[ CORNELL

UNIVERSITY LIBRARY

BARNES BIBLICAL LIBRARY

THE GIFT OF

ALFRED C. BARNES

1889

10100

ܚܝ

DATE DUE

ara? VWs 3 O

GAYLORD

Cornell University

The original of this book is in the Cornell University Library.

There are no known copyright restrictions in the United States on the use of the text.

http://www.archive.org/details/cu31924092359698

EVANGELION DA-MEPHARRESHE

INTRODUCTION AND NOTES

London: C. J. CLAY anp SONS, CAMBRIDGE UNIVERSITY PRESS WAREHOUSE, AVE MARIA LANE.

Glasgow: 50, WELLINGTON STREET.

Deipsig: F. A. BROCKHAUS. Pew Work: THE MACMILLAN COMPANY. Bombay and Calcutta: MACMILLAN AND CO., Lrp.

[All Rights reserved.]

ܡܝ

EVANGELION DA-MEPHARRESHE

The Curetonian Version of the Four Gospels, with the readings of the Sinai Palimpsest and the early Syriac Patristic evidence

edited, collected and arranged by

F. CRAWFORD BURKITT, M.A.

University Lecturer in Palaeography.

VOLUME II INTRODUCTION AND NOTES

CAMBRIDGE, at the University Press, rgo4.

ܐ Ke‏

Cambridge : PRINTED BY uv. AND 0. F. CLAY,

AT THE UNIVERSITY PRESS.

CONTENTS OF VOLUME II.

\ PAGES IntRoDUcTION . ܢ‎ : : : : : : 0 ‘ . : . 1-6 Cuaprer J. DescrIPTION OF THE MSS... : : 8 : . : . 7-33 Copex C: Later History . ܪ‎ : 8 0 . 2 8 ‫: = 7 a Composition of Quires, etc. . 3 2 ܘ 7 . ܕ‎ 9 0 Date. é : : ܘ 0 . : . ܗ‎ 8 : 13 0 Colometry . 3 . . ‘ . . 0 2 ‘ : 14 8 Cureton’s Edition : ‫. ‘ : ‘ 3 . . 8 16 CopEx S: Recent History . : ‘ | . j ‘ : : : 17 ܪ‎ When and where the upper writing was transcribed . ‘ 18 7 The original ss. ܪ : : : . : : ܪ‎ A 21 :ܪ‎ Composition of Quires, etc. ‘ i . 4 . ܘ‎ a 23 Titles, Subscriptions and Colophons in C and 8. ܘ ܘ‎ : . ‘ 30 Nore on the line and paragraph divisions‘in C and § . ܘ ܕ‎ : 34—38 CHaprer 11. GRAMMAR AND Synvrax. i : _ ܕ‎ 0 : : 39—84 General remarks on the style of Cand S . ܪ‎ 1 . : : ‘ 39 Spelling . : 0 : : ‘ A : 8 A : : : 40 Pronouns 0 : : ܕ‎ : : i : : : : . 5 41 Nouns. . : : : : . . 0 : . 1 : : 43 Numbers and Particles : ‘ : : : . . : : : 49 Verbs. : 0 ‘ : : 3 : ܘ‎ . : : 2 ; 51 Syntax . >+ ©. Ww, Sl RS ܨ ܨܡ 3 ܐܦ‎ © $: @€ 7 . Vocabulary. : : : : $< ® ܨ‎ +e ݁ܕ‎ a ia . 78 Appenpix: (A) Syriac renderings of etOvs, etdéas . : j 5 ‘ . 85 ‘5 (B) Syriac renderings of ܐܐܘ‎ 2220 20020 sD 7 (C) Syriac equivalents for ‘answered and said’ . : : 5 90 7 (D) The names of 8. Peter 02 3 ® :8& .ܘ ܘܐܗܗ )ܐ‎ 989

7 (E) Renderings of “Ijcods and of xvpuos ; : . . : 97

Contents of Volume LI.

PAGES

Cuaprer 111. Tsar Pesuirtra New TESTAMENT AND ITs Rivas. . 100—165

The Acts of Judas Thomas . : : : ] ‘ ‘ £$ ® 101

The Doctrina Apostolorum ܘ : : : ܒ‎ : ‫. : ‘ . 107

Aphraates > . : : : ; . : . . 9 : . . 109

©, Ephraim . i : i : ‘ ‘ ‘ . . : . 112

The Commentary of Abba . : 1 a ‫. , : F : . 149

Cyrillona . : ‘ : F : : . : : : . 150

The Doctrine of Addai . ; . . : : . : : : 189

The Martyrdom of Habbib . : ‘ : . . : . , 155

Marutha . : 0 . 1 . : : 3 ‘ ܘ‎ . : ` 155

The Armenian, Georgian and Ethiopic Versions . ; ; : ; . 160

The Life of Rabbula > . : : : : :ܕ‎ . 160

Rabbula’s Revision identified with the N.T. Peshitta : ‘ : 101

162 . ‘ ܘ ‘ : 0 ܘ : :ܕ : Objections considered.‏

The survival of Sand 6 .. ‫. : : : : ܘ ܘ‎ . 165

AppENDIX: The Quotations in the Theophania of Eusebius . . 166—172

CHarter IV. Tur DIATESSARON AND THE OLD Syriac . 0 1 . 173-212

Notices of the Diatessaron in Syriac Literature . : : : : 2 173

Addai. ‫: F F ; . : . : ; : . . 174

175 » . : ; . : ܕ : ܕ ܨ : ‘ . Eusebius‏

Theodoret : : : : é 1 ܹ : . 176

Notices of the Evangelion da- Ginna ܕ‎ : : : , 177

06 ܀ ܘ : ܵܕ . : . : ܘ : ܕ . Rabbula‏

Bar Bahlul, Barsalibi. ܇ 8 : : : ܕ 1 ܕ‎ 178

Ignatius Philoxenus 0 : ܘ‎ : : ‘ : : . . 179

On the Quotations of Aphraates . 2 2 : : . ` . 180

©. Ephraim’s use of the separate Gospels . 3 : : . . 186

References to the ‘Greek’ in S. Ephraim’s Commentary . 1 189

Internal evidences of the date of the Evangelion da-Mepharreshe . 191 Differences between the Hv. da-Mepharreshe and the Diatessaron

in the underlying Greek text . 8 : : :ܘ .‫ ܘ‎ TOD

7 in rendering Greek words . 8 : : : : : . 196

49 in the Nativity Story : ܕ ܕ‎ . 198

The Hv. da-Mepharreshe influenced by the Old Testament Peshitta . . 201

The Diatessaron influenced by the Old Testament Peshitta . : . 905

Serapion of Antioch and Palut of Edessa . ܪ‎ : ‘ ‘ i - 207

The Hv. da-Mepharreshe dated about 200 A.D. . : . : . 09

Dr Hjelt’s Theory . 210

Contents of Volume TT. PAGES CHapter ¥. THE TEXTS oF S AND oF C. : 4 0 0 : . 213—254 Later revisions of our mss. from the Greek 214 Absence of the ‘Greater Interpolations’ from 8. 216 Signs of patchwork in the text of C 217 S, C, and the Diatessaron : 0 220 The textual affinities of the Old Syriac version . 223 I. The later Antiochian text 224 II. The xB-text : : 996 The ‘Western Non-Interpolations’ 228 Later Elements in ܠܐ‎ and B 233 111. Western texts 234 S or C with Western texts against the Diatessaron 235 S or C with Western texts including the Diatessaron 238 Passages where the Diatessaron reading is not known . 242 IV. Secondary Greek mss. : : 946 ¥. Noteworthy singular readings of S and C 251 Notes on SELECT READINGS S. Matthew 257 S. Mark . 280 8. Luke . 286 8. John . 306 GENERAL INDEX 318 BIBLICAL REFERENCES . 321

ERRATUM. P. 8, 11. 9, 20 for Habibai read Habbib (see Wright’s Catalogue, p. xxxv)

vii

CopEx ܗܣ‎ Dr CurEtTon Coprx § CopEx ܣ‎

”

Coprx S

LIST OF ILLUSTRATIONS.

Mk xvi 17—20 followed by Joh i 1—7

Lk xix 32—45 Joh vi 41—53 Joh vi ® Mk xvi 1—8 followed by Lk i 1—3

to face p. 7

” ” p- 16 ” ” Pp. 28

facing each other between pp. 38 and 39

to face p. 257

®

INTRODUCTION.

Durine the greater part of the first nine centuries of our Era the language commonly used in the Valley of the Euphrates and the neighbouring provinces was the dialect of Aramaic which we call Syriac. The literary headquarters of the Syriac-speaking Church was the city of Edessa (in Syriac Urhdi), which also had been the centre from which Christianity spread in all that region. The beginnings of Christianity at Edessa are lost in legend, but it is certain that the new religion was well established there before the city was absorbed into the Roman Empire during the reign of Caracalla (Ap 216). The political independence of the little state accounts for the early translation of the Scriptures into the vernacular of the Euphrates Valley.

About the year 420 ap the Gospel was extant in Syriac in three forms, viz :-—

1. The present Syriac Vulgate, now called the Péshittd.

2. The Diatessaron of Tatian.

3. A translation of the Four Gospels, called by the Syrians

Evangelién da-Mépharréshé.

The Evangelién da-Mépharréshé ܝܐܘܢܓܠܝܘܢ ܕܡܦܪ̈ܫܐ)‎ Le. ‘Evangel of the Separated ones’) derives its name by contrast with the Diatessaron, which is a Harmony containing the substance of our Four Canonical Gospels arranged in one narrative. This Harmony, besides the naturalised Greek name Diatessaron ܕܝܛܣܪܘܢ)‎ (, was also called Hvangelién da-Mchallété ܐܘܢܓܠܝܘܢ ܕܡ̈ܚܠܛܐ)‎ ie. ‘Evangel of the Mixed ones’).

The main object of the following pages is to trace the history of the Evangelion da-Mepharreshe, and to determine as far as possible its relations to the Diatessaron and to the Peshitta.

B. II. 1

2 Introduction.

The Péshittd ܕܡܦܩܬܐ ܦܫܝܛܬܐ)‎ , ie. ‘The Simple Edition’) does not seem to have acquired this name earlier than the 9th century’. It is called Simple to distinguish it from the later versions of the Old and New Testaments made by Paul of Tella and Thomas of Harkel, both of which were provided with an apparatus of critical signs inserted in the text. The name Peshittw is never used by Syriac writers to distinguish the Syriac Vulgate either from the Evangelion da- Mepharreshe or from the Diatessaron, but the term is distinctive, and it is sometimes convenient to continue its use. In the same way we speak of the Latin Vulgate as opposed to pre-Hieronymian texts, although the term Vulgata editio was originally used by the Council of Trent to contrast S. Jerome’s work with the new translations of Erasmus, Beza, and others.

The Peshitta is the only version now used in the Church services. It is so used by all branches of Syriac-speaking Christendom, whether Nestorian, Monophysite, Maronite, or Malkite. This fact alone is enough to prove that its general acceptance is older than the great split between the Nestorians and Monophysites after the middle of the 5th century”, In this version Mss of the Gospels are very numerous and a few are themselves as ancient as the 5th century, but they all represent the same type of text, the variations being considerably less important than those exhibited by the better mss of the Latin Vulgate. There aré several editions of the Peshitta New Testament, none greatly differing from the editio princeps of Widmanstadius or Widmanstetter (Vienna, 1555): a useful small edition was published in 1880 at New York and subsequently reprinted. A critical edition is now being prepared at the Clarendon Press by the Rev. G. H. Gwilliam, B.D., who has collated for the purpose all the oldest codices : of this edition, the volume containing the Gospels (called Tetra- evangelium Sanctum) appeared in 1901.

The wide use of the Diatessaron in the early Syriac-speaking Church is undoubted. This work (to quote Dr Wright) “certainly gained

1 Péshittd is the pronunciation according to the ‘Nestorian’ System, which preserves the older sound of the vowels, as in Talitha and Maranatha. The Monophysites and Maronites say Péshitté. Theword is a fem. adj. in the ‘definite’ state, agreeing with mappaktd, ie. ‘ Edition,’ but Bar Hebraeus sometimes uses it by itself in the ‘absolute’ fem., hence the spelling Peshito. The form Peschzio is merely an adaptation to German orthography.

2 The Nestorian School at Edessa was finally broken up in 489.

Introduction. 3

great popularity in the early Syrian Church, and almost superseded the Separate Gospels. Aphraates quoted it; Ephraim wrote a com- mentary on it; the Doctrine of Addai (in its present shape a work’ of the latter half of the 4th century) transfers it to the apostolic times ; Rabbula, bishop of Edessa (411-435), promulgated an order that ‘the priests and deacons should take care that in every church there should be a copy of the Separate Gospels (Evangelion da-Mépharréshé), and that it should be read’; and Theodoret, bishop of Cyrrhus (423-457), swept up more than two hundred copies of it in the churches of his diocese, and introduced the four Gospels in their place: ra rv terrdpav evayyehiotav avtecnyayor evayyédua.”!

The policy of Rabbula and Theodoret was only too successful. Not a single copy of the Diatessaron has survived in anything approaching its original form—that form, I mean, in which it was known to and used by Aphraates and Ephraim. The discoveries of the last twenty years have enabled us to determine with considerable accuracy the order followed by Tatian, but it is only here and there (and generally by way of inference rather than direct testimony) that we can re- construct the actual text of the Diatessaron.

The chief sources of information about the Diatessaron now available are :—

(i) The Commentary of S. Ephraim.

This work is preserved in an Armenian translation, which has been printed in vol. 11 of ©, Ephraim’s Works (Venice, 1836). A Latin translation of the Armenian was made by the Mechitarist Aucher and edited in 1876 by G. Mésinger, but the passages quoted by Ephraim from the Diatessaron are more accurately given in an English version revised by Dr Armitage Robinson in Dr Hamlyn Hill's Harliest Life of Christ, pp. 333-377 ?.

Some fragments of the original Syriac of 8. Ephraim’s book, which moreover include a few important readings from the Diatessaron itself, are imbedded in later Syriac writers, notably the commentators Isho:- dad the Nestorian (ff. 852) and Dionysius Bar Salibi the Monophysite

± Wright’s Syriac Literature, p. 9: for further details, see Chapter 4 of this volume. 2 Repeated in Dr Hill’s Dissertation on the Gospel Commentary of S. Ephraem, pp. 75-119. The pages of Mosinger’s edition, by which the Commentary is always quoted, are to be found in

Dr Hill’s margin.

4 Introduction.

({ 1171). Neither of these somewhat voluminous compilations has as yet been published, but most of the quotations from 8. Ephraim have been collected in Dr Rendel Harris’s Fragments of the Commentary of Ephrem Syrus on the Diatessaron (Cambridge, 1895).

(ii) The quotations from the Gospel in Aphiaates.

The Homilies of Aphraates were written between the years 337 and 345. In his numerous Evangelical references and allusions he never mentions either the Diatessaron or the evangelists by name, but it is universally recognised that some at least of his quotations are from the Diatessaron rather than from the Separated Gospels. This is notably the case with the rapid survey of our Lord’s ministry at the end of Homily II (Wright's Aphraates, pp. 41-43).

(iii) The Arabie Diatessaron.

This is a careful translation of the Diatessaron from Syriac into Arabic made by the Nestorian monk Ibn at-Tayyib (+ 1043). It was edited from two mss by A. Ciasca of the Vatican Library in 1888. A Latin translation was given by Ciasca, and an English one 8 to be found in Dr Hamlyn Hill’s Lar liest Infe of Christ, published 1 in 1894. Unfortunately the Syriac text of the Diatessaron from which the Arabic was translated had been subjected to a revision which very seriously lessens its worth for critical purposes.

In its original, or at any rate earlier, form the Syriac Diatessaron was very closely akin in its renderings to the Hvangelion da-Mepharreshe. The causes which led to this textual resemblance are the subject of con- troversy, but the fact is undisputed. Moreover the Evangelists were not named in the text of the Harmony. But in Ciasca’s Arabic the text is conformed to the Peshitta, and every clause is labelled ‘ Matthew,’ ‘Mark,’ ‘Luke,’ or ‘John.’ In other words the Syriac Diatessaron from which the Arabic version was made had been prepared by identifying the Gospel passages out of which Tatian’s Harmony had been con- structed, and substituting clause by clause the corresponding passages as given in the Peshitta?.

The three documents above mentioned supply our main information about the text of the ancient Syriac Diatessaron. The Peshitta, as has

± The Latin Harmony prepared by Victor of Capua, preserved in the Codex Fuldensis, was

constructed in the same way out of what seems to have been a Latin text of Tatian’s Diutessuron.

Introduction. 5

been already stated, is preserved in many ancient mss, some as old as the 5th century. The third form of the Gospel in Syriac, the Evangelion da-Mepharreshe, is represented by the two ancient 8 called © and S in the following chapters.

Besides these primary authorities we have to reckon with the scattered quotations from the Gospel in the scanty remains of Syriac literature before the second quarter of the 5th century. It is an obviously delicate task to distinguish between quotations from the Gospels and quotations from the Diatessaron, when (as often happens) the wording of the Gospel and of the Diatessaron coincides, and it is only in the case of the Acts of Thomas that we can be sure that the writer is using the Separate Canonical Gospels. The quotations have been collected in the present work, and are discussed in Chapter 3.

The main conclusions to which I have been led may be summarised as follows :—

(1) The Peshitta is a revision of the Evangelion da-Mepharreshe, undertaken mainly with the object of conforming the translation more closely to the Greek text as read at Antioch early in the 5th century. It was prepared by Rabbula, bishop of Edessa from 411-435 ap, and published by his authority as a substitute for the Diatessaron,

(2) The Diatessaron is the earliest form of the Gospel in Syriac. It was made originally in Greek, probably at Rome, by Tatian the disciple of Justin Martyr, and translated into Syriac during Tatian’s lifetime, about 170 ap. As might be expected from a document geographically Western in origin, the Gospel text of the Diatessaron is very nearly akin to that of Codex Bezae (D) and the various forms of the Old Latin version.

(3) The Lvangelion da-Mepharreshe dates from about the year 200 ap. It was the earliest rendering of the Four separate Gospels into Syriac, but the translator was familiar with the Diatessaron and often adopted its phraseology. There is great probability that the Evangelion da-Mepharreshe was prepared under the auspices of Serapion, the bishop of Antioch who is mentioned in the Ecclesiastical History of Eusebius as having suppressed the apocryphal Gospel of Peter, and there is some reason to identify the translator with Palut, the third bishop of Edessa.

6 Introduction.

(4) In text, the Evangelion da-Mepharreshe, so far as it is a direct translation from the Greek, reproduces for us the Greek text current in Antioch at the end of the 2nd century, a text of great critical value which is often very slenderly represented in extant Greek mss. But the use of the Diatessaron by the translator has often introduced readings which really belong to the texts current in Western lands. Moreover both S and C, our two ss of the Evangelion da-Mepharreshe, contain readings which have been assimilated to the Diatessaron by transcribers ; and further, C' represents a text that has been partially revised by later Greek mss.

“L—1 1 gop Aq panoz70f OS@—LI ܬܐ‎ ± A suenquos abod ܘ‎ ‘azis [8:11480 ‘16g ‘7of fo qund saddn ‘Q xaaog

PS ae oe 7 we ngow' wor ܢ‎ Neca 0] ܢ‎ yee

Wie 3 aia 0 ܣ‎ 0 ` 50

CHAPTER LI.

DESCRIPTION OF THE MSS.

Copex C

CopEex NITRIENSIS CURETONIANUS, called in this book C consists in its present state of 825 leaves in the British Museum, numbered Add. 14451, and of three leaves at Berlin, forming the fly-leaves of Orient. Quart. 528. The British Museum leaves are described in Wright’s Catalogue, p. 73, No. cxtx. The ms came from the great Library of the Convent of 8. Mary Deipara in the Natron Valley, west of Cairo.

Later history of C.

The preservation of C appears to be the result rather of a happy accident than of reverence for antiquity. Eighty of the surviving leaves reached England in 1842 as part of a volume of the Gospels made up in the year 1222 ap from various Mss of the same size ; the other leaves of the volume were taken from copies of the Peshitta, and the binder hardly seems to have been aware that the text of 0 was different from the rest. The remaining leaves came to Europe as fly-leaves to strengthen the bindings of other books. The leaves thus used are fol. 53, containing Lk i 48—-iii 16; and the Berlin leaves, containing Joh vii 37—viil 19, Lk xv 22—xvi 12, xvu 1-23. Two more detached leaves reached the British Museum in 1847: fol. 52, the half-leaf containing fragments of Joh xiv; and fol. 72, containing Lk xiv 35—xv 21.

Of the earlier history of C we know very little. On the blank recto of the first leaf is written in a hand of about the 10th century the following note of its presentation to the Library of 8. Mary Deipara :—

8 Description of the MSS.

܀ ܐܝܬܘܗܝ ܟܬܒܐ ܗܢܐ ܕܚܒܝܒܝ ܕܝܪܝܐ ܕܫܟܢܗ | ܠܕܝܪܐ ܩܕܝܫܬܐ dasa‏ ܝܠܕܬ ܐܠܗܐ ܕܗܘܪ̈ܝܝܐ. | ܕܒ ܡܕܒܪܐ ܕܐܣܩܝܛ+ܐ. ܕܐܠܗܐ Xo‏ ܖܪ̈ܚܡܹܐ ܘܪܘܚܦܐ | am‏ ܕܡܛܠ ܫܡܗ ܡܫܒܚܝܐ ܦܪܸܲܫ ܘܝܗܼܒ | ܗܝܡܬܐ ܗܕܐ ܪܘܚܢܝܬܐ am‏ ܢܚܣܐ ܚܘ̈ܒܘܗܝ | ܘܢܫܒܘܩ ܒܘܨܪ̈ܘܗܝ ܘܢܡܢܘܗܝ (sic)‏ ܒܓܒ̈ܝܐ ܕܝܠܗ | ܒܝܘܡܐ ܕܚܝܝܢ ܖ̈ܚܡܼܘܗܝ ܒ̈ܨܠܘܬܐ ܕܟܠܗ ܚܘܕܪܐ | ܕܩܕܝܫ̈ܐ ܐܡܢܢ ܐܡܒܢܝܢ ܀ ܒܪܶܗ ܕܐܠܗܐ ܚܝܐ ܚܘܣ ܒܫܥ ܬܐ ܕܟ As‏ | ܚܛ ܝܐ ܕܣܗܪܛ ܐܡܝܢ ܀ “This book belonged to the monk Habibai, who presented it to the‏ holy monastery of the Church of the Deipara belonging to the Syrians‏ in the desert of Scete, that God, abounding in mercy and compassion,‏ for the sake of whose glorious Name he separated and gave this‏ spiritual treasure, might pardon his faults and forgive his shortcomings‏ and number him among His own elect in the day that His mercy‏ cometh to life, by the prayers of all the circle of the Saints. Amen,‏ amen !‏ “Son of the Living God, have pity in the hour of Thy judgement on‏ the sinner that wrote this. Amen!”‏

Whether C was perfect when it was presented to the Nitrian Library by the monk Habibai we do not know, but there are some indications that it was in a tattered condition before the rebinding in 1222 ap. The table of the quires given below shews a large proportion of loose leaves, and some marks on foll. 75 ¥, 76 ¥, 77 ,ܐ‎ shew that fol. 72 was once lying loose between 76 and 77, while at the same time fol. 79 was facing 75. The conjugates of foll. 77 and 79 are now at Berlin: no doubt they were loose detached leaves when they were used to strengthen the binding of the book in which they now rest.

After the rebinding in ap 1222 a few Church-lessons were marked in the margin, and a misguided person corrected some of the pages containing the Sermon on the Mount to the Peshitta text. But the original reading can in all cases be made 0111.

± The only word which presents any difficulty is 4m in Matt v 39, where Cureton failed to decipher the original reading and edited ܐܠܦܐܕ‎ (the Peshitta reading) between square brackets. In Matt ¥ 41 the word ܥܡܗ‎ has been entirely retraced by this late corrector.

Composition of C. 9

Composition of Quires, 6.

The quires of C were arranged in quinions or gatherings of five conjugate pairs. These were originally 18 in number, but two of them are now represented only by single detached leaves and six have altogether perished. The original signatures seem to have been placed at the beginning of each quire on the lower margin, but so near the right-hand edge of the leaf that all have disappeared except that upon x 1 (now fol. 43 r), which is signed ». The binder in ap 1222 signed the beginnings and ends of the quires with Syriac letters, so that e.g. the second quire has ܒ‎ on fol. 9r and ܒ‎ again on fol. 18 v. The inner leaves of Quire 11, now foll. 12-15, have been supplied by a late hand from the Peshitta. They are hardly earlier than the rebinding. The last leaf, fol. 88, is of the same period as foll. 12-15.

Present Original Quire Numeration Contents Headlines and Leaf of Folios [on verso only] Il lost 3 9 [fly leaves] 3 1 Matt i 1— Title 4 9 5 3 6 4 ܕܡܬܝ.‎ 7 5 8 6 ܕܡܬܝ.‎ 9 7 10 8 aaa 1 ܐܘܢܓܠܝܘܢ ̇ ܕܡܬܝ.‎ II 1 9 Matt vi 21—— 2 10 3 11 —viii 22 ܕܡܬܝ.‎ [4] [12] [viii 23— [5] [13] (Later ܐܘܢܨܠܝܘܢ ܕܩܠܬܝܢ‎ [6] [14] supplement) [7] [15] —x 31] 8 16 x 382— 9 17 10 18 -—xii 29 whens. ܐܘܢܓܠܝܘܢ‎

10

Original Quire and Leaf

111 1

V—VIII

2

ܚ Doe‏ ¥ ܗ ܒ

10

ܢܝܙ wp‏ ܫܙܼܝܙ ܗܡ ܗ 8¥- ܣܝ ܒ

< ܒ

ܢܝܙ ܟܬ FP w&‏ ܗ ܗ ܓ ܗ ܩܒ

ܒܙ oO‏

Description of the MSS.

Present Numeration of Folios

19 20

bo em Whe

ܨܝ

ܡܐ wb ww‏ ܐܢܬܐ ܠܬܐ ܐܐ @) ܐܝܡ

ܗ

all lost

lost lost lost lost 39 40 lost lost 41 42

Contents

Matt xii 29—

—xvili 3

Matt xviii 3—

25 ܥܕܢ

[Matt xxiii 25—

—Mk xvi 17] Mk xvi 17-20, then Joh i 1-42 [Joh i 42—iii 5] Joh iii 5— —iv 10

Headlines [on verso only]

. ܕܡܬܝ.

. ܕܡܬܝ.

ܐܘܢܓܠܝܘܢ . ܕܡܬܝِ

. ܕܡܬܝ ِ

ܕܡܬܝِ

olson i‏ . ܕܡܬܝ.

[No Headline]

ܐܘܢܓܠܝܘܢ . ܕܝܘܚܢܢ.

Original Quire and Leaf

a

XI—XIV

8 2

4 <j ܝܙ‎

ܐܐ

=

‘pel

ܟܬ ܘܢ AN OOF‏ ܒܩ ܗ

DH‏ ܗܢ ܥܙܼܙ ܡܓ ܗ ON‏ ܒܩ ܗ

OS w DOH‏ ܗ ܝ ܗ ܩ ܘܒ

Present

Composition of C.

Numeration Contents of Folios

43? 44 45 46 47 48 49 50 51

Joh iv 10—

Berlin 3° —viii 19

all lost, [Joh viii 19—Lk vii 33]

except fol.

52 Joh xiv (fragments)

and fol. 53° Lk ii 48—iii 16

lost 54

55 56 57

58 59 60 61

62

63 64 65 66 67 68 69 70 71 72

Lk vii 33—

—x 39 ‫:

Lk x 39—

—xv 21

1 This leaf is signed 4 by the original scribe.

2 Orient. Quart. 528, fol. 129. 3 Fol. 53 was probably the second leaf of Quire xiv.

11 Headlines [on verso only] . ܡܝܘܢܐܢܠܢ‎ vasa, . ܐܘܢܓܠܝܘܢ , ܕܝܘܚܢܢ‎

[No Headlines]

. oaks,

. nals,

ܐܘܢܓܠܝܘܢ ݂ ܕܠܘܩ݀ܐ

ܼ ܕܠܘܩܐ ِ

.aoals,

ܐܘܢܓܠܝܘܢ ̇ ܕܠܘܩ̈ܐ.

12 Description of the MSS.

Present Original Quire Numeration Contents Headlines and Leaf of Folios [on verso only]

XVII 1 Berlin 1! Lk xv 22— 2 lost Berlin 2° oan, 73 74 75 76 ِܿ ܕܠܘܩ݀ܐ‎ . 77 78 79 —xxi 12 ܿ ܐܘܢܓܠܝܘܢ ̇ ܕܠܘܩ݀ܐ‎

ܩܢ OP‏ ܗ MN‏ ܣܒ ܣ

ܝܐ

80 Lk xxi 12—

81

82 _vnlon. 83

84

85

86 wala. 87 —xxiv 44 lost * 10 lost

It will be noticed that the headlines are inserted on the verso of the 3rd, 6th and 10th leaves of each quire in 8. Matthew and 8. John, but in 8. Luke they occur on the 3rd, 7th and 10th leaves. This fact, even apart from the preserved signature on fol. 43r, is enough to shew that the true order is Matt Mk Joh Lk, and not Matt Lk Mk Joh. No headline occurs on fol. 40 ¥, which (on the assumption that 44 leaves are lost between fol. 38 and fol. 39) is the 6th leaf of a quire: probably the title to Joh on fol. 39r was regarded as an equivalent.

XVIII

TO oF WDD eH

ܗ ܩܒ

Codex © in its original state contained the Four Gospels in the unusual order Matt Mk Joh Lk, the beginning of 8. John following the end of 8. Mark on the same page. The portions still extant are :—

Matt i 1—vili 22, x 832—xxin 5

Mk xvi 17°-20 followed by

Joh 1 1-42°, ui 5°—vili 19*, xiv 10°-12%, 15°-19*, 215-24, 96-292,

Lk ii 48°—iii 165 vii 33°—xvi 12, xvii 1»—xxiv 44%

The colophon at the end is not preserved, but a title is prefixed to 8. Matthew which will be discussed later in connexion with the

1 Orient. Quart. 528, fol. 1. 2 Td., fol. 128. 3 The present fol. 88 is a late supplement containing Lk xxiv 44 to the end of the Gospel.

Date of C. 13

colophon of codex S. There are no lectionary marks by any early hand or any numeration of chapters, but the text is divided into sections, a new line marking the beginning of each section. At present the text is divided into short sentences by red points, but it will be shewn that these were inserted by a later hand. The writing is a large and beautiful Estrangela, the work of a practised scribe: the freedom with which the curves of the letters are formed points to the early part of the 5th century as the latest date that can be assigned to the writing. The vellum also suggests an early date, as it is very smooth and exceedingly white where not stained or otherwise damaged.

Each page contains two columns of writing, vertical lines for which were ruled in the vellum with a fine point, As is often the case with ancient Syriac Mss, horizontal lines were ruled only at the top and bottom of the columns, and the number of lines in a column consequently varies from 22 to 26. The usual number is 24 or 25. Each leaf is about 11? in. by 9 in. There are generally only three words to a line. Headings and subscriptions to the Gospels are written in red ink, as is also the first word of each Beatitude in 8. Matthew. The sign < is used in Matt viii 17 and Joh vi 32 to fill up the ends of lines accidentally left blank ; in Lk iii 11 the sign ܗ‎ is used for this purpose.

The text of C suffered very little from subsequent correction previous to the rebinding in 1222 ap. The words ܠܘܬ ܢܘܪܐ‎ in Lk xxii 56 bave been added above the line, perhaps by the original scribe, and the word sas ‘forsooth’ (which is only met with in ancient Syriac writings) has been washed out in the three places where it occurs, viz. Lk xvi 11, xx 17, xxii 70". In Joh iv 35 ܐܢܬܘܢ‎ (2°) and in Lk ix 12 ܐܥܝܪܘܗܝ‎ have been cancelled with small red dots. The word ܐܥܝܪܘܗܝ‎ is a mere slip in writing, but ܐܢܬܘܢ‎ is apparently correct and should have been left standing. The same. remark applies to ܕܐܟܘܠ‎ in Joh iv 32, which has been washed out though it. is read in Sand the Peshitta in accordance with all other authorities.

Several other words and lines have been washed out by the scribe and then written over, owing to various ordinary accidents of tran- scription, but neither in writing nor in correction is there the slightest trace of the use of a second exemplar.

1 In all three instances S has the word.

14 Description of the MSS.

The Colometry of C.

The text of C ig now divided into short sentences by a reddish- brown point, which is placed about as often as a comma or larger stop would come in an English book. At the end of paragraphs the same hand has added larger and more ornamental stops, substituting for the

© plain point or :ܘ‎ of the original scribe figures such as 0-0 or 900.

Where the paragraph ends a line, so that no room for a stop is left, tha punctuator inserted ee ܗܗ ܗܘ‎ between the lines, though frequently there was but little room for this, the original scribe having left no space vacant between the lines except where he intended a paragraph to end.

It is obviously difficult to fix the date of a scribe whose work is confined to mere dots; but a fortunate accident makes it clear that the dots were inserted after the sheets were already sewn, and that the hand was not that of the regular rubricator who wrote the occasional headlines. For the dots are entirely absent from /fol/. 48 ¥, 49 r (Joh vi 30°-53*). The only mark of punctuation visible when the book is open at this place is a single point at the end of vi 51 ܕܥܠܝܡܐ.)‎ ,mais), where the paragraph ends. As soon as the leaf is turned over the points begin again. Now this could hardly have happened otherwise than through careless turning over of two leaves at the same time: in other words, 0 was already a bound book when the punctuator was at work. Moreover he must have written currente calamo, with much the same haste as according to the common story the New Testament was divided into verses ; we cannot imagine that he would not have found out his mistake if he had been copying the punctu- ation from another exemplar. Thus the colometry of the Curetonian Syriac represents nothing more than the individual judgement of a reader.

That this reader was not the rubricator of the Ms appears from the fact that fol. 48 ¥ is one of the pages which has -,asasx- written in red upon the upper margin as a headline. The colour also of the red ink is different, that of the punctuation dots being sensibly browner than the headlines.

Colometry of C. 15

The recto of the detached half-leaf containing fragments from Joh xiv is also without punctuation, but it appears on the verso. The same accident therefore of turning over two leaves together occurred here as well as in Joh vi.

For some reason which I am quite unable to divine the punctuation throughout the whole of the first chapter of S. Matthew has been carefully washed out. The result may be clearly seen in the photograph of C published in F. G. Kenyon’s Our Bible and the Ancient MSs, facing p. 155.

What punctuation was inserted in C by the first hand is very difficult to determine, as most of the places where punctuation is natural have been covered by the red dots. A small black dot was generally placed at the end of paragraphs, e.g. at the end of Matt i 23, vi 23, etc.; in other places a slightly more elaborate stop was used, @.g. ܒܐܝܕܗ ...ܘ‎ Lk xiii 17; but often no room was left for any stop at all, e.g. in Matt xxiii 14 etches comes quite to the end of the linet. There is a dot by the first hand at the end of the short interrogative sentence in Lk vili 45 (.A ܩܪܒ‎ ars), but if we may judge by the pages left untouched by the hand who inserted the red dots, the original scribe of C hardly gave any punctuation at all. This is very. uncommon in Syriac Mss, but much the same state of things once obtained in cod. B of Aphraates. Besides Lk viii 45, the only places I have noted as having stops certainly by the original hand are ܕܫܒܬܐ.‎ Lk xiii 14 and ܥܝ̈ܢܝܗܘܢ‎ Lk xxiv 31: both instances occur at the ends of lines. On the other hand it is clear that no stop was intended by the original scribe after ܓܒܳܪܶܝܢ‎ Matt xv 38, after ܕܐܝܬܝ‎ Matt xvi 15, or after ܘܒܫܘܒܝܝܐ‎ Lk xxi 27.

In editing CI have inserted the dots of the punctuator, as they are on the whole a satisfactory division of the text. But it must be repeated that they have no claim to represent a traditional Old Syriac colometry.

1 The ܘܘ ܘܘ‎ oo is here inserted by the later punctuator between the lines.

16 Description of the MSS.

Cureton’s Edition.

Codex C derives its name from Dr Cureton, who edited the text in full in 1858. The title of his work is Remains of a very antient Recension of the Four Gospels in Syriac, hitherto unknown in Europe ; discovered, edited, and translated by William Cureton, D.D., F.RS....- London, 1858. The three Berlin leaves form part of a ms bought in Egypt by Dr Brugsch the Egyptologist : they were edited by Roediger in the Monatsbericht der Kéniglich Preussischen Akademie der Wissenschaften zu Berlin for July, 1872, p. 557. A small edition of 100 copies was printed for private circulation to range with Cureton’s book by Dr W. Wright.

Cureton’s edition gives the Syriac text of C with great fidelity. His introduction contains much interesting and curious matter, then to a great extent new and unfamiliar, but now almost entirely superseded by the labours of two generations of Syriac scholars. Cureton’s theory that the text of S. Matthew's Gospel in @ retained to a great extent “the identical terms and expressions which the Apostle himself employed” (p. xciii) attracted a good deal of attention, but gained few converts: it is suficient refutation to point out that Edessene Syriac is quite a different dialect from the Aramaic of Palestine. The same claim had been advanced for the Peshitta by Widmanstadius in 1555: Ex, quibus omnibus, he says, conecturam non leuem capi posse arbitror, et Mathewm Euangelium suum, & Paulum ad Hebreos Epistolam sermone Syro, Hebraici populi vulgari vsu trite, vt & Iudeis passim omnibus intelligerentur, scripsisse, eaque in Syrorum Eeclesys vam vsque a temporibus Apostolorum conseruata fuisse (Preface to the Ed. Prine. of the Peshitta, fol. ® #x##3).

000 ܨ

WILLIAM CURETON, D.D., F.R.S.

Assistant-Keeper of MSS at the British Museum and Canon of Westminster. b. 1808; d. 1864.

Recent history of the Sinai Palimpsest. 17

CopEx S.

Coprx Parrupssstus Srvarricus is No. 30 among the Syriac Mss in the Convent of 8, Catharine on Mount Sinai. A description is given in Mrs Lewis's “ Catalogue of the Syriac mss...on Mount Sinai” (Studia Sinaitica 1), pp. 43-47.

Recent history of S.

The Syriac Library at the Convent on Mount Sinai was first explored by Mr (now Dr) J. Rendel Harris and Mr Bliss in 1889, when Dr Harris discovered the Syriac translation of the early Christian Apology of Aristides. In 1892 the Convent was visited by Mrs Lewis and her sister Mrs Gibson, of Cambridge, who saw S among the other volumes of the Library. Struck by the antique appearance of the lower writing of the palimpsest, which they knew from the still visible headlines to be a ms of the Gospels, these ladies took photographs of the whole volume. On their return to Cambridge a few of the more legible pages were deciphered by the late Prof. R. L. Bensly and the present writer!. Early in 1893 the palimpsest itself was transcribed at Sinai by Prof. Bensly, Dr Rendel Harris and myself; on the same occasion some more photographs were taken by Mrs Lewis, who was also of the party. Our transcript was published at Cambridge in 1894, after Prof. Bensly’s lamented death?. Mrs Lewis visited Sinai again in 1895, and transcribed some more passages with the help of a reagent: these were published in 18969, In 1897 Mrs Lewis went

1 It was not always an easy matter at first to discover from what part of the Gospels any given page was taken, and when that was done it was not always a page where @ was extant and

characteristically different from syr.vg. I remember that the first reading we made out where the photograph clearly agreed with the Curetonian against the Peshitta was ܘܢ‎ (instead of (ܬܫܒܩܘܢ‎ at the end of Matt xxiii 23.

2 The Four Gospels in Syriac transcribed from the Sinaitic Palimpsest by the late Robert L. Bensly, M.A., and by J. Rendel Harris, M.A., and by F. Crawford Burkitt, M.A., with an Introduction by Agnes Smith Lewis; edited for the Syndics of the University Press, Cambridge, 1894. I quote this volume as “Syndics’ Edition.”

3 Some Pages of the Four Gospels re-transcribed from the Sinaitic Palimpsest...by Agnes Smith Lewis; London, 1896. I quote this volume as “Some Pages.”

B, IL. 3

18 Description of the MSS.

for the fourth time to Sinai, bringing back. several corrections, which she published in the Hxpositor for August, 1897, pp. 111-119, and also a series of excellent photographs much clearer than any previously taken. Complete sets have been presented by her to the Cambridge University Library; to Westminster College, Cambridge; to the University Library, Halle ; and to the Rylands’ Library, Manchester.

Thus the editor of S has to take into consideration at least three publications, and he should also consult the photographs. In printing the text I have adopted the rule of silently following the latest published reading; when for any reason another reading is given, the reader is expressly warned in the notes. This chiefly occurs where I have succeeded in correcting the printed text from the photographs ; in this way the text of S as now given differs in over 250 places from what has been hitherto published. The corrections are registered in Appendix 111 to the first volume of this work.

When and where the wpper writing was transcribed.

The upper writing of the Sinai Palimpsest is dated in the year of the Greeks 1090, 1.6. 778 ap (fol. 181 v). The determination of the place where it was written is of interest to students of the Hvangelion da-Mepharreshe, as giving some indication of the home of S in the 8th century.

In its present state the volume consists of 182 leaves of vellum, including a leaf pasted into the cover; the quires are quinions, except the last, which has six pairs. These quires are numbered doubly, wiz. with Syriac letters running from right to left and with Georgian signatures running the opposite 1877. The Georgian signatures were first observed by Mrs Gibson and should have afforded some clue to the place of writing, for it is not everywhere that we come across traces of Georgian communities in Syria.

The contents of the book are chiefly taken up with a collection of twelve Lives of Female Saints. The scribe gives his name three times

Mzeto ܝܘܚܢܢ ܚܒܝܫܐ ܕܒܝܬ ܡܪܝ‎ (fol. 2v) John the anchorite of Beth Mari the Saint.

1 Abridged from Mrs Lewis’s description in Syndies’ Edition, p. vi.

Upper writing of the Sinai Palimpsest. 19

(fol. 165v)‏ ܝܘܚܢܢ ܚܒܝܫܝܐ ܕܒܝܬ ܡܪܝ ܘܢܘܢ ܒܕܝܫܐ ATI‏ ܡܨܪܝܢ

ܡܕܝܢܬܐ

John the anchorite of Beth Mari Qanon the Saint of Matarrath Mesrén city. ܕܡܥܪܬ ܡܨܪܝܢ‎ a ܝܘܚܼܢܢ ܐܣܛܘܢܝܐ ܕܒܝܬ ܡܪܝ ܘܢܘܢ‎ (fol. 1811)

ܡܕܝܢܬܐ ܟܘܪ ܕܐܢܛܝܘܟܝܐ

John the stylite of Beth Mart Qanon, a monastery of Ma‘arrath Mesrén city in the district of Antioch}.

The natural interpretation of these sentences is that the book was written by a stylite monk called John in a certain monastery at Ma‘arrath Mesrén (.,, 46 sje), Which is a small town in Lat. 36°y, about equidistant from Antioch and Aleppo. This place is mentioned in Ydqit 1v 574 and in the various authorities for the history of the Crusades, but I have entirely failed to discover any mention of the seribe’s monastery. It is curious that J ass does not appear in the sentence quoted from fol. 2 ¥, The natural meaning of ܡܪܝ ܘܥܘܢ‎ dus is ‘the House (ie. Convent) of 8. Qanon, but we can hardly sup- pose the scribe to have accidentally dropped the Saint’s name in the first lines of his preface. ,t therefore seems to be a proper name. There is a village called Bét Merri near Beyrout, which also contained a monastery.

But wherever our palimpsest was transcribed, it is certain that it reached Mount Sinai not alone, but as part of a considerable library. In the first place there is a likelihood that all the numerous mss at Sinai which are either in the Georgian language, or (like our palimpsest) contain Georgian writing, came originally from the same collection. Still clearer is the case with regard to cod. 588 of the Arabic mss at Sinai, a vellum book containing a Prophetologion in Arabic written over various Syriac Apocrypha. One of these fragments appears to have belonged to the same ms of the Transitus Mariae (Kotpunous

17 am sure the last word but one is 4as ‘district, not sas ‘star’ (as read by Mrs

Lewis in Studia Sinattica IX, p. xxiv, at the suggestion of Dr Nestle). The word occurs at the end of a line, and there is a small gap between the o and the 4 which I think was left blank by

the scribe, a being a letter which cannot be prolonged. ‘ass is derived from the Greek and so can be used in the absolute state, e.g. santa Aas Lk iii 1 SC; but ܕܧܐܐܕܢܛܝܘܦܝ̈ܐܕ‎ oan is grammatically improbable.

20 Description of the MSS.

Mapias) as was also used for this palimpsest by John of Ma‘arrath Mesrén. “This identification rests (1) on the similarity of the vellum, (2) on the probable agreement in size, the leaves of the Syriac Apocryphal text used by John the Recluse having been slightly cut at the edges in order to harmonize with the rest of the volume, (3) on the character of the script, and (4) on the coincidence of the contents?.” Furthermore this same cod. 588 is a double palimpsest, four leaves of the Syriac Apocrypha being written over fragments of the Third Book of Kings in the Palestinian Syriac dialect? This brings the book written by John of Ma‘arrath Mesrén into connexion with the ancient Palestinian Syriac fragments at Sinai and St Petersburg, which also are in some instances covered with Georgian writing.

It may be permitted, in the absence of certain information, to conjecture the fate of this Library of mss in Edessene and Palestinian Syriac, in Georgian and in Arabic, to which our palimpsest once belonged. The town and district of Ma‘arrath Mesrén was a portion of the Latin Principality of Antioch granted by Alexius to Bohemund about 1100 ap*. But in granting the temporal dominion to the foreign Crusader the Emperor expressly reserved the appointment of the Patriarch of Antioch, who was to be chosen from the Constantinopolitan clergy. This meant the predominance of ‘ Melkite’ influence throughout the Principality ; the Monophysite Patriarch had already left Antioch to take refuge in a Mohammedan land, and we may assume that the monks of Ma‘arrath Mesrén were not long in making their submission to Constantinople and Chalcedon. But the one fact which comes out from historical sources about Ma‘arrath Mesrén is that it was continually exposed to the chances of war; it was often raided, often retaken, and must have become more and more unsuitable as a vesting place for stylites and anchorites. In the same way therefore as Monophysite monks took refuge in the Natron Valley we may suppose that the monks of Ma‘arrath Mesrén migrated to the great Orthodox sanctuary of Sinai, bringing their books with them. At least there is nothing improbable in the conjecture.

1 Syndics’ Edition, p. xvii. The identification is due to Dr Rendel Harris. ° Discovered by my friend Mr J. F. Stenning, and edited by him in Anecdota Oxvonensia, 1896. 3 Alexiad xiii 12.

Conwposition of the Sinai Palinypsest. 21

The original MSS.

John of Ma‘arrath Mesrén wrote his book of Saints’ Lives on portions of five older mss. These are

1. 142 leaves of the ms of the Evangelion da-Mepharreshe, here called S.

2. 4 leaves, containing fragments of the Gospel of 8. John, in fine square Greek uncials of the 4th or 5th century.

3. 20 leaves, containing fragments of the Acta Thomae in a Syriac hand of the 5th century.

4. 4 leaves, containing fragments of the Transitus Mariae (Koipnous Mapias), in a Syriac hand of the 5th or 6th century.

5. 12 leaves, containing fragments of Homilies in fine sloping Greek uncials of the 6th century or earlier.

Nos. 4 and 5 were taken from mss much larger than the others, aud each leaf has been seriously cut to make it fit, but the leaves of S were very little trimmed when they were written over.

All the leaves not taken from S were arranged by John of Ma‘arrath Mesrén in the last four quires of his book, so that we may regard them as a makeshift, resorted to when the supply from S was exhausted. The whole of S is preserved except 22 leaves, and these 22 include the blank fly-leaves originally placed at the beginning and end of the volume. These leaves and their conjugates are generally the first to go in neglected codices, so that their absence in this instance is not surprising. Probably therefore S was taken to pieces for the express purpose of supplying vellum for the existing palimpsest, and consequently we learn that the 22 leaves were already missing in 778 AD—a small number for so ancient a ms. The 12 leaves from the ms of Greek Homilies occupy the final quire, while the other miscellaneous leaves were used to make up the 15th, 16th and 17th quires.

Portions of two leaves of the Greek Gospel fragmeuts have been made out, which formed the outside pages of a quire containing Joh vii 6—ix 238. The text is given in Studia Simantica rx, pp. 45, 46. In the same volume, pp. 23-44, the present writer has edited eight of the more legible pages of the fragments of the Acts of Thomas. These fragments are at least four hundred years older than any other known

29 Description of the MSS.

text of the Acts of Thomas, and in view of the importance of the Acts for the criticism of the Evangelion da-Mepharreshe itself I add here a complete Table of the three quires of the palimpsest in which the fragments occur.

Quires xv—xvil of the Sinac Palimpsest (Studia Sinaitica 1x, p. 25)

2 Pages of Wright’s Quire

and Leaf _-—-Photographs nee 1 Text) XV 1 281, 282 Acts of Thomas 2997—3017 2 283, 284 Greek Gospel (Joh vii 6— ) 3 285, 286 Syriac Gospel (Mk xiv 64—xv 19) 4 287, 288 Greek Gospel 5 289, 290 Acts of Thomas 317*— 319° 6 291, 292 Acts of Thomas 315*— 317 7 293, 294 Greek Gospel 8 295, 296 Syriac Gospel (Mk xii 19-42) 9 297, 298 Greek Gospel (Joh —ix 23) 10 299, 300 Acts of Thomas 301%—303” XVI 1 301, 302 Transitus Mariae 2 303, 304 Acts of Thomas 321—323 3 305, 306 Acts of Thomas 185°—187 4 307, 308 Acts of Thomas 305—307 5 309, 310 Transitus Mariae 6 311, 312 Transitus Mariae 7 313, 314 Acts of Thomas 295%—2975 8 315, 316 Acts of Thomas 198” 202" (sic) 9 317, 318 Acts of Thomas 3118—313° 10 319, 320 Transitus Mariae XVII 1 321, 322 Acts of Thomas 209—211"° 2 323, 324 Acts of Thomas 297°—299° 3 325, 326 Acts of Thomas 309’—3118 4 327, 328 Acts of Thomas 2534—255 5 329, 330 Acts of Thomas 319°—321 6 331, 332 Acts of Thomas 313°—3154 7 333, 334 Acts of Thomas 237 —2390 8 335, 336 Acts of Thomas 323—325 9 337, 338 Acts of Thomas 303"—305 10 339, 340 Acts of Thomas 211%_213 XVIII 1—12 =. 341364 Greek Homilies (still unidentified)

Composition of 8. 23

We come at last to S itself, the ms of the Evangelion da-Mephar- reshe. In its original form S was a vellum codex consisting of 166 leaves, on which were written the Four Gospels in the usual order Matt Mk Lk Joh. The following Table shews the original arrangement, together with the numeration of Mrs Lewis's photographs!, which follows the paging of the ms as it now is.

The original composition of S.

es Content 0 Il missing 2 ܝ‎ missing 3 Matt i 1—17 163, 164 4 i 17—ii 15 112, 111 5 ii 15—iii 17 83, 84 6 iii 17—v 1 97, 98 7 ¥ 1—26 110, 109 8 v 26—vi 10 177, 178 9 [vi 10-- missing 10 ܐ]‎ 1 —viii 3] missing 11 1 ܐ‎ Matt viii 3—30 20, 19 2 viii 30—ix 23 149, 150 3 ix 23—x 15 229, 230 4 x 15—xil 74, 73 5 xi 1—30 192, 191 6 xii 1—31 190, 189 7 xii 31—xili 5 68, 67 8 xiii 6—31 231, 232 9 xiii 31—xiv 1 151, 152 10 xiv 1—31 2,1 111 1 Matt xiv 31—xv 27 47, 48 2 xv 27—xvi 15 61, 62 3 [xvi 15—xvii 11] missing 4 xvii ]1—xvili 8 141, 142 5 xvili 9—xix 3 34, 33 6 xix 3—28 28, 27 7 xix 28—xx 24 159, 160 8 [xx 24—xxi 20] missing 9 xxi 20—43 79, 80 10 ܒ=‎ xxi 43—xxii 27 53, 54

1 See above, p. 18.

Description of the MSS.

Ancient Gir Cont < Iv 1 ܒ‎ Matt xxit 27—xxiii 15 89, 90 2 xxili 15—xxiv 2 269, 270 3 xxiv 2—31 225, 226 4 xxiv 31—xxv 12 200, 199 5 xxv 12—37 147, 148 6 xxv 37—xxvi 7 153, 154 7 xxvi 17-—44 182, 181 8 xxvi 4467 235, 236 9 xxvi 67—xxvii 19 271, 272 10 xxvii 20—47 91, 92 ¥ 1 Matt xxvil 47—xxviii 7 267, 268 2 [xxviii 7—Mark i 12] missing 3 Mark i 12—44 60, 59 4 [i 44—1i 21] missing 5 ii 21—iii 21 201, 202 6 iii 21—iv 17 219, 220 7 [iv 17—41] missing 8 iv 41—v 26 42, 41 9 [v 26-—vi 5] missing 10 ܓ‎ vi 5—28 273, 274 VIIA Mark vi 28—54 106, 105 2 vi 54—vii 23 49, 50 3 vii 23—viii 14 81, 82 4 viii 1438 187, 188 5 vili 388—ix 25 193, 194 6 ix 25—x 2 99, 100 7 x 9- 7 51, 9 8 × 27—xi 1 116, 115 VII 1 Mark xi 1—27 56, 55 2 xi 27—xii 9 128, 127 3 xii 19—42 296, 295 4 xii 42—xiii 23 96, 95 5 xiii 24—xiv 10 23, 24 6 xiv 10—36 37, 38 7 xiv 36—63 86, 85 8 xiv 64—xv 19 286, 285 9 xv 19—47 134, 133 10 oa xvi 1—Luke i 16 46, 45

Ancient Quire and Leaf

VIII 1

ܒܙ 8 bd‏ 8 = ܢܝܙ ܟܬ ܘ ܥܫ ܗ ܗ ܒ¬- ܤܣ ܩ ܗ ܢܝ ܟܬ ܘ ܥܙܼܝܙ ܗ ܗ ܒ ܤܣ ܩܘ ܗ

ܢܝܙ ܙܐ ܩܢ ܦܙܝ ܗܓ ܗ ܢ ܣ ܒ ܗ

ܫܒ

ܬܚܐ

ܫ

_

WD‏ ܥܝ ܗ ܗ ܕܝ ܗ ܩ ܗ

[>]

ܗ

[ܗ]

]ܘ[

Composition of 8.

Contents

Luke [i 16—38] i 38—73 i 73—ii 18 ii 18—40 ii 40—iii 9 iii 9—iv 1 iv 1—26 iv 26—v 6 v 6—28 [v 28—vi 11]

Luke vi 12—35 vi 35—vii 6 vii 7—28 vii 28—viii 1 vili 1—22 viii 22—40 viii 40—ix 6 ix 6—27 ix 27—49 ix 49—x 11

Luke x 11—33

x 33—xi 13 xi 13—32

xi 32—52

xi 52—x1i 21 xii 21—42 xii 42—xiii 3 xiii 3—22 xili 22—-xiv 5 xiv 5—23

Luke xiv 24—xv 12 xv 13—xvi 2 xvi 2—21 xvi 21—xvii 9 xvii 9—33

xvii 33—xviii 16

xviii 17—40

xviii 40—xix 22

xix 22—45 xix 45—xx 21

Photographs (and pp. of ms)

missing

183, 184 276, 275 157, 158 101, 102 119, 120 143, 144 266, 265 197, 198

missing

214, 213 94, 93 205, 206 7,8

8, 4 17, 18 13, 14 215, 216 88, 87 208, 207

243, 244 9, 10

125, 126 132, 131 221, 222 239, 240 130, 129 135, 136 11, 12

257, 258

103, 104 162, 161 25, 26 65, 66 176, 175 166, 165 75, 76 35, 36 180, 179 117, 118

ܝ

Ancient Quire and Leaf

XII

1 9

wr wre‏ ܗ ܕ ܗ ܩܒ

ܐܡܪ ܗ

XIII

DANAIanFwh ܢܝܙ‎

[9]

]ܐ[

]ܐ[

Description of the MSS.

Contents

Luke xx 21—44

John

xx 44—xxi 23 xxi 23—xxii 8 xxii 8—34

xxii 34—59

xxii 59—xxiii 14 xxiii 14—38 xxiii 838—xxiv 5 xxiv 5—26 | xxiv 26—fin.

[i 1—25]

1 25—47

[i 47—ii 15] ii 16—iii 11 iii 11—31 iii 31—iv 15 iv 15-37 [iv 38—v 6] ¥ 6—25

[v 25—46]

John v 46—vi 19

John

vi 20—44

vi 44—69

vi 69—vii 21 vii 21—39

vii 39—viii 21 viii 21—41 viii 41—ix 1 ix 1—21

ix 21—x 1

x 1—23

x 23—xi 5

xi 5—31

xi 31—48

xi 48—xii 7 xii 7—28

xii 28—49

xii 49—xiii 18

Photographs (and pp. of ms)

58, 57 941, 9 109, 0 174, 3 78, 7 64, 63 168, 167 171, 172 259, 260 44, 43

missing 203, 204 missing 280, 279 185, 186 195, 196 262, 261 nvissing 217, 218

missing

124, 123 245, 246 249, 250 247, 248 32, 31

30, 29

253, 254 251, 252 255, 256 138, 7

145, 146 209, 210 107, 108 121, 122 139, 140 113, 114 211, 212 155, 156

Quire signatures in 8. 27

Ancient Quire Photographs and Leat Contents (and pp. of ms) XVI 1 [a] John xiii 19—xiv 1 21, 22

2 xiv 1—24 5, 6

3 xiv 24—xv 15 233, 234 4 xv 15—xvi 10 238, 237 5 xvi 10—32 224, 223 6 xvi 32—xvii 20 227, 228 7 xvii 20—xviii [24] 15, 16

8 xviii 14—31 39, 40

XVII 1 John [xviii 31— missing

2 — missing

3 — xix 40] missiny

4 xix 40—xx 17 2638, 264 5 xx 17—xxi 2 69, 70

6 xxi 2—17 71, 72

7 xxi 17—fin. 277, 278 8 # missing

9 ® missing 10 [ܛ]‎ * missing

A Table similar to the preceding is given in the Syndics’ Edition, pp. xxx ff, but at that time not enough of the original signatures had been found to make clear upon what system they were inserted. The system actually employed is very curious. Instead of signing the first quire with ,ܐ‎ the second with ,ܒ‎ and so on, either uniformly at the end or uniformly at the beginning of a quire, the scribe signed Quire 1 at the end with ܐ‎ and Quire 2 at the beginning with pw, Quire 3 at the end with ܒ‎ and Quire 4 at the beginning with ,ܒ‎ and so on. The signatures are in each case placed in the inner corner of the lower margin, thus :

first leaf of Quire 4 last leaf of Quire 3

28 Description of the MSS.

In the Table I have indicated all the places where signatures ought to occur, adding square brackets where the letter is not actually visible in the photograph.

In general appearance S is not unlike C. There is no title at all prefixed to 8. Matthew: at least no reagent used has brought up even the faintest indication of a letter’. The writing begins on a verso, and the recto of that leaf (Photograph 163) was absolutely blank. At the end of 8. John there is a colophon written by the original scribe, followed by a note in another almost contemporary hand, now unfortu- nately illegible: these will be discussed later in connexion with the heading to 8. Matthew in C. There are no lectionary marks in S nor any numeration of chapters, but as in @ the text is divided into sections, each section beginning with a new line. In many places the original hand has marked the end of sentences and even single clauses with a small point, but it is evident that this punctuation was somewhat irregular, even when due allowance is made for the loss of detached dots under the upper writing of the palimpsest. The writing is a very beautiful Estrangela, even more rapidly formed than that of C: it cannot be later than the beginning of the 5th century and is not inconsistent with an earlier date still. The vellum is now somewhat crinkled and in places rather brittle, partly owing to the washing process undergone in preparing the leaves to receive the later writing, partly owing to assiduous thumbing by readers of the lives of Female Saints. The washing process must have made the detached sheets of vellum quite limp, as in six instances (v1 1, 8; vil 2, 9; vu 4, 7; X1 2,9; x1v 1,10; xv13, 6) the conjugate leaves are now folded the re- verse way. On one or two pages the surface has a tendency to scale off’.

Each page contains two columns of writing, vertical lines for which were ruled with a stylus; the sharp point has often made a round dot at the end of these vertical lines, giving the appearance of punctua- tion at the beginning or end of the first and last lines of the columns’.

± We should have expected at least ܐܘܢܓܠܝܘܢ‎ for a headline, since ates occurs as the headline to the following verso (Photograph 112). See below, p. 33.

% I cannot agree with Dr Harris’s opinion (Syndics’ Edition, p. xxxv) that some pages have been scraped with a knife. But however this may be, it should be clearly stated that there is no evidence at all that S has been intentionally defaced in particular places for dogmatic reasons.

% Eg. Matt i 1 (ܦܬܒܐ)‎ and Mk vii 30 ܐܫܦܚܬܗ:)‎ In each case the dot should be omitted.

ܘ ܡܢ iu‏ SOG,‏ ܛܘܢ ܢܐ

ܓ

abe < aude Ser

Copex 5, fol. 907, as now bound (Lk xix 32—45).

Orthographical signs in 8. 29

1 cannot be sure whether horizontal lines were ruled to join the ends of the vertical lines, but certainly no other horizontal lines were ruled, and Consequently the number of lines in a column varies even more than ܬܐ‎ C—from 29 in 8, Matthew to 21 in 8. John. I think that the Whole ws is the work of one scribe, but the writing gradually became “tger and the lines in a column fewer as the work progressed. The caves now measure 88 in. by 64 in. There are generally only three words toa line, though in the earlier parts of the ms four words a line and 111 the later parts two words a line are not uncommon. The subscriptions to the Gospels and the colophon are in red. The headlines consist of ܐܘܢܓܠܝܘܢ‎ on the verso and ܕܡܬܝ‎ (or the corresponding name) on the recto ; 1t seems to have been intended to insert them on every page, but in many instances they are no longer legible’. The sign % 18 placed in Lk xii 18, xxi 15, xxiv 4, to 1 up blank spaces at the end of lines where no pause was intended, but in Matt ix 25 and Joh xiv 31 a horizontal stroke is used for the purpose.

The very few corrections visible in S seem all to have been made by the original scribe and arise from the ordinary accidents of transcription, not from the use of a second exemplar. Words accidentally repeated or inserted are deleted by means of dots in Mk vi 18, Lk xiv 12; Joh ii 8 and x 18. In Lk xxi 24 ܕܚܘܿܪܒܐ‎ , the dot indicates that the word should be ܕܚܪܒܐ‎ , Words and letters accidentally dropped are inserted by the original scribe between the lines in Matt ¥ 48 ܐܝܦܢܐ‎ Lk xii 43 ܢܐܬܐ‎ , Joh xiii 16 ܐܡܪܢܐ‎ 8110 two steps in the Genealogy

ܗ 4

Lk iii 33; also Mk xii 1 ܡܥܨܬܐ‎ , Lk xvii 23 ܬܪܛܘܢ‎ . If I have

rightly conjectured, the word ܡܝܢ ܠܐ‎ was inserted in this way between

Lk ix 17 and 18. A few letters have here and there been corrected by the first hand: instances occur Matt ix 20, xx 28, Mk 111 14.

The only orthographical signs found in S, with the exception of

(sic) in Mk xv 29, are the seydmé marks for the plural. These seem‏ ܐܘ hardly ever to have been dropped, though now the dots are often‏ illegible. In addition to their regular use over nouns they are used for‏ sometimes‏ ܐܙ ܡܝܢ the numerals and for the present participle, especially‏ Lk viii 56, and‏ ܢܐ ܡܪܘܢ they occur where they are not wanted, e.g.‏

1 At the head of x1 1 r (Photograph 103) ܐܘܓܓܠܝܘܢ‎ is written, apparently by mistake.

30 Description of the MSS.

even ܕܐܝܣܪܖ̈ܐܝܠ‎ Mk xv 32. The other dots usually found in Syriac 8 are absent, even the point which distinguishes the fem. suffix -dh from the masc. suffix -eh: eg. in S both leh ‘to him’ and lah ‘to her’ are written ol. In this S differs from C, which has the dot for the fem. sutix and sometimes also to distinguish am haw (‘that’) from ܗܘ‎ hi (‘he’)!. Neither in S nor C is there any sign to distinguish the various parts of the verb which are written with the same consonants.

Titles, Subscriptions and Colophons in C and 8.

The end of 8. Mark with the beginning of the following Gospel is preserved both in Cand S. We there read

C S ܫܠܡ ܐܘܢܓܠܝܘܢ ܫܠܡ. ܐܘܢܓܠܝܘܢ̇‎ ܕܡܪܩܘܣ. ܝܨ. ܕܡܪܩܘܣ ܢܤ‎

1 6:68: splat . ܐܘܢܓܠܝܘܢ ܕܠܘܩܐ. ܐܘܢܓ ܠܝܘܢ‎ Endeth Evangel Endeth Evangel of Mark. of Mark. en te Sica aware are: Evangel of Sohn. Evangel of Luke.

Similarly, at the end of 8. Luke S has (at the bottom of a column)

| ܫܠܡ ܐܘܢܓܠܝܘܢ | ܕܠܘܩܐ | ܐܘܢܓܠܝܘܢ ܕܝܘܚܢܢ‎ Endeth Evangel | of Lnke. | Evangel of John. |

The beginning of 8. John’s Gospel no doubt stood at the head of the next left-hand column, but the leaf which contained it is unfortunately missing.

These simple colophons, found both in @ and in S, differ from those in the codices of the Peshitta, which contain the peculiar phrase

ܐܘܢܓܠܝܘܢ ܟܪܘܙܘܬܐ ܕ[ܠܘܩܐ] The Evangel, the preaching of [Luke].‏

1 Hg. Matt ¥ 12 € Lk xv 4 € The point for the fem. suffix 844 is sometimes missed

even in ¢ ag. ܠܪܚܿܡܬܗܿ ܘܠܫܒ̈ܒܬܗ‎ Lk xv 9.

Colophon of 8. 31

This is not mere verbiage, but an attempt to render the Greek preposition kard. The Evangelion da-Mepharreshé had been content to render Evayyéuov kata Aoveav by Evangel or Luke, but the Peshitta wishes to emphasize the fact that it is not the ‘Gospel of Luke’ but the ‘Gospel according to Luke.’ Besides this, most codices of the Peshitta prefix ‘ Holy’ to ‘Gospel’ and add the reputed places where the several Gospels were composed, after the manner of many Greek minuscules}.

The end of C is not preserved, but we read in S at the end of 8. John

ܫܠܡ ܐܘܢܓܠܝܘܢ | ܕܡܦܪ̈ܫܐ ܐܪ̈ܒܥܐ | ܣܦܪ̈ܝܢ. ܫܘܒܚܐ | ܠܐܠܗܐ ܘܠ ܡܫܝܚܗ | ܘܠܪܘܚܗ ܩܕܝܫܬܐ | ܟܘܠ ܕܩܪܐ ܘܫܡܥ | ܘܢܛܪ sana‏ ܢܨܠܐ | ܥܠ ܚܛܝܐ shar‏ | ܐܠܗܐ ܒܖ̈ܚܡܘܗܝ | ܢܫܒܘܩ ܠܗ ?ܚܛܗܘܗܝ| ܒܬܪܝܗܘܢ cals‏ ¡ ܐܡܝܢ ܘܐܡܝܢ | . -o‏ |

Endeth the Evangelion da-Mépharréshé, Four 0 Glory to God and to His Messiah and tv His holy Spirit. Every one that readeth and heareth and keepeth and doeth it pray for the sinner that wrote; (fod in His compassion forgive him his sins in both worlds. Amen and Amen.

This colophon is noteworthy for two reasons. In the first place it distinctly describes S as a copy of the Evangelion da-Mepharreshe, the Evangel of (or, ‘according to’) the Separated, ie. the Gospels divided into the four volumes of Matthew, Mark, Luke, and John, and not mixed together as in the Diatessaron. The importance of this will be especially felt when we come to emend the not fully preserved title in C. But the way the Three Persons of the Trinity are mentioned is even more remarkable. It is not exactly unorthodox, for the co-ordination of the Three Persons by a simple AND was the watchword of orthodoxy +, but it is unusual to find the Holy Spirit treated as feminine in these

1 Matt is said to have been written in Hebrew in Palestine; Mk in Latin at Rome; Zk in Greek at Alexandria; John in Greek at Ephesus: see Gwilliam’s Tetraewangelium, pp. 194, 314, 478, 604.

2 The scribe apparently wrote ܢܛܘܗܘ ܗܝ‎ , by a slip of the pen.

3 That we should read sefrin ‘volumes,’ and not séfrin ‘scribes,’ is clear from the colophons to the Theophania in Lee’s ms (B.M. Add. 12150, dated 411 ap). Thus at the end of the Theophania we find ܝܫܠܡ ܠܡܦܬܒ ܚܡ̈ܫܐ ܣܦܪ̈ܝܢ ܕܐܪܘܣܒܣ‎ ie. Lndeth writing the Five Volumes of Husebius.

4 Compare Gibbon ii 382.

89 Description of the MSS.

ascriptions of praise. The 8. Petersburg Codex of Eusebius's Ecclesias- tical History, dated ap 462, has (ed. Wright and M*Lean, p. 412):

ܫܘܒܢܢܐ ܠ ܐܒܐ ܘܠܒܪ̈ܐ ܘܠܪܘܚܐ ܕܩܘܕܼܝܫܐ ܠܥܠܡ ܥܠܡܝܢ ܐܡܝܢ ܘܐܡܝܢ. wis‏ ܗܘ ܐܠܗܐ ܘܡܫܝܚܗ Awa‏ ܠܥܒܕܗ ܒܨܝܪܐ ܢܛܝܐ ܐܝܣܚܩ . . . .

Glory be to the Father and to the Son and to the Holy Spirit for

ever and ever, Amen and Amen.

Blessed be God and His Messiah, who strengthened his servant. the wretched sinner Isaac...... .

But not one of Mr Gwilliam’s Peshitta mss has anything like the doxology in S. Nevertheless I think we should do the scribe a wrong if we looked for the explanation of his language in any of the Greek heresies of the 4th century. It is rather an example of the conservatism of Syriac-speaking Christianity in its earlier stages. We may compare it with the great Creed of Aphraates (On Faith §19), where he speaks of belief in “God, the Lord of all,...Who sent of His Spirit in the Prophets, and then sent His Messiah into the world.” In this Creed, as in the colophon of S, we find the sacred names spoken of as ‘God’ and ‘ His Spirit’ and ‘His Messiah,’ instead of ‘the Father,’ ‘the Son,’ and ‘the Holy Spirit.’ But it is easy to see how such language would give offence after the epoch of Arian controversy. The unconventional terms in which the doxology of Sis couched can hardly be used to prove the formal heresy of the scribe, but they certainly suggest that the ms was written before the Christological disputes had greatly troubled the Syriac-speaking Church. In other words S is earlier than the 5th century.

It is a singular fact that there seems to be no title in S to S. Matthew. The first two leaves are now missing, but they were probably intended as guards and left blank, as is usual in handsome mss. The Gospel of 8. Matthew begins on the verso of the third leaf, the recto of that leaf being entirely blank®. The ecto of the next leaf has ܕܡܬܝ‎ as a headline, so we should expect at least ܐܘܢܓܠܝܘܢ‎ as a headline on the preceding verso, but as a matter of fact nothing can be seen there and no re-agent has brought up any sign of a title. C, on

1 It is possible, of course, that this is a piece of intentional Nestorianism, and that the scribe Isaac really meant to distinguish between the Eternal Son and God’s Messiah.

2 Of course this leaf, which is numbered jo. 82 of the present ms, is now covered on both sides by the later writing.

The title in C. 33

the other hand, has a title. It is unfortunately mutilated by two small holes, and before the term Evangelion da-Mepharreshe had. been well established as the native name of the ‘Old Syriac’ version some controversy had been raised as to the reading of the words and their interpretation.

The title in C, as now preserved, 8

The writing is in vermilion ink which has turned silver through damp. There is no room for a complete letter between the second and third word, but as the first and the second are separated by a dot, it is almost certain that a dot stood also in the gap between the second and third. The dots over the middle of the first and third words are merely placed there for ornament, but the dot over the ܗ=‎ in the second word is unsymmetrical. It is therefore not placed for ornament but for use : we must therefore suppose that it is part of the plural sign (seydné). It is one of the peculiarities of the scribe of @ that he frequently avoids making the seyd@mé dots coalesce with the dot on a 4, ie. he was as likely to have written ܕܡܵܦ. ܪܫܐ‎ as the more regular form ܕܡܦܪ̈ܫܐ‎ 5 I have therefore no hesitation in restoring the title thus :

ܐܘܢܓ ܠܝܘܢ : ܕܡܰܦܪܫܐܐ.] ܡܬܿܝ. Evangelion. da-Mépharréshé. Matthew.‏

The construction is the same as in the colophon of S, where we read “ Ev. da-Mepharreshe, Four Books,” not “ according to the Four.” Before it had been clearly made out that Hvangelion da-Mépharréshé was the name given by Syriac-speaking Christians to the Four Gospels as distinguished from the Diatessaron, it was supposed that ܕܡܦܪܵܫܐ‎ in the title to C had some special reference to ©, Matthew or to the Gospel of Matthew. But although the construction given above is certainly somewhat harsh there can be little doubt that it is correct, now that the Sinai Palimpsest definitely speaks of all four Gospels

under the name of ܐܘܢܓܠܝܘܢ ܕܡܦܪ̈ܫܐ‎

1 Eg. in Matt iii 5 C writes ܝܘܪܡܢܢ‎ waa

34 Description of the MSS.

Note on the line and paragraph divisions in C and S.

On pp. xx, xxi of the Syndics’ Edition a theory of the line and paragraph divisions in Cand S is briefly indicated by Mrs Lewis and Dr Rendel Harris, which, if well grounded, would have an important bearing on the history of the Gospel text in Syriac. I was unconvinced of the truth of the theory at the time, and subsequent investigations do not seem to bear it out: I will therefore very shortly put down the reasons why I consider the line-divisions of Syriac mss have no significance whatever, and why it is extremely improbable that any system of stichometry should have had a Syriac origin.

Mrs Lewis says (p. xx), ‘‘ With regard to the punctuation, it is important to observe (with Mr Harris) that the division into paragraphs in our text and in the Cureton Ms is often identical. Not only so, but in many places there is a line for line agreement.” Twenty passages are then given in illustration. Mrs Lewis goes on to say (p. xxi): “Mr Harris says that the paragraphs are divided into short sentences by stops, which, where they can be traced in the palimpsest writing, agree frequently with the red stops in the Cureton ms. This shows us that the linear and colometric arrangement of the ms from which they were both taken, may be of very high antiquity.”

Here there are three distinct theories indicated. S and C are said to shew a close resemblance in their division of the text (1) into Paragraphs, (2) mto Lines, (3) into Sentences or Clauses. Let us take these three theories in the reverse order.

First, as to the Sentences or Clauses. It was assumed by Dr Rendel Harris that the red dots in C were by the first hand, or that they were at least part of the original contents of that ms. But it has been already shewn that this is not the case and that the dots were added after @ was a bound volume, perhaps many years after the codex had been written. Moreover these red dots occur very frequently, almost at every place where an English comma or higher stop would be set. It was therefore inevitable that their position should frequently agree with the points found in S, or indeed in any other Ms punctuated on a rational system. But they do not so frequently agree in anomalous punctuations ; and imperfectly as the punctuation of S is preserved, there are not wanting instances where S has a point though there is none in (. For instance, in Matt ii 9 S has a point after <uasas, but there is none in C. In Matt ii 13 S has “the angel of the Lorp appeared to Joseph in a dream, and said to him Arise,” but in C there is no stop after “dream” and in compensation there is a stop after “ said to him.” Similarly in Matt 1115 S has “spoken...by the mouth of Isaiah the prophet, who had said From Ligypt I have called my son,” while C has “ said...through the prophet who had said, From Egypt [ have called my son'.”” Other instances may be found in Lk viii 48 ܥܠܘܗܝ.‎ §, Lk ix 82 ܕܥܡܗ.‎ §, Lk ix 61 ܡܪܝ.‎ S, Lk xii 48 ܡܤܓܠܝ ̇ ܡܠܝ‎ 8 In all these places there is no stop in C corresponding to that in S. Yet on the whole there can be little doubt that the existing punctuation in Cis much fuller and more regular than that of S. I conclude therefore that they are practically independent.

± Tn Syriac ܢܒܐ ܕܐܡܪ ܗܘܐ ܡܢ ܡܨܪܝܢ ܩܪܝܬ ܠܒܪܝ܂‎ anew arn... ܕܐܬܡܠܠ‎ & wis\ ܘ ܕܐܬܐܡܪ.. ̣. ܒܝܕ ܢܒܝܐ ܕܐܡܪ ܗܘܐ. ܕܡܢ ܡܨܪܝܢ ܩܪܝܬ‎

Line divisions in Syriac MSS. 35

There are however a few cases where S and C agree in an interesting punctuation, notably in having a stop after ܘܠܫܬܐ‎ , at the end of Joh vii 87. In this they differ from the earlier Latin tradition, represented by the Epistle of the Churches of Vienne and Lyons, by Codex Bezae, by Cod. Palatinus (e) and by the Speculwm p. (700). According to this Latin interpretation, the first words of Joh vii 38 qui credit in me go with bibat at the end of the preceding verse, and the ‘living waters’ flow not from the believer but from Christ Himself. The fact that S and 0 have a stop at the end of ver. 37 shews that the Old Syriac Version rendered the verse as it is rendered in the English Bible and by the great majority of interpreters, ancient and modern. ‘The agreement of S and C in so widespread an interpretation proves very little as to the origin of their punctuation.

As to the second point, the alleged agreement of S and C in their division into lines, it might seem suthicient to call to mind the great textual differences between these two mss. Sand C differ on matters of such fundamental importance, that it is difficult to believe that there can be any significance in their occasional agreement in the division of the text into lines. But the matter does not rest upon general probabilities. We have to consider the nature of Syriac script and its important differences from Greek script. Greek mss of the 4th and 5th centuries were written with no spaces between the words, and no scruple was felt about dividing the words at the end of a line. Thus the number of words in a line varied greatly, but the number of syllables was approximately constant, e.g. cod. & averages about six syllables in a line, cod. B about seven or eight. In a pair of such Mss it would be exceedingly improbable that the lines should coincide. There are so many ways of lawfully dividing most Greek words that marked examples of coincidence for several lines together could not be regarded as the result of a mere accident. But Syriac writing is quite different. The words are separated by spaces, as in modern English, and they are never divided at the end of a line. Consequently the average number of syllables in a short Syriac line varies very greatly, but the average number of words is constant. Now S and C, like all other known early Syriac Mss, are written in short lines, with only three words on the average to a line. Whenever there- fore they start level, as at the beginning of a paragraph, there is every probability that they will go on agreeing line for line, unless an actual variation in text occurs, such as the omission of one or two words. For example, to take the first passage on Mrs Lewis’s list, it is true that both S and C give Matt ¥ 3 thus :—

ܒܪܘܚܗܘܢ ܕܕܝܠܗܘܢ ܗܝ

ܡܠܦܘܬܐ ܕܫܡܝܐ Lo:‏

But how otherwise should the lines be divided? If the scribe did not end the first line at ܠܡܣܵܦܢ̈ܐ‎ he would have to get ~amvoi into the space. And indeed, as if to shew that this line-division in Syriac Mss is a matter of absolute indifference, we find the very same division of this verse in such codices of the Peshitta as B.M. Add. 17117, although naturally B.M. Add. 17117 reads ܒܘܚ‎ instead of ܒܪܘܚܗܘܢ‎ , Moreover, when once the line-agreement of S and C is disturbed, it remains disturbed, e.g. in Matt ¥ 10 we

have

36 Description of the MSS.

ܠ C‏ ܛܘܒ̈ܝܢܗܘܢ ܠ ܐܝܥܠܝܢ ܛܘܒܝܗܘܢ ܠ ܐܝܠܝܢ ܕܪܕܝܦܝܢ ܕܪ̈ܕܝܦܝܢ ܡܛܠ ܙܕܝܩܘܬܐ ܡܛܠ hoon‏ ܕܕܝܠܗܘܢ ܕܕܝܠܗܘܢ ܗܝ ܡܠܦܘܬܐ ܗܝ ܡܠܦܘܬܐ ܕܫܡܝܐ ̇ܘܼܿ. ܕܫܡܝܐ ܘ .

The fact that S and C often take exactly the same number of lines to get through three or four verses is thus merely another way of stating that the lines in each ms are of much the same breadth, but the nature of Syriac script robs this circumstance of the significance which it might have had in a pair of Greek mss. I have therefore not thought it worth while in this edition to keep any record of the line-divisions either of S or of ¢

Had the occasional agreement of S and @ in their division of the text into lines possessed the significance suggested by Mrs Lewis, we should have been compelled to regard SN and C as direct descendants of the same exemplar. Besides this, it would have been reasonable to assume that these agreements were intentional and that they had something to do with an early system of Syriac stichometry. Dr Harris had found at Sinai a 9th cent. ms of miscellaneous contents which contained among other reckonings an enumeration of the number of <S:\ he in the Four Gospels. Now there are two systems of reckoning found in various Greek mss of the Gospel; we have enumerations sometimes of the or/xor, sometimes of the pyyara, sometimes of both. A ܘܐܗ‎ is a line of a given length, or rather of a given average number of syllables, e.g. a half-hexameter. If therefore the number of e7éyou in a work is known, und the number of lines in « column be constunt, it is easy by counting the number of columns at once to ascertain whether a Ms is approximately complete. In fact, the use of the numeration of o7ixor (Lat. wersus) is to guard the buyers of books against fraud. The meaning of fyyara in enumerations is not so clear. The word ܧܦܬܠ ܡܐ‎ may stand either for oréyos or pyya, and Dr Harris brought forward a theory, which found many supporters, that in the case before us ܦܬܓܡ̈ܐ‎ was a translation of ܕܬ )ܐܗ‎ and pyyara a retranslation of ܦܬܓܡ̈ܐ‎ In other words, the Greek enumeration of oriyou had been adapted to the Syriac version and the numbers so transformed had been retranslated into Greek under the name of pypara. An additional confirmation of the theory seemed to come from the presence of certain readings which agreed with the Old Syriac in the ‘Ferrar group’ of Greek mss, and the ‘Ferrar group’ were among the mss that contained the reckoning of pypata!,

Attractive as this theory appears at first sight, it breaks down under close examin- ation from the Syriac side. If the pyyara system had a Syriac origin it must obviously have been a system intended for the ‘Old Syriac,’ ie. the Hvangelion du-Mepharreshe. The coincidences noted by Dr Harris between the Ferrar group and Syriac readings were with readings of the Leangelion da-Mepharreshe or of Tatian’s Diatessaron, not of the Peshitta. But the Syriac mss which contain the ܧܦܬܓܡ̈̈ܐ‎ are all of them late—the earliest witness is not anterior to the 9th century—and in at least two cases these lists contain figures for the full Greek Canon of the seven Catholic Epistles, a circumstance

1 J. R. Harris, On the origin of the Ferrar Group (1893), pp. 9, 17 fh

Paragraph divisions in S and 4 37

which suggests a Greek origin for the whole system’. Moreover early Syriac mss, and especially S and C, are singularly ill adapted to form the basis of stichometrical caleula- tions. Neither in S nor in @ are horizontal lines ruled for writing, so that the number of lines in a column, even in parallel columns on the same page, varies considerably, and to ascertain the number of lines occupied by either of the Gospels in ¢ or Cit would be necessary to inspect each page of the ms.

The fyara system appears in some Latin mss of the Vulgate about the same time as the earliest evidence for it in Syriac: the same reckonings seem to underlie the famous 4th century Cheltenham List, so that pyyata might equally well bea translation of wersus as of ܧܦܬܓܡ̈ܐ‎ , But whatever be the origin of these fyyara there is no tangible evidence to connect them with the Hvungelion da-Mephurreshe ; ov the contrary, there are many indications that the scribes of S and C were not familiar either with the pypara or any other system of stichometry.

With regard to the third point raised by Mrs Lewis, viz. the division of S and C into paragraphs, the evidence is not as clear as might be wished. ‘here would be every probability that the paragraph system in a pair of mss such as S and C’ should go back, in part at least, to a common origin. But here again we are hampered by the ambiguity of Syriac writing. Both in S and in Ca fresh paragraph starts at the beginning of a line. The preceding line will in many cases be short, and the end of the paragraph is thus clearly marked either by the blank space at the end of the short line or bya stop. When the end of a paragraph happened to occur at the end of a full line it seems to have been the general intention of the scribes to leave a blank space, e.g. after Matt xiii 43S. But this blank space is not always left. When therefore the end of a paragraph comes at the end of a line it is often impossible to determine whether the next line was intended to begin a new paragraph or not. ‘This happens very frequently, for on the average there are only three words to a line. In something like one case out of ten therefore the same word will end a line both in /¥ and in C, whether a paragraph be intended to end there or not. And if a word is known to end the line in the one ms, the chances are one to two that it will also end the line in the other. This being the case it needs clear instances of agreement in quite peculiar divisions to prove that the paragraphs in S and C belong to the same system.

Now as a matter of fact S and C very often do not agree in their paragraph divisions. If we take from Matt i1 to vi 10 (where S breaks off owing to the loss of a couple of leaves) we find that paragraphs clearly end in C, but not in S, before

Matt ii 14 (Now Joseph arose...) 16 (Then Herod, when he saw...) 22 (Now when Joseph heard...) iii 1 (And in those days came John the Baptist...) iv 17 (From then Jesus began to preach...) 21 (And when he removed thence...) v 25 (Be agreed with thy adversary...) 48 (Become therefore perfect...)

1 See the Note by the present writer on the Gospel pyyara found in Syriac mss (Journ. of Theol. Studies, ii 429 tf).

38 Description of the MSS.

On the other hand paragraphs end in S, but not in C, before Matt iv 1 (‘Then Jesus was led by the Spirit...) 25 (And when there was a great multitude...) ¥+ 31 (It hath been said ‘He that leaveth his wife....’) 43 (Ye have heard that it hath been said...) vi 9” (Our Father in heaven...)

It will hardly be denied that here we have considerable variation. It is indeed quite surprising that S should not make Matt iii 1 begin a paragraph and that C should not make Matt iv 1 begin a paragraph’. Much the same state of things is to be found in other parts of the Gospels. For instance Lk xii 32 begins a paragraph in C, but not in S; on the other hand, Joh viii 12 begins in Con the same line as the last words of vii 52, but in S there is a fresh paragraph. ‘These grave divergences suggest that the systems of paragraph division in S and C may have been developed quite independently*

In conclusion I will quote some words of Mr G. H. Gwilliam upon the paragraph divisions in Mss of the Peshitta, which have a real bearing upon the question before us. Mr Gwilliam says “evidence of the independence of these mss [of the Peshitta] appears in the different arrangement of the paragraphs of the sacred text in the different copies. In some the paragraphs are numerous; in others few, and differently placed. For example, in S. Matt x the Cod. 14459 makes a break in our Lord’s discourse at the end of verse 10, and seems to stand alone in so doing. In 0. xi it makes its division at the end of verse 1, thus not so distinctly connecting the message of the Baptist with the preaching of Christ recorded in verse 1, as do other authorities. And similarly in other Mss divisions are constantly made, more or less arbitrarily, according to the fashion of some scribe or school” (Studiw Biblicu 1 166). “ Besides the formal sections, the text is often interrupted in the best mss, where the sense requires a break; and these paragraphs are often made with much judgement” (Studia Biblicu 1181). If even the mss of the Peshitta Gospels shew independence in their paragraph divisions we shall not be surprised if the mss of the Hvangelion da-Mepharreshe exhibit wide divergence.

1 It is perhaps worth notice that the Latin Cod. Bobiensis (4) has no fresh paragraph at Matt iii 1, although like S it has a fresh paragraph at the Puter noster. But elsewhere S and & do not exhibit much in common in their system of text division.

2 The chief instance of marked agreement between S and @ in this matter is the division of

each Beatitude in Matt ¥ into a separate paragraph. Each parable in Matt xiii is similarly divided.

ܫܡܝܐ ܕܢܐܟܘܠ ܐܐ ܡܢܗܟ ܘܢܡܘܬ ܐܢܝܐܕܐܢܐ 7 39

ܕܐܐ ܐܙܬܠ ans‏ 18

ܬܘ ܠܨܝܢ ܗܘܘ ܝܗܘܕܝܐ

soaks‏ ; < ܐܡܝܐ

ܐܡܪܝܢ A 00d)‏ ܗܘܐ ܗܢܐ ܝܩܠ

Copex © fol. 49x (Joh vi 41--68(.

e with a small black dot at the end of a paragraph, but no other punctuation.

0 x

i

eat 00

A aN eae

1 Sak RS nar AN ad PN had ste

4

Copex C, fol. 48v (Joh vi 30—41).

A page with a headline and corrections, but no punctuation dots

CHAPTER IT.

GRAMMAR AND SYNTAX.

GRAMMATICAL PECULIARITIES OF S AND C.

Ir was long the fate of the ‘Curetonian’ to be praised for its defects and slighted for its merits, and in no point was this treatment more marked than in the matter of its grammatical features. The defenders of its antiquity too often based their case upon points of grammar and spelling which C’ shares with ancient Peshitta mss, while many of the real peculiarities have remained hardly noticed to this day. Even more absurd was the contention that the ‘Curetonian’ was older than the Peshitta on the ground that the rougher and less polished version, as the ‘Curetonian’ was declared to be, must on this ground have preceded the more regular and grammatical. It is true that the Peshitta follows Greek idiom much more closely than its rival, though without the pedantic servility of the Harclean ; and this circumstance undoubtedly renders it extremely unlikely that it should be older than the text represented by Cureton’s ms. But it will be well at the outset clearly to state that neither the text of the Curetonian nor that of the Sinai Palimpsest can be described as in any degree barbarous or uncultured. Apart from a few corruptions in the text, such as might befal mss of Demosthenes or Cicero, both S and C are written in the most idiomatic Syriac. Our two mss are themselves very ancient, and the text which they present is doubtless much more ancient still ; hence the transmitted text of the Evangelion da-Mepharreshe is full of peculiarities of grammar and spelling which are hardly to be met elsewhere in Syriac literature, or are found only in the oldest and best preserved works. But there is no question of dialectical variety or of rustic idiom. On the contrary, every indication shews that the translator of the Evangelion da- Mepharreshe used the vernacular Syriac of Edessa with the simplicity

40 Grammar and Syntax.

and ease which come only from literary training, coupled (as one is tempted to add) with the prevalence of a sound standard of taste. Few prose translations surpass the Hvangelion da-Mepharreshe in literary excellence.

In this chapter I have followed the order of the sections of Noldeke’s Syriac Grammar, noting the more important peculiarities under each section. In transliterating I have used Greek letters for the

. ܒ ܓ ܕ ܟ ܦ ܬ aspirated sounds of‏

Spelling.

Sand Care on the whole regular in spelling apart from words where the form is really different from that generally found.

[Noldeke §4.]| ܟܘܠ‎ and ܡܛܘܠ‎ occur frequently for ܟܠ‎ and ܡܛܠ‎ especially when there is plenty of space in the line. On the other hand S often writes ܠܩܕܡ‎ , (aol, ܠܩܒܠܗ‎ (Matt xxvii 61), at the end of a line ܐܘܪܚܐ‎ is three times written ܐܪܺܚܝܝܐ‎ (Mk x 32, Lk xix 36, Joh xiv 6)', and we even find ܬܬܒܘܢ‎ for ܬܬܘܒܘܢ‎ in Lk xiii 3. In this point C is more regular, though ܐܪܺܫܠܶܡ‎ is found for ܐܘܪܺܫܠܡ‎ in C, e.g. Lk xiii 4 (at the end of a line). The defective spelling ܘܐܫܛܘ‎ Matt xxii 19 S (sve, for (ܘܐܘܫܛܘ‎ must be a mere slip of the pen?.

[ Noldeke §35.| Radical ܐ‎ is not unfrequently dropped in S, and sometimes also in C, when the letter is practically unpronounced. Thus we find ܦ ܦܝܐ‎ in Matt xxiv 2 S, Mk xiii 2 S, Lk viii 51 S, Lk xii 41 C ete., and in Seven Aq for ܢܐܙܠ‎ in Lk [ii 15, and | ix 51. In Lk xiv 16 C, Joh vii 31 C, Joh xviii 2 S ܣ̈ܓܝܐ‎ occurs for ܣ̈ܓܝܐܐ‎ at the end of a line. But the insertion of ܐܙ‎ in such words as ܬܐܥܠܘܢ‎ , Lols, never occurs either in S or C, with the single exception of ܡܓܐܪ | ܓܐܪ‎ Lk xvi 18 S, where the intrusive ܐ‎ in the first word is perhaps only due to a slip? The curious form ܡܝܝܐܘܬܐ‎ 04 in Mk iii 11, Lk x 34S, is noticed below on Néldeke $77.

1 yAed occurs Matt xx 30 in Mr Gwilliam’s cod. 36.

2 The same ix probably true of ܘܢܥܠ‎ for ܘܢܥܘܠ‎ in Joh x 98. It is noteworthy that many of the words from which a ܘ‎ is dropped in S are words in which another a occurs. Misspellings of another kind are ܦܘܬܘܢܝܢ‎ Mk vi9 Sand ܥܘܠܘܒܘܬ̇ܕ‎ Mk vii 93 8.

5 This incorrect insertion of ܐܡ‎ is not much found before the 9th century and then only in Jacobite mss, e.g. Cod. Crawfordianus 2 of the 12th cent. Along with many other late and in- correct forms it found its way into early editions of the Peshitta.

Spelling. 41

[Noldeke § 50 B.] The unpronounced a and ܝ‎ at the ends of words are very rarely left out in Sand C. In Joh iv7 Chas A om, but the correct spelling A 55 is found in v. 10 and in S?/,. The majority of the mss of syr.vg have ܠܝ‎ sm: probably the phrase was regarded as a single word. In Matt xiv 16 ܗܒܘ ܠܗܘܢ‎ is correctly written in all the mss.

[Noldeke §51.] Prosthetic ܐ‎ before r is common both in S and in C, eg. in Matt i 5 ‘Ruth’ is spelt ܐܪܶܥܘܬ‎ in C, while S has hast. In Joh xx 12 the curious phrase for ‘at the foot-place’ is written ܡܢ ܐܕ̈ܓܠܐ‎ in S, ie. min arydlé. In the Peal of ܪܰܟܒ‎ ‘to ride’ we find ܘܐܪܟܝܒ.‎ Matt xxi 5 Cand ,wals ܘܐܪܟܒ.‎ Mk xi7 8. This use of prosthetic ܐ‎ even after a explains the spelling of ܘܐܪܥܘܬܐ‎ Lk 1 14S: ܐܪܥܘܬܐ‎ (=ctdoxia) was probably so spelt in this passage in order more effectually to distinguish it from Whas4, 1. .ܘ‎ × 4 ‘shepherds,’ a word which occurs four times in the immediate context}.

It may be mentioned here that S has ܐܪܶܩܘܥܬܐ‎ in Matt ix 16 and ܪܩܘܥܬܐ‎ in Mk i 21 for the word which is usually spelt hasta ‘a patch.’ In each case S is supported by Gwilliam’s cod. 23 and some other evidence. C' is not extant in either place.

In Matt vii 13 C has esata (hiat S), but Mr Gwilliam has adopted ܘܐܪܘܝܚܐ‎ on excellent authority.

Pronouns.

[Noldeke § 63.] assarv is the form commonly used both in S and in C. ti» only seems to occur twice in S, viz. Lk vii 20, xix 14, but it is found more often in C) e.g. Matt xix 27, xx 18, xxi 16; Lk xi 4.

[Néldeke §64.] The contracted forms of the Ist pers. sing. mase. such as atm’, Miss, are more common both in S and in C’ than the uncontracted forms ܒܥܐ ܐܢܐ , ܐܡܪ ܐܢܐ‎ . In the fem. we find ܝܕܥܢܐ‎ Joh xi 22 S, Joh xx 14 S, but ܝܕܥܐ ܐܢܐ‎ Joh xi 24 S: either form is of course pronounced ydd‘dnd?, while the masc. is yddsaind. ܡܗܝܡܢܐ ܐܢܐ‎ occurs Joh xi 27 8.

1 The pronunciation of hana I suppose to have been ard: comp. my) Ezr ¥ 17, vii 18, and in Syriac haze. 2 So also hw amd A* 365, agreeing with S, but A? has ܪܘܒܥܬܐ‎ 3 Mr Gwilliam edits ܝܕܥܐܐ ܠܐܡ‎ in Joh xi 24, wet efx, in Joh xx 14, B. Il. 6

42 Grammar and Syntax.

In the 2nd sing. mase. the forms with ܐܢܬ‎ written separately are almost universal. As exceptions I have noted ܝܕܥܬ‎ Matt xv 12 6 ܨܒܝܬ‎ Matt xix 21 S (middle of line), ܐܡܪܬ‎ Lk xxii 60 S, ܚܟܡܬ‎ Lk xxiv 18S, ܒܥܝܬ‎ Joh iv 27 S ܒܥܬ)‎ C), ܫܐܠܬ‎ Joh xi 22S; ܚܙܬ‎ in Lk vii 44 S appears to be a scribe’s blunder for de, similar to that made by Cin Joh iv 27. In C ܩܪܺܝܬ‎ occurs Lk x 26 and ܚܝܒܬ‎ Lk xvi 5. But in the 2nd sing. fem. the rule is reversed: S never has the separated form and @ only twice. We find ܕܨܒܝܬܝ‎ Matt xv 28 S ܨܒܝܬܝ ,)0 ܕܨܒܝܐ ܐܢܬܝ)‎ Matt xx 21 S (due ܨܒܝܐ‎ 0 sic); ܐܡܪܶܬܝ‎ Matt xxvi 70 S, Mk xiv 68 S: ܫܪܺܝܬܝ‎ (pass. part.) Lk xiii 12 ® C and Pesh., ܝܕܥܬܝ‎ Lk xix 41 SC and Pesh. In Matt xxiii 87 S, Lk xii 34 S C we find ܕܩܛܠܬ ܝ . . ܘܪܶܓܡܬܝ‎ but C has aliios, just as ܐܢܬ‎ is written in C for ܐܢܬܝ‎ in Matt xi 23, xx 21, an error which occurs again twice in Joh iv 10 C ܗܘܝܬ)‎ WAcz...duam ean). It is not unlikely that ܝܨ ܦܬܝ‎ in Lk x 41 Cis a corruption of ܝܝ ܦܬ ܝ‎ Le. ydspat ‘thou ) 71 ( art anxious.’

The forms ܗܘ‎ am and asam are about equally common. am is never written enclitically after participles.

[Noéldeke §66.] The irregular forms of the verbal suffixes will be found under the verbs.

[ Néldeke §67.] ܗܢܐ ܗܘ‎ and ܗܢܘ‎ are about equally common. In C am hau sometimes has the ordinary point, e.g. ܝܘܡܐ‎ ams Matt ¥ 12, aco ܡܢ‎ LE xvii 14.

The rare and ancient form wala is found in Matt xv 22 ¢ xx 9 C, xxi 40 C, xxii 7 C; Lk vill 13 S C, xii 37 C; Joh iv 38, 43 C (lat ©. Except in Lk vii 13, S where extant has in each case ܗܢܘܢ‎ for ܗܠܘ‎ It appears to be used with no distinction of meaning and ܗܢܘܢ‎ occurs in @ side by side with ,ܗܠܘܟ‎ e.g. 1 Matt xx 10. The fem. ܗܠܝܟ‎ occurs Matt xv 24 CS.

[Noldeke §68.] ܡܢܘ‎ is frequently written for ܡܢ‎ and am ܡܢ‎ and also for ܡܢܐ ܗܘ‎ both in Sand ¢ ܡܢ ܗܝ‎ is written » in S where- ever it occurs, vz. Matt xii 48, Mk 111 33, Lk vii 39: this spelling is apparently unknown elsewhere in Syriac?

1 931 and 59 are both found in the Babylonian Talmud, as in the story of Rabbi Chasda (Shabb. 140b), who said to his daughters, ‘‘When any one knocks at the door 5 nds ܡܙ‎ won xd, ie. do not say ‘Who is there?’ as if to a man, but as if to a woman.”

Plural Forms of Nouns. 43

Nouns.

[Néldeke §71, 4.] The fem. pl. emph. of ܗܓܝܐܐ‎ is generally spelt ܣܗܓܝܐ̈ܬܐ‎ in S as in all other Syriac mss. But in Lk xxii 65 the word is very distinctly written ܣܘܓܝܐܝܬܐ‎ 1, This might represent a pro- nunciation saggtydéa (instead of saggt’d0d), but it is probably a mere ship of the pen.

[Noldeke §74.] The plural of ܫܠܝܛܐ‎ is uniformly ܫܠܝ̈ܛܢܐ‎ when used as a substantive, including Lk xii 11 S. But as a predicate we find ܫܠܝܛܝܢ‎ in Matt xx 25 C (haat S) and in Lk xx 25 S: in the last passage Chas even ܫ̈ܠܝܛܐ‎ . In Mk iii 15 S has ܡܫܠܛܝܢ‎

[ Noldeke §77.] The plural of ܡܢܝܘܬܐ‎ ‘wound’ should be ܡܚܚ̈ܘܬܐ‎ ie. mahwd0d, as in Lk x 84 C, vi 21 SC, xii 48 SC. But S in Mk iii 11, Lk x 34, has ܡܚ ܐ̈ܘܬܐ‎ , ie. m’hawdd, like s’lawd6d from $7004 ‘prayer.’ That it was not treated as masc. appears from the context of Lk x 34S, where we read... pols ira ܒ ܡܚ ܐܘ̈ܬܗ‎ The grammatical interest of this form, thus doubly attested for S, is that it throws some doubt upon the complete sufficiency of the grammatical tradition of the Syrians. Ancient Syriac Mss are not vocalised, and the grammatical traditions upon which our Syriac grammars are based is essentially a series of directions for the pronunciation of the Bible in Church. In the same way the Massoretic vocalisation of the Hebrew text gives the Synagogue chant. But these official schemes do not allow for all varieties of pronunciation and they have a tendency to reduce what was varied and fluid in the living language to the uniformity of a cast-iron pattern.

Many of the niceties of Syriac grammar depend on the vocalisation alone, and divergent forms may be hid under the regular consonantal spelling of ancient MSS. Thus in the present case ܡܚܘ̈ܬܐ‎ may repre- sent equally well mahwdéd and mhawdéd. But ܡܢܝܐ̈ܙܘܬܐ‎ cannot stand for mahwdéd ; there must be with this spelling a vowel between / and w. But if the scribe of S spelt the word ܡܥܚܐ̈ܘܬܐ‎ in Mk iii 11 and Lk x 34, he probably pronounced ܡܢܚܝܘ̈ܬܐ‎ as mhawdbd in Lk xii 48

1 See facsimile to the Syndics’ Edition, last line of col. a.

44 Grammar and Syntax.

and elsewhere. It would be interesting to trace the usage of this word and others of the same sort in the earlier poets.

[Noldeke §79 B.| The plural of ܠܒܐ‎ in S Cis whaal, except in Lk xxi 13 (14) where both mss have ܒܠܒܝܟܘܢ‎

[Noldeke § 87.1 ܡܠܬܐ‎ ‘ word’ is always fem. in the Hvangelion da- Mepharreshe. By a usage which derived its authority from Theology rather than Grammar, the Peshitta always makes ܡܠܬܐ‎ masc. when it means ‘the Word’ in the Johannine sense. SS is unfortunately missing for the opening verses of the Fourth Gospel, but in C, supported by Aphraates and by Ephraim, we read ܘܐܓܢܬ ܒܢ‎ ham ܡܠܬܐ ܦܓܪܐ‎ ‘the Word became a body and She sojourned with us®.’ So also in the Acts of Thomas 241 we find ܡܠܬܐ ܡܪܬ ܟܠ‎ ° the Word, Mistress of all.’

‘spirit’ is also fem. in S and C. In the Peshitta it is‏ ܪܘܚܝܐ frequently treated as masc. when it means the Holy Spirit. But‏ the Holy Spirit is fem. in S and C: even in Joh xiv we find‏ vat ‘the Spirit, the Paraclete...She‏ ܕܝܢ ܦܪܩܠܛܐ . . . ܗܝ ܬܠܦܟܘܢ will teach you?” That this usage was not wholly a matter of‏ grammatical inflexion, but had also an influence in theological teaching‏ is clear from Aphraates 354, where that ancient writer asserts that the‏ Father and Mother whom a man leaves when he marries a wife are God‏ and the Holy Spirit.‏

In one passage, ܪܘܚܐ ܛܢܦܐ‎ acl Lk viii 29 S C, an unclean spirit is treated as masc., but probably in this passage sat is a correction for ܫܐܕܐ‎ ‘devil’: 7@ Samoviw instead of ܐ‎ mvevpate is read in D and in e of the Old Latin‘.

The word ܫܡܝܐ‎ is used in Syriac both as sing. and plur. for ‘heaven’ and ‘heavens,’ and in the sing. it is sometimes masc. and

1 Alone of course it would be in any case a word of three syllables. But the spelling of ¥ would be greatly supported if an instance of ܕܡ ܐܢܘܬܐܕ‎ or <haiisa\ could be found where the verse required four syllables. Possibly the irregular spelling indicates that the third radical still had some consonantal force. The root is ND for ,/ YND, corresponding to as. and YN.

? For the exact meaning of ܐܓܢܬ ܒܢ‎ see the Note on the passage. 3 Cis here partly defective but ܗܝ ܬܠܦܟܦܘܢ‎ is clear. ± In Joh iii 6 C adds at the end of the verse ܡܛܘܠ ܕܐܠܗܐ ܪܘܚܐ ܗܘ . ܘܡܢ ܐܠܗܐ .ܐܬܝܠܕ.‎

But § has only ܢܐܐ‎ am ܡܛܠ ܕܐܠܗܐ ܪܘܢܚܵܐ‎ ‘because God is a living (m.) Spirit.’ The word way is unfortunately not very clear, but I think it is correctly read, otherwise we might conjecture aia, In any case it is a sentence where a feminine adjective would be very harsh.

‘Word, ‘Spirit, ‘Heaven.’ 45

sometimes fem. In S and C ܫܡܝܐ‎ is always 11886. pl. when it is the subject of a verb. The ‘ heavens’ are opened (Matt 111 16, Lk in 24) or are shut (Lk iv 24), and ‘ they’ are the throne of God (Matt v 34): in the last instance syr.vg has the sing. masc. But in other cases ܫܡܝܐ‎ is sing. masce. in S and sing. fem. in C. Thus the phrase “from one end of the world to the other” is rendered

Mk xii 27 S‏ ܡܢ ܪܫܗ ܕܐܪܥܐ ܥܕܡܐ ܠܪܶܫܗ ܕܫܡܝܐ Lk xvii 24 8‏ ܡܢ ܪܫܘܗܝ ܕܫܡܝܐ ܘܥܕܡܐ ܠܪܫܗ

Here as elsewhere S does not insert the diacritic point which distin- guishes résheh (m.) from réshdh (f.), but the suffix of ܪܫܘܗܝ‎ in the second passage shews that ܫܡܝܐ‎ is treated as masc. In Mk C 8 missing, but we find

Lk xvii 24C‏ ܡܢ ܪܫܐ ܕܫܡܝܐ ܘܥܕܡܐ ܠܗܘ̈ܦܝܗ

in which ܫܡܢܐ‎ is sing. fem. In Matt xi 25 ® C and syr.vg (with syr.vg also in Lk x 21) ‘Lord of heaven and earth’ is rendered ܕܫܡܝܐ ܘܕܐܪܥܐ‎ WI, whereby the gender of ܫܡܝܐ‎ is not indicated ; in Lk x 21 S and C* have ܕܫ ܘܕܐ‎ atm, but a corrector of C has added a dot in red over the om, thereby making ܫܡܝܐ‎ fem.

Elsewhere the gender and number of ܫܡܝܐ‎ is carefully left indeterminate. The ‘kingdom of God’ is sometimes ܡܠܟܘܬܐ ܕܐܠܗܐ‎ sometimes ܡܠܟܘܬܗ ܕܐܠܗܐ‎ , but the ‘kingdom of Heaven’ is always ܡܠܟܘܬܐ ܕܫܡܝܐ‎ , except in the few instances where the construct state is used, an idiom which also leaves the gender and number of ܫܡܝܐ‎ doubtful? Neither ܡܠܟܘܬܗ ܕܫܡܝܐ‎ nor ܡܠܟܘܬܗܘܢ ܕܫܡܝܐ‎ ever occurs in Syriac literature, so far as I know.

Similarly the Birds of Heaven are ܦܪ̈ܚܬܐ ܕܫܡܝܐܙ‎ Matt vi 26, viii 20, etc.; the Clouds of Heaven are ܥܠܥ̈ܝ ܫܡܝܐ‎ Matt xxiv 30 S, xxvi 64 S, or ܕܫܡܝܐ‎ iis Mk xiv 898 ; the Angels of Heaven are ܡܠܐ̈ܟܐ ܕܫܡܝܐ‎ , or ܡܠܐܟܝ ܫܡܝܐ‎ Matt xxiv 36, or ܥܝܪ̈ܝ ܫܡܝܐ‎ Lk xv 10 A; and the Powers of Heaven are ܚܝܠܘܬܐ ܕܫܡܝܐ‎ Matt xxiv 29 8, Mk xiii 25 S (sic)3, or ܚܺܝܠܝܐ ܕܫܡܝܐ‎ Lk xxi 268 €

It is worthy of note that the construction of ܫܡܝܐ‎ is the same for

1 ܪܝܫܗ‎ (so I read the photograph) : S* has v<~z5, In this verse syr.vg makes ܝܡܐܪ‎ fem. sing.

2 .ܘܢܝܐ‎ hhaslss occurs Matt vii 21 ¢ xiii 52,8, xviii 3 4221, xix 23 ¢ A399.

3 Not ܚܝܠܘܬܗ‎ as S. Syr.vg has ܙܰܢܺܝܠ ܐܕ‎ in Matt and Lk, hola in Mk.

46 Grammar and Syntax.

rendering ovpavds and ovpavoit. Thus in the Greek of Matt ili 16 the word is plural; in the parallel passage Lk 111 21, in Lk iv 24, and in Matt v 34, the word is used in the singular.

The word ܣܘܟܬܐ‎ ‘branch,’ pl. Made, is not marked in Néldeke § 87 as of common gender. But both in Matt xxiv 32 and in Mk xiii 28 Shas ܣܘ̈ܟܝܗ‎ asia ܡܐ‎ ‘when its branches have become tender.’ C unfortunately is not extant for either passage, and in Matt xxiv 32 syr.vg has wi, Le. fem. pl. This is also the reading adopted by

Mr Gwilliam in Mk xiii 28, but the Nestorian Massora (Mas. 1), with one of the Jacobite Massoretic codices (Mas. 4) and three ancient mss (7 11 21) all read aad as in S (src), thereby making ܣܘܦܬܐ‎ mase. in the plural.

[Néldeke §98 c.] ܟܝܢ‎ occurs Matt i 19 S and ܟܐܢ‎ Joh ¥ 30C. ܟܢܘܬܐ‎ and chairs are about equally common, as is usually the case in ancient Syriac MSS.

[Noldeke §100.] For ,ܡܚܝܐܘܬܐ‎ see on Noldeke §77.

[Néldeke §103.] ܡܢ ܪܘܚܩ‎ occurs Mk vii 3 S, Lk xvi 23 8. The absolute and construct of ܪܘܚܩܐ‎ are otherwise unknown. On the analogy of star and star, we may suppose the pronunciation to

have been ruhag. [Noldeke §117.] sats Matt viii 12 S must be a mere scribal error

for pitas, like ܣܥܘܪܢܟܝ‎ Lk xix 44 S for ܣܘܥܪܙܟܥܝ‎ .

[Néldeke § 127*.] For the formation of ܐܪܥܘܬܐ‎ (=evdoxia) Lk ii 14 S, see on Noldeke § 51.

[ Néldeke §128 a.] sastasmo Lk xix 44 S must be a scribal error for ܣܘܥܪܢܟܝ‎ . The very peculiar rendering »ahast for rhs ܡܗ )ܬ*¢‎ $ gov is attested by 0 and Aphraates 412, and it is possible that the misspelling in S may come from an unskilful correction in its exemplar.

[Noldeke § 135.) The words for ‘Galilee’ and ‘Galilaean’ are irregular in S. In C, as in other Syriac documents, ܓܠܝܠܐ‎ is ‘ Galilee,’

‘a man of Galilee’ But in Matt xxvi 69, Mk xiv 70, Lk xxii 59, S has ܓܠܝܠܐ‎ for ‘Galilaean.’ The plural is spelt willy. in Lk xii 1, 2, and also in Lk xvii 11 apparently. To make the irregularity yet more conspicuous S has ܓܠܝܠܝܐ‎ in Lk xxiii 5, 6, for

` Galilee.’ [Noldeke § 145.] The possessive suffixes to Nouns offer in S no

peculiarities, but the spelling ,asssa\h occurs in C twice (Matt xi 2,

‘Abba, Father, 47

Joh iv 8) for ܬܠܡ̈ܝܕܘܗܝ‎ This spelling is also found in most varieties of Palestinian Aramaic, so that possibly it was once not uncommon in Edessene.

[| Noldeke § 146 : the anomalous Nouns. |

is usual both‏ ܐܒܝ ordinary form for ‘my father,’ viz.‏ ܘ{ ܐܒܐ in Sand C. But in Matt x 32 C,xv 13 S, Lkii49 C, Joh vi 32 C, we‏ although no Greek ms omits pov. In Matt vi 4 S also we‏ , ܐܒܐ find‏ It is true‏ ,ܐܒܘܟ where C in agreement with the Greek has‏ , ܐܒܐ find‏

that in none of the cases are both S and C agreed, but on the other hand the same interchange of ܐܒܐ‎ and ܐܒܝ‎ is occasionally found in the Mss of syr.vg. I have noted ܐܒܐ‎ for ܐܒܝ‎ Joh vi 32 (cod. 14), x 17 (cod. 9), xii 27 (cod. 40), xii 50 (codd. 3 4 14 17 28 40), xiv 26 (cod. 3*), xvi 17 (cod. 12), xvii 25 (cod. 9(. ܐܒܝ‎ for ܐܒܐ‎ only occurs Joh xii 49 (cod. 8), xiv 12 (codd. 4 9 23 36 37). In Joh vi 32 C there- fore has some outside support, and in Joh xii 27 ܐܒܐ‎ has the support of S and the Greek. It is, I venture to think, not unlikely that ܐܒܐ‎ i.e. Abba, was once used for ‘my father’ in Edessene, as in most forms of Palestinian Aramaic, and that these variations are the last trace of a vanishing idiom’. In Mk xiv 36 for ABBa 6 warip we find ܐܒܝ‎ in S, Le. ‘my Father,’ without any addition; syr.vg has ܐܒܐ ܐܒܝ‎ ie. ‘Abba, my Father,’ in agreement with the Greek.

‘hand’‏ ܐܝܕܐ in all the Biblical translations the plural of‏ 48 .ܐܝܕܐ It only occurs in the phrase ‘to lay hands on’ a‏ ,ܐܝܕܐ not‏ , ܐܝ̈ܕܝܐ is‏ person ^.‏

1 In Christian Palestinian Aramaic, as in Jewish Aramaic, <>< means ‘my father’ even where the emphasis is on the suffix, e.g. Joh xx 17 ‘unto my Father and your Father’ is rendered ܠܘܬ ܐܒܐ ܘܐܒܘܦܘܢ‎

2 This phrase curiously illustrates the occasional fixity of Syriac idiom. ‘They laid [their] hands upon him’, ie. arrested him, is in Syriac ܐܪܡܝܘ ܥܠܘܗܝ ܐܝ̈ܕܝܐ‎ the word ܐܝ̈ܕܝܐ‎ being without a suthix and at the end of the clause. This is the case even in Matt xxvi 50, Lk xxii 53, where the Greek adopts the order usual in English. In Matt xxvi 50 the Greek has éméBadov ras yeipas emt Tov "Incovv Kat é€xpdtnoav aire. This does not go conveniently into Syriac, for . ܥܠ ܝܫܘܥ‎ is a little too heavy to be put between ܐܙܪܡܝܘ‎ and ܐܝ̈ܕܝܐ‎ Accord-

ingly S has ot ܥܠܘܗܝ ܐܝܕܝܐ ܘܐܚܕܘܗܝ ܠܝܫܘܥ‎ ami

But syr.vg is corrected back to the Greek, It has ܥܠ ܝܫܘܥ ܘܐܚܕܘܗܝ‎ Lamia ܐܪܡܝܘ‎ in accordance with the Greek order of the words. ܐܙܝ̈ܕܝܐ‎ being no longer at the end of the

clause, receives a suffix, as in Matt xix 13, 15, and thus the distinction between laying one’s hands on a person for blessing and laying them on for violence is obliterated.

48 Grammar and Syntax.

réasax..—See on Néldeke § 87.

is the constant spelling in S but in‏ ܒܠܺܝ ܐܢܫܐ to and‏ ܐܢܫܐ - ܐܢܫ occur occasionally? .‏ ܒܢܝ̈ܢܫܐ and‏ ܒܪܢܫܐ C‏

pterv.—The sing. abs. of this word is spelt ܐܚܪܢ‎ in S: the same spelling is found in Cat Joh iv 87, v 82, vii 33, always at the end of a line. ܐܚܪ̈ܢ‎ also occurs in cod. A of Aphraates (e.g. Wright, p. 156), and in cod. B (Wright, p. 48), but I have only noticed it once in the ss of syr.vg, vz. Matt xi 3 in Mr Gwilliam’s cod. 36.

pines, therefore, which stands at the end of a line for addoic 1 Mk x 40 S, is almost certainly meant for a plural, the scribe having probably intended to write ܠܐ ܚܪܢܝܢ‎

The emph. is always spelt ater in S, but ܘܚܪܵܢܐ‎ occurs in C at Lk vil 8 and a few other places.

rhato.—The form ܩܘܪܝܬܐ‎ Joh vil 42 S” is a mere error of transcription: ܩܪܺܶܝܬܐ‎ is quite clear in the photograph.

ܡ̈ܛܠܝܢ pl.‏ ܡܛܠܬܐ ‘roof, ‘dwelling,’ and‏ ܡܛܠܠܐ. ܡܛܠܠܐ ‘booth,’ have been confused in my translation, and the words are‏ certainly confused in syr.vg. But I now believe they were kept distinct‏ in Sand C. The ‘dwelling’ of the Centurion (Matt viii 8, Lk vii 9)‏ and the ‘dwelling’ of the birds of the heaven (Matt viii 20, Lk ix 58)‏ But the ‘booths’ which 8. Peter wished to set up at the‏ . ܡܛܠܠܐ is‏ Transfiguration (Matt xvii 4, Mk ix 5, Lk ix 33) the everlasting‏ ‘habitations’ of the parable (Lk xvi9)?, and the ‘Tabernacles’ of the‏ Feast (Joh vii 2, 14) are et\\sa, corresponding to oxnvai, oxnvornyia,‏ in the Greek.‏

Among anomalous nouns may be put s5az ‘seven’ Mk vii 5, 65S, Lk xx 29 S, ܫܘܒܥܬܝܗܘܢ‎ Mk xii 23S, ܫܘܒܬܐ‎ ‘Sabbath’ Lk xiii

14 S*”, and ܫܘܒܥܝܢ ܘܬܪܝܢ݀‎ Lk x 17 6", In the last two passages the photograph is illegible, but all four spellings may be held to hang together. These variations of the ordinary ܫܒ̈ܥܝܢ , ܫܒ̈ܥܐ‎ , “hax, are all the more interesting because they are found in the Christian Pales- tinian Aramaic, as well as in various forms of Jewish Aramaic.

Equally suggestive of early forms of Aramaic is saa Matt xiii 35 S. It occurs at the beginning of a line, where there was plenty of room for saa, so there is no reason to regard it merely as asdefective spelling.

1 E.g. Matt. xix 6, Lk ix 25». 2 In § read ܒܡ̈ܛܠܝܗܘܢ‎ as in ¢; with one ܠ‎ 01-

Anomalous Nouns and Particles. 49

D5 is always written without ( in the Aramaic portions of Daniel, and with a suffix we find M253 (Dan vii 5). ‘pS appears actually to occur

in the old Aramaic inscription from Nerab near Aleppo.

Numbers.

[Noldeke § 148.] The numbers in S and C call for few remarks. ܫܘܒ̈ܥܐ‎ for ܫܒ̈ܥܐ‎ (Lk xx 29 S) has been noticed above. In two places S appears to have a false concord, for we find ܫܥܐ ܥܣܖ̈ܐ‎ Joh 139 and ܫܬ ܝܘܡ̈ܝܢ‎ Mk ix 2. Such false concords are extremely uncommon: I have not found any example noted in Mr Gwilliam’s apparatus to the Peshitta Gospels.

Decapolis is rendered ܡܕܝܢ̈ܬܐ‎ dims in SC, as in syr.vg. But »ܘ‎ Tov Tecodpwr avéuwv Matt xxiv 31 is rendered Maat asin ܡܢ‎ in © where syr.vg has ܡܢ ܐܪܒܥܬ ܪ̈ܘܚܐ‎ ; in Mk xili 27 ܡܢ ܐܪܒܥܬܝܗܝܢ‎ ܖ̈ܘܚܐ‎ is found both in S and syr.vg.

Particles.

| Noldeke §155.] A Adverbs of Quality : w for dAws does not occur. In its place we find the very curious

locution ܩܥܘܡܟܘܢ‎ Matt v 34 SC (so also Aphraates 505); and in Joh ix 34 S has wears. For the meaning see Notes on Matt ¥ 34. hal, occurs Matt xv 32 C, but not in S or syr.vg.

The adverbial termination -ܐܝܬ‎ is written fully in S and C almost always, but we find ܡܪܺܝܪܺܝܝܬ‎ Lk xxii 62 C, ܫܪܺܝܪܺܝܝܬ‎ Lk xxiii 47 C in each case at the end ofa line. dhster occurs Matt xxi 29 C, but S has ܒܐܚܪܝܬܐ‎ . In Matt xxi 37 (where syr.vg has (ܐܚܪܶܝܬ‎ we find ܒܐܚܪܝܬܐ‎ in S, hates in €

B. Adverbs of Time and Place:

occurs in Lk xiii 9 SC for eis 7d wéddov, as in syr.vg. I have‏ ܠܡܢܚܝ translated it ‘next season,’ but the precise meaning of the phrase is as‏ doubtful as its derivation.‏

rita’ ‘where?’ and ܐܝܟܐ ܕ‎ ‘where’ is found in S, but side by side with it is found another form alors, especially in 8. John, which appears to be otherwise unknown. ܐܦܝܐ‎ occurs in S Matt xxiv 28, xxvi 13, 17; Mk ix 189, xiv 12, 14 bis; Lk viii 25, xii 17, xvii 37°,

II. 7‏ .ܡ

50 Grammar and Syntax.

moat 117"; Joh yu 35, vin 2207), xiv 4, ey. 6, xe 9, 18, 15, wx 5, On the other hand wale occurs Matt ii 2,4; Mk xv 47; Lk xvii 7, 875, scan 975 11”; Joh 1 28, 88, 89, (11:89, vin 11, war 145 19, 1 19, xi 34, xii 35, xiii 86, xiv 5. It will be seen that the two forms are used indiscriminately, even in the same verse. am ܐܝܦܐ‎ occurs Mk xiv 14 S: aase occurs Lk xix 23 € Lk xx 5 SC.

Among the linguistic peculiarities of S is a curious preference for ܡܢ ܐܝܡܟܐ‎ , Le. ‘from whence, instead of ܐܝܡܦܟܐ‎ ‘whence. In eighteen passages where ܐܝ .ܡܟܐ‎ occurs in the extant parts of the Evangelion da-Mepharreshe, ܡܢ‎ is prefixed 13 times in S, only 8 times in syr.vg. In C ܡܢ‎ 18 prefixed 8 times out of fifteen, six of them being in 8. John.

For the use of ܗܟܝܠ‎ ete. in rendering ,ܘ‎ see the Appendix at the end of this Chapter.

Among the words for ‘immediately’ ܒܫܥܬܐ‎ ca> is much the most frequently used in S and Cin Matt and Mk, but ܒܪ ܫܥܬܗ‎ and Wass are also found. ܡܚܕܐ‎ is written ܡܢ ܚܕܐ‎ in Mk vi 45 S, and ܒܝܝܕܐ‎ oceurs Mk vi 25 S, Joh xiii 32 8. tule ܡܢ ܬܚܝܬ‎ occurs in SC for ܘܐ )ܘ‎ Lk 1x 39 and for aidviduos Lk xxi 34}.

rast ܕܩܠܝܠ‎ (sic) for ܼܬܐ ܧ‎ ypdvov Lk iv 5 S is worth notice as an idiomatic rendering.

C. Adverbs of Quality and Conjunctions.

occurs Lk xvi 11, xx 17, xxii 70, both in S and in C, as the‏ ܥܘܕ equivalent of ody, in Lk xxii 3 S (not C’) and in Joh xi 37 S (heat C)‏ without any Greek equivalent. This word is only found in the oldest‏ Syriac literature, and seems to mean ‘forsooth, always with a‏ touch of contempt. I suppose it was considered too lively a particle‏ to be retained in Scripture. It has been allowed no place in the‏ Peshitta, and it has been diligently washed out in each of the three‏ passages where it occurs in C. This does not appear to have been‏ the case in S, for where a letter has been washed out in S by a cor-‏ rector before it was turned into a palimpsest it is now totally illegible.‏ But what has happened in C’ has happened also in Aphraates, who‏

1 The same phrase occurs in syr.vg for eEaidpvns Ac ix 3, xxii 6, but in the Gospels only the ordinary phrases ale. ܡܢ‎ and ܡܢ ܫܠܝ‎ have been allowed to stand. 2 E.g. the first letter of [ܡ]ܦܝܠ‎ Lk xii 31.

Adverbs and Conjunctions. 51

when quoting 1 Cor xv 29 has ܠܐ ܩܝܡܝܢ.‎ xan ܐܢ ܡ̈ܝܬܐ‎ and in quoting 1 Cor xv 32 ܢܐܟܘܠ ܥܘܕ ܘܢܫܬܐ. ܕܡܚܪ ܡܝܬܝܢ ܚܥܢ.‎

(Wright, p. 157). At least this is what the two mss originally had, as I was able to see after a careful examination in a good light. But in the former passage both mss now have ܓܝܪ‎ ; and in 1 Cor xv 32 the one Ms (A”) has ܓܐܪ‎ (ie. tX_), and the other (A*) has ܡܕܝܢ‎ , Both sets of corrections appear to be quite late.

The particle ,ܠܡ‎ used in quoting other people’s words, very much after the manner of ‘says he’ or ,)ܗܗ‎ is found in Matt xvi 13 SC, Lk xiv 17 SC, and Joh xii 34 S (heat C). ,ܠܡ‎ like sas, has been banished from the Peshitta Gospels, but in the Epistles it is very suitably retained in Col ii 21, 2 Thess 11 2. In this point, as in others, the Gospels were more drastically revised than the rest of the N.T.

in Matt xii 33 S*, Lk xviii 25 S, but possibly‏ ܓܪ is spelt‏ ܓܝܪ (=ydp) is however the‏ ܓܪ these were merely slips in writing.‏ spelling found in Palestinian Syriac documents.‏

never occurs in the Evangelion da-Mepharreshe, simple ܠܐ‎ or ram ܠܐ‎ being used instead. x occurs Mk xi 13 S, as a rendering of € dpa.

[Noldeke §156.} The Prepositions in S and @ shew few peculiarities of form: for syntactical uses, see on Noldeke §§ 246-252.

In Lk ix 23 ܢܝܐܬܐ ܒܐܬܪܝ‎ in S, followed by ,ܢܝܐܬܐ ܒܬܪܝ‎ appears to be an attempt to distinguish between the ܘ ܗܐܘ‎ pov épyer Oar and dkoovbeirw jot of the Greek. I have not met with any other resolution of ܒܬ ܪ‎ into its original elements.

Verbs.

[Noldeke §158.] The longer form of the 3rd pers. pl. mase. of the Perfect occurs in ܗܝܡܢܘܢ‎ Joh xi 40 S (sec). The 3rd pl. fem. Perf. is identical with the 3rd sing. masc. in S and C, as in all other ancient Edessene Mss, except in verbs tertiae »'. But in the Imperative pl.

1 Te, ‘my eyes were opened’ is ܐܬܦܬܚ ܠܝܠܟ‎ not ܐܬܦܬܢܝ ܠܝܠܟ‎ In the case of a palimpsest like S, often very difficult to read, there might in some case be a doubt whether a

52 Grammar and Syntax.

fem. the longer form in én is used, as in the Peshitta, e.g. perme Lk xxiv 6 SC. In Mk xvi 7 for ܘܗ‎ € ei’tate Dr Harris edited ܐܙܠܝ ܐܡܪܝ‎ , my transcript has ܐܡܪܝܢ‎ AY. It is therefore probable that the former of the two verbs has no nén at the end: the photograph is unfortunately illegible.

[Néldeke §160.] ܝܩܝܪܘ‎ Lk ix 32 S must be a mere orthographical miswriting of atas, not a relic of an intransitive 2-form. In several other places S has a twist too many, identical in shape with the small letter yod, e.g. ܐܫܟܢܢܝܢ‎ (for (ܐܫܟܝܝܢܢ‎ Matt xvii 19, rascal (for (ܠܡܚܐ‎ Matt xix 25, ܚܡܫܝܝܢ‎ (for ܚܡܫܝܢ‎ ( Joh vill 57.

For ܝܗܒ‎ instead of sms, see on Noldeke § 183.

[Noldeke § 172.] ܟܐܐ‎ Lk xvii 3 Sis no doubt a scribe’s blunder for ,ܦܐ ܝ‎ Both Sand C correctly read ܦܐܝ‎ in Lk xix 39.

[Noldeke §173.] For the Imperative of t\a, S has ata\, Joh xiv 15 and ܛܪ‎ Joh xvii 11. For ܛܪ‎ Matt xix 17 C we find ܢܛܪ‎ in S, ie. the Pael. The Imperative Peal of ܢܩܡ‎ ‘to knock’ occurs Lk xi 9 S Cin the ordinary form azas, and thé Perfect occurs in Lk xii 36, but in Lk xi 10, xiii 25 (where C has the Participles rms, ,(ܢܩܫܝܢ‎ we find rom and ܡܩܫܝܢ‎ in S, ie. the Aphel participles magqqesh, maqgéshin. These appear to be the only passages where the Aphel of this verb is used in Edessene, but ܡܩܫܝܢ‎ occurs in the Christian Palestinian Lectionary at Lk xiii 25. Here again therefore a peculiarity of S finds illustration from other Aramaic dialects.

[ Néldeke §174.] For ܐܙܠ‎ and whe, see on Néldeke § 183.

[Noldeke §177.] ܢܕܫܘܢ‎ Lk xii 1S, ܬܬܒܘܢ‎ Lk xiii 3S, and ame (imptv.) Lk ix 44 S, are simply instances of seriptio defectiva for ܬܬܘܒܘܢ , ܢܕܘܫܘܢ‎ and ܣܝܡܘ‎ : they do not imply any difference of grammatical form.

] Néldeke §183: the anomalous Verbs. }

is‏ ܐ and \ie.—In the Imperative of these verbs the initial‏ ܐܬܐ sometimes not dropped in S, particularly in ©. Matthew. The instances‏

mere twist were added ut the end of a word like ܐܬ ܠܦܬ ܙܕ‎ or not; what makes it almost certain that such a twist is never added is that verbs ending in ܟܙ‎ and ~p never add the yod. Thus in Matt xxv 5 we have ܠܡ‎ 4 a, not ܢܡܝܢ‎ and »Aspa: both in Matt xxiv 32 and in Mk xiii 28 S reads asa not ,ܪܰܦܟ‎ In zpMe xxvii 572 (Cyrillona ii 194) Bickell edited ܨܢܣܦܝ‎ but I have ascertained that the original reading of p.m. Add. 14591 is .ܠܦܘ ܟܙ‎ Curiously enough, Gwilliam’s cod. 36 reads say for tay in Matt xiii 6, where the word is masc. sing.

Anomalous Verbs. 53

are (1) for ܐܬܐ‎ and ܐܬܘ‎ , Matt xi 28, xix 21, xxii 4; Lk ix 59; (2) for ܐܙܠ‎ and ܐܙܠܘ‎ , Matt ii 20, v 41, vili 4, 9, ix 6, 18, xvii 27; Mk vii 29 (Maw); Lk vii 22, x 3, xiii 31, xxii 10. It will be seen that no example of this spelling occurs in §. John.

The fem. pl. Imperative of ܐܬܐ‎ is spelt ܬܝܢ‎ in Matt xxviii 6 S, the only passage where it occurs. ܗ‎

so written, from‏ ܠܐ ܬܪܛܘܢ Lk xvii 23 S we find‏ ܙ -. ܪܗܛ was unpronounced,‏ ܗ which perhaps we may infer that the vowelless‏ and so became liable to be dropped in writing.‏

>m..—In three places we find ܝܗܒ‎ written in S, where the sense is passive and we should expect msm.. The passages are

(1) ܝܗܒ ܐܪܙܐ ܕܡܠܟܘܬܗ ܕܐܠܗܐ‎ am ܠܦܘܢ‎ Jo you is given the mystery of the kingdom of God (Mk iv 11 S) ;

(2) ܕܠܝ ܝܗܒ‎ MN... ܠܟ ܝܗܒ ܐܢܐ ܫܘܠܛܢܐ ܗܢܐ‎ To thee I

give this authority...because to me it 1s given (Lk iv 6 S);

(3) ܠܗ ܟܬܒܐ‎ seal? ܘܥܠ ܠܟܢܘܫܬܐ.... ܐܝܟܢܐ ܕܡܥܕ ܗܘܐ‎ ܕܐܫܥܝܐ‎ And he entered the synagogue...as he was wont. 1 And there was gwen him the book of Isaiah (Lk iv 16, 17 S).

The same spelling occurs in the codex of Aphraates called by Wright A and cited in this book as A*% In Wright, p. 355, we find ܠܗ̇. ܐܠܐ ܡܶܢ ܕܝܗܒ ܠܗ‎ nam ܠܐ ܗܘܐ ܟܠܢܫ‎ Not every one is sufficient for wt, but he to whom it rs given (Matt xix 11(. The other ms of Aphraates A> has ܕܝܗܝܒ‎ instead of omer, and ܕܝܗܝܒ‎ is also found here in S @ 8110 the Peshitta!, but the correction was so obvious that we may be pretty sure that A* preserves the true text of Aphraates. And again (Wright, p. 114), according to the same ms A’, we find ܫܢܢܐ ܕܚܪܒܐܐ ܫܩܝܠ ܡܢ ܐ̈ܦܝ ܐܝܠܢܐ ܕܢܳܢܳܝܳܐ܂ ܘܝܗܒ ܡܦܘܠܬܐ ܠܡܗܝ̈ܡܢܐ܂‎ ian le? ܘܦܪܕܝܣ̈ܐ ܡܠܝܟ ܠܛܘ̈ܒܢܐ. ܘܠܒ̈ܬܘܠܐ ܘܩܕܝܫ̈ܐ܂ ܘܝܗܒܝܢ ܦܐܪ̈ܝ‎

ܡܦܘܠܬܐ ܠܡܗ̈ܝܡܢܐ ܘܠܒ̇ܬܘܠ̈ܐ,

(At the coming of Christ) the edge of the sword is taken from in front of the Tree of Life, and it is given for meat to the faithful ; and Paradise is promised to the blessed and to the virgins and the holy, and the fruits of the Tree of Life are given for meat to the Juithful and to the virgins.

The other ms A has ܘܝܗ ܒ=‎ and ܆ ܘܝܗܝܒܝܢ‎

These five instances of the use of ܝܗܒ‎ for ܝܗܝܒ‎ shew that we are

1 One ms however, Mr Gwilliam’s 14, actually has a2 here, and similarly Mr Gwilliam’s 9 has ms Joh xix 11.

54 Grammar and Syntax.

dealing with something more than a slip in writing. It would be intolerably harsh to translate the passages impersonally, and at first I believed that they contained a relic of the old Passive formed by internal vowel change, instances of which survive in Biblical Aramaic. 37’ actually occurs in Dan vii 14. But in the passage just quoted from

Aphraates 114 the word ,ܝܗܒܝܢ‎ however pronounced, must be a Participle : it cannot be a Perfect tense. Moreover, as all the instances of this defective spelling concern the verb om, the explanation must be specially suited to that word. I therefore conjecture that when ܝܗܒ‎ is written for the Pass. Part. the ܗ‎ is intended to be elided, and that just as in the Perfect original yZhaB has become yaB, so in the Participle original yzhiB became yi®B or 78: that is to say, we should

point the word .ܝܗܒ‎

a

The spelling ܝܗܒ‎ for ܝܗܝܒ‎ is also found in Christian Palestinian documents, viz. Matt xix 11 codd. asc, Lk vii 25 codd. Be, Joh vi 66 cod. ,ܡ‎ Joh vil 39 codd. Bc, Joh xix 11 codd. Bc; also in Mrs Lewis’s Praxapostolos Gen ix 17, Exod xi 5 (sic)}.

[Noldeke § 184 ff: the Verbal Suffixes.| The Verbal Suffixes are regular in C, so far as the consonantal writing is concerned, but in S there are found some rare and some otherwise almost unattested forms, especially in the 3rd sing. masc. suffix to the Imperfect.

§§ 188, 189. The forms found in S are

ܢܩܛܠܘܢܝܗܝ ܢܩܛܠܝܘܗܝ ܢܩܛ ܠ ܘܢܗܝ ܢܝܩܛܠܝܗܝ

ܢܩܛܠܘܢܗ ܢܩܛܠܘܗܝ

with corresponding forms for Saad, ,ܐ ܩܛܘܠ‎ etc. The form ܢܩܛܠܗ‎ does not occur either in S or C’; on the other hand ܢܩܛܠܘܢܗ‎ is much commoner than ܢܩܛܠܘܢܝܗܝ‎ both in S and in C. Examples of the irregular forms in S are given below. It will be noticed that they are

less frequent in 8. Matthew than elsewhere. (1) Forms in ,oaa- (see also on § 195) :-̈ ܕܫܟܥܚܝܗ ܝ‎ Matt xxiv 46 (contrast Lk xii 43); ܐܫܐܠܝܗܝ‎ Mk vi 24 (contrast Matt xiv 7), ܕܬܘܒܕܝܗܝ‎ Mk ix 22, ܕܢܫܠܡܝܗܝ‎ Mk xiv 10, 11 (contrast Matt xxvi 16),

1 In the last instance sm, means simply ‘is’ or ‘is to be found,’ Just as ܝܗܒܝܢ‎ in Lk vii 25 corresponds to imapyovtes.

Anomalous Verbal Suffixes. 55

Lk xii 44,‏ ܕܢܫܠܛܝܗܝ ;)61 Mk xiv 58 (contrast Matt xxvi‏ ܐܣܬܪܝܗܝ ܕܢܦܪܩܝܗܝ ,18 Lk xx‏ ܬܫܚܩܝܗܝ ,461 Lk xt‏ ܘܢܦܠܓܝܗܝ .... ܘܢܣܝܡܝܗܝ ܕܐܣܝܡܝܗܝ. . . . ܐܫܩܠ .ܗܝ ;34 Joh iv‏ ܘܐܫܠ ܡܝܗܝ ;21 Lk xxiv‏ Joh x 18 (both apparently meant for fem. suffixes, as in Lk xii 46),‏ Joh xiii 2,‏ ܕܢܫܠܡܝܗܝ ,48 ,47 Joh xii‏ ܕܐܕܘܢܝܗܝ.... ܬܕܘܢܝܝܗܝ Joh xiv 21.‏ ܐܪܚܡܝܗܝ ܘܐܚܘܝܗܝ

(2) Forms in ,ena- (see also on §195):—,ma\\eax Matt xxiv 47?; ܕܢܩܝܡܘܗܝ‎ Lk xii 42°, ܕܬܫܕܪܘܗܝ‎ Lk xvi 27, ܐܫܒܩܘܗܝ‎ Lk xxii 16, 22; ܕܢܕܘܢܘܗܝ‎ Joh ili 17, ܐܩܝܡܘܗܝ‎ Joh vi 40, 54, ܕܐܥܝܪܘܗܝ‎ Joh xi 11.

(3) The Plural forms in msa- and -ܘܢܝܗܝ‎ need no illustration ; it is sufficient to observe that msalsy and ܕܢܣܝܡܘܢܗ‎ occur Lk v 18 ® but ܕܢܥܠܘܢܝܗܝ‎ Lk v 19S. The following instances of -ܘܢܗܝ‎ are found in S:—,malezian Mk ix 32, ܕܣܬܟܠܘܢܗܝ‎ Lk xxiv 16, ܕܢܥܒܕܘܢܗܝ‎ Joh vi 15, ܬܪܺܝܡܘܢܗ ܝ‎ Joh vil 28, ܢܩܛܠܘܢܗܝ‎ Joh xii 10, ܕܢܫܠܡܘܢܗܝ‎ Joh xviii 284. ܢܩܛܠܘܢܗܝ‎ also occurs in Lk xxii 2 C, where S has ܘܢܝ ܗܝ‎ - .

§ 190. The regular form of the Impt. mase. pl. with suffix occurs in ܕܘܒܪܘܗܝ‎ Joh xviii 31 S (sic), but for verbs with initial ܐ‎ we find ܐܡܪܘܗܝ‎ ‘say ye it’ Matt x 27 S, ܐܚܕܘܗܝ‎ ‘take ye him’ Matt xxii 13 S, Mk xiv 448.

In the Sing. we have the regular forms ܙܩܘܦܝܗܝ‎ Mk xv 14 S (and in Lk xxii 21 C), also ܫܩܘܠܝܗܝ‎ Lk xxii 18 S, but in Lk xxiii 21 S has ܙܩܦܘܗܝ ܙܩܦܝܗܝ‎ for cravpov oravpov.

[Noldeke £192 ff. Verbs with final ܐ‎ and suffixes. |

§ 194. In the forms of the 3rd pl. masc. Perf. with suffixes we find -aa- always written in S for -ar-, and generally in C. Thus we have ܚܙܘܘܗܝ‎ Matt 110, 11 SC, Matt xxi 28 SC, Lk xx 14 SC; but smarts occurs Matt xiv 26 C, where S has avs without a 85 With a fem. we find ܡܠܘܘܗ‎ Matt xiii 48 S,

In Mk vi 49 S ܚܙܘܢܝܗܝ‎ is the 3rd pl. masc. Perf. in -én- with suffix. The word was so read by the late Professor Bensly and myself at Sinai, but the form is said to be otherwise unknown in these verbs and I

1 Possibly these words were meant for the fem., i.e. ܘܢܦܠܓܝܗ ܡܢܬܗ ܘܢܣܝܡܝܗ‎ : cf Joh x 18.

2 Iam not quite sure that the true reading of S may not be ܕܢܫܠܛܝܘܗܝ‎ On the other hand in Matt xxi 38 the photograph of S appears to me to suggest um ܢܒܛܠܘ‎

3 Photograph illegible.

4 In ܢܢܢܝܒܘܢܝܗ‎ Mk x 33 S the ܥ‎ is no doubt intrusive.

5 ܠܙ ܐܕܘܗܝ‎ 8180 occurs in the ancient palimpsest fragments of the Acts of Thomas (= Wright 3121).

56 Granmar and Syntax.

confess that the photograph now suggests to me ,waars» as the reading of the ms, a form which actually occurs in the following verse, Mk vi 50.

§ 195. Instances of the irregular suffixes to the Imperfect are given below from S.

(1) Forms in ܝܗܝ‎ - (see above on §§ 188, 189) -—,malah ܠܐ‎ Matt ¥ 42 (sic), Lk vi 29; ܐܚܙܝܗܝ‎ Mk xii 15; ,oastxan Lk xxili 20, ܐܪܺܕܝܗܝ‎ Lk xxiii 22; ܕܐܚܙܝܗܝ‎ Joh xii 47, ܘܐܚܘܝܗܝ‎ Joh xiv 21.

(2) Forms in ,ma-:— ,moate Lk xxiii 161. :

§ 196. In the Imperative, as in the Perfect, we find in S -ao- instead of -arv-: e.g. ܐܝܬܘܘܗܝ‎ Lk xix 30 S (but ܐܝܬܐܘܗܝ‎ C) ; ܩܪܘܘܗܝ‎ Matt xxii 9 S, ,naat Joh xi 44 ܝܗܘܝܬܝܐܘ‎ Mk x12 S is probably a mere slip in writing. The Imptv. pl. of ,ܚܘܝ‎ with suffix of Ist pers. sing., is saa Matt ii 8, xxii 19, Lk xx 24 in S, but C has the regular form »saras in all three passages.

In the Infinitive of these verbs, besides the regular forms we find ܠܡܚܙܝܗܝ‎ ‘to see him’ Lk xxiii 8 S.

These irregular suffixes, which are one of the most striking gram- matical peculiarities of S, have some slight attestation in other Syriac documents. An instance occurs, curiously enough, in the dedication prefixed by Habibai to Codex @ itself where we find ܢܡܢܘܗܝ‎ for ܢܡܢܝܘܗܝ‎ : this may however have been a mere error of the writer. Of much more importance is the occurrence of ܕܐܥܝܪܘܗܝ‎ in Aphraates cod. A (i.e. A*, Wright, p. 169), in a quotation of Joh xi 11, the same passage where it is found in S. 16 is therefore clear that these peculiar sufhixes are not merely due to some accident of transcription in the Sinai Palimpsest. They are doubtless genuine remains of that early stage of Edessene Syriac, of which the Evangelion da-Mepharreshe is the only well preserved monument: of these suffixes, as of other features, we may say with Wellhausen, “man gewinnt den Eindruck, dass solche Rarititen stehn gebliebene Reste sind, dass schon im Sin. und Cur. die stilistische Korrektur begonnen hat, die in der Peschita (namentlich des Neuen Testaments) entschiedener, wenngleich auch nicht systematisch durchgefiihrt ist?.”

± ܐܕ ܢܙܘܗܝ‎ also occurs in Cyrillona i 12 and in the ancient palimpsest fragments of the Acts

of Thomas (corresponding to umaswrt Wright 3158).

2 J. Wellhausen in Vachrichten der k. Gesellschaft der Wissenschaften zu Gottingen. Phil.-hist. Klasse, 1895, i, p. 5.

Uses of the Absolute State. 57

Syntax.

The following remarks do not profess to be anything like a complete account of the Syntax of S and C, or even of their peculiarities of Syntax. As was remarked at the beginning of this Section, the Evangelion da-Mepharreshe is written in idiomatic Edessene Syriac : the most notable peculiarity of S and C is not the presence of this or that idiom, but their freedom from that imitation of Greek construc- tions which pervades so much of the later Syriac literature.

[Noéldeke §2028.] ܩܘܪܺܝܬ ܕܡ‎ (ie. “Bloodfield”) Matt xxvii 8 S is an interesting example of the Absolute state in proper names. It is curious that the Syriac should have avoided the obvious rendering ܕܡܐ‎ Ans, found in the Latin and in the Palestinian Syriac texts’.

Equally noteworthy is ܡܠܟܘ‎ Joh xviii 10 S (sec), because it shews that the translator of the Evangelion da-Mepharreshe recognised the genuine Semitic name which appears in the Greek as MaddAyos. The name is actually written yobp in a number of Palmyrene inscriptions of the 1st cent. aD and is doubtless identical with the common Arabic

name Sue Manik". The Peshitta has Maléy, without the final a :

probably by the 5th century the a had become unfamiliar as an ending to masculine names, and ܡܠܟܘ‎ only sounded like the abs. sing. of ܡܠܟܘܬܐ‎ . Similarly in Nehem vi 6 192 becomes wasX_in syr.vg.

[§ 202c.] The curious phrase aS), wala ܓܒܝܘ ܢܘ̈ܢܐ‎ Matt xiii 48 SC appears to mean “they chose out the fishes whatever good ones there were.” A doubled adverb is normal in Syriac, as in the phrase ܥܒܝܕܝܢ‎ wo mon, used for of kaxas €xovres in the Peshitta as well as in SC. But the use of a plural adjective in this distributive sense is very uncommon and seems to have been a puzzle to the trans- lator of the Armenian version of the Gospel. A good example of the distributive use of the absolute state is paar ܐ[ܦܝܢ‎ for car’ dy Joh vii 24,8 C. The Peshitta has Warts «ammo.

1 The Palestinian Lectionary has for Matt xxvii 8 ܛܘܪܐ ܗܐܘ ܕܩܠ ܐܕܡܐ ܥܕ ܡܛ ܐ ܠܝܘܡܕܝܢ‎ sinh oa) Land’s ancient cod. Petropolitanus (a Ms of the continuous Gospel text) has

ܠܟܢ Sse woh‏ ܕܝܬܗ eat‏ ܕܩܠ ܕܡܐ ܡܳܐ ings‏ ܕܐܕܡ ܥܕܡܐ

1 ܨ ܠܝܘܡܕܢ.

B. IL, 8

58 Granunar and Syntan,

[Noldeke § 208 4.[ ܗܘ ܐܬܪܐ‎ yin Lk xv 15 S is an instance of the somewhat careless construction mentioned by Duval (Grammaire Syriaque § 357 f), who quotes ܪܘܡܝ ܐܢܬܬ ܗܘ ܡܠܟܐ‎ from B.O, i 365. In Lk xv 15 C has am ܒܢܝ ܐܬܪܐ‎ , but in Lk xix 39 both S and C have ܟܢܫܐ‎ am ܡܢ ܒܝܢܝ‎ .

[Noldeke § 210.1 ܟܠ ܩܪܶܐ ܕܓܠܝܠܐ‎ ‘every village of Galilee’ Lk ¥ 17 8 is a good instance of the use of the absolute state before a genitive.

[Noldeke § 212.] In Lk xxiii 2 kat Aéyovra 6:7076 ypiorov Bacrhéa civat is translated in SC ܘܐܡܪ ܥܠ ܢܦܫܗ ܕܗܘܝܘ ܡܠܟܐ ܡܫܝܚܐ‎ The last two words agree letter for letter with the Jewish NMwWh xobp, commonly translated ‘King Messiah.’ But Dr Dalman (IVorte Jesu 240) has shewn that it is a mistake to treat ‘Messiah’ in this phrase as a proper name, and that we should render it ‘the Anointed King.’ And this also agrees with the usage of the Evangelion da-Mepharreshe and the Peshitta, for the Syriac not only has ‘ Herod the king’ in Matt ii 3, Mk vi 14, where the better Greek texts have 6 Baoweds “Hpddns, but also ‘Augustus Caesar’ in Lk ii 1, although ‘Caesar Augustus’ is the order found in all other authorities. In Lk xxiii 2, therefore, ܡܠܟܐ ܡܫܝܚܐ‎ should be translated ‘an anointed king’ rather than ‘King Messiah.’

[Noéldeke §217.] ܟܠ‎ for ܟܠ ܡܕܡ‎ is now attested by S in Joh iil 35, as well as Aphraates 123. On the other hand the Peshitta has As in Johi 3, where ܟܠܡܕܡ‎ is read by C and by Ephraim (Rom. iv 18 ,ܐ‎ Lamy 11 513).

[Noldeke § 2208.] By the dropping of a repeated ܡܢ‎ the enclitic ܐܢܘܢ‎ occupies an unusual position in ܡܢܝ ܐܡܝ ܐܘ ܐܢܘܢ ܐܙܚܢ‎ ‘“Who’s my mother or who are my brothers?’ Matt xii 48, In the parallel passage Mk 111 33 the ܐܢܘܢ‎ is not present.

[Noldeke § 222.] The idiom of a preposition followed by 8 pro- nominal suffix and a is ingeniously used in Joh xi 32 S, where 7\ev ܘܐܘ‎ yv “Inoods is rendered ܡܛܬ ܠܘܬܗ ܕܝܫܘܥ‎ , Similarly in Joh vii 42 ris ܢܳܐ‎ (¢ Omov qv Aaveid is rendered ܩܪܝܬܐ ܕܝܠܗ ܕܕܘܝܕ‎ by the Peshitta as well as SC. In each case the clumsy Johannine peri- phrasis is indicated without being allowed to hamper the movement of the sentence,

Pronominal Phrases. 59

[Néldeke § 223.] ‘mas-o with suffix is twice used for ddws, viz. ܩܢܥܘܡܟܘܢ ܠܐ ܬܐܡܘܢ‎ = p17) :ܘܗܘ ܘْ‎ dws Matt ¥ 34. SC and A 505, and again ܒܚ̈ܛܗܐ ܐܬܝܠܕܬ‎ main ܐܢܬ‎ = ev ܗ‎ 076 av eyevuynOns

odos Joh ix 34 S, where ܘ ܘܬܘ‎ is read for ddos in 1-118-131-209 and some other Greek mss, as well as the Armenian vulgate. The same idiom has been left standing in 1 Cor vi 7 syr.vg, where ddws yr7qHa ܐܬܐܐܐ‎ éoriv is rendered ܝ ܩܦܢܥܘܡܟܘܢ ܚܒܬܘܢ ܠܟܘܢ‎ ie. ‘ye your own selves are guilty.’

[Noldeke § 224*.| To render 6 Sapouobeis Mk v 18 S has the characteristic Syriac idiom ܕܫܐܕܗ‎ am, and this reappears in the Peshitta as ܕܫܐ̈ܕܘܗܝ‎ am. But in Lk viii 36 6 dayomoOels is only rendered in S and C by ‘that man’; accordingly in the Peshitta we find ܕܝܘܢܐ‎ ac etaX_‘that demoniac man.’ Evidently aw ܓܒܪܵܐ‎ was taken over from syr.vt, and ܕܝܘܢܐ‎ added to give the sense of the Greek. In other words the text of the Peshitta in Lk viii 36 is con- flate; we catch the reviser at work and see how his style differs from the idiomatic Syriac of the Evangelion da-Mepharreshe.

[Noldeke § 225.] Asn is occasionally used in S and C to give - emphasis, e.g. ܡܛܠ ܫܡܝ ܕܝܠܝ‎ Matt v 118 (om. ܕܝܠܝ‎ C), where syr.vg has ܡܛܠ ܬܝ‎ in agreement with the Greek &vexey guod. Cases like ܠܐܘܨܪ̈ܐ ܕܝܠܝ‎ Matt xiii 80 C (om. ܦܬܘܪܐ ܕܝܠܝ ,(§ ܕܝܠܝ‎ Lk xxii 308 ܦܬܘܪܝ)‎ C’), where ܕܝܠܝ‎ appears to have no real force at all, are very uncommon and seem to be due to some accident of revision.

[Noldeke § 228.| It is here very truly remarked by Néldeke that the difference between ܗܢܘܢ‎ ‘those’ and ܗܠܝܢ‎ ‘these’ is often neg- lected in Syriac. Thus in Matt xxii 23 ratra...«axetva is rendered by ܗܠܝܢ.... ܘܗܠܝܢ‎ in the Peshitta as well as in S C, although in other more important respects the Peshitta text of this verse has been conformed to the Greek.

[Noldeke § 233. ] ܡܢ ܫ‎ stands for ¥) cou dvoud éorw; (Lk

viii 30) in syr.vg as well as SC. This is no doubt the old Semitic idiom: it is found in the Hebrew text. of Judges xiii 17 JOY 'D, and in the Targum of Onkelos to Gen xxxii 27 we find SY ܐ‎ although the Massoretic Text here has ]2¥ 72. The use of the phrase ‘Who is thy name?’ is doubtless connected with that identification of the name with the personality, whereby in Semitic idiom the Name of God is

60 Grammar and Syntax.

practically used for His personal character, as known to the wor- shipper.

[Noldeke § 2360.] In several passages, e.g. Matt xx 15, Lk x 28, Joh iv 22, S has ܡܐ ܕ‎ where C and the Peshitta have .ܡܕܡ ܕ‎ In a few places the Peshitta keeps ܡܐ ܕ‎ with S, while C has the ordinary ,ܡܕܡ ܕ‎ e.g. Matt xvii 30.

[Noldeke § 240 4.] ܫܒ̈ܥ ܫܒܥ‎ [paar] ܥܠ ܫܒ̈ܥܝܢ‎ Matt xviii 225 © and 4?/,, literally ‘on 70 [temes] seven seven, does not mean 3430 times but 490, as Aphraates especially declares (Wright, p. 298). ܫܒܥ ܫܒܥ‎ means ‘seven by seven,’ Le. ‘in rows of seven. When therefore the acts of forgiveness are piled in rows of seven upon seventy, we get 70 x 7, which is four hundred and ninety.

[Noldeke § 243.) In Lk iv 40 8vvovros tov nAlov is rendered in S by the idiomatic ܡܥܪ̈ܒܝ ܫܡܫܐ‎ , without ܒ‎ prefixed, and this also is the reading of the Peshitta, as edited by Mr Gwilliam from the mss. In Mk i 82 ܡܥܪ̈ܒܝ ܫܡܫܐ ܕܝܢ‎ in S corresponds to dias 8¢ yevouevys ore eOvoev 6 ܘܬܐ‎ ¢ and in remedying the apparent deficiency syr.vg inserts the ܡܥܪ̈ܒܝ ,ܒ‎ no longer being the first word in the sentence ܒܪܡܫܐ ܕܝܢ ܒܡܥܪ̈ܒܝ ܫܡܫܐ)‎ (. Here again the effort to conform the Syriac to the Greek has resulted in the disappearance of a characteristic Syriac idiom.

[ Noldeke § 244.| For the omission of ܠܕ‎ in short descriptive clauses, see on Noldeke § 275.

[Noldeke ܛ‎ 249 £. [ The curiously slack construction sometimes used after ܡܢ‎ in comparative clauses is found in Matt xviii 13 S C’ as well as in the Peshitta and Aphraates 142, where we read that the shepherd who has found the lost sheep “ rejoiceth over it more than the ninety and nine which have not gone astray” ܡܢ ܬܫܥܝܢ....)‎ tedu .(ܚܕܐ ܒܗ‎ The Greek is paddov 7 eri followed by a dative, but no Syriac authority reads the logically more correct ܡܢ ܕܥܠ‎ teks. The same construction is found in Lk xv 7 C, but S followed by syr.vg reads As are instead of ܝܬܝܪ ܡܢ‎

Similarly in Matt xxvii 9 there is nothing but the context to decide whether the prophet held the Christ dearer than he held the sons of Israel, or whether he held the Christ dearer than the sons of Israel did.

Usages with Prepositions. : 61

The use of ܐܘ‎ instead of ܡܢ‎ in comparisons, in imitation of (or at least corresponding to) the Greek 7, occurs several times in the Gospels, e.g. Matt xix 24 SC, confirmed by Aphraates 392. I see no reason to doubt that S in Lk xv 7 preserves the original rendering of the Evangelion da-Mepharreshe: there are many characteristic variations in the early part of Lk xv where S differs both from C' and the Peshitta, and it is more likely that @ borrowed ܡܢ‎ tc in v. 7 from the parallel passage Matt xviii 13 or from the Diatessaron itself than that S should have adopted the one variation ܐܘ ܥܠ‎ from the Peshitta.

The compound preposition ܡܢ ܠܘܬ‎ corresponds to the French de chez, e.g. 111 Lk viii 37, and also in Matt vii 34S, the Gadarenes beseech Jesus to depart ܡܢ ܠܘܬ ܗܘܢ‎ . Similarly Judas comes with a multitude mamas ܡܢ ܠܘܬ ܪ̈ܒܝ‎ Matt xxvi 47, Mk xiv 48, the Greek being dé in Matt but apa in Mk. It is a little less direct than the simple ܡܢ‎ 4 gives the impression that the place of departure is, so to speak, a little more complex and less personal. Accordingly it is used of God, in exactly the same way as DP | is used in Jewish Aramaic. Our Lord is thus said to have come forth ܡܢ ܠܘܬ ܐܠܗܐ‎ in Joh xiii 3 S (=a $eod) and in Joh xvi 27 S-vg (=apa rod Oeod). But there can have been very little significance in the periphrasis, for wapa ܘܗ‎ 0 is rendered

in S by ܡܢ ܠܘܬܟ‎ in Joh xvii 7 and by ܡܟ‎ inv. 8. In Matt xxi 42

and Mk xii 11 (‘ From the Lorp this came to pass’) all the Syriac texts have ruts ܡܢ ܠܘܬ‎ , but the Peshitta of Ps exviii (exvii) 23 has the characteristically Jewish ܡܢ ܩܕܡ ܡܪܝܐ‎ .

[Noldeke § 250.] For ‘demoniac possession’ in the Synoptic Gos- pels the Syriac vulgate uses the preposition ,ܒ‎ e.g. ‘a demon is wm him’ Matt xi 18, ‘22 whom was a demon’ Lk viii 27; in S. John édaiporiov ܕ :)ܗ‎ is literally translated ܕܝܘܐ ܐܝܬ ܠܟ‎ ‘thou hast a demon.’ But

in S and C'this ܠ‎ is never used and its place is supplied by ܒ‎ or As, eg. ܕܝܘܐ ܐܝܬ ܒܟ‎ Joh vii 20 SC, ܕܝܘܐ ܐܝܬ ܥܠܝܟ‎ Joh viii 48, 52 S (hiat C), and ܕܐܝܬ ܗܘܐ ܥܠܘܗܝ ܫܐܕܐ‎ ‘on whom was a devil’ Lk viii 27 SC. What was meant by this is evident from the picturesque

expression ܕܪܟܝܒ ܠܗ ܫܐܕܐ‎ } whom a devil was riding’ (or, as we say, ‘ devil-ridden’), found in Matt ix 32 S for dapomdpevor!.

1 This use of ܥܠ‎ is also found in the Palestinian Syriac Lectionary, e.g. ܫܝܕ ܥܠܝܟ‎ Joh vii 20,

62 Grammar and Syntax.

[Noldeke § 251.) In Lk x 30 S agrees with C’ in translating Apwbarns by ss\ dura dus ‘between dead and alive. The Peshitta is widely different. In Lk xvii 11 the addition of ܠܐܝ ܪܝܚܘ‎ by C to the phrase ܫܡܖ̈ܝܐ ܠܓܠܝ̈ܠܝܐ‎ dus is not found in S, and it probably means ‘to Jericho,’ It cannot therefore be brought forward as a parallel to ܠܪܘܚܐ‎ tal ܐܒܐ‎ dus Ephr. Overbeck 147 ult., qaoted by Néldeke : see further the Note on Lk xvii 11. |

[Noldeke § 263.] The tenses are employed normally in S and C, and call for no special remark here except as regards the Pluperfect, ie. the Perfect followed by ,ܗܘܐ‎ In my translation I have ventured uniformly to translate these Syriac Perfects with ܗܘܐ‎ by the English Pluperfect, in spite of the occasional harshness, as I believe the reader will thereby be better able to seize the point of view taken by the Syriac narrator in telling his tale. In a plain historical narrative we find in Syriac a series of verbs in the Perfect, varied occasionally by Perfects followed by ܗܘܐ‎ or ܗܘܘ‎ as the case may be. These latter Perfects with ܗܘܐ‎ often occur in positions that obviously require us to use a Pluperfect in translation, but sometimes it is not so obvious and most translators then simply leave the ܗܘܐ‎ untranslated. But to do this obliterates the march of the action as conceived by the Syriac mind. If I am right, we may regard a Syriac narrative as a series of tableaux vivants. The simple Perfects describe the action, the movement, which we are invited to witness; the Perfects with ܗܘܐ‎ , on the other hand, describe the anterior action, the actions which we are not actually supposed ourselves to observe, but which have brought the dramatis personae into the required situation.

Thus in ©, Mark’s story of the cursing of the barren Fig-tree and the cleansing of the Temple, as told by S, the narrator wishes to invite us to see and hear the following actions. Our Lord hungers, sees a fig-tree, comes to it, finds nothing but leaves, utters a curse on it. The curtain then lifts on another scene: He begins to put out the buyers and sellers in the Temple, and during some time we see Him stopping the passengers, teaching and saying ‘My House is a House of Prayer.’ Then again the disciples see the fig-tree withered, and 8. Peter says ‘The fig-tree is withered,’ and Jesus answers ‘ Have faith in God.’ Then in another scene we see the chief Priests come to Jesus, and they ask for His authority and He answers them (S. Mark xi 12-29).

Pluperfect, Historie Present. 63

Here we have four scenes, two on the way and two in the Temple. The events which we are supposed to see and hear are told us in the simple Perfect. But the connecting links, the subsidiary, though necessary, actions that bring the actors into the required situations, are told us in the Pluperfect. How is it that Peter remembered (v. 21)? Because the disciples had heard ܫܡܥܘ ܗܘܘ)‎ v. 14°), How is it that we find our Lord busy with the merchants in the Temple ? Because He had entered the Temple (Kam ܥܠ‎ +. 15"). How is it that in the fourth scene our Lord is found in controversy with the chief priests? Because He and the disciples had come again to Jeru- salem (aam ܐܬܘ‎ v. 27°), and the chief priests had heard of His doings in the Temple (aam ܫܡܥܘ‎ v. 18%).

The point is, that this tense describes a past scene. It may break the thread of the narrative to bring in a detail, but it does not carry the narrative forward. Wherever it appears there is a break of continuity!. A good example of this is Lk x 17 where S and C have ܗܘܘ‎ asada, suggesting the break which is logically demanded after v. 16. The previous verses give one scene, containing our Lord’s parting instructions to the Seventy-two: the following verses describe what was said when the Seventy-two had returned*. It is a question of pictorial effect, of the subordination of phrase. To neglect this . subordination turns a Syriac narrative into a monotonous chain of statements and takes the life out of the action.

Naturally the proper grouping and subordination of the incidents in a story is a matter of individual taste, at least to some extent. We therefore find that editors often inserted or cut out the ܗܘܐ‎ or aam. A series of instances will be found in Lk vii 19 ff, where S three times has a simple Perfect when C has the Pluperfect : Matt xxi 46 is another example of the same variation.

[ Noldeke § 274.] The ‘ historic present’ is rare in Syriac, but several clear instances occur in S: e.g. Matt xx 11 when the Labourers saw, they murmur (galt, Gr. éysyyvlov); Matt xxiv 1 when...the disciples drew near, they shew Him the buildings (passa, Gr. émidetEar) ; Matt xxvii 19 Pilate’s wife sendeth word to him ,ܫܠܚܐ)‎ Gr. daéoredev) ;

1 It is, in fact, the exact opposite of the Arabic ܩܢ‎ or the Hebrew strong -}. 2 A similar break in narration is to be,found in Lk i 62 5S,

64 Grammar and Syntax.

Mk vi 5 on a few infirm folk He /ayeth His hands and they were healed ܣܗܐܡ)‎ , Gr. ܕ( :»݀ܧ‎ [...€0epdmevoev]); Lk viii 4, for ovmévros S has ܟܕ ܡܬܟܢܫ‎ , Chas ܐܬܬܢܫ‎ aa, syr.vg has .ܦܕ... ܟܢܫ ܗܘܐ‎

These examples, in all of which the tense used is clear from the consonantal writing, raise the question whether we ought not sometimes to point verbs as Participles rather than Perfects in cases where the consonantal writing does not distinguish between them. The point which distinguishes Mya killing from ܩܛܠ‎ he killed is never found in S or C, so that we are entirely dependent on analogy and such tradition as is afforded us by the transmitted vocalisation of the Peshitta’.

[Noéldeke § 275.| Circumstantial clauses are expressed in Syriac by the Participle preceded by ܦܕ‎ or some other particle such as as, or by a relative. The Participle, or participially used adjective (§ 244), is rarely allowed to stand alone, except after Imperatives (§ 272). Thus in Matt xi 18 (John the Baptist came nezther eating nor drinking) Shas whe ܠܐ ܐܟܠ ܘܠܐ‎ , But this is altered in the other texts: Chas ܟܕ ܠܐ ܐܟܠ ܘܐܦ. ܠܐ ܫܬܐ‎ and syr.vg has whe ܕܠܐ ܐܟܠ ܘܠܐ‎ In the parallel passage, however, Lk vil 33, all three texts agree with that of S in 8. Matthew.

A somewhat similar instance is Lk xviii 11, where S has ‘That Pharisee standeth by himself praying...’ ,)ܩܐܡ...... ܡܨܠ .̈ܐ)‎ but C and syr.vg have ‘was standing......and thus was praying’ (ram ܗܘܐ... ܘܗܠܝܢ ܡܨܠܐ‎ yrs). In this way the historic present and the independent participle of SS are both made to disappear.

After Imperatives and some other expressions, such as ܡܥܕ ܗܘܐ‎ ‘he was accustomed,’ we find the bare Participle used, as is noted in Noldeke § 272. Thus Lk xii 13 ‘Speak to my brother to divide (uepioacOar) the inheritance with me’ is rendered in syr.vg, as well as SC, lit. ‘Speak to my brother dividing with me the inheritance.’ It is this construction which I believe to be intended in Matt xv 26 S,

ܠܐ ܘܠܐ ܠܡܣܒ ܠܚܡܐ ܕܒܢܝܐ ܪܡܝܢ ܠܟܠܒܐ It is not fitting [for /0/0[ to take the sons’ bread to cast it to the dogs.‏

1 For the inconsistencies exhibited by the Peshitta text in the phrase ‘answered and said, see the Appendix at the end of this chapter.

Participle, Infinitive. 65

Here instead of ܪܡܝܢ‎ rdmén we find in C and syr.vg ܘܠܡܪܡܝܘ‎

‘and to cast it,’ in accordance with cat Badeiv in the Greek. But the construction of S sounds to me idiomatic and original, although the antecedent to ܪܡܝܢ‎ has to be wholly inferred from the context}.

[Néldeke § 286.] The Infinitive is used, as Dr Néldeke says ‘als eine Art Epexegese,’ e.g. Matt ii 20 ܒܥܝܢ ܗܘܘ ܢܦܫܗ ܕܛܠܝܐ ܠܡܥܕܝܘ‎ ‘they were seeking the lad’s life to snatch away,’ where ‘to snatch away’ is omitted by S. Here ܠܡܥܕܝܘܬܗܿ‎ ‘to snatch it away’ would have been possible, but with transitive verbs the addition of the suffix is not necessary. When however the verb requires after it a construction with a preposition the suffix is necessary after the pronoun, e.g. in

Ps civ 26 39 pnd myo innd where we in English can say ‘Leviathan, that Thou hast formed to laugh at, the Syriac like the Hebrew must say ‘ Leviathan, that Thou hast formed to laugh at zt.’

This will explain the phrase ܠܡܠܥܣ ܒܗ‎ in Joh xxii S. Verbs of

eating, such as ܐܠܠ‎ and ws\, usually govern an accusative; followed

1 Dr Merx, in his always interesting notes on the text of S (Dee Vier Kanonischen Evangelien: .. Erliiuterungen..von Adalbert Merx, i 248 ff), takes a widely different view. Deliberately dis- regarding the Greek, and even the text of S in the parallel passage Mk vii 27, he considers ܪܡܝܢ‎ to refer to ~4si=s and makes ܪ‎ a relative: the saying of Christ thus becomes ‘ Js zt not fitting to take the bread that the sons cast to the dogs?’ i.e. ‘is it not fitting that I, cast out as I am by the Jews, should help the Gentiles?’ To this question the woman replies by an eager affirmative. According to Dr Merx, the ordinary text of S. Matthew and also the parallel passage in S. Mark have been corrupted by a Judaistic re-editing ( Verjtidelung), which S alone has escaped.

It is undoubtedly much easier to construe § in the way advocated by Dr Merx, but I cannot believe that his translation gives the sense intended by the scribe. When I find rév dprov rav réxvev in the Greek and ܠܥܢܡܐ ܕܒ̈ܢܝܐ‎ in the Syriac translation, I cannot but believe that ܪ‎ denotes the genitive and that the phrase means ‘the bread of the sons.’ Moreover Aphraates 149, in an allusion which I omitted to quote in voli, pp. 88, 89, as being too paraphrastical for textual purposes, says that those who assiduously beg for mercy are the dogs that receive the sons’ bread and they cast to them ܘܪܡܝܢ ܠܗܘܢ)‎ usa ܠܚܡܐ‎ ems), Dr Merx wishes to emend this also and to cut out the a before ,ܪܡܝܢ‎ but as it stands it attests the expression the sons’ bread, and a very little imagination is needed to believe that the phrase in Aphraates is a somewhat mechanical reminiscence of the text of S, understood as I have understood it and not as Dr Merx has done.

If it be necessary to choose an antecedent to ܪܡܝܢ‎ in S, I should be inclined mentally to supply ܠܢ‎ after ,ܘܠܐ‎ Similarly in Mk vii 27 the Palestinian Lectionary has Jt ¥ not good that we should take the sons’ bread and (that) WE should cast it to the dogs. But no word is really required, seeing that in Joh xviii 8 S renders apere rovrovs trayew by ܫܒܘܩܘ ܐܙܠܝܢ‎

ܠܘܝܬܢ ܗܢܐ ܕܒܪܝܬ ܠܡܓܚܟ ܥܠܘܗܝ 2

B. I, 9

66 Grammar and Syntar.

by ܡܢ‎ they signify ‘to eat part of a thing.’ But ‘to eat of a joint’ is ,ܐܟܠ ܒܗ‎ The expression is fairly common in Hebrew, e.g. Judg xiii 16, but it also occurs in Syriac, e.g. Job xxi 25 in imitation of the Hebrew. Hence ܐܝܬ ܠܟܘܢ ܡܕܡ ܠܡܠܥܣ ܒܗ‎ means ‘ Have ye anything to eat of?’ The choice of the preposition to be used was no doubt due to the fact that the Greek is wy te tpoodaycov € ;

The confused construction of Lk ii 8 °C, is not supported by S, which has fais ܢܩܝܡ‎ arta ܡܫܟܚ ܐܠܗܐ ܕܡܢ ܗܠܝܢ‎ where ¢ has ܠܡܩܡܘ‎ for mass. The text of C (noticed in Néldeke § 286) appears to be nothing more than an unskilful mixture of the phraseology of Matt ui 9 with that of S in 8. Luke.

A good example of the Infinitive used without a finite verb to express ‘must’ is Joh ix 30, where S has ܒܗܕܐ ܠܡܬܕܡܪܘ ܒܗ‎ ‘this is something to wonder at!’ The Peshitta inserts am after ܒܗܕܐ‎ and omits eas.

[ Noldeke § 290.| A noteworthy example of a double accusative is to be found in ܘܠܢܦܫܟܝ ܕܝܠܟܝ ܬܥܒܪܝܢܗ ܪܘܡܚܐ‎ Lk ii 35 S, cor- responding to kat cod [de] adrys thy Wuyny duelevoerar popdaia. The text is supported by a fragment of the original Syriac of 8. Ephraim’s Commentary on the Diatessaron preserved by Isho‘dad, who has .ܒܢܦܫܟܝ ܬܥܒܪܝܢܗ̇ ܪܘܡܚܐ‎ The meaning is apparently ‘And through thine own self thou shalt cause a spear to pass,’ but no other authority has the verb in the 2nd person. For the use of tax. with an accusa- tive instead of with ܒ‎ see Lk xix 1, where we find ܐܝܪܺܝܚܚܕܘ‎ taxa in S, instead of ܒܐܝܪܝܚܘ‎ Jana.

[Noldeke § 295.] The Infinitive absolute is much more commonly used in the Evangelion da-Mepharreshe than in the Peshitta. ܡܫܢܐ ܫܢܐ‎ Joh x 20 (Gr. paivera) appears to be the only instance in the Peshitta Gospels where the idiom is not actually indicated in the Greek. In one instance, ܡܗܝܡܢܘ ܗܝܡܢ‎ Lk viii 50, where it occurs in C but not in S or the Peshitta, the reading of C is supported by Aphraates, by the Commentary of S. Ephraim, and by the Acts of Thomas. It is possible, however, that this last quotation may be based on Mk v 36, a passage for which S is unfortunately not extant.

Syntax of Particles. 67

[Noldeke § 304.] In Lk xviii 3 S has ܐܪܡܠܬܐ ܚܕܐ ܐܝܬ ܗܘܐ‎ where C' and the Peshitta have ܐܝܬ ܗܘܬ‎

[Noldeke § 328 B.] Both S and C are among the ‘ancient docu- ments’ that invariably use ܠܐ ܗܘܐ‎ and not al. The Peshitta on the other hand contains ܠܘ‎ several times, e.g. Joh vii 25.

[§ 828 F.] Besides the use of ܕܠܐ‎ for ‘without,’ and also as a conjunction meaning ‘lest’ (almost like ,([ܕܠܡܐ‎ it is found several times in the Evangelion da-Mepharreshe in the sense of ‘else’ or ‘otherwise’: in such cases it is always followed by a noun, so as to distinguish it from ܕܠܐ‎ meaning ‘lest.’ This usage is obscured in the Peshitta and does not seem to be recognised in the Syriac Grammars, so I here set down the instances I have observed®. Matt vil SC: ‘Do not your righteousness before men, else ye have no reward with your Father’ ܐܓܪܐ ܠܝܬ ܠܟܘܢ ܠܘܬ ܐܒܘܟܘܢ)‎ win); Matt vi 24 C (hiat S), Lk xvi 18 S (hiat C): ‘No man can serve two lords, else the one he will hate and the other he will love’ ܕܠܐ ܠܚܕ ܢܣܢܐ)‎ mets ;(ܘܠܐܚܪܢܐ‎ Matt ix 16, 17, Mk ii 21, 22 S (hiat C): ‘No man putteth a new patch on a worn-out garment, else the fulness of the new pulleth away the weakness of the worn-out part...... neither new wine into worn-out wine-skins, else the wine teareth the wine-skins’ ܕܠܐ ܡܠܝܘܬܗ ܕܚܕܬܐ ܡܢܬܦܐ... . ܕܠܐ ܚܡܪܐ ܡܨܪܐ ܠܗܝܢ ܠܙ̈ܩܐ)‎ . In Matt vi 24 and Lk xvi 13 the corresponding Greek is 7 ydp, in the other instances it is €! d€ wy or ei Oe prjye.

The Peshitta has ܘܐܠܐ‎ instead of ܕܠܐ‎ in Matt vi 1, and ܐܘ ܓܝܪ‎ in Matt vi 24 and Lk xvi 18. In Matt ix 16, 17, and Mk ii 21, 22, it has ܕܠܐ‎ but except in Mk ii 22 the construction is changed. In the other passages the verb comes immediately after ,ܕܠܐ‎ and in Matt ix 16 this is actually put in the future ܕܠܐ ܬܬܘܦ ܡܠܝܘܬܗܿ...)‎ ‘that the fulness of it may not pull,’ etc.). We may remark in passing that the unusual divergence of the Peshitta from the Greek in this verse receives a natural explanation when viewed as a stylistic cor- rection of the Evangelion da-Mepharreshe. The Greek here has aipe. yap TO mhypopa x.7.d., not € Sé py, aiper 70 TAypwpa as in 8. Mark. The Evangelion da-Mepharreshe rendered both passages

1 Here am ܐܝܬ‎ is read by Mr Gwilliam’s cod. 40, and by his 14 36 in Lk vii 37. 2 Other instances of ܕܠܐ‎ meaning ‘else,’ are to be found in asp 47}, 6817, 6916 ; Aphraates 1852, (R.H.K.)

68 Grammar and Syntax.

alike, as its custom often is, employing in each passage the idiomatic use of ܕܠܐ‎ now under consideration. The Peshitta alters the idiom in Matt ix 16, but in so doing departs much further from the Greek. It is of course quite likely that in this particular phrase the official Peshitta text is simply reproducing a previous stylistic correction : we are even at liberty to conjecture that C, which is here missing,

itself read .ܕܠܐ ܬܬܘܦ‎

[Noldeke § 338c.] In more than a dozen passages the Peshitta begins a paragraph with ܕܟܕ‎ rama ‘And it came to pass that when...,’ corresponding to kai éyévero ove in Matt, and Kat éyévero as or éyévero 6€ &v t@ in Lk. The formula occurs in S in Matt xi 1 and xix 1, and perhaps also in Matt xxvi 1, but elsewhere it is avoided in the Evangelion da-Mepharreshe. Usually the éyévero is left un- translated, as in Matt xiii 53 and Lk x 38: in other places we find ܟܕ‎ rama ‘And it came to pass when..., eg. Matt xi 1 C, xix 1 C, Lk i 41 S, ix 51 C. In the last passage S has saa without ܗܘܐ‎ . It is possible that the original translation had regarded sax kama as an exact equivalent of kai éyévero ore and aa ܘܗܘܐ‎ as an exact equivalent of kat éyévero ws. But as above remarked the éyévero is usually dropped in the Syriac rendering.

The idiom specially mentioned in § 338¢ (wiz. ‘And it came to pass...and’) occurs in Lk ix 28 SC, but in the Peshitta the intrusive and has been corrected out.

Here may conveniently be noticed the very curious anacoluthon introduced by the Kvangelion da-Mepharreshe and the Peshitta into their rendering of 8. Matthew’s phrase ‘Now all this is come to pass that it might be fulfilled,’ ete. (robro dé [ddov] yéyovey wa x.t.d.). The phrase occurs in Matt 1 22, xxi 4, xxvi 56, and in the Greek the construction is perfectly straightforward. But the Syriac has ܗܕܐ ܕܝܢ‎ x ܕܗܘܬ‎ , just as if the translation had read 0 instead of édov. That it is a native idiom and not a translator’s mistake is shewn by the retention of the phrase in the Peshitta: even in Matt xxvi 56, where S has in the plural ܕܝܢ ܟܠܗܝܢ ܕܗܘܝ ܕ‎ aloo, the Peshitta has ܗܕܐ ܕܝܢ ܕܗܘܬ ܕ‎ . In Matt xxi 4 a few ancient Peshitta codices

1 <Z\n meaning ‘else’ also occurs in Lk x 6, where reads ܐܢ ܕܠܐ ܥܠܝܟܒܘܢ ܢܗܦܘܟ‎

(sic, see the List of Errata), i.e. Lf (tt be) otherwise, upon you it will return.

Apodosis introduced by the Copula 69

(Mr Gwilliam’s 15 17 19 20 36) remove this anacoluthon by omitting the ܕ‎ before ܗܘܬ‎ 1. This construction is also noticed in Néldeke § 358 B.

[Noldeke § 339.] In the ordinary Edessene Syriac, as known to us in writings dating from the 4th century onward, the conjunction ‘and’ is not used to introduce the apodosis. But in S and C' there are several instances of this thoroughly Semitic idiom. The passages may be conveniently arranged under the two heads of Temporal Sentences, containing xa in the protasis, and Conditional Sentences, containing ܐܢ‎ 6 ale in the protasis.

(A) Temporal Sentences.

Matt ii 16 Banriobets S€ 6 “Inaods ¢ܐ0ܗ‎ avéBn amo Tov vdaTos Kal idov...

SyI.vg‏ ܦܕ ܥܡܕ ܕܝܢ ܝܫܘܥ ܡܚܕܐ ܣܠܩ ܡܢ ܡ̈ܝܐ ܘܐܬܦܬܚܘ ܠܗ ܫܡܝܐ. Now when Jesus was baptized, immediately He came up from the water‏ and the heavens were opened to Him.‏ but‏ © ܘܟܕ ܥܡܕ. ܒܗ̇ ܒܫܥܬܐ ܕܣܠܩ ܡܢ ܡ̈ܝܐ. ܘܗܐ ܐܬܦܬܚܘ ܫܡܝܐ. And when He was baptized, in the same hour that He came up from‏ the water, [and] lo, the heavens were opened.‏

S‏ ܘܟܕ ܥܡܕ ܘܣܠܩ ܡܢ ܡܝܐ ܘܗܐ ܐܬܦܬܚܘ ܫܡܝܐ And when He was baptized and came up from the water, [and] lo,‏ the heavens were opened.‏

The [and] which I have put in square brackets simply serves to introduce the apodosis. Notwithstanding important differences, S and

@ agree in the general cast of the sentence against the Greek and against the Peshitta, which follows the Greek.

Lk. xiii 10ff The Greek text tells the story of the woman with the spirit of infirmity in the following manner: “'Now He was teaching... Mand lo, a woman having a spirit... 1Now Jesus... (spoke) land laid His hands on her.... ‘*Now the ruler of the synagogue answered and...(complained). 1° But (8€) the Lord answered him and said ‘ Hypocrite ! 7

1 A similar anacoluthon may be intended by the insertion of a before ahs in Lk xxii 37 S,

but I incline to think the sa mere scribe’s blunder. A similar construction after ܦܠܗ ܦ‎ is found in Addai 234, 443, as my friend Canon Kennett points out.

70 rrammar and Syntax.

S and C without substantial variation have: “'And when He was teaching... “and there was there a woman that had a spirit... Rand Jesus...(spoke) and laid His hands on her.... ‘And the ruler of the synagogue answered and...(complained). Jesus answered and said to him: ‘ Respecter of persons ! ` 7

It is surely unreasonable to take all the clauses in vv. 10—14 inclusive as introductory to the reply of our Lord in v. 15. Is it not more natural to regard v. 11 as the apodosis to v. 10? When Jesus was teaching in a synagogue one Sabbath, a certain woman (we are told) was present.

The Peshitta retains the ‘when’ of SC in v. 10, but omits the introductory ‘and’ in v. 11, so that it reads “Now when Jesus was teaching... "there was there a woman, etc.”; 8 new sentence begins at v. 12 with “Now Jesus saw her.” Evidently therefore the a in the Hvangelion da-Mepharreshe at the beginning of v. 11 was taken as the introduction to the apodosis, and is accordingly omitted in the Peshitta in agreement with the later theory of Syriac syntax.

Joh iv 1—8. This is another case where the Evangelion da- Mepharreshe had ܠܦ‎ in the protasis and a in the apodosis, and where the Peshitta in getting rid of this antiquated construction departed still more widely from the Greek. The Greek has ws ody ܘܗ‎ in v. 1, followed by a rather long parenthesis, but v. 3 takes up the principal narrative, beginning a¢nxev rHv “lovdaiay. This is properly and idiomatically rendered in S by ܟܕ ܕܝܢ ܝܕܥ‎ followed in ®, 3 by ܘܫܒܩܗ ܠܝܗܘܕ‎ . C is here mutilated, but from the space it 1s evident that v. 1 began with sa as in S, and v. 3 with ܫܒܩܗ‎ or ܘܫܒܩܗܿ‎ The Peshitta has manza like S, but it omits sa at the beginning of v. 1. The fact that the Peshitta has the a at the beginning of v. 3 is a clear indication that its insertion is no mere peculiarity of S but a characteristic reading of the Hvangelion da-Mepharreshe.

Other instances of ܠܕ‎ followed by a are Matt ix 18 S, Matt xii 9 C, (not S), xviii 8, 9 S ܘܬܐܙܠ)‎ sic :ܨܐ‎ not C); Mk x 46 S, xi 15 S (tna, cf syr.vg); Lk xix 1, 2 S (not C), xix 36 S (not C).

Somewhat similar is Joh xvi 6. Here S has ܡܛܠ ܕܐܡܪܬ ܠܟܘܢ ܓܝܪ ܗܠܝܢ ܘܐܬܬ ܟܪܝܘܬܐ ܘܡܠܬ ܠܒܘܬܟܘܢ‎ For because I have said to you these things [and] sorrow hath come and filled your hearts.

Apodosis introduced by the Copula. 71

Here again it is obvious that we must omit the [and] in translating into English. The Peshitta also has ܘܐܬܬ‎ , but it omits ܡܛܠ ܕ‎ although the Greek is )ܬܘ‎ dru radra ܬܡܐ ܬ ܐܘܬ‎ dulv, Wan wemdjpoKev UPOV THY Kapdtlav.

(B) Conditional Sentences.

Lk xii 45, 406. édv 8€ etary 6 Soddos Exetvos ev TH Kapdia adTov...Kal ܫ‎ , A ~ 467 ܕ ܟ ܘ ܣ , ܨ‎ , apEnta TUTTEW TOS Tmatoas K.T.r., HEEL 6 KbpLos TOD SovAoV Exetvov...

Here S and C have ܥܒܕܐ ܒܠܒܗ ..... ܘܢܫܪܐ ܠܡܡܝܝܐ ܠܥܒ̈ܕܐ...‎ am tac ܐܢ ܕܝܢ‎

1.00 ܘܢܝܐܬܐ ܡܪܶܗ ܕܥܒܕܐ‎ Now if that slave shall say in his heart...and shall begin to beat the slaves... *®[and] the lord of that slave will come...

The apodosis is thus clearly introduced by a, which we have to leave untranslated, or render by ‘then.’ The Peshitta agrees in the rendering of these verses with SC, but it omits the characteristic a before ܢܐܬܐ‎ at the beginning of v. 46, in accordance with the accepted syntactical theory.

The evidence of Lk xii 45, 46, is especially cogent, because S and C' agree in inserting the ܘ‎ and there is no variation in the Greek. In other instances we have only the evidence of single ss.

Matt xviii 12 8

ܐܢ ܢܗܘܝܢ ܠܓܒܪܐ ܡܐܐ ܥܢܐ ܘܬܛܥܐ ܚܕܐ ܡܢܗܝܢ ܘܠܐ par‏

ܬܫܥܝܢ ܘܬܫܥ

Tf a man shall have a flock of one hundred and one of them be strayed,

doth he not leave the ninety and nine...? (lit. ‘and doth he not leave...?’)

Here C and the Peshitta have another word for sheep, and they also omit the a before ܠܐ‎

Matt xx 28 fin. C ܗܢ ܕܝ ܬܗܬܡܟ ܒܕܘܟܬܐ ܒܨܝܪܬܐ. ܘܢܐܬܐ ܡܢ ܕܒܨܝܪ ܡܢܟ.‎ armas ܘܢܐܡ ܡܪܐ ܐܚܫܡܝܬܐ ܐܬܩܪܒ ܘܐܬܥ ܠܐ‎ ܣܡ̈ܝܟܐ.‎ psd Witham ܘܬܗܘܐ ܬܫܒܘܚܬܐ‎

But if thou sit down to meat in a lesser place, and there come one less than thou, and the lord of the supper say to thee ‘ Bring thyself and come up and sit down to meat, then thou shalt have more glory in the eyes of the guests,

72 Grammar and Syntax.

This sentence occurs in the well-known interpolation attested by Codex Bezae and the Old Latin version, but not by current Greek texts or the Peshitta. S is here deficient, owing to the accidental loss of a leaf, but it is evident from the space required that it could never have contained this long insertion. I have made the apodosis in the above translation come at the last clause: it might equally well be put at ܘܢܐܬܐ‎ or ܘܢܐ ܡܪ‎ , and D and the Latins actually set it at ܘܢܐܡܪ‎ , reading éped and not ܬܘܐ‎ épet. But in any case the clause which contains the apodosis begins with ‘and.’ As the evidence of S shews us that the interpolation does not belong to the earliest form of the Evangelion da-Mepharreshe, it is all the more interesting, from the point of view of the history of the language, to find in it an instance of the idiom now under discussion.

Matt xxi 21 8‏ ܐܢ ܬܐܡܪܘܢ ܠܛܘܪܐ ܗܢܐ ܕܐܫܬܩܠ ܘܦܠ ܒܝܡܐ mama‏ ܗܟܢܐ.

If ye shall say to this hill ‘ Be taken wp and fall into the sea, then it shall be so.

Here ¢ has ܢܗܘܐ ܠܟܘܢ ܗܟܢܐ‎ without a. The Peshitta has

but curiously enough three‏ , ܐܦܢ ܠܛܘܪܐ ܗܢܐ ܬܐܡܪܘܢ.... ܬܗܘܐ

of Mr Gwilliam’s Mss (15 20* 36) read ܘܬܗܘܐ‎

Lk ×! 8 8‏ ܐܢ ܡܛܠ ܪ̈ܚܡܘܬܐ ܠܐ ܢܬܠ ܠܗ ܘܡܛܠ ܚܨܝܦܘܬܐ ܢܩܘܡ ܘܢܬܠ wal‏

Though because of friendship he will not give to him, ‘yet because of umportunity he will rise and give to ham.

The a before the second ܡܛܠ‎ is not in Cand the Peshitta, and it is not visible in the photograph of S, though there is room for it and the rest of the passage is quite clear. I have not seen the passage myself at Sinai, but I should not wonder if a careful examina- tion shewed not only the a read by Dr Harris in 1893, but also signs that the letter had been intentionally washed out before the whole ms was broken up and converted into a palimpsest. In such cases the washed out letters are sometimes illegible in the photograph : a good instance of this is to be found in Lk xii 31, where I have little doubt that the first hand of S* had Massa, not ܗܟܝܠ‎ ,

Apodosis introduced by the Copula. 73

The illegibility of the a in Lk xi 8 is all the more to be regretted, as the sentence is of a somewhat different type to those we have been examining. The ܐܢ‎ has a concessive force, ie. ‘ though’ or ‘even if,’ and in the apodosis the first word is not a verb, but a noun preceded by a preposition.

Lk xviii 4,5 8

ܐܢ ܡܢ ܐܠܗܐ ܠܐ ܕܚܠ ܐܢܐ ܘܡܢ ܒܪ ܐܢܫܐ ܠܐ ܡܬܟܚܕ ܐܢܐ

«sma‏ ܐܪܡܠܬܐ ܕܗܟܢܐ ܡܠܐܝܐ ܠܝ ܐܬܒܥܝܗ

Though of God I am not afraid and for man I have no reverence, * yet this widow who thus fatigueth me I wall requite.

C has ܒܟܠܙܒܢ‎ instead of ܕܗܟܦܢܐ‎ ; the relative thus being dropped, ܘܗܕܐ‎ now begins a third conjunctive clause of the protasis, and the apodosis consists of the single word ܐܬܒܥܝܗ‎ ‘1 will requite her.’ For [ܘ]ܗܕܐ.. .. ܠܝ‎ the Peshitta has

ܐܦܢ ܡܛܠ ܕܡܠܐܝܐ ܠܝ ܗܕܐ ܐܪܡܠܬܐ

so that the sentence runs ‘ Though of God I am not afraid and of men I have no reverence, even though it 1s because this widow fatigueth me, I will requite her. This contains an admirable rendering of dud ye TO Tapexew por ܗܘܐ‎ THY xypav TavTyY, but it will be noticed that in this rendering also the a before the apodosis has disappeared.

Lk ix 58 8‏ ܐܢ ܠܬܥܠܐ ܢܩܥܐ ܐܝܬ ܠܗܘܢ ܘܠܦܪܚܬܐ ܕܫܡܝܐ as‏ ܘܠܒܪܗ ܕܐܢܫܐ ܠܝܬ ܠܗ ܐܬܪ ܝ ܪܫܗ

Though the foxes have dens and the birds of the heaven have nests, yet the Son of Man hath not where He may rest His head.

In agreement with the Greek, Cand the Peshitta omit ܐܢ‎ and read ܠܒܪܗ ܕܝܢ‎ instead of mzala, besides one or two minor variations.

The word ܐܝܢ‎ ‘yea’ is written ܐܢ‎ in Lk xii 5 S, but there can be little doubt that ܐܢ‎ here stands in S for ‘if’ or ‘though!’ An ‘if’ ig more than once inserted in S where the construction seemed to require it, e.g. Matt xii 28, Mk xii 87. But the half dozen instances

which I have given of conditional sentences, in which the apodosis

1 So also Dr Merx in his translation, p. 133. B. IL. 10

74 Grammar and Syntax.

is introduced by a, will I trust sufficiently prove that the idiom was really used in the earlier stages of Syriac literature.

Now and then the apodosis of conditional sentences, especially such as contain dv in the Greek, are introduced by .ܕܝܢ‎ The classical instance is Gen xlii 10 syr.vg

ܐܠܘ ܠܐ ܐܫܬܘܚܪܢ ܟܒܪ ܕܝܢ ܗܦܟܢ.

Had we not delayed, perchance we should have already returned. ܕܝܢ‎ tas here corresponds to Any 3.

The same construction occurs in Matt xi 21, Lk x 13, in S Cand the Peshitta; ‘if the mighty works had been done in Tyre and Sidon per- chance they would have already repented.’ yaa implies a slight doubt: the speaker is morally certain, but it remains a hypothesis‘: When jaa is absent, ܕܝܢ‎ can equally well introduce the conclusion without any expression of doubt. Thus in Lk xix 23 S C, and apparently also in Matt xxv 27, the lord asks the lazy slave ‘Why didst thou not give my money to the bankers ?’—equivalent to a conditional sentence —‘and I then ) (ܘܐܢܐ ܕܝܢ‎ had come and required mine own.’ 6 Peshitta both in Matt and Lk omits ,ܕܝܢ‎ as we might have expected.

But the clearest example is Matt xvi 20 S, where we read

ܐܠܘ ܐܝܬ ham‏ ܒܟܘܢ ܗܝܡܢܘܬܐ ...ܿ ܐܡܪܬܘܢ ܕܝܢ ܠܛܘܪܐ ܗܢܐ ܕܐܫܬܢܐ ܘܢܫܢܐ.

If there had been in you faith...ye would have said to this hill ‘Be removed, and it would remove.

Here C'and the Peshitta read ܕ]ܬܗܘܐ‎ acm] ܐܢ‎ at the beginning and substitute ܬܐܡܪܘܢ‎ for ܐܡܪܬܘܢ ܕܝܢ‎ .

[Noldeke § 349 B.] The Greek construction of syr.vg and C in Lk ix 4 (aam ܬܡܢ‎ [ml] ܐܢܬܘܢ‎ alsa ܐܝܢܐ‎ wdual) is found also in S, and the same phrase recurs in Matt x 11, Mk vi 10, Lk x 5, 8, 10. But such a construction would hardly be possible in S and € with any preposition but \, which in Syriac is so intimately connected with the mere sign of the accusative after a transitive verb. When another preposition is used in sentences of this kind the Syriac begins with a ‘nominative absolute,’ e.g. Lk xx 18 éf’ dv dv méon ܐܬܬܬ‎ ܘܐܘ‎ is rendered in S ¢ ܡܢ ܕܬܦܠ ܥܠܘܗܝ ܬܫܚܩܝܘܗܝ‎ As, and

± gas ܡܢ‎ would have expressed the purely temporal sense of ‘already.’

Various dependent Clauses. 75

similarly in syr.vg with the substitution of ܬܕܪܝܘܗܝ‎ for the last word. Only in the Harclean do we find ܕܬܦܠ ܬܕܪܝܘܗܝ‎ ac As.

In view of the concession to Greek idiom made by the Evangelion da-Mepharreshe in Matt x 11, it is a remarkable circumstance that in D and that excellent minuscule cod. 28 the clause runs 7 moAus eis Nv av eich Onre eis ;]ܵܘ‎ «7.4. Why do Western authorities thus syriacize at a point where Syriac texts graecize ?

And in Mk vi 10, where the Greek is é7ov édv eiaédOyre eis otkiar, a phrase which reads like a translation from a Semitic original, why are S and syr.vg assimilated to the phraseology of the other Gospels ?

It is easier to ask these questions than to suggest a satisfactory solution.

[Noldeke § 355.] In my translation I have uniformly left the relative in short adjectival or adverbial phrases untranslated, wherever the verbal construction is left unexpressed. Thus ܐܝܠܝܢ ܕܒܫܡܝܐ ܐܢܝܢ‎ really means ‘the things that are in heaven,’ but ܐܒܘܢ ܕܒܫܡܝܐ‎ is better rendered into English by ‘our Father in heaven,’ than by the full verbal statement ‘our Father, who art (07, who is) in heaven".’ The mere fact that the relative in Syriac is a light unaccented half- syllable has doubtless helped the tendency of the language to insert it where the sense would be over-expressed by the English relative.

[Noldeke §3588B.] See above, on § 338 ¢

[Noldeke § 373.] In three passages ܕܠܡܐ‎ seems to stand ellip- tically at the beginning of a sentence. The use of ܠܡܐ‎ for ܠܐ‎ in prohibitions is not found in the Evangelion da-Mepharreshe, unless we regard Joh v 45 C as an exception, so it is improbable that the ܕ‎ in ܕܠܡܐ‎ stands for ‘inverted commas,’ like 67 recitativum. The instances are

(i) Matt vili 4 (Jesus saith to him “Opa pndevi etrys) ܐܡܪ ܠܗ... ܕܠܡܐ ܠܐܢܫ ܐܡܪ ܐܢܬ.‎ SC

where the Peshitta has ܚܙܝ ܠܡܐ‎ instead of ܕܠܡܐ‎ , ie. ‘See, do not say to any one. Thus in SC ܕܠܡܐ‎ corresponds to dpa and the negative contained in pmdevi.

1 Of course atne5 o> could not be used in Syriac for ‘our heavenly Father’ (of Matt xviii 10 ‘their-angels in-heaven-behold my-Father-in-heaven’).

76 Grammar and Syntax.

(ii) Matt xvii 9 (Jesus commanded them, saying Mydevi etayrte TO Gpapa) ܡܦܩܕ ܗܘܐ ܠܗܘܢ ܝܫܘܥ ܘܐܡܪ ܠܗܘܢ̇ ܕܠܡܐ ܠܐܢܫ ܐܡܪܝܢ‎ C

ܐܢܬܘܢ ܚܙܘܐ. has‏ ܕܠܡܐ... ܚܙܘܐ S is missing at this point. The Peshitta for‏ ܠܥܝܢ ܐܢܫ ܠܐ ܬܐܡܪܘܢ iim Wow‏

Here again ܕܠܡܐ‎ stands in C for ‘Beware lest.’

(ili) Matt xxv 9 (The wise virgins reply Myjore otk dpkéon ypiv Kal vty)

S‏ ܐܡܪ̈ܢ.... ܚܟܝܡ̈ܬܐ ܕܠܡܐ ܠܐ ܢܣܦܩ ܠܢ ܘܠܦ̈ܝܢ.

Here ܕܠܡܐ‎ practically stands for ܕܠܡܐ‎ A, ie. ‘Nay, lest....,’ but the omission of the direct negative at the beginning of the sentence both in Greek and in Syriac gives a more courteous turn to the refusal. The Peshitta substitutes ܠܡܐ‎ , Curiously enough, in Matt xiii 29, where the Greek has ov, pymore... and S C have ܠܐ ܕܠܡܐ‎ the Peshitta has ܕܠܡܐ‎ alone, like Sin Matt xxv 9.

It is noteworthy with what persistence the Peshitta avoids ܚܙܝ ܕܠܡܐ‎ and ܕܠܡܐ‎ ows. This is the case in Matt ix 30, xviii 10, xxiv 5, Mk i 44, Lk xxi 8, in all of which places the phrase is used

by the Evangelion da-Mepharreshe. The only exception I have noted is Mk xii 5 ܚܙܘ ܕܠܡܐ ܐܢܫ ܢܛܥܝܟܘܢ‎ , where S has ܠܡܐ‎ .

[ Noldeke §374 B.] ܐܢ ܗܘ ܕ‎ is generally avoided in S. Out of 15 places where the phrase occurs in one or other Syriac text, S has it only in Mk vii 3, 36. It occurs seven times in C; and its infrequency in S appears to be the result of stylistic correction. At least this is what is suggested by the occurrence of x ܐܢ‎ Matt x 18 S, am ܐܢ‎ (without x) Matt xii 10 S, and the reading ܗܘ ܬܟܝܠ ܗܘܐ‎ Matt xxvii 43 S. In the last mentioned verse the ordinary text has ܘܡ‎ alone, but € wéroev is read by D, 1-118-209, the Old Latin, the Egyptian versions, the Armenian and the Ethiopic: I venture to think it probable that the Evangelion da-Mepharreshe once read in agreement with these authorities

ram ܕܬܟܝܠ‎ am ܐܢ‎

and that am in S is a relic of this reading.

Hypothetical Clauses. 77

An expression similar in construction to 1 am ܐ ܢ‎ is to be found in Matt xiii 10, where S has ܡܢܐ ܗܘ ܕ‎ while ¢ and syr.vg have ܠܡܢܐ‎

[§ 374, Note at end.] The use of ܐܢ‎ to expres dy, common in later Syriac translations from the Greek, is naturally absent from the Evangelion da-Mepharreshe. But S preserves two instances of the use of ܐܢ‎ in alternatives )700/0606 § 3728), corresponding to I in Arabic’. Thus Mk xiii 35 S ܕܒܝܬܐ ܠܐ ܐܢ‎ mim ܠܐ ܓܝܪ ܝܕܥܝܢ ܐܢܬܘܢ ܐܡܬܝ ܐܬܐ‎ ܒܪܡܫܝܬܐ ܘܠܐ ܐܢ ܒܡܨܥܝܬܐ ܘܠܐ ܐܢ ܒܢܘܓܗܝܬܐ ܘܠܐ ܐܢ‎ For ye know not when the master of the house cometh, not whether

at be in the evening-time, nor whether in the midnight, nor whether in the morning twilight, nor whether at the dawn.

The Peshitta has ܐܘ‎ in each case in place of ,ܐܢ‎ to agree with the Greek 7.

Similarly in Matt xviii 8 S* has ‘2t as better for thee to enter life ܐܢܬ ܐܘ ܟܕ ܚܓܝܣ‎ Xe ܟܕ ܐܢ‎ whether lame or halt. The ܐܢ‎ has been apparently washed out of the text by a corrector and does not appear in C or the Peshitta, but the occurrence of the word in Mk xiii 35 inclines me to believe it genuine here also.

The same use of ܐܢ‎ is retained in the N.T. Peshitta outside the Gospels, e.g. in Rom i 16, where “Iovéatm te mp@tov Kat "EdAyve is rendered

ܐܢ ܡܢ ܝܗ̈ܘܕܝܐ ܠܘܩܕܡ ܘܐܢ ܡܢ ܐܪ̈ܡܝܐ.

[Néldeke §375 a.] The use of ܐܠܘ‎ in Sand 0:18 the same as in other Syriac documents, i.e. it introduces a hypothesis which is regarded by the speaker as impossible. Thus ܐܠܘ ܠܐ ܐܬܝܠܕ‎ is ‘if he had not been born. The use of alr, therefore, in the Hvangelion da- Mepharreshe for the sayings in Matt xvii 20 and Lk xvii 6 about faith as a grain of mustard-seed adds a tone of impatience and regret which

1 According to Wright ii § 166 the Arab grammarians distinguish between alternatives sepa- rated by ܩ‎ | and by .ܐܦ‎ In the former case one of the alternatives is known by the speaker to be

true, in the latter no knowledge is assumed.

78 Grammar and Syntax.

is lost in the Peshitta. ܐܠܘ‎ is practically our ‘if only’ :—‘if only,’ says Christ to the apostles, ‘ye had the grain of faith which ye have not, ye would have said....’

[§ 375 B.] Just like our ‘if only,’ the sense of ܐܠܘ‎ passes into that of a wish, and Lk xii 49 Cis worth quoting here as a parallel to the passage of the Julian Romance quoted by Dr Néldeke. In Julian 23” the Jews say ܘܟܡܐ ܨܒܝܢ ܗܘܝܬܘܢܿ. ܐܠܘ ܡܢ ܟܕܘ ܥܪܒܬ.‎ ie. “And

how much ye would have been pleased if only our star had set!’ This is in form exactly like ,ܘܡܐ ܨܒܐ ܐܢܐ ܐܠܘ ܡܢ ܟܕܘ ܚܒܬ‎ Le, ‘ And how I should be pleased if only the fire had been already kindled!’ For the first clause S has ,ܡܐ ܕܨܒܐ‎ perhaps a slip for ܡܐ ܕܨܒܢܐ‎ but the construction of ܐܠܘ‎ is the same as in C. The Peshitta agrees with C but omits ,ܡܐ‎

Vocabulary.

The Vocabulary of the Evangelion da-Mepharreshe is characterised by a straightforward simplicity, which is to some extent preserved in the Peshitta. There are certain standing locutions which are used constantly to the exclusion of others that might be thought equally good Syriac. Of these the most remarkable is the use of ras ‘life’ for cwrTypiov and cernpia as well as for Cwy. This extends to the verbal forms: ܐܢܢܥ‎ is used for ‘to save.’ Salvation and Life were thus identical terms in Syriac theological language, a different word (1n ta) being reserved for ‘ deliverance.’

Similarly ‘to come (or go) after’ stands both for dxodovfety and for €pyeoOo. 40, and among nouns ‘field’ is ܩܪܝܬܐ‎ and ‘boat’ is .ܣܦܝܢܬܐ‎ To each of the last there is one exception: eas is used for ‘field’ in Lk xvii 7 S, and ܐܠܦܐ‎ (‘ship’) is used for ‘boat’ in Matt xiv 22 @ a verse where S is illegible.

Notable Greek words are :—

occurs in Mk xu 38 S, Lk xx 46 SC, where the Greek has‏ ܒܐܣ̈ܛܘܐ otohais. The Hv. da-Mepharreshe apparently understood that‏ ܐܗ the Scribes wished to walk é Sroais like the Philosophers : cf Joh x‏ for orody occurs in Mk xvi 5 S, Lk xv 22 SC.‏ ܐܣܛܠܐ .23

Notable Greek Transliterations. 79

mia is used for 4 :ܐܩܘ«‎ ¢ Lk vi 17 S. The Peshitta has ܝܡܐ‎ Tw.

occurs in Lk iv 29 S as the name of the hill, from which ths‏ ܦܪܶܣ people of Nazareth wished to throw Christ down. The Greek is‏ by S. Well-‏ ܥܕܡܐ wal‏ ܛܘܪܐ ws dppvos Tod dpous, rendered‏ hausen (Nachrichten der K. Ges. der Wiss. zu Géttingen, 1895, p. 4)‏ suggests that wa is a transliteration of [6|¢pvos: possibly the word‏ was understood as an equivalent of ®apos, spelt elsewhere in Syriac‏ .ܦܘܪܘܢ and‏ ܦܐܪ̈ܘܣ

ma is used for rdoxa Mk xiv 1 6, Lk 11 41 SC, and also in Joh ¥1 4 ¢

See on ܦܛܝܪ̈ܐ‎ below, and the Note on Joh vi 4.

is used for kepdria Lk xv 16 S: see below.‏ ܩܖ̈ܛܐ

is used for Oapoetre Matt xiv 27 S, Mk vi 50S. In the other‏ ܬܪܣܘ C is only extant‏ , ܠܐ ܬܕܚܠ passages where Odpoe occurs S has‏ every-‏ ܐܬܠܒܒ for Matt xiv 27, where it has alusde; syr.vg uses‏

where.

The following words are noteworthy, as being adaptations from the Greek which are used to render other words than those of which they are adaptations.

used for ei dpa Mk xi 13 8.‏ ܕܛܟ : (from taxa)‏ ܛܟ

réin\ (from exdvy): ܠܩܢܐ ܕܫܝܓܬܐ‎ ‘a dish for washing,’ used for vurtyp Joh xii 5 S, supported by Aphraates 226 and Ephr. Lamy i 657. The Peshitta has ܡܫܝܓܬܐ‎ .

(from mirrdxvov): used for érvypagy Lk xxiii 38 SC, and also‏ ܦܛܩܐ by Ephr. Lamy i 667. The Peshitta has toda.‏

Lk xxiii 19 Cis quite obscure. It is obviously connected‏ ܗܛܖ̈ܘܬܐ with wwim ver. 25 C. In both places S has chaz, and the‏ corresponding to ordovs in the Greek. The‏ ,ܐ ܣܛܣܝܣ Peshitta has‏ reading of S is obviously a correction for some misunderstood or‏ in Mk xv 7, where‏ ܒܝܫܬܐ miswritten word. Similarly we find‏ syr.vg has pasa wore.‏

It is highly probable that the original word was some adaptation

of ordous, and I venture to conjecture that ms ܥܒܕ‎ Mk xv 7* 8 is a substitute for ܣܛܣܝܪܐ‎ (i.e. ,ܬܬܗ ܘܐܗ‎ Bar Hebr. Chr. Keel. ii 725%, Nold. § 140), and that ܒܝܫ̈ܬܐ‎ in Mk xv 7° S, Lk xxiii 19,

80

Granmar and Syntax.

25 S, stands for ܣܛܗܝܪܽܘܬܐ‎ (i.e. sedition, the crime of a cracudpvos), a word that actually occurs elsewhere in Syriac (Guidi, Statute. de Nisibi osai iv 1833). It is easy to see how ܣܛܣܝܪܘܬܐ‎ could be cor- rupted into the ܗܛܪܘܬܐ‎ found in C, while wim appears to come from a further confusion of this word with wm\m, ie. asalsor!,

More frequent than transliteration is the use of genuine Semitic

phraseology to render technical terms. The list that follows is arranged in alphabetical order. ܡܩܕܫܐ‎ dus ܐܝܩܪ‎ Joh x 22 8 (=7a evkaivia). The use of ܐܝܩܪܵܐ‎

lit. ‘honour, for Dedication is curiously illustrated from the in- scriptions on Palmyrene tombs. In addition to the inscription of ownership, which usually begins ‘This grave (S37 N99) was made by So-and-so,’ there is in at least two cases a tablet under an ornamental niche, which says ‘This memorial (or this statue), which is ܐܐܬ‎ na 4p’, was set up by So-and-so to the honour of his family’ (De Vogiié i pp. 40, 41, and p. 47). The setting up of the ornamental statue of the family genius or of the guardian of the tomb was the Dedication of the building to sacred purposes, just as the setting up of the Altar by Judas Maccabaeus (1 Mace iv 56) was the Dedication of the Temple. In any case it is interesting to find the same technical term used by the Christians of Edessa as was used by their heathen cousins at Palmyra about a century earlier.

a bier, Lk vii 14 8S". The original form of this Semitic word is‏ ܐܖܪܢܝܐ

preserved in the Arabic ht irdn ‘a bier.’ This became in Hebrew jms, the long d@ becoming 6, as usual. The meaning is ‘box’ or ‘ark,’ and so was used for the ‘ Ark of the Covenant.’ The Hebrew word in this technical sense passed over into Jewish Aramaic and also into Edessene Syriac, but the 6 of ’avénd marks the word as borrowed trom Hebrew. The Christian Palestinian (in Lk vii 14) has retained the word with the genuine Aramaic vowel, the spelling varying between ܪܶܢܐ‎ and rate.

Lk ii 14 S (=evdoxia). The word corresponds exactly to‏ ܐܪܥܘܬܐ

my Ezr v 17, vii 18. See above, on Néldeke § 51.

1 Somewhat similar to these words is ܫ ܡܛܘܠܪ̈ܝܐ‎ (i.e. kverriovdpior, quaestionariz), used in

the Peshitta as the equivalent of xoverwdia Matt xxvii 65ff. Here S has ܩܣܤܣܛܘܕܝ‎ and ܡܒܣܛܘܕܝܐ‎ . The word inal con (always in the plural as here) is, however, common in Syriac literature.

Syriac Vocabulary. 81

Lk xxiii 48 § 0 (=ocvrmapayevdpevor): of Acts of Thomas‏ ܐܫܬܘܪܝܘ The word exactly corresponds to ‘se trouver’ in French.‏ .178

to be excited (of persons), used by all Syriac versions for‏ ܐܬܥܙܙ euBpyracOar Joh xi 33, 38; also for dvaorevalew Mk viii 12 S, and‏ for ducxupiferba. Lk xxii 59 SC. The word seems to be a‏ metaphor taken from the stirring up of a storm.‏

praters Joh vii 49 §"@ (see vol. i, p. 554), corresponding to‏ ܒܕܝܐ . ܩܘܛܢܐ dyNos otros. Here C has‏ 6

in Matt vi 7 S has ‘do not be saying battdld0d, i.e. idle‏ : ܒ̈ܛܠܬܐ things, to render py Batradoyjonre, and a similar rendering is‏ found in the Palestinian Lectionary. C and the Peshitta have‏ ‘be not stammering’ (mépaqgéqin), i.e. wy Battodoyjaoynte. Is it‏ possible that the word Barradoyet is actually an early Christian‏ coinage from the Aramaic ?‏

the under-hair of camels, Matt iii 4 S C. The word also‏ ܒܥܘܐ occurred in Ephraim’s Commentary on the Diatessaron, according to‏ Ishé‘dad (Harris, p. 22).‏

ܐܬܒܨ stands for ¢ éEnpdvOn, but de‏ ܒܨ ܘܝܒܫ 6 .06 Lk vit‏ ܒܨ

means ‘to be emaciated, wizened (from drought)’ Ephr iv 491 .ܡ‎

wild mint Matt xxii 23 S(=ydvocpov). Cand syr.vg have‏ ܗܪܙܡܐ the ordinary word esas, and so also S in Lk xi 42. The‏ is obscure, but it does not seem to be a mis-‏ ܗܪܶܙܡܐ derivation of‏ writing of 7dvecpor in Syriac letters, as the 4 was quite clear in S.‏

seed of the Gentiles, used in Joh vii 35 SC‏ ܙܪܥܐ ܕܐܪ̈ܡܝܐ for 7 Swuacmopa tov “Eddjvev. etasaies, lit. ‘Aramaean,’ is the‏ regular conventional equivalent in Christian Syriac for ‘a heathen,’‏ without ethnographic signification, eg. Rom i 16. But the way‏ is used suggests that the technical sense of diacmopa for‏ ܙܪܥܐ ‘the Dispersed Jews’ was not familiar to the translator. The‏ Peshitta here has ‘the countries of the nations.’ In James i 1‏ ܕܙܪ̈ܝܥܢ) Siaamopa is rendered ‘those sown among the nations’‏ and in 1 Pet i 1 the ‘ Diaspora of Pontus’ is translated‏ ,(ܒܥܡܡ̈ܐ ‘those sown 7m Pontus.’‏

ras. The use of ras, life, to render cwrnpia and cwrrpiy, instead of some word meaning ‘ deliverance,’ together with the corresponding

B, II, 11

89 Grammar and Syntax.

equivalence of ܚܝܝܐ‎ and calerOa, belongs rather to theology and philosophy than to linguistics. It is noteworthy that this re- markable usage of the Hv. da-Mepharreshe whereby ‘salvation’ is identified with ‘life,’ was retained in syr.vg. The same definition of ‘salvation’ is given also by Clement of Alexandria: wrypia 009 TO emecOar Xpiot@: 6 yap ܘܗ ܘܟ‎ & atta lan éeoTw (Paed 1 vi 27).

dus the coning-to-life of the dead, used for dvagracis Matt xxiii‏ ܡ̈ܝܬܐ in S and partly also in C and syr.vg. It corresponds‏ ,30 ,28 ,23 nvnn. The ordinary equivalent to‏ ܐܐܐ exactly to the Jewish‏ it is noteworthy that the specifically Syriac‏ : ܩܝܡܬܐ avdotacis is‏ term for the Resurrection, vz. esasai, well used in the Peshitta‏ of Joh xi 24, 25, does not oceur in the Hvangelion da-Mepharreshe.‏ meant dvapvéis (Ac iii 19) rather‏ ܢܘܚܡܐ Perhaps in early times‏ than dvaoracis.‏

to be acquainted with, Mk xiv 68 S, Lk xxii 60 S (not C),‏ ܢܚܟܫܡ ܡܢܝܟܘܡܬܐ Lk xxiv 18 SC, where other texts have sas. Hence‏ ܗܘ acquaintance, used for ot yyworot Lk xxii 49 SC, for‏ used for “ Boaz (09 7‏ ܕ ܡܚܟܘܡܬܢ Joh xviii 15,168. Similarly‏ in Ruth ii 2 syr.vg. In all three places S spells the word‏ (cf Noldeke § 126 B).‏ ܡܚܟܡܬܐ

carol-pods of the sca, used for kepdtva Lk xv 16 C,‏ ܚܪ̈ܘܒܐ ܕܝܡܐ where S has wis, ie. ‘S. John’s Bread, the Carob-tree bean :‏ see art. ‘Husks’ in Encyclopaedia Biblica. But the addition of‏ is very puzzling.‏ ܕܝܡܐ

ial, i.e. the hills, is used for dypés in the sense of ‘the open-country’ in Lk xii 28 SC, and péde 0:6 ܹ6 is translated ܕܒܫܐ ܕܛܘܪ̈ܐ‎ Matt iii 4S. dak, Aull, is used for this sense of dypés in the Palestinian Lectionary.

cama Mk xii 14 S evidently differs from mtxt amas, the‏ ܕܪ̈ܫܐ ordinary Syriac equivalent for ‘poll-tax,’ in order to indicate‏ emxepadhavoy rather than kivoos.‏

ܗܐ ܐܦ Joh ix 21 S. For atros puxiay dyer S has am‏ ܡܪܫܢܘ̈ܗܝ ‘lo, he also hath become master of his years. 6‏ ܡܪܫܢܘ̈ܗܝ ܗܘܐ ‘he also hath entered‏ ܐܦ am‏ ܥܠ ܠܗ ܠܫܢܰܘܗܝ Peshitta here has‏ his years,’ an almost equally idiomatic phrase.‏

Lk xvii 10 ¢ The word means ‘ the sweepings of a threshing-‏ ܡܫ̈ܝܐ

Syriac Vocabulary. 83

floor,’ ‘chaff,’ e.g. Amos viii 6. I have therefore translated rato ܥܒ̈ܕܐ‎ by ‘slaves and riff-raff.’ But it is not unlikely that the translator confused dypetou useless with a&xupor chaff-heaps. The use of the rare Syriac word ܡܫܝܐ‎ in this forced and un- natural connexion probably led to its omission in S, followed by the Ethiopic version.

the morning-twilight watch, Mk xiii 35 S, is fem. to agree‏ ܢܘܓܗܝܬܐ .»ܘܩܘ »€ understood. The Greek has‏ ܡܓܪܬܐ with‏

rméoos Joh ui $ S, wes Joh iv 48 C, well known as a Jewish Aramaic term for ‘miracle.’ In Syriac it seems only to be used in the plural, chiefly in the phrase ܢܣ̈ܐ ܘܐܬܘ̈ܬܐ‎ (i.e. onpeta Kal répara, as in Joh iv 48 and in Aphraates 181, 273). But in Joh ii 28 ܢܤܣܐ‎ corresponds to oneta alone.

mar to cast lots, Matt xxvii 35 S, Lk xxiii 34 S C, where the Greek has Baddew kdnpov. The ordinary Syriac phrase is WMA ܐܪܡܝ‎ as in Mk xv 24.8; but ܢܦܣ‎ also occurs in Joh xix 24 syr.vg, where however the Greek has \ayydvew without «dypov.

)ܐ unleavened bread (ra dlvpa), used in 8. John to render‏ ܦܛܝܪ̈ܐ and the Note on Joh vi 4.‏ ,ܦܣ ܟ See above on‏

malas the mob, the common people, Joh vii 49 CL The Greek is only 6 dxXos ovros, but the comparatively rare Syriac word exactly hits the sense required. S has ܒܕܝܐ‎ praters: the ܒ‎ and the ܐ‎ are quite clear in the photograph taken by Mrs Lewis in 1902 (see above, p. 81). Possibly the original Syriac was sat > outsiders, as in Mk iv 11.

mais (with suffix) for dos, Matt ¥ 34 SC, Joh ix 34 S: see on Noldeke § 155.

puddle, shallow pool, Matt xiii 5 S (as I read the photograph).‏ » ܪܩܩܐ The word occurs in a gloss on 7a metpaédy, inserted apparently to‏ shew that there was there a little moisture, though without ‘depth‏ .ܬܬ of earth. mot is used in Exod 11 3 syr.vg to translate‏

letting the hands hang down (i.e. ‘ helplessness’), and‏ ܪܘܫܠ ܐܝ̈ܕܝܐ clasping the hands together (i.e. ‘ perplexity’) are‏ ܦܘܫܟ ܐܝ̈ܕܝܐ Lk xxi 25. The former is in S, the latter in‏ »)ܘ used to render‏

Cand syr.vg.

84 Grammar and Syntax.

Matt vi 5 S, Lk xiv 21 SC, is used to render pvuy ‘lane,’‏ ܫܩܩܐ ‘bazar,’ a word which corresponds‏ ܫܘܩܐ as distinguished from‏ and to mdareta. Like some other words for ‘small‏ ܘܘ both to‏ means by etymology ‘a fissure.’‏ ܫܩܩܐ street’ in various languages‏

blue-purple, used as a technical term for the ‘ ribband of blue’‏ ܬܟܠܬܐ worn on a Jew’s dress in accordance with Nu xv 38. It corresponds‏ to kpdomedov in Matt xiv 36 C, Matt xxii 5 SC (and syr.vg).‏ In Matt ix 20 S has rasa (as also in xiv 36), while syr.vg has‏ was‏ ܬܟܠܬܐ it may therefore be conjectured that here also‏ : ܩܪܢܝܐ the original reading of the Hv. da-Mepharreshe. Was there a‏ prejudice against representing our Lord as dressed in a distinctively‏ Jewish garb ?‏

ie. Hebr. porn, used by syr.vg in Matt xxiii 5 for dudaxrypia,‏ ܬܦܠܐ ois ‘the straps of their phylacteries.’‏ ܕܬܦܠܝܗܘܢ but S C have‏

an insipid fig-tree, Lk xix 4S Cand syr.vg (= ovko-‏ ܬܬܐ ܦܟܝܗܬܐ ܦܦܝܗܐ popea). There does not seem to be any other instance of‏ ܬܬܐ in the sense of ‘wild’: in fact, the natural rendering of‏ is ‘a fig that has gone bad,’ and it is difficult to resist the‏ ܦܟܝܗܬܐ obvious explanation that the translator did not know what tree was‏ popav. In‏ ܟܐܗ meant and translated the word as if it were‏ ie.‏ ܬܘܬܐ Lk xvii 6 S C and syr.vg translate cuxdpwos by‏ ‘mulberry.’‏

The Diatessaron, on the other hand, translated ov«opopéa in

Lk xix 4 by Na, the tree whose name forms part of the word ‘Bethphage, and if we may trust the Arabic (Diat xxxiii 10)

it had ‘fig’ instead of mulberry in the passage corresponding to Lk xvii 6.

are ܒܫܥܬܐ‎

8110 statim.

(A) Renderings of ei0vs, eiOéws, and kindred eapressions.

Tabulated Renderings.

APPENDIX TO CHAPTER IL.

88

The renderings for ev6¥s which we commonly find in the Hvangelion da-Mepharreshe

a>, 7

Matt iii 16 ܐܗ‎ 6

iv 20 evééws

and simple omission.‏ ܡܚܕܐ ,ܒܪ renderings (including simple omission) are found, such as continuo, confestim, protinus,‏ The lists which follow were originally drawn up in order to ascertain‏ whether the Latin and the Syriac renderings shewed any tendency to agree inde-‏ pendently of the Greek, but such is not the case, even with regard to omissions.‏

22 evOéws (om. lat.vt)

viii 3 ܐܘ‎ ¢ (om

. &*)

© > ܐ ܚ‎ 9 ‫ܹ 13 év TH wpa exetvy

ix 25 (wapaxpyya ®)

30 (after Kat: xili 5 evbéws 20 evOus (om.

6 ܘ 21

of xx 34)

2)

xiv 22 edOéws (om. N*C*al)

27 evOUs 31 evOéws

xx 34 evdéws

wo

xxi 2 evOéws (om 3 evOus

19 rapaypyyo.

20 mapaxpypa

xxiv 29 ܘܢܘ‎ ¢

xxv 15 ad fin. «6

xxvi 49 >¢

. lat. eur)

/ €ws

74 ܘ ܙ06ܧ‎ (or -éus)

XXVil 48 edGews

wa - illegible om. ܒܗ ܒܫܥܬܐ‎ ܒܗ ܒܫܥܬܐ‎

ܣܨ as‏ ܒܫܥܬܐ <hass aa‏

hiat

hiat buxamiam

In Latin also several

Syr.vg

as‏ ܒܫܥܬܐ

om, (= rell.)

86

Grammar and Syntax.

Mk i 10 ܘ ܐܧ‎ (om. D lat. cur)

i

11

=r

me

iv

Vv.

ܝܙ

19 6 18 ebdOvs (om. c) 20 ebOus (om. bt) 21 ܘܐܗ‎ (om. c) 23 edBds (om. © D latt) 28 edfus (om. N* 1 28 lat.vt) 29 dvs (om. Dec fr) 30 edfus (om. be ff qr) 31 (cd0éws, om. & Bale arm) 42 edOds (om. ber) 43 edOvs (om. ¢ bc aeth)

2 (ci0éws, om. N Bal)

8 ܐܧ‎ (om. D 28 565 al) 12 edOvs (om. lat.vt)

6 ܐܧ‎ (om. DLbcq)

5 edbus 15 ed6vs (om. 1-209 arm) 16 etOus (om. D 28 ܘ‎ £ ( 17 ܐܧ‎ (om. 1-209) 29 eddus (om. ec)

2 (dOus, om. B lat.vt arm) 13 (cd6éws, om. NBal) 29 ܘܐܘ‎ 30 cb6Us (om. lat.eur)

36 (edOews, om. NBD al latt. ) exc. a)

42 066

25% ed6us (om. 17 L 1 lat.vt) 25> égauris (om. Def)

27 edOvs (om. off vg)

45 evOus (om. c)

50 e’Ous (om. D 33 ¢ ?(

S hiat

hiat

ܒܗ ܒܫܥܬܐ

ܒܗ ܒܫܥܬܐ

om. (om.) om. om.

om,

ܒܗ ܒܫܥܬܐ ܒܗ ܒܫܥܬܐ

om. hiat hiat hiat

ܒܪ ܫܥܬܗ

om.

om. hiat hiat

hiat

hiat

<a>

ans‏ ܫܥܬܐ

om.

ܡܢ ܚܕܐ

ܒܗ ܒܫܥܬܐ

Tabulated Renderings. 87

S ܨ‎ Syr.vg Mk vi 54 ed6Us (om. q) ܒܗ ܒܫܥܬܐ ܗܝ‎ hiat ܒܪ ܫܥܬܗ‎ vii 25 etfs (om. an qg) om. 0 <a 35 (cits, om. BDlatvt) ܒܗ ܒܫܥܬܐ‎ : haze as viii 10 ed@us (om. D lat.vt exe. «) om. ‘i um ix 8 édmuva (ebOis Dlateur) ܡܢ ܫܠܝܐ . ܡܢ ܫܠܝܐ‎ 15 tbs ܒܪ ܫܥܬܗ .‫ ܒܗ ܒܫܥܬܐ‎ 90 ܘܗ‎ (om. Diateur.) hse ܒܪ ܫܥܬܗ . ܒܗ‎ 94 66 ܒܪ ܫܥܬܗ 5 ܒܗ ܒܫܥܬܐ‎ × 52 ets ܡܚܕܐ . ܒܗ ܒܫܥܬܐ‎ x1 2 06¢ܧ‎ (om. k) ܒܪ ܫܥܬܗ 0 ܒܗ ܒܫܥܬܐ‎ 3 dbus ܡܚܕܐ‎ 0 <aum xiv 43 eb6vs (om. D113 565latt) — om. 0 om. 45 ebfis (om. D 565 lat.vt) am ܒܪ ܒܫܥܬܗ ܕ‎ 72 etGvs (om. =) om. 0 ܒܗ ܒܫܥܬܐ‎ × 1 edOus (om. ac) om. 0 au Lk i 64 rapaxpqya (om. @) ܒܗ ܒܫܥܬܐ‎ hiat ܡܚܕܐ‎ ii 38 ܫ ܙܚܘ‎ ‡ Spa ܒܫܥܬܐ ܒܗ ܒܫܥܬܐ‎ as iv 39 ܡܚܕܐ .‫ ܒܪ ܫܥܬܗ )ܘܩ ܘܐ‎ ¥ 13 eibéws ܒܪ ܫܥܬܗ 0 ܒܪ ܫܥܬܗ‎ 35 <» ܡܢܢܕܐ 7 ܒܗ ܒܫܥܬܐ ¥ ܘ‎ 40 (ahd 90990 ܡܕܐ "