Plantinga's Proper Function and the Evidential Landscape of Belief — Epoche C2
The charge against Alvin Plantinga's religious epistemology is easy to state: by holding that belief in God can be properly basic — held rationally without being inferred from anything else one believes — he has excused it from the demand for evidence, and Reformed epistemology is therefore fideism, the view that religious belief answers to faith alone and not to reasons, written out in analytic prose. The charge is wrong, but it is not stupid, and the standard rebuttals are too quick. What Plantinga actually does is not to dismiss evidence but to move the evidential question twice: from the content of a belief to its aetiology, and from epistemology to metaphysics. Both moves are defensible, both are costly, and the second is the one his critics ought to press. What "warrant" is, and what the word is not doing Everything turns on a term of art, so it is worth being exact about it first. Plantinga defines warrant stipulatively as whatever quantity it is that, added in sufficient measure to true belief, yields knowledge. The definition is deliberately empty at the point of substance — it does not say that the quantity is justification, or reliability, or evidential support — and so does not prejudge the debate it is meant to settle. His substantive answer, in Warrant and Proper Function (1993), is a conjunction of four conditions. A belief $B$ has warrant for a person $S$ when, and only when: the cognitive faculties that produced $B$ were functioning properly — that is, not subject to dysfunction, damage or disease; the cognitive environment in which they operated is sufficiently similar to the environment for which they were designed; the segment of the design plan governing the production of $B$ is aimed at the production of true belief, rather than at survival, or comfort, or the ability to go on; the design plan is a good one, in the sense that there is a high objective probability that a belief produced in that way, in that environment, is true. Condition (3) does more work than it appears to. The optimism of the gravely ill patient who believes he will recover may well be produced by a mechanism selected for precisely because such belief aids recovery; the belief can then be true, and properly produced, and still fail to be knowledge, because the mechanism is not in the truth business. Warrant also comes in degrees, increasing with the firmness of the belief. One thing must be corrected here, because compressed versions of this argument — including the earlier version of this one — commonly say that the design plan is divine. It is not part of Plantinga's analysis of warrant that it is. The analysis is officially neutral about the source of the plan, and the neutrality is load-bearing rather than diplomatic: if "proper function" simply meant "functioning as God intended", the claim that theistic belief satisfies conditions (1) to (4) would presuppose theism, and could not be offered to anyone not already convinced. Plantinga argues separately , later in the same book, that naturalistic reductions of proper function to selection history fail and that naturalism conjoined with evolutionary theory is self-defeating; those are further theses, and one may reject them while keeping the four conditions. Why the demand for evidence had to be dismantled first The objection Plantinga takes as his target is not a vague suspicion of religion but a specific structure, which he traces through Locke to W. K. Clifford's dictum of 1877 that it is wrong, always and everywhere, to believe anything upon insufficient evidence. That dictum has teeth only when paired with an account of what may be believed without evidence, and the account historically supplied is classical foundationalism: a proposition is properly basic for a person only if it is self-evident to him, or incorrigible for him (a report of his own current mental states), or evident to his senses; everything else must be believed on the basis of propositions that are. The proposition that God exists belongs to none of the three basic categories, so if the theistic arguments also fail, the believer stands convicted of an intellectual fault. Plantinga's reply in "Reason and Belief in God" (1983) is a single argument, worth stating in full because the whole later structure rests on it. Let (F) be the foundationalist's criterion itself: a proposition is properly basic for a person only if it is self-evident, incorrigible, or evident to the senses . Apply (F) to (F). It is not self-evident: competent philosophers have denied it, which is not something that happens to genuinely self-evident propositions. It is not incorrigible, being no report of anyone's mental states. It is not evident to the senses. So by its own lights (F) is not properly basic, and whoever accepts it must accept it on the basis of propositions that are properly basic by its standard — and no such derivation has been produced. The criterion is self-referentially incoherent: anyone who believes it violates it. The result is narrower than it is usually taken to be: not that evidentialism is false, nor that there is no criterion of proper basicality, but that the particular criterion which made "believed without argument" equivalent to "irrational" cannot be sustained. The question of what may properly be believed in the basic way is thereby reopened, not answered — which invites the obvious retort. Plantinga names it himself: the Great Pumpkin objection. If belief in God may be properly basic, why not belief in the Great Pumpkin, who returns to the pumpkin patch every Hallowe'en? The reply has two parts. Criteria of proper basicality are arrived at as criteria in philosophy generally are, by assembling clear cases and framing hypotheses that fit them; one need not possess a general criterion before rejecting a candidate, any more than the foundationalist needed one before rejecting groundless beliefs about tomorrow's weather. And proper basicality is indexed to persons and circ