An Evidentialist Turn in Plantinga's Epistemology — Epoche C2
The account at issue, and the reputation it has acquired Alvin Plantinga's Warrant and Proper Function of 1993 offers a four-condition analysis of what has to be added to true belief to yield knowledge, and none of the four conditions mentions evidence. A belief has warrant for a subject, on that analysis, if and only if it is produced by cognitive faculties that are functioning properly; in a cognitive environment sufficiently similar to the one for which those faculties were designed; according to a segment of the design plan that is aimed at the production of true belief rather than at some other benefit; and where beliefs produced in those circumstances have a high objective probability of being true. Plantinga adds that warrant comes in degrees, varying with the firmness with which the belief is held. On the strength of this analysis Reformed epistemology has been read for three decades as the standing alternative to evidentialism in Anglo-American philosophy of religion: a way of holding that religious belief can amount to knowledge without being based on argument. The contention here is that this reading, while not wrong about the letter of the four conditions, misdescribes where the disagreement lies, and that Plantinga's system contains an internalist component which he himself names and concedes. Establishing that requires three things the earlier and more compressed version of this argument omitted: what Plantinga is actually attacking, since 'evidentialism' names his target only loosely; what the word 'evidence' is doing in each of the positions being compared, since the compressed version's four-way classification conflated two independent distinctions; and where in Plantinga's own text the concession is made. What Plantinga is attacking, stated as an argument rather than a label The evidentialist objection to theistic belief, in the form Plantinga addresses, has a premise that is rarely made explicit by those who use it, and his strategy is to make it explicit and then refuse it. The objection runs: belief in God is rational only if there is sufficient propositional evidence for it; there is not; therefore such belief is irrational. Its canonical statement is W. K. Clifford's essay of 1877, which opens with a shipowner who quiets his doubts about a vessel's soundness without inspecting it, and concludes with the maxim that it is wrong, always and everywhere and for anyone, to believe anything on insufficient evidence. Clifford's maxim is deontological in form: it makes belief the subject of a duty, and a believer who breaches it is not merely mistaken but culpable. Plantinga's reply, given at length in his 1983 essay 'Reason and Belief in God', does not dispute the second premise. It attacks the first, by identifying the epistemology that makes it look obvious. That epistemology is classical foundationalism, which holds that a proposition is properly basic for a person — believable without being inferred from anything else — only if it is self-evident, or incorrigible, or evident to the senses, and that every other belief must be supported by inference from propositions of that kind. Belief in God is none of the three, so on this criterion it needs an argument. The refutation is a self-referential one, and it is worth setting out because its form determines what survives it. Consider the criterion itself: the proposition that a belief is properly basic only if it is self-evident, incorrigible, or evident to the senses. It is not self-evident, since competent philosophers have denied it. It is not incorrigible, since one can be mistaken about it. It is not evident to the senses, since it is not the sort of thing the senses report. And no one has produced a derivation of it from propositions that are. So by its own lights the classical foundationalist may not hold it either basically or on the basis of anything else — the criterion fails the test it imposes. The conclusion Plantinga draws is not that criteria of proper basicality are impossible but that they must be arrived at inductively, from examples of beliefs we take ourselves to be entitled to hold basically; and once the procedure is that one, the theist who takes belief in God as an example is proceeding no differently from the foundationalist who takes memory beliefs as one. The target, then, is not evidentialism as such but a particular criterion of proper basicality plus the deontological framing that turns a failure to meet it into a fault. Plantinga's companion volume of the same year, Warrant: The Current Debate , prosecutes the second half of this: it traces the identification of epistemic justification with the fulfilment of epistemic duty to Descartes and Locke and through Chisholm, and argues that justification so conceived is neither necessary nor sufficient for warrant. The demonstration that it is insufficient is what matters below. A subject may violate no epistemic duty whatever — may be doing, with complete conscientiousness, everything a rational person could be asked to do — and still hold a belief that is not knowledge, because something has gone wrong in the machinery that produced it. Plantinga's illustration is a lesion in the brain which, among its other effects, causes its victim to believe that he has a brain lesion. The belief is true; the process that produces it is, in the narrow sense, perfectly reliable, since it never yields a false verdict; the victim is blameless, having no way to detect the malfunction. It is still not knowledge, and what it lacks is proper function. Four labels, and the two distinctions they run together The compressed version of this essay distinguished strong, weak, internalist and externalist evidentialism as four species. They are not four species; they are the cells of two independent distinctions, and treating them as a list makes the subsequent argument look weaker than it is by hiding which cell Plantinga occupies. The first distinction concerns what may coun