prior) knowledge of such a positive indication. How in that case would one know of one’s wakefulness? Perhaps one would know it by believing it through the exercise of a reliable faculty. Perhaps one would know it through its coherence with the rest of one’s comprehensive and coherent body of beliefs. Perhaps both. But, it may be urged, if these are the ways one might know of one’s wakefulness, does not this answer commit us to a theory of the form of A below? (A) The proposition that p is something one knows (believes justifiably) if and only if one satisfies conditions C with respect to it. And if so, are we not caught in a vicious circle by the question as to how we know – what justifies us in believing – (A) itself? This is far from obvious, since the requirement that we must submit to some test procedure for wakefulness and know ourselves to test positively, before we can know ourselves to be awake, is itself a requirement that seems to lead equally to a principle such as (A). At least it is not evident why the proposal of the externalist or of the coherentist as to how we know we are awake should be any more closely related to a general principle like (A) than is the (foundationalist?) notion that in order to know we are awake we need epistemically prior knowledge that we test positive in a way that does not presuppose already acquired knowledge of the external world. The problem of how to justify the likes of (A) is a descendant of the (in)famous ‘problem of the criterion,’ reclaimed in the sixteenth century and again in this century (by Chisholm, Theory of Knowledge, 1966, 1977, and 1988) but much used already by the Skeptics of antiquity under the title of the diallelus. About explanations of our knowledge or justification in general of the form indicated by (A), we are told that they are inadequate in a way revealed by examples like the following. Suppose we want to know how we know anything at all about the external world, and part of the answer is that we know the location of our neighbor by knowing the location of her car (in her driveway). Surely this would be at best the beginning of an answer that might be satisfactory in the end (if recursive, e.g.), but as it stands it cannot be satisfactory without supplementation. The objection here is based on a comparison between two appeals: the appeal of a theorist of knowledge to a principle like (A) in the course of explaining our knowledge or justification in general, on one side; and the appeal to the car’s location in explaining our knowledge of facts about the external world, on the other side. This comparison is said to be fatal to the ambition to explain our knowledge or justification in general. But are the appeals relevantly analogous? One important difference is this. In the example of the car, we explain the presence, in some subject S, of a piece of knowledge of a certain kind (of the external world) by appeal to the presence in S of some other piece of knowledge of the very same kind. So there is an immediate problem if it is our aim to explain how any knowledge of the sort in question ever comes to be (unless the explication is just beginning, and is to turn recursive in due course). Now of course (A) is theoretically ambitious, and in that respect the theorist who gives an answer of the form of (A) is doing something similar to what must be done by the protagonist in our car example, someone who is attempting to provide a general explanation of how any knowledge of a certain kind comes about. Nevertheless, there is also an important difference, namely that the theorist whose aim it is to give a general account of the form of (A) need not attribute any knowledge whatsoever to a subject S in explaining how that subject comes to have a piece of knowledge (or justified belief). For there is no need to require that the conditions C appealed to by principle (A) must be conditions that include attribution of any knowledge at all to the subject in question. It is true that in claiming that (A) itself meets conditions C, and that it is this which explains how one knows (A), we do perhaps take ourselves to know (A) or at least to be justified in believing it. But if so, this is the inevitable lot of anyone who seriously puts forward any explanation of anything. And it is quite different from a proposal that part of what explains how something is known or justifiably believed includes a claim to knowledge or justified belief of the very same sort. In sum, as in the case of one’s belief that one is awake, the belief in something of the form of (A) may be said to be known, and in so saying one does not commit oneself to adducing an ulterior reason in favor of (A), or even to having such a reason in reserve. One is of course committed to being justified in believing (A), perhaps even to having knowledge that (A). But it is not at all clear that the only way to be justified in believing (A) is by way of adduced reasons in favor of (A), or that one knows (A) only if one adduces strong enough reasons in its favor. For we often know things in the absence of such adduced reasons. Thus consider one’s knowledge through memory of which door one used to come into a room that has more than one open door. Returning finally to (A), in its case the explanation of how one knows it may, once again, take the form of an appeal to the justifying power of intellectual virtues or of coherence – or both. Recent accounts of the nature of thought and representation undermine a tradition of wholesale doubt about nature, whose momentum is hard to stop, and threatens to leave the subject alone and restricted to a solipsism of the present moment. But there may be a way to stop skepticism early – by questioning the possibility of its being sensibly held, given what is required for meaningful language and thought. Consider our grasp of observable shape and color properties that objects around us might have. Such grasp seems partly constituted by our discriminatory abilities. When we discern a shape or a color we do so presumably in terms of a distinctive impact that such a shape or color has on us. We are put systematically into a certain distinctive state X when we are appropriately related, in good light, with our eyes open, etc., to the presence in our environment of that shape or color. What makes one’s distinctive state one of thinking of sphericity rather than something else, is said to be that it is a state tied by systematic causal relations to the presence of sphericity in one’s normal environment.
A light now flickers at the end of the skeptic’s tunnel. In doubt now is the coherence of traditional skeptical reflection. Indeed, our predecessors in earlier centuries may have moved in the wrong direction when they attempted a reduction of nature to the mind. For there is no way to make sense of one’s mind without its contents, and there is no way to make sense of how one’s mind can have such contents except by appeal to how one is causally related to one’s environment. If the very existence of that environment is put in doubt, that cuts the ground from under one’s ability reasonably to characterize one’s own mind, or to feel any confidence about its contents. Perhaps, then, one could not be a ‘brain in a vat.’ Much contemporary thought about language and the requirements for meaningful language thus suggests that a lot of knowledge must already be in place for us to be able to think meaningfully about a surrounding reality, so as to be able to question its very existence. If so, then radical skepticism answers itself. For if we can so much as understand a radical skepticism about the existence of our surrounding reality, then we must already know a great deal about that reality.
See also ACADEMY , CLOSURE , DESCARTES, EPISTEMOLOGY , FOUNDATIONALISM , JUSTIFI — CATION , SKEPTIC. E.S.