After a break to follow some distractions and another to earn some money, I am returning to things that struck me in reading Nicholas Perrin’s Thomas, the Other Gospel. As part of his doctoral work, Perrin did a reconstruction/retroversion of GosThom in Syriac and first argued there for a Syriac original which is dependent on Tatian’s Diatessaron. He published this research as Thomas and Tatian: The Relationship Between the Gospel of Thomas and the Diatessaron (Atlanta: Society of Biblical Literature, 2002). During the course of this work, he identified 269 catchwords in Coptic Thomas, 263 in a reconstructed Greek version and 502 in his Syriac reconstruction.
I have no Syriac, so I won’t attempt to make any comment on that part of his work, although I would take issue with this comment:
While there is inevitably some guesswork in reconstructing the Syriac, thankfully I was able to make use of some controls. When Thomas parallelled the NT scriptures, I supplied the word-for-word equivalent of the oldest extant Syriac copy of those scriptures. (Other Gospel, 86)
It seems to me that this would only be a valid methodology in passages where there is verbatim correspondence between Coptic Thomas and the Greek Canon. Except where we have the Oxyrhynchus passages, reconstructing the Greek text is not an exact science, and it seems to me that there are very few passages where Coptic Thomas is exactly the same as one of the Greek canonical gospels. Thus assuming that any hypothetical Syriac Thomas would be exactly the same as the Syriac canonical gospels seems problematic to me. I would see using a word-for-word equivalent of the Syriac scriptures as having more potential to introduce bias rather than control, but perhaps I have misunderstood what Perrin actually did? If he simply means that when translating, for example, the parable of the mustard seed, when faced with several possible Syriac words, he chose the one that appeared in the Syriac Scriptures, I have fewer concerns.
It would certainly seem that being able to nearly double the number of catchwords and create links that don’t exist between sayings in Coptic or Greek by translating back into Syriac provides some quite strong evidence for a Syriac original rather than a Greek one, even allowing for some bias in the retroversion.
The step that surprises me in Perrin’s argument about catchwords, however, is this:
First, the consistency of the catchword pattern indicates a literary as opposed to an oral background. Even for Bultmann, who was loath to attribute synoptic material to the editorial activity of the gospel writers, long chains of sayings indicated the presence of an editor who worked hard to put those saying together. (p 93, quoting Bultmann’s History of the Synoptic Tradition, New York, Harper and Row, 1963/1921, 322)
Unfortunately, I don’t have immediate access to Bultmann’s book, but it seems to me rather unlikely that he was comparing oral versus literary transmission when he said this. While catchwords can certainly be used as a literary device, they are also used by oral tradents as a means of remembering the ordering of complex oral material and if we are prepared to talk about “oral texts” then this kind of work is the oral equivalent of editing.
Perrin then quotes Stephen Patterson (The Gospel of Thomas and Jesus, Sonoma, Polebridge Press, 1993, 102):
The significance of such a pattern in Thomas may be assessed variously. For example, an editor might have organized the collection in this way to facilitate its memorization. The utility of this for the street preacher, who would compose his or her speeches ad hoc in the busy colonnades of the agora, is obvious. Alternatively, one could well imagine an editor assembling these sayings simply as he or she remembered them, cacthwords triggering the recollection of each new saying. In this case the catchwords will not have been part of any conscious design on the part of the editor, but simply the result of his or her own process of remembering. The occasional gaps where no catchwords are to be Found suggest the latter. (Italics added by Perrin)
From this he argues (p 94) that since Patterson says that gaps in catchwords = no intentional editing, no gaps in catchwords must necessarily mean that editing has taken place, so the fact that his Syriac retroversion has no gaps in catchword links means that Patterson must conclude that there has been editing. This may not be the case, because I think that the issue is more complex than this. I don’t think, however, that demonstrating that there has been intentional editing necessarily makes a watertight case for the document having been written all at one time by the person who transcribed the text.
As I indicated earlier and as Patterson points out, catchwords were important tools for oral tradents who needed to be able to remember long pieces of oral text. I think that this would have been particularly important for someone who was trying to memorize the material in GosThom because it doesn’t contain any narrative that would help the tradent to keep it in order in his or her mind. Most of the studies on oral transmission of long pieces of text are on narratives about the careers and mighty deeds of heroes, where there is a particular sequence of events that makes memorization easier. GosThom is simply a series of sayings of Jesus, so a chain of catchwords would have been particularly important if an oral tradent was to remember all of it. Thus, it seems to me that a careful chain of catchwords is just as likely to support the argument made by DeConick and others that GosThom moved in and out of oral transmission stages during the course of an extended composition as it does Perrin’s theory that the text was composed late and all at once. It seems to me that the careful chain of catchwords and the rather random order rather better fits the notion of a corpus of sayings that was designed to be communicated orally than it does the carefully crafted work of someone who was producing a written text. If an editor was going to go to all that trouble, why not also try to knock the sayings into a more logical order for the reader? If this is the case, then I think that Quispel’s theory of a shared source explains the commonalities between GosThom and the Diatessaron at least as well as does Perrin’s that GosThom is dependent on the Diatessaron, given the dating issues that Perrin points out on p 97.
In other words, I don’t think I’m convinced, but I am not prepared to be admant about it, either.