Epistemology and the Psychology of Human Judgment [Paperback]

Great book! I am what the authors call a "Standard Analytic Epistemologist", and I'm finding it hard to escape their arguments. The book is intelligent, fair, and beautifully written. It is also funnier than most philosophy books (not very hard to do, I know). Epistemology and the Psychology of Human Judgment is not written to preach to those already converted to naturalism. Rather, it is an honest experiment in naturalism, in all of its multi-disciplinary splendor. In developing their version of naturalistic epistemology - called strategic reliabilism -- they expose contemporary epistemology for the insular field that it is. Highly recommended.

An extremely good book that describes what epistemology could be, as compared with the tedious academic exercise it has unfortunately become. As it stands, if traditional epistemology disappeared tomorrow, nobody outside of its small group of practitioners would notice or care.

Bishop & Trout's argument for a reorientation of epistemology is based in the "Aristotelian Principle" that in the long run, poor reasoning tends to lead to worse outcomes than good reasoning. That simple assertion licenses empirical testing of the relative goodness of competing reasoning strategies. It is also "a necessary precondition for the practical relevance of epistemology," because if better reasoning doesn't lead to better outcomes than bad reasoning does, then it wouldn't matter how we arrive at our beliefs, and epistemology in any form would be a pointless enterprise.

Reliabilism is the theory of epistemology that holds a belief to be justified if it results from some process of demonstrated reliability, one that has been shown to yield true beliefs. The purpose of this book is to make the case for something the authors call Strategic Reliabilism. SR is an epistemological theory that defines epistemic excellence as (1) efficient allocation of cognitive (reasoning, problem-solving) resources (2) to robustly reliable strategies, (3) applied to significant problems.

The goal is a more prescriptive, reason-guided epistemology, relevant to problem solving in the real world by offering practical recommendations about how to reason better and thus achieve better outcomes. It would be based in empirical research on the limitations and foibles of human reasoning and how to avoid or compensate for them. Strategic Reliabilism is not a theory of justification; its focus is on comparing reasoning strategies to identify the better ones so they can be used instead of ones that are less good. The importance of this effort can be seen in the enormous costs connected with poor reasoning. A patient with a positive result on a diagnostic test for cancer or AIDs needs good information about the likelihood that he actually has the disease. In fact, one study showed that only 18% of faculty and staff at Harvard Medical School were able to reason to the correct interpretation of a positive diagnostic test; the average estimated likelihood was 28 times too high.

"Ameliorative Psychology" is Bishop & Trout's term for the empirical discipline that searches for better reasoning strategies and, from the results of empirical tests, makes normative recommendations about how to reason better. The specific techniques described that have so far been found to improve reasoning include: Statistical Prediction Rules; thinking in terms of frequencies vs. probabilities in Bayesian problems; and recognition of common sources of bias and error (overconfidence, the interview effect, the generalized attribution error, lack of comparison in causal attribution, regression effects, leaping to plausible but ungrounded causal fictions for rare events). The many examples of these common problems in reasoning are highly interesting and troubling, especially as they show up in professionals who you assume would know better.

Traditional epistemology (referred to SAE: Standard Analytic Epistemology) emphasizes knowledge defined as justified true belief. But it provides no guidance about how to arrive at that worthy goal. SAE emphasizes a priori intuition as the method of evaluating clever examples involving strange hypothetical situations. B&T argue that these word games make little contact with real-world problems, that such intuitions are not, in fact, a priori, and that there is no evidence that the intuitions of philosophers produce trustworthy judgments about epistemic matters. SAE has failed to deliver.

This is a terrific book, one that offers a meaningful future for epistemology, joining it with scientific psychology to create a specialty with something of real value to offer the world. Whether it will be received that way by philosophers is another question entirely, although one can always dream.

This effort is a good offering for naturalistic epistemology as it wishes for the impossible (reasonable and practical modern epistemic outlook). Bishop & Trout beg the moral question as they assert that improper reasoning leads to bad results forasmuch as "bad" presupposes a standard that the authors, as naturalists, cannot provide. Epistemic tests to determine good or bad presuppose a fixed moral ground.

They offer fine critical judgments on a variety of epistemic approaches and if one doesn't guard one's pre-commitments too closely, one will be converted to their view.

This is an enjoyable read, even for non-philosophers, and it will expand anyone's epistemic horizons even if one disagrees with their rational (empirical) approach.