Saron (Hastings' Dictionary)

The above sketch of the history of the Diatessaron proves beyond question: (1) That this form of the Gospels was very widely, if not ex- clusively, used in the Syrian Church during the 3rd and 4th cents. ; and (2) that the work continued to be known and read by Syriac writers down to the beginning of the 14th century.

But the evidence of the existence and influence of Tatian’s Harmony is, as we have pointed out, not confined to the Pah Hons and references of Syriac commentators ; or though the Syriac Diatessaron has unfortu- nately not been preserved to us, yet we possess both Latin and Arabic translations of Tatian’s work, together with an Armenian version of St. Ephraem’s commentary upon it.

These versions in themselves furnish incontrovertible proof of the great esteem in which the Harmony was held, and in that respect form a most important addition to the evidence set forth above; but their chief value lies in the fact that by their means we are enabled to obtain some conception, not only of the order and arrangement of Tatian’s work but also of its actual text. (a) The Armenian version of St. Ephraem’s com- mentary on the Diatessaron.—lt has been already suggested that St.

Ephraem’s commentary on the Diatessaron contributed in no slight measure to the preservation of the latter work. For the honour and esteem in which that writer and his works were held by the Syrian Church naturally extended to his Gospel commentary, and ensured the survival of Tatian’s work at least in that form.

We may even go further, and assume that the example set by so prominent a writer as Ephraem exercised a considerable influence on his successors, who were thus led to study—and so to preserve— a work which otherwise seemed destined to dis- appear.

But, whatever its influence in the past, it is undoubtedly true that in modern times the publication of a Latin translation of the Armenian version of this commentary has been the means of once more arousing the interest of scholars in the Diatessaron, and of rescuing it from that oblivion to which it had been so long assigned. The Ar- mInenian version of the commentary first appeared in the edition of St. Ephraem’s works issued in four volumes by the Mechitarist Fathers of St. Lazzaro in 1836.

A Latin translation of the com- mentary was prepared by J. B. Aucher, one of the editors, as early as 1841, but was not published. The work was finally made accessible to scholars by Prof. Moesinger, who in 1876 published Aucher’s translation, which he had revised and corrected by the aid of another MS, under the title: Hvangelw Concordantis Expositio facta a Sancto Ephraemo Doctore Syro.

Both the MSS on which this trans- lation is based date from the year 1195, but the version itself is assigned to the 5th cent. (Moesinger, . xi). That the Armenian version was made rom the Syriac commentary of St. Ephraem seems to be fully established,* and we are therefore justified in treating the work as genuine.

Among the first to recognize the great import- ance of Moesinger’s translation was Professor Zahn, to whom, indeed, all NT scholars are largely in- debted for a knowledge, not only of a considerable part of the Diatessaron itself but also of a large number of facts bearing on its history and char- * Zahn, Forsch. i. 46f.; J. Hamlyn Hill, A Dissertation on the a Convnentary of S. Ephraem the Syrian, Edinburgh, acter.

Those portions of the text which the author was able to restore with the aid of the new trans- lation (and also of the Homilies of Aphraates) were incorporated in the first volume of his Forschungen zur Geschichte des neutestamentlichen Kanons und der altkirchlichen Literatur (Erlangen, 1881).

In this work Zahn further gave a full and complete account of all that was then known of the Diates- saron and its author, and, to a large extent, solved the many complicated and difficult questions which are connected with its origin and history. The main interest of the volume, however, centres in his brilliant restoration of the text, and in the evidence which he has adduced in support of his reconstruction. We therefore nopend.

a brief de- scription of the methods employed by Zahn in recovering the lost text. The task of reproducing the order and arrange- ment of the Diatessaron, as Zahn discovered, was materially lightened by the character and form of Ephraem’s commentary. For the latter consists of a series of lectures or discourses, which are largely homiletic both in form and substance, and appear to have been delivered orally.

Moreover, each discourse was apparently preceded by the readin of the Gospel section which formed the subject o discussion ; and though, unfortunately, the text of the section was not included in the commentary, the discourse itself affords sufticient evidence for identifying the passage of Scripture thus com- mented on. In reconstructing the text itself Zahn had to fall back upon the Gospel quotations contained in the commentary as translated by Moesinger,* and in the Homilies of Aphraates.

t+ The Gospel quotations that occur in the Latin translation of Ephraem’s commentary naturally form the basis of the text. Those quotations,t however, which occur —chietly by way of illustra- tion—out of their context, i.e. in other discourses than the one to which they belong, are inserted in square brackets, as also the quotations from Aphraates, the latter being given, for the sake of clearness, in German instead of Latin.

Further, all quotations, whether in Ephraem’s commentary or in the Homilies of Aphraates, which are not given literally, but freely reproduced, are printed in italic type ; and, lastly, all Zahn’s own additions, e.g. references, etc., are enelosed in round brackets.

In the voluminous notes appended to each section Zahn has compared the Curetonian and Peshitta versions, and, in many cases, also the Harklean ; while in the more important passages reference is made to the Greek MSS (8, B and D) and to the Itala MSS. It is no slight tribute to the skill and ingenuity of Zahn to say that he has by these means succeeded in restoring not only the broad general features of the Diatessaron, bat also, to a relatively large extent, its actual text.

The former we are able to control by means of the Latin (Codex Fuldensis) and Arabic translations of the Diatessaron, which confirm in the most striking manner the accuracy of Zahn’s deductions ; but these versions, unfortu- nately (see below) afford but little assistance in restoring the actual text. Enough, however, of the original Diatessaron * A more accurate English translation from the Armenian MSS by Canon Armitage Robinson is given in Appendix x. to Dr.

Hamlyn Hill’s he Earliest Life of Christ ever compiled from the Four Gospels, being the Diatessaron of Tatian, etc.; and in the same writer’s Dissertation, p. 75f. + Cf. Baethgen, Hvangelienfragmente, p. 62f., who points out that, though Aphraates knew and used the Diatessaron, his Aha quotations are not taken exclusively from that work. { The doubt expressed by Zahn as to whether these quotations were taken from the Diatessaron or from the Peshitta has now been dispelled once and for all.

Since the publication of Mr. Burkitt’s work on S. Ephraem’s Quotations from the Gospel, it may be regarded as certain that Ephraem did not use the Peshitta. 456 DIATESSARON has in this way been restored to enable us to make out both the object of the author and the methods which he followed.

It is clear that the object of the Harmony was not to detract from, or impair, the authority of the four canonical Gospels, which undoubtedly form the basis of Tatian’s work, but rather to put together a single connected account of the life of our Lord, which should contain all that was essential in the narratives of the Gospels.

It was thus a popular rather than a learned work, and was designed to obviate those difficulties to which the fourfold form of the Gospels was only too apt to give rise. But, though the author was fully convinced of the genuineness of his sources, he did not adopt, as Zahn puts it, a ‘superstitious attitude’ towards them.

He rightly perceived that divergent accounts did not necessarily imply more than a single occurrence(of the same incident, and acted accordingly; while in cases of actual (diserepancy or contradiction he boldly followed one authority to the exclusion of theothers. Thus he followed St. Mark’s(10**) narrative of the healing of one blind man after leaving Jericho, in prefer- ence to that of St. Matthew (20 8°), who speaks of two blind men, and to that of St.

Luke (18%), who laces the miracle before the entry into Jericho. n this payer he appears to treat all four Evan- gelists as of equal authority ; but, in the main, his scheme of our Lord’s public ministry, which ex- tends over three Passovers, is based on the Fourth Gospel. In detail, however, the latter is treated with the same freedom as the Synoptists. Thus the purification of the temple (Jn 2'4-) and the discourse with Nicodemus (31-2!)

are transferred to the Feast of the Dedication at Jerusalem in the last winter of our Lord’s life. The following ex- ample will perhaps give a better illustration, not only of the boldness with which Tatian treated his sources, but also of the keen insight and judgment displayed by Zahn in tracing out and determining the principles which appear to have guided him :— We find in the Harmony that Jn 6!-*! (§ 34. Feeding of the 5000 just before the second Passover) pre- cedes Jn 444 (§ 38.

Discourse with the woman of Samaria) and Jn 5 (§ 40. Visit to Jerusalem and cure at the Pool of Bethesda), the two latter being separated by § 39 (the healing of the leper, Mt 8*-4, Mk 14-4, Lk 52-4), But § 38 (Jn 4“:), which forms part of a journey from Judea to Samaria, seems to be inconsistent with the preceding sections (14-37), which (with the exception of § 25) describe the Galilean ministry.

A further difficulty is pre- sented by the isolated position of § 39, which is the only incident belonging to the Galilean ministry which, presumably, stands in its proper position between the journey to Galilee (Jn 4*) and the visit to Jerusalem (5!) The correct explanation is, no doubt, that offered by Zahn, who points out that Tatian has reversed the order of St. John, and assigned the two days’ sojourn at Samaria (Jn 4%, § 38) to a journey from Galilee to Jerusalem.

Tatian’s procedure is bold, but it involves no alteration of Jn 4442, since these verses include nothing which requires that the sojourn in Samaria should form part of a journey from Judza to Galilee. The remaining verses of ch. 4, it is true, clearly point to such a journey, but they form no part of § 38. For vy.) had already been given in § 13; of vv.*-46 Tatian had only utilized v.“ in § 32, while it is doubtful if v.“ ever formed part of the Harmony : according to Zahn, vv.

*>-®4 also were omitted by Tatian, their place being taken by § 39.+ It naturally follows from this alteration of Ct. § iii. c. According to the Arabic version, no account of the first Passover is given by Tatian; this, however, does not affect the length of our Lord’s ministry. t These verses, however, occur both in the Latin and the 4rabic translations, though in different contexts. DIATESSARON Tatian that the scene of § 39 is transferred from Galilee to Judea.

This second change, however, is certainly an improvement from the point of view of the history, for Mt 84, Mk 1“, Lk 5™ appear to presuppose easy access to the temple and its priestly ritual. The complete chronological scheme underlying the Diatessaron, which has thus been restored, is as follows :— Sections 1. The Logos, Incarnation and Childhood of our. Lord. ‘ : ; : 1-7 The first Manifestation . fs 0 . 8-12 The beginning of His public ministry.

First Passover (Jn 2%) P : 0 13 Jesus in Galilee : ‘ 4 . 14-37 Journey through Samaria. Second Passover (Jn 5) . = c = . 388-40 Sojourn in Galilee. : : 5 . 41-51 Visit to Jerusalem. Feast of ‘aber- nacles (Jn 77) , : ‘ ; 5 §2 Journey to (Perza or) Galilee and back 53-58 59-71 : 6 PAE . From the Triumphal Entry to the In- stitution of the Lord’s Supper . 74-89 Feast of the Dedication in Jerusalem . 12. Passion, Resurrection, and Ascension . 90-100 SIG ERIE eo i Raising of Lazarus.

Sojourn in Eph- raim and return to Bethany i] _ (6) Codex Fuldensis.—That the Latin Harmony of the Gospels discovered by Victor, bishop of Capua, about the year A.D. 545, and ascribed by him, on the authority of the statements contained in Eusebius (HZ iv. 36), to Tatian * does actually represent the Iiatessaron, may now be regarded as proved.

t A comparison of the two documents clearly demonstrates that they are closely allied, and chat at least as regards the order, they are, with few exceptions, in remarkable agreement.

This agreement, unfortunately, does not extend to the text, for the copy of the Latin Harmony which Victor inserted (in place of the four Gospels) at the head of his edition of the New Testament is not so much a translation, whether of the original Syriac or of an intermediate Greek version, as a transference of the language of the original text into the language of the Latin Gospels as revised by Jerome.

In other words, the form exhibited by the Latin Harmony of the Codex Fuldensis is that of the Diatessaron, the text is that of the Vulgate. We cannot, however, follow Hemphill} in attributing the form of the Latin text to Victor or to the scribe working under his direction. Such a alec is incon- sistent with Victor’s own introductory remarks,§ which convey no hint of such a laborious task, and is directly excluded by an examination of the descriptive capitulation prefixed to the Harmony.

For though the latter frequently disagrees with the enumeration of the chapters as given in the text, and was clearly, therefore, composed before the Harmony had assumed its present form, it has been preserved unchanged by the copyist.

The fol- lowing instances of this disagreement, taken from the commencement of the Harmony, are the more interesting as they serve in a large measure to ex- * ‘Ex historia quoque ejus comperi, quod Tatianus, vir eru- ditissimus et orator illius temporis clarus, unum ex quatuor compaginaverit evangelium cui titulum diapente conyposuit’ (Codea Fuldensis, ed. KE. Ranke, 1868, p.1f.) The origin of the curious title ‘Diapente’ (d:% wévre) for ‘ Diatessaron’ has long perplexed the minds of scholars.

It is not found either in the original Greek of Eusebius or in the Latin (of Rufinus) and Syriac translations of that work. Further, it is expressly excluded by Victor's statement that the Harmony was compiled from the four Gospels (the numeral is also inserted in the trans- lation of Rufinus). Zahn (Forsch, i. 2f.) is probably right ia regarding it as a dapsus calami either of Victor or of his scribe. H. Wace, Eapositor, 1881, 2, p. 128f,; Zahn, Forsch. 1 pp. 1-5, 298 f.

t The Diatessaron of Tatian, p. xxiv, § Cf. Zahn, Forsch. i. p. 3f. DIATESSARON DIATESSARON 457 plain the opposition with which Victor’s identifica- tion of the Harmony was for so long received. The table of chapters commences: Prefatio I. In principio verbum, deus apud deum, per quem facta sunt omnia. From this heading we see that the Harmony commenced with Jn 1), and that Lk }*4, which now precedes it in the Harmony, formed no part of the original work.

In the same way we can explain the presence of the genealogies of the Lord, which, as we know, were omitted in Tatian’s Diatessaron. The capitulation runs: V. de generationem (sic!) vel nativitate Christi. Here the word generatio is clearly identical with na- tivitas, and does not refer to the genealogies (Mt }-16, Lk 3-88, Mt 117) but to Mt 18% (Christi autem generatio sic erat), which is given in the Codex at the beginning of ch. 5.

Thus we see that the Latin Harmony originally commenced with Jn 1', and did not contain the genealogies, the omission of which is so characteristic of Tatian’s work. The elimination of these later additions to the Latin Harmony undoubtedly removes two of the strongest objections that were urged against the identification of Victor’s discovery with the Diatessaron.

There still remains, however, the question of the lanyuage, since, in the opinion of many scholars, the Latin translation can have been made only through the medium of a Greek version, whether that of the original Diatessaron or of a translation made from theSyriac. But the evi- dence which we have examined affords no support, or rather is entirely opposed, to the theory of an original Greek Diatessaron, while the researches of Zahn (Forsch. i. 311f.)

have shown conclusively that the supposed need of a Greek intermediary translation lacks historical support. Thus a con- temporary of Victor, the African Junilius, who was Questor sacri palatu at Constantinople about A.D, 545-552, made a Latin translation (Jnstituta reguiaria divine legis) of an introduction to the Scriptures, composed by the Syrian Nestorian Paul, a pupil and teacher of the school of Nisibis, and sent it to Primasius, bishop of Adrumetum.

* Even at an earlier date Cassiodorus and the Roman bishop Agapetus (d. 536 A.D.) conceived the idea of founding a theological school at Rome on the model of those at Nisibis and Alexandria. Still more important is the testimony of Gen- nadius of Massilia, who wrote a continuation of Jerome’s de Viris Illustribus. In the first chapter of this work (written about A.D.

495) he discusses at some length the Homilies of Aphraates, whom he identified with Jacob of Nisibis, and explains Jerome’s silence with regard to this writer on the ground that his works had not been translated ; for the works of the few Syriac writers which are included in Jerome’s category were, by his own testimony, known to him only through Greek trans- lations.

Gennadius mentions further a Syriac thronicle of Jacob of Nisibis, the writings of two of Ephraem’s pupils, and also those of Isaac of Antioch. It would seem, therefore, that the ignor- ance of Syriac, which prevailed among Western writers at the time of Jerome, had largely dis- ered during the interval between the date of the latter and that of Victor of Capua.

ep teely as Zahn suggests, this change was chiefly brought about by the Syriac monks who settled in Sinai, Palestine, Egypt, and Constantinople, and there came into contact with Western scholars, Hence it is by no means improbable that some Latin scholar in the 5th or at the beginning of the 6th tent. should have compiled that Latin form of the Syriac Diatessaron which has been preserved to us in the Codex Fuldensis.

, It is obvious from what has been said above that the Codex Fuldensis can add nothing to our know- * Kihn, Theodor von Mopsuestia und Junilius Africanus. ledge of the text of the Diatessaron. It is, how- ever, an important witness to the general structure and arrangement of its Syriac original, though even in that respect it seems to have suffered from revision.

Its chief value for our purpose consists, as we shall see, in the fact that it supplies us with the means of controlling the far more trustworthy evidence of the Arabic version. (c) The Arabic Version.—This version was first published by A. Ciasca, one of the guild of scriptors at the Vatican Library, under the title: ‘Zhe Diates- saron which Tatian compiled from the Four Gospels (in Arabic), sew Tatiani Evangeliorum Harmonie Arabice.

Nune primum ex duplici codice edidit et translatione Latina, donavit P. Augustinus Ciasca, etc., Rome, 1888.’ Of the two MSS which form the basis of this edition, one (Cod. Vat. Arab. xiv.) had been brought from the East by Joseph Assemani as early as A.D. 1719, and had been definitely described by its discoverer as ‘Tatiani Diatessaron seu quatuor Evangelia in unum re- dacta’ (Bibl. Or. i. 619).

A statement to the same effect contained in the colophon* was also quoted by Stephen Assemani ; nevertheless the MS was left unnoticed, except by Zahn, until the publication of Ciasca’s De Tatiani Diatessaron Arabica Versione + in 1883. In this essay Ciasca gave a full description of the MS which, like Assemani, he assigned to the 12th century. He further defended the statement of the colophon, despite the fact that the Arabic Harmony com- menced with Mk 1!

(instead of Jn 11), and con- tained the genealogies according to Matthew and Luke. His conjecture that these additions were a later interpolation was substantially confirmed by a comparison with the second MS, which shortly afterwards came into his hands. This MS was presented to the Museo Borgiano in 1886 by the Copt, Galim dos Gali. It is probably to be assigned to the 14th century. It displays a less correct orthography than the Vatican MS, but presents the text in a more original form.

Both the intro- ductory notice and the colophon describe the work as the Diatessaron. It clearly begins with Jn 1, Mk 1!’ forming a sort of title, while the gene- alogies are not included in the text, but have been inserted before the colophon. For his Arabic text Ciasca has mostly followed the Vatican MS, but there are a number of passages in which he has adopted the readings of the later MS; the variants are in every case added in the footnotes.

The usefulness of the Latin translation is a good deal impaired by Ciasca’s attempt to adapt it to the style and character of the Clementine Vulgate ; apart from this fact, it also contains too many inaccuracies to be of much critical value. This translation has been followed by Dr. Hemphill in his English edition of the Diatessaron (1888), and forms the basis of another English translation, which has been compared throughout with the original Arabic, ublished by Dr.

Hamlyn Hill in Lhe Earliest Life of Christ. A more literal and entirely independent English rendering of the Arabic has also been published by the Rev. Hope W. Hogg.t An interesting statement as to the origin of the Arabic translation has been preserved both in the introductory notice and in the colophon§ of the In fine fol. 123 hxc a librariis adnotata reperies: Explicit auxilio Dei Sacrosanctum Evangelium quod ex quatuor Evan- geliis collegit Tatianus, quodque Diatessaron vulgo dicitur.

Et laus Deo’ (Mai, Script. Vet. Nova Coll. iv. 2. 14). t Pitra, Analecta Sacra, iv. 465-487. t Ante-Nicene Christian Library: Additional Volume. 1897. § These are given in full by Hjelt, op. cit. p. 63 n., together with a cliscussion of the difficulties presented at the commence. ment of the Harmony by the various readings of the two MSS. Hjelt adopts the view (cf. Zahn, Gesch. ii.

588) that the material common to both MSS is probably the remains of an original title which perhaps ran as follows: ‘The Gospel of Jesus Christ, 458 DIATESSARON Borgian MS, according to which it was made by the ‘excellent and learned priest’ Abul Faraj ‘Abdullah ibn at-Tayyib. he colophon adds further that the Syriac exemplar was written by ‘Isa (MS _ wrongly Gubasi) ibn ‘Ali al-Mutatabbib, the pupil of Honain ibn-Ishak.

By means of these notices, the correctness of which we have no reason to doubt, we are able to fix the date both of the Arabic translation and of its Syriac original. Ibn at-Tayyib was a well-known writer of the 1lth cent. (d. 1043), who commented on the writings of Aristotle, Galen, and Hippocrates, translated the Gospels of SS. Matthew and John from Syriac into Arabic, and also wrote an Arabic commentary on the Gospels.

The scribe who was responsible for the Syriac exemplar is most prob- ably none other than the famous lexicographer Jesus bar- Ali (see above, § ii. 6), who flourished in the latter half of the 9th cent. (his teacher Honain died in A.D. 873). Thus the date of the Syriac MS used by the translator of the Borgian MS is the latter half of the 9th cent., and that of the Arabic translation itself the first half of the 11th century.

It is disappointing to find that, though the Arabic translation has preserved the outward form and characteristics of Tatian’s Harmony, and in that respect is a most important witness to the order and arrangement of the Diatessaron, the text which it exhibits has throughout been accommodated to that of the Peshitta, and is therefore of no value for restoring the original Syriac version.

The data at our command are perhaps insufficient for determining whether this accommodation had already been effected in the Syriac exemplar which was used by the Arabic translator, or is to be assigned to the latter him- self. Zahn * maintains that the translator entirely recast the Syriac in accordance with the existing Arabic versions of the Gospels, and that the his- tory of the Arabic Harmony thus presents a close analogy to that of the Latin (Codex Fuldensis).

In support of this view he urges that, when once the Diatessaron had been banished from the public worship of the Church, it would soon cease to have an interest for any but the learned, and the latter would have no motive in introducing any altera- tions.

As evidence of such learned interest in the Diatessaron he points to the marginal references, attached to both the Arabic MSS and presupposed by the introductory notice in the Borgian MS, by which the source of each passage was indicated : these, presumably, already existed in the Syriac copy of the 9th cent., since the writer of the notice is silent on the subject ; and they naturally formed no part of the original Diatessaron.

It seems, however, more probable that the later type of text preserved in the Arabic version reflects the result of a process of revision by which the Syriac Dia- tessaron had been gradually brought nearer and nearer to the authoritative text of the Peshitta. For the Arabic, unlike the Latin Harmony, shows evident signs of its Syriac origin; and this fact alone makes it difficult to imagine that its text was entirely recast in a similar manner to that of the Codex Fuldensis.

Moreover, it is not only, as Zahn admits, a faithful witness to the order and arrangement of the Diatessaron as a whole, but also reproduces in many cases the finer de- tails which determine the internal composition of the individual sections. These arguments would naturally receive strong confirmation if we could viz. the Diatessaron.’ In the Syriac exemplar, used by the Arabic translator, the original title was probably obscured by the substitution of ‘the Son of God’ (Mk 11) for ‘viz.

the Diatessaron,’ while later still the insertion of another marginal gloss from Mk 11 (ex Marco dic: Initiwm) caused even further tonfusion. * Gesch. il, p. 680f. DIATESSARON follow Hjelt in his theory—which is undoubted! supported by the fact of this llth cent. Arabic translation—that the Diatessaron was retained in church use by tie Nestorians down to the Middle Ages.

For, had such been the case, the accepted text of the Peshitta could not have failed in course of time to exercise a marked influence on the older text. Hjelt’s further suggestion, that the refer- ence in the colophon of the Borgian MS to the work of ‘Isa ibn ‘Ali al-Mutatabbib, z.e. Jesus bar- ‘Ali, possibly contains a hint as to the authorship of that final revision of the Syriac Diatessaron, which is embodied in the Arabic version, can only be described as an ingenious conjecture.

But, whatever its genesis, it is clear that the Arabic translation possesses far greater value for restoring the original work of Tatian, more especially in é relation to its internal structure, than the Latin | Harmony of Victor. Moreover, since a comparison | with the quotations of Ephraem and Aphraates | attests its trustworthiness in those parts for which the Syriac writers are available, we should be justi- fied in admitting its evidence, even where the | latter are silent.

In such cases, however, some {| doubt would naturally exist, and it is therefore in this connexion that the importance of the sister Latin version is most apparent. For if, as can be shown, the Codex Fuldensis is in entire agreement with the Arabic version in fassages which are otherwise attested by Syriac evidence, we may infer that their agreement elsewhere is also due to the fact that both have preserved the original form of the Diatessaron.

This is the more certain, as the Arabic translation of the 11th cent. cannot be dependent on the Latin version of Victor in the 6th cent., while the theory of a common source — for both is excluded by a comparison of their variations from the original.* The testimony of the Arabic Harmony has natur- ally enabled Zahn to supplement his former work to a very considerable extent, while at the same time contirming in the most striking manner his reconstruction of the Diatessaron.

one respect only is a correction necessary in the chronological scheme (see above), viz. the omission of the first Passover (Jn 2!%) +: otherwise the changes involved are contined to a few cases in which the order of the individual sections varies in the Arabic version.t Of these the majority are accepted by Zahn when the order of the Arabic Harmony is confirmed by that of the Codex Fuldensis.§ | iv. RELATION OF THE DIATESSARON TO THE — OLD SyriAc.

—The term ‘ Old Syriac’ is here used to denote that early form of the Syriac Gospels which existed alongside of the Diatessaron down to the beginning of the 5th cent., but which was then revised in conformity with the Greek by Rabbila, with the result that the new version, the | Peshitta, speedily became the accepted ecclesias- tical text.|| We have already seen (§ ii.

6) that this version, with its separate Gospels, was fre- quently described as Hvangelion da-Mépharréshé (Gospel of the Separated [ones]), to distinguish it — from the Diatessaron or Ev. da-Méhalléte (Gospel — of the Mixed). Of this version we now possess two codices, viz. that called the Curetonian Syriac (Se), a Nitrian MS of about the middle of the 5th cent., which was published by Cureton in 1858, and the © Sinai palimpsest (Ss), dating from the beginning of the 5th cent.

, which was edited by the ata Prof. | Bensly, J. Rendel Harris, and F. C. Burkitt in | * Zahn, Gesch. ii. 535 f. + Horsch. i. 250. ¢ Hamlyn Hill, Harliest Life of Christ, App. ix. § In tio cases at least it would seem that Zahn has insisted — too rigidly on the absolute agreement of the Arabic and Latin versions ; see Hjelt, op. cit. p. 70f. | For a fuller account, see an article by the present writer in The Church Quarterly Review, April 1903, pp. 143-171. DIATESSARON 1893.

The two codices agree so closely—though they also display some important textual varia- tions—that they may be suitably described as two recensions of one and the same translation. Of the two texts Ss is decidedly the purer, and may be regarded as a faithful witness to the text of the 2nd cent. ; Sc, on the other hand, probably represents a later recension of the 3rd century.

The question that naturally arises from a con- sideration of these two MSS of the Old Syriac is that of their relation to the Harmony of Tatian, which also involves the further question of the relation of the Old Syriac to the Diatessaron. As the result of his exhaustive investigations in connexion with the Diatessaron, Zahn concluded that Tatian had based his Harmony on the text of the Old Syriac, which must have been made about the middle of the 2nd cent.

; but had also made use of a Greek text of a similar type to that of the Codex Beze and of the oldest Itala MSS. This conclusion was, in reality, based on two miscon- ceptions. For from a comparison of the text of Se with that of the Diatessaron (T), Zahn argued that the former had influenced, and was therefore prior to, the latter ; but, since he also identified Sc with the Old Syriac, the two errors did not affect the correctness of his main contention, that the Old Syriac was prior toT.

The real relation of T to Se was first established by F. Baethgen,* who, as Zahn admits, has shown conclusively that T clearly influenced, and was therefore prior to, Sc.

‘The arguments adduced by Baethgen are, briefly: (1) the presence of an extraordinarily large num- ber of harmonistic readings in Sc, which must be derived from the Harmony; (2) the numerous eases of abridgment which are to be explained in a similar way ; (3) the specifically ‘ Alexandrine’ readings, which point to a later date than the 2nd cent. ; (4) the great freedom of rendering, which frequently lapses into | he amr and may be recognized as due to Tatian; (5) the dogmatic character of Se.

(1) Even if we exclude those cases in which a harmonistic reading is attested by either a single Greek MS or one of the old translations, or a Patristic quotation, Baethgen has shown that there still remain some 150 cases in which Sc stands alone, except for the frequent agreement of T.

This is the more remarkable, since such mixed texts are necessarily confined to passages for which there is a parallel account; while Zahn’s restoration of the text of T, of which Baethgen makes use, is of a very fragmentary nature.

(2) It is of the very essence of a Harmony such as that comyiled by Tatian that it should omit not only those incidents and sayings which are repeated by one or more of the Evangelists, but also many of the small clauses and words which, without affecting the sense, serve to characterize the narrative.

But, though these omissions might naturally be expected to be restored in a transla- tion of the four separate Gospels, nearly a half of the (roughly) 270 readings which are peculiar to Se (excluding the harmonistic readings) belong to this category of abridgment or omission : hence we may infer that Sc has made use of the shorter text of T wherever such a course does not affect either the meaning or the context.

(3) As the result of an exhaustive examination, Baethgen pronounces the text of Sc to be of a decidedly ‘Western’ type, as is shown by its affinity with D, in part also with % and some minuscules (especially 69), and, lastly, with the ‘ African’ text of the Old Latin. But Sc also dis- plays traces of ‘ Alexandrine’ influence, which seem3, in fact, to be due to Origen. Hence the translation must certainly be assigned to a later * Evangelienfragmente, Leipzig, 1885.

DIATESSARON 459 —— date than that of the Diatessaron, and eannot be earlier than the 3rd century. (4) In his discussion on the method of the trans- lator of Sc, Baethgen (pp. 13-23) classifies the various expedients adopted in order to reproduce the sense of the Greek text.

But similar examples occur with even greater frequency in T; and though naturally some of the latter’s more striking trans- lations are no longer to be found in Sc, yet the two so often agree that the dependence of the one on the other can hardly be denied (cf. Baethgen, p. 87). In view, therefore, of the priority of T which has been already established, it is highly probable that in this respect also Sc is dependent on the Diatessaron.

(5) It is, however, especially with regard to its ‘dogmatic’ character that Sc betrays the influence of Tatian. Clear traces of the latter’s anti-Jewish or universalistic views appear to be preserved in Mt 17 (mundum for rdv dadv airod), Jn 6” (rdvras for rods dw&dexa), 7° (the omission of judy after 6 véuos).* The omission of the possessive pronoun with ‘Father’ in Mt 6% 10%, Lk 2%, Jn 6% 147! is also, according to Baethgen, to be ascribed to a similar point of view.

Further, Tatian’s Encratite views seem to be reflected in those renderings of Se which are clearly due to a desire to establish the perpetual virginity of the mother of our Lord. This is especially noticeable in Mt 138%, The above arguments do not all possess the same evidential value, but the cumulative evidence which they supply is more than sufficient to justify Baethgen’s conclusion as to the relative dates of Se and T.

It by no means follows, however, that the evidence which was conclusive in the case of Se necessarily applies to its archetype, viz. the Old Syriac: hence Baethgen’s further conclusion, that the Diatessaron was the earliest form of the Syriac Gospels, cannot be accepted without addi- tional proof. In this connexion the discovery of the Sinai palimpsest is of the greatest importance.

For if it can be shown that this codex agrees with Se in exhibiting the same traces of T’s influence, we can only infer that this agreement goes back to their common source, 2.e. the Old Syriac, and that the latter is therefore posterior to Tatian’s Harmony. Thatsuch is actually the case is main- tained, among others, by Zahn and Nestle; but, in view of the arguments brought forward by Burkitt t and Hjelt, + the contrary opinion seems to be the more probable.

For a comparison of the text of Ss with that of Se shows that those peculiar features of the text which clearly pointed, in the case of the latter, to the influence of T, are by no means so strongly marked, if not entirely wanting, in the former. This divergence of text is especi- ally noticeable in respect to the harmonistic and ‘dogmatic’ readings which undoubtedly form the main support of Baethgen’s arguments as to the relation of Sc to T.

With regard to the former, Burkitt notes that sixteen, § or more than one- third, of the forty-three examples (quoted by Baethgen), where Se stands alone (or with the Diatessaron), are not shared in Ss. Hence it is clear that, though Ss undoubtedly contains a large number of mixed readings, these by no means form such a distinctive feature of its text as they do in the case of Sc, and need not therefore be ascribed to the same cause.

Their presence is more than sufficiently accounted for, whether we assign it to the well-known tendency of scribes to harmonize parallel passages unconsciously, or to the actual influence of the Diatessaron, which, as * To these we may add Mt 10% (the omission 3f rot "Iepafa). +t Guardian, Oct. 31, 1894. t Op. cit. p. 107 f. § Hjelt, op. cit. P 108 n.

, corrects this to fifteen, and points out that in four of these passages Ss is defective DIATESSARON we have seen, was the accepted text of the Syriac Church during the 38rd and 4th centuries. In re- spect of their ‘dogmatic’ character, the difference between the two codices is even more strongly marked, since Ss exhibits no traces of Tatian’s influence. A striking illustration of this fact is furnished by an examination of the two texts in the crucial passage, Mt 12; y.

°Ss her husband— Sc omits; v.” Ss thy wife—Se thy betrothed; v.™ Ss and thou shalt call —Se (his name) shall be called; v.A Ss his wife—Se Mary; v.> Ss and she bore him a son, and HE called his name Jesus— Se and he lived with her purely, until she brought forth the son, and SHE called his name Jesus.

(In this rendering of ovk éylywoxer airjv [omitted by Ss] Se follows Tatian): But it is obvious that the lack of agreement between Ss and Sc on these points, which formed the basis of Baethgen’s argument for the priority of T to Sc, materially weakens the case for the similar relation between I’ and Ss, since the other points of contact are not decisive in themselves and may have arisen equally well from the dependence of T on Ss.

Moreover, we are not without positive proofs of the priority of the Old Syriac (as represented by Ss). The most important of these is the omission in Ss of the last twelve verses of St. Mark, which, as Burkitt has pointed out, can only be a Greek variant, and must represent, therefore, the origi- nal form of the Old Syriac. But we know that Tatian included Mk 16°-?° in his Harmony : hence its omission by Ss clearly points to the priority of the latter.

Other omissions which point to the same conclusion are those of Mk 15%, Lk 22%. #4 (the Bloody Sweat) and 23%4 (the Prayer on the Cross), all of which are given in the Diatessaron.

Similar evidence is also afforded by the curious mistranslations of Ss which occur in Mt 2216, Mk 778 105°, Lk 4% In Mt 2236 rods uabnras atrod (Se Tod "Inood) for atréy (Se trav Papicalwy); in Mk 776 Ss renders from the border of Tyre in Phenicia, clearly identifying the Zupo in Zupodowlkicoa with Tyre (ms); in Mk 10° dmofaddéy is misread as drohaBdv ; in Lk 4” its rendering presupposes dare KaTaxpeudoat avrév instead of dere Kataxpyuricar airév.

* 'To these we should probably add Mt 5+, Lk 2%, where Aphraates renders rapaxdnOijoovra and mapdxAno.s according to the ordinary meaning of the verb, viz. to pray. Aphraates, as we know, made use of the separated Gospels, and has prob- ably done so in the present case ; for Tatian, who was well acquainted with Greek, would hardly have fallen into such an error.

Ss, it is true, translates rapdxAyovs correctly in Lk 674, but makes a similar misrendering in Lk 2%; possibly, as Hjelt suggests, Ss does not represent the original oe of the Old Syriac either in Lk 6% or in t 54.

But, apart from these omissions and mistransla- tions, which clearly attest the independence, and therefore the priority, of Ss, a comparison of the two texts shows no less clearly that ‘in those cases in which they differ from one another, the former, as a rule, oer a form of text which appears to be the older and more original’ (Hjelt, p- 155), while in many cases the rendering of T seems to be directly based on that of Ss, or of one similar to it.

Space forbids a complete discussion of all the divergences of the two texts, but the following instance will afford sufficient illustration of this statement. We have already seen from a comparison of the ~ texts of Ss and Sc in Mt 1'- that the former gives a plain unbiassed rendering of the passage, which is in marked contrast to that of Sc. But the * Tatian apparently inferred that our Lord was actually cast over the cliff, but was miraculously preserved from harm (Moesinger, pp. 180f.

, 212; Hamlyn Hill, Dissertation, p. 93). DIATESSARON variants of Sc in this passage reflect, to a large extent, that desire to emphasize the virginity of our Lord’s mother, which is even more apparent in the Diatessaron—e.g. in the rendering (v.18) ante. quam data est viro for ply 4 cuvedOeiv, the omission of airfs, v.12; the rendering in sanctitate habitabat cum ea for ox éylywoxev, v.”, and the transposition of v.

% and >, Possibly the clearest indication of the priority of Ss to T is given by v.”, in which the rendering of T and (Sc) obviously presupposes that of Ss. Other passages which point to T’s immedi- ate dependence on Ss, or on a text similar to it, are Mt 10% 14%, Lk 133, Jn 334» 358, In Mt 10% Ss renders civitatem by {2;2 ; while in Lk 10”, which is here combined with the Matthew passage, themore usual |A14,S0 is found.

This difference of render- ing, however, is preserved by T, who uses the latter word for civitatem in the introductory sen- tence which he has incorporated from Lk 10, but renders the mé\es of Mat. by 1o;2. In Mt 14% he translates éxéracev by two synonyms, the one (QM) being taken from Matthew, and the other (ad.») from the parallel Mk 6°.

The dependence of T is no less clear in Lk 12%, where Ss renders é:6r: elanxotaOn % dénols cov quite freely by ‘ For, behold, God has hearkened to the voice of thy prayer’ ; for, though he has restored the passive construction of the Greek, he has also retained the addition of Ss in the form ‘ exaudita est deprecatio tua ante Dewm.

’ Jn 3% offers a number of interesting Syriac vari- ants ;* but of these the rendering of T (od yap é« pérpov dldwor [7d mvedua 6 rarnp|t 7H vig abrod ayarg 5¢ avrév) seems-to be based on that of Ss (ob yap éx pérpov dldwoww 6 Oeds 6 warip, dyamrg dé Tov vldw avrod), the order of which he has slightly varied.

Other passages in which Ss has preserved the more original reading are: Mt 4° (for Bade ceavrov kdrw), 5% (the order of the second and third Beatitudes), 10° (for apis 7a mpéBara ra amrokwhéra Tov olkod "Iopajd), 16 17% (for dua rhy_dmioriay iuav), 18% 2178-81 (Ss here presents a ‘ Western’ text), Mk 74 9%, Lk 2 (the rendering of evdoxia), 6*° 1721 (evrds tudv), Jn 2 (omission of rodroy after rov vaov), 4° (omission of 4 yuv7), 6° 8 111. 5 (omis- sion of cal 7 fw7).

Fe In consideration of these facts we are justified in ignoring a large number of those passages which were formerly brought forward by Zahn in support of his theory of the afte es of T to Ss. For, though the majority of these attest the close affinity of the two texts, they do not of themselves furnish any evidence as to the ~ origin of this affinity, i.e. as to the priority of T or Ss.

Zahn’s view, however, finds its main support in those traces of harmonistic poets which he discovered in Ss, which he natural y ascribed to the influence of T. Briefly, Zahn's theory was as follows: Ss and Se are un- doubtedly closely related, and may be described as two recensions of a single version.

Their variations, in which Se, as a rule, agrees with the Peshitta, are for the most part of a grammatical, lexical, and stylistic character; in others the agreement of Sc and P against Ss can be explained only by the supposition that the free, or less accurate, translation of Ss was altered in Se and P, and brought into closer conformity with the Greek text. But Se has also been shown to have much in common with T: hence it was natural to expect that Ss and T should be closely allied.

The conclusion arrived at by Zahn, after an examina- *See Burkitt, S. Ephraem’s Quotations from the Gospel, p 60f.; Zahn, Forsch. i. p. 129. + So Aphraates, 123 ; Moesinger, 105, omits these words, t Theol. Littbi, 1895, Nos. 1, 2, 3. DIATESSARON tion of the text of Ss, was that the latter was even more closely allied to T than Sc.

The instances cited by Zahn in proof of his contention for the ay of T have been carefully examined by jelt,* who has shown that in the majority of cases the alleged dependence of Ss on T rests on insufficient evidence. A few traces of harmoniza- tion, it is true, are to be found in Ss, but these are probably to be ascribed to later interpolation.

_Our examination, therefore, of the relation of the Diatessaron to the two codices of the Old Syriac version leads to the following conclusions as to the history of Tatian’s Harmony. The two texts are closely related to each other, but a com- “ane of the two shows clearly that the Old yriac is the earlier version : hence the latter must have been in existence before A.D. 172.

At this date Tatian compiled his Diatessaron, or Harmony of the four Gospels, in Syriac from the older version, which it quickly superseded, revising it with the help of a ‘ Western’ copy of the Greek text, and introducing a number of arbitrary changes in accordance with his theological views. Down to the end of the 4th cent.

the Diatessaron was universally accepted by the Syriac Church, the extent of its influence Maine reflected in the later recension of the Old Syriac version repre- sented by Sc. In A.D. 411 the Old Syriac version was revised, in conformity with the current Greek text, under the auspices of Rabbila, who forcibly removed the Diatessaron from. church use in order to make room for his new version, viz. the Peshitta.

As the result of Rabbila’s action, the Diatessaron practically disappeared from the knowledge of the Syrian Church, the references to it in later writers being mainly connected with the better known commentary of St. Ephraem. All interest, however, in Tatian’s work did not cease with its banishment as a service-book, for about the begin- ning of the 6th cent. it was translated, or rather transferred, into Latin byanunknownauthor.

This translation, as preserved by Victor of Capua in the Codex Fuldensis, probably formed the basis of the German version made c. 820-830 A.D., and this again was utilized by the author of the Old Saxon poem known as Heliand. The last stage of the history of the Diatessaron was not reached until the llth cent., when it was once more rescued from: obscurity, this time in the form of an Arabic translation, —~ LirERATURE.—(1) General :—O. A. Credner, Beitrdge zur Ein- leitieng in die bibl.

Schrijgten, 1832, p. 437 ff., Gesch. der neutest. Kunons (herausg. von G. Volkmar), 1860, p. 17 ff.; H. A. Daniel, Tatianus der Apologet, 1837; C. A. Semisch, Tatiani Diatessaron, 1856; Th. Zahn, Forschungen zur Geschichte des neutest. Kanons, i. (1881, ‘Tatian’s Diatessaron’), ii. 286 ff. (1883), iv. 225-246 (1891, ‘Der Text des von A. Ciasca heraus- gegebenen arabischen Diatessarons von Dr. Ernst Sellin’), Geschichte des Kanons (1888), i. 369 ff. (cf.

Harnack, Das Neue Testament wm das Jahr 200, p. 90ff.), ii. 530-556, art. ‘ Zur Geschichte von Tatian’s Diatessaron im Abendland’ in Neue Kirehliche Zeitschrift, 1894, pp. 85-120, art. ‘ Evangelien- harmonie’ in PRE® y. 653 ff. (1898); A. Hjelt, Die altsyrische Evangelieniibersetzung und Tatian’s Diatessaron, Leipzig, 1901 (=Zahn, Forschungen, vii. 1, 1993); Fr. Baethgen, EHvan- gelienfragmente, Leipzig, 1885; J.

Rendel Harris, 7'he Diates- saron of Tatian, Cambridge, 1890, Contemporary Review, Aug. 1895, pp. 271-278 (a reply to W. R. Cassels’ article in Nineteenth Century, April 1895, p. 665 ff.), Fragments of the Commentary wv Ephraem Syrus upon the Diatessaron, London, 1895 ; Isaac H. Hall and Professor Gottheil, Journal of Biblical Literature, xi. 11, pp. 153-155 (1891), xii. 1, pp. 68-71 (1892) ; Goussen, Studia Theologica, Fasc. 1 (1895); J. B. Lightfoot, Hssays on Super- natural Religion, pp.

272-288; H. Wace, Expositor, 1881, 1882; Westcott, History of the Canon’, pp. 325-333; Fuller, art. in Dictionary of Christian Biography, ii. p. 140 ff.; Holzhey, Der mneuentdeckte Codex Sinaiticus untersucht, Miinchen, 1896; Bewer, The History of the NT’ Canon in the Syrian Church, Chicago, 1900; F. C. Burkitt, S. Ephraem’s Quotations from the Gospel (=Texts and Studies, vii. 2), Cam- bridge, 1901 ; art. in Church Quarterly Review, April 1903, pp. 43-171. » Op. cit. pp. 113-180; see also Ch.

Quarterly, April 1903, pp. 167-170, JOSEPHUS 46] (2) Texts.—A. Ciasca, ‘De Tatiani Diatessaron arabica ver. sione’ (in S. P. Pitra’s Analecta sacra spicilegio Solesmenst parata, Paris, iv. pp. xxviiiff., 466ff., 1883), Tatiani evan- geliorum harmonie arabice, Rome, 1888; Hemphill, The Diatessaron of Tatian, Dublin and London, 1888; J. Hamlyn Hill, The Earliest Life of Christ, being the Diatessaron of Tatian, Edinburgh, 1894, A Dissertation on the Gospel Com- mentary of S.

Hphraem the Syrian, Edinburgh, 1896; Hope W. Hogg, ‘The Diatessaron of Tatian’ in Ante-Nicene Christian Library, Add, vol., Edinburgh, 1897; G. Moesinger, Hvangelit Concordantis Eaxpositio ; in Latinum translata, etc., Venice, 1876 (Professor A. Roninson’s more accurate English translation of the Gospel quotations occurring in the Commentary is given in Appendix x. of Hill’s Earliest Life, and in his Dissertation, pp. 75-119); E.

Ranke, Codex Fuldensis: Novum Testamentum latine interprete Hieronymo ea manuscripto Victoris Capuani, Marburg and Leipzig, 1868. JOHN F. STENNING.