Boismard on the Magdalene Gospel

M-E Boismard, LE DIATESSARON: DE TATIEN A JUSTIN, Gabalda, Paris, 1992.

[Last edited in May 2002. This was posted originally to various discussion groups in July 2000.]

Dear friends,

I have now prepared a short synopsis of Boismard's _Le Diatessaron: de Tatien a Justin_. I have reread this book recently, and was again impressed with some of his detailed arguments. Because of Boismard's rather startling and controversial conclusions, and because of their very significant implications for the study of early Christian texts, I thought it was a good idea to provide a convenient outline of Boismard's arguments in English. I have already provided a preliminary overview of this subject before. This and quite a few other articles about the Magdalene Gospel can be found at this link.

Now, I begin by presenting a brief overview of Boismard's book with my own comments, and also referring from time to time to the critique of Boismard's views as supplied by William Petersen. Petersen provided a review of Boismard's book in his TATIAN'S DIATESSARON, Brill, 1994: 348-356. This is the only English language review of Boismard's book that I'm aware of, and because of this, I'm also planning to provide a critical review of it some time later.

In the following, I will go along with the assumption, made by both Boismard and Petersen, that Tatian wrote the Diatessaron. Although, for the reasons that I've already provided, I think that this was extremely unlikely in the real life, this matter shouldn't affect so much the basic arguments in the following review. So, for the purposes of the following review, it's enough to assume that "Tatian's Diatessaron" will mean a later version of the Diatessaron, as opposed to its earliest version.


Essentially, Boismard's book is a detailed study of the Magdalene Gospel (Boismard refers to it as "l'harmonie Pepys", or "l'harmonie P"). Boismard thinks it is very important, because it seems to be a reasonably good witness to Justin's Harmony (JH).

When boiled down to basics, what Boismard is proposing in his book is actually quite simple. He begins by arguing that there was another, earlier gospel harmony before Tatian's, and this was the harmony that Justin Martyr used. (And Petersen actually seems to agree with him on this.) Justin probably did not produce it, rather he provides an attestation of it for us. Presumably Justin had this harmony in Greek, but it also may have existed in other languages. Boismard found a lot of support for it existing in the Old Syriac (the language that is very close to that of the Historical Jesus), and I think it is reasonable to suppose that it also existed early on in Latin. One way or another, this harmony was compiled on the basis of early gospel texts -- the texts that were subsequently re-edited to produce the final canonical versions. This is why this whole matter seems so important, of course. Indeed, JH, if indeed the Magdalene Gospel (MG) is a good witness of it, as Boismard argues, can provide us with a unique window to the history of composition of New Testament gospels.

So, according to Boismard, Tatian merely reworked this previous harmony to produce his Diatessaron (DT). In the process of his work, Tatian introduced some changes to the narrative order, and changed some of the texts included into his DT. One can assume that his DT also existed in Syriac, Latin, and Greek, although the Greek still seems to be rather poorly attested. It seems like both of these harmonies had a wide circulation both in the East and in the West, and Boismard presents a lot of evidence for this.

So what Boismard is really saying is that the Magdalene Gospel is a reasonably good witness to Justin's Harmony. But, importantly, MG also should not be seen as a "nail that sticks out and should be hammered down". In fact Boismard stresses that very strong textual parallels exist between MG and some other Western harmonies, especially the Liege DT. Indeed, independent of Boismard, I have found much further evidence that, indeed, numerous often unique parallels exist between MG and Liege. (In this case, it is possible that Petersen may not be appreciating the full complexity of Boismard's argument, as can be shown later.)

Boismard does not mention the Old High German DT in his book, but here is another distant relative of MG that should be added into the picture together with the Liege.

So these two clearly seem like the textual relatives of MG in the West. And all three seem to be derived at least in part from JH, and thus to preserve it for us, if Boismard's arguments are to be accepted.

But in the East, it seems, JH was not quite as lucky, because the two complete Eastern harmonies that have survived to our time, the Persian Diatessaron, and the Arabic Diatessaron, have not preserved the early material quite to the same extent. The Persian Diatessaron, that is believed to have derived from an Old Syriac Diatessaron, seems to be the more valuable of the two. But both the Persian DT, and the Arabic DT, have been Vulgatized (i.e. "corrected" based on the standard canonical texts) very substantially.

Otherwise, among the extant Eastern witnesses to the Diatessaron, we have only the bits and pieces that have been included into the commentaries, mostly by Aphrahat and Ephrem.  But, still, in the West, there are a lot more such commentaries (although of much later dates, to be sure) that also appear to preserve very substantial parts of JH, and Quispel's TATIAN AND THE GOSPEL OF THOMAS, 1975, provides large amounts of additional supporting material in this area. (It should be noted that in his book Boismard does not make any references to this work by Quispel, that may have actually helped Boismard somewhat to make his case.)

Perhaps Boismard's theory, as he formulates it, is a little too complicated. While I previously outlined it in 3 stages, Petersen expands them to 5. In my own view, there's no real need to postulate an additional "Syro-Latin Harmony" like Boismard does. Boismard sees this Syro-Latin Harmony as an intermediate stage between JH and the Western and other Harmonies, but I think this only complicates the picture that is already complicated enough as it is.

So now I will provide a synopsis of the arguments that Boismard outlines in his book, while also occasionally referring to the comments of Petersen on Boismard's findings.


-- by Yuri Kuchinsky --


After a general introduction, and an overview of both the Western and the Eastern Diatessaronic witnesses that survived to our time, Boismard focuses on the narrative order of MG. He compares its order with Fulda DT (Codex Fuldensis), and with Arabic DT, because it is generally believed that these two witnesses preserve best the order of Tatian's DT. After presenting and surveying a detailed comparative chart of the sequences of these three Diatessaronic witnesses, Boismard comments as follows,

"Contrairement a ce que l'on aurai attendu, l'harmonie P es beaucoup plus proche de l'harmonie A que de l'harmonie F." 39

"Contrary to what one would have expected, [from the point of view of its narrative order] MG is a lot closer to the Arabic harmony than to the Fulda harmony."

(By "l'harmonie P" Boismard generally means a harmony of the type represented by the Magdalene Gospel. And the same applies to his "l'harmonie A[rabe]" and "l'harmonie F[ulda]".)

And later he also notes that,

"..l'harmonie P, malgre quelques divergences, suit fondamentalement le meme ordre que l'harmonie A, ordre commande avant tout par l'evangile de Luc tandis que l'harmonie F suit un ordre mattheen." 41

"MG, in spite of some divergencies, basically follows the same order as that of the Arabic harmony -- this order staying very close to Lk, whereas Fulda harmony follows the order of GMatthew."

And later on in this passage, he also finds agreements in order between MG, the Liege, and the Venice DTs, together with the Arabic, on the one hand, against Fulda on the other. (Such wide agreements against Fulda are quite significant, of course, because this indicates that they derive from some other harmony more primitive than Fulda. One should note that the extent, and the exact nature of Fulda's influence on the later medieval vernacular Western harmonies are still disputed by some Diatessaronic scholars today, and in particular by Baarda.)

At this point Boismard also introduces some evidence from the gospel citations found in Peter Comestor's HISTORIA SCHOLASTICA (ca 1170 CE), that he also thinks is quite a significant textual witness. (And it is also used by Quispel, to be sure).

His general conclusion so far is that MG, also paralleled in part by Peter Comestor, preserves the form, or the narrative sequence, of the earliest gospel harmony better than any other western harmony.

"Une conclusion s'impose. En occident, l'harmonie P, dont dependent Pepys et, en partie, Petrus Comestor, represente la forme la plus ancienne d'harmonies evangelique." 42

It should be noted that these important conclusions that Boismard arrives at, about the very unusual and possibly quite primitive narrative order of MG, are in general accepted by Petersen.


After this consideration of the narrative sequence of MG, Boismard proceeds to the business of detailed down-to-earth textual comparisons, pericope by pericope, between MG and the other likely witnesses to JH from both the East and West. Because of space considerations, I will only summarise here his procedure and his findings as follows.

Boismard begins with "the Healing of the Paralytic" incident (Lk 5:17-19) as presented in MG, and he compares to this the texts of the Fulda, Toscan, Venice, and Liege DTs. After a detailed comparison, he concludes that, in this case, MG preserves best the mosaic of texts from which the earliest harmonistic narrative was composed, and notes some special contacts of MG with the Liege and the Venice DTs (45).

Next is "the Healing of Peter's Mother-in-law". Boismard notes the similarities in order between MG and the Arabic DT, and then gives the parallel texts of Venice DT and of Liege, and analyses them together with the text of MG. Then he adds a quotation from Severus of Antioch. After a lengthy analysis, Boismard concludes that, while MG generally follows the Arabic DT in this narrative, there's also an important divergence from the Arabic, where MG is also supported by Liege and by Severus (51). This close contact of MG with Liege and with Severus seems to indicate that all these follow JH rather than Tatian's DT.

Next is "the Temptation of Christ". After various comparisons, Boismard concludes that, while MG follows the order of Arabic DT pretty closely in this narrative, again it diverges from the Arabic in some important respects, which indicates that it may be following an earlier harmony. Also MG is supported in this to some extent by Liege and by Venice DTs against the Arabic. And Boismard concludes that this MG narrative has no connection with Fulda,

"Il n'y rien qui le rapproche de Fulda." 53

Next there's a detailed analysis of "Jesus' return to Galilee" pericope (Jn 3:22-26). Again, after various comparisons, Boismard concludes that MG has many parallels with the Arabic (both prefer the Lukan sequence over against Mt), but also he finds there some important divergencies -- the divergencies that, importantly, are also supported by Liege DT, indicating the antiquity of MG version (56).

Next, Boismard analyses the inaugural sermon of Jesus (Mt 5:1ff/Lk 6:12ff) in various harmonies. He concludes after his analysis of these texts that there seem to be two textual traditions present in them. One is that of Fulda, which favours the Matthean version of this narrative, and the other, and more primitive, is the tradition of the Arabic DT, and of MG. But also, among the latter two, MG seems to preserve the earlier text better (59).

Analysis of the scene of Jesus' arrest follows. Here, besides the other usual sources, Boismard also brings in the Himmelgartner fragments. These seem to agree with MG against Fulda in some respects, which supports the earlier thesis of Baumstark. Also, Boismard finds additional support in the Didascalia for some of the details in MG's narrative.

"Women at the Tomb" (Mt 28:1ff) is the last text analysed in this section of the book. In this narrative, Boismard finds important differences in MG compared to Tatian's DT as it can be reconstituted. He concludes that MG does not depend on the later Diatessarons in this case,

"Ici, il n'y a vraiment aucun rapport entre Pepys et le Diatessaron." 65


On pages 67-72, Boismard makes his case that Justin followed a harmony quite different from, and more primitive than Tatian's DT. (In this, Petersen generally agrees with Boismard.) He includes towards the end of this section a large chart where the sequences of various texts of the baptism of Jesus are compared. On the left MG, Fulda, and the Arabic are represented, and on the right the citations from Justin. This chart demonstrates clearly, according to Boismard, and I agree, that Justin in this case seemed to follow a harmony that was much closer to MG than to Fulda, and "especially to the Arabic". (72)

Next, Boismard goes on to the detailed analysis of these same texts which generally confirm his previous findings. He concludes on p. 82 that MG is indeed very close to Justin's Harmony. Thus, according to Boismard, the textual tradition of MG is in a number of respects earlier than Tatian's DT,

"En raison de tous ces exemples, il est difficile d'echapper a la conclusion que l'harmonie P[epys] devait deriver de celle que Justin avait utilisee et non du Diatessaron de Tatien." 82

"As a result of all these comparisons, it is difficult to escape the conclusion that the Pepys Harmony must have derived from the one utilised by Justin, and not from Tatian's Diatessaron."

And he adds to this that the Liege and Venice DTs should also be seen, along with MG, as witnesses of this "harmonie archaique" of Justin, in so far as they diverge from Fulda.

Next, Boismard analyses in some detail the rather unusual Dura Europos fragments, that may be preserving for us some fragments of Tatian's DT in Greek (pages 83-91). This analysis is very complex. According to Boismard, these rather special fragments basically stand apart from much else that we understand about Tatian's DT, because they feature considerable divergencies from all other witnesses to Tatian's DT. And yet Boismard also finds there certain indications of some similarities with MG.


On p. 93 Boismard begins his analysis of the Syriac Diatessaronic traditions. First, he deals with Ephrem (?-373) and his COMMENTARY. This text is preserved for us in its Syriac version, as well as in Armenian, and Boismard compares the two of them. He notes that in some cases the Armenian version may be more authentic.

According to Boismard, Ephrem's work may be a fusion of two commentaries, one of which was composed based on Tatian's DT, while the other was based on a harmony close to MG. As Boismard writes, Ephrem's COMMENTARY,

"..fusionne deux commentaires differents, l'un compose a partir du Diatessaron de Tatien, l'autre a partir d'une harmonie proche de l'harmonie P[epys]." 95

Boismard proceeds to examine the evidence in some detail, and to build his case for Ephrem's text including in itself two commentaries, one using a more primitive harmony than the other. (For some passages, Boismard actually postulates also the existence of yet a third commentary as attested within Ephrem's work! I will not deal with this yet additional complication here.)

Certainly this sort of a comparison needs to be extremely complex by its very nature, and perhaps even too complex... After all, we don't really have the text of Tatian's DT, so then how can it be compared so easily to any other document? As it's well known, scholarship in this area seems to have been basically stalled already for some time, and, apparently, we are not any closer to reconstructing Tatian's DT than we were 100 years ago.

As for Petersen, he seems to be rather sceptical in regard to this area of Boismard's work, although he also does not reject these theories out of hand.

From my own point of view, it makes very little difference for me if Ephrem's Commentary is homogeneous or not. This is not the question that has any big importance in this area. The important thing, on the other hand, is that Boismard does find considerable parallels between Ephrem and MG. In this, he draws on some previous work of Plooij, who also noted some of these parallels many years ago in his analyses of the Diatessaronic traditions (especially in his massive and magisterial edition of the Liege DT).

And, for some passages in this part of his book, Boismard also draws on the evidence from the Clementine Homilies, which is generally quite an obscure source that seems to preserve some unique Jewish-Christian material. Boismard outlines here some shared parallels between Ephrem, the Clementines, MG, as well as Venice DT. (99)

Boismard considers next the episodes of "the Marriage in Cana", and "the Call of Matthew" in Ephrem, and he finds additional evidence to support his thesis that Ephrem's COMMENTARY preserves considerable remnants of JH (100-104).


Next, Boismard begins his consideration of the Syriac gospel manuscripts, especially SyrS. Because it is widely believed that some of these texts were influenced by Tatian's DT, the relationship between the Old Syriac, Tatian's DT, and JH appears to be extremely complex. And so, Boismard admits that, in some cases, he's himself uncertain how these relationships really work. And yet Boismard often finds confirmation that some of these Syriac texts may preserve elements of JH, in so far as considerable parallels seem to exist between them and the textual tradition represented by MG, also supported by other Western versions of DT.

He begins the detailed analysis of these parallels by considering "the Multiplication of the Loaves" pericope (105-113). At the end of his analysis, Boismard's conclusion is as follows,

"De ce resume, il ressort que SyrS offre des affinites beaucoup plus marquees avec Pepys qu'avec les autres harmonies." 113

"From the [above] summary, it emerges that SyrS offers much clearer affinities with MG over above the other harmonies."

Because this pericope was singled out by Petersen for his criticism of Boismard's work, I can provide later a more detailed overview of it, together with an analysis of what Petersen has to say about it. In general, some of Petersen's critique may indeed be valid, but I think he misses the much bigger point that Boismard is making.

After this, Boismard deals briefly with "the Baptism of Christ" narrative (he also returns to this subject later on to deal with it at length). At this point, his conclusion is that SyrS supplies in this case a harmonised text that offers few contacts with Tatian's DT, but, on the other hand, its text is quite similar to MG. (114)

Next, Boismard looks at the gospel quotations found in Aphrahat, a Persian exegete who was a slightly older contemporary of Ephrem. In this connection, he considers in detail the following three narratives.

In the case of "the Tree and its Fruits" narrative, Boismard finds that Aphrahat in this case follows a harmonised text that is quite different from Tatian's DT. (115)

For the incident of "a Rich Young Man", Boismard supplies a long analysis which finds considerable agreements between Aphrahat, the Old Syriac texts, and the Western harmony tradition, including MG, against Tatian's DT (122).

And similar conclusions are reached on the basis of the analysis of "the Message To Those Who Are Persecuted" pericope (Lk 12:11ff).

Boismard's general conclusion stresses that often there are wide agreements between Ephrem, Aphrahat, and the Syriac texts, which indicate that all of them follow a harmony that is pretty close to that of Justin (125).

And finally, Boismard devotes pages 127-154 of his book to a very detailed analysis of the narrative of "the Baptism of Christ". He analyses all the usual sources to try to determine the earliest harmonised version of this narrative. He finds some remarkable and unique agreements there between MG and Peter Comestor's commentary. Some of the text of MG is also supported by Juvencus (ca 330 CE) (136). His conclusions are similar to those he reached earlier in the book, i.e. the version of this text as contained in MG seems very close to the gospels harmony as used by Justin.

His general conclusions and summary follow on pp. 155-157.

The general picture of these textual developments and interrelationships as outlined by Boismard seems entirely reasonable to me. Indeed, it does appear like there was another, yet earlier gospel harmony before Tatian's DT, and this was the harmony that Justin used. This harmony seems to be preserved quite well by MG, but also by some other Western harmonies. And Boismard provides large amounts of support for all of this in his book. (Of course, from my own point of view, "Tatian's DT" simply means "a later, more developed DT".)

I don't necessarily agree with Boismard on absolutely everything he says, but I welcome his study as a real tour-de-force by a highly skilled and qualified scholar. Also, I think there's a lot of additional material that Boismard hasn't even considered in this book that would also support his general analysis.

This is one of the more important works in the biblical field that I have read for a long time. Admittedly, his arguments are highly complex, and his presentation demands familiarity with quite a few languages, and with this whole rather difficult area of scholarship. It would be a very good idea for this book to be translated into English, so more scholars would get easier access to all this abundant and often unique information that he collected on this important subject.

With best wishes,


Click here to return to Yuri's New Testament Research Page.