'M3\ 
 
 
 
 
 w 
 
 ^Mm.^^ 
 
 m '^ 
 
CORNELL 
 
 UNIVERSITY 
 
 LIBRARY 
 
Date Due 
 
 t^ C«T. NO i3233 
 

FOUE LECTURES 
 
 THE WESTERN TEXT 
 
 OF 
 
 THE NEW TESTAMENT. 
 
aonDon: C. J. CLAY and SONS, 
 
 CAMBRIDGE UNIVERSITY PRESS WAREHOUSE, 
 
 AVE MARIA LANE. 
 
 damtoiUsc: DEIGHTON, BELL AND CO. 
 
 Ecipjis: F. A. BEOCKHAUS. 
 
 i^Eijj gorlt: MACMILLAN AND CO. 
 
FOUE LECTURES 
 
 ON 
 
 THE WESTEEN TEXT 
 
 OF 
 
 THE NEW TESTAMENT 
 
 BY 
 
 J. RENDEL HARRIS, M.A., D.LiTT.(Dubl.) 
 
 FELLOW OF CLARE COLLEGE, -AND LECTURER ON PALAEOGRAPHY 
 IN THE UNIVERSITY OP CAMBRIDGE. 
 
 LONDON : 
 
 C. J. CLAY AND SONS, 
 
 CAMBKIDGE UNIVEESITY PRESS WAREHOUSE, 
 
 AVE MAEIA LANE. 
 
 1894 
 
 [All Rights reserved.'] 
 

 PBINTED BY C. J. CLAY, M.A. AND SONS, 
 AT THE CNIVEKSIXY PKESS. 
 
" Nee sine voluptate audias eniditos viros, amicissime et 
 jucunda verborum verecundia dissentientes, ut non disputationi 
 quales re vera esse pleraeque solent, sed colloquio qualia veteres 
 fingere consueverimt philosophi, interesse te putes." 
 
 J. D. MiCHAELIS, 
 
 {Curae in Versionem Syria/iam, 1755, p. 160). 
 
Cornell University 
 Library 
 
 The original of tiiis book is in 
 tine Cornell University Library. 
 
 There are no known copyright restrictions in 
 the United States on the use of the text. 
 
 http://www.archive.org/details/cu31924029305393 
 
PREFACE. 
 
 The lectures contained in the following pamphlet are con- 
 cerned with that problem of problems, the interpretation of the 
 Western text of the New Testament. Since my tract on the 
 Codex Bezae was published in 1891, there have been a number 
 of weighty contributions to the solution of the problem, which 
 have added much to our knowledge of the textual history of 
 the New Testament, and made it necessary for me, before con- 
 tinuing my imperfect examination of the Cambridge MS., to 
 estimate the results of the other workers in the same field. 
 
 The fact is there are now four separate and distinct theories 
 before the world ; Resch's theory' that the bifurcation in the 
 primitive text of the New Testament is due to independent 
 translations from a Semitic document (probably Hebrew) ; 
 Chase's", that all the variants are due to reflexion from an Old 
 Syriac translation ; Blass's theory', that in the Lucan writings 
 they are due to the issue of two separate drafts from the hand 
 of the original writer — a statement which supports itself, in 
 certain points, on a powerful programm of Peter Corssen, in 
 which was demonstrated the antiquity and wide diffusion of 
 that part of the Latin tradition of the text of the Acts which 
 is in agreement with the quotations of Cyprian*; and my own, 
 
 ^ Besch, Aussercanonische Paralleltexte. Leipzig, 1892. 
 
 2 Chase, The Old Syriac element in the text of Codex Berne. Cambridge, 
 1893. 
 
 ' Studien u. Kritiken, 1894, pp. 86—120, Die zioeifache Textilberlie/emng in 
 der Apostelgeschichte. 
 
 * Der Cyprianische Text der Acta Apostolortim. Berlin, 1892. Corssen says- 
 some significant things about the modern edited texts, which he calls 'der 
 destillierte Text, den die Modernen aus einigen grieohischen Unoialen gewonnen 
 haben,...nur ein Spiegelbild einer willkiirlich fixierten Eeeension des vierten 
 Jahrhunderts' (p. 24). 
 
Vlll PREFACE. 
 
 which claimed that there had been a reaction on the Greek 
 text from the primitive Latin translations, as well as, occasion- 
 ally, from the Syriac version. Further we have a, remarkable 
 chapter on the Codex Bezae in Ramsay's recent work The 
 Church in the Roman Empire, in which numerous changes of 
 the text are assigned to the hand of a Greek reviser acquainted 
 with the geography of Asia Minor. T shall set my own theory 
 for the present on one side, not because I have abandoned 
 it, but merely remarking that my critics were probably 
 right in saying that I had exaggerated the sphere of Latin 
 influence, and I believe, equally right in conceding that a 
 certain amount of Latinisation did exist. 
 
 The first lecture which discusses Resch's views was de- 
 livered more than a year ago, and printed in the Classical 
 Review for June, 1893; it was not meant to be a final exami- 
 nation of Resch's theory, but merely to point out that a closer 
 acquaintance on his part with the actual text of the Codex 
 Bezae was necessary, and a consequent restatement of the 
 arguments, before Resch was likely to meet with the exhaustive 
 treatment, which I have no doubt his hypothesis deserves. 
 
 The other lectures are concerned chiefly with Mr Chase's 
 theory that the Codex Bezae is under Old Syriac influence, 
 and Dr Blass's view, that it is an original document, in good 
 Lucan Greek. 
 
 Both of these writers have added much to the subject upon 
 which they treat ; and Mr Chase's is a theory, which in spite 
 of certain peculiarities in its presentation, challenges the 
 fullest scrutiny, and will certainly, if sustained, greatly ad- 
 vance the subject in hand. The lectures are, of course, an 
 incomplete treatment of the questions at issue ; but I feel 
 hopeful that they, too, will do something to speed us towards 
 the goal which the critics have been so long striving to attain, 
 the complete explanation of the primitive variation and 
 bifurcation in New Testament texts. 
 
CREDNER AND THE CODEX BEZAE'. 
 
 In a work just published, entitled Aussercanonische Far- 
 alleltexte', by Alfred Resch, who is already knowa to the world 
 of Biblical criticism by his treatise on the Agrapha of the New 
 Testament, will be found certain criticisms of my tract on the 
 History of the Codex Bezae, which was published last year in 
 the Cambridge Texts and Studies, just as Resch's researches 
 are in Haruack's Tetcte und Untersuchungen. The parallel be- 
 tween our publications is not merely an external one, though I 
 think it is fair to admit that we are distinctly imitating, in our 
 little Cambridge series, the German research and enthusiasm 
 which Harnack has done so much to crystallize : we are also 
 working internally on parallel lines, and especially Dr Resch 
 and myself are engaged on the very same questions, viz. the 
 origin of the variant forms of the Gospels, only we are working 
 from opposite ends ; I am working up stream, and Resch is 
 working down ; I follow the readings of variant MSS. up 
 stream until I find, as I suppose, their origin; Resch has 
 divined, as he supposes, their origin and has only to read the 
 facts in the light of his hypothesis ; and we shall meet by and 
 bye somewhere between our two starting-points, and it would 
 be presumptuous at present to anticipate whether the meeting- 
 point is nearer to my end of the line of action or of his. 
 
 But we may at least be grateful that each of us is able to 
 appreciate the industry of the other, and not disposed unduly 
 to depreciate the results which are brought forward ; on my 
 
 ' This Lecture was delivered in the Divinity School, Cambridge, Nov. 19, 
 1892. 
 
 ^ Aussercanonische Paralleltexte zu den Evangelien: textcritische wnd quellen- 
 critische Grundlegungen ; von Alfred Eeseh, Leipzig, 1892. 
 
 HA. 1 
 
4 CREDNER AND THE CODEX BEZAE. 
 
 Credner's Beitrdge zur Einleitung in die Biblischen Schriften, 
 pp. 452 — 518. It is true that Kesch points out a weakness in 
 Credner's work, in that he did not recognize sufficiently the 
 relations between the Codex Bezae and the Old Latin and Old 
 Syriac texts, which constitute with it a distinct textual family. 
 But I think we may say without injustice that Resch, with this 
 single exception, endorses Credner's investigation and its 
 results. For instance, the view that the MS. has a Judaeo- 
 Christian origin is endorsed^; the 'unknown authority' of 
 Credner, which lies at the back of the Western text as one of 
 its sources, is identified with a secondary translation of the 
 original Hebrew GospeP; and the stages of development which 
 Credner indicates for the Western text, as we find it in 
 Codex Bezae, are accepted and tabulated **- 
 
 ' E.g. p. 33. 'Hatte der Redaktor jenes Archetypus, jenes altesten Evan- 
 gelienoanons, ohne selbst Judenchrist zu sein, die — auch von Justin getheUte, 
 vermittelnde Stellung dem gemassigten Judenchristenthum gegeniiber zum 
 Ausdruck gebraoht, sofern er das judenchristliohe Evangelium an die Spitze 
 dieses im TJebrigen echt katholischen Bvangeliencanons gestellt hatte, so ver- 
 danken wir speciell die weitere Ausbildung der Handschrift, die uns jetzt im 
 Codex Bezae vorliegt, ausschliesslich judenchHstlichen Ereisen, welche die 
 Apostelgesohiohte und die katholischen Briefe, nioht aber das paulinische 
 Schriftthum, der Handschrift einverleibten. Auch die weitere Conservirung 
 der Handschrift im Laufe der nachsten Jahrhunderte wird, wie Credner ganz 
 richtig gesehen hat, auf dieselben judenchHstlichen Kreise zuriiekzufiihren sein. 
 Denn wahrend in der orthodoxen Eirohe in Folge der oauonisohen Textrecension 
 die Exemplare jener Torcanonisehen Evangeliensammlung langst verdrangt und 
 verschwunden und nur in Uebersetzungen erhalten waren, blieben diese juden- 
 christlichen Kreise fon der Textrecension der Grosskirche unberiihrt, und 
 konnten so ein griechisches Exemplar jener vorcanonischen Evangeliensamm- 
 lung fiir ihren gottesdienstlichen Gebrauoh bewahren und in jene spatere Zeit 
 hiniiberretten.' 
 
 ■' P. 144. 'Bei der Besprechung dieser wichtigen Handschrift... habe ioh 
 bereits darauf hingewiesen, dass die aussercanonisohe Textrecension, mit 
 weloher speciell das Lucas-evangelium in diesem Codex auftritt, zu erklaren sei 
 aus dem Einfluas einer Uebersetzung des Urevangeliums, verwandt derjenigen, 
 welche -von dem ersten Evangelium beniitzt worden ist, und dass eben hierin die 
 von Credner gesuchte "unbekannte Autoritat" zu finden sei, "auf welche die 
 kiihne Textrecension des Lucas-evangeliums nach dem Codex D sioh stutze, 
 indem hieraus auch die zahlreiohen seheinbaren Conformierung des Lucastextea 
 nach dem Matthaustexte sich erklaren." ' 
 
 = P. 35. 
 
 'A. Archetypus. 
 Griechischer Evangelienoanon spatestens um 140. 
 
CREDNER AND THE CODEX BEZAE. 5 
 
 It is further stated, in accordance with Credner's views ('hat 
 Credner jedenfalls richtig bezeichnet '), that in the history of 
 the origin of the bilingual text we must allow that the Latin 
 text was added to the Greek as early as 500, in order to allow 
 for such corruptions as have arisen from the interaction of the 
 Greek and Latin upon one another. But before this time the 
 tradition of the text involved many marginal annotations, such 
 as the Ammonian sections, and apparently the Sabbath lections, 
 while at the time when this redaction was made, the lections of 
 Euthalius were introduced, and the stichometric division of the 
 text. For we know for certain (according to Resch) that the 
 stichometric division of the Acts is due to Euthalius. And it is 
 natural to assume that if at this time (about 600 a.d.) the 
 Latin text, stichometrically divided, were added to the Greek, 
 the Latin text would remain free from the previously existing 
 Greek annotations of a liturgical character. And it is these 
 liturgical notes, together with their Sabbath lessons, which more 
 than anything else {mehr als alles Andere) entitle us to refer the 
 origin and use of the Western text to Judaeo-Christiau circles, 
 and enable us to approve Credner's suggestion that the text 
 was brought into Southern Gaul, in its later form, by some 
 Syrian Jewish-Christian, probably a trader, and that it was 
 finally dictated to a scribe, not very well acquainted with Greek, 
 towards the end of the sixth century. 
 
 It is sufficient to present this brief summary, to show that 
 Resch has absorbed Credner's views almost without modifica- 
 tion; and since he has rarely added any reason for their 
 reception except, by reference and implication, the reasons 
 already given by Credner, we are entitled to conclude that he 
 
 B. Evangelienoanon 
 mit Apostelgesch. u. katliol Sriefen 
 vor 200. 
 
 C. Neue Bedaktion tou B, 
 Beifiigung dea lateinisohen Textes, 
 
 urn 500. 
 
 D. Letzte Abschrift des 
 biliugualen Codex 
 
 gegen Ende des 6. Jahrhunderts.' 
 
6 CEEDNER AND THE CODEX BEZAE. 
 
 considers those reasons to be valid, and not to need much 
 further enforcement. 
 
 The best way to see the error which Resch has made in 
 thus endorsing Credner will be to follow the method which I 
 adopted in my tract on the Codex Bezae, viz. to begin with the 
 marginal annotations. 
 
 On p. 27 Resch has copied the following marginal note from 
 the Codex Bezae, and given an elucidation of it : — 
 
 rNOCMA ) 
 
 
 ■ [avajyvoa/Ma 
 
 piToycA 
 
 
 [7re]/at tov era 
 
 TOOTHC 
 
 
 [,S/8a]T0U T7?9 
 
 AKOyNI 
 
 
 \^i\aKOVvi, 
 
 MOY 
 
 
 { [cTLlflOV 
 
 i.e. it is a lection for the Sabbath which precedes the Sunday 
 after Easter, which is called the SiaKivrjai,/MO';. And it is in- 
 ferred that since the lectionary direction is given in this 
 imperfect form, it must have been copied from a previous MS. 
 in which the direction had become partly illegible '- 
 
 In this Resch was simply following Credner, who had taken 
 the lection from Kipling, and had remarked that 'in unsere 
 Handschrift konnten die verstiimmelten Worte aber nur 
 dadurch gelangen, dass sie, Buchstabe fiir Buchstabe, aus einer 
 andern Handschrift eingetragen worden sind, welche schadhaft 
 geworden war, dergestalt, dass die Anfangssylben fehlten.' — 
 (Beitrdge, p. 500.) 
 
 The mistake made (for as we shall see presently the 
 explanation is erroneous from stem to stem) was partly due to 
 Kipling, who had printed these liturgical notes on the margin 
 of his text, in the same type as the text. But Kipling did not 
 venture to make the liturgical note coeval with the MS., as 
 must be the case, if the theory is to hold that it was to be 
 found in the tradition of the text at an earlier date than the 
 Codex Bezae itself. What Kipling said of it was as follows 
 
 ' 'Die lifcurgischen Eandbemerkungen, welche dureh den fortgesetzten 
 Gebrauch des friihereu Kirohenexemplars defekt geworden waren, wurden in 
 ihrer verstummelteu Gestalt der neuen Handschrift von einer andern Hand 
 einverleibt.' 
 
CREDNER AND THE CODEX BEZAE. 7 
 
 (p. XV.): 'notae liturgicae,...non a prima quidem manu, at certe 
 tamen, ut mihi videtur, ante saeculum septimum appositae,' 
 and Kipling's statement was copied by D. Schulz (Disputatio, 
 p. 10) in the words ' Haec glossemata antiquissima, si minus a 
 primel quidem manu, at certe ante saeculum septimum iamiam 
 adscripta esse.' How then did Credner come to attribute an 
 artificial antiquity to such a liturgical note ? Obviously it was 
 the incomplete form in which the note occurs, which suggested 
 that it had been copied from a previous Codex. But in this 
 Credner was misled by Kipling, and did not see that what 
 Kipling was trying to reproduce was an annotation on the 
 margin of a MS., where a part of the MS. had been cut away. 
 
 And it is unfortunate that Eesch, who has read through 
 the Codex Bezae, both in Kipling's edition and in the edition of 
 Scrivener, did not see the mistake that Credner had made, nor 
 correct it, either by Kipling's preface or by Scrivener's preface 
 and annotations. If he will turn to Scrivener's edition, p. 450, 
 he will find the following note : 
 
 423 b. 11. 11 — 15... ayvo<rfjia..pi tov a-a...Tai T7]<;...aKovvi.... 
 fiov {i.e. SiMKivrjaifiov sive e/SSo/i. a, margine abscissa L etc.); 
 
 and if he will further turn to the preface of Scrivener (p. 
 xxvii.) he will find conclusive reasons for dating this corrector 
 (whom Scrivener calls L) not earlier than the ninth century. It 
 is, to say the least, unfortunate that Eesch had not taken the 
 trouble to verify such an easy point as the date of an annotator. 
 
 What then becomes of Credner's Judaeo-Christian liturgical 
 notes, which, according to Resch, more than anything else, lead 
 us to believe that the Codex Bezae goes back to a Judaeo- 
 Christian origin ? Not only does it appear that this particular 
 case, on which so much has been built, is a delusion ; but the 
 whole of Credner's remarks on the liturgical annotations in the 
 Codex Bezae are at fault : first, they do not belong to the date 
 to which Credner wishes to refer them ; secondly, they have 
 nothing to do with the Jews. 
 
 As Resch has challenged an appeal to Credner, to 
 Credner he must go ; but he must not go without the MS. 
 or some trustworthy edition of its text. And if Credner wishes 
 
8 CREDNER AND THE CODEX BEZAE. 
 
 to carry these liturgical notes back into remote antiquity, or 
 even to the time of production of the Codex Bezae or the 
 century before that time, he must be met with a stern 
 palaeographical negative. None of these annotations are as 
 early as the ninth century, and some of them are as late as the 
 twelfth. Consequently it is hard to regard Credner's study of 
 the Codex Bezae as the most instructive that has yet appeared; 
 it is unjust to Scrivener, to say the least. 
 
 The number of errors which Credner has fallen into in his 
 account of these marginal hands is simply appalling. I shall 
 give one or two instances, if only to justify myself for having 
 neglected Credner. 
 
 Would any one believe it possible that a lection marked 
 against Matt. xvi. 28 — xvii. 9 with the words fieTat.fiop(f>o<; 
 avvayvoa/jba could be interpreted in the following manner ? — 
 
 ' Ich weiss das selfsame erste Wort nur aus eine Ver- 
 mischung des Griechischen mit dem Lateinischen zu erklaren. 
 fjLerafiopcfio'i soil heissen, eigentlich : fiera morbos. Das Ein- 
 dringen Lateinischer Wbrter in die Griechischen Sprache des 
 gemeinen Lebens ist aus dem N.T. bekannt. Hiemach sollte 
 der bezeichnete Abschnitt als Gebet und Trost fur Kranke und 
 Genesende verlesen werden, und dazu passt auch der[Inhalt.' 
 The lesson is, as the matter shows, the regular one for the feast 
 of the Transfiguration (t'^? /jLeTafxopcfxoaewi). The date of the 
 annotation is, as before, of the ninth century, yet Credner does 
 not hesitate to say (p. 505) 'auch der Umfang und die liturgische 
 Beschaffenheit dieser Eandbemerkungen fuhren uns auf 
 Judenchristen.' 
 
 A more striking case still is in a marginal annotation 
 attached to John v. 18; which reads 
 
 epi<>,Ni\nAY 
 AMeNOc 
 
 and is rightly given by Credner in the form -jrepl dvairavaa- 
 /jievov;. That is, we have here a church lection pro defimctis. 
 But according to Credner, who wishes to find traces of Judaeo- 
 Christian usage, we are to see in the words an allusion to those 
 persons who rest on the Jewish Sabbath ; for according to his 
 
CREDNEK AND THE CODEX BEZAE. 9 
 
 view (p. 506), 'Dies bezieht sich auf die Jiidische Feier des 
 Sabbathes, welche avd7ravac<i heisst. Joseph. Antiq. 3, 12, 3. 
 Derselbe contra Ap. 2, 2. Epiphanius Haer. 42, 3. Thilo, 
 Acta Thomae pp. 146, 223. ' Kva-n-avarafjuevoi sind folglich 
 diejenigen, welche den Sabbath nach Jiidischer Weise als 
 Ruhetag feiern. Die Absicht des Verfassers dieser Bemerk- 
 ung kann nun keine andere gewesen sein, als das Unrecht- 
 massige der Jiidischen Feier des Sabbathes auf Johannes v. 18 
 gestiitzt hervorzuheben u.s.w.' And it follows that, if the 
 allusion to those persons who rest on the Sabbath be attached 
 to the passage in which Christ is charged with breaking the 
 Sabbath, then we are in the circle where Jewish beliefs are in 
 process of being antagonised; that is, the text belongs to a 
 time when the Church is being withdrawn from its half- Jewish 
 state into one more distinctly Christian. All this pyramid-on- 
 apex-building depends on the curious interpretation which 
 Credner makes of the marginal reading. Nor does he stop 
 here, but realizing that the case was not dissimilar to the one 
 which we previously discussed, he maintained that the im- 
 perfect form in which the marginal note is transmitted is again 
 a case where marginal annotations have been taken from a 
 copy in which they had become partly illegible. 
 
 But, as before, the marginal reading, which is by the very 
 same hand, refers to lectionary usage of the ninth century; 
 the two missing letters have been cut from the edge ; and the 
 lesson is the proper one to be read over the departed. And 
 the reason for the selection of the passage is to be sought, not 
 in any allusion to Sabbath breaking, but in the doctrine that 
 the ' Son quickeneth whom he will &;c.' It is the more strange 
 that Credner should have missed the meaning, since he cites 
 lower down (p. 511) a Roman burial inscription in the form 
 ToTTOS dvatravaaica'! 'A/ji,/j,oviov koL Evrv^e^ov dpiirTov. But 
 the fact is that very little which Credner wrote will stand an 
 appeal to the manuscript, and for this reason I am sorry that 
 Resch has endorsed so much of his work, and especially that 
 he has laid such stress on the demonstration of a Judaeo- 
 Christian origin which Credner detected in the marginal 
 liturgical annotations. A reference to Resch (p. 34) will show 
 
10 CREDNER AND THE CODEX BEZAE. 
 
 that he has carried up to the date A.D. 500 the Ammonian 
 sections, the pericopes for the Sabbath, the lections of Eu- 
 thalius in the Acts, and the stichometry (= colometry) of the 
 MS. Let us then see what Scrivener says on the subject of 
 the Ammonian sections in the Codex Bezae. He tells us 
 {God. Bezae, p. xx.) that 'The Ammonian sections, without 
 the Eusebian Canons, are inserted in the side margin of Codex 
 Bezae by a scribe whom we shall hereafter show to have 
 lived several centuries later than this manuscript was written'; 
 and again (pp. xxvi. xxvii.) 'it is evident from a careful 
 comparison of the marginal numerals of the Ammonian 
 sections with the great body of the liturgical annotations 
 (written in thick, clumsy letters with ink of a purple hue), 
 especially in the Gospels, that they are the work of one scribe, 
 
 whom we shall call L A bare inspection of Facsimile PL 
 
 iii. no. 12 will prove that L cannot be dated before the 
 
 ninth century.' I suppose that Credner was here misled by 
 Kipling's edition in which the Ammonian numbers are printed 
 with the text, though Kipling did not assume them to be 
 coeval with the MS.: and Resch must have followed Credner, 
 though a reference to Scrivener would have kept him from this 
 hydra-headed catalogue of errors. We have now disposed of 
 the argument from the Ammonian sections, as well as that 
 from the pericopes for the Sabbath (which are a mere relic 
 of Galilean usage in the ninth century, as I think might have 
 been gathered from my own modest little tract). We come, 
 next to the question of the lections of Euthalius in the Acts, 
 which Credner, followed by Resch, carries back to the time 
 of Euthalius, very nearly. Here again we are dealing with 
 marginal references of a later date, by two separate hands 
 of the twelfth century. I do not even believe that we can 
 make out an identification of the lections with the Euthalian 
 system ; but, even if we could, there would be nothing gained, 
 for no result follows from the marginal ascription of Euthalian 
 lections in the twelfth century. And I am only sorry that 
 Resch did not see that he was treading on the thinnest of thin 
 ice in following Credner. 
 
 But, it may be said, the stichometry of the MS. is surely an 
 
CREDNER AND THE CODEX BEZAE. 11 
 
 integral part of the MS. itself; and while on the one hand it 
 cannot be more ancient than Euthalius, who invented it; on 
 the other hand it must be early, for the Codex Bezae was 
 obviously transcribed from a copy similarly divided to itself. 
 But neither can this be made out, for a reference to Euthalius' 
 reckoning of the crrt'^ot into which he divided the text agrees 
 closely with the conventional book-measure obtained by dividing 
 the text into breadths of sixteen syllables. And even if it be 
 argued that Euthalius colified (to coin a word) his text into 
 short sense-lines as well as measured it, there is no proof 
 that the Codex Bezae contains his system, and, as far as the 
 Gospels go, the line division can be carried back much farther, 
 probably into the second century. So that all the details of 
 the description of the MS. which Eesch characterized as 'most 
 instructive ' are shown to be errors, arising from an insufficient 
 acquaintance with the MS. itself But, further, Eesch has 
 (p. 34) adopted from Credner the theory that the Codex was 
 written from dictation. There is nothing in this which bears 
 closely upon the problem of origins; and yet, as I have been 
 directed to study my Credner, and Resch gives no other 
 evidence than what is found in Credner, it may be worth 
 while to look at the instances, and see whether the palaeo- 
 graphical argument is a just one. 
 
 Credner gives ten instances of errors introduced by aberra- 
 tions of the ear in a scribe writing from dictation. Most of 
 these would be rejected at once by any one familiar with the 
 copying of uncial texts ; e.g. 
 
 Luke xvi. 26, D has neicei for iwicei (written iweicei in early 
 texts). 
 
 Luke vi. 20, D has eriApAc for en^pAc. 
 
 Acts vi. 5, 4, D has mscon coi for mgnon coi, where the eye 
 has wandered two or three letters. 
 
 Acts iv. 29, D has read AneiAAc as *v^ll^c, which is certainly a 
 palaeographical error. 
 
 But what need to go further ? for if these are copyists' errors, 
 the scribe was reading the book for himself, and not writing 
 from dictation. The only instance which Credner gives that 
 
12 CREDNER AND THE CODEX BEZAE. 
 
 has any verisimilitude is John xiv. 21, where i/xcfiavla-cc has 
 been read as ev(pcov-)j<TQ). But the human mind is quite capable 
 of such confusions, without the introduction of a dictator. 
 
 It seems, then, that all the particulars of Credner's theory 
 which Eesch has taken over are invalid and unsustained by 
 an appeal to the MS. We do not mean to say that Credner's 
 theory of a Judaeo-Christian origin of the Western text is 
 an impossible theory, or the closely related theory of Resch. 
 Only the reasons which have been brought forward thus far 
 are out of harmony with the palaeographical facts, and new 
 reasons must be found. lion tali auxilio nee defensorihus istis. 
 We have no wish to disparage Resch's work on the text, 
 which is extremely interesting and may lead to some very 
 important conclusions. Only we must ask him to neglect 
 Credner, and to allow other people to neglect him, when they 
 can show good reasons for doing so. There is too much 
 interesting work on hand for us to be justified in spending 
 precious time in correcting the multitudinous errors of a critic 
 of the last generation. Probably it would also be wise of 
 Eesch not to lay undue stress on certain other points, which 
 he has borrowed from Credner, in reference to the theory of 
 the Canon; for they do not affect his new theory of Gospel 
 origins, and may cause it to be unfairly discredited. I mean, 
 to take an instance, the statement that the original Judaeo- 
 Christian Canon contained only the Four Gospels, the Catholic 
 Epistles, and the Acts : and not the Pauline Epistles (Resch, 
 p. 33). We are all agreed that the Codex Bezae does not 
 contain the Pauline Epistles, and we are also satisfied that it 
 contains one page of a single Catholic Epistle (probably the 
 last of three originally extant Epistles of John); but we are 
 not agreed that Codex Bezae is the Canon, nor that it is a 
 Judaeo-Christian or Petrine book. Surely it would be better 
 to reserve our judgments in the case of a MS. which is im- 
 perfect in the middle and at the end, and of which we have not 
 even the right to affirm that it existed without a companion 
 volume. 
 
 It is fair to make this last suggestion, because Resch has 
 made a mistake of this very kind with regard to the com- 
 
CREDNER AND THE CODEX BEZAE. 13 
 
 pleteness of a MS. in his notes on the Oodex Sangallensis (A), 
 which he thinks may go bacii into a very early base, partly 
 because it is bilingual, and partly because it is limited to the 
 Four Gospels. It is quite true that the Codex Sangallensis 
 does contain many very early readings, but, as to its limitation 
 to the Four Gospels, it is well known that the companion 
 volume containing the Pauline Epistles is extant, and is known 
 in the critical apparatus by the sign G """', and to librarians as 
 the Codex Boernerianus. 
 
 I have said nothing so far in defence of my own theory 
 about the Codex Bezae. The reader of Resch's little book 
 will see that my explanations are not considered adequate, 
 and that there are many readings where Codex Bezae deviates 
 from the Canonical text, which are not easily explained by the 
 hypothesis of Latin influence. I think this is quite possible 
 and have no objection to make, if Resch can establish his 
 contention ; for I hope to see my way some day to all necessary 
 corrections and expansions of my first statement. But perhaps 
 I may remark that there are some weak spots in Resch's list, 
 and in his deductions from them. For example, if it be true 
 that the reading of D in Matt. x. 6 (inrdyeTe for iropeveffOe) 
 cannot be due to any retranslation, as the words are wholly 
 indifferent in meaning, why does it follow that in Luke ix. 57, 
 where D reads indyei'; for izTripxy, we are entitled to infer 
 that the Itala MSS. also must have had vTrdyea or virdyr)^ in 
 the texts from which they were translated, because they now 
 agree with Cod. Bezae in reading ieris (wenn nun alle Itala- 
 Handschriften mit der Vulgata ieris lesen und also vtrdyeK... 
 voraussetzen) ? Or perhaps the retranslation would be invalid 
 in support of Harris' theory, but good when employed in de- 
 monstrating the unity and antiquity of the Western text, 
 which, of course, I hold as strongly as Resch ? 
 
THE OLD SYRIAC TEXT OF THE ACTS'. 
 
 Mr Chase's theory of the Western text is that the pecu- 
 liarities of the Codex Bezae are due to retranslation from an 
 Old Syriac version ; and apparently to this cause only, since he 
 states that the demonstration of the Syriacization of the Greek 
 text excludes the theory of its Latinization, and I suppose 
 would equally, from his point of view, exclude the theory of 
 every other form of textual reaction and bifurcation. In other 
 words, the hypothesis of re- translation from the Syriac is an 
 adequate one to explain the peculiarities of the Bezan Text^ 
 
 The hypothesis here presented is, in one sense, not a new 
 one. It is, in fact, the theory of J. D. Michaelis in the last 
 century and D. Schulz in the present century. Michaelis' state- 
 ment is as follows': "an alteration of the Syriac from the Latin 
 cannot possibly be supposed ; ... in Syria, where Greek was 
 understood, no man could have thought of correcting the 
 Syriac Testament from a Latin translation, and those Syrians, 
 who were acquainted with Greek, were undoubtedly ignorant 
 
 of Latin More probable is the supposition that the Syriac 
 
 has had influence on the Latin, especially in those examples 
 where an error is committed, that might happen more easily 
 to the Syrian than the Latin translator. The Latin text is 
 properly a composition of several ancient versions, one of 
 
 1 This lecture was delivered in the Divinity School, Cambridge, January 
 19th, 1894. 
 
 2 It even explains, from Mr Chase's point of view, the itaoisms of the scribe 
 (d5fX0r) for dde\<poi.) and the Alexandrian verb-forms (as iOopv^ovaav = iBopv^ovv 
 ■q(Tav). But we must not judge a theory by the extravagances into which it may 
 lead its promoter. 
 
 3 Marsh's Michaelis, Vol. ii., part 1, p. 25, 
 
THE OLD STKIAC TEXT OF THE ACTS. 15 
 
 which must have been made by a native Syrian', as appears 
 from the Syriasms found in the Latin text of several ancient 
 MSS., which greatly exceed in harshness the Syriasms of the 
 Greek Testament : this Syriac translator was probably guided, 
 in obscure passages, by the version of his own country, the 
 effects of which appear to be felt at this very day in the 
 
 Vulgate But the foregoing hypothesis is very insufScient 
 
 to account for that general coincidence observed between the 
 Old Syriac", the Old Latin, and those ancient Greek manu- 
 scripts, which were undoubtedly written in the West, as ap- 
 pears from the Latin translations with which they are accom- 
 panied. The wonderful harmony between the two most ancient 
 versions of the New Testament, one of which was spread 
 throughout Europe and the North of Africa, the other propa- 
 gated from Edessa to China, could have no other cause than 
 a similarity of the Greek manuscripts in the West of Europe 
 and the East of Asia, which must have deviated in an equal 
 degree from our printed text and the manuscripts of what is 
 called the Greek edition'." 
 
 From the foregoing it appears that Michaelis had attempted 
 to explain the Western readings by means of reactions from 
 the Syriac (and he refers his readers for further details of the 
 theory to his Gurae, pp. 169 — 173), but that he clearly held 
 the theory very doubtfully, regarding it at best as a partial ex- 
 planation, and that he settled down into what has been, since 
 his day, the accepted theory, that the Western readings are a 
 bifurcation in the primitive Greek text. Apparently he did 
 not exclude the idea of some Syriac reaction on the Greek ; 
 for he says (Vol. i. p. 321), "It is not improbable that the 
 Syriac and Coptic versions have had some influence on the 
 Greek copies of the New Testament." 
 
 1 This does not seem to agree with the previous statement that " those Syrians 
 who were acquainted with Greek were undoubtedly ignorant of Latin." 
 
 2 By the Old Syriao, Michaelis does not mean what is implied in that term 
 to modern ears; he is speaking of the Peshito, in contradistinction from the 
 Philoxenian version. 
 
 ' He means the Eeoensio Oonstantinopolitana of Griesbach (as Marsh 
 explains). 
 
16 THE OLD SVEIAC TEXT OF THE ACTS. 
 
 His commentator, Marsh, observes that Michaelis' theory 
 would require that the Syriac version must have been made 
 before the end of the first century (in order that it might be 
 employed in the structure of the primitive Latin rendering): 
 but he regards the premises as resting on very unstable ground, 
 denies the necessity for supposing that a primitive Syriac ver- 
 sion earlier than the Latin existed, or the probability that, if it 
 existed, it would have been employed so as to colour the 
 early Latin translations. No doubt the connexion between 
 the Peshito and the Old Latin was not an easy one to esta- 
 blish'; nor does it seem that the attempt to forge the critical 
 link between the two versions has been successfully re- 
 attempted since the publication of the Curetonian fragments. 
 
 Now in what respect does Mr Chase's theory differ from 
 the obscure and somewhat self-contradictory statements of 
 Michaelis, or the suggestion of Schulz that the Codex Bezae 
 has been under direct Syriac influence? 
 
 In the first place, it involves the substitution of the Old 
 Syriac (in the modern sense) for the Peshito ; this step was 
 an obvious one, if the text of the Gospels was to be handled 
 in the light of a proposed theory of Syriac reaction ; for it is 
 in the Curetonian fragments and in the recovered Tatian 
 Harmony that we find those decisive proofs of the agree- 
 ment between the ancient Eastern and Western texts, which 
 was at first suggested by the comparison between the Peshito 
 and the Old Latin versions. 
 
 But, in the second place, Mr Chase does not choose as the 
 ground of his re-statement of the theory of Syriacization, the 
 text of the Four Gospels, in which it was possible to reason 
 from the Old Syriac of Cureton, and the Old Syriac quotations 
 in Ephrem, Aphraates and other Syriac writers to the early 
 Western text of the Codex Bezae and the Latins ; but he 
 chooses for his ground of debate the text of the Acts of the 
 
 ' The support brought to Michaelis' theory by D. Schulz in 1827 {Disp. 
 de Cod. Cant.) was not sufficient to bring the hypothesis into public favour or 
 reception; it consisted chiefly in laying emphasis on coincidences in reading 
 between D and the Peshito version. Michaelis refers also to a work by Storr 
 which I have not seen {Observ. sup. ver. Syr. Stuttgart, 1772). 
 
THE OLD SYRIAC TEXT OF THE ACTS. 17 
 
 \.postles, where, up to the present time, no evidence of an Old 
 3yriac text has been forthcoming, and starts on his enquiry 
 vith the hypothesis that there once existed an Old Syriac 
 rersion of the Acts. Such a hypothesis is, in his view, adequate 
 o explain the Western readings which are so thickly strewn in 
 he Bezan text of the Acts, which readings bear in themselves, 
 iiccording to him, the marks of derivation from a Syriac 
 )riginal. 
 
 The boldness of this hypothesis is evident; and it has 
 laturally provoked opposition. The question is immediately 
 isked, ' Why do you not test the theory of Syriacization in the 
 jospels where the Old Syriac does in great part exist [and 
 ive might add, where it has lately come to light in an almost 
 ;omplete form], instead of flying off in search of an Old Syriac 
 ;ext which is not yet known to exist ? ' Mr Chase's answer is 
 iwo-fold ; first, that he is especially interested in the text of the 
 A-cts ; second, that the intrusive phenomena in the text which 
 le has to explain are more decided in the Bezan Acts than in 
 ihe Bezan Gospels. We might add that the wisdom of the 
 choice of ground is also seen in the fact that we are in the 
 A.cts of the Apostles free from some of the disturbing factors 
 ivhich occur in the text of the Gospels ; the assimilations of one 
 Grospel to another do not obtrude themselves on the reasoning, 
 md the probability of Aramaic elements in the sources (that 
 leath-trap for the man who is calculating Syriac influences) is, 
 ;o say the least, much smaller in the Acts than in the Gospels, 
 md may, perhaps, be entirely absent. 
 
 In this sense, then, the ground is wisely chosen ; but what 
 )f the hypothesis, which is to explain the phenomena of the 
 ;ext, and to be accepted as a true hypothesis on the ground 
 ;hat it does so explain them ? Naturally the first question 
 ;hat would be asked by a critic would be whether there was 
 my tangible evidence for the existence of an Old Syriac version 
 )f the Acts of the Apostles; it ought to be possible, for instance, 
 io demonstrate the existence of such a text, either from the 
 [notations in the Homilies of Aphrahat, or from the works 
 )f Ephrem, or by making a scientific demonstration that either 
 ;he Peshito, or the primitive form of the Philoxenian version, 
 HA. 2 
 
18 THE OLD SYRIAC TEXT OF THE ACTS. 
 
 leans on an earlier text of which it constitutes the revision. 
 Unfortunately no attempt seems to have been made either 
 by Mr Chase or any one who preceded him to clear up these 
 points. The five places where Aphrahat quotes the Acts do 
 not furnish any satisfactory evidence on the point ; not a single 
 one of the 190 glosses which in my first studies I selected from 
 the text of the Acts for special examination has, if we may 
 judge from Tischendorf's apparatus, the attestation of either of 
 the great Syrian fathers in question ; and as to the known 
 versions, we have not yet succeeded in getting behind the text 
 of Thomas of Heraclea to that of his predecessor Philoxenus of 
 Mabug, much less have we been able to analyze this important 
 textual nucleus into its primitive parts : and it is the same 
 with that imperfectly studied version, the Peshito. 
 
 It is, however, certain that an Old Syriac text of the Acts 
 did exist, and that Mr Chase's hypothesis can be removed into 
 the region of facts. We are in a position to prove that 
 the Old Syriac text is, in a certain sense, extant, and has been 
 before the public for more than half a century. The demon- 
 stration which we are going to give of the existence of this 
 ancient text is of great critical importance, and while we must 
 not conclude from the fact of the text's existence that it was 
 necessarily the source of the Western variants it will certainly 
 help us towards the final solution of the question. 
 
 Precisely as the commentary of Ephrem on the Harmony of 
 the Gospels, which now attracts such constant study and is the 
 centre of such lively critical interest, lay dormant in the 
 Armenian text published by the fathers of the Monastery 
 of S. Lazaro at Venice, until it was made accessible by the 
 Latin translation of Mosinger, so have certain other works of 
 Ephrem been blushing unseen and wasting their sweetness on 
 what is, critically at all events, a desert air. 
 
 The first to be noticed amongst these translated works of 
 Ephrem which the Armenian has preserved are his commen- 
 taries on the Pauline Epistles ; which have, in the course of 
 the last few months, appeared in a Latin dress, and so have 
 become accessible to general criticism. As soon as the book 
 appeared it was made the subject of review in three brilliant 
 
THE OLD SYHIAC TEXT OF THE ACTS. 19 
 
 articles in the Theologisches Literaturhlatt by Zahn, who had 
 long ago petitioned for its publication in Latin' 
 
 Zahn pointed out the peculiar features in the commentary 
 which identified it as a work of the same hand as had com- 
 mented on the Diatessaron, and the peculiar features of the 
 text which characterized it as belonging to what is called the 
 Western tradition, and as being an older form of the Syriac 
 than that which is found in the Peshito. 
 
 The most interesting of all the peculiar readings in the new 
 text is perhaps the expansion of the anti- Judaic verses in 
 Gal. iv. 21—27. 
 
 The text and commentary of these verses is as follows (we 
 print them without distinction of type on account of the diffi- 
 culty of distinguishing, in a text which has gone through some 
 process of glossing, the commentary from the text). 
 
 Hae vero fuerunt symbola duorum testamentorum. Una 
 populi .Judaeorum secundum legem in servitute generans ad 
 similitudinem eiusdem Agar. 
 
 Agar enim ipsa est mons Sina in Arabia ; est autem ilia 
 similitude huius Jerusalem, quia in subjectione est, et una cum 
 filiis suis servit Romanis. 
 
 Superior autem Jerusalem libera est, sicut Sara ; et eminet 
 supra omnes potestates ac principatus. Ipsa est Mater nostra, 
 Ecclesia Sancta, quam confessi sumus. 
 
 When we compare this passage with the current Greek 
 text, or with the critical apparatus of the New Testament, 
 there is not at first sight anything that suggests a very different 
 text to the common text of the epistle ; but when we turn to 
 Tertullian's fourth book against Marcion, or to Zahn's recon' 
 structed text of the epistle as used by Marcion, we find, to our 
 surprise, that a large part of the apparent commentary is part 
 of the text of Marcion ; and since there is no reason for doubt- 
 ing Tertullian's tradition of this text, nor for supposing that 
 Ephrem's text has had any connexion with a faulty interpreta- 
 tion made by Tertullian, we have no other alternative than to 
 conclude that Ephrem is commenting on a Marcionized text. 
 
 ' Tlieol. Lit.-Blatt. for Septr. 29, 1893 and two following weeks. 
 
 2—2 
 
20 THE OLD SYRIAC TEXT OF THE ACTS. 
 
 That this is really the case appears, as we have said, from the 
 language of Tertullian (Adv. Marc. IV. 4) : 
 
 " unum a monte Sina in synagogam Judaeorum secundum 
 legem generans in servitutem, aliud super omnem principatum 
 generans, vim, dominationem et omne nomen quod nominatur 
 noa tantum in hoc aevo sed et in futuro, quae est mater nostra 
 in quam repromisimus sanctam ecclesiam." 
 
 It will be seen that TertuUian's text, like Ephrem's, had 
 incorporated a passage from the Epistle to the Ephesians (i. 21), 
 describing the Church, in sufficiently bold language (but which 
 can be justified by a little interpretation), as seated, with the 
 Lord, far above all principality and power. And the explana- 
 tion is also added, both by Tertullian and Ephrem, that this 
 Upper Jerusalem who is our Mother is the Holy Church, whom 
 we have confessed. True, TertuUian's text differs at first sight 
 in that it uses the word 'repromisimus'; but this may very 
 well have been due to a variant rendering of a primitive thfidXo- 
 yt^KUfiev, which may equally mean ' to promise ' and ' to con- 
 fess'.' 
 
 This remarkable passage then formed a part of Marcion's 
 text; it may well startle us as a textual phenomenon, not only 
 for its. own sake as indicating a very free handling of the 
 biblical text for dogmatic ends, but also as containing a 
 reference to the Symbol of the Faith, in the Old Roman form 
 (credo in sanctam ecclesiam). That it is the hand of Marcion 
 we do not doubt, not only because we have Tertullian directly 
 in evidence on the subject, but also because there is no passage 
 in the Epistles that would be more satisfactory to his anti- 
 Judaic mind. There can be little doubt that the verses in 
 Galatians were stock quotations with Marcion and his followers. 
 We find also that the passage affirming the true Mother and 
 
 ' For a similar instance take Acts vii. 17 where the original Greek appears 
 to have been 
 
 TTJs ^irayyeXias ^s ufioXoyriffev 
 for which the Bezan Latin is 
 
 promissionis quam poUicitus est, 
 and the Bezan Greek shows the alternative reading iTrjyyel^aro 
 
THE OLD SYRIAC TEXT OF THE ACTS. 21 
 
 real Jerusalem to be the Church became one of the watchwords 
 of the Paulician heresy which derives so much from Marcion. 
 When the question came up amongst them as to the degree of 
 honour to be given to the Mother of God, they used to say, 
 " the true Theotokos is the heavenly Jerusalem, the Mother of 
 believers." 
 
 But while we do not doubt that we have here the hand of 
 Marcion, we hardly expected to find biblical evidence from 
 before the middle of the second century for the currency of one 
 of the Articles of the Old Roman Symbol, in the form which 
 preceded the conventional ' holy catholic church '.' 
 
 While we are drawing attention to the newly-published 
 commentary and to its Marcionite reading it may be worth 
 while to examine whether in a consecutive commentary on the 
 Epistles, which shows Marcionite influence, we find any sugges- 
 tions of the same arrangement of the Epistles as was found in 
 Marcion's Apostolicon. If we may judge from Tertullian, the 
 epistles stood in the order 
 
 Galatians, 1 and 2 Corinthians, Romans, 1 and 2 Thess., Ephes., 
 Col., Phil, Philem. 
 
 The Armenian Commentary is arranged according to the 
 current usage, which Zahn calls the Alexandrian usage, but 
 there are suspicious traces of its having been re-arranged by 
 the Armenian translator. In the opening of the epistle to the 
 Romans, Ephrem adds to the text " that I may impart unto 
 you some spiritual grace " the words " as I have done to your 
 
 1 The natural suggestion was made by Zahn (1. c. col. 465) that the text of 
 the Epistles upon which Ephrem is commenting was, like that of the Gospel, a 
 text which had passed through Tatian's own hands, and which may have been 
 brought back by him from Eome to Edessa, and have furnished the text from 
 which the first Syriac version of the Pauline Epistles was made. " Wenn 
 Marcion um 145 in Eom einen catholischen Text der Paulusbriefe in der Hand 
 gehabt hat, weloher in sehr auffaUigen Punkten mit dem altesten erreiohbaren 
 syrischen Text zusammentrifft, so weiss ioh dafiir keine andere annehmbare 
 Erkliirung, als dass der erste syrische Uebersetzer der Paulusbriefe eine im 
 Abendland geschriebene Handschrift seiner Arbeit zu Grunde gelegt hat. Am 
 einfaehsten bleibt die Annahme, dass der von Rom nach Mesopotamien 
 heimgekehrte Tatian seinen Landsleuten den ersten 'Apostolos' wie das erste 
 ' Evangelium ' in ihrer Sprache gegeben hat." 
 
22 THE OLD SYRIAG TEXT OF THE ACTS. 
 
 fellows the Qalatians and Corinthians." The suggestion is that 
 the commentator, who frequently poses as Paul, observes the 
 order, Galatians, Corinthians, Komans, and the suspicion is 
 confirmed when we find at the beginning of the epistle to the 
 Hebrews the words : 
 
 " Cum nee in epistolis scriptis ad Galatos, nee Corinthios, et 
 ad proximos quos viderat, id fecerit, neque in epistolis ad 
 Eomanos datis, et ad caeteros quos non viderat, tale quoddam 
 egerit." 
 
 We may fairly conclude that the Epistles, at all events the 
 first ones, stood in the order which they occupied in Marcion's 
 Apostolicon. But Zahn points out that in the preface to 
 Philippians, Ephrem intimates that Colossians is to follow, 
 whereas in Marcion's order, Colossians preceded. I cannot, 
 however, in view of what has been noted as to the priority of 
 Galatians, believe he is right in saying that the order of the 
 Epistles in the Armenian text is certainly that of Ephrem. 
 
 Leaving now on one side the demonstration which Zahn 
 makes of the thoroughly Western character of the text of the 
 Epistles commented on by Ephrem, we pass on to the question 
 which we proposed to examine : viz. the existence or non- 
 existence of an Old Syriac text of the Acts. Are there any 
 traces of such a text in the Commentary on the Epistles ? 
 
 We must premise that in dealing with a question of this 
 kind which has to be resolved by the study of an Armenian 
 translation, we shall never be safe in concluding from the 
 existence of certain readings in the Armenian text to their 
 existence in the lost Syriac original unless the text vary from 
 the popular Armenian; for the simple reason that the 
 translator accommodates his translation sometimes consciously 
 and sometimes unconsciously to the Armenian Vulgate, which 
 has undergone revision from the Latin. We need, therefore, to 
 be very careful with our steps in those cases where the 
 Armenian book before us agrees with the Armenian Vulgate, 
 Let us, then, ask the question whether Ephrem in this 
 recovered commentary makes any quotations from the Acts of 
 the Apostles, and what sort of text is involved in the Syriac 
 
THE OLD SYBIAC TEXT OF THE ACTS. 23 
 
 of such quotations. We will begin by taking a passage from 
 2 Tim. iii. 11, where S. Paul reminds Timothy of the persecu- 
 tions which happened to him in Antioch, Iconium, Lystra. 
 Ephrem begins by explaining that it is Antioch of Phrygia, not 
 Antioch of Syria that is meant: Antiochia autem non ista 
 Syriae, sed ilia Phrygiae ; ubi excitarunt Judaei rectores 
 civitatis, et mulieres divites, et fecerunt tribulationem magnam 
 super eos, expulsis eis extra fines suos. 
 
 For this account we turn to Acts xiii. 50, where the text is 
 
 01, Be 'lovBaloi Trapunpvvav ra? (re^ofieva'i yvvalica<i ra? evayrj- 
 /Movwi Koi roii^ irpwrovi rrji; TroXew?, Kal eTnjjeipav Stcoyfiov 
 i-jrl rov HavXov Kal BapvdlSav Kal i^e/Sa'Xov avToi/'i diro t&v 
 opmv avToov. 
 
 Ephrem's text agrees with the Peshito in translating 
 eva'XVfJ'Ova'i by divites ; but the expression ' fecerunt tribula- 
 tionem ' appears in the Bezan version of the passage which has 
 
 6\eiyjreiv fieyaKrjv Kal Siaiy/iov. 
 
 Ephrem goes on to tell us that " Iconii autem post 
 anteriorem tribulationem suscitarunt persecutionem, Judaei et 
 Gentiles, et lapidantes eum ac Barnabam, ejecerunt illos a 
 civitate." 
 
 The common text of the Acts knows nothing of two persecu- 
 tions at Iconium, nor of any actual stoning of Paul and Barna- 
 bas, yet something like a previous persecution is implied in the 
 common text of Acts xiv. 2, where ' the Jews that were dis- 
 obedient stirred up and evilly affected the minds of the Gentiles 
 against the brethren.' Neither does the account say that Paul 
 and Barnabas were expelled from the city. 
 
 When we turn to the Bezan text, we find first of all that 
 two distinct Iconian persecutions are given, the first being con- 
 cluded by the intimation that ' the Lord promptly gave peace ' ; 
 the second stage of the persecution does not shew any actual 
 stoning on the part of the Iconians, at least not in the Greek 
 text, but when we turn to the Latin, which so often is superior 
 to the Greek in archaism, we find 
 
24 THE OLD syRIAC TEXT OF THE ACTS. 
 
 ut autem factum est impetus gentilium 
 et iudaeorum cum magistribus ipsorum 
 et iniuriauerunt et lapidauerunt eos 
 
 It may, of course, be said that the last line is simply a tran- 
 scriptional error for 
 
 ut iniuriarent et lapidarent; 
 
 but we notice that it is in agreement with the text of Ephrem, 
 and suspect the Bezan Latin to be more archaic than its 
 Greek. (Cf. also the Laudian Latin : et contumeliis adficerunt 
 eos et lapidarent (sic).) 
 
 In 1 Cor. xiv. 23 (p. 77) Ephrem says " de Apostolis 
 dixerunt eos musto plenos inebriates esse " ; the combination of 
 the two passages involved is perfectly natural, but there is 
 reason for believing it to have been in the Old Syriac, since a 
 slightly different form of the combination is in the Peshito (of 
 Acts ii. 13), which reads " They have drunk new wine and are 
 intoxicated." 
 
 A more striking case will be Ephesians iv. 10 (p. 150), 
 where the writer has not only quoted the text of the Acts, but 
 incorporated two of the famous Western glosses (cf. Acts i. 5 in 
 Cod. Bezae) : 
 
 ' Qui descendit, ipse est et qui ascendit super omnes caelos, 
 id est, super omnes altitudines caelorum ; ut impleret omnia 
 quae dixit; istud est, quod dixerat; quain recipitis vos non post 
 multos dies, sed usqite ad Pentecosten.' 
 
 The addition of these glosses can hardly be due to a later 
 hand than Ephrem ; moreover there is no sign of them in the 
 text of the Acts in the Peshito nor in the Armenian Vulgate ; 
 they are among the glosses for which no Syriac evidence has 
 as yet been forthcoming; in fact their whole attestation, 
 outside the Codex Bezae, seems to consist of certain passages 
 of Augustine and of the Sahidic version. 
 
 The occurrence of this famous gloss from Acts i. 5 in the 
 text of Ephrem must be considered very significant : moreover, 
 the conjunction (sed) by which the gloss is connected with the 
 text is important; the clause occupies the same place in 
 
THE OLD SYRIAC TEXT OF THE ACTS. 25 
 
 Ephrem's text as it does in the Codex Bezae, and it looks as if 
 the motive for the gloss had been in the peculiar Greek order 
 
 oil fiera ttoWo.? ravraf ■^fiipai;, 
 
 instead of /ler oi TroXXas ravTw; rjfjiipa^. 
 
 Whether its origin be in a Greek antithesis or in a trans- 
 lator's expansion or the remark of a commentator, we will not 
 discuss further at present. "What we are occupied with is the 
 existence of "Western elements in the Syriac used by Ephrem. 
 
 Enough has been said to demonstrate this from the Com- 
 mentary of Ephrem on the Pauline Epistles. Mr Chase's 
 hypothesis of the existence of an Old Syriac text of the Acts 
 is therefore a good one whatever may be the origin of the text. 
 We will now pass on to prove it still more conclusively and in 
 another way. 
 
 It would naturally suggest itself to any one who was in 
 search of the Old Syriac of the Acts, to examine the com- 
 mentaries on the Acts made by the Old Syrian fathers. The 
 question then arises, Did Ephrem write any connected com- 
 mentary on the Acts ? If so, why should we spend our time in 
 hunting out stray references to the Acts in commentaries on 
 other books ? 
 
 Unfortunately, though there is reason to believe that 
 Ephrem wrote a commentary on the Acts of the Apostles, 
 there are no traces of it in his published works, as far as 
 I know; there remains, then, the possibility that fragments of 
 it may be preserved in Greek or Syriac Catenae. As far as 
 I have been able to make a search, no Syriac catena on the 
 Acts has come to light ; but happily for our investigation, the 
 Armenians at Venice have published for us^ a complete Catena 
 on the Acts which is either a translation from the Syriac, or 
 was made from materials existing in Armenian, which were 
 derived from the Syriac by translation, and this Catena 
 contains a large number of extracts from Ephrem. The greater 
 part of the book is, however, taken from the writings of 
 Chrysostom. If we assume that the Catena was made, as 
 seems likely, late in the eleventh century, it is probable 
 
 1 Comm. on Act. Apost. Venice, 1839. 
 
26 THE OLD SYEIAC TEXT OF THE ACTS. 
 
 that it was made up out of works existing in Armeaian ; 
 in that case it is not unreasonable to hope that the complete 
 text may some day be recovered in Armenian. But whatever 
 may have been the manner of its composition we shall be 
 able (by the kind help of my good friend Mr Conybeare who 
 has responded to my appeal for a translation) to extract a 
 good deal of Epbrem from its pages, and to come to a decided 
 opinion as to the nature of the text upon which Ephrem 
 worked. 
 
 The first thing that will be noticed is that the compiler of 
 the Catena is largely under the influence of the Armenian Vul- 
 gate, so that we shall have to be careful in our interpretations 
 on account of the difficalty of discriminating between the sources 
 of the Armenian text. We will, therefore, bracket the texts 
 from the Catena where they agree closely with the Vulgate. 
 
 The Catena is divided into chapters, and the initial excerpt 
 of each chapter is left unascribed ; perhaps the scribe meant to 
 illuminate the first names and afterwards omitted to do so ; 
 these sections had better be laid on one side. They are very 
 likely Chrysostom's in view of the preponderance of extracts 
 from that writer. 
 
 We shall first discuss some of the more striking sections 
 from the lost Commentary, and then we will subjoin the text 
 of the major part of the Ephrem fragments as an Appendix. 
 
 We will first draw attention to the account of Paul's visit 
 to Philippi, which is, as is well known, much expanded in the 
 Bezan text, and often with great appearance of originality. 
 
 The commentary on Acts xvi. 35 begins as follows (from 
 Ephrem ?) : 
 
 p. 300. " Perhaps the heads of the army knew all the great 
 wonders which had occurred ; and so they did not venture of 
 themselves to release them, but sent to the gaoler to dismiss 
 them, as it were, by stealth. 
 
 p. 301 (Ephr.). The Astaritai were afraid and full of fear, 
 they the mighty of the city, of the earthquake, and knew truly 
 that this earthquake happened on account of them, but they 
 did not undertake to avow it. They sent secretly to bring 
 them out." 
 
THE OLD SYRIAC TEXT OF THE ACTS. 27 
 
 Here it seems clear that the text before Ephrem must have 
 contained a statement very like that in Codex Bezae : 
 
 CYNHAGON OIC CTpATHfOl 
 enl TO AYTO eic thn AfOpAN 
 KAI ANAMNHCeeNTGC 
 TON C6ICM0N TON refONOTA e0OBH9HCAN. 
 
 (Astaritai is probably a misreading of a transliterated 
 Syriac r *^ v , ^'.' \ A ly. w^ ; in fact the word is so transferred in 
 the Peshito.) 
 
 The Commentary then proceeds with vv. 35 — 37 as in the 
 Armenian Vulgate ; then follows (p. 302) 
 
 0. xvi. 39 (Ephr.). " So then that this favour might be 
 unto them, they came and besought of them, saying, We knew 
 not that ye were just, even as the earthquake indeed presaged 
 about you. So then we ask of you this favour, depart from 
 this city, lest the same men gather together after the earth- 
 quake against you, (the same) who before the earthquake were 
 gathered together." 
 
 Cf with this the verse as current in the Codex Bezae : 
 
 €(t)0BHeHCAN K<M ni\pAr6N0M6N0l 
 
 MeT*. (fiiAcoN noAAcoN eic thn c|)yAakhn 
 HApeKAAecAN AYTOYC eieAGeiN emoNTec 
 
 HrNOHCAMEN TA KA9 YMAC 
 OTI 6CTAI ANipeC ilKAIOI 
 
 KAI elArAfONTec 
 
 HApeKAAecAN AYTOYC AeroNTEC 
 
 6K THc noAecdc TAYTHC eleAOATe 
 
 MHnOTe HAAIN CYNCTpACt)CO>CIN HM6IN 
 
 eniKpAZONTec kaB ymoon. 
 
 It is clear that some text very like that of Codex Bezae 
 must have been before Ephrem. 
 
 Turn in the next place to c. xvii. 15 (p. 310) : 
 
 (Ephr.) "So he came as far as the shore, receding. But 
 the Holy Spirit prevented him from preaching lest they should 
 slay him. [And those who conducted Paul, led him as far as 
 
28 THE OLD SYEIAC TEXT OF THE ACTS. 
 
 Athens and having received] from Paul [a command to Silas 
 and Timotheus that they should at once come to him] at 
 Athens. [And they went] to him when they received the 
 command." 
 
 The Bezan text with which we may make comparison is as 
 follows (the chief expansions being bracketed) : 
 
 TON M€N OYN nAY^ON 
 
 01 <\AeAct)oi elAnecreiAAN 
 
 *,neA6eiN eni thn GaAaccan 
 
 YneMeiNSN Ae o ceiAAc km o timo960c eKei 
 
 01 he KATACTAN0NT6C TON HAYAON 
 Hr^rON 6(JC>C AeHNCON 
 
 [nApHAeeN Ae thn eecc^AiAN 
 eKooAYOH r^p eic <\ytoyc 
 
 KHpY5<!>l ton AoTOn] 
 
 AaBontsc Ae eNToAHN [nAp<\ hayAoy] 
 
 npOC TON C6lA*,N KAI TIMOGeON 
 
 onooc eN tax^i eAecociN 
 
 npOC AYTON €lH6CAN. 
 
 It is clear, then, that Ephrem had before him an expanded 
 text like that of D ; the statement that Paul was prevented by 
 the Holy Spirit from preaching in Thessaly must have been in 
 his copy. One of the smaller glosses in the Codex Bezae was 
 also present {irapa TravXov) and perhaps the words eh Ta<; 
 'Ad'^va<: were also in the text. The peculiar expression of 
 Ephrem that ' Paul came to the sea, receding,' is obscure. The 
 words mean literally ' giving way ' (? = ava'^wpcov). It is 
 curious that the Latin of Cod. Bezae has 
 
 abire ad mare uersus 
 
 where ad mare uersiis is perfectly good Latin'. 
 
 Is it possible that this versus has been understood as con- 
 versus or reversus ? Whatever be the origin of the statement 
 there can be no doubt that Ephrem had a Bezan text. 
 
 1 Cp. Caesar, B. G, vi. 33, Labienum ad Ooeanum versus in illas partes 
 proficisci iubet. 
 
THE OLD STKIAC TEXT OF THE ACTS. 29 
 
 One more proof shall be given : from Acts xix. 38, 39, 
 p. 352 [Ephr. ?], " This Demetrius, vile and shameless, he says, 
 he and the children (-TratSe?) of his craft, if they have any suit 
 with one another, let them stand forward and make it clear to 
 the hegemon. And [if there be] any other [enquiry let] it be 
 [pronounced on in the lawful assembly]. 
 
 We compare as before 
 
 ei MeN GYN iHMHTplOC [OYTOc] 
 01 KAI CYN AYTCO T€XN6ITe 
 
 6X0YCI npoc [aytoyc] tina AoroN 
 
 ArOpAIOI ArONT*.! KAI ANeYnATOI 6ICIN 
 6NK(\AlTC0C*,N AAAhAOIC 
 
 ei Ae nepi erepcoN eniZHTeire 
 
 eN TOO NOMOO €KKAHCIA 8niAY9HC€TAI. 
 
 Here Ephrem has the added ovro^ of the Bezan text ; he 
 has also the added avrov<i which has been understood as 
 eavToixi, as if Demetrius and his fellows might have quarrels 
 inter se ; further he has read irepanepo) (which surely must 
 be original) as Trepi erepcov which we find in D. (This variant 
 is therefore a primitive Greek error on the part of a copyist.) 
 It is clear, therefore, and instances might be abundantly 
 multiplied, that Ephrem's text was in the later chapters of 
 the Acts closely connected with that of the Codex Bezae^ 
 
 Another passage in the commentary on the Pauline Epistles 
 which invites study, but from which it is not easy to draw very 
 definite conclusions, is 1 Cor. ii. 8, where Ephrem remarks : 
 
 Id est, quod apostoli dixerunt, " Scimus, quia per errorem 
 deceptionis haec fecistis ; convertimini igitur, et poenitemini, 
 et nemo id vobis reputabit ad peccatum." 
 
 ' The last extract seems to be acephalous, but it evidently belongs to the 
 same fabric as the others. The Peshito helps us to restore the original Syriac 
 
 It looks as if oBros were due to a wrong line-division ; we should read 
 
 61 MEN OYN AHMHTplOC 
 
 OYTOC KM 01 CYN AYTW TeXNeiTAI 
 
30 THE OLD STEIAC TEXT OF THE ACTS. 
 
 The reference is to Acts iii. 17: and although it is not quite 
 easy to detach the text of the passage from its setting, there 
 seems to be no reason to doubt that the text of Ephrem had the 
 Western reading which appears in Cod. Bezae as eTna-Td/Meda 
 (d" = scimiis) and instead of olBa (where Irenaeus has scio). 
 
 The origin of the variant appears to be Greek, and to be an 
 attempt to avoid the confusion caused by the presence in the 
 text of fih after oIBa, which might be read as olBa fiev or as 
 oiSafiev. The existence of this fiev is evidenced by the double 
 fact that Cod. Bezae has carried it into the next line (v/xel<; ixev), 
 and instead of olBafiev has the equivalent iwicrTd^eda. 
 
 We notice also that the text used by Ephrem did not 
 contain the expansion at the end of the expression 
 
 Kara dyvoiav eTrpd^are [Trovrjpov^, 
 
 which is found in the Codex Bezae, in Irenaeus and elsewhere, 
 but some simpler expansion, probably the same as occurs in the 
 Peshito, 'ye did this (p^.ttn).' 
 
 The rest of the verse as quoted by Ephrem is obscure and 
 paraphrastic, and may be from his own hand. On the whole 
 the text seems to be Western in character, but not as decidedly 
 as we should have expected. 
 
 Acts XX. 29 is also quoted by Ephrem in the introduction 
 to the apocryphal 3rd Epistle to the Corinthians, but appa- 
 rently in the terms of the Peshito'. 
 
 Probably we should also notice the commentary of Ephrem 
 on Rom. viii. 7 (p. 26), in which he contrasts the imperfection 
 of the law with the fulness of the Gospel : he says that Chris- 
 tians really do obey the law, 
 
 " etsi circumcisi, ac sabbati observatores, existiment nos 
 adversarios esse legis, eo quod superflua ilia legis soluta sunt 
 desuper. Si autem ilia occidisset atque salvasset, oportuisset 
 
 1 But observe that ' ut convertant auditores ad sequendos se ' where the 
 Greek text is toS dirocrirdv is in agreement with Irenaeus ' ut convertant ' and 
 Cod. Bezae toO iwo<!Tpi(t>eiv as well as with the Peshito 
 
 .__oeniix3 »_^\\p0.i K'ri^baX^ ,.__OAacn_a."i vyrti* 
 
 We have also an allusion to the story of Elymas on p. 247, but without any Old 
 Syriac traces, as far as I see. 
 
THE OLD SYRTAC TEXT OF THE ACTS. 31 
 
 ea.m prius in cordibus inscriptam fiiisse. Tu autem vide, quia 
 neque justitia plena est ilia, neque in ea est ilia justitia, quae 
 dicit, Quod tibi malum videtur, proximo ne facias." 
 
 The argument is that circumcision is a superfluous part of 
 the law which has been abolished. If it had been a part of the 
 true law which kills and makes alive, it would have had to be 
 heart-circumcision. But the old imperfect statement of the 
 law did not contain the precept to do nothing to the neighbour 
 which we should ourselves dislike. 
 
 It seems not unreasonable to enquire whether Ephrem in 
 his text of Acts xv. (vv. 20 and 29) may not have had the 
 addition of the famous negative precept to the Jerusalem 
 Concordat. This is an important and interesting question, 
 inasmuch as the reading is perhaps the oldest reading extant 
 of those which are called Western. It has been pointed out by 
 Seeberg' that the interpolation in Acts xv. 29 must have been 
 in the text of the Acts used by Aristides the Apologist: for 
 Aristides tells us in his summary of the early Christian ethics, 
 that " they do not worship idols in the form of man ; and what- 
 ever they do not wish that others should do to them, they do 
 not practise towards any one ; and they do not eat of the meats 
 of idol sacrifices, for they are undefiled." The apparent want of 
 sequence in the precepts is explained at once by a reference to 
 the interpolated passage in the Acts in which the negative 
 Golden Rule is made a pendent to the regulations against 
 eating idol-meats, &c. Accordingly Seeberg says, and I do 
 not see that exception can be taken to his reasoning (except by 
 denying the genuineness of the Syriac text) that 
 
 " Hieraus folgt deutlich, dass Aristides den Spruch nicht in 
 der Form der Didache, sondern in der in das N. T. iibergegan- 
 genen Form gekannt hat. Da er nun den Spruch mit der 
 Enthaltung von den elBcoXoOvra zusammen anfurt, so kann 
 nicht bezweifelt warden, dass er in seinem Text der Apostel- 
 geschichten diesen Spruch, wie Irenaus, bereits gelesen hat. 
 
 Dann ist Aristides der alteste Zeuge fiir diese Interpolation 
 
 Bald darauf folgt Ubrigens bei Aristides die Enthaltung von der 
 1 Die Apologie des Aristides, p. 213. 
 
32 THE OLD SYRIAC TEXT Or THE ACTS. 
 
 avvovaia dvo/jLo<; (cf. die iropveia der Apgesch.). Diese Stelle 
 erweist also sowol die kirchliche Beniitzung der Apgesch. zur 
 Zeit des Aristides als das Vorhandensein der Interpolation in 
 der Mitte des 2. Jarh." 
 
 The genuineness of the Syriac text appears further to be 
 established by the consideration that no reason can be assigned 
 for the insertion of the precept by a translation at a point 
 where its connexion with the context is not at first sight obvious, 
 as well as by the reflection that the passage, if genuine, would 
 be out of date and almost unintelligible to a literary pirate in 
 the seventh century \ 
 
 Whatever, then, be the date of the first appearance of the 
 variant in the text of the Acts, we are sure that it was extant 
 very early, and need not be surprised if we should find it 
 current in the text commented on by Ephrem. We do not, 
 however, wish to speak too positively as to the source of the 
 quotation in Ephrem : and that for various reasons : the 
 negative precept turns up everywhere in the early Church, 
 having been absorbed, in the first instance, from Jewish ethics. 
 Moreover it seems likely that it was not only interpolated into 
 the Acts, but, if we may judge from certain remarks of Ter- 
 tullian against Marcion, it also was current in Marcionized 
 copies of the Gospel of Luke. Further the form in which 
 
 1 Mr Chase, I observe, quotes the incorrect Greek of the Apology, and so 
 avoids the conclusions of the foregoing argument ; and explains the occurrence of 
 the negative precept in Aristides as a case of apologetic absorption from the text of 
 some form of the AiSaxv- But even in the incorrect text, the connexion between 
 Aristides and the interpolated Acts is so close that he is forced to admit that 
 " from such an apologetic passage the saying naturally passed into a similar 
 context in Acts xv." [This is dangerously near to the admission of a Greek 
 original for the gloss. Did the Apologists write in Syriac?] He then makes 
 a laboured and obscure argument to prove that after the passage had been 
 absorbed into the text of the Acts from Aristides or some similar Apologist, it 
 passed into the text of Theophilus of Antioch from the text of the Acts through 
 the medium of a Syriac version (cp. Theoph. ad Autol. ii. 34). The difference 
 in the treatment of the two cases is, we may conjecture, due to the fact that 
 Mr Chase wishes to go to Antioch for the origin of his textual corruptions ; and 
 does not wish to go to Athens ! He deals in a somewhat similar manner with the 
 Western text of the Acts which is quoted in Polyoarp's Epistle to the Philippians : 
 if it had only been Ignatius ! (Cf. Chase, p. 21 on Acts ii. 24.) 
 
THE OLD STRIAC TEXT OF THE ACTS. 33 
 
 Ephrem quotes is not the exact form in the Acts ; he uses the 
 expression "quod tibi malum videtur" which is much nearer 
 to the Talmudic form of the precept : and the same peculiarity 
 appears where Ephrem quotes the precept in Romans iii. .21 ; 
 where it is expressly called, in opposition to the statement 
 previously quoted', a precept of the law (aut ipsam legem 
 docere mansuetudinem et fidem ; ut exempli gratia quum 
 dicit : quod tibi malum videtur ne aliis feceris) : we should not 
 then feel justified in employing the passage quoted from Rom. 
 viii. as a proof that Ephrem had the famous interpolation in 
 his copy of the Acts. 
 
 From two separate lines of enquiry, therefore, we have 
 discussed the question of the existence of an Old Syriac text 
 of the Acts, and have removed Mr Chase's hypothesis into the 
 region of fact. Setting on one side the question as to what 
 the result of this discovery will be upon the criticism of the 
 text, and it cannot fail to be far-reaching, we can only most 
 cordially congratulate Mr Chase on the complete and thorough 
 verification of the assumption with which he commences his 
 investigation into the peculiarities of the Western text. It is 
 not often that a speculation is so rapidly justified from un- 
 expected quarters". It remains to be seen whether the reason- 
 
 1 Ephrem is no model of consistent interpretation ; he loves alternatives : 
 the aXXws whose equivalent is employed so often in his works is his own, and 
 not the suggestion of a later hand. 
 
 ^ Of course I am aware that Mr Chase desiderates in the working out of his 
 theory, not merely one old Syriac text, but many : in one single passage he 
 requires sometimes as many as three separate versions ! He justifies this view 
 of the variety of the primitive Syriac texts by quoting the following remarks of 
 Dr Hort with regard to the Curetonian text of the Gospels. " The rapid varia- 
 tion which we know the Greek and Latin texts to have undergone in the earlier 
 centuries could hardly be absent in Syria; so that a single MS. cannot be 
 expected to tell us more of the Old Syriac generally than we could learn from 
 any one average Old Latin MS. respecting Old Latin texts generally." Mr 
 Chase does not notice that when he has assigned the Syriae version as the cause 
 of the Greek and Latin Western Variants, these remarks of Dr Hort no longer 
 apply. The comparison in that case between the progressive changes of the 
 Syriac and those of the Graeco-Latin texts must be made between the Syriac 
 version and the Western texts considered as unaffected by the Syriac version, if 
 any analogy between the two sets of phenomena is to hold good. But on Mr 
 Chase's theory the variation of Graeco-Latin texts is almost nil when the 
 Syriac reactions are removed, 
 
 HA. 3 
 
34 EPHREM 
 
 ing which he has based upon the hypothesis can also be 
 justified. 
 
 We will now add as an appendix the more important of the 
 extracts which we have been able to collect of the commentary 
 of Ephrem on the Acts. 
 
 APPENDIX. 
 
 Extracts from the Armenian Version of Ephbem's 
 COMMENTARY ON THE AcTS. Translated by F. C. CoNY- 
 BEARE, with some additional notes. 
 
 p. 13 (Ephrem). The author of the Acts of the Apostles 
 was Luke the Evangelist. He was not indeed with Christ 
 from the commencement of his preaching, but he attached 
 himself to the apostles of Christ from the very first descent 
 of the Spirit and before. And although his gospel was only 
 written by him according as he heard from the apostles of 
 Christ, yet of the Acts of the Apostles which he wrote he 
 was himself an eye-witness. He wrote his Gospel, because he 
 saw that certain impostors had written out of their heads a 
 gospel under the name of " the infancy of Christ our Lord," 
 and other books of questions (hartzouadzots, but ? read herd- 
 zouadzots = of heresies) under the name of Mary and of the 
 disciples of Christ, in which they say that after the resurrection 
 that first-born one ascended after 18 months; whereas the 
 disciples write about him that he after the fortieth day exactly 
 ascended into heaven. Luke then in order to hinder the false 
 books of heterodox writers by {? or from) the gospel of our Lord 
 Jesus Christ, who concoct about the Lord Jesus an old age 
 and a youth of works (narrating some things before his baptism 
 and others after his ascension on the fortieth day) — therefore 
 he sets in his book of the Acts of the Apostles a beginning 
 and an end of the works of our Lord, in imitation of the other 
 Evangelists, beginning from the baptism of the Lord by John, 
 and continuing to his ascension on the 40th day : in order 
 to shew that every work whatever ascribed to Christ earlier 
 than his baptism and subsequent to his ascension after 40 
 days, is a work alien to Christ our Lord, And it is clear from 
 
ON THE ACTS. 35 
 
 the fact that Christ himself said to his disciples, " If I go not, 
 the Comforter will not come." And the Comforter came on 
 the completion of Pentecost, the 50th da}' after his resurrection. 
 It is, therefore, manifest that on the completion of the 40 
 days' term, as the Apostles say, Jesus ascended. And those 
 impostors are false who say he ascended after 18 months. 
 So Luke wrote about the resurrection of our Lord, about his 
 Ascension and the Coming of the Spirit and the increase of the 
 disciples and about all that followed^. 
 
 p. 19. Acts i. 2 (Ephrem). [Until the day of commanding 
 the Apostles by the Holy Spirit.] As I said above, at the 
 beginning of the Acts of the Lord he also sets a term, saying 
 ' until the day of commanding,' which is the day of his ascent, 
 — in order to silence the liars. 
 
 p. 20 (Ephrem). Now he shewed that he remained after 
 the Cross not without miracles (cnjfiela), but in many miracles 
 (arj/ieia) and many signs (T€Kfji,ijpia) which he wrought in the 
 forty days: as he appeared to them in all likenesses, now 
 known and now unknown: according as in another place it 
 saith: "Their eyes were holden that they should not know 
 him," and "He was made known to them." 
 
 p. 21. Acts i. 4 (Ephrem). Not as having any natural 
 wants, therefore, of food, but making a concession in order to a 
 convincing demonstration of the resurrection. 
 
 p. 22 (Ephrem"). And because they were frightened, first 
 he led them forth into Galilee, that without suspicion they 
 might hear what was said. And when they heard, lo ! for 
 forty days he tarried with them, and commanded them not to 
 leave Jerusalem nor to go forth to preach before receiving the 
 Spirit. As no one allows soldiers to engage in battle before 
 being armed, so he did not allow them to enter the affray and 
 conquer (? be conquered) before the Coming of the Spirit. 
 
 ^ For the doctrine of an Ascension after 18 months see Iienaeus (ed. Mass. 
 p. 14) where the helief is given as a peculiarity of the Valentinians. " xai tous 
 Xot7roi)s SsKOQKTW KlQvas ^avepoui7$aty dta tqv fiera. ttjv ^k veKpup dvatrraffiv SeKaoKTuj 
 Hrjcrl \iyav SiareTpiipivai airbv ffiv rots iiaBTp-ah." See also Ascensio Isaiae 
 (ed. Dillmann, c. ix. p. 43). 
 
 " Almost all of this section will he found in Chrysostom in loc. (ed. Savile, 
 p. 611). 
 
 3—2 
 
36 EPHEEM 
 
 And again because of the many who were about to believe in 
 Jerusalem, he made it necessary for them to abide there; 
 and again that the Jews might not say that they left alone 
 those whom they knew and went forth to strangers, because of 
 their hatred or of glory, and lest being attacked they might 
 run away from them, even for sake of the very erucifiers they 
 give out the tidings of the resurrection in that city, where 
 the unjust slaying of Christ by them took place, in order that 
 the outer heathen might easily believe, seeing the slayers of 
 Christ come to believe in Him, and the erucifiers become 
 preachers of his resurrection. 
 
 But that the disciples might not say : ' how shall we 
 remain among the cruel slayers ? ' nor flee after his removal he 
 dissolves their suspicion by the promise of the Spirit; to 
 first bestow it on them there. For by this hope as with a 
 chain, he will detain them in Jerusalem, sitting and awaiting 
 there the promise of the good news of the Father, who by the 
 prophets saith ' I will pour out of my spirit upon all flesh.' 
 
 p. 24. Acts i. 5 (Ephrem, Chrysostom, sic !). And not 
 only doth he avow himself to be great beyond comparison, but 
 *he shews bis disciples to be greater than John, saying. Ye shall 
 be baptised (? baptise), for they were destined to baptise even 
 others in the Holy Spirit. And he did not say, I baptise you, 
 but. Ye shall be baptised, teaching us to be humble. But that 
 he himself it was that baptised them by the Spirit, is clear 
 from the testimony of John ; for he said ; " He shall baptise 
 you with the Holy Spirit and with fire."* 
 
 And that they received the Spirit in the upper-chamber is 
 clear. * But how saith he, ' ye shall be baptised,' there being 
 no water in the upper-chamber ? I answer that the Spirit is 
 supreme, by which the water also energises (ivepjei). In like 
 manner he himself is called anointed, not with sensible oil 
 indeed, but with the Spirit of joy. And in another fashion 
 (we may explain it) : they had long before been baptised with 
 water by John : for if publicans were baptised, much more they, 
 
 * The words between asterisks are from Chrysostom as may be seen by refer- 
 ence to his published Commentary. The double heading is therefore doubtful. 
 
ON THE ACTS. 37 
 
 whose destiny it was to be baptised and to baptise with the 
 Holy Spirit. For though in our time it is possible to be bap- 
 tised at once with water and Spirit, it was then in the time of 
 the disciples (only possible) in separate times.* 
 
 p. 28 (c. i. 7) (Ephr.). And this with so much firmness, 
 because he willed not to reveal to them these days of his 
 ascending, which they saw with their own eyes. 
 
 p. 30 (c. i. 8) (Ephr.). [For ye shall receive power] and 
 courage at the coming of the Spirit on you. And ye shall go 
 out from the upper-room, and shall be manifest to the world, 
 witnesses of my resurrection and of what ye heard and saw 
 from me not only in Jerusalem, city of crucifiers, where indeed 
 ye are afraid, but also among the Samaritans, and all races. 
 
 p. 31 (c. i. 10) (Ephr.). [And the cloud] hid {or covered) 
 [him from their eyes.] 
 
 p. 34 (c. i. 12) (acephalous but probably Ephrem). 
 
 Then [they returned to Jerusalem from the mountain called 
 of Olives... which is near to Jerusalem according to a Sabbath's 
 journey]. 
 
 (c. i. 13). [And when they entered] Jerusalem, as they 
 received a command not to leave Jerusalem, [they went into 
 the upper-room, where the lodgings of course were, etc.]... But 
 Simeon (Shmawon) the Zealot is by Matthew and Mark called 
 Simon the Cananaean. Perhaps in the Hebrew tongue he is 
 called Zealot. And it is averred by many that he is son of 
 Joseph father of the Lord, and brother of the Lord. Moreover 
 Judas (brother) of Jacob, was brother of the same Simon and son 
 of Joseph, who also was brother of the Lord. This one wrote the 
 Catholic epistle which in his name is called the epistle of Judas, 
 in which at the beginning out of humility instead of calling 
 himself brother of the Lord, he writes brother of Jacob. And 
 hence it is clear that he is the same whom Matthew and Mark 
 call Lebaeus and Thadaeus, so that they and Luke do not 
 respectively name different persons, but only one and the same 
 person by different names. And no wonder if in Hebrew there 
 was a plenty of double names and multiple names, whence the 
 ambiguity in question of the Evangelists as to Thadaeus and 
 Judas is one of name only, not of persons. For of the first set 
 
38 EPHREM 
 
 chosen by Christ not one perished, but only Judas the traitor. 
 It is certain then that the other Thadaeus who was with 
 Abgar was of the Seventy, to which fact their tombs also 
 testify. For Thadaeus, one of the Seventy, died in Armenia in 
 the region Artazon ; but Judas of Jacob who in Matthew and 
 Mark is Thadaeus, one of the Twelve, died in Ormi in Ar- 
 menia. Thus the agreement of the Evangelists as to the names 
 of the Apostles is confirmed. 
 
 p. 38 (c. i. 17) (Ephr.). [because he was in our number along 
 with us and there had fallen to him the lot of this service.] 
 
 p. 42 (c. i. 25) (Ephr.). [From which passed away Judas to 
 go to his own place]... not to that which is full of light, which 
 the Lord promised him, but into darkness. 
 
 p. 45 (c. ii. 2) (Ephr.). A violent sound of a wind came 
 about in the house where were gathered together the disciples 
 of Jesus and a sweet smell was wafted from the violence of 
 the wind and filled all the house'. 
 
 p. 45 (c. ii. 2) (Nyss. Ephr.). [And filled all the house in 
 which they were sitting.] And how did the wind fill the 
 house ? Manifestly with a sweet smell and with a bright 
 light. 
 
 p. 47 (c. ii. 3) (Ephr.). [And it sat upon each of them.] 
 That is to say, the tongues appeared and sat upon one by one 
 of them. It is clear that they severally (eKaa-rai) sat on each, 
 the whole of the parts sitting on one by one of them. For 
 which reason and because of the sameness of the nature, he 
 gathers the whole of the parts into one, and says in the 
 singular : It sat upon each of them. 
 
 p. 49 (c. ii. 6) (Ephr.). [When there was this voice, there 
 came together the crowd and was confused.] The voice which 
 came from heaven was audible to all the citizens. And the 
 smell, which from the violence of the wind was wafted, gathered 
 and brought thither the many. This is the voice which there 
 was. 
 
 1 Compare the following section and ii. 6, also ii. 32. We may suspect 
 that there was something in the text which provoked the comment about 
 the sweet smell. Was it an assimilation to Isaiah vi. ' The house was filled 
 with smoke,' viz. of incense? 
 
ON THR ACTS. 39 
 
 p. 49 (c. ii. 6) (Ephr.). These then are those whom the 
 terrible^ voice moved to fear and the smell of fragrance brought 
 and mustered together — when they saw the Galileans talking 
 in all tongues, were amazed as he says : [For they heard them 
 speak in their own tongues]. 
 
 p. 52 (c. ii. 14) (aceph.). [Ye men etc not as ye think] 
 
 that we are filled with new wine. [For it is but the third 
 hour of the day.] 
 
 p. 55 (c. ii. 20) (Ephr.). For as the dawn is sign of the 
 rising of the sun, so the signs on the day of the cross of Christ 
 are prognostics of the pouring out of the Spirit of God. 
 
 p. 56 (c. ii. 20) (Ephr.). Whose light was given to the 
 heathen and the vapour of smoke for the exacting from them 
 of the requital of the blood of Christ and of the just. And 
 there is darkened upon them the sun before the taking of 
 them into a lake of fire, of which he says, [Until there be 
 come the day of the Lord great and famous]. 
 
 p. 58 (c. ii. 22) (Ephr.). He proclaims him man, that as with 
 milk he may feed them with the Gospel, and so that when 
 they be perfected, they may proclaim him judge, creator and 
 God. 
 
 p. 62 (c. ii. 32) (Ephr.). [To whom all we] are witnesses. 
 And to us are witness the violent voice which breathed and the 
 sweet smell which was wafted and the strange tongues which 
 we speak. 
 
 p. 66 (c. ii. 38) (Ephr.). For the remission which is hidden 
 in his baptism absolves you from lawlessness, for you crucified 
 him. And when ye are absolved and pure, then ye become 
 worthy of the gift of the Spirit which ye saw in us, ye also. 
 And he confirms his argument and says [For to you is the good 
 news and to your children]. Manifest is that good tidings 
 given by Joel, ' I will pour out of my spirit.' 
 
 p. 78 (c. iii. 1) (Ephr.). But some say, because he was 
 inexperienced, and did not know bow to walk, for he had never 
 walked. 
 
 p. 94 (e. iv. 26, 27) (Ephr.). [Because of the Lord and his 
 anointed.] Because in dishonouring Christ they dishonoured 
 ' Probably a misreading of a Syriao text ' the voice of power.' Cp. p. 62, 
 
40 EPHREM 
 
 the Father whom they did not dishonour (?). [For there were 
 gathered indeed in this city against thy holy Son Jesus whom 
 thou anointedst, Herod and Pontius Pilate with tribes and 
 multitudes of Israel to do whatever thy hand and will afore- 
 time determined to come to pass.] 
 
 p. 102 (c. V. 1) (Ephr.). Thus were slain the house of 
 Ananias, not only because they thieved and hid, but because 
 they feared not, wishing to trick those in whom the Holy 
 Spirit that searches all was dwelling. 
 
 p. 115 (c. V. 37) (Ephr.). [After him, he says, arose Judas 
 a Galilean in the days of there being a district-writing, and 
 caused to revolt a great multitude after him.] Satan then 
 raised them up before the birth and at the birth of our Lord. 
 For he heard about his birth from the words of the angel who 
 was with Zachariah and Mariam, and beheld that Simeon the 
 old man was prevented, so as not to taste death till he should 
 see our Lord Jesus Christ, and he was eager by this revolt 
 to damage the plan of Christ. But through his haste as [he], 
 so also this one [was destroyed] and those who [complied] with 
 him [were scattered]. 
 
 p. 127 (c. iv. 13) (Ephr.). But because they ridiculed the 
 apostles as being simple and unlearned, he began to repeat to 
 them the Scriptures, beginning from Abraham he summarises 
 down to Christ and to their shamelessness. 
 
 p. 144 (c. vii. 43) (Ephr.). [Ye took, he says, the tent of 
 Moloch] that is the cause of sacrifice, [and the star of your god 
 Hrempha]...[the images which ye made to worship them]... 
 For because thereof [I will transplant you to the other side of 
 Babylon '].... But even [the tent of witness was with our fathers 
 in the wilderness, as he commanded who spake with Moses, 
 to make it according to the model which he saw]. ...He declares 
 then that all this was so, and they had no temple. Nay more, 
 there being the tent, there were no sacrifices. [Surely ye did 
 not bring to me] victims [and offerings] he says. Mark how, 
 although they had the tent of witness, it helped them nothing, 
 nor the signs that were previous and subsequent. But all the 
 
 1 It will be noticed that here the text [ = Arm. Vulg.] is against that of Cod. 
 D which for (wiKeiva BafivXiSmi has ^Tri to nipii B. 
 
ON THE ACTS. 41 
 
 bones were destroyed and fell to the ground'. And he adds 
 [in the desert,... which our fathers received and bore] suc- 
 ceeding one another [along with Joshua] in the possession of the 
 Canaanites,...[wbom] also [God drove out, from the face of our 
 fathers. . .Until the days of David], he says, there was no temple. 
 He [found grace before God and prayed to find a home for the 
 God of Jacob... Furthermore Solomon built to him a house.] 
 But not that the Highest' dwells in temples made by hands... 
 [the heavens are my throne and the earth the footstool of my 
 feet... What sort of house will ye build me], he says, [or what 
 place of my repose ? For all this did my hand make.] 
 
 p. 146 (c. vii. 51) (Ephr.). [0 ye stiff-necked], he says, [and 
 uncircumcised in heart]. 
 
 p. 152 (c. vii. 59) (Ephr.). [They stoned Stephen who was 
 crying aloud and saying : Lord Jesus, receive my spirit] 
 
 p. 153 (c. viii. 1) (Ephr.). And it is similar that on this day 
 he took their goods as spoil; which things the apostle praises: 
 " Receive with joy the plundering of your goods^" And they all 
 were dispersed into the villages of Judaea and Samaria, except 
 the disciples'. 
 
 p. 154 (c. viii. 3) (Ephr.). [But Saul was doing harm to the 
 Church. From house to house he went, dragged off men and 
 women, threw them into prison.] 
 
 p. 155 (c. viii. 5) (Ephr.). Philip then went down thither and 
 at the power of his signs he filled the land of Samaria with his 
 teaching, on such a scale that Simon Magus also, who startled 
 the Samaritans with his magic, undertook to come down with 
 the Samaritans for the washing of the font, as the Evangelist 
 relates in due order. 
 
 p. 158 (c. viii. 14) (Ephr.). And therefore they sent Peter and 
 John that by their laying on of hands the Samaritans may 
 receive the Spirit of signs and may astonish the children of 
 Jerusalem by the works of the Spirit which the Samaritans 
 
 1 An allusion to 1 Kings xiii. 3 ; or is it the equivalent of ' whose carcases 
 fell in the wilderness '? 
 
 2 Heb. X. 34. 
 
 » Here we should have looked for the Western gloss 'who remained in 
 Jerusalem.' 
 
42 EPHREM 
 
 performed. [Then they laid hands on them and they received 
 the Holy Spirit.] It is clear that making prayer (as has been 
 said) they laid their hands. For the Holy Spirit was not 
 simply given nor could they give it, but there was need of much 
 asking. For it is not the same thing to get healing and to get 
 the power of healing. [Simon having seen that when the 
 Apostles laid on] hands [there was given the Holy Spirit, he 
 gave them money and said : Give also to me this power, that 
 on whomsoever I may lay hands, he may receive the Holy 
 Spirit]. Simon, he says, having seen. But perhaps he did not 
 see that no signs were manifested by the Samaritans... (p. 159) 
 he laid silver before them : why ?... wherefore Peter says to him : 
 [Thy silver be with thee unto destruction] for thou dost not use 
 it as it is right:... [because thou hast thought to obtain the gifts 
 of God by money] thinking little of the freedom of God's gifts... 
 [there is not for thee part and share in that matter... thy heart] 
 he says [is not right before God... Repent thou] he says [of 
 those evil] thoughts [of thine, and pray the Lord that there 
 may be remission to thee of the sinful thoughts of thy heart]... 
 he said that there may be remission to thee of the deceitful 
 thoughts of thy heart and from the bitter bonds of greed in 
 which thou art entrammelleil...[For unto the bitterness of 
 wrath and unto the entanglement of unrighteousness I behold 
 thee]... the magician said [Do ye pray for me unto the Lord, 
 that there come not upon me aught of the things of which ye 
 have spoken']. 
 
 p. 16.3 (c. viii. 27) (Ephr.). But it is likely that on this 
 account he came, for that he received it in succession from the 
 tradition of the queen of the South who came to worship in 
 the temple in the days of Solomon. 
 
 p. 1 66 (c. viii. 40) (Ephr.). Wherefore as he went up out of 
 the font of baptism, there settled forthwith on him the Spirit of 
 the power of works. That by works of the Spirit which he 
 wrought in India, the cross which he preached might be faith- 
 
 1 Here there do not seem to be any signs of the influence of the Western 
 text which we should expect, such as the addition of the word 'evil' in the 
 last line, or the account of Simon's ceaseless weeping. 
 
ON THE ACTS. 43 
 
 fully reverenced^ And an angel of the Lord snatched up Philip 
 and the Eunuch no longer saw him. But in the old copies of 
 the translation he says : the Spirit of the Lord snatched up 
 Philip'': and often he repeats 'the Spirit'; I think because he 
 would make it clear that in the snatching up by the angel of 
 Philip he became invisible to the Eunuch, lest the angel 
 appearing in gross form, as to many in human shape, the 
 Eunuch should think him to be a man. 
 
 p. 168 (c. ix. 2) (Ephr.). But he, as if no one sent him, him- 
 self with obstinate will, [having come to the High Priest, asked 
 of him letters to Damascus to the synagogues; in order that if 
 he find anyone of that way, men or women, he may bring them 
 bound to Jerusailem.] 
 
 p. 169 (c. ix. 3) (Ephr.). With the light then he blinded him 
 and so frightened him and with awful fear of his glory he ex- 
 tinguished his rage, and with gentle voice he mollified him, in 
 which also he was persuaded to confide. And because he feared 
 to contemn the humility of our Lord, who appeared to him with 
 so gentle an utterance, and he was struck with fear of dishonour- 
 ing his might, who by the mighty light startled him. And while 
 he lay prone on the earth, dazed not after the voice but before 
 the voice, lost in wonder as to who from heaven blinded him, 
 for Jesus was not risen from the dead as he thought. But when 
 he said to him in censure ; [Saul, Saul, why persecutest thou 
 me ?] what wrong hast thou suffered from me that thou doest 
 this to me ? he fainted (?) in his mind, saying, I persecute, 
 because of the Lord of heaven ; and not, I persecute him who 
 dwells in heaven. So he asked : Who art thou. Lord, who in 
 thy heavens art persecuted ? For I persecute Jesus who is 
 among the dead, along with his disciples. 
 
 p. 171 (c. ix. 7) (Ephr.). But the strong illumination they 
 saw not, lest they too be blinded and there be confusion. But 
 he blinded Saul strictly, but pitied them out of his grace. 
 
 1 Lit. ashamed, but there is probably an error in the text. 
 
 ° It is probable that the Armenian translator has confused and perhaps 
 amplified the passage. The printed Yulgate has 'the angel of the Lord,' but 
 a 12th century Codex of Paris, written by a certain Nerses, has ' the Spirit of 
 the Lord.' 
 
44 EPHREM 
 
 p. 171 (c. ix. 8) (Ephr.). For that reason he then raised him 
 up into the third heaven in an inscrutable way and taught him 
 atr6ppr)Ta in supernatural wise... but instantly Saul rose afresh 
 from the ground and [with open eyes, saw no one... by his hand 
 they took him and brought him to Damascus] whither he set 
 out to go so proudly... 
 
 p. 172 (c. ix. 10) (Ephr.). The Lord was revealed in a vision 
 by night to Ananias, that without fear he might come and 
 baptise the persecutor. It (?) again was revealed to Saul, 
 that without hesitation he might awake in presence of his 
 physician. 
 
 p. 184 (ix. 27) (Ephr.). But as he was persecuted by the 
 Jews who were there, and as he was not trusted by the 
 disciples who were there, for they did not, he says, believe that 
 he was a disciple ; then Barnabas presented him to all his 
 companions who were in Jerusalem, took him by the hand and 
 led him to the Apostles. 
 
 p. 195 (c. X. 11, 12) as in the Armenian Vulgate. 
 
 p. 201 (c. X. 34 from elirev to 35) as in Armenian Vulgate, 
 then the comment " that also among the heathen who to us 
 seemed despicable, if there be found one who worships him 
 truly, he is acceptable before him." 
 
 p. 205 (c. X.) (Ephr.). While then Peter having come in, 
 recounted the preaching of our Lord, whence and where he 
 began and where he finished by the Cross, and about his 
 resurrection and about the 40 days that he remained and 
 afterwards ascended, and that all the prophets witness to him, 
 and that every one is forgiven who believes and is baptised in 
 his name ; so on the spot the Holy Spirit came by means of 
 tongues and settled on all the hearers of the word, and they 
 began to speak with tongues, as the course of the history shews. 
 
 p. 230 (c. xii. 19) (Ephr.). [But Herod, when he sought him 
 and found him not, having asked the guards ordered them to be 
 slain.] 
 
 p. 256 (c. xiv. 20) as in the Arm. Vulgate. Then the 
 comment "when the day declined and it became dark, the 
 disciples brought him into the city*. 
 
 ' Cf. Fleury "cum rccessisset populus vespere," and the Sahidic version. 
 
ON THE ACTS. 45 
 
 p. 257 (c. xiv. 23) as in Arm. Vulgate, and the following 
 comment : 
 
 Mark the power of the Gospel. For in those very cities 
 whence they drove them out along with their gospelling which 
 they preached, lo ! they made elders and deacons fearlessly. 
 
 p. 262 (c. XV. 6) and since there was a great dispute between 
 the synagogue and the heathen' and with the Apostles and 
 their friends, the Apostles came and gathered together and the 
 priests along with the multitude to see what issue would come 
 forth about this subject (Xoyov). [And after much discussion 
 Peter stood up and said to them] : for Paul stood forward 
 in Jerusalem before Simeon and his companions against the 
 law, as also he spoke in Antioch before them against the keep- 
 ing of the law. But this Simeon, who was silent in Antioch, 
 when Paul came forward and spoke against the law in Jeru- 
 salem, there dwelt in him the Holy Spirit^, and he began to 
 speak against the upholders of the law thus :... 
 
 p. 277 (c. XV. 29 ?). For as you shall keep faithfully all this 
 without circumcision and observation of the law, ye shall receive 
 the Holy Spirit to speak all tongues'; even as your companions 
 received, the party of Cornelius, who were chosen before you. 
 
 p. 289 (c. xvi. 9). So then that they may hasten to come to 
 Macedonia, where things were ready for them, there appeared 
 to Paul as it were' a man of Macedonia, for he came and prayed 
 and besought him to come and help in Macedonia (after which 
 c. xvi. 10 — 12 as in Arm. Vulg. except Philippopolis for 
 
 Philippi'). 
 
 p. 294 (c. xvi. 19). And instead of the price of healing which 
 
 1 It looks as if this were meant for ' the synagogue of the Gentiles,' in which 
 case we have a suggestion of the double deputation from Antioch to Jerusalem 
 which appears in Codex Bezae. 
 
 " We have here something like the text of Codex Bezae 
 
 ANecTHceN EN THsTi TTeTpoc. 
 
 ' Does this imply the equivalent of <pep6fiei'oi hi T(f crflcfi TveifMTi in the 
 text? 
 
 * The liirel of Codex Bezae. 
 
 ^ So some MSS. of the Arm. Vulg. 
 
46 EPHREM 
 
 the masters of the girl owed to Paul, they stirred up the city 
 against him and accused him of the laws of the Jews which he 
 did not preach to them. 
 
 p. 296 (c. xvi. 22, 23) (Chrys. Ephr.). From koX oi aTparri'yol 
 to TTipeiv avTov<; ace. to Arm. Vulg. (but add avrov^ after 
 pa^Si^eiv). Then the commentary thus : The heads of the city 
 in rending their garments wished to quiet the tumult of the 
 crowd. They tried to prevent it. Because they saw the crowd 
 set on in fury they wished to quiet their rage by the scourging. 
 And to please the will of the crowd which was collected they 
 pinioned the party of Paul and had them cast into the prison, 
 and gave order to keep them carefully, wishing later on to hear 
 about their cause. 
 
 p. 299 (c. xvi. 27) (Ephr.). Then there was a shock in the 
 city and the doors of the prison were opened, and the bonds of 
 the prisoners fell off them. But that there might be no sorrow 
 to the gaoler who was about to believe, none of them fled. 
 For because of this the gaoler deserved the baptism of the font 
 along with his household as he says [then c. xvi. 27 — 31 as in 
 Arm. Vulg.]... 
 
 Then follow vv. 31—35 ace. to Arm. Vulg. On v. 35 the 
 comm. is as follows : 
 
 Perhaps the heads of the army knew all the great wonders' 
 which had occurred, and so they did not venture of themselves 
 to release them, but sent to the gaoler to dismiss them as it 
 were by stealth. 
 
 p. 301 (Ephr.). The Astaritai (= a-TpaTrjyoi) were afraid 
 and full of fear, they the mighty of the city, of the earthquake ; 
 and knew truly that this earthquake happened'' on account 
 of them, but they did not undertake to avow it ( ? = ofiokoyeiv 
 iv avrm). They sent secretly to bring them out. 
 
 Then vv. 35 — 37 ace. to Arm. Vulgate. 
 
 p. 302 (c. xvi. 39) (Ephr.). So then that this favour might 
 
 ' It seems to be Chrysostom (ed. Savile iv. 811), but cf. Cod. D, 
 
 K&l ANAMNHCGeNTeC 
 
 TON CeiCMON TON rGfONOTA e(|)0BH9HC«iN. 
 
ON THE ACTS. 47 
 
 be unto them, they came' and besought of them, (saying), We 
 knew not that ye were just": even as the earthquake indeed 
 presaged about you. So then we ask of you this favour, 
 depart from this city, lest the same men gather together after 
 the earthquake against you, (the same) who before the earth- 
 quake were gathered together °. 
 
 p. 310 (c. xvii. 14) (Ephr.). So he came as far as the sea 
 shore, receding*. But the Holy Spirit prevented him from 
 preaching, lest they should slay him'. [And those who con- 
 ducted Paul, led him as far as Athens, and having received] 
 from Paul" [a command to Silas and Timothy that they should 
 at once come to him] in Athens. [And they went] to him 
 when they received the command'. 
 
 1 Cod. D. 
 
 2 Cod. D. 
 
 8 Cod. D. 
 
 * Cod. D. 
 
 » Cod. D. 
 
 « Cod. D. 
 
 nApArsNOMeNoi. 
 
 HrNOHC&MeN T& KA9 '^Md.C 
 OTI eCTM ANApeC AlKdlOI. 
 
 Ttd,peKd,Kecd.n aytoyc AeroNTec 
 eK THC noAecoc t&ythc eleA9<VTe 
 
 MHTTOTe TT&AlN CYNCTpiS><J)COCIN HMGIN 
 
 eniKpAzoNTec kaO yMO^N. 
 
 (MTeAGeiN em thn GaAaccan 
 abire ad mare uersus. 
 
 n&pHAGeN Ae thn eecc&AiAN 
 gkcoAyOh r'S'P eic aytoyc 
 
 KHpYiAl TON AorON. 
 
 AaBontec Ae eNToAHN nAp& ttiSvyAoy- 
 ' The words 'and they went to him' are due to the arrangement of the text 
 in the Bezan Greek, which runs 
 
 npoc TON ceiA&N km TiMoOeoN 
 oncoc eN taxsi eAGcociN 
 npoc AYTON elHecAN 
 =ad eum proficiscebantur. 
 
 The last line has been detached from the previous ones by the reader or 
 translator, and made into a separate sentence. If this is the correct explana- 
 tion (and it is almost certain in view of the absence of the verb i^-ijfcrav in the 
 rendering of the previous sentence), then we have again an instance of the early 
 currency of the Bezan line-division, 
 
48 EPHREM 
 
 p. 329 (c. xviii. 17) (Ephr.). "The believing Greeks beat 
 Sosthenes the head of the Synagogue"... and that the governor 
 when he saw it might not require vengeance of the in- 
 dignity offered to him, i.e. to Sosthenes, he became as though 
 he saw not, that their blows might be redoubled yet more'. 
 
 p. 331 (c. xviii. 19, 20) (Ephr.). So Paul came and arrived 
 at Ephesus and with him Aquila's party and he spoke there in 
 the Synagogue. And they asked him to remain with them; 
 but he did not choose to stay longer with them ; because 
 whither he was conceraed (to go) thither he had to go. How- 
 ever he did not simply leave them, but with a promise again 
 that they should expect his coming by the will of God. This 
 he says, that they may pray God for his coming, that He may 
 will it. 
 
 p. 334 (c. xix. 1) (Ephr.). Paul wished of his own will to go 
 to Jerusalem ; but the Spirit sent him back to Asia'', as he 
 relates; [it came to pass], he says [while Apollo was in Corinth, 
 Paul went round the upper regions and came down to Ephesus 
 and found certain of the disciples and said to them : If ye 
 received the Holy Spirit in believing. But they say, But not 
 even if the Holy Spirit is have we heard etc.] as far as v. 7, 
 inclusive ace. to Arm. Vulg., only reading k. Irjaov XpiaTov v. 5. 
 
 p. 340 (c. xix. 15) (Ephr.). You, he says, mutilated and 
 broken-backed by the devils, who are you who bid the devils 
 depart? And the devil straitened ( = o-ui^eo-reWe) them right 
 and left and drove them forth from the house. 
 
 p. 352 (c. xix. 38, 39) (Ephr.). This Demetrius', vile and 
 
 ' This involves the reading of D tunc gallio fingeiat earn nan uidere. Cf. the 
 Flenry text : et gallio simulabat se non uidere. 
 » Cod. D. 
 
 eeAoNTOc Ae toy nAY^oy 
 
 KATA THN lAlAN BoyAhN 
 
 rropeYece&i eic lepocoAYMA 
 
 eincN AYTw to ttna YrrocTpe<})eiN eic thn acian. 
 s Cod. D. 
 
 61 MEN OYN AhMHTPIOC OYTOC 
 01 K<M CYN AYTCO TeXNEITe. 
 
ON THE ACTS. 49 
 
 shameless, he says, and the children (TratSe?) of his craft, if 
 they have any suit with one another', let them stand forward 
 and make it clear to the hegemon. And [if there be] any other' 
 [enquiry, let] it be [pronounced on in the lawful assembly]. 
 
 p. 354 (c. XX. 3) (Ephr.). According to the Armenian 
 Vulgate, and then as follows : For that the Jews plotted 
 against him, he wished to go into Syria, but the Spirit turned 
 him back into Macedonia^. 
 
 p. 356 (c. XX. 7) (Ephr.). For as Paul was speaking from 
 dawn until midnight, the youth fell asleep and fell from three 
 stories, because he was sitting there; and gave up the ghost. 
 But Paul went down, fell on him and restored him to life, as 
 he relates. After which vv. 8 — 11 inclusive according to the 
 Armenian Vulgate. 
 
 p. 363 (c. XX. 24) (Ephr.). When he began to tell truly what 
 he was to undergo in Jerusalem from priests and scribes, but he 
 for the sake of the saints who were in Jerusalem was going to 
 comfort them ; and again, that he taught, if ill-treatment does 
 not appal him, then without scruple or fear, without shrinking, 
 he was hastening forward to meet difficulties. Nevertheless he 
 adds this saying: non aestimata mihi anima mea pretiosior quam 
 labores Evangelii vitae et quam ministerium Evangelii verbi 
 quod a domino nostro recepi, id est, testimonium dabo Judaeis 
 et Graecis^ 
 
 p. 410 (c. xxiv.) (Ephr.). When the Rhetor spoke about the 
 peace of their people, and about the disturbance which in all 
 places Paul excited against them, then an order came to Paul 
 to make a defence of himself (Then c. xxiv. 10 as far as eTrt- 
 
 1 Cod. D. 
 ■-= Cod. D. 
 
 exoyciN npoc AyToyc tina AopoN. 
 ei Ae rrepi erepcoN enizHreire. 
 
 » Cf. D. 
 
 K&i reNHGeic iS>YTW eTTiBoyAHC ytto twn ioyAaiwn 
 HeeAHceN anaxGhnai eic cypi«.n 
 einsN Ae to tTn*. aytco YrrocTpectjeiN 
 
 AlA THC MAKeAONIAC. 
 
 '' Observe the agreement in the concluding words with Cod. Bezae : 
 AiAMApTYP«iceAi ioyAaioic km eAAHCiN. 
 HA. 4 
 
50 EPHREM 
 
 ardfievo'! ace. to Arm. Vulg.) Therefore Paul was do stranger 
 or new comer, if he had known the judge for many years. 
 
 p. 410 (c. xxiv. 11) (Ephr.). But he stood forward and said : 
 They have accounted me a raving maniac and a disturher of 
 the synagogue. Be cognisant, hegemon, that in this city I 
 am but a few days and not any considerable number. And in 
 the temple, as I was worshipping, they came and found me, I 
 will not say a crowd mustered apart ; I was teaching. So then 
 if in their synagogue outside the city or here in the city they 
 could not catch me and find me teaching a crowd mustered 
 together, how where all these events were not found, do they 
 come and accuse me as an impostor' ? 
 
 p. 411 (c. xxiv. 14) (Ephr.). However though I were even a 
 Christian, as they say, yet I also worship the God of our fathers, 
 the family of Abraham who without the law worshipped God. 
 So that I believe in the law and in the prophets, whatsoever 
 is written. 
 
 p. 439 (c. xxvii. 23) (Ephr.). Paul told them about the angel 
 who appeared to him and said to him : Before the Caesar thou 
 art to stand and your ship is broken, and not one man of the 
 270 men in it shall be lost. 
 
 After which vv. 27 — 32 ace. to Arm. Vulg. 
 
 p. 454 (c. xxviii. 30) (Ephr.). And he was a space of two 
 years in all at his own expense, and received all who came in 
 unto him. So when (a)?) he discoursed to Jews from dawn till 
 night about Christ from the law and prophets and reiterated 
 about the unfaithful who received not the words of Isaiah; 
 Luke turned and remembered his actions and the labours 
 of his hands, which he gave as hire of his house for one 
 biennium''. And that he ceased not to discourse about Christ 
 
 ' The Armenian literally. 
 
 2 This apparently inexplicable sentence means, I suspect, that there was a 
 gloss in the text concerning the cloak and books, which Paul left in Troas (2 Tim. 
 iv. 13), and used them to pay for the rent of the Eoman lodging. The word 
 'actions' stands for the Syriac rCllixio^, and this is a, misreading of the 
 tranaUterated Greek ^lUvv (The Peshito, however, makes it a book-case 
 r^ia^ Au3 ; as of coui-se, it might be ; of. Birt, Das antike Buchwesen, 
 p. 65.) The Philoxenian transliterates ^< <\\rn«\ 
 
ON THE ACTS. 51 
 
 to Jews and heathen, who went out and came in to him. And 
 he was saying that Jesus Christ is the son of God'. For for his 
 sake we labour and attain to crowns by means of Christ. 
 
 That Ephrem really imagined the house was to be paid for by the proceeds 
 of the cloak and books, may be seen from the following extract from the 
 prologue to 2 Tim. in Ephrem's Comm. on the Pauline Epp. 
 
 " Penulam (phighon, evidently from a Syriao transliteration) antem et libros 
 iussit afferre, aut ut venditis illis, penderet pro domo conducta : aut ut haereditare 
 faceret, oui iustum esset." 
 
 This reference in Ephrem on the Pauline epistles renders almost certain our 
 explanation of the obscure passage in Ephrem on the Acts. 
 
 ' Cf. the Philoxenian text, and the Latin codd. tol. and demidov. 
 
 4—2 
 
CORSSEN AND BLASS ON THE WESTERN TEXT 
 OF THE ACTS. 
 
 Let us now try, before proceeding to examine Mr Chase's 
 theory of the Syriac influence on the Western text, to get 
 some idea of the results which have been arrived at, and the 
 prospects of further conclusions which are being held out to us, 
 by two distinguished German scholars, one of whom approaches 
 the subject from the side of the Latin versions, and the other 
 from that of the recensions of the Greek. We will begin with 
 Corssen's Programm, entitled ' The Cyprianic text of the Acts 
 of the Apostles". As there is probably no living scholar who 
 is more familiar with the Old Latin texts of the Bible, nor one 
 who knows better how to present bis reasonings and results, we 
 shall be sure to get some daylight on the Western question 
 from this tract, although it is only a preliminary notice of 
 further enquiries and is contained in less than 30 pages. 
 Corssen begins by drawing attention to the importance of the 
 Fleury palimpsest from which Sabatier published in 1743 a 
 couple of fragments of the Acts'" and which has, after various 
 attempts by other transcribers, been lately published in what is 
 probably a final textual form by Samuel Berger'. The value of 
 this version (of which one can only deplore that more leaves 
 have not been preserved) lies in the fact of its being an Old 
 Latin rendering, presumably of a Greek text which must have 
 been in singular agreement with the text of Codex Bezae. It 
 
 ' Peter Corssen: Der Cyprianische Text der Acta Apostolomm. Berlin, 
 Weidmannsche Buchhandlung, 1892. 
 
 ^ Quoted as reg in Tisohendorf s Apparatus. 
 
 " Berger : Le Palimpseste de Fleury. Paris, 1889. 
 
CORSSEN AND BLASS ON THE WESTERN TEXT OF THE ACTS. 53 
 
 becomes, therefore, a textual authority of the highest im- 
 portance, not only on account of the support which it gives to 
 the Bezan text, and the help which it occasionally furnishes 
 where that text is in lacuna, but also because it is suspected of 
 being often more archaic than Codex Bezae itself 
 
 Further, as Oorssen points out, the Fleury Latin is in the 
 Acts of the Apostles in close agreement with the quotations of 
 Cyprian, and other Latin fathers whose text is related to that 
 of Cyprian, so that it becomes possible to restore by skilful 
 editing a large part of what Corssen calls the Cyprianic text of 
 the Acts. We thus obtain a Latin text of the Acts, not merely 
 of the sixth century, to which period the Fleury MS. may be 
 referred, but at least, according to Corssen, of the middle of the 
 third. The importance of this is obvious, and we shall probably 
 be able to shew that Corssen's estimate of the age of the 
 restored text is too modest. It must go back farther than 
 Cyprian. 
 
 The argument by which the antiquity of the text is de- 
 monstrated depends upon a comparison of the readings of the 
 Fleury text with (i) the quotations of Cyprian, (ii) with similar 
 quotations in two works of Augustine entitled De Actis cum 
 Felice Manichaeo and Contra epistulam Manichaei, (iii) with the 
 quotations made in a work of the fifth century, wrongly at- 
 tributed to Prosper, and entitled De promissionibus et prae- 
 dicationihus Dei. From a comparison of these various texts, 
 Corssen shews their derivation from a common Latin primitive, 
 which he calls the Cyprianic text. And this common primitive 
 was a text which had an internal unity and sequence which we 
 look for in vain in the text of Codex Bezae, however much we 
 may be persuaded that the Cyprianic text agreed in the main 
 with the peculiarities of the Bezan Greek. That is to say, the 
 restored Cyprianic text is a Western witness of greater worth 
 than even the Greek of Codex Bezae. 
 
 Corssen then proceeds to suggest that the text of Codex 
 Bezae is composite, and can be resolved into an original Western 
 text plus certain contaminations and insertions which are due 
 to the influence of the Common Greek text. In order to make 
 this clearer we will reproduce Corssen's first and leading illus- 
 
54 CORSSEN AND BLASS ON THE 
 
 tration. The opening verses of the Acts in the two tracts of 
 Augustine against the Manichees agree in the rendering 
 
 in die quo apostolos elegit per spiritum 
 sanctum et praecepit praedicare evangelium. 
 
 This, according to Corssen, was the primitive rendering; its 
 influence may be seen in an extract from Vigilius Contra 
 Varimadum 
 
 in die qua apostolos elegit per spiritum 
 
 sanctum praedicare evangelium, 
 
 and, with some re-action from the common text in Augustine's 
 De Unitate ecclesiae, 
 
 usque in diem, quo apostolos elegit per spiritum 
 sanctum mandans eis praedicare evangelium. 
 
 But if this be the primitive form, we arrive at the important 
 conclusion that it cannot have been made from the text of D as 
 it now stands, for while the restored Cyprianic text is the 
 equivalent of a Greek text 
 
 d')(pi ^9 j^fj,epa<! 
 
 i^eXe^aro rov<; diroaToKov; 
 
 Bid irvev/MaTO'} drylov 
 
 Kal eKeXevce 
 
 KTjpvaa-eiv to eiiayyeKiov 
 the text of D is 
 
 d'X^pi ^? rifiepa<i 
 
 dveXijfi<j)6rj ivTeiXdfievo'i toI<; diroaToKoi,'; 
 
 Bid 7rvev/j,aT0<; dyiov 01)9 e^eXi^aro Kal eKeXevae 
 
 K7)pva(Teiv TO evajyeXiov. 
 
 This latter text is, according to Corssen, due to the influence 
 upon the equivalent of the Cyprianic text of the common 
 Greek text 
 
 d-xpi ^9 rjijjipa^ 
 
 ivTeiXdfievo^ T0t9 a7ro(7To\oi9 
 
 Bid irvevfiaTOi dyiov 
 
 01)9 e^eXe^aro 
 
 dveXrifiipdrj. 
 
WESTERN TEXT OF THE ACTS. 55 
 
 We are therefore to regard the text of Codex Bezae as 
 made up out of (i) a primitive Western text (call it the 
 /3-text) corrupted by mixture with the common text (which we 
 will call the a-text). According to this theory the words icaX 
 iKeXevae Kripva-aetv to evajyeXiov are not a mere gloss on the 
 common text from some outside source, as they certainly seem 
 to be at the first examination, but a part of an early redaction 
 of the text of the Acts which does not agree with the common 
 text'. 
 
 The first thing that we have to reflect upon is the way in 
 which Corssen restores the Cyprianic text in this passage. We 
 will add something to the demonstration of the antiquity of 
 the words which have been apparently appended to the 
 primitive text by Codex Bezae, and we will do something to 
 justify the apparent absurdity of the text restored by Corssen, 
 which makes the choice of the Apostles take place on Ascen- 
 sion day. In the first place, there is evidence earlier than 
 Cyprian of the currency of the words 
 
 Koi, i/ciXevae Kr)pv(T(reiv to evwyyiXiov. 
 
 In the 21st chapter of his Apology, Tertulliaa speaks of the 
 Ascension as follows : 
 
 ad quadraginta dies egit docens eos quae docerent. Dehinc 
 ordinatis eis ad officium praedicandi per orbem, circumfusa 
 nube in caelum est receptus. 
 
 We have here the trace of the commission to preach in an 
 account of the Ascension and notice that it is like the added 
 words in Codex Bezae, in that, although the account is in the 
 main that of the Acts, it is connected with the terms of Mark 
 xvi. 15 ; and there can be little doubt that Tertullian in his 
 text of the Ascension had at least the equivalent of the word 
 Krjpva-a-eiv. But in the second place, the peculiar language 
 
 ordinatis eis ad officium praedicandi 
 
 1 Corssen points out in a foot-note that I have, in my Study of Codex 
 Bezae, missed the explanation of the genesis of the variants, through a failure 
 to detect the reconstruction which D exhibits of an archaic text under the 
 influence of the received text. Possibly this may be so, but we are not yet 
 quite convinced on this point. 
 
56 CORSSKN AND BLASS ON THE 
 
 seems to imply something more than the equivalent of eV- 
 TetXdfj,evo<; or SKeXeva-ev; it is almost as strong as elegit ad 
 praedicandum; but, if this be the case, Tertullian's text would 
 be uncommonly like that which we have found restored by 
 Corssen, for it involves the election of the Apostles on the 
 day of the Ascension. We may say, then, that there is a very 
 high probability that Tertullian had in his text of the Acts 
 the injunction to preach; and a lesser probability that he had 
 words which implied the Apostolic election as occurring on the 
 day of the Ascension. So that something can be said in favour 
 of the antiquity of Corssen's restoration ; and I, for my pai-t, 
 am quite satisfied that the supposed gloss in the Western text 
 is earlier than the time of Tertullian. We will give another 
 little illustration of the way in which the Fleury text and its 
 companions can be put in evidence for the existence of a 
 Latin translation of the story of the Ascension earlier than 
 Tertullian. 
 
 The manner in which Tertullian quotes the verses Acts i. 
 10, 11 implies, at all events in certain touches, something more 
 than his own habitually free renderings of the text : we have 
 as follows : 
 
 De Resurr. c. 22. Quis caelo descendentem talem conspexit, 
 qualem asceudentem apostoli viderant secundum angelorum 
 constitutum ? 
 
 Ibid. c. 51. Idem tamen et substantia et forma qua 
 ascendit, talis etiam descensurus, ut angeli adfirmant. 
 
 Adversus Praxeam c. 30. Hie et venturus est rursus super 
 nubes caeli, talis qualis et ascendit. 
 
 It will be noticed that in these three passages Tertullian 
 harps upon the words talis and qualis. I infer that they were 
 known to him in a Latin translation of the Acts : (the Vulgate 
 and most other fathers have sic veniet quemadmodum, which is 
 a more literal rendering). If we now turn to the Fleury 
 palimpsest, and examine its reading of Apoc. i. 7, we have 
 
 et uidebunt eum omnes tribus terrae talem. 
 
 The added word talem is not a mere blunder of the Fleury 
 text : the Fleury text is in the Apocalypse the same as that of 
 
WESTERN TEXT OF THE ACTS. 57 
 
 Primasius and Cyprian, and is used by Haussleiter to restore 
 the so-called Oyprianic Apocalypse, just as Corssen wishes to 
 use it for the editing of the Cyprianic Acts. When we turn to 
 the text of the Acts used by Primasius, as edited by Haussleiter, 
 we find 
 
 et uidebit eum omnis terra talem. 
 
 It is impossible to resist the conclusion that the primitive 
 Latin Apocalypse in very early days had talem in its text, 
 perhaps as the remains of a primitive gloss, something like 
 Tertullian's talem qualem ascendentem apostoli viderant or 
 talem qualis ascendit. But such a text must be held to be 
 assimilated to that of a passage in the Latin Acts. The com- 
 mentary of Primasius does not help us to restore the missing 
 clause ; it has talem hidden away amongst the explanation of 
 the author, but it does not seem as if more than this were in 
 Primasius' text, for he has made an expansion of his own, cum 
 talem viderint qualem non credebant^. 
 
 But even if it should be maintained that talem is a corrup- 
 tion of amen, which occurs a little later in the text, the error is 
 still due to the influence of the passage in the Acts, and 
 implies the word talem in Acts i. 11. If this reasoning be of 
 any force, we must again say that the restored Cyprianic 
 'text is earlier than Cyprian. It might just as well be called 
 Tertullianic. 
 
 Even where we cannot carry a gloss back with certainty into 
 the Latin of Tertullian, we can shew sometimes that the gloss 
 itself was known to him, without determining the form in which 
 he read it. For example in Apology c. 39 we have 
 
 itaque qui animo animaque miscemur, nihil de rei com- 
 municatione dubitamus; omnia indiscreta apud nos praeter 
 uxores. 
 
 The first sentence is an obvious paraphrase of Kaphia xal 
 ■^vvv /"■'« in Acts iv. 32 and the last is the equivalent of dXX' 
 
 1 The whole passage is "Et uidebit ilium omnis terra. Omnes etiam ter- 
 reni uel omnes generaliter mali, maxime Judaei, qui eum ut minimum putavere 
 neoandum, oum talem uideriut qualem non credebant, sera semetipsos poeni- 
 tentia lamentabuntur, et plangent se adveniente eo. Ita amen." 
 
58 CORSSEN AND BLASS ON THE 
 
 ^v avTol<; nravra Koiva in the same verse. The intermediate 
 sentence is, therefore, a reproduction of the Bezan gloss 
 
 KoiX ovK rjv Bi<iKpiai,<; iv avToi'! ovhefiia, 
 
 but whether it comes from the Greek, or from a Latin rendering 
 
 et non erat dubitatio in eis ulla 
 
 we cannot tell. 
 
 But the gloss must have been known to Tertullian, and 
 have stood in the same place which it occupies in the Bezan 
 text. 
 
 Other instances of the same kind may readily be brought 
 forward ; for example in De Pudicitia, c. 21 
 
 In ilia disceptatione custodiendae necne legis primus 
 omnium Petrus spiritu instinctus et de nationura vocatione 
 praefatus : Et nunc, inquit, cur tentastis etc. 
 
 Here spiritu instinctus is due to the gloss in the Bezan text 
 of Acts XV. 7 
 
 dvicTT'qa-ev iv irvi, Trerpo? 
 
 primus omnium may be due to a similar gloss in Acts ii. 14 
 where the situation is similar, iTr-fjpev tt/scuto? rrjv (fxovT/v aiiTov. 
 Even where the text of Tertullian is not under the influence _ 
 of Bezan glosses, it sometimes gives us the key to the genesis 
 of the glosses. 
 
 For example, in De Baptismate c. 10 
 
 quoniam qui Joannis baptisraum habebant non accepissent 
 spiritum sanctum, quem ne auditu quidem noverant, 
 
 the language of Tertullian suggests to us the following primi- 
 tive and idiomatic rendering of Acts xix. 2 
 
 sed ne auditu quidem spiritum sanctum accepimus. 
 
 Now compare with this the Bezan confusion, 
 
 AAA oyAe TTNA ApoN [AamBanoycT 
 
 TINec] HK0YCAM6N 
 
 = sed neque spm sanctum [accipiunt 
 quidam] audivimus. 
 
WESTERN TEXT OF THE ACTS. 59 
 
 And it is not unreasonable to suppose that the accipiunt 
 quidam of the Bezan text goes back into quidem (iccepimus, 
 and that the attempt to bring the translation and text into 
 closer harmony with one another has led to the Bezan con- 
 flation and confusion. 
 
 Keturning now to Corssen, and his theory that the original 
 form of the Western text has been mixed with the common 
 text, we remark that the theory is an attractive one in so far 
 as it furnishes an explanation of the structure of a number of 
 passages that appear double in D, but it is encumbered by 
 the difficulty that it does not furnish a single explanation of 
 any residual Western variations. The text is thus, from 
 Corssen's view, a double redaction, the reason for whose varia- 
 tion must be sought in the sources. 
 
 The following specimens are given by Corssen to illustrate 
 the double redaction : 
 
 Acts iii. 7, 8. 
 Text. rec. 
 Kal •7nd(ra<; avrov rrj<; Se^id<; %etj009 rjyeipev avrov 
 Trapaxpv/J'Ct Se ea-TepewO-qaav 
 al /3d(reK avrov xal to, crcpvSpd 
 Kal e^aX\6/j,6vo<! effrr] 
 Kal •KepieirdTei 
 
 with which cf. D 
 
 Kal 7nd(ra<{ avrov t^? Se^id'; %etjOo? Tjyeipev 
 
 Kal irapa-xprj/ia iarddTj 
 
 Kal iarepewdija-av avrov 
 
 al iSdaeK Kal ra a-cf)vpd [kol e^aXXo/iei'o? ecrrr]] 
 
 Kal •jrepieirdrei ■xaipofievo'i 
 the bracketed words being assumed to be borrowed from the 
 common text. 
 
 The Fleury text agrees closely with D, as follows : (italics 
 where the text is illegible or doubtful) 
 
 Et adpraehensa manu eiws dextera 
 excitauit eum 
 et continuo stetit 
 
60 CORSSEN AND BLASS ON THE 
 
 con&rvaatiqvs sunt 
 
 gressus eius et laccania 
 
 et ambulabat gaudens et exultans 
 
 where observe that the Fleury text has no trace of the words 
 which Corssen considers an interpolation from the common 
 text. 
 
 Or we may take Acts xviii. 8, where Text. ree. has 
 
 KOi TToXXot TWV K.optv6i(ov 
 aKovovTei iiriaTevov 
 
 and D kol ttoXKoI twv Kopivdimv 
 
 aKovovTe<; [eirLCTTevov 
 Kol] i^aiTTi^ovTo -TTiaTevovTei; reS Oea 
 Sia Tov ovo/MUTO^ Tov KV rjfjkwv i7]v 'xpv 
 
 and the Fleury text 
 
 et quomodo multa plebs corinthiorum 
 audierant uerbum dni 
 unti sunt credentes 
 in nomine ihu XP'' 
 
 where again the text of the palimpsest has not the repetition 
 which characterises D, and may perhaps be regarded as an 
 earlier form of the Western text. 
 
 There can be no doubt that the Bezan text is marked by 
 such doublets as Corssen points out, whatever may be their 
 origin, and that whether they consist of whole sentences or 
 occasional words, one of the first things necessary for the 
 clearing of the Bezan text is the separation of the doublets. 
 
 For example Acts xix. 8 (D) 
 
 elaekOmv Be 6 IlaOXo? et? rrjv crvvaywyTjv 
 iv Svvdfiei /MeyaXr) eTrappTja-id^eTO 
 
 is probably the resultant of two texts ; the (a) text 
 
 elaekdwv he 64? Trjv cruvaycoy^v 
 enapp7]cri,d^eT0 
 
WESTERN TEXT OF THE ACTS. 61 
 
 and the (/3) text 
 
 €iaeX0a)v Se 6 IlavXos et? rrfv crvvayayriv 
 iv Svvdfjbet fieyaXr) iXaXei (?) 
 
 So in Acts iii. 13 the text of D 
 
 Kara irpoaunrov TleiKdrov 
 rov KpeivavTO<i iKeivov 
 diroXveiv avrbv OeKovTO<; 
 
 should be resolved into the a-text, 
 
 Kara irpocrwrrov YieCKaTov 
 KpLvavTO<i eKeivov dirokveiv 
 
 and the y8-text 
 
 Kara irpoawirov YleCkdrov 
 Qekovro<; eKeivov 
 diToKveiv aiiTov 
 
 The difficulty here would be that the Fleury text has ' cum 
 ille iudicaret eum dimittere'; the text avoids the doublet, but 
 sides with the a-text, instead of presenting the desiderated 
 /8-text. 
 
 In the same way Acts xx. 18 (D) 
 
 w<; Be irapeyevovTO 7rpo<i aiirbv 
 
 o/jt.m<! eovTcov avrdov elirev tt/so? avTow 
 
 should be resolved into the a-text 
 
 &)9 Se irapeyevovTO 7rpo<; avrov 
 eltrev avToi<; 
 
 and the /8-text 
 
 Kal '6fia)<; eovrcov avrwv 
 el-irev 7rpo<i avTov<;. 
 
 (Here again the Fleury text is absent ; it is significant that 
 the Gigas MS., which represents the text of Lucifer, is double 
 and reads " cum convenissent ad eum simulque essent.") 
 
 It is not surprising that after suggesting this means of 
 purifying the text of D, of whose base he has justified the 
 antiquity, Corssen should say some significant things about the 
 
62 COKSSEN AND BLASS ON THE 
 
 modern edited texts of the Acts, which he declares to be the 
 reflected images of a fourth-century recension, which itself is 
 arbitrary in character. We build, says he, on the great uncials, 
 as if they were rocks ! Such language as this would, in England, 
 be credited to a disciple of Burgon, and set aside as mere 
 fanaticism. But it would hardly be wise to treat Corssen with 
 that kind of contempt. 
 
 Upon the whole we should say that the value of Corssen's 
 Programm consists chiefly in the demonstration it furnishes 
 of the antiquity of the Western Latin, but that further investi- 
 gation is necessary before we could say decidedly that the 
 /3-text had existed in Greek, unmixed with the a-text. It 
 needs to be shewn that the /3-text is something more than the 
 possible contamination of a text by a version*. 
 
 Precisely at this point of difficulty, Blass comes forward 
 with a learned and acute investigation of the essential double- 
 ness of the early text of the Acts in which he tries to detach 
 the Bezan base from the received text with which it has been 
 encumbered, and to justify the text so detached as in linguistic 
 harmony with the language of the New Testament in general 
 and of S. Luke in particular. 
 
 Blass's theory is that it is significant that the Western 
 variations are most intense in the writings of S. Luke, and that 
 it is a natural assumption to make that two copies of the 
 author's works got into circulation, the rougher and earlier 
 of which (the /3-text) is represented to us approximately in the 
 Codex Bezae, the Fleury-text, and the marginal annotations of 
 the Heraclean Syriac (which are translated from a Greek MS.) 
 together with sundry survivals in other texts and versions. 
 
 The supposition that the Western text makes no material 
 addition to our knowledge is held by Blass to be a mistake ; 
 we cannot explain all the Western expansions of the Acts by 
 carelessness and harmonistic corruption ; but if this be so, the 
 Western tradition acquires at once a certain independence. 
 
 1 Corssen's attempt to restore the text of the vision of Ananias in c. ix. 
 10 — 12, and to reduce it to a mere subjective impression of S. Paul's own 
 inner experience, does not need any serious treatment; for, as Blass says, 
 ' Who performed the baptism? ' 
 
WESTERN TEXT OF THE ACTS. 63 
 
 The scribe and translator give place to the author. Accordingly 
 Blass proceeds to analyse and justify the Western peculiarities, 
 beginning with the tenth chapter of the Acts and working 
 forward through the tract where the glosses are most signifi- 
 cant, and then returning to read the first nine chapters in the 
 light obtained frona the study of the rest of the book, the 
 linguistic analysis of which is often extremely suggestive. 
 
 The originality of some of the Western expansions of the 
 text ought, I think, to be at least provisionally conceded; it 
 is difficult to believe that a mere transcriber invented the 
 'seven steps' of the prison at Jerusalem which S. Peter and 
 the angel descend together ; the detail which is given in the 
 visit of Peter to Cornelius that when they came near to 
 Cesarea, one of the slaves ran forward to announce Peter's 
 approach, and that Cornelius sprang forward to greet him, 
 is as life-like as anything we could wish, and agrees with the 
 statement that Cornelius had sent two slaves along with a 
 devout soldier. 
 
 The story of the sons of Sceva the priest is free, in the 
 Bezan form, from the contradiction which is involved in 
 speaking of seven sons in the beginning of the story and 
 calling them d/jbcjjoTepoi, in the conclusion of it; yet it does 
 not seem as if the Bezan text were a corrupt and late recon- 
 struction reducible on examination to the form of the received 
 text. 
 
 The visit of S. Paul to Jerusalem in company with Mnason 
 is obscure enough in the received text, which makes it one 
 day's journey, and yet implies that S. Paul carried his host 
 with him into a city where he had many friends. The Bezan 
 text knows that it is two days' journey (68 miles, as Blass 
 points out), and makes it intelligible why Paul's entertainer on 
 the road should travel with himS 
 
 The statement that S. Paul lectured every day in Ephesus 
 from 11 a.m. to 4 p.m. (the hours when Tyrannus' school was 
 
 1 Many of these Bezan singiilarities are discussed by Prof. Eamsay in his 
 Church in the Roman Empire, and some of them are approved by him as bearing 
 the stamp of originality. 
 
64 CORSSEN AND BLASS ON THE 
 
 disengaged ?) has every appearance of being a statement based 
 upon direct knowledge ^ 
 
 More remarkable still is the fact which Blass points out 
 that the Western glosses occasionally add to what are called 
 the We-sections of the Acts. 
 
 In Acts xi. 27 the current text tells us that iv rauTat? 
 Se ral<i i^/j,epai<; KaTrjX6ov airb 'lepoaoXv/u.cav Trpo^rJTai et? 
 ' Avrio'^eiav dvaaTo,^ Se eh ef avr&v ovofiari "Aya/So? kts., 
 to which Codex Bezae adds as follows after 'Avrioxeiav, 
 
 fjv he iroXKr] ayaWiaaK 
 (TVvecrTpafi/jievaiP Se rjp,S)v 
 e<fyr] el? e^ avTmv 6vo/j,aTi "Aya/So? 
 
 where the peculiar form of the second line will be noticed ; 
 whoever the writer is, he has at all events thrown himself 
 pretty vigorously into the situation. Nor is this the only 
 case of the kind, for we find Irenaeus (ed. Mass. 201) quoting 
 amongst the proofs that Luke was the inseparable companion 
 of Paul a passage from the Acts as follows : 
 
 " Quoniam autem is Lucas inseparabilis fuit a Paulo et 
 cooperarius eius in Evangelio, ipse facit manifestum, non 
 glorians, sed ab ipsa productus veritate. Separati enim, 
 inquit, a Paulo, et Barnaba et Johanne qui vocabatur Marcus, 
 et cum navigassent Cyprum nos venimus in Troadem." The 
 whole argument of the passage turns on the occurrence of the 
 words nos venimus in the text ; and Irenaeus goes on to quote 
 passage after passage from the We-sections in order to shew 
 that Luke was the constant companion of S. Paul, much in the 
 same way as if he were writing an introduction to the Acts for 
 the Cambridge Bible for Schools. I do not see how we can 
 refuse to recognise the existence of the words 
 
 KaTTivTijcrafiev et? Tp^dSa 
 
 in the text of Acts xvi. 7 in Irenaeus. Nor can the reading be 
 very well dissociated from the previous case, since the tendency 
 
 1 Mr Chase's theory is that it is an assimilation to the darkness at the 
 Crucifixion, with an hour put on at each end ; I suppose, to make the darkness 
 of S. Paul's teaching more visible I Are we to take this as ajeu d'esprit? 
 
WESTERN TEXT OF THE ACTS. 65 
 
 in both is to intensify the personal element in the narrative. 
 A similar personal touch is preserved in Acts xxvii. 35, where, 
 Cod. 137, the obelized Heraclean and the Sahidic add after 
 iaOUiv the words einBiSov<; koI rffuv. 
 
 Where shall we find a reviser or glossator earlier than the 
 days of Irenaeus, whose critical faculties were so highly de- 
 veloped as to make all these alterations and expansions in the 
 text, who had travelled so far as to correct and expand the 
 geography at every point, and had studied the history so care- 
 fully that he was able to illustrate the text with fresh meanings, 
 and to remove inconsistencies which would hardly have been 
 alluded to in an uncritical age ? Can Tatian or the first Syriac 
 translator or reviser be responsible for all this ? We admit 
 that if the glossator be a separate person from the author, he 
 must have had the soul of a harmonist, but he must also have 
 been gifted with some of the trained instincts of a modern 
 critic. Such explanations as that of the double persecution at 
 Iconium (Acts xiv.) or the remark that when Paul took up his 
 lodging at the house of Justus, he had left his old friends 
 Aquila and Priscilla (Acts xviii. 7), are of the nature of a 
 modern commentary. Now I am willing to admit that Tatian 
 was something of a commentator, and in my notes on his 
 Diatessaron have shown that this wonderful work may be 
 described from one point of view as the earliest commentary 
 on the New Testament; but the rehandling of the text of the 
 Acts is so much more extensive than the explanatory touches 
 in the Gospels, that one hesitates to say that the Western text 
 is due entirely to Tatian, and to his translation of the New 
 Testament into the Syriac language. It may be so, but the 
 theory of Blass is much easier, which throws back a part at all 
 events of the textual changes upon the author and his sources. 
 And the theory demands the more consideration inasmuch as 
 it is now practically certain that the so-called Cyprianic Latin 
 text cannot be later than the second century, so that any texts 
 or versions which lie behind this must be not very remote from 
 the actual sources. 
 
 On every account therefore the Blass-Corssen theory needs 
 to be carefully looked into ; even if it be not the true solution 
 HA. 5 
 
66 CORSSEN AND BLASS ON THE 
 
 of the textual bifurcation, we must at least allow this much, 
 that the bifurcation itself is demonstrably so early, that it 
 would be very unreasonable to suppose that none of the 
 Western readings were genuine. It may be justified in select 
 readings even where it cannot be justified as a whole. And 
 this means that there is nothing against which we need to be so 
 much on our guard as the seductive supposition that the cause 
 of certain variants is necessarily the cause of the remainder, or 
 that we can, because we have explained two or three obscure 
 changes in the text, use the Newtonian vera causa over the 
 remainder. The Bezan text and all other Western texts will 
 remain complex until their simplicity has been demonstrated 
 in a sufficiently broad and comprehensive manner. 
 
 The actual test of the correctness of Blass' reconstructions 
 will be best made by a close study of individual passages ; and, 
 where it is possible, by setting up the a-text and /8-text side by 
 side and comparing them with the earliest versions. 
 
 For example in Acts xvi. 10 the current text is 
 
 0)9 Se TO opafia elSev 
 
 ev6ea)<; e^TjTija-a/Mev i^ekOelv ei<; M.aKeSoviav 
 crvv^i^d^ovTe<i '6ti irpouKeKKrfraL ■^fia'i 6 d<; 
 evayyeXiaaarOai avTOVi. 
 
 But Codex Bezae reads 
 
 SteyepOeh ovv ScriyrjcraTO ro opa/ia ■^/uv 
 KoL evoriaafiev '6tl TrpoaKeKktjrai -^fia^ 6 /cs 
 evajyeXvaraerOai rovf ev ttj MaKeBovia. 
 
 Here Blass justifies Sieyepdeh as a Lucan expression by 
 Luke viii. 24, and observes that ivoTjaa/iev, though not Lucau, 
 is common in the language of the N.T. He points out that the 
 Sahidic version supports in part the expansion of D, though it 
 soon falls in line with the received text (cum autem surrexisset, 
 narravit nobis visionem ; statim quaesivimus egredi in Mace- 
 doniam, ostendentes iis quod Dominus vocaverit nos ad an- 
 nuntiandum iis evangelium). Further, at the commencement 
 
WESTERN TEXT OF THE ACTS. 67 
 
 of the next verse, the y8-text gives us a note of time which is 
 wanting in the current text, 
 
 tt} Se iiravpiov a')(9evTe<; a-rro TpwaSo?. 
 
 The question is, whether all this precision and expansion is due 
 to scribe and commentator. We shall in our next lecture be 
 able to throw a little light upon the subject, though not, 
 perhaps, to come to a complete solution of the questions at 
 issue. 
 
 0—2 
 
CHAEACTER OF THE GLOSSES IN THE WESTERN 
 TEXT OF THE ACTS'. 
 
 We have, in a preceding lecture, completely justified 
 Mr Chase in his hypothesis of an Old Syriac text of the 
 Acts behind the venerable Peshito, and we must now turn 
 for a brief survey of his contention that the hypothesis which 
 we have verified for him is an adequate one to explain the 
 peculiarities of the Western text in general, and of the Bezan 
 Acts in particular. 
 
 It will not be possible for us to make a complete exami- 
 nation of a theory which is only partially brought out, and, 
 as we shall show, not very adequately thought out. But we 
 will place ourselves as far as possible in Mr Chase's position, 
 and see whether the glosses look natural and explicable in 
 the light of his hypothesis. When we have done that, we 
 must make a similar attempt to appreciate Dr Blass's view 
 that the Bezan text is a true Lucan text, and the forcible 
 arguments by which he vindicates so many of the expressions 
 in the much abused Codex. Or better, since the investigation 
 is a series of examinations of various passages, we shall try to 
 look at them one by one from every point of view and see 
 whether they suggest Greek, Latin or Syriac individualisms. 
 
 Now with regard to the question of Syriasm in the Codex 
 Bezae, I feel sure that no one who had a good knowledge of 
 Semitic languages would read the MS. through without ad- 
 mitting that there were a number of things which were im- 
 mediately explicable by Syriac influence. Some of these were 
 pointed out in my tract on the Codex Bezae. But at the same 
 time I have no doubt that there would arise a suspicion that 
 other explanations were possible ; and further, there would be 
 
 ' This lecture was delivered in the Divinity School, Cambridge, January 24, 
 1894. 
 
CHARACTER OF GLOSSES IN WESTERN TEXT OF THE ACTS. 69 
 
 a number of textual expansions of which the assumed Semitic 
 scholar would say, "This cannot be credited to any Syriac 
 hand"; and others of which he would say that if they stood 
 for Syriac, it was very bad Syriac. To take an instance under 
 each head, the first that occur to one's mind, no one would be 
 willing to credit the Syriac language with the addition of the 
 word ia-Tiv to the Greek question in Mark v. 9, ti a-oi ovofid 
 ia-Tiv, because a formula of that kind would be in Syriac 
 either 
 
 v^az. ^ or vcox. reli&^re' 
 
 and while we may freely admit that a complete retranslation 
 of the Syriac into Greek or Latin might give the added 
 vocable, the process of glossing an already existing text so as 
 to make up its deficiencies from another text would not give the 
 desired expansion. Eveu if we explain it as an assimilatioa to 
 Luc. viii. 30, the connexion cannot be made through the Syriac. 
 Such a case as this, and numerous parallel cases might be cited 
 for which I refer to my book, must either be a case of primitive 
 variation in the Greek, or it must be a case of Latinisatioo. 
 Blass would perhaps say the former, though I do not quite 
 know whether he is prepared to extend his theory of a primitive 
 dual text beyond the limits of the Acts and the Gospel of Luke. 
 I should, probably, adhere to my view that it is a case of Latin 
 influence. But in any case it can hardly be Syriac, unless we 
 assume that a complete new Greek or Latin version was made 
 from the Syriac. 
 
 In a similar manner, if we draw attention to the very 
 interesting gloss in Acts xiv. 2, 
 
 o Ae KC eAcoKCN taxy giphnhn, 
 
 we should hesitate to say 'this is a Syriac gloss,' because the 
 expression in Syriac 'to give peace' means 'to say farewell.' 
 For this reason the translators of the Old Testament have 
 sometimes felt obliged to change the literal rendering of the 
 corresponding Hebrew expression and substitute another word 
 that would be clear of misunderstanding. For instance, in 
 Numbers xxvi. 6, instead of transferring "the Lord lift up 
 
70 CHARACTER OF THE GLOSSES IN THE 
 
 His face upon thee and give thee peace," the Syriac says 
 * and make thee peace.' In the same way in 1 Chron. xxiL 9 
 the words 'and I will give peace and quiet in his days' are 
 translated 'and peace and tranquillity shall be in his days.' 
 No doubt these changes in the translation were intended to 
 remove possible misunderstanding. At first sight then, this 
 gloss in the Bezan Acts does not look like good Syriac. 
 Probably Mr Chase would say that the occurrence of the gloss 
 in the companion document (Oodex E) in the form " God made 
 peace" (o Se 6<; elprjvjjv iirotija-ev) furnishes the clue to the 
 underlying Syriac: and of course this may be so; I am only 
 pointing out that there are a number of things on the face of 
 the Bezan text which do not look like Syriac at all. Perhaps 
 that is inevitable in such a complicated enquiry, and where we 
 know next to nothing of the changes which the documents 
 have passed through. 
 
 But let us assume for a while that Mr Chase's explanation 
 of the Western variants is the true one, and read the text in 
 the light of it, and examine carefully some of his illustrations 
 of the theory. Since all the glosses are in his view Syriac, we 
 may begin anywhere. 
 
 The first thing that we desire to draw attention to is that a 
 number of the glosses in the Codex Bezae are wrongly inserted, 
 and the non-appreciation of this palaeographical feature of the 
 MS. has misled Mr Chase (as well as some other writers) into a 
 variety of impossible explanations of the meaning and origin of 
 the glosses. We are to prove first that this displacement of 
 the glosses is a fact. Let us begin with the interesting gloss at 
 the close of Acts vi. in which we are told by the Bezan text 
 
 KAi eiAoN TO npoccionoN aytoy 
 C0C6I npocwnoN A^^£'^OY 
 
 eCTCJOTOC 6N MeCCO AYTCON 
 
 where the last line is supposed to be a gloss. 
 
 The Latin of the text and gloss is as follows : 
 
 et uiderunt faciem eius 
 quasi faciem angeli 
 stans in medio eorutn. 
 
WKSTERN TEXT OF THE ACTS. 71 
 
 The gloss appears also in the Fleury palimpsest' in the 
 form 
 
 uidebant uultu eius tamquam 
 uultum angeli di stantis inter illos 
 
 where tjhe variations from the Latin of Codex Bezae must be 
 carefully noted. The Fleury text is generally held to be an 
 independent translation from a Greek text which is in sub- 
 stantial agreement with that of the Codex Bezae; and to 
 furnish evidence that the Bezan Greek is not the solitary 
 phenomenon which it sometimes seems to be. In the present 
 case it follows the Greek of Codex Bezae in translating 
 eo-TWTo?, but it does not render the supposed iv /ietrm avrmv 
 literally, and it adds (if indeed it be an addition) the word dei 
 after angeli. 
 
 Prof. Bernard has recently pointed out that the same gloss 
 on the text is contained in a citation of the passage in an Irish 
 MS., the Leabhar Breac or Speckled Book'. He remarks as 
 follows : 
 
 'In Acts vi. 15 we have uidebant... faciem angeli stantis 
 inter illos which is almost the same as the reading of the 
 Fleury Palimpsest (h) ...angeli dei stantis inter illos. The 
 verbal variations here from Codex Bezae {d)= faciem angeli 
 stans in medio eormn seem to indicate that this gloss, at all 
 events, originated in the Greek and not in the Latin text.' 
 
 We have then two fresh Latin authorities to add to the 
 evidence of Codex Bezae (the only authority for the gloss 
 known to Tischendorf). 
 
 Mr Bernard is naturally staggered at the thought that the 
 apparently impossible and ungrammatical reading of the Codex 
 Bezae can be responsible for the grammatical Greek, and for 
 the variant forms of the Latin tradition. We will justify the 
 Bezan Latin presently. 
 
 Having now registered the variant forms in which the gloss 
 has come down to us, we ask the question, what possible 
 
 1 reg of Tischendorf; h of Hort. 
 
 " Transactions of the Royal Irish Academy. Vol. xxx. Pt. viii. , p. 322. 
 
72 CHARACTER OF THE GLOSSES IN THE 
 
 motive can be assigned to the expansion in question ? If we 
 cannot answer the enquiry 'Who made the gloss?' we ought 
 at all events to be able to assign some motive for the 
 conduct of the anonymous scribe and to say why he made it. 
 To this enquiry two answers are made. Prof. Blass counts 
 the passage as one of those in which the primitive dual re- 
 daction of the Acts can be recognized ; and praises the gloss, 
 which is really a part of St Luke's own text, on account of the 
 vividness which it obviously imparts to the account'. 
 
 The other explanation is that of Mr Chase and is as follows: 
 
 "The last line is an interpolation, due to assimilation to 
 xxiv. 21 where the Greek is iv avroK eo-rto? but the Syriac 
 Vulgate has 
 
 The word ^^^ocnoujjs is used to translate ev fieaw avTwv 
 in Matt, xviii. 2, 20, Lc. xxiv. 36. Note the order of the words 
 in the Bezan gloss and in the Syriac of xxiv. 21. The Greek 
 gloss must therefore be the rendering of an old Syriac gloss." 
 
 The reader will, of course, turn to Acts xxiv. 21 in order to 
 see whether there is any possible reason for the reference made 
 by Mr Chase's Syriac glossator. He will look for ' the angel ' 
 who is glossed and not find him ; nor will he be able to make 
 any connexion between the language of St Paul about himself 
 iu the xxivth chapter, and the language of St Luke about 
 St Stephen in the vith. In fact, there is no connexion between 
 the two passages, and one can only conclude that Mr Chase 
 turned the gloss back into Syriac, and then looked for the 
 scriptural words which in other places most closely agreed with 
 the language of the glossator. The rest of his identification is 
 equally devoid of force. 
 
 For, be it observed, the gloss does not belong in the last 
 sentence of the vith chapter of the Acts at all, but in the 
 next sentence at the beginning of the viith chapter : where the 
 Bezan text is 
 
 ' Stud. u. Krit. 1894, p. 115, " anaohaulioher als in a." 
 
WESTERN TEXT OF THE ACTS. 73 
 
 emeN Ae o Apxiepeyc tco ZrecJjANca 
 ait autem pontifex Stephano. 
 
 We proceed to show that the gloss belongs after the word 
 Apxiepeyc in the Greek, and after pontifex in the Latin. 
 
 The words are added in the sense that ' the high-priest 
 stood in the midst and said to Stephen'; we have the exact 
 parallel in the language of the Gospel in Mark xiv. 60 
 
 Koi dvaaTa<i 6 dp)(^iepev<; eli ixecrov. 
 
 If then there has been assimilation, this must be the passage 
 to which reference has been made. The situations are similar, 
 and the language of the unglossed passage furnishes the 
 requisite link with the passage from which the addition has 
 been made*; Stephen before the high-priest has been equated 
 with Christ before Gaiaphas, and the judicial examination of 
 the disciple has been coloured from that of his Lord. 
 
 So much being clear, it is also clear that (1) the Bezan 
 Latin is more archaic than the Bezan Greek^, for it has 
 preserved the necessary stons where the Bezan Greek has 
 followed the exigencies of grammar and replaced eo-raj? by 
 k<j-Ta>TO'; ; (2) that the expansion made by the Fleury text 
 angeli dei is probably a later addition and not a part of the 
 original Western gloss ; (3) that the Bezan Greek and Latin (eV 
 fiea-m avTwv) are nearer to the primitive form than the inter 
 illos of the Fleury palimpsest and the Leabhar Breac ; (4) that 
 Mr Bernard was wrong in praising the Greek of the gloss as 
 nearer to the source than the Latin ; (5) that Prof Blass was 
 wrong in calling it a part of the primitive text; (6) that 
 Mr Chase was wrong in the source which he assigned to the 
 gloss and in everything he said about it ; (7) that I myself am 
 an idiot for not having seen all this sooner. 
 
 1 The importance of this consideration seems often to have escaped 
 Mr Chase; it is not sufficient to establish assimilation between passages 
 after glossation ; the assimilation must be found in nucleus before glossation ; 
 something must suggest it before it is made. 
 
 2 Chase, p. 5. 'The formation of the Bezan Latin must be independent 
 of and later in time than the formation of the Bezan Greek.' 
 
74 CHAHAOTER OF THE GLOSSES IN THE 
 
 So much being premised we have still to ask the question 
 as to the language in which the gloss was made. 
 
 It appears from the Fleury text as if there had been more 
 of the passage in S. Mark borrowed than now appears, or as if 
 the assimilation had been carried further by some later hand. 
 For we have instead of the blunt text of the edited Acts 
 
 elwev 8e 6 dp'x^iepev'; 
 
 the longer form 
 
 et interrogauit sacerdos Stefanum, 
 
 where notice that both D and E (Cod. Laudianus) are in 
 evidence for rm '%Te(f>dva>, and that the text of Mark is koI 
 dvaaTii^ 6 dpxi'epev<; el<; fieaov eTrr)pwT7]a-ev tov Irjcrovv, Xeymv' 
 
 [d interrogabat ihm dicens], 
 
 and further the Peshito shows signs of the existence of some- 
 thing answering to iirripcoTrjaev, since it reads the opening 
 sentence 
 
 rdlcn^ .s>\ cnAr<lx.O 
 = et interrogavit eum summus sacerdos. 
 
 The genesis of the passage seems, therefore, pretty clear. 
 
 But it may be asked, was not the gloss originally added in 
 Syriac rather than in Greek or Latin ? Possibly it may have 
 been ; but while there are some things which look like it, 
 there are others which are doubtful. The strongest argument 
 would be the replacement of the primitive dvaardi; by iarox; 
 which might easily have been accomplished through the Syriac 
 •pr<n ; on the other hand the Syriac would almost certainly 
 have broken up the participial construction ; moreover we miss 
 the (^Li^jjGor^lA in the Syriac where it must have been per- 
 fectly natural, and we notice also that the text of Mc. xiv. 60 
 has the equivalent of ev ixea-cp (as also the Diatessaron), and 
 not the Bezan eV /A€<r&) avrwv. Probably Mr Chase can clear 
 up the matter, now that we have given him the clue. I may 
 remark that the displacement of glosses (of which this is our 
 first instance) is more natural in Syriac MSS. than in Greek 
 Codices, on account of the glosses being often written along 
 
WESTERN TEXT OF THE ACTS. 75 
 
 the margin, vertically; this custom carries the gloss along so 
 much of the margin that the risk of displacement is much 
 increased. The occurrence of frequent displacement in the 
 glosses is, therefore, a palaeographical feature which is more in 
 harmony with Syriac originals than with Greek texts^ 
 
 Our next passage shall be the difficult gloss in Acts xv. 29 
 
 ACj) CON AlATHpOYNTeC 6AYT0YC 
 
 ey npdJdrre ct)6poMeNoi 
 eN TOO Afico UNI eppcocee 
 
 = a quibus conuersantes uos ipsos 
 bene agitis ferentes 
 in santo spo ualete. 
 
 The gloss is more interesting than the previous one, on 
 account of the antiquity of its attestation. It is found in 
 Irenaeus and in Tertullian ; and found in forms which suggest 
 either a primitive difficulty or an initial roughness of trans- 
 lation. 
 
 Irenaeus has amhulantes in spiritu sancto, the Greek of the 
 gloss being problematic'', and Tertullian has either 
 
 rectante uos spiritu sancto, 
 
 or vectante uos spiritu sancto, 
 
 either of which is conceivably correct, vectante being near to 
 the Greek cjiepo/xevoi, and rectante being a good representation 
 of the idea of guidance which underlies the word. 
 
 The occurrence of three such distinct Latin forms shows, as 
 we have said, that there was either a difficult word somewhere 
 to be translated, which has met with varied treatment at the 
 hand of translators ; or that the first translator into Latin made 
 a bad or obscure rendering, which others have had to emend. 
 
 ^ Note however that the ooourrenoe of 'stans' in the Bezan Latin shows 
 that the gloss was meant to go into the next line in the Latin : which almost 
 implies that it had stood outside on the margin of a [Graeoo-] Latin text. 
 
 * We are perfectly safe in reasoning from the existence of the gloss in the 
 Latin to its existence in the Greek. But whether the Greek was aybixaoi or 
 (jjep6iji,(voi or irfpiTaToSnTes, we cannot, at present, undertake to say. Probably a 
 comparative study of the Latin of L-enaeus with the extant Greek would throw 
 some light upon the question. 
 
76 CHARACTER OF THE GLOSSES IN THE 
 
 What are we then to say of the gloss : is it a part of the 
 primitive dual redaction ; a sentence truly belonging to the 
 first period of genesis of the document ? Or is it an insertion 
 resulting from the tendency of scribe or commentator ? 
 
 I believe the first suggestion vfith a view to explaining the 
 gloss was that which I made in my ' Study of Cod. Bezae,' that 
 it was the attempt of a later hand to refine upon the obvious 
 unspirituality and insufficiency of the Jerusalem Concordat. 
 
 No doubt such a feeling would exist, both amongst the 
 more spiritual people, like the Montanists, and amongst the 
 more anti-Judaic like the Marcionites. The former would 
 naturally desire some reference to the Paraclete, the latter 
 would resent an apostolical communication which so partially 
 removed the yoke of Judaism from the religious world. 
 
 I am sorry, however, that I made this suggestion as to the 
 need for reforming and expanding the Jerusalem decrees 
 because it has misled my friend Mr Chase, who has adopted 
 the reason given by me for the gloss, and endorsed it by a 
 number of arguments which are, I am afraid, all wide of the 
 mark. His language is as follows : 
 
 " The Old Syriac' has ' well-ye-shall-be, be strong in 
 our Lord.' Probably the Bezan ev Trpd^are (true text ev 
 irpd^eTe) represents an Old Syriac reading ' well-be-ye.' 
 But what of the Bezan interpolation ? I believe that the 
 desire to make the apostolic decree more spiritual led to 
 the introduction into the Old Syriac text of a phrase from 
 a Pauline epistle which dealt with thaJudaistic controversy. 
 See Gal. v. 18. ' But if in the Spirit led (^vs^AiSo) are 
 ye, ye are not under the law '; and compare Rom. viii. 14, 
 Jn. xvi. 13, Lc. iv. 1 : ' There led him (cnixis.i) the 
 Spirit into the wilderness ' ; it will be remembered that 
 we saw reason to think that the context of this last passage 
 suggested the gloss in v. 26^. The rendering of the Syriac 
 ' led ' by (jtepofievoi, is quite natural (see the use of the 
 Greek word in Mc. xv. 22, John xxi. 18, Acts xiv. 13), 
 
 ' He means the Feshito. 
 
 ° We shall show, bye and bye, that this is a delusion. 
 
WESTERN TEXT OF THE ACTS. 77 
 
 especially as the Bezan scribe in translating Syriac glosses 
 frequently avoids the most obvious Greek word. The 
 choice of the word was possibly influenced by 2 Pet. i. 21 
 (where the Syriac has another word). 
 
 This suggestion as to the source of the gloss is strongly 
 confirmed by the fact that Irenaeus (iii. 17) preserves 
 another Pauline form of the gloss 'ambulantes in spiritu 
 sancto ' (Gal. v. 16). It would appear that in this passage 
 Irenaeus, like Cod. E in v. 39, vi. 10, preserves an Old 
 Syriac reading different from that implied in God. D. My 
 position that these are Syriac glosses is confirmed by the 
 fact that the Syriac Vulgate preserves yet another ex- 
 pedient for spiritualizing the decree. In place of the 
 simple ' be-strong ' (eppaxrOe), it has the phrase (see above) 
 ' be strong in our Lord.' With this compare Eph. iv. 
 15, -jiiix. ^ocn This Bezan gloss.... together with that 
 
 preserved in Irenaeus, implies an Old Syriac version of 
 the Pauline Epistles." 
 
 I have quoted this passage almost at length on account of 
 the illustration which it furnishes of Mr Chase's methods. Of 
 the whole of this imposing mass of arguments, the only thing 
 that is correct is the statement, which might have been made 
 in half a line, that <j>ep6fj,evoi may conceivably be equated with 
 the Syriac ^^.lioo. 
 
 For the gloss does not belong where Mr Chase imagines 
 and where I first thought it to belong, but is a part of the 
 following sentence, describing the Apostolic Mission to Antioch. 
 The current text of this passage is 
 
 oi fiev ovv dTroXvdevTet KaTrjXdov eh AvTio')(eiav, 
 
 with which we must compare the parallel passage (xiii. 4), 
 
 oi fiev ovv eK'ireix(f>6evTe<: vtto tov dyiov 7rv€Vfj,aTo<; 
 KUTrfKOov el's 'Ze\evK€iav. 
 
 Accordingly, the sentence in Acts xv. 30 should run, 
 'So they were led by the Holy Spirit, and came down 
 to Antioch.' 
 
78 CHARACTER OF THE GLOSSES IN THE 
 
 The gloss, inserted here, makes perfect sense ; it is in fact 
 merely an expansion, or, if you will, merely an explanation of 
 d7ro\v6evTe<;. There is no need to invoke Montanus or Marcion, 
 or any anti-Judaic commentator, nor to make references to 
 Galatians and parallels in Romans, John and Luke. Such 
 references could not in any case have been in the mind of the 
 glossator, unless indeed he were writing a sermon instead of 
 transcribing a text. 
 
 Neither is there any reason to assume three separate forms 
 of the Old Syriac in order to meet the exigences of the theory, 
 one for Beza, and one for Irenaeus, and one for Tertullian'. 
 Three separate Syriac origins, with perhaps a fourth for the 
 Peshito, is rather a large order to explain so early a corrup- 
 tion. Neither is it necessary to explain the translation of the 
 supposed Syriac word by (f>ep6/j,evoi on the theory of translation 
 by unlikely words, and a possible influence of 2 Pet. i. 21, 
 where the Syriac has another word. 
 
 I am not quite sure whether I understand Mr Chase at this 
 point. Does he mean that there was an Old Syriac text of 
 2 Peter ? If so, he ought certainly to make a definite state- 
 ment of his discovery of that lost text. But even so, he will 
 still have to assign a reason why a passage from 2 Pet. should 
 have any influence in the rendering of another passage in the 
 Acts with which it is not in verbal agreement. If the reference 
 to 2 Pet. is a good one, why is it not adequate to explain the 
 gloss as from a Greek original ? and does not the comment of 
 Ephrem (p. 277) "ye shall receive the Holy Spirit to speak all 
 tongues," imply the equivalent of a primitive ^epofievot which 
 has been interpreted as in 2 Pet. i. 21 ? It will be seen that 
 we entirely dissent from Mr Chase's methods, but that is not 
 the same thing as proving his theory invalid at every point, nor 
 is it a reason for discarding it at all. Let us ask the question 
 whether the gloss, as replaced in the sequence of Acts xv. 30, 
 can be referred to its original language. 
 
 When we turn to the text of the Peshito, in the two 
 passages which we have ventured to compare, we find the 
 
 1 See the note on p. 95 of Chase. "It is quite possible that this (Tertullian'a 
 form) is to be traced to a Syriac gloss derived directly from Lc. iv. 1." 
 
WESTERN TEXT OF THE ACTS. 79 
 
 desiderated word Ta.ioca in neither, nor any trace of the 
 gloss in XV. 30. But we do find, and this is worthy of note, 
 that the two passages are in harmony, as far as the principal 
 verb is concerned, in the Peshito : compare Acts xiii. 4 
 
 and Acts xv. 30 
 
 But whether any conclusion can be drawn from this we are 
 not able to say. For the very same approximation of the 
 accounts appears in Cod. Bezae which reads in xiii. 4 
 
 ■ ipsi vero disraissi ab spo sancto 
 
 and in xv. 30 
 
 illi quidem dismissi. 
 
 A translation from a richer language into a poorer vocabulary 
 often results in an approximation of similar accounts : and as 
 we have said, there does not seem any trace of the word 
 ^is.l^sa which Mr Chase suggests: but perhaps this sugges- 
 tion also is a mistake. Possibly Mr Chase may be able to 
 find the traces of the missing or requisite word. We have no 
 prejudice against his theory : for the sooner the problem is 
 solved the better for all persons concerned. 
 
 Let us turn in the next place to the passage Acts v. 38, 
 and see whether the supposition of a Syriac original will throw 
 light upon the glosses, and in particular upon the curious gloss 
 which appears in Cod. D in the form 
 
 MH MIANANTeC TAG )(eip*,C 
 
 = non coinquinatas manus 
 and in Cod. E 
 
 MH moAynontgc tag x^'P''^'^ ymwn 
 
 = non coinquinantes manus vestras. 
 
 The passage is one to which I drew special attention (1) on 
 
80 CHARACTER OF THE GLOSSES IN THE 
 
 account of the appearance of an accusative absolute in the 
 Latin, which seemed to me to be original, (2) on account of 
 the coincidence between D and E in the verb used in the 
 Latin, although the Greek from which they were supposed to 
 be taken was different ; I inferred that the gloss passed into 
 the Greek from the Latin. 
 
 The gloss certainly adds to the force of the narrative ; and 
 so does the repetition of aTroo-T^re nV avrcov ; and although 
 the two Greek forms cannot, of course, be original, that does 
 not mean that one of them may not be the original: and on 
 the other hand, neither of them need be original. If they do 
 not come from a primitive Greek or Latin, they may be, as 
 Mr Chase suggests, translations of a Syriac phrase. The coin- 
 cidences between the two Latin forms would in that case be 
 accidental. On the supposition of a Syriac original, the first 
 thing that we should do is to look for any peculiarities or 
 irregularities conserved in the rendering of the Peshito. The 
 two verses (v. 38, 39) are as follows : 
 
 And now I say to you : Remove from these men, and let 
 them alone ; for if this device and this work be from men, they 
 will dissolve and perish ; but if it be from God, it does not 
 reach to your hands to annul it. 
 
 The reader will observe the curious translation of ov 
 Swrjcrea-de = ^_^aA^^*r<isi rf i^ra pcl^ 
 
 If the gloss comes from the Syriac, I suspect that this 
 passage must be the cause of it ; we have only to exchange two 
 letters, and read, with a slight change, 
 
 to give the sentence " no pollution in your hands," which may 
 perhaps be recognized as an adequate original for the glosses in 
 the Western texts'. 
 
 But if this explanation be correct, and it is the only one 
 which I can think of on the Syriac hypothesis, for Mr Chase's 
 
 ' I do not spend time to prove that r<lSo!^ may rightly be rendered /uabu 
 or ^o\6v(o or coinquino. 
 
WESTERN TEXT OF THE ACTS. 81 
 
 explanation seems to me quite impossible, then the gloss has 
 again got into a wrong place in the text', as we saw in the two 
 previous cases. If then the gloss be Syriac, we have drawn 
 attention to three cases of displacement in the inserted glosses. 
 We thus arrive at the conclusion that ' the glosses in the Codex 
 Bezae show signs of having been inserted from the margin, 
 either from the margin of the ancestry of the Codex or from the 
 margin of some other Codex, the language of which Codex is not 
 yet quite clear'; and further 'the displacement which is observ- 
 able in certain of the glosses, is a strong though not a conclusive 
 argument, against the theory that those glosses formed a part 
 of a primitive redaction of the text.' Prof. Blass's theory will be 
 much weakened if we can show that any of the glosses on which 
 he relies are out of their true place. It is true that the theory 
 of contamination of one of the forms of the primitive redaction 
 by the other may help him to evade the difficulty ; for if there 
 were two primitive forms of the text, the second form (which 
 Blass calls /3, and which represents what Hort calls the Western 
 text) does not exist unmixed, but, as we have it in D, in the 
 form of an a-text contaminated by a /3-text, or conversely ; and 
 this leaves open the possibility that sentences of the /3-text may 
 be wrongly affixed, as it can be shown that single words are 
 often wrongly restored from some companion document. But it 
 is doubtful whether this hypothesis would bear the strain of 
 the cases of displacement which we are able to bring forward. 
 
 Keturning to the gloss which we were just now discussing, 
 we do not think that the explanation given of the displacement 
 is decisive in the same sense as in the two previous cases, 
 where there is really no room for doubt. But it would seem to 
 be much more natural to explain the added words in this way, 
 than to translate the sentence back into Syriac as Mr Chase 
 does, and then search for obscure parallels to the language in 
 the Old Testaments 
 
 ' Or rather it has moved away from its right place. 
 
 2 Mr Chase's ai-gument is "In Isaiah lix. 3 there occur the words 'your 
 hands are defiled with blood.'. ..From this verse came, I believe, an Old Syriac 
 gloss ( = and do not defile your hands)." Credo, quia impossibile est. There is 
 no motive to take either a transcriber or glossator into Isaiah. 
 
 HA, 6 
 
82 CHARACTER OF THE GLOSSES IN THE 
 
 Having now shown that certain of the glosses do not lend 
 themselves to Mr Chase's elaborate explanations, nor to Prof. 
 Blass's justifications, on account of their not standing where the 
 explanations assume them to stand, we pass on to make a few 
 remarks on the question of the assimilations of the text, as 
 when one passage of the Acts is conformed to or expanded from 
 another passage of the Acts or as when the text of the Acts 
 is similarly expanded from the Gospels, the Diatessaron, the 
 Epistles, the Prophets, the writings of Papias (in Syriac), and 
 the like, according to Mr Chase's explanations. 
 
 The question between Blass and the rest of the critics really 
 turns upon the extent to which assimilation can be detected in 
 the supposed glosses : if they are merely textual parallelisms 
 developed in a version, and projected back upon the original 
 text, the position defended by Blass becomes a hopeless one. 
 But here, also, much will depend upon the extent to which the 
 corrupting influence can be proved to work ; it is quite possible 
 that there may be an occasional assimilation in the text with- 
 out our being obliged to condemn as non-original those many 
 vivid historical and geographical details with which the Codex 
 Bezae abounds. No assimilation will explain the tarriance at 
 Trogyllium, the fright of the Philippian magistrates over the 
 earthquake, the visit of S. Paul to Myra (to which Kamsay has 
 drawn especial attention on account of its influence on the 
 story of Paul and Thekla), or the statement that Paul spent 
 the night somewhere on the way from Cesarea to Jerusalem. 
 The geographical details must be original or else they are the 
 work of a man who was thoroughly familiar with the naviga- 
 tion of the Levant' and the topography of the Holy Land. 
 
 But, then, on the other hand, it is quite possible that many 
 of the apparently original details are due to assimilation. We 
 have pointed out two trifling instances already. Let us take 
 one of the most striking cases, the passing by Thessaly in Acts 
 xvii. 15. The man who added it, if it is an addition, was a 
 Levantine traveller. Was he the original writer of the book, or 
 a later wanderer ? 
 
 ' It is in such questions as these that Prof. Ramsay's criticism is so valuable 
 and suggestive. 
 
WESTERN TEXT OF THE ACTS. 83 
 
 The added passage in Cod. Bezae is 
 
 HApHAQeN Ae THN GeCCAAlAN 
 
 eKC0AY6H r^p eic AYTOYC 
 
 KHpY^AI TON AOfON 
 
 and we have shown that something like it was current in the 
 Old Syriac. 
 
 The obvious parallel is Acts xvi. 6 where the current text is 
 
 AirjXdov Be T7]v ^pvyiav koX TaXaTiK-rjv '^copav, KtoXvdevre'i 
 viro Tov CL'^lov •JTvevfiaro'; (D adds fji,iqhevX) XaXrjcraL tov Xoyov 
 
 (D adds TOV 6eov) iv rfi 'Acrla TrapeKd6vT6<i (D SteX^WTes) 
 
 Se Trjv ^va'iav Kare/37jaav (D KaTrjvr7)ijav) et? T/scoaSa. 
 
 The suggestion that the expansion in c. xvii. 15 was made 
 by a reviser in D, under the influence of the parallel passage 
 in xvi. 6, leads to the supposition that he found in his text 
 which he was reproducing the word ■rrapeXOovTe'i and not the 
 SieXOovTe'i which now stands in the Codex Bezae. That is 
 he was revising from what Blass calls the a-text. But this 
 is too hasty a conclusion, for the reviser, who is nothing if not 
 a geographer, may be using the word iraprjXdov because he 
 wishes to imply that St Paul went to Athens by sea (as we 
 may be morally certain that he did go), and so did not go 
 through Thessaly, but coasted by it\ 
 
 The precision of the language is such that it seems difficult 
 to ascribe it to a translator, who is merely incorporating an 
 assimilation in a Syriac or other version. And this same pre- 
 cision of geographical knowledge seems to come out in the 
 counter-change in Acts xvi. 6 where the D text substitutes 
 Si,eX66vTe<; because he imagines St Paul to have gone through 
 Mysia. Here again the change is so delicate that it hardly 
 seems likely that casual retranslations from the Latin or Syriac 
 could effect so much''. 
 
 1 Hence Prof. Bamsay is perhaps incorrect in saying {Church in the Roman 
 Empire, ed. 3, p. 160) that the reviser did not observe that Paul probably sailed 
 direct from the coast of Macedonia to Athens. 
 
 ^ Prof. Eamsay points out to me that " there was in reality no need for the 
 latter change, because in the Travel-document Si^pxoMO' is used with an accusative 
 of locality to indicate ' going from point to point along a road or over a country ' 
 
 6—2 
 
84 CHARACTER OF THE GLOSSES IN THE 
 
 Moreover when we turn to the Syriac of Acts xvi. 6, we find 
 
 n\\~«< r^.l r<lx..1Cla.l t<1*»OT ,.__OJf< AUaO 
 
 fSn CUi&J SAO (v. 8) . . . . r^-xflor^la r<'co»K'.'1 ooovlsa 
 
 " and the Holy Spirit forbade them to speak the word of God in 
 Asia;... and when they came out of Mysia," and this construc- 
 tion, which has every appearance of being close to the original 
 rendering (for it agrees with D in the insertion of the "word of 
 God' and implies the journey through Mysia in a form which 
 certainly suggests dependence upon the text of D) can hardly 
 furnish us with the material for the gloss in xvii. 15: for this 
 gloss does not imitate the construction, nor take up the equi- 
 valent of Tov Oeov, nor translate the equivalent Syriac word 
 into 'KaX-ija-ai. It will be seen, therefore, that there is a good 
 deal to be said against the hypothesis of Syriac assimilation. 
 Was it then a Greek assimilation ? This has more probability ; 
 the language is similar; the passage to which assimilation is 
 made stands as in the a-text without any addition to tov Xoyov ; 
 and it is possible that this may be the explanation. But I can 
 believe that Blass would find it possible to defend the original- 
 ity (and I may say the Lucauity) of the gloss on the ground 
 that -n-aprfKOev in Acts xvii. 15 could equally be taken in the 
 sense of 'neglect Thessaly,' especially when the explanation 
 was made that ' he was forbidden to preach the word to them.' 
 
 On the whole the examination of the passage leans rather 
 to the side of Blass, and in favour of an original element in the 
 Bezan text, than to the theory of a revision; and certainly there 
 is little to be said in support of Chase's position. We must 
 remember, also, in dealing with these parallel passages, that 
 Luke is a writer who repeats himself occasionally with variations, 
 
 in the performance of one's purpose, viz. preaching. This construction is found 
 only in the Travel-document and in 1 Cor. xvi. 5 — and nowhere else in Greek. 
 Luke uses bUpx. did or KaTa...€iayy£\t^6fj,6vos both in the Gospel and in 
 Acts i — xii. In contrast to this iraprj\$ov means ' neglect ' ' leave nnevan- 
 gelized, ' without laying any stress on the topography of the journey. The 
 narrator of the Acts has never the faintest interest in topography and geo- 
 graphy : he thinks only of the missionary character of the journey." 
 
WESTERN TEXT OF THE ACTS. 85 
 
 SO as often to suggest a double use of some source, or the exist- 
 ence of a document which has been translated. One has only 
 to compare two such passages as Acts ii. 44 sqq. and Acts iv. 32 
 to see what I mean. One account has kuI elxov oiravra Koivd, 
 the other a\X' rjv avroh iravTa koivo. ; one account says Kal 
 Siefiepi^ov, the other Kal SteSt'Sero ; one account has Tracri, 
 the other ivl eKaa-TO) ; while they both agree in the expres- 
 sion KaOoTi av Tt? ^pet'ai' el')(€v. What are we to say to 
 this ? is it the twofold use of some primitive document, or is it 
 the natural variation of a writer recording similar phenomena 
 twice over ? If it is the latter, then we must not be hasty in 
 assuming assimilations nor in determining their origin. For 
 Luke repeats himself with slight variations'. So much may be 
 said on behalf of Blass's theory. 
 
 Resuming now our discussion of the subject from Mr Chase's 
 point of view, we shall show briefly and rapidly that, as in the 
 case of the glosses, Mr Chase interprets the assimilations 
 wrongly, and assigns incorrect and unnecessarily obscure origins 
 to them, in those cases where a theory of Syriac assimilation 
 is most plausible. 
 
 I am going to draw his attention to what I consider to be 
 the best example extant of a Syriac assimilation in the text 
 of the Acts : and then to ask him to compare his own theory 
 of the genesis of the error, and see if he adheres to it. The 
 passage is Acts xv. 26 which stands in the Codex Bezae 
 
 nAp^AEACOKACIN THN ^fYXHN <\YTCON 
 
 ynep toy onomatoc toy i<Y hmcon ihy XPY 
 eic nANTA neipACMON. 
 
 The gloss is in the last line ; it should come a line higher up, 
 
 ' Compare also the case which w£ discussed previously where xiii. 4 and 
 XV. 30 are contrasted and note the variation in the similar accounts. Observe 
 further that the reviser (if he be different from the author) can be proved not 
 to make his assimilations in a verbatim manner. Take the great gloss, for 
 instance, which the Heraclean margin has preserved for us in Acts xxv. 24, and 
 compare such sentences as "dicebam ut sequeretur me in Caesaream ubi 
 oustodiebatur...C[Uum autem dicerem, Vis judicari cum iis Hierosolymae, " with 
 the parallels in the received text, and it will be seen how freely the author of 
 these sentences handles his materials. Assimilation does not mean agreement. 
 
86 CHARACTER OF THE GLOSSES IN THE 
 
 but that does not sensibly affect the sequence of the passage. 
 The necessity for a gloss arose from the ambiguity of the 
 translated word nApAAeAwKAciN, which would be much better 
 for a word or two of explanatory matter. In the present case 
 the source of the gloss is not obscure ; it is due to the influence 
 of a famous passage in Sirach ii. 1 : " My son, if thou draw 
 near to the service of the Lord, prepare thy soul for temp- 
 tation," which the Syriac renders 
 
 "thou hast surrendered thy soul to all temptations." 
 
 Everything would seem to be adequate for the calling up of 
 the gloss : it was necessary to add something for clearness ; 
 and the rendering of the text in Acts »__oeooiiaj OSaix.r^.i 
 was sufficient to make the connexion at once with the famous 
 verse in Sirach'. 
 
 Mr Chase's solution of the passage is as follows : 
 
 " What of the gloss et? iravTa ireipaa fiov which is also 
 found in Cod. E ? The only other passage in the N. T. where 
 the expression 'every temptation' occurs is Lc. iv. 13 avvTe\ea-a<s 
 iravTa TreipaafLov. In this latter passage, the Syriac Vulgate 
 has ' all-of-them his-temptations,' but the Arabic Tatiau ' cum 
 consummasset diabolus omnem tentationem ' (Ciasca, p. 8) is 
 good evidence that the Old Syriac had a literal translation of 
 the Greek. In the Greek there is no point of contact between 
 Lc. iv. 13 and Acts xv. 26. But in the Syriac it will be 
 noticed that ^nlx. = o-wreXe'cra? ; in Acts OSolx.r<'=7ra|Oa8e- 
 BwKaai ; the same verb, that is, is used in the two passages 
 though not in the same sense. I would suggest, therefore, 
 that, when some Syriac scribe wished to qualify the words 
 ' men who surrendered their soul,' the sound of the verb carried 
 his thoughts to Lc. iv. 13, therefore he appended the phrase 'to 
 (A) every-temptation".' " 
 
 1 A verse by-the-bye which is the heading of a famous chapter in the 
 Imitatio Ghristi. If the explanation be correct, it is in evidence for the early 
 translation of Sirach. 
 
 ' The A is, of course, not in the Syriac of Luke I 
 
WESTERN TEXT OF THE ACTS. 8t 
 
 It is a hard exercise of faith to he told to connect two such 
 unlike situations by means of the sound of a single ambiguous 
 word which is admitted to be used in two opposite senses ! We 
 will make Mr Chase a present of our interpretation, for the 
 better propagation of his theory. His assimilation to the Gospel 
 is a false assumption, and must be abandoned. 
 
 It would also be wise to remove from his pages such cases 
 as make the prophets of the Old Testament responsible for 
 ordinary turns of speech in Syriac. For example, Acts vii. 43 
 appears in the ordinary text as 
 
 KOi fieroiKiw Vfidt; i-rreKeiva Ba^vKrSvo<;, 
 
 which the Codex Bezae, as far as we can judge, altered into 
 
 = in illas partes Babylonis. 
 
 The passage is an interesting one, because the peculiar word 
 iireKetva inspires confidence, and is the very word used in the 
 LXX. of Amos v. 27, from which the quotation comes. If it is 
 not genuine, we should have a prophetical assimilation indeed, 
 but on the wrong side of the house and, as far as Mr Chase 
 is conceroed, in the wrong tongue ! But it must surely be 
 genuine. 
 
 Now Mr Chase suggests that eVt to, jMeprj is a re-translation 
 of a Syriac A.a.3.1 K'AxoiA^rtL^, and under the influence of 
 a number of passages in the prophets. No one could I suppose 
 object to a translation of the name of a country which either 
 in Syriac or in Latin took the form in partes Babylonis. The 
 idiom is sufficiently common in both languages. It occurs con- 
 stantly in Syriac, at all events, and here are a couple of cases 
 from the recently published Commentary of Ephrem on the 
 Pauline Epistles : 
 
 ICor. xi. 16: nos, in partibus nimirum SjTiae, 
 
 talem consuetudinem non habemus, 
 
 where the idiom is Ephrem's own = we Syrians ; and 
 
 2 Cor. i. 8 : de tribulatione nostra, quae facta 
 est in partibus Asiaticorum, 
 
88 CHARACTER OF THE GLOSSES IN THE 
 
 where the idiom is that of the primitive translator of the 
 Pauline epistles, rendering ev r^ 'Aa-la. 
 
 Surely we need not spend time in proving further that 
 either in Syriac or in Latin in partes Bahylonis is idiomatic. 
 Now let us transcribe Mr Chase's explanation, p. 73 : 
 
 " The Vulgate Syriac does not help us ; it has the phrase 
 used in the Syriac of Amos v. 27 (comp. Matt. viii. 30, 2 Cor. x. 
 16). It is, however, through the Syriac that a solution of the 
 problem comes. The word rf'i^ri' (= the place) is used to 
 translate to, fiipn in Matt. ii. 22, xvi. 13 by the Old and 
 Vulgar Syriac, and by the latter version (the Curetonian 
 fragments here failing us) in Mc. viii. 10. In Acts ii. 
 10, xix. 1, XX. 2 the same Greek phrase is represented by 
 t<'AiO'iA»t<' (=the places). But this word 'places' takes us 
 back to a series of passages in the Prophets : Jer. viii. 3 
 'all the residue... which remain in all the places whither I 
 have driven them,' Jer. xxiv. 9, xxix. 14, xl. 12 'then all the 
 Jews returned out of all places whither they were driven,' Ezek. 
 xxxiv. 12 'I will deliver them out of all places whither they 
 have been scattered in the cloudy and dark day.' Thus in the 
 Prophets of the Captivity 'the places' is almost a technical 
 expression meaning 'the foreign countries of exile'.' The 
 Syriac 0. T. has r^^ai^r^ in all these passages (comp. Jer. 
 xlv. 5, Amos iv. 6) except Jer. xxix. 14 where the Syriac is 
 varied, possibly because the word in the singular occurs later 
 in the verse (' I will bring you again unto the place whence I 
 caused you to be carried away captive'). We may therefore 
 with some confidence believe that the Old Syriac read in Acts 
 vii. 43 \-i-i ^ t^^oi^rell (to-the-places of-Babylon) and we 
 can see the rationale of the reading. The Old Syriac is more 
 suo harmonising, embellishing a quotation from one Prophet 
 with a characteristic expression of other Prophets who deal 
 with the same subject of the exile." 
 
 I forbear to criticise a passage where I can only praise the 
 industry of the author; I am afraid it will offend Syriac 
 
 1 I suppose like ' in partibus ' in ecclesiastical circles for ' in partibus iufide- 
 lium ' ! 
 
WESTERN TEXT OF THE ACTS. 89 
 
 scholars, and prejudice them against the theory which 
 Mr Chase is propounding, and indeed I doubt whether a single 
 one of the many references to the Syriac Old Testament which 
 he detects, either here or elsewhere, will be accepted. 
 
 There are, no doubt, passages which will lend themselves 
 to elucidation on the hypothesis of glossing in and from the 
 Syriac ; and there are passages which will not do anything of 
 the kind. We have done something, in the foregoing argu- 
 ment, to remove obstacles from the pathway of Mr Chase's 
 readers : but the more we think of it, the more complex does 
 the Bezan text become. It has passed through the hands of 
 a number of people of active mind, whose remarks are stratified 
 in the Western text ; they are not all of them Syrians ; and it 
 is not yet even proved that there are no Western expansions 
 which are original. The whole history of the text requires 
 renewed and careful inquisition, without prejudice in favour of 
 the solvent power of a single hypothesis. 
 
 I shall conclude my discourse, in the interests of poetical 
 justice, by showing that there is one important gloss in the 
 Acts which Mr Chase was unable to explain, and which does 
 not seem to yield to any theory except that of a primitive 
 Latin reaction. 
 
 In Acts iv. 31, we have in the Codex Bezae 
 
 KAI GAaAOYN ton AOrON TOY eV MCTA UAppHCIAC 
 
 UANTI Tco eeAoNTi niCTeyeiN, 
 
 = et loquebantur uerbum di cum fiducia 
 omui uolenti credere. 
 
 The gloss is an old one, occurring in Irenaeus, both in Greek 
 and Latin, and in a less pronounced form in Augustine, who 
 omits the last word. 
 
 Its origin is evidently an attempt to assimilate the fulfil- 
 ment of the prayer to the prayer itself which is in v. 29 
 
 METd, UACHC nAppHCIAC AaAEIN TON AOfON COY 
 
 cum fiducia omni loqui uerbum tuum. 
 
 Hence we expect naturally the addition of Trao-jj?, and a 
 number of MSS. show it. (For example, the Gigas reads 
 
90 CHARACTER OF GLOSSES IN WESTERN TEXT OF THE ACTS. 
 
 loquebantur verbum del cum omni fiducia.) This is the cause 
 of the omni at the beginning of the gloss : but this omni 
 separated from fiducia by the line division has been read as a 
 dative, and turned back into Greek as wavrt with the result 
 that it has itself become the subject of expansion, in order 
 to limit the extravagance of the statement and to round off 
 the sentence. (The Syriac, of course, has no parallel pheno- 
 mena; it cannot translate Trapprjcria except by a circumlo- 
 cution, nor, when it does translate, can it make any distinction 
 between Trapprjaia and iraaa irappTfaia.) Such is, we believe, 
 the origin of the gloss. But as we have used so much 
 ■7rapp7](TLa in explaining to other people the errors into which 
 they have fallen, we must also promise to be modest, and to 
 retract this opinion, and any others which we hold, as soon as a 
 better explanation can be offered, and we can fairly be shown 
 to have made an erroneous statement '. 
 
 ' In the foregoing remarks I have avoided the discussion of certain test 
 passages which Mr Chase considers decisive, because they are not, at all events 
 as presented by him, of the nature of proof. It is not fair, for example, to 
 quote the reading "their sons and their daughters" in Acts ii. 17, in proof of a 
 Syriac origin of the Bezan text of the Acts, and to support the statement by 
 reference to Tertullian {adv. Marc. v. 8), without at the same time informing the 
 reader that Tertullian is expressly, and from the necessities of the case, quoting 
 Joel against Marcion, and that the Bezan text shows signs of having been 
 corrected to the text of Joel ! The argument needs re-statement, to say the 
 least. 
 
NOTE. 
 
 ON A POSSIBLE CONNEXION BETWEEN 
 CHRYSOSTOM AND EPHREM. 
 
 The foregoing pages have been largely occupied with a discussion 
 of the character of the text of the Acts which underlies the com- 
 mentary of Ephrem, some of whose fragments we have collected from 
 the Venice Catena. And it has been shown that the text upon 
 which Ephrem was working was a text which, at all events in the 
 latter half of the book, was closely related to that of the Codex 
 Bezae. 
 
 We are now going to ask the question whether the lost com- 
 mentary of Ephrem has had any effect upon the extant commentary 
 of Chrysostom ; did Chrysostom know Ephrem's commentary, either 
 in Syriac or by a translation, and was he under the influence of if! 
 The question is interesting, not only on account of the explanation 
 which it may possibly furnish of certain textual phenomena in 
 Chrysostotn's writings, but also because there has been a wide-spread 
 suspicion of some foreign element in Chrysostom's commentary on 
 the Acts. Many scholars have gone so far as to deny that the work 
 is rightly ascribed to Chrysostom, on account of the repetitions and 
 other literary imperfections which characterise it ; and while, on the 
 other hand, a comparison of the general method of interpretation 
 with that employed on the Pauline epistles would seem to justify the 
 belief in a common authorship, there still remains the suspicion 
 which is provoked by the internal evidence of the commentary, 
 which leads us to say that it cannot be wholly due to Chrysostom, 
 and certainly not to Chrysostom at his smoothest and best. 
 
 Now it will be interesting to see whether the Venice Catena 
 throws any light upon the question of authorship suggested in the 
 
92 CHRYSOSTOM AND EPHREM. 
 
 foregoing remarks. It will not be easy to draw conclusions from 
 the study of a Catena which is largely made up out of Bphrem 
 and Chrysostom, on account of the confusions and errors to which 
 Catenae are peculiarly liable. If, for example, we were to remark 
 that a certain section was ascribed to both Ephrem and Chrysostom, 
 we should not be justified in assuming that the Catenist found similar 
 matter in the commentaries of both writers, and therefore assigned a 
 double authorship ; for, in the first place, it is not an uncommon 
 thing for a Catenist to dovetail together extracts from different 
 writers; and in the next place, we have no collateral evidence for 
 Ephrem's Commentary to which we can appeal in order to show that 
 the Catenist's ascription of authorship is correct. Unless we can 
 find, then, cumulative evidence of a number of passages ascribed to 
 Ephrem-Chrysostom, of which the whole matter is substantially 
 Chrysostom, we can hardly argue that there was common matter in 
 the two commentaries. 
 
 Nor would the difficulty be much less in cases which were ascribed 
 to Ephrem alone, but which were shown, by a reference to the 
 printed text, to belong to Chrysostom. The heading might be an 
 error of the original Catenist or one of his transcribers. So that we 
 should again demand cumulative evidence, before we could discard 
 the suggestion of error in the titles, and say, this passage is both 
 Ephrem (as is proved from the Armenian Catena) and Chrysostom 
 as is shown by his printed works. 
 
 Probably the best test would be to examine the Commentary of 
 Chrysostom for evidence of another biblical text underlying his 
 remarks, which should not only differ from his own text upon which 
 the remarks are ostensibly made, but should agree with the text of 
 Ephrem, or, which is the same thing, with those peculiar features of 
 the Bezan text which we have shown to characterise the text of 
 Ephrem. If such an analysis led to a satisfactory conclusion, we 
 could then with some confidence go over the sections which had a 
 double ascription, and those which seemed to have an incorrect 
 ascription, and perhaps justify the headings in doubtful cases. 
 
 The following remarks are meant to suggest the line which such 
 an enquiry should take. We will begin by examining Chrysostom's 
 remarks on certain passages (in which our references are made to 
 Savile's edition). 
 
 p. 845 (c. xix. 21) WiTO 6 IlavXos iy t(5 Trveu/xari 8ieX0u)V ttJv 
 MaKeSoviav koX 'Axaiav, iroptvccr^at cis 'Iepoa-6\vfii.a. The text is the 
 
CHRYSOSTOM AND EPHREM. 93 
 
 current one; but Chrysostom goes on ovk 2ti avOpwirLvoi? ivravOa ttoici, 
 a\Xa irvtvfiaTi, w kw. vpociXero SieX^ctv. The suggestion that Paul had 
 at one time been wishing to act dvOpioirivui^, in his prospect of the 
 journey to Jerusalem, takes us back to the gloss at the beginning of 
 the chapter (Acts xix. 1) where the Bezan text tells us 
 
 eeAoNTOC Ae toy n^Y^OY 
 
 KATA THN lAlAN BoyAHN 
 
 TTopeYecGM eic \epocoK-fmd., 
 
 aud the Armenian Catena expressly assigns this peculiar reading to 
 the text of Ephrem. We must therefore say that, as the demon- 
 stration has already been clearly made of the agreement of the 
 Ephrem text with that of Cod. Bezae, we have traces of an ac- 
 quaintance of Chrysostom with a famous Bezan-Ephrem reading. 
 There is no question here of an error in the heading of the Armenian 
 Catena. The commentary, then, made by Chrysostom belongs to the 
 Ephrem-text rather than to his own. 
 
 If now we keep before our minds the term which Chrysostom 
 uses {dvOpoyiTLvoi^) to describe the action of Paul when he plans in 
 his own will and has to be restrained by the Spirit, we can throw 
 some light on the following passage, p. 816 (c. xvii. 14) w's im Trjv 
 6<iXa<7crav' virifxivov Se o re %lXa^ kox 6 Tip.oOeo'; ekei. Opa airov Kac 
 VTro)(Uipovi'Ta Koi ivLdToijjLtvov Kou TToWa d.v9p(xnrLvu><i iroiovvTa. 
 
 The passage is one in which the Bezan glossator has told us that 
 Paul was acting under the peculiar restraint of the Spirit, 
 
 TTApHAeeN Ae thn eeccAAiAN 
 eKwAYSH r^P eic aytoyc 
 
 KHpYiAl TON AOrON. 
 
 It is implied, as a comparison with parallel passages shows, that 
 Paul had been planning to do something else than go to Athens : 
 he had been acting dvdpunrivta';. And Chrysostom points out what 
 he had been doing; he had been retreating and pressing forward ; 
 now if we refer to Ephrem's remarks on this passage, we shall find 
 that Chrysostom's word wox<op<uv exactly corresponds to 'giving 
 way' or 'receding' which we detected in the extract of Ephrem; and 
 it was shown, also, that at this very point Ephrem must have had 
 the Bezan gloss quoted above. 
 
 Either then this word vTroyi^piw is a part of Ephrem's commentary, 
 in which case Chrysostom's use of that commentary is demonstrated ; 
 or it was a part of its text ; in which case we have a second time 
 
94 CHRYSOSTOM AND EPHREM. 
 
 shown that the Bezan-Ephrem text underlies the remarks of Chry- 
 sostoin. 
 
 We will next show that certain remarkable Bezan additions 
 were known to Chrysostom, for which we have not yet recovered the 
 evidence of Ephrem. Take the case of the tears of Simon Magus, 
 the antiquity of which gloss is, as Mr Chase has pointed out, demon- 
 strated by Tertullian's allusion {Simon Samarites.../ru,itrajlevity. 
 
 Chrysostom says that his tears were merely formal; p. 714, 
 Acoi/ aTTO KopSias ixeroMorrjaai, Beov KX.av(rai koI TrivByjaaf 6 h\ d<f)0- 
 eriaxrei jxovov toCto Troia" d apa d(j>i6rj(TeTaL a-oi' tovto tl-Trev, ov)( (us 
 ov (Tvyxo)prj6ivroi av avTiS, el iKXav(Tev ktL It is, therefore, certain 
 that Chrysostom knew the tradition about Simon Magus' tears. 
 
 The importance of this piece of evidence is that we are almost 
 obliged to equate the frustra of TertuUian with the d<^ocnui<Te.i of 
 Chrysostom and to carry their equivalent into the Western text ; 
 now if this could be done thi-ough the Syriac coA 'A -^ r^Ao 
 (=and he did not really care), we could perhaps connect the Bezan 
 reading by cnA -^X^ rdio (= and he did not cease); we should 
 thus find an original for the curious koI ov SieXlfLTravev of the Bezan 
 Codex. The question would then arise as to the primitive form 
 and language of the gloss, which was extant originally in the sense 
 that ' Simon Magus pretended to weep^' 
 
 By the foregoing instances, we have done something to prove 
 that there underlies the Commentary of Chrysostom a glossed text 
 of the Acts something like the Bezan-Ephrem text, and differing 
 from the text that accompanies the commentary. Bearing this in 
 mind, let us now turn to Chrysostom on Acts xviii. 19 (Savile, 
 p. 831): he tells us 
 
 8ia TOVTO EKtoXueTO £is Trjv 'Acriai/ ikdiiv, Trpos to. KareTrtiyovTa ot/xat 
 iXavvofi.evo^, opa yow ivravda koI TrapaKaXov/jLevov avTov fielvai, koL 
 OVK avi^ofi,ivov, iireiS-ij ijTrciyeTO d-mXdelv ov fj.i]v d7rX(3s avTOVs etaaev, 
 aXXd jxerd vrroc^eVcws. 
 
 1 De anima, c. 34. 
 
 2 The somewhat similar ease is that of the peculiar description of GaUio's 
 conduct (Acts xviii. 17, "Gallio cared for none of these things"), which appears 
 in the Bezan Latin as 
 
 tunc gallio fingebat eum non uidere 
 and in the Fleury text as 
 
 et gallio simulabat se non uidere. 
 
CHKYSOSTOM AND EPHREM. 95 
 
 If we compare this ■with the extract of Ephrem, printed on p. 48 
 of the present tract, we find the words 
 
 " He did not choose to stay longer with them ; because whither 
 he was concerned to go, thither he had to go. However he did not 
 simply leave them, but with a promise that they should expect his 
 coming by the will of God." 
 
 The exact coincidence both in thought and language between the 
 two extracts is obvious ; but there is enough diflference in the hand- 
 ling of the argument to show that we cannot correct the ascription 
 of the Catenist from Ephrem to Chrysostom : on the other hand the 
 sentence Spa yovv ktc is exactly in Ohrysostom's manner; and we 
 can only conclude that the two commentaries were substantially 
 giving the same thoughts in the same language : and as we are 
 clearly dealing here with commentary and not with text, this means 
 that Chrysostom has been incorporating the idea of Ephrem. 
 
 Turn in the next instance to Ohrysostom's note on xvi. 39 
 (p. 812): 
 
 <j>ol3ovvTai OTi 'Pioyttatot elaiv, ov)( otl dStxcos ive/SaXov Kai ■qpuyrqirav 
 avTOVi l^eXOuv d-KO njs iroAecos' X"P"' 'QT<)<^'>-v ravrqv ktL 
 
 and a reference to Ephrem in loc. shows that the concluding words 
 are either a gloss in the text, or borrowed from the commentary of 
 Ephrem : " So then we ash of you this favour, depart from this city, 
 lest the same men, etc." 
 
 But a reference to the Bezan text renders it tolerably certain that 
 the words in question are merely an equivalent to the irapcKaXecrav 
 of the text : so that we are obliged again to conclude that Chryso- 
 stom has been borrowing from Ephrem. 
 
 Next let us compare p. 721 (c. ix. 4) : 
 
 aA.Xa TOVTOV jj-ovov i-mjpiacre /cat ecr/Secrcv avTov tov dvp.6v t(S <^o/3a) 
 <3crT£ avTOv aKOvcrai ra Xeyo/neva 
 
 where Ephrem's remark is (if we may trust the Catena), " With 
 the light he blinded him and so frightened him and with awful fear 
 of his glory he extinguished his rage " ; here, again, the dependence 
 of Chrysostom on Ephrem seems to be established. 
 
 Our last instance shall be from p. 713 (c. viii. 19) : 
 ircus ovv, <f)r](rL, iryevjua ovk IXaySov ovtol; irveuixa. iXajSov to -rijs 
 ad)OT£(i)s" TO 8e Ttoi/ o-rjju.ei'fcji' ovw<i> ^(rav Xa|SovTes' Kai ot(, tovto otti, koL 
 
96 CHRYSOSTOM AND EPHBEM. 
 
 TO Twv (TrjixiLwv irvtv/ia ovK iXapov, opa ttuJs iSwv o 26/icuv Trpoff^Xpt 
 
 TovTO aiTeiv. 
 
 We notice here the curious distinction between the Spiiit of 
 grace and the Spirit of gifts, by which distinction Chrysostom tries 
 to explain the fact that baptized persons had not received the Spirit. 
 The expression which he uses, the Spirit of signs, is taken from 
 Ephrem, as may be seen from the extract quoted on p. 41 ('they 
 sent Peter and John that by their laying on of hands the Samaritans 
 may receive the Spirit of signs'). 
 
 Reviewing the instances which have been brought forward, we 
 think a good case has been made out for the theory that the rough- 
 ness of Chrysostom's commentary on the Acts is due, in part, to the 
 fact that it is based upon the previously existing commentary of 
 the great Syrian father. 
 
 ADDENDUM. 
 
 The Maroionite reading of Galatians iv. 27, referred to on p. 19 
 of the present work, will be found again, though in a less obvious 
 form, in Ephrem's Commentary on the Diatessaron (p. 34, ed. 
 Mcisinger), as follows : ' Vide quomodo isti filii tui locum princi- 
 palem acceperunt in Jerusalem, quae sursum est, matre nostra, 
 quam laudamus {more correctly confessi sumus) quae Moysi apparuit 
 in monte.' 
 
 cameeidqe; pbintkp bt c. j. olav, m.a, anp sons, at the cnivkksity press 
 
WORKS BV THE SAME AUTHOR. 
 
 Stichometry. Demy 8vo. 5s. 
 
 On the Origin of the Ferrar-Group. A Lecture on the 
 Genealogical Relations of New Testament MSS. Delivered at Mans- 
 field College, Oxford, on Nov. 6th, 1893. Demy 8vo. \s. ed. 
 
 Biblical Fragments from Mount Sinai. Demy 4to. 10s. Qd. 
 
 Fragments of Philo, newly edited. With two Facsimiles. 
 Demy 4to. 12s. 6d 
 
 Leicester Codex of the New Testament, the Origin of. 
 
 With three Facsimiles. Demy 4to. Ws. Qd. 
 
 Teaching of the Apostles. Newly Edited, with Facsimile 
 (Autotype) Text and a Commentary, for the Johns Hopkins Univer- 
 sity, Baltimore, from the MS. of the Holy Sepulchre (Convent of the 
 Greek Church), Jerusalem. Demy 4to. £1. Is. 
 
 The Codex Sangallensis (A). A Study in the Text of the 
 Old Latin Gospels. Royal 8vo. 3s. 
 
 Diatessaron of Tatian. A preliminary Study. Royal 8vo. 5s. 
 
 The Acts of the Martyrdom of Perpetua and Felicitas ; 
 
 the Greek Text, now first edited by J. Rbndel Harris, and 
 Seth K. GiFroED. Royal 8vo. 5s. 
 
 The Rest of the Words of Baruch : A Christian Apoca- 
 lypse of the year 136 a.d. The Text revised with an Introduction. 
 Royal 8vo. 5s. 
 
 Some Interesting Syrian and Palestinian Inscriptions. 
 
 Royal 8vo. 4s. 
 
 Texts and Studies : contributions to Biblical and 
 Patristic Literature. 
 
 Vol. I. No. I. The Apology of Aristides on behalf of 
 the Christians. Edited from a Syriac MS., with an Introduction and 
 Translation by J. Rbndel Harris, M.A., and an Appendix containing 
 the chief part of the Original Greek, by J. Armitage Robinson, B.D. 
 Second Edition. Demy 8vo. 5s. Net. 
 
 aonHon : 0. J. CLAY and SONS, 
 
 CAMBRIDGE UNIVERSITY PRESS WAREHOUSE, 
 
 AVE MARIA LANE. 
 
 ffilaSgoto : 263, ARGYLE STREET.