The Arabic Diatessaron and Codex Bezae: the Case of Mark's Long Ending
Mina Monier — 06.01.2020▶
This E-talk comes to you as part of the SNSF-funded MARK16 project, hosted in the Swiss Institute of Bioinformatics, Lausanne, Switzerland. In analysing evidence behind the different endings of the Gospel of Mark, today’s e-talk will shed light on the presentation of this ending in the Arabic manuscripts of the Diatessaron, in comparison with other surviving translations of this important work. Written probably in the third quarter of the second century, by Tatian in Rome, the Diatessaron is a synthesis of the four canonical Gospels of the New Testament: Matthew, Mark, Luke and John. The title “Διατεσσάρων” which means via-four is not really the name of the work but a description of the Gospel’s composite nature. For instance, in his Letter to Carpianus (PG 22:1276-1277), Eusebius of Caesarea tells us that Ammonius of Alexandria (2nd-3rd century) produced a harmony that influenced his own gospel canon tables, which he refers to as τὸ διὰ τεσσάρων εὐαγγέλιον.
So, the nature of the Diatessaron as a mixture of Gospels is not itself an act of innovation, for it is a synthesis inasmuch as the Gospels it is made of. However, the work became at a time the question of canon was subject to the heated debates in Rome, when Marcion introduced the first known New Testament canon, which was made of one Gospel and a selection of Pauline letters. The Diatessaron was widely circulated around the world. In the west, it may have circulated in an Old Latin version, reflected in a mid-6th-century Latin Diatessaron, found in the Codex Fuldensis. Discovered in the Fulda monastic library in Germany, it was commissioned by Bishop Victor of Capua, who provided a preface, thus also known as the Victor Codex.Meanwhile, in the East, it was popular. In Syriac, the Diatessaron – the Gospel of the Mixed – was the text used for the lectionary in numerous churches. It was quoted in the works of Aphrahat and was so entrenched in Syriac-speaking communities that in the 4th century the celebrated Syriac scholar Ephrem wrote an extensive commentary on it (extant in Syriac in Chester Beatty 709 and in Armenian). With the rise of the Peshitta, however, it was pushed aside, and actively destroyed. In the 5th century, Theodoret of Cyrus noted that the Diatessaron was widely circulated in his diocese, and he had to collect some 200 copies (Theodoret, Haer. 1.20).
The earliest piece is also found in the east: a Greek fragment discovered in Dura Europos and identified and published by Carl Kraeling. The fragment has just 14 lines, but dates before 256 CE (the year of the destruction of the site).However, the bulk of Diatessaron manuscripts is found in Arabic language: 7 complete codices and a fragmentary one. This is more than all other translations combined. The translation was done by Abū ’l Faraj Ibn al-Ṭayyib, from Syriac in the early 11th century. The Arabic Diatessaron appears to derive directly from a reasonably early form of the largely lost Syriac text. Unfortunately the Arabic Diatessaron was not considered worthy of investigation in terms of constructing the original Diatessaron. This depreciation is based on consulting dated and inaccurate translations. Since the publication of Ciasca’s Latin translation in 1888, his edition became the standard reference for subsequent studies. Even later translations in other European languages were either produced directly from his Latin translation or influenced by his textual choices, inheriting its deficiencies. Accessing the only two surviving manuscripts of the Arabic Diatessaron at his time (A and B), Ciasca produced his translation with a clear preference to the readings of MS A, despite the fact that the exemplar of MS A had a clear tendency to conform the text with canonical readings. This includes, for example, adding the genealogies of Jesus which are specifically known to have been absent in the original Diatessaron (Theodoret, Haer. 1.20). What can the Diatessaron’s Arabic manuscripts tell us about Mark 16? Let’s have a look at a selection of examples.
Mark 16:1-2: The final chapter of Mark starts with the women’s journey to the tomb in order to anoint the body of Jesus. The women arrive when “the sun had risen.” This is also what we find in the Latin and Dutch Diatessaron editions (Fs 174.2 and Lg 234.6). But the Arabic edition makes no mention of the sun, the scene “was still dark.” The Arabic edition, as Joan E. Taylor of King’s College London noticed, agrees with the scene as depicted on the wall of the Dura baptistery, where the earliest and only surviving Greek fragment of the Diatessaron was discovered [Joan E. Taylor, ‘What Did Mary Magdalene Look Like? Images from the West, the East, Dura and Judaea,’ in Kelly Olson and Alica Batten, Dress and Religion in Mediterranean Antiquity (London: Bloomsbury T&T Clark, 2020)]. Further, while the Latin and Dutch editions have two to three women visiting the tomb, the Arabic has many, which also agrees with the painting.
This is the first verse attested in full agreement between the Arabic, Latin and Dutch. We find that they all do not have Mark 16:4a (καὶ ἀναβλέψασαι θεωροῦσιν ὅτι ἀποκεκύλισται ὁ λίθος - When they looked up, they saw that the stone, had already been rolled back). This construction is found in Codex Bezae (D), which is an important remark. The women’s report to the disciples comes in a verse composed of Luke 24:9 and Mark 16:10, in both Arabic and Latin editions. However, a closer look into the Arabic edition shows some interesting features. As you can see in this table, the Arabic Diatessaron omits “from the tomb.” The only witnesses that lack this reference are codex Bezae and the Vetus Latina. The other case is how ‘those who were with him’ of Mark 16:10 is written in Arabic. The majority reading ‘τοῖς μετ᾽ αὐτοῦ’ is literally ‘to those with him.’ This is what we also find in the Fs: ‘qui cum eo fuerant,’ which is the standard reading found in Latin manuscripts, and Peshitta (ܠܗܢܘܢ ܕܥܡܗ). In Ar, we have a dative demonstrative pronoun that precedes the article: ‘to those لأوليك who الذين were with him.’ This exact structure is found in Codex Bezae only: ‘αὐτοῖς τοῖς μετ᾽ αὐτοῦ.’Mark 16.15 is a good example of how dated translations could lead to wrong conclusions. It is also of special importance since it is the only verse in the Long Ending that is cited by Ephrem the Syrian in his commentary on the Diatessaron. The verse comes in harmonisation with Matt 28.19. The Fuldensis reads: ‘Going into the whole world, preach the good news to every creature. Teach all the nations, baptising them in the name of the Father and of the Son and of the Holy Spirit.’ The part in italic is Mark 16.15b, while the second part is Matt 28.19. Fs follows the Markan majority text faithfully, using the participle euntes for ποροθέντες, and in the Matthean part it also follows the participle βαπτίζοντες by using baptizantes. Ephrem’s edition reads: ‘Go to the whole world and baptise in the name of the Father, the Son and the Spirit.’ Apparently, Ephrem’s text is shorter but it is important to observe that, unlike Fs and the majority of readings, it has the Markan ‘go’ (ܙܠܘ), and the Matthean ‘baptise’ (ܘܐܥܡܕܘ) in imperative form. This is the case with Arabic, which reads: ‘Go now into all the world, and preach my gospel to every creature, and teach all the nations, and baptise them in the name of the Father and of the Son and of the Holy Spirit.’ The verbs امضوا (go) and أعمدوهم (baptise them) are both in the imperative form. But where did the Syriac and Arabic readings come from? The two imperative verbs are also found in the Syriac Peshitta. However, in Ar the imperative ‘go’ is followed by الآن (now). Both Ciasca and Marmardji translated it: ‘go therefore,’ in order to conform the Arabic with the Peshitta of Matthew 28:19, which reads: ܙܠܘ ܗܟܝܠ (go therefore). This wrong translation of a clear word like ‘now’ shows their tendency to vulgatise the odd readings in Arabic. While the imperative verbامضوا (go) agrees with Ephrem’s ܙܠܘ, the whole امضوا الآن (go now) is in verbatim agreement with Matt 28.19 only in D and Vetus Latina: ite nunc.Right, what is to make of this? Well, these samples offer some significant insights. Of course, as I said earlier, for a verse-by-verse analysis, an upcoming article will be offering that, but the samples were selected to provide us with the overall results of this analysis. The result before us shows close affinity between the Arabic Diatessaron reading and a rather unexpected partner: old Latin tradition (Vetus Latina) and particularly Codex Bezae.
Here I would like to quote Petersen’s exposition on his third criterion of authenticity: “Geographic diversity is also required. This guards against ‘local texts’ in either the East or the West. Given the wide influence of the Vetus Latina and Vulgate in the West, and the Syriac versions in the East, it is possible that these ‘local texts’ might have influenced vernacular harmonies. […] However, it is unlikely that a variant found in the Vetus Latina should have crept into the Arabic harmony […] unless the cross–fertilization took place very early–possibly via the Diatessaron” [Petersen, Tatian’s Diatessaron, 374].
If we follow Petersen’s sensible point, we should consequently think that the Arabic Diatessaron preserves for us Diatessaronic readings that should be taken seriously, for its Syriac antegraph must have been close enough to the stage of what Petersen called “the cross fertilization” by the Diatessaron itself. Putting this in the context of Tatian’s historical profile, one would wonder how the Syriac behind the Arabic translation became so close to the Bezan text, and old Latin manuscripts unless this is explained by the fact that Tatian, an Assyrian by birth, produced this text in Rome.
This could also tell us something important about the text of Mark’s ending as it stands in its earliest comprehensive citation known to us. In the light of this analysis, we can see that parts of Mark 16 were possibly added by later copyists of the Diatessaron in order to leave no canonical text behind. But a core of specific verses, particularly but not exclusively verses 14-18, is not only attested across the Diatessaron translations in the east and west, but may have also been handed down to Tatian in a form known to us from Bezan and Arabic texts. Was this the earliest form of the Long Ending? Was this the long ending itself?
This is Mina Monier, from the Swiss Institute of Bioinformatics, Lausanne – Switzerland