Harman, Gilbert (1938–)

views updated

HARMAN, GILBERT
(1938–)

Gilbert Harman was born in 1938, graduated from Swarthmore College in 1960, and received his PhD from Harvard in 1963, where W. V. Quine was his dissertation advisor. He is distinctive in being a leading contributor across a broad range of subdisciplines of philosophy: epistemology, ethics, philosophy of language, philosophy of mind, and metaphysics. This entry reviews only a few of his many important contributions.

Harman has been perhaps the most significant contemporary defender of moral relativism. According to Harman, moral right and wrong are akin to motion: They are relative to a framework and no framework is privileged. Harman appeals effectively to two sorts of consideration in developing his position. First, like J. L. Mackie, he is impressed by the degree of moral diversity across and even within populations. Second, complementarily, Harman defends moral naturalism—the view, roughly, that morality is fundamentally continuous with the natural sciences.

In rejecting moral nonnaturalism, Harman claims that the postulation of nonnatural moral properties is unjustified: They would be explanatorily impotent. Could not a sui generis fact of, say, torture's being bad explain, at least, our belief that torture is bad? Harman argues that there are other better ways of explaining such a belief, in terms of conventions and other social arrangements—arrangements to which we are normally exposed as we develop.

Holding that moral disagreement is widespread, Harman accordingly argues—as a kind of inference to the best explanation—to the conclusion that there are no absolute moral facts, beyond the facts about what holds relative to one or another framework. Different people can, without ignorance (or related independently specifiable failings), find one or another ultimate moral demand inapplicable in their own case. But because a moral demand is said to apply only if the agent either accepts it or rejects it only out of ignorance, then ultimate moral demands may apply only selectively, to some agents and not to others.

A related aspect of Harman's moral relativism is thus his motivational internalism: If morality is understood as the product of a framework constituted by psychological states (and, in particular, by the agreement, plans, and conventions emerging therefrom), then it will be easier to understand how morality could have that motivational force.

It should be noted, too, that, although he distinguishes them, Harman embraces forms of each of normative moral relativism, moral judgment relativism, and metaethical relativism. Normative moral relativism holds, roughly, that people can be subject to different ultimate moral demands. Moral judgment relativism claims, in effect, that moral judgments implicitly refer to a person, group, or set of moral demands. And according to metaethical relativism, conflicting moral judgments about a particular case can in a way both be right.

In epistemology, Harman has long defended a view that has elements of foundationalism and elements of coherentism. Harman's is a kind of "foundations" theory in which everything a person accepts at a given time is foundational and needs no justification except when there are conflicts. Accordingly, knowledge is best understood "when skepticism is turned on its head": starting from what we know, we diagnose what goes wrong with arguments for radical skepticism. For Harman, a key insight is that knowledge is essentially inferential: Inference is a matter of increasing the coherence of one's overall state—reasoning consists in trying to obtain a reflective equilibrium (though he is concerned about possible instabilities in this process)—and coherence is partly a matter of explanation.

If inference to the best explanation is to have the central role in our cognition that it appears to have, it will have to be understood as a sort of explanatory inference. So Harman defends the adequacy of that sort of inference. In enumerative induction we generalize observed regularities; but according to Harman all such cases of induction are really cases of inference to the best explanation. So, at a minimum, inference to the best explanation is not in general any less legitimate than induction.

This emphasis on the role of inference in knowledge is related to Harman's focus on inference in its own right. He draws a sharp division between logic and inference. For Harman, there is, for example, no such thing as deductive inference and the search for an inductive logic is the product of confusion. Reasoning is change in view; logic is the theory of implication.

Consider modus ponens. This exceptionless rule of logic cannot serve as a principle about how to change one's view: sometimes, when one believes P and believes that if P, then Q, what one should do is to give up one's belief that P. Moreover, although according to the rules of logic inconsistent premises imply any proposition, it's not the case that inconsistent beliefs permit one to infer any proposition. The distinction between inference and logic coheres well with Harman's views about the fundamentally explanatory character of inference to the best explanation (see above).

In philosophy of mind, Harman was a seminal proponent of what has come to be known as intentionalism about experiential states. Harman holds that there is no phenomenal difference between such states without an intentional difference. They have accordingly come to be viewed as individuated by their representational or intentional character: what makes the state what it is is that it is about what it is about. That perceptual experience should not be understood as individuated by (what others would call) its qualitative character instead is seen as sustained by the transparency of experience. Introspection does not seem to reveal the nature of experiences themselves, only that of their external objects.

This brief review cannot do justice to the richness and range of Harman's work. We have not so much as touched on his important discussions of meaning and analyticity (and his brilliant exposition—really, development—of Quine's views on the subject), on his contributions to conceptual-role functionalism in philosophy of mind, or on his more recent argument that work in social psychology supports eliminativism about the central posits of any virtue theory. Still, some sense of the scope and significance of Harman's contributions should have emerged.