certainty the property of being certain, which is either a psychological property of persons or an epistemic feature of proposition-like objects (e.g., beliefs, utterances, statements). We can say that a person, S, is psychologically certain that p (where ‘p’ stands for a proposition) provided S has no doubt whatsoever that p is true. Thus, a person can be certain regardless of the degree of epistemic warrant for a proposition. In general, philosophers have not found this an interesting property to explore. The exception is Peter Unger, who argued for skepticism, claiming that (1) psychological certainty is required for knowledge and (2) no person is ever certain of anything or hardly anything. As applied to propositions, ‘certain’ has no univocal use. For example, some authors (e.g., Chisholm) may hold that a proposition is epistemically certain provided no proposition is more warranted than it. Given that account, it is possible that a proposition is certain, yet there are legitimate reasons for doubting it just as long as there are equally good grounds for doubting every equally warranted proposition. Other philosophers have adopted a Cartesian account of certainty in which a proposition is epistemically certain provided it is warranted and there are no legitimate grounds whatsoever for doubting it. Both Chisholm’s and the Cartesian characterizations of epistemic certainty can be employed to provide a basis for skepticism. If knowledge entails certainty, then it can be argued that very little, if anything, is known. For, the argument continues, only tautologies or propositions like ‘I exist’ or ‘I have beliefs’ are such that either nothing is more warranted or there are absolutely no grounds for doubt. Thus, hardly anything is known. Most philosophers have responded either by denying that ‘certainty’ is an absolute term, i.e., admitting of no degrees, or by denying that knowledge requires certainty (Dewey, Chisholm, Wittgenstein, and Lehrer). Others have agreed that knowledge does entail absolute certainty, but have argued that absolute certainty is possible (e.g., Moore). Sometimes ‘certain’ is modified by other expressions, as in ‘morally certain’ or ‘metaphysically certain’ or ‘logically certain’. Once again, there is no universally accepted account of these terms. Typically, however, they are used to indicate degrees of warrant for a proposition, and often that degree of warrant is taken to be a function of the type of proposition under consideration. For example, the proposition that smoking causes cancer is morally certain provided its warrant is sufficient to justify acting as though it were true. The evidence for such a proposition may, of necessity, depend upon recognizing particular features of the world. On the other hand, in order for a proposition, say that every event has a cause, to be metaphysically certain, the evidence for it must not depend upon recognizing particular features of the world but rather upon recognizing what must be true in order for our world to be the kind of world it is – i.e., one having causal connections. Finally, a proposition, say that every effect has a cause, may be logically certain if it is derivable from ‘truths of logic’ that do not depend in any way upon recognizing anything about our world. Since other taxonomies for these terms are employed by philosophers, it is crucial to examine the use of the terms in their contexts.
See also EPISTEMOLOGY, JUSTIFICATION, SKEPTICIS. P.D.K.