Duncan Pritchard is Chancellor’s Professor of Philosophy at the University of California, Irvine, and Professor of Philosophy at the University of Edinburgh. We invited him to answer the question “What norms or values define excellent philosophy of religion? as part of our “Philosophers of Religion on Philosophy of Religion” series.

Wittgenstein taught us that the resolution to philosophical difficulties often lies in questioning the presuppositions that generate the puzzle in the first place. It can seem that our everyday practices are generating contradictions, but in fact these conundra are the result of the illicit import of dubious philosophical claims. Once these philosophical claims are eliminated, the puzzle disappears.

Now one might think that philosophers of religion are not well-placed to exploit such a methodology, given that religious conviction is no longer orthodoxy (at least in the western world anyway), and hence isn’t obviously ‘everyday’ anymore. But the methodology is nonetheless applicable, at least in certain cases. I want to canvass support for one such case, which is the debate regarding the rationality of religious belief.

Here is a common way of setting up the issue. Religious belief, unlike other kinds of belief that are epistemically paradigmatic (like everyday perceptual belief, for example), presupposes basic religious commitments (such as in God’s existence). Those basic religious commitments, however, lack any independent rational basis. Hence, there is a fundamental problem with the rationality of religious belief.

Standard responses to the problem of rationality of religious belief effectively concede this set-up. This is clearest with evidentialist responses to this problem, for example, since they respond by arguing that there is an evidential basis for the basic religious convictions in play (e.g., that there are logical proofs for God’s existence). But other standard responses to this issue also concede the set-up, at least implicitly. Reformed epistemology, for example, agrees that basic religious belief is in need of an adequate epistemic grounding, but supplies this grounding via appeal to an externalist epistemology. Fideism, in contrast, agrees that basic religious belief lacks an adequate epistemic grounding, unlike belief in general, but argues nonetheless that this is not a bar against religious belief (since, unlike belief more generally, it does not stand in need of an epistemic grounding).

But suppose we question the framework behind this puzzle? Wittgenstein provides us with the resources to do just this. In his final notebooks, published as On Certainty, Wittgenstein argues for a radical conception of the nature of rational evaluation. A staple part of radical scepticism, at least since Descartes’ Meditations, has been to rationally evaluate all of our beliefs at once and find them wanting. In contrast, anti-scepticism has been characterised by the project of evaluating all our beliefs at once and thereby determining their positive epistemic status. Both projects seem entirely coherent. Wittgenstein argues, however, that they are chimeric.

Working through a series of examples, Wittgenstein shows that the very idea of a universal rational evaluation is simply incoherent, and not at all rooted in our ordinary epistemic practices (where all rational evaluation is essentially local). Rational evaluations instead always take place against a backdrop of basic ‘hinge’ commitments that are needed in order for rational evaluation for occur, and which as a consequence cannot be themselves rationally evaluated. Moreover, Wittgenstein shows that this is not an incidental feature of our epistemic practices, as if we could engage in fully general rational evaluations if only we were more consistent, imaginative, intelligent, and so on. Rather it is in the very nature of a rational evaluation that it takes place relative to these arational hinge commitments. It follows that both radical scepticism, and for that matter traditional forms of anti-scepticism, both trade on a faulty philosophical picture. In particular, they both presuppose the idea that fully general rational evaluations are coherent, something which Wittgenstein has shown to be in fact a dubious philosophical claim that is entirely disconnected from our everyday epistemic practices.

Notice how the Wittgensteinian claim alters our understanding of the supposed problem of the rationality of religious belief. That puzzle posed a challenge to show how basic religious belief could satisfy the epistemic standards that other, epistemically paradigmatic, forms of belief enjoy. So evidentialism is concerned with showing that basic religious belief enjoys an independent rational basis. Reformed epistemology contends that although basic religious belief lacks an independent rational basis, this doesn’t matter because (like other forms of belief) it enjoys an externalist epistemic basis that isn’t specifically rational. Fideism concedes that religious belief lacks an adequate epistemic basis, unlike epistemically paradigmatic forms of belief, but argues nonetheless that religious belief is not irrational. If Wittgenstein is right, then all of these responses to this puzzle about the rationality of religious belief are misguided.

In particular, all of these standard responses to the puzzle rest on the same mistake, which is to suppose that it is in the nature of epistemically paradigmatic everyday belief that it doesn’t presuppose ariatonal hinge commitments. Once we recognise that our everyday beliefs presuppose hinge commitments in this way, then that changes how we view the putative arationality of religious belief. It cannot now be a complaint against religious belief that it presupposes arational religious hinge commitments if it is true of all belief that it incorporates arational hinge commitments. Reformed epistemologists famously offer a parity argument in favour of religious belief, to the effect that belief of this kind is on an epistemic par with everyday belief, where the latter is to be reconceived along broadly epistemic externalist lines. What Wittgenstein is offering—which is a line of argument that I think he gets from earlier work by John Henry Newman, most notably his Essay in Aid of a Grammar of Assent—is a very different kind of parity argument. Religious belief, like belief more generally, can be fully rational even though, like everyday belief, it presupposes arational hinge commitments. The point is that there’s no need to bring in epistemic externalism, or indeed any kind of epistemic revisionism, in order to account for the rationality of religious belief. Rather, the very problem that was thought to afflict this kind of belief trades on an implausible account of our everyday epistemic practices. Once this faulty picture is rejected, then the puzzle dissolves.