Gathercole on the composition of Thomas (3)

In Chapter 3, Gathercole works through 77 areas in the Coptic text of Thomas that have been proposed by various authors as Semitisms. He looks at those identified by Quispel and Guillaumont and listed by DeConick in her The Original Gospel of Thomas in Translation but adds a number of others presented elsewhere in the scholarly literature. He quite rightly says that problematising the proposal of a Semitic background requires a large sample size.  He does not, however, attempt to include all the 502 Syrian catchwords proposed by Perrin. Of the aim of the chapter, he says:

It is hoped that the present chapter will show that in almost every instance, alternative explanations are readily available, and to suggest that, as a result, the case for a Semitic Vorlage underlying our Greek and Coptic texts has been greatly exaggerated and is in fact very vulnerable. In addition to the immediate concern with the original language, this chapter is also significant for the question (which will loom large later, in Part II) of Thomas’ s independence, since we will treat here a number of alleged cases of Aramaic Vorlagen translated differently (and thus independently) by Thomas and the Synoptics. (pp 43-4)

As can be seen from the previous post, Gathercole’s argument about the Semitisms in Thomas looks at three main areas: the identification and classification of Semitisms in the text; the identification of mistranslations or wooden translations which could are explained by an underlying Semitic text; and the identification of divergent translations that occur in either the Greek and Coptic Thomas or  in canonical parallels to Thomas and which could be explained by a common Semitic Vorlage.

I think he demonstrates quite credibly that quite a few of the pieces of text identified by other scholars as Semitisms are either acceptable Greek or acceptable Coptic idiom and that others, while clearly arising as a result of translation from a Semitic language can be classified as Septuagintisms, rather than what might be termed de novo Semitisms. He also demonstrates that a significant number of those passages which have previously been considered to be the result of mistranslation are actually the result of problematic exegesis and are acceptable was they stand; and that in a number of situations where real problems exist with the text, there are Greek explanations that are equally as likely as the Semitic ones that have been proposed, or they could be explained by textual corruption. Thus, he raises significant doubt in the first two areas.

In looking at the third area, that of divergent translations, he notes that in a number of situations the parallel texts are so different that they could only be considered to be loose translations at best, so do not provide convincing evidence that they are translations of a common Semitic Vorlage. In other situations, the divergences are translations of conjunctions and prepositions which are acknowledged by scholars  to be translated unpredictably between other languages and Coptic. Finally, there are places where the explanations offered by other scholars require a translation directly from a Semitic language to Coptic, which he finds untenable.

He finishes with: “These conclusions do not, of course, mean that it is impossible that various sayings in Thomas go back to Semitic originals  . . . The analysis in this chapter does emphasise, however, how difficult it is to conjure up evidence which can only be explained on the basis of a Western Aramaic or Syriac Vorlage.” (p 104)

Again, however, while the evidence he provides is well researched, it seems to me that the conclusions he draws from it are an overstatement of the case, for several reasons. First, even though it may be possible to provide an individual explanation of every Semitism proposed that does not require that it comes from a Semitic language original, the people who are proposing a Semitic original are saying, in effect, that when you put all these pieces together the overwhelming ‘feel’ of the text is that there is a Semitic language underlying it. To provide a modernt parallel: When I read the English translation of Uwe-Karsten Plisch’s commentary on Thomas, there is nothing in it that is incorrect, but there are definitely segments which someone who speaks English as their first language would have worded differently. The fact that I can provide perfectly acceptable explanations for each one of them in English does not take away the “germanic” feel which is caused by their presence in numerous places in the text. Of course, as Gathercole himself notes once or twice, it is difficult to identify the source of this Semitic feel. Given that Jesus did most, if not all, of his teaching in Aramaic and Thomas consists almost entirely of  Jesus’ words, it may come from Jesus. If the person who translated the text into Coptic spoke a Semitic language as his/her first language s/he may have introduced Semitisms that did not exist in the original text, and the notion of LXXisms of course makes sense, too. Thus, while a Semitic Vorlage may not be the only explanation for many of the Semitisms detected in the text, it must one of the possible explanations remain at this stage and I don’t think it is necessary that it be the only possible explanation in order for it to be the best one.

From my perspective, however, there is a more significant gap in Gathercole’s treatment when he deals with the divergent translations. While I concur that we are on shaky ground trying to demonstrate a common Vorlage for material that appears only to be a loose translation of the original, much of the problematic material seems to me to have an explanation which Gathercole does not explore. The more I read about eyewitness testimony and human memory, the more likely I think it that divergences such as those between Thomas and the canon come from the accounts of different eyewitnesses to the same events, exacerbated by the fact that the transmission for the first decade or two was oral. Most of the divergence theories were proposed several decades ago, before much research had been done about eyewitness testimony and when biblical scholars were largely unaware of research on human memory and oral transmission that was being done in other fields. The variations seem to me to be better explained as gist transmissions by several eyewitnesses through different trajectories than by loose translations of one underlying text. Whether they became stable as part of community tradition (see eg Bailey) or as a result of Jesus teaching his disciples in the manner of the rabbis of his time (see eg Gerhardsson) is unclear.

In chapter 4, which will be the subject of the next post, Gathercole moves on to positive evidence for a Greek-language origin. This may build a case which outweighs the difficulties that I’ve raised above.

Gathercole on the composition of Thomas (2)

The first section in Gathercole’s book deals with the original language of Thomas and consists of four chapters. This post will deal with the first two.

In chapter 1, which is very short, he outlines the various theories that have been advanced about the language in which Thomas was originally written. When Puech announced the discovery of the gospel in 1957, he was sure that the original had been in Greek. In 1958, Guillaumont noted a significant number of mistranslations and argued that the sayings had been translated from Aramaic. In 1960, Garitte proposed a Coptic original. The notion of a Coptic original has not persisted, but both other options are still argued by scholars today, while other scholars have added arguments for a Syriac original. A significant factor in the various arguments for an Aramaic or Syriac original is the presence of many Semitisms in the text, but Gathercole indicates that he plans to present a criticism of the proposal for an Aramaic or Syriac original and demonstrate that a Greek original is ‘much more likely’. This he will do in three parts – first looking at problems with gathering evidence for a Semitic original; second examining all the proposed Semitisms systematically; and third providing positive evidence for a Greek original. ‘This conclusion in favour of a Greek original will pave the way for seeing a closer relationship to the New Testament Gospels than is often seen in current scholarship’ (p 23).

Chapter 2 looks at methodological problems with Semitic theories under a number of headings.

  1. The need to eliminate Greek and Coptic explanations before arguing for a Semitism: this is the basic, common sense idea that even though a particular turn of phrase might be common in Semitic languages, if an explanation for it can be found in Greek or Coptic, it should not be claimed as a Semitism. Gathercole offers five possible scenarios where this might be the case: (a) where a particular turn of phrase which is common in Semitic languages is also acceptable Greek or Coptic; (b) where the original text is corrupt (c) where there was a scribal error in copying; (d) where the phraseology in question has been misunderstood (e) in Thomas, it may simply be that the author’s intention is to be obscure.
  2. The need to establish the linguistic base for identification of Semitisms: here, he draws on the work of Wilcox and Davila on the need to be comparing the text with Syriac, Hebrew and Aramaic texts from the same period rather than from a century or two afterwards.
  3. The difficulty of classifying Semitisms: some Semitisms come from Greek translations of the Hebrew scriptures and make good sense in the text as it stands – they are examples of ‘biblical expression’. Gathercole cites two other similar cases where, he argues, the Semitic origin of an expression does not indicate any particular linguistic background and distinguishes between these and those cases where the Greek makes little or no sense because it appears to be a literal translation of a specific Semitic idiom.
  4. The difficulty of assessing the significance of Semitisms for the original language of composition: texts such as the Didache provide examples of texts written in Greek but with a huge number of Semitisms. Gathercole quotes Davila’s argument that the presence of such Semitisms cannot be taken as decisive proof of translation from a Semitic Vorlage. He also argues that the fact that part of a composition might have a Semitic origin does not indicate that the whole has a Semitic origin – citing as an example the parts of the Gospels that can be traced back to the OT.

Gathercole then proposes that there are two principle ways to identify a Semitic Vorlage to Thomas – identification of mistranslations which make little or no sense in Coptic or are very wooden, but which make sense as idiom in a Semitic Vorlage;  and of divergent translations of parallel passages in Thomas and the canonical gospels or in Greek and Coptic Thomas which might be accounted for by a common Semitic Vorlage, although these also are not without problem, the possibility of bilingual interference being one that is common to both possible explanations.

Gathercole then lists four additional problems associated with therories of a specifically Syriac original.

  1. The paucity of Syriac literature in the relevant period- we have virtually no evidence of Syriac being used as a literary language in the first two centuries CE, so the earlier Thomas is dated, the more surprising it would be for it to have originally been written in Syriac.
  2. The rarity of translation of Syriac works into Greek
  3. The possibility of bilingual composition – A late Syrian origin might well have resulted in two versions, one in each language, having been written mor or less simultaneously, as Klijn has suggested for the Acts of Thomas
  4. The difficulty of the ‘catchword’ theory – here, Gathercole criques Nicholas Perrin’s identification of catchwords as being too uncontrolled to be convincing.

I think that Gathercole has successfully made a case for need not to be too hasty in identifying Semitisms in Thomas; and in reminding us that there are other explanations for the presence of Semitisms in the text than that the text was originally written in a Semitic language. I am rather surprised, however, that he hasn’t suggested what to me is an obvious possibility – that the author spoke a Semitic language as his/her first language and was not sufficiently fluent in Greek to eliminate entirely the traces of that language from the text ie that the author was not, in fact, truely bilingual – although perhaps he uses a looser sense of  ‘bilingual’.

I am not sure, however, that he makes a strong enough case to justify his conclusion to the chapter:

In sum, these caveats may lead us to wonder whether an Aramaic or Syriac original is identifiable; at the very least they should mean that the burden of proof lies heavily on those who would argue for such a Semitic Vorlage.  It is surely such factors as the above which led even such an enthusiast as Ménard to compare the terrain of the study of Semetisms to quicksand ( L’évangile Selon Thomas, 1973, p 23). As we proceed to investigate the particular instances, we will see that the terrain is uncertain indeed. (p 42)

I continue to be uneasy when dealing with early Christian material when people start talking about ‘proof’ and ‘burden of  proof’. I simply don’t think that the material we are working with enables us to be more confident that anything is more than ‘highly likely’ – empirical proof is beyond us.

Simon Gathercole – The Composition of the Gospel of Thomas

My copy of Simon Gathercole’s The Composition of the Gospel of Thomas – Original Language and Influences (Cambridge University Press, Cambridge, 2012) arrived last week and I started on it this weekend. It looks as though it is going to be a very good read!

The book is divided into three parts:

  1. The original language of thomas
  2. The synoptic gospels and Thomas
  3. Thomas and other early Christian literature

In the introduction (the only bit I’ve read so far), as well as indicating the scope of the work with respect to language and influences, Gathercole highlights some incongruities in current Thomas scholarship for which he proposes an alternative approach. The incongruities he flags are:

  1. the revival of Semitic theories of Thomas’s composition in light of recent scholarship on Semitisms
  2. continued attachment to form-critical “laws” in the light of exposure of their subjective nature and even falsification
  3. confident assessments of oral factors in Thomas in light of scepticism elsewhere about their predictability and distinctiveness
  4. the assumption of detailed knowledge of Q in light of recent “unfreezing of the Synoptic problem”
  5. persistent polarisation of “independent oral tradition” vs “literary dependence” despite some questioning within Thomas scholarship

(Note that the above are in the exact wordings of the subheadings in the introduction pp 4, 5, 6,8 & 9 respectively – I felt that individually footnoting them might be just a tad tedious.)

Issues 4 & 5 have been of particular interest to me, too, and I have given passing attention to the others as well, so I am looking forward to seeing what he has to say. I am particularly interested in his expansion on his article  ‘Luke in the Gospel of Thomas’, New Testament Studies, 57 (01) (2011), pp. 114-144 (he indicates on p 15 that he does this) because I had some questions when I read it and was hoping that he addressed it more fully in this book. I agree with some of the positions he outlines and not with others, but the points he makes on issues where we disagree have in general been thought-provoking. I am wondering if he will cause me to change my mind. :-)

Robert McIver – Memory, Jesus and the Synoptic Gospels

Last year, SBL published Robert McIver’s book Memory, Jesus and the Synoptic Gospels. I ordered it immediately but it sat on my shelf for several months before I had time to look at it. McIver is a fellow Australian and I met him at the SBL international conference in Auckland, New Zealand in 2008.  He has published two papers on oral and written transmission and the gospels with his colleague Marie Carroll* which I found useful in writing my JBL article on eyewitness testimony, so I was interested to see what he had to say.

The book provides a quite comprehensive coverage of the psychological research on eyewitness testimony and human memory, which makes it an excellent resource for anyone who is interested in an overiew of the field. I found the chapter on personal event memories particularly helpful. It draws attention to the work of David Pillemer in Momentous Events, Vivid Memories (Cambridge, Mass.: Harvard University Press, 1998) which extends the classic work of Brown and Kulik on ‘flash bulb memories’ to memories of momentous events in general – something I had not come across before.

The book has limitations, however. First, I would question McIver’s decision to limit his discussion of the problems of human memory in Daniel Schacter’s The Seven Sins of Memory (Boston, Houghton Mifflin Company, 2001) to transience, suggestibility and bias. I think that both misattribution and absent-mindedness can also have significant impact on eyewitness memories of events in some of the circumstances in which the gospels came into being.

Second, and far more importantly, I do not understand how he has moved from the evidence he has provided to the conclusion he draws. He provides careful evidence that eyewitness memory is not at all good at producing verbatim reproductions of what was said; that people often see and remember what they expect or want to see and remember; that there is a rapid deterioration of memory at the level of detail in the first hours, days and weeks after an event and it is not until about five years out that the rate of forgetting slows almost to a stop; that people fill gaps in their memories with material that is likely to have happened, given their knowledge of the people and circumstances involved so that ‘what is recorded in the Gospels is highly likely to be consistent (emphasis McIver’s) with what actually happened’ (p 156); and that while about 80% of what is remembered by eyewitnesses is accurate we have no way of determining which 20% in inaccurate.  He notes that apophthegemata and aphorisms are more likely to be remembered almost verbatim or not at all than are extended narratives and argues that parables, as coherent stories with a punch line, are ideal for accurate transmission, but the overall picture he provides is the reliability of reproduction is at the level of gist and overview rather than verbatim repetition and fine detail.

He also argues with Birger Gerhardsson et al that Jesus is likely to have trained his followers to remember his teachings in the same way that rabbis trained their disciples. This is really the only way that one could be confident that the words put into Jesus’ mouth by the authors of the gospels are the words Jesus spoke. The evidence from psychological research points to gist-only levels of accuracy.

McIver finishes by saying:

So it can be concluded that, like most products of human memory and despite all the frailties of such memory, the Gospels should be considered to be generally reliable. If the evidence presented thus far may be relied on, then  -  at least for the apophthegmata, the parables and the aphorisms – the burden of proof should lie with those who wish to claim that a saying found in the Gospels in not from Jesus or that an incident reported about him did not happen, not with those who assume its authenticity. Human memory is a remarkable facility, and the traditions found in the synoptic Gospels may be considered to be a product of its effectiveness. (p 187)

Unfortunately, incidents are not apophthegmata, parables or aphorisms, and the evidence provided by the psychological research is not, at least in my opinion, nearly strong enough to make this confident a statement about events, regardless of what you may or may not believe about Jesus as rabbi. While a careful reading of the book makes it clear that McIver is not saying “see, see, we can prove that we have Jesus’ actual words and blow-by-blow, accurate accounts of his actions”, I think he significantly overstates his case and I will not be at all surprised if others use the book in this way.  I think what the book does is rather to provide some helpful understandings of how the variations between the gospels have arisen – due to the frailties of human memory rather than a deliberate attempt on the part of the authors to push their particular theological barrow at the expense of accuracy.

* McIver, Robert K. and Marie Carroll ‘Experiments to Develop Criteria for Determining the Existence of Written Sources, and Their Potential Implications for the Synoptic Problem’, Journal of Biblical Literature, 121 (4) (2002), pp. 667-687; and McIver, Robert K. and Marie Carroll, ‘Distinguishing Characteristics of Orally Transmitted Material When Compared to Material Transmitted by Literary Means’, Applied Cognitive Psychology, 18 (9) (2004), pp. 1251-1269.