This is a preview of subscription content, log in via an institution to check access.
Notes
For a very helpful historical survey, see Chapter 1 of Robert Amico, The Problem of the Criterion, Rowman and Littlefield (Lanham, MD, 1993).
Though the claim is not explicitly made in the chapter, an assertion of this dependence is clearly intended. The intention is made clear in two places. First, it is explicitly assumed that skeptical arguments work by contending a circularity is involved in meeting some necessary condition for knowledge (p. 116, n.2). And second, a Cartesian approach is endorsed as a explanation of how the (PC1) condition on knowledge is met. As the conclusion of a summary of the Cartesian approach, the point is made that it is not vitiated by any circle (p. 132). Thus, some closed circle of requirements must loom. Part of the problem must be the claim that we have to rely on already knowing full well in order to justify the reliability belief.
I discuss more extensively the threat from the original problem to meeting the justification condition on knowledge in Part 2 of the “Getting Started” section of “First Things First,” in Earl Conee and Richard Feldman, Evidentialism: Essays in Epistemology (Oxford, Oxford University Press) 2004. (Part 1 of the “Getting Started” section discusses a different version of the problem of the criterion, attributed there to Roderick Chisholm. It is a problem for acquiring knowledge of knowledge.)
In the cited note 5, the plausibility of principles like (PC1) is traced to Ascent and a principle called “Transfer.” There is no such labeled principle in the chapter. Closure is clearly the other principle intended (I speculate that “Closure” is a re-labeling of Transfer, re-labeled to avoid any suggestion that the principle implies a transfer of justification from antecedent to consequent.).
Another principle of the criterion is formulated (PC2). It is much like (PC1), except that (PC2) it does not require any consideration (pp. 121–122).
The characterization in the chapter of foundationalism is this. “[S]ome ultimate reasons are justified non-inferentially, are justified in a way that does not require the support of some ulterior reasons” (p. 123). This claim about ultimate reasons is indirectly related to BonJour’s Generalization, which is about fully justified belief. No Justified Support is a foundationalist thesis along much the same lines that more directly engages with BonJour’s Generalization.
In fact the support seems to go in the other direction. Justified Probability implies, with the assumption that whatever is known full well is fully justified, that we our have justification for the probability of whatever we know full well. This implication at least mildly assists the claim of Ascent that reflection on what is known full well implies justification for its being known full well. Likewise, by implying justification for the probable truth of the fully justified, Justified Probability at least mildly assists the claim of Closure that the fully justified implications of the fully justified are themselves fully justified.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Conee, E. Criterial problems. Philos Stud 143, 417–426 (2009). https://doi.org/10.1007/s11098-009-9341-6
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11098-009-9341-6