Here’s the second installment of my interview with Bill Arnal on the Gospel of Thomas. Those who are both interested in and skeptical of “Q” will be interested in what he has to say here. Enjoy.
(CWS) 3. You are widely recognized for your work on Q. As you well know, there is a strand within North American scholarship that sees Q and Thomas as the earliest strata of the Jesus-sayings tradition. However, opposition to Q has been growing in recent years (largely due to the work of Mark Goodacre) and much of the recent work on Thomas insists that Thomas was familiar with the Synoptic tradition (viz., Gathercole, Goodacre, Meier). I am interested in hearing your reflections on and responses to these two currents within Q and Thomas studies.
(WA) I have a very hard time taking Q denial seriously. It’s easy to poke holes and find weaknesses in any source-critical theory whatsoever; to find this exception, that problem, and so on. What’s harder is to find an alternative theory that doesn’t suffer from just as compelling problems (or worse problems). And there’s a reason for this: the actual process of composing Mark, Matthew, and Luke (and Thomas) was necessarily and undoubtedly more complicated than any useful source hypothesis will reflect – source theories are maps, not territories, and they do not and should not include or encompass all of the literary details of the texts they discuss. But this means that there will always be material that a source theory does not quite grasp, and thus that any source theory can be criticized from this perspective. How to deal with this problem? By assessing source theories in terms of their overall persuasiveness, and in terms of broad patterns, rather than in terms of individual and isolated problems. Does a hypothesis offer a compelling and realistic overarching explanation of the general shape of things, or does it force us to rely on ad hoc explanations that appeal to facts not in evidence?
Since Q denial has mainly taken the form of simply throwing criticisms at the Q hypothesis (many of which turn out to be wrong [Q’s genre is unattested], or not logically sound [“Q is hypothetical!”], or are easily accounted for [many of the “minor agreements”]), without developing a strong alternative hypothesis, it strikes me that Q denial, in the long run, doesn’t have legs. Yes, of course, the main contender (but not the only one!) for an alternative hypothesis these days is FG [=Farrer-Goulder], but the real question must be: is FG a strong and better-evidenced hypothesis than 2DH [= Two Document Hypothesis]? Does it raise fewer problems than Q? And the short answer is no. Indeed, FG multiplies the problems we have understanding synoptic relationships. As just one simple example, it defers the problem Q is invoked to solve: if Mark was the first gospel, where did Matthew get his additional (i.e., the stuff we currently regard as M and as double-tradition) information from? We’re still forced to hypothesize some sort of non-extant source for Matthew (even if it’s just oral tradition). And we’re left with Luke’s use of Matthew as a source, which per FG requires us to imagine that Luke treated his two sources (Mark and Matthew) in completely different ways, and for no persuasive reason. The 2DH is the more compact and economical of the two hypotheses: it explains where double tradition (and maybe a very small amount of M and L stuff) comes from, and it imagines that Matthew and Luke are consistent in their treatment of sources. Aside from that, the 2DH, and specifically the postulation of Q, has an astonishing degree of what I like to call experimental validation. That is to say, we posit Q on the basis of the double tradition. Then we subsequently discover, independently of the grounds for hypothesizing it in the first place, that this material hangs together, that it has a coherent ideological perspective, that it consistently lacks certain types of concerns or vocabulary, that it has coherent reasons for lacking those concerns and consistent alternative vocabularies, that it represents a known ancient literary genre, and so on. Meanwhile, Q scholarship continues to produce interesting work on, especially, Q’s social context, including important forthcoming books on Q and peasant societies by Sarah Rollens, and on Q and scribal ideologies by Giovanni Bazzana.
In the case of Thomas, things are more complicated. We are dealing here – and everyone agrees, even if they don’t put it that way – with a redacted document. So we need to achieve some clarity on what is at issue here: is it that our manuscripts of Thomas show knowledge of the synoptics? Or that the final form of Thomas (insofar as we can accurately reconstruct it, given our dismal MS evidence) depends on the synoptics, and does so globally and exclusively? Or that our final text was influenced by the synoptics, but without global or detailed dependence? Or that, perhaps, an initial collection of chreiai were independent of the written synoptics, and later redacted from a synoptic perspective? Or that some initial chreia collection was dependent on the synoptics, and redacted from a non-synoptic perspective? And then there is the question – a question of importance no matter what one thinks of the sources for Thomas’s synoptic-like material – where did all the other stuff come from? The simplistic either-or approach to the question of sources strikes me as evidence that both sides are guilty of making thinly-disguised value judgments. What really seems at stake is whether Thomas is good (= early and independent) or bad (= late, dependent, Gnostic). Of course none of this follows: we recognize the canonical Gospel of Luke to be late, dependent on the synoptic(s), and agenda-driven, but no one takes this to mean that Luke isn’t worth studying, or is easily dismissible, or the like.
A more serious source-criticism of Thomas would not peck through the document looking for strings of synoptic-like wording, or, conversely, strings of wording that (somehow) couldn’t be synoptic; it would look at overall patterns, and try to understand what kind of constellation of sources, and what kinds of uses of those sources, best account for such patterns. After all, even if we could show that Thomas knew (one or more of) the synoptic gospels, that hardly proves that they were his only sources, even for material paralleled in the synoptics. Indeed, a clear case for dependency would raise a whole of fascinating ancillary questions about Thomas’s literary techniques, about his other sources, and about the dissemination and circulation of synoptic tradition in the late first or second centuries. In this sense, I think that John Kloppenborg’s rejoinder to Gathercole and Goodacre in the recent JSNT is exemplary. It identifies, among other things, many of the weaknesses in their arguments. But more importantly, it suggests an alternative that is productive, socially grounded, and sophisticated. Given the complexity of Thomas’s traditions, and the complexity of their relationship with their synoptic parallels, Kloppenborg suggests a context for Thomas drawing on a fairly extensive commentary tradition on the sayings of Jesus. This suggestion allows for the possibility of some synoptic influence on Thomas (as we have it), but at the same time recognizes that many of (I would say, the vast majority of) Thomas’s versions of this material do not appear to be drawn from the synoptics, and that there is a host of material on Thomas that cannot come from the synoptics (because it has no synoptic parallels). Whether he’s right or not, this argument has the virtue of taking Thomas seriously as a historical and textual datum, rather than either dismissing it or valorizing it.
But in the end, I don’t think these kinds of questions matter as much as people think they do. Yes, the idea of Q, and of an early Thomas, are mutually reinforcing; but neither requires the other. If you get rid of Q, it has no direct bearing on Thomas. And if you say that Thomas is a late document, and/or that it used the synoptics as sources, we are still left with a real document that both requires explanation, and that provides us with a host of new data for antiquity.
(CWS) 4. Not too long ago, you reviewed my book, What Are They Saying About the Gospel of Thomas? (JAAR 4 : 1113-1116), and along with your comments about the book, you also spent an entire paragraph lamenting the current state of Thomas research. Specifically, you suggested that scholars have been spinning in circles for years, asking the same questions over and over. To your mind, what areas are most in need of further research and what questions need to be raised that aren’t being asked by scholars? What are the most interesting or substantive areas currently being investigated, and in what other directions would you like to see Thomas research move?
(WA) Off the top of my head, there are two questions being addressed in current scholarship that are of serious interest, and that I would like to see pursued further. One of them is the interest in Thomas and middle Platonism – and especially Jewish traditions informed by Middle Platonism – which has in the past been flagged, inter alia, by Elaine Pagels, Arthur Droge, and others. Steve Patterson is currently investing a lot of energy in this question. Such an investigation will shed light not simply on the ideology and philosophical presuppositions undergirding Thomas, including insight into its treatment of Adam, androgyny, and the material world (none of which can be accounted for via synoptic parallels); but will also illuminate Thomas’s parallels with Paul and with the Gospel of John; and will allow Thomas to serve as additional evidence for the impact of Middle Platonism on Jewish mythological speculations. This in turn can help illustrate the cultural cross-fertilization between Greek materials and indigenous ANE traditions in places like Judea, Egypt, and Syria.
A second area of interesting research is basically into the social conditions of possibility for a text like Thomas. We have here a document whose genre is strongly associated with teaching, and with the inculcation of morals, a document packed with proverbial wisdom, and governed by an insistence on transformation, a document that contains at least one extensive elaboration of a saying (21), extensive glosses on others (45), and clusters of proverbs (e.g., 31-36). It is a literate document, and, more than this, a literary document, i.e., one that assumes proficiency in both reading texts and in interpreting them. So the most straightforward supposition is that Thomas is an ancient school product. The implications of this social setting for understanding Thomas, and for comparing it to other ancient social practices, is especially being pursued right now by Ian Brown, one of John Kloppenborg’s doctoral students. I think there is tremendous potential here for making sense of Thomas.
And then there are other areas that I wish received more attention than they do. For instance, the circulation and transmission of Thomas merits serious attention. The text is quite widely attested, so it is definitely worth trying to determine just who was interested in this document, and why. I also think that the parallels we find in Thomas with materials in Philo, in Paul, and in the Mishnah, among others, suggest that Thomas was participating in a conversation that some Jews were having in the first or second centuries. It would be very intriguing to try to work out the nature of this conversation, albeit without invoking a putative Aramaic or Syriac Thomas for which there is little evidence (on which point I am thoroughly persuaded by Gathercole).
Of even more importance, to my mind, is the esoteric character of Thomas. We need to understand this much better than we currently do. The document’s esotericism is quite obviously artificial and manufactured by the redaction of the text. “A man cannot mount two horses” is quite clearly not a mystical saying, and requires no great insight to make sense, nor does it seem to have any metaphysical ramifications. So it’s not the inclusion of stuff like this that makes Thomas esoteric. Thomas’s opening, of course, imposes an esotericizing framework on all the material that follows, but the actual substance of that following material reinforces the sense that this teaching is all mysterious, metaphysical, and in need of fairly serious interpretive skills. How does Thomas accomplish this? This is something that I think is a sine qua non of making sense of Thomas as a document, and that I’ve tried to contribute to in more recent work, including my contribution to Stowers’ FS, where I look at a variety of ways Thomas imposes an esoteric sensibility on his material, and so shapes both the document as a whole and its constituent units. This in turn raises the issue of, again, the economic, social, and cultural conditions of possibility for a text like Thomas. We can contextualize it in terms of both the social circumstances of ancient schools, and the production and the treatment of esoteric texts in antiquity, thus placing Thomas within a range of relatively ordinary ancient practices and wider human doings than is afforded by the very restricted comparisons available to “Christian theology.” To a considerable degree, I think Thomas was as popular as it was simply because it made possession of a fairly basic knowledge of Middle Platonic tropes feel like a big deal. It thus addressed the social aspirations of the recently- or modestly-literate.
There’s another piece I published some years ago, that I don’t think gets nearly the attention it deserves. It’s called “The Rhetoric of Social Construction” (in Rhetoric and Reality in Early Christianities, 2005), and in it I argue – I would say, I show – a literary feature of Thomas that does not seem to be widely recognized. Thomas uses metaphors copiously, and he uses them in loaded ways: that is, a given notion, say, “poverty,” has a particular negative or positive connotation. What several analysts of Thomas have noted is that sometimes, these metaphors are applied in contradictory ways. The clearest example I can think of is Thomas’s use of “drunkenness.” In saying 13, being drunk is a positive metaphor for having the capacity to grasp Jesus’ teaching, whereas in saying 28, it’s a negative metaphor for lacking the capacity to receive Jesus’ words. This kind of thing is sometimes held up as evidence of Thomas’s incoherence, or of layers of tradition. But it turns out that this inconsistency is thoroughly consistent across the Gospel of Thomas: nearly every single metaphor used in the text is used multiple times and with opposite valences. This is true, for instance, of: drunkenness, poverty, wealth, merchants, usury, duplicity, thieves, maleness, and others besides. The inconsistency of Thomas is so extraordinarily consistent that it indicates a very clear redactional perspective that unifies the document across a wide range of sayings. This article has been cited a couple times, but to the best of my knowledge, no one has engaged it directly, or indicated whether they find my explanation of the phenomenon to be adequate or not. But what I think is especially interesting here and perhaps fodder for additional strands of research, is how this literary feature of Thomas plays in connection to theories about Thomas’s sources, or stratification hypotheses. It seems to me that this phenomenon of opposing metaphors could be used to confirm, or disconfirm, those hypotheses by seeing how this feature is distributed across the putative sources or layers of the text. Take April DeConick’s stratification of Thomas, for instance. We actually have a means to test this thesis: do one set of metaphors appear consistently in her kernel, and the opposing metaphors in a particular accretion layer? Or perhaps all of them stem from one set of accretions? Or at least, does positive usage of a specific metaphor consistently dominate one accretion layer, with a negative use dominating the other? Such observations would tend to confirm DeConick’s thesis. Or again, take the stratification I proposed in 1995: same questions. And if the metaphors do not line up by strata, or appear to stem from one stratum in particular, this would tend to disconfirm my thesis. The same kind of thing applies to source theories: does the pattern of conflicting metaphors mesh somehow with the shape of the putative sources, or not? So I would really like to see someone run with this, although I have no plans to do so myself.
Again, I really appreciate the substantive and detailed way Bill has answered these questions. Stay tuned for part three!