What was later called expressivism about ethics, and what we can call Wittgensteinian approaches to religion, had their origin in the same place: empiricist theories of meaning, which ruled out from “descriptive” discourse anything that was not empirically verifiable. In both cases, not everybody was willing to consign the discourse to the oblivion of pure nonsense; rather, the speech acts of asserting within the discourse were reinterpreted, and attention focused on the function of ethical or religious language, rather than its truth conditions. In ethics, this general viewpoint evolved into the work of Blackburn, Gibbard, Horgan & Timmons, et al, while in religion it had a good mid-century run as “Wittgensteinian philosophy of religion” and became ensconced in liberal Protestant theology.
I think it is fair to say (I’m willing to be corrected on this point) that most philosophers, and not least philosophers of religion, no longer regard Wittgensteinian approaches as the way to go in philosophy of religion. Rather, there are a lot of philosophers who think there is no god, and a fair number who think there is, and both are pretty clear about the truth-conditions of their claims. Reinterpreting obviously theistic discourse as not actually referring to a deity—as was characteristic of the Wittgensteinian approach—would likely be taken, among philosophers these days, as a form of unreasonable obscurantism, or perhaps bad faith. Since both sides of the debate now think of (some) religious discourse as descriptive, we ought to ask what made non-descriptive approaches so popular for a while, and why they fell out of favor. In particular, I wonder why they fell out of favor in religion much more precipitously than they have in ethics, where non-descriptive approaches are still very much in vogue.
It can’t simply be that, for a while, no one had realized the technical hurdles that non-descriptive approaches present to the semanticist, and when they did realize this the Wittgensteinians dropped their project, because the hurdles are no harder in religion than they are in ethics and the expressivist project in ethics is still going strong. In fact, I don’t think anyone has pursued Frege-Geach style objections to Wittgensteinian philosophy of religion, though clearly they could. There is some other explanation at work.
Let me try a sociological explanation. (I should be careful to note that I am not one of those who think that explanations of beliefs in terms of non-rational factors are necessarily derogatory explanations.) There was a while, in the middle of the twentieth century, when respect for and attachment to religious institutions was high among educated people, while belief in their metaphysical underpinnings was low. This produced cognitive dissonance, which was resolved by turning to non-descriptive ways of understanding religious discourse. The alternative was thinking of too many of your friends and relatives and colleagues, not to mention other cultures, as whacked; also, perhaps, giving up a practice you were happy to continue in. But as these sociological obstacles to thinking of religious discourse as straightforwardly false have disappeared, so has the attraction to Wittgensteinian philosophy of religion.
Suppose something like this is the right story. Turn now to expressivism in ethics. The parallel explanation (which, again, I do not intend derogatorily) would be that expressivism is what you get when people remain attached to the practice of making moral/normative/evaluative claims, but have lost belief in any underlying metaphysics that would make those claims true. Unlike religious discourse, however, there is no chance whatsoever that people are ever going to disentangle themselves altogether from normative/evaluative discourse. So they continue to resolve their cognitive dissonance, aka philosophical problem, with expressive interpretations of ethical claims.
I am not sure what, if anything, these ruminations provide an argument for. Maybe it breaks down like this. The parallels between the cases of ethics and religion support moral descriptivism: we’ve come around to descriptivism about religious discourse, so we should about ethical discourse too. The differences in the two cases—in particular, the impossibility of abandoning ethics in the way one might abandon religion, even if/when one thinks that the metaphysical status of the two realms is on a rough par—support moral expressivism of some kind.
Thoughts?
Broadly expressivist approaches to religion have deeper sources that have nothing to do with empiricist conceptions of meaning nor any naturalistic world-view. Consider Erasmus’ privileging piety over rational theology, to take just one prominent example. For Erasmus, religion wasn’t a primitive cosmological theory the way contemporary philosophers sometimes tend to conceive of it; rather religious practice was a matter of a perspective on the world that importantly involved having the appropriate sentiments. Early twentieth century expressivist treatments of religion at least had the virtue (whatever its other limitations) of emphasizing the affective dimension of religious faith.
Expressivist treatments of religion needn’t be the result of cognitive dissonance. Perhaps it was for Ayer, but I doubt it was for Wittgenstein, even given his evident religious ambivalence.
Moreover, I suspect that it is the difficulty that we moderns have in sustaining core theological sentiments (such as regarding things as sacred or holy, at least literally) that is the real source of sustaining conviction in an expressivist treatment of religion. Notice there is less of a correlative problem in our sustaining core ethical sentiments, thus making ethical expressivism (whatever technical problems it may face) at least a plausible alternative.
I don’t think theological noncognitivism is anywhere near as dead as the post seems to suggest. Indeed, cries of “you’re being too literal, it’s all just a beautiful metaphor” have positively bedeviled contemporary atheist popularizers like Sam Harris and Richard Dawkins whenever they give interviews or hold public speaking engagements. The difference is that sophisticated Wittgenstinian mid-20th-century philosophers were open and explicit about what they took their religious claims to amount to — and this has gone out of style — while the doctrine as it survives today is held in reserve by liberal Christians and Jews, to be deployed as a “rescue operation” in the event that some such busybody starts challenging their claims. Religion becomes nonliteral at precisely the instant it would otherwise be forced to make contact with the evidence. Even Hilary Putnam has waffled on this.
Another contributing and complementary sociological factor to LW’s own formulation in terms of philosophy of language is the latent movement of demythologization that had already been going on among German theologians like Rudolf Bultmann and Paul Tillich — who were, for all practical purposes, atheists. Since we’re doing cultural anthropology here, it may be helpful to employ another methodological caveat from that discipline: I suspect that LW’s “informants” when he looked about the European Academy to hear what this “religion” thing was all about were precisely these kind of theologians, cut off from the hoi polloi. It’s not at all surprising given this that he might come to the conclusion that expressions of awe and wonder were what religion was “really all about”
I think that error theory is not inconsistent with our first order moral practice and even the practices we use to pick our moral principles. Maybe we can abandon our belief in ethics without facing the dire consquences you seem to expect.