I've been reading Scott Atran's In Gods We Trust: The Evolutionary Landscape of Religion; I'm enjoying it immensely. The primary irritation created by the book is that I can make neither heads nor tails of some of the things Atran says. The most egregious offender is his account of quasi-propositions, which doesn't even appear on the surface to be coherent. It's possible, and even likely, that there just are some terms being used technically in ways that I'm not familiar with. To name an example, Atran talks about beliefs that are 'materially false'. Now, in philosophy, 'materially false' is a technical term: a proposition is materially false if it would be formally false taken materially. Thus, "I manipulated the object's center of gravity" is materially false, since centers of gravity are abstract objects that don't admit of manipulation. But, of course, it can be taken in a sense that is formally true (materially false is not formally false, and only formally false is actually false). But this is clearly not what Atran has in mind, so if I were to take it in that sense, that would be a merely verbal problem (I don't know what Atran means by 'materially false', but he certainly isn't using it in the standard technical sense). But with the account of quasi-propositions I am much more skeptical and not as inclined to pass it off as merely verbal. Some of the things he says:
* "Quasi-propositional beliefs may have the superficial subject-predicate structure of ordinary logical or factual propositions, but they can never have any fixed meaning because they are counterintuitive." (p. 113)
* they evoke logically and factually impossible worlds (p. 113)
* "a religious quais proposition p has no fully specifiable or fixed content, even in principle" (p.110)
* "necessarily counterfactual, because they violate innate, modular expectations about basic ontological categories" (p. 96)
* "inconsistent with commonsense knowledge" and "dramatically contradict basic commonsense assumptions" (p. 96)
* a quasi-proposition is at least analogous to Ayer's pseudo-propositions
* "only look like propositions insofar as they may take a subject-predicate form" (p. 290n4)
* "they are 'category violations'" (p. 290n4)
* "non-propositional" (p. 94)
* "like a metaphor in poetry" (p. 95)
* have truth value, but are not truth valuable [i.e., verifiable by observation or logical analysis] ('have truth value' could be understood as in 'believed to have truth value, but 'truth valuable' certainly has to be a fact about quasi-propositions themselves) (p. 95)
* "taken on faith and emotionally validated with little reasoning required for support" (p. 95)
* involves acceptance of what is "materially false" as true or what is "materially true" as false (p. 5)
* counterfactual insofar as they are anomalous, implausible, and counterintuitive (p. 4; indeed, he goes so far as to say that everyone implicitly knows it, p. 5)
Unless Atran is using terms in odd ways that I'm just not getting, this is a horrible mish-mash of logical inconsistency; to take just the most obvious example, if something is non-propositional it can't have truth value, if something is not truth valuable it can't be known to be counterfactual, if something only looks like a proposition it can't contradict anything, etc. Perhaps some of this is just due to looseness of phrasing; but it leaves me entirely unclear as to what these quasi-propositions are supposed to be. Under some characterizations, any first principle would seem to count as a quasi-proposition; under others, any quasi-proposition would seem to be false; under others, any quasi-proposition would seem to be senseless.
It also glosses over important facts, e.g., Ayer's account of pseudo-propositions is rarely held today because it collapsed in a miserable heap, unable to withstand serious analysis, and the principle of verifiability, which clearly makes a show here in Atran's discussion of truth valuability, is held by virtually no one because no coherent formulation has ever been found that is not obviously false (indeed, it is perhaps partly Atran's association of his account with logical positivism that inclines me to think his account is really inconsistent rather than attributing my puzzlement to verbal misunderstanding); science also regularly comes up with things that are implausible, anomalous, and counterintuitive in the sense of being schema-inconsistent (think of the discovery of meteorites or fossils; or just about all of modern cosmology); most religious propositions don't involve category violations in an Aristotelian sense; there are good reasons to think that not all violations of Aristotelian categories lead to unverifiable quasi-propositions; and so forth. These puzzles aren't of immense importance, I imagine, but they aren't quibbles, either: these are serious problems with the characterization of quasi-propositions given by Atran in the book, and would have to be distinguished from minor quibbles I would have (e.g., the Kierkegaard quote on p. 5 needs to be read against the background of the German Idealists who had views of absurdity that would generally be considered odd today; since I assume Atran is not advocating German Idealism, it doesn't actually support his claim) that don't affect the basic point, or interfere with how the argument runs. I'm also a little puzzled that, in this account of religious cognition, very little is said about actual philosophical and theological thought. It seems very doubtful to me, for instance, that Spinoza (to take one instance) can really be said to be assuming things as true on the basis of emotional validation rather than rational support. At least, it would take a very good argument to convince me otherwise. But that's a side issue.
Again, it's possible that I'm just missing something; but I don't see any way to salvage the account Atran seems to be giving.
Despite my bewildered complaint, I am, as I said, enjoying the book quite a bit. In a sense, the primary argument of the book is just that it's very difficult to provide a good cognitive account of religion; it's in looking at various aspects of this issue that Atran lays out his own view (which might possibly be a reason for the difficulty of reading him; it's like trying to figure out Malebranche's views from The Search after Truth). The discussion of counterintution itself, for instance, is quite fascinating, and I'm looking forward to the critique of memetic accounts.