27:23, 171

23:26, 181–82

26:14, 116

27:24, 74

23:28, 207, 213

26:17, 106, 114

27:26, 65–66

23:35, 211

26:18, 100

27:28, 227

23:37, 194–95

26:19, 106, 112,

27:38, 82, 83,

78

113

171, 209

24:3, 28, 106, 113

26:20, 116

27:40, 152–53

24:4, 260

26:23, 74, 258

27:44, 209

24:6, 129

26:24, 195

27:46, 72

24:8, 204

26:25, 33

27:50, 72, 83

24:10, 207, 213

26:26, 52, 106

27:57, 134

24:24, 184

26:28, 52

27:58, 66, 171

24:30, 194

26:33, 260

27:60, 134, 181

24:32, 82, 165

26:35, 106

27:64, 106, 115

24:33, 204

26:36, 106, 115

28:7, 106

24:34, 204

26:40, 106, 115

28:8, 106

24:37, 156

26:45, 106, 115

28:11, 251

24:1, 106

328 28:13, 106

INDICES

3:1, 27, 74, 134– 35

6:5, 74–76, 193 6:7, 115, 119

Mark

3:3, 74

6:13, 55

1:1, 142

3:5, 79

6:14, 236

1:2, 142

3:7, 107

6:19, 171

1:4, 4, 142

3:7–8, 268–72

6:27, 41

1:5, 186–88

3:9, 107

6:29, 107

1:9, 186–87

3:16–19, 158

6:29–32, 198

1:15, 120

3:31, 179, 228

6:31, 223

1:16, 44, 48, 59–

3:33, 99

6:35, 107, 165,

60, 121

3:35, 99

167

1:18, 120–21

4:2, 279–80

6:37, 258

1:19, 121

4:4, 160

6:38, 151

1:20, 235

4:10, 107, 119

6:41, 107, 115

1:24, 194

4:11, 27

6:44, 144

1:26, 72

4:33, 27

6:45, 107, 200

1:27, 250

4:34, 107

6:56, 27

1:31, 79

4:37, 165–67

7:1, 141, 229

1:34, 273

4:38, 27

7:2, 74, 107, 141

1:37, 131–32, 201

5:2, 134

7:3, 74, 177, 207,

1:38, 197

5:4, 186

1:39, 131

5:9, 182

7:5, 74, 177, 229

1:40, 134, 207,

5:13, 143

7:6, 27, 82, 240,

212, 218

5:14, 251

213

258

1:45, 27

5:17, 276

7:8, 177, 223–24

2:5, 60–61, 100,

5:23, 75, 79–80,

7:9, 177

122, 199

152

7:10, 93

5:24, 26, 27

7:11, 93

199

5:25, 139

7:13, 177

2:15, 106

5:31, 107

7:14, 28

2:16, 107, 228

5:34, 96

7:17, 28, 107

2:23, 27, 107

5:40, 97

7:19, 81–82, 154–

2:25, 28

6:1, 107, 263

2:9, 60–61, 122,

55

329

INDICES

7:24, 27, 134, 162, 276

9:25, 208, 212, 218

10:38, 186 10:39, 186

7:25, 137

9:28, 27, 107

10:46, 27, 107

7:26, 278

9:29, 208, 211,

10:51, 258

7:28, 258

213, 215, 217,

10:52, 36, 265–66

7:30, 96

219–20, 221–

11:1, 28, 107

7:31, 276

22

11:2, 54

7:32, 74, 134

9:31, 107

11:3, 192

8:1, 107, 115

9:33, 181

11:4, 163

8:4, 107, 257

9:35, 27, 115

11:7, 27, 199

8:6, 107

9:36, 135

11:9, 266

8:9, 143–44

9:38, 263

11:11, 116, 165,

8:10, 107, 180,

9:39, 36

167 11:14, 107, 258,

278

9:41, 86, 169

8:17, 164

9:43, 76, 272

8:23, 75

10:1, 276

8:25, 75

10:2, 101

274–75

8:27, 107

10:3, 258

11:17, 209

8:29, 27, 258, 260

10:5, 260

11:18, 232–33

8:33, 107

10:10, 107, 113

11:20, 83

8:34, 27, 107,

10:11–12, 19

11:25, 60

10:13, 76, 107,

11:29, 257

266–67 8:35, 120 8:38, 120, 207, 213, 218

113

274–75 11:15, 208, 232,

11:29–30, 256

10:16, 74–76

11:30, 258

10:17, 252

11:33, 32, 258, 262

9:2, 168

10:19, 98, 235

9:4, 31, 207, 212,

10:20, 260

12:1, 139, 140

213, 218

10:21, 264

12:2, 99

9:5, 212, 258

10:23, 107

12:3, 55–56

9:8, 107, 113

10:24, 100, 107

12:6, 164

9:9, 250

10:29, 120

12:19, 101

9:14, 107

10:30, 169–70

12:33, 82, 83

9:22, 182–83

10:32, 115, 266

12:34, 260

330

INDICES

12:35, 258

14:41, 208, 217

16:8, 208

12:38, 279–80

14:43, 116

16:9–20, 8

12:41, 33–34

14:44, 55–56

16:13, 251

12:43, 107

14:46, 77

16:15, 120

13:1, 28, 107

14:48, 209, 260

16:16, 186

13:3, 27

14:50, 108

16:17–20, 8

13:8, 153

14:57, 251–52

16:18, 74, 76

13:9, 208, 213,

14:63, 192, 196

16:19, 33–34,

218

14:64, 87, 196–97

108, 113

13:10, 120

14:65, 170

13:12, 96, 99, 100

14:67, 236

Luke

13:27, 120

15:1, 56

1:5, 283

13:28, 165

15:4, 253

1:11, 38, 39, 42–

13:30, 204

15:10, 38

14:1, 54

15:12, 258

1:12, 42–43

14:3, 27, 135

15:13, 59

1:22, 38

14:4, 108

15:14, 59

1:23, 159

14:8, 150

15:15, 66

1:26, 235

14:9, 120

15:17, 227

1:35, 260

14:10, 28, 116

15:24, 178

1:42, 38

14:12, 38, 43,

15:27, 83, 209

1:54, 283

108, 191

15:32, 169

1:56, 38, 143

14:13, 108, 135

15:34, 72

1:59, 238

14:15, 135

15:37, 72

1:60, 258

14:16, 108, 115

15:41, 156–57,

1:61, 38

14:17, 115

263

43

1:62, 250

14:20, 116

15:42, 165, 167

1:65, 130–31

14:22, 35

15:44, 165, 167

1:66, 131

14:23, 129

15:45, 84

1:74, 96–97

14:32, 108

15:46, 44–45, 48,

2:1, 235

14:37, 151 14:40, 151, 256– 57, 258

56, 57

2:2, 235

16:6, 191

2:4, 235

16:7, 108

2:11, 239

331

INDICES

2:15, 159, 197–98

4:11, 74–75

6:8, 74

2:16, 208, 212,

4:12, 261

6:13, 108

4:16, 204

6:14–16, 158

2:19, 130

4:17, 27, 239–40

6:17, 27, 108, 273

2:20, 208, 215

4:18, 198

6:20, 108

2:22, 171–72

4:23, 27, 197

6:24, 273

2:25, 135, 282–83

4:25, 204, 240

6:25, 208, 213,

2:26, 250

4:27, 42–43

2:27, 99, 236

4:28, 49–51, 129

6:29, 82

2:28, 38, 84

4:29–30, 273

6:36, 156

2:33, 45, 98

4:33, 72

6:38, 63, 273

2:37, 208, 211,

215

214, 215, 218

4:40, 75, 273

6:41, 192

215, 218

5:1, 159

6:43, 176

2:39, 38, 236

5:2, 121

6:44, 176

2:43, 239

5:3, 252

6:45, 84

2:44, 27

5:4, 182

7:1, 193

2:48, 100

5:5, 121, 260

7:6, 97, 165

3:4, 66–67

5:8, 244

7:7, 283

3:5, 129

5:12, 27, 82, 135

7:9, 50

3:7, 186–88

5:13, 27, 74

7:11, 108

3:9, 176

5:14, 27

7:20, 27

3:11, 258, 260

5:18, 135

7:22, 251

3:12, 186

5:20, 60–61, 122,

7:29, 186

3:14, 156, 223 3:16, 208, 211, 215, 217, 258, 261

199 5:23, 60–61, 62, 121–22, 199 5:30, 108, 229

7:30, 184, 186– 88, 208, 229 7:40, 258, 260 7:41, 138, 208, 216, 217

3:19, 38–40, 41

5:31, 192

3:20, 40

6:1, 27, 79, 108

7:43, 257

3:21, 159, 186

6:3, 262

7:45, 196

3:23, 124, 145

6:4, 156

7:47, 60, 122, 199

4:4, 258, 261–62

6:5, 156

7:48, 60, 122, 200

4:8, 260

6:6, 28

7:50, 27

332

INDICES

8:1, 28, 115, 122–

9:14, 27, 108,

10:18, 29

23, 159, 223

144–45

10:21, 35

8:2, 208, 213, 215, 216 8:8, 250

9:15, 108, 113

10:22, 108, 115

9:16, 27, 108, 113

10:23, 108, 113

9:18, 108, 113,

10:25, 229

8:9, 108

159

10:27, 258, 261

8:19, 27

9:19, 258

10:28, 257, 258

8:20, 251

9:21, 55

10:29, 32

8:21, 260

9:23, 129–30

10:30, 31, 209

8:22, 108

9:25, 208

10:31, 53

8:23, 28–29

9:26, 93, 120

10:32, 53

8:27, 178

9:28, 145, 153,

10:33, 53

8:28, 72

160, 228

10:34, 223

8:29, 29

9:36, 251

10:36, 209

8:32, 135–36

9:37, 159

10:38, 138, 160

8:37, 29

9:38, 135

10:41, 258, 261

8:40, 160

9:41, 208, 212–13

10:42, 192

8:41, 137

9:42, 164

11:1, 108, 138,

8:42, 27, 143

9:43, 36–37, 108

8:43, 128

9:47, 135–36

11:2, 29

8:47, 251

9:49, 139, 265

11:4, 60, 124

8:48, 96

9:51, 122, 159

11:5, 27, 97

8:49, 164

9:52, 135–36

11:6, 97

8:50, 259

9:54, 108, 115,

11:7, 97, 165

8:51, 29, 97, 228

236

159

11:10, 126–27

9:1, 115, 116

9:57, 139, 160

11:14, 124–25

9:6, 116, 208,

9:58, 82

11:15, 38

217, 218

9:62, 79

11:17, 27, 127

9:7, 185

10:1, 108, 113

11:24, 171

9:8, 141

10:8, 126–27

11:26, 171

9:12, 113, 115,

10:11, 84

11:27, 72–73, 159

10:13, 280

11:28, 27, 55

10:17, 119

11:33, 208

117

333

INDICES

11:37, 139–40

12:58, 65

15:3, 27

11:38, 186

13:1, 251

15:6, 98

11:39, 169

13:3, 129

15:7, 192

11:42, 280

13:5, 129

15:9, 97, 98

11:43, 280

13:6, 139

15:10, 128

11:44, 145–46,

13:8, 258

15:11, 27

13:11, 135

15:12, 93, 99, 128

11:45, 229

13:12, 55

15:13, 99

11:46, 229, 280

13:13, 74–76

15:22, 93

11:49, 208

13:15, 261

15:24, 153

11:51, 208, 211,

13:18, 27

15:25, 208, 215,

273–74

215

13:20, 27

218, 221

11:52, 229, 280

13:23, 139, 254

15:29, 93–94, 258

11:53–54, 225,

13:23–24, 181

15:31, 100

13:24, 27

16:1, 27, 108, 115

12:1, 27, 108, 247

13:26, 171

16:2, 64

12:13, 139

13:27, 171

16:13, 208, 215

12:14, 29

13:29, 234

16:19, 208, 213,

12:15, 108, 113

13:32, 29

12:22, 84, 108,

13:34, 83, 194

228–30

218, 219, 221 16:22, 160, 185,

14:1, 159

190

12:25, 140–41

14:3, 229

16:24, 122

12:28, 278

14:4, 56

16:25, 100

12:33, 209

14:5, 230–31, 258

16:31, 41

12:37, 97

14:9, 171

17:1, 27, 108

12:39, 209

14:10, 97, 171

17:5, 117

12:42, 64

14:12, 99

17:8, 66

12:47, 66

14:15, 50

17:11, 27, 159

12:49, 165

14:21, 251

17:13, 72, 83

12:50, 186

14:23, 99

17:14, 159

12:53, 93, 98, 99,

14:26, 95–96, 234

17:15, 72

14:35, 191

17:16, 82

15:2, 229

17:17, 258

115

101–102 12:56, 230

334

INDICES

17:20, 29, 185,

19:37, 72, 109,

22:5, 65

258, 261

130, 165

22:7, 191

17:22, 108, 113

19:40, 258

22:12, 66, 135

17:23, 208–209

19:42, 169

22:14, 29, 116–19

17:37, 258

19:43, 281–82

22:23, 122

18:4, 178

19:44, 146–47

22:33, 38

18:11, 48–50,

19:45, 209, 217

22:34, 29

19:46, 209

22:37, 164

20:3, 258, 260

22:38, 27

20:7, 257

22:39, 109, 113,

153–54 18:15, 76, 108, 113

263, 264

18:20, 92, 98

20:9, 139–40, 178

18:22, 48, 50, 264

20:10, 99, 246

22:41, 82

18:29, 29

20:11, 99

22:45, 109, 113,

18:31, 115

20:15, 53

153

18:32, 234

20:16, 50

22:47, 116

18:33, 234

20:19, 77, 231

22:49, 109, 117

18:35, 139

20:28, 101

22:52, 209, 234

18:37, 251

20:39, 258

22:53, 77–78

18:39, 27

20:42, 124

22:58, 45

18:40, 190, 280–

20:45, 109, 115

22:59, 145

21:2, 138

22:62, 163

19:2, 135, 209,

21:4, 129

22:64, 255

213, 215

21:5, 27

22:70, 27

19:3, 41

21:11, 231, 246

22:71, 124, 164,

19:4, 27

21:12, 77, 211

19:15, 159

21:16, 96, 97, 99,

81

19:28, 27

185

192 23:2, 238 23:3, 253, 258

21:20, 171

23:5, 155

244

21:21, 171

23:7, 43

19:30, 56

21:24, 127

23:8, 178

19:31, 192

21:30, 165

23:9, 27

19:34, 192

21:36, 130

23:16, 55

22:3, 116

23:21, 59

19:29, 108, 113,

335

INDICES

23:22, 55

24:49, 236

3:3, 258, 260

23:23, 72

24:50, 162–63

3:5, 209, 210,

23:25, 223

24:51, 159

23:32, 190

211, 215, 217, 226, 257

John

3:7, 182

1:6, 282–83

3:8, 210, 211, 226

23:34, 89, 178

1:14, 84

3:9, 258, 260

23:44, 145, 166

1:17, 23, 209,

3:10, 258, 260

23:33, 82, 83, 135–36

23:46, 51, 72

213, 215, 216

3:15–16, 9

23:49, 27

1:18, 93

3:16, 99

23:50, 135, 209,

1:21, 252, 257

3:17, 99

1:25, 255

3:18, 165

23:53, 57

1:26, 258, 261

3:22, 109

23:56, 221

1:29–6:5, 23

3:23, 186

24:4, 159

1:37, 32, 266

3:25, 135–36

24:5, 82

1:38, 32, 266

3:26, 132

24:6, 146

1:39, 143, 266

3:27, 258, 261

24:9, 109, 111,

1:45, 23, 38, 40

3:32, 209, 215

1:48, 258

4:1, 113, 114

1:49, 259, 260

4:2, 109, 168

1:50, 258

4:5, 135

1:51, 28, 222

4:6, 143

260

2:2, 109

4:7, 140

24:19, 231

2:11, 109

4:8, 109

24:25, 27, 171

2:12, 96, 109, 115

4:10, 31, 258, 260

24:29, 165, 166–

2:17, 109

4:12, 236–37

67, 200–201

2:18, 258

4:13, 258, 260

24:30, 159

2:19, 260

4:16, 34

24:32, 159

2:20, 42

4:17, 258

24:36, 50–51

2:22, 109

4:27, 109

24:38, 27, 171

2:23, 27

4:31, 109, 113

24:39, 158

2:25, 192

4:33, 109

24:40, 74, 82

3:1, 282–83

4:34, 32

215, 218

131, 204, 251 24:15, 123, 157, 159 24:18, 168, 258,

336

INDICES

4:35, 164, 165

6:16, 109

7:20, 258, 259

4:36, 158–59

6:17, 165, 167

7:21, 258, 260

4:42, 124, 238

6:19, 144

7:22, 93

4:43, 27

6:22, 109, 113,

7:28, 158

4:46, 35

180

7:30, 77 7:32, 45, 48, 50–

4:48, 215

6:24, 32, 109

4:51, 99, 165, 283

6:26, 258, 261

4:52, 19

6:27, 236

7:34, 57–58

4:53, 94

6:29, 258, 260

7:36, 58

5:2, 135

6:32, 94–95

7:42, 246

5:3, 234

6:37, 162–63

7:44, 77

5:6, 166

6:42, 170

7:45, 45

5:7, 258

6:43, 31, 258, 260

7:46, 146, 257

5:11, 257

6:46, 236

7:47, 257

5:12, 253

6:47, 127, 142

7:50, 29

5:15, 23, 251

6:52, 84

7:52, 258

5:17, 257

6:54, 195

8:11, 169

5:19, 93, 258, 260

6:57, 90

8:14, 258, 260

5:20, 99

6:58, 94

8:18, 90

5:22, 99

6:60, 109

8:19, 257, 261

5:46, 38

6:61, 109

8:21, 33, 177

6:3, 109

6:65, 90

8:25, 38, 43

6:6, 254

6:66, 109

8:26, 50

6:7, 257

6:67, 115

8:29, 168

6:8, 109

6:68, 257

8:31, 171

6:9, 127, 140

6:70, 116, 257

8:33, 257

6:10, 23, 31, 37,

6:71, 116

8:34, 257

7:3, 27

8:35, 99–100

7:5, 27

8:38, 52, 90, 94,

143–44 6:11, 109, 155,

51

147

171

7:12, 43

6:12, 109

7:13, 43

8:39, 31, 258

6:13, 144

7:14, 165

8:48, 258

6:14, 35

7:16, 258

8:49, 257

337

INDICES

10:31, 43

11:52, 193

257

10:32, 257

11:54, 109, 115,

8:58, 31

10:33, 257

162

8:59, 172

10:34, 257

11:55, 42

9:1, 98, 135

10:38, 224, 283–

12:4, 109, 111,

8:54, 94–95, 142,

9:2, 109, 253

84

240

9:3, 32, 257

11–12, 42

12:6, 51, 209

9:4, 183

11:6, 38

12:9, 27, 168–69

9:8, 98–99

11:7, 109, 115

12:11, 39

9:11, 192, 261

11:8, 109, 113,

12:14, 240 12:16, 109, 110,

9:15, 192

115

9:19, 253

11:9, 257

171

9:22, 43, 165, 230

11:11, 29

12:21, 253

9:25, 257, 261

11:12, 109

12:22, 179

9:27, 113, 165,

11:13, 38

12:23, 258

11:16, 113

12:24, 122–23

9:30, 258, 260

11:17, 38, 166

12:30, 258, 261

9:32, 98

11:18, 144

12:31, 163

9:34, 258

11:19, 96

12:34, 257

9:35, 27, 163

11:29, 38–39

12:37, 127

9:36, 258, 259

11:30, 164

12:40, 82

9:37, 158

11:33, 83

12:42, 27

9:40, 49–51

11:37, 98, 156

12:47, 58

9:41, 31, 169

11:39, 165

12:49, 209, 216

10:1, 209, 213,

11:40, 31

13:1, 93

11:41, 31, 86, 89

13:2, 165, 284

10:3, 83

11:43, 72, 264

13:3, 33–34, 74

10:8, 209

11:44, 74, 82,

13:4, 83

257

214

10:10, 209

171–72, 211

13:5, 83, 110, 113, 115

10:14, 188–89

11:45, 27

10:16, 127

11:47, 171

13:6, 45–46

10:22, 160

11:49, 39

13:7, 258, 260

10:25, 257, 262

11:50, 127

13:8, 257

338

INDICES

13:9, 82

14:16, 91

18:8, 257

13:10, 35, 82

14:21, 94, 186

18:11, 90–91

13:12, 155

14:22, 19, 97

18:12, 171

13:13, 211

14:23, 258, 260

18:12–13, 58

13:14, 97, 100,

14:24, 91

18:15, 110, 141

14:26, 91–92,

18:19, 110

209, 211, 218 13:21, 83

237–38, 284

18:20, 257

13:22, 110, 115

14:28, 91–92

18:22, 82

13:23, 31, 32, 110

14:31, 91

18:24, 27

13:26, 27, 32,

15:3, 165

18:25, 110

15:8, 94

18:26, 39

13:27, 171

15:15, 97

18:29, 162–63

13:28, 110, 113,

15:20, 227

18:34, 257

117

15:22, 177

18:35, 257

13:29, 192

15:24, 177

18:36, 257

13:30, 171

16:8, 121, 177

18:37, 257

13:33, 100

16:9, 177

18:40, 209

13:36, 31, 97,

16:10, 90

19:3, 82

16:12, 164

19:5, 41–42, 46

13:38, 257

16:16, 90–91

19:6, 48, 58–59

14, 91

16:17, 90, 110

19:7, 257

14:1, 171, 172–73

16:23, 127

19:11, 257

14:5, 97

16:29, 110

19:14, 144

14:6, 91

16:31, 257

19:15, 55–56, 257

14:8, 97

16:32, 170

19:21, 43

14:9, 90–92, 127,

17:1, 99

19:22, 257

17:6, 245

19:26, 97–98

14:9–10, 90

17:8, 123

19:28, 165, 167

14:10, 91, 182

17:11, 150

19:30, 35, 82, 83

14:11, 91

18:1, 110

19:33, 165

14:12, 91, 156,

18:2, 110

19:38, 33

18:5, 35, 257

19:39, 33, 144

18:7, 253

19:40, 192

150, 257, 259

257, 262, 267

182

182 14:13, 91, 99

339

INDICES

20:2, 183

21:17, 27, 37, 39

Daniel 6:8, 205

20:3, 179

21:18, 83

Daniel 6:25, 205

20:6, 264

21:19, 39

Daniel 7:14, 205

20:10, 110, 111

21:20, 265

Daniel 8:26, 205

20:11, 163

21:23, 110, 111

Daniel 9:7, 205

20:12, 82, 85–86

Daniel 9:16, 205

20:16, 171

Other Biblical Texts

20:17, 31, 89, 90–

Genesis 44:32, 89

Acts 3:22, 126

Deuteronomy

Acts 8:18, 75

20:15, 32

91 20:18, 110, 171, 251 20:19, 110 20:20, 80–81, 110 20:21, 35–36 20:23, 61–62 20:25, 110 20:26, 110, 113 20:28, 258 20:30, 110, 115 21:1, 110, 115 21:2, 110, 113, 127 21:4, 110, 165

32:14, 176 2 Samuel, 237 1 Kings 8:29, 205, 210 1 Kings 8:59, 205, 210 1 Kings 22:32, 157 Isaiah 40:3, 66– 67 Jeremiah 1:10, 205 Jeremiah 11:7–8, 226

21:5, 37, 257

Daniel 2:31, 205

21:6, 121

Daniel 2:35, 205

21:7, 83, 171

Daniel 2:38, 205

21:8, 110, 144

Daniel 2:40, 205

21:11, 121, 127

Daniel 3:27, 205

21:12, 110

Daniel 4:19, 205

21:14, 110, 115,

Daniel 4:33, 205

165

Daniel 5:23, 205

21:16, 27

Daniel 6:5, 205

Daniel 10:9, 205 Acts 1:13, 158

Romans 1:7, 233 Philippians 1:11, 176 1 Timothy 4:14, 75 2 Timothy 1:6, 75 Hebrews 6:2, 75