Descartes, and many of the other great epistemologists of the modern period, looked to epistemology to put science and intellectual inquiry generally on a secure foundation. Epistemology’s role was to provide assurances of the reliability of properly conducted inquiry. Indeed, its role was nothing less than to be czar of the sciences and of intellectual inquiry in general. This conception of epistemology is now almost universally regarded as overly grandiose. Nonetheless, Descartes and the other great epistemologists of the modern era were not completely mistaken. Epistemology does have a foundational role to play, but not that of a guarantor of knowledge. Its role, rather, is the less flamboyant but nonetheless theoretically crucial one of providing a philosophically respectable foundation for a general theory of rationality.

We can describe languages in which no words refer to objects. Such languages may contain sentences equivalent to any sentences of English, and hence may allow for as much objectivity as English does. It is wrong to try to deal with such languages by claiming that there are more objects than those accepted by common sense ontology. The correct move is rather to acknowledge a sense in which the concept of an object might have been different. A consequence of this position is that we cannot have a general semantics applicable to every describable language in which words are referentially connected to objects. The point here is not that reference may be inscrutable, but that different concepts of ‘referring to an object’ may be required for different languages.

As the Pyrrhonians made clear, reasons that adequately justify beliefs can have only three possible structures: foundationalism, coherentism, and infinitism. Infinitism—the view that adequate reasons for our beliefs are infinite and non-repeating—has never been developed carefully, much less advocated. In this paper, I will argue that only infinitism can satisfy two intuitively plausible constraints on good reasoning: the avoidance of circular reasoning and the avoidance of arbitrariness. Further, I will argue that infinitism requires serious, but salutary, revisions in our evaluation of the power of reasoning. Thus, reasoning can not provide a basis for assenting to a proposition—where to assent to a proposition, p, means to believe that we know that p. A non-dogmatic form of provisional justification will be sketched. Finally, the best objections to infinitism, including those posed by the Pyrrhonians, will be shown (at least provisionally!) to be inadequate.

The most powerful skeptical challenge to knowledge and justification is Pyrrhonian. It challenges nonskeptics to identify non-question begging warrant for their beliefs whereby they will not simply assume a point needing support in light of skeptical questions. The skeptical challenge is comprehensive, bearing on warranting conditions in general. Any answer given to such a comprehensive challenge apparently relies on a warranting condition being questioned. From this two questions emerge. First, is the skeptical challenge itself question begging in a way that undermines its epistemic significance? Second, is question begging necessarily an epistemic defect? This paper answers no to the first question, and identifies the problem facing skeptics who presuppose an affirmative answer to the second question. The problem stems from the availability of certain conceptions of epistemic rationality that do not prohibit question begging unqualifiedly.