π -ness Envy? The Irrelevance of Bayes’s Theorem

In a recent post responding to a blog review of Sources of the Jesus Tradition by atheist blogger Richard Carrier, I made the point that his own contribution to the book does not rise above the level of pedantic lecturing on a theorem of dubious value to engage the literary matter.

Carrier has claimed on a number of occasions that his approach is revolutionary, a tour de force and essentially over the heads of New Testament scholars.  Apart from the naivety of saying anything like this in a field littered with the corpses of dead theories and “discoveries,”  this is scarcely where you’d want a revolution to be fought.

Strauss

There are numerous critical issues attached to using a theorem that is primarily about probability to assess material that isn’t.  It is, however, a common feature of forensic (i.e. controversialist) approaches to the Bible on both the fundamentalist side and the atheist side to engage the material on a literal level.  This is so because both sides have to meet on the field at that point where literal claims about the text are being made, if not with the claim that the texts themselves are designed to propose facts–though most biblical literalists would say that they are, and most of their opponents would say that they are defeasible at this level.

Accordingly, a particular way of reading the text has been the main source of “rationalistic” critiques of the Bible since the Enlightenment though many of those critiques were superficial and almost all have been improved by serious academic study in the last two and half centuries.

Reasonably speaking,  it is analytically impossible to assess claims of “factuality” without assessing the texts on which such claims are based, even if we begin with relative certainty or skepticism (for example) about the occurrence of miracles or the reliability of a written tradition. Despite the fact that the multiplication of “conditions” and “assumptions” violates the sensibility of most post-modernists who deplore looking at things like authorial intention, criticality, or audience (community) in discussing narrative,  the character of the text is the conditio sine qua non–the starting point—for all discussion.  It is a condition a literalist believer, on the basis of his epistemological suppositions, tries to avoid since his prior assumption is that the text has a particular integrity, and the skeptic, guided by opposite beliefs, no less literal, is often able to ignore.

Modern New Testament scholarship emerged precisely in response to the impasse between credulity and skepticism, neither of which seemed a sufficient answer to the “problem of the text.” Forensically speaking, text is text.  Critically speaking, it isn’t.

Generally speaking, the biblical literalist feels he is under no  compulsion to defend his confidence in the text; he assumes he has warrants for his confidence.  He can invoke a number of interdependent subordinate claims to support his position–arguments from antiquity (the age of the text or its distance from the reported events), reliability (a kind of mock-psychological assessment of the trustworthiness of “reporters”), self-consistency (whether the text is basically coherent within itself and among variants, where they exist), inspiration and inerrancy  (the belief that the text is autonomous as a product of revelation and thus superior to any methods used in its evaluation). Most subordinate claims have been savaged by modern critical approaches that have grown organically out of the study of the gospels and cognate literature, though some are still of interest to historians.

A forensic approach to the Bible means that key debating points like the six-day creation story, the resurrection of Jesus and (perhaps) the existence of Jesus have to be treated as historical assertions to the same extent they are asumed to be true by the most literal readers.  This is a severe limitation to forensic approaches since they initiate discussion with the question of whether a text is vulnerable as a truth-claim, using a formula more suitable for modal logic than for history: Is something possible or probable?  Are events described in a text more likely to have happened or not to have happened?

Hume

Even the study of the text for both literalists and skeptics will be subordinated to the modality of claims.  Texts that assume propositional value for the literalist (even if that value has to be manufactured) are the very texts the skeptic needs to find defeasible.  Scholars have understandably winced at this level of discussion because it’s easily seen as a branch of apologetics rather than as a field of serious literary and cultural study. Its preoccupation is not with what the text has to tell us, but with whether you or your opponent is right or wrong about a relatively small number of events.

Gunkel

Debaters like Carrier have suggested that the critical methods developed for dealing with the Bible in the nineteenth and twentieth century are insufficiently rigorous. But that is simply not the case.  In fact, the methods grew in tandem with evolving perceptions of what the character of the text actually was, how it was formed, and what its creators thought about the world. In the language of an older school of criticism, what its “life situation” was. They continue to evolve and to adapt in an organic way.  Only if the sole question to be answered is whether the description of an event corresponds directly and generically to “what really happened” (if it were possible to answer that question, as it isn’t in many cases) would the modality of a forensic approach be useful, and its usefulness would still depend on prior questions.

“Conventional” and revisionist approaches remain central to academic study, however, if we assume that the New Testament is not making its case propositionally, event by event, but narratively.  If Genesis was not intended to teach astronomy, the New Testament was not intended to teach medicine. Neither of those statements tells us what the Bible was intended to do, yet such a determination would be essential for answering questions about how it fulfills its purposes.

Beyond the forensic approach, the question about the kind of literature the New Testament literature represents remains absolutely prior and absolutely crucial.  As an example, the amount of material that can be removed into the category of “myth” (a great deal, from most of Genesis to all of Revelation) can never be determined by modal assessment of the truth properties of a text, since analytically myth is not amenable to modal analysis and only a wrong definition of myth as a kind of rhetorical lie or pre-scientific error–a definition that flies in the face of modern anthropology–would make such analysis possible. The forensic approach does itself a huge disservice by paying insufficient attention to the history of criticism, where the general mythological character of much of the material is almost taken for granted, and focusing instead on a discounted view of myth as non-factuality.

What is true of myth, moreover, is true of the other “forms” (literary and historical genres) that exist within the Bible and the New Testament especially. So much of the Jesus story is myth, in the sense of μυθογραφία (writing of a fabulous story), that I have no objection to the phrase “the Jesus Myth.” –But a great deal to object to in the sentence “Jesus ‘was’ a myth,” implying absolute non-historicity and a method designed simply to document his irreality.  In Sources, this is the subject of two essays, one of which (“On Not Finding the Historical Jesus”) suggests that the gospel writers bore no interest in the “question” of the historical Jesus but had a profound interest in his reality.

For the forensicists, “Was the cosmos created or was it not created in six days”; “Did Jesus or did he not rise from the dead?”;  “What did he really say?” and “Did he exist?” are primary questions that should not be swept under the rug of literary analysis: they are questions of right and wrong.  The text exists primarily to settle these questions.

In my view, this is an impoverished way to approach the Bible since the book (taken as a kind of religious artifact rather than an accident of editorial history) was not construed to answer such questions and the methods that have been devised to explore it have been driven by different phenomena and concerns: what communities believed; how they understood society; how they manipulated history and politics religiously to provide social coherence; why ideas like salvation and redemption gained ascendancy in the first century and how they evolved to become something quite different in the second.  Put flatly: the questions asked by the forensic approach are not primary questions at all because they do not arise from the text.

Not unless you accept the prior assumption that the literature of the Bible puts itself forward as hard fact (and most scholars in the present century would say, it doesn’t) all operations on the material should derive naturally from what it is.  Certain techniques like hermeneutical suspicion, mutiple attestation, “dissimilarity,” and redaction, source (and various other) criticisms and linguistic distribution are simply code for ways of testing how the tradition developed and how the sources evolved over time.  If anything, the “factuality” or modal probability of events in using any of these methods is held in suspense in the same way Coleridge describes the willing suspension of disbelief (and for not altogether different reasons) in the Biographia Literaria.

Back to Bayesics?

I was reminded of the danger and potential irrelevance of imposing non-literary templates on the biblical material by a former student, whose comments on the use of Bayes’s theorem are significant because (a) he is not critiquing the use of this device as a New Testament scholar: he is a PhD candidate in mathematics and is properly reckoned a prodigy in pure mathematics; and  (b) he is not a Christian.

I personally find his comments devastating to the use of the theorem as an assist to the modal approach to the Bible.  But I’ll leave it to others to decide:

“Is this insistence [Carrier] of trying to invoke Bayes’ theorem in such contexts a manifestation of some sort of Math or Physics envy? Or is it due to the fact that forcing mathematics into one’s writings apparently confers on them some form of ‘scientific’ legitimacy?

The fact of the matter, as far as I know, and as I thought anyone would realize is that Bayes’ theorem is a theorem which follows from certain axioms. Its application to any real world situation depends upon how precisely the parmeters and values of our theoretical reconstruction of a real world approximate reality. At this stage, however, I find it difficult to see how the heavily feared ‘subjectivity’ can be avoided. Simply put, plug in different values into the theorem and you’ll get a different answer. How does one decide which value to plug in?

Secondly, is it compulsory to try to impose some sort of mathematically based methodological uniformity on all fields of rational inquiry? Do there exist good reasons to suppose the the methods commonly used in different areas that have grown over time are somehow fatally flawed if they are not currently open to some form of mathematization?

If this kind of paradigm does somehow manage to gain ascendency, I assume history books will end up being much more full of equations and mathematical assumptions etc. While that will certainly make it harder to read for most (even for someone like me, who is more trained in Mathematics than the average person) I doubt that it would have any real consequence beyond that.

The fatal flaw in Carrier’s misuse of the theorem therefore is that the “real world conditions” he finds described in the gospels are not real world conditions. Thus its application does not flow from axioms designed for its use.  The gospels are the complex record of the reactions of communities to conditions that are extremely difficult to assess.  Even though Carrier may know and accept this premise, he finds it unimportant to address its consequences.

It may be that in further work Carrier will lay the theoretical groundwork, justifying his use of Bayes as a cipher for understanding the gospels.  But even if his mission is not that–even if it’s just a game-playing exercise for debunking their historicity in front of believers–it seems to me that Ayez has raised a fatal objection:  Bayes is for apples and the gospels are oranges. And Carrier’s persistent defense that no one is really on the same page–or able to “get” the page he’s writing–is becoming increasingly difficult to swallow.