Fred Dretske

(1932 - 2013)

Melnyk complains that my account of the way reasons explain behaviour cannot be extended to cover novel behaviours. I admit that I did not extend it, but deny that it is not extendible. This, indeed, is what Chapter 6 of Dretske (1988) was all about. Noordhof finds faults with my account and claims there is another account (partial supervenience) that does a better job. I acknowledge one of the defects—a defect I was aware of when I wrote the book‐but deny that the partial supervenience of content on intrinsic properties represents a better theory of the explanatory role of content.

Materialist explanations of cause and effect tend to embrace epiphenomenalism. Those who try to avoid epiphenomenalism tend to deny either the extrinsicness of meaning or the intrinsicness of causality. I argue that to deny one or the other is equally implausible. Rather, I prefer a different strategy: accept both premises, but deny that epiphenomenalism is necessarily the conclusion. This strategy is available because the premises do not imply the conclusion without the help of an additional premise—namely, that behavior explained by reasons is caused by the reasons that explain it—and this premise is false.

There are two ways to think about knowledge: From the bottom-up point of view, knowledge is an early arrival on the evolutionary scene; it is what animals need in order to coordinate their behavior with the environmental conditions. The top-down approach, departing from Descartes, considers knowledge constituted by a justified belief which gains its justification only in so far as the process by means of which it is reached conforms to canons of sciemific inference and rational theory choice. Keith Lehrer's epistemology is analyzed as a top-down intemahst position and criticised with examples that show that in certain cases obviously knowledge is attained without meeting the standards for reliability of sources and processing of the required informantion.