Chisholm and coherence

✍ Scribed by Richard Foley

• Publisher: Springer Netherlands
• Year: 1980
• Tongue: English
• Weight: 580 KB
• Volume: 38
• Category: Article
• ISSN: 0031-8116
• DOI: 10.1007/bf00354525

✦ Synopsis

Perhaps the dominant theory of empirical justification today is the coherence theory. To be sure there are objectors who argue that relations of coherence are not sufficient to understand empirical justification. However, even these objectors have generally conceded that relations of coherence are at least necessary in order to understand justification. Indeed, Roderick Chisholm as well as many other of the most famous noncoherentists of this century have claimed that a principle of coherence is needed in order to give a complete account of justification. 1 And not surprisingly, some coherentists have seized upon this purposed necessity to help defend coherence theories. Laurence Bonjour, for example, has recently claimed that any lack of clarity concerning what coherence might be is not a particularly serious problem for coherentists, because any adequate theory of justification will include a principle of coherence and thus any adequate theory of justification will be afflicted with the same unclarity. 2

Against this prevailing view, I want to suggest that coherence relations are not needed in order to provide an adequate account of justification. However, I will not argue for this particular claim. Instead, I will concentrate my attention on the slightly less controversial claim that, contrary to what the most prominent of foundationalists seem to think, foundationalist theories of justification can be developed quite nicely without recourse to a principle of coherence.

Of course, in order to establish such a claim, it is necessary to have some idea of what coherence is, and unfortunately there are about as many different conceptions of coherence as there are epistemologists. However, everyone seems to agree that a set of beliefs is coherent only if the propositions believed are consistent. That is, it must be logically possible for the conjunction of such propositions to be true. The problem is in saying what in addition to consistency is needed, for coherence is ordinarily thought to be something more than consistency and something less than mutual deductibilityJ How-