Millikan, Ruth Garrett (1933–)

views updated

MILLIKAN, RUTH GARRETT
(1933–)

Born December 19, 1933, and raised in Swarthmore, Pennsylvania, where her father taught physics, Millikan received her Ph.D. from Yale University in 1969. She began her career as a self-described "faculty housewife," raising four children before publishing her first book. Internationally recognized, Millikan has made significant contributions to philosophy of biology, animal cognition, philosophy of language, mind, and ontology. A unifying theme is the importance of the fact that humans are products of evolution. (Millikan's mother held a Ph.D. in paleontology—perhaps influencing Millikan's orientation to Darwinism.) A student of Wilfred Sellars, Millikan rejects epistemic givens and takes meaning talk to have the function of helping speakers bring their use into conformity with others; unlike other Sellarsians, Millikan sees the sort of function that underwrites intentional content everywhere, not just in linguistic creatures. Her first book (Millikan 1984) is a detailed articulation of teleosemantics, a Darwinian account of both mental representations and language.

Millikan's work reaches far beyond her account of intentionality, as a small sample of her conclusions shows—among them: that dogs have perfectly good concepts, that some thoughts have two directions of fit at once, that understanding language is a form of direct perception. Difficult to summarize, Millikan's program can nonetheless be seen to be framed by three questions: In the philosophy of mind, What is it for one's thoughts to be of something?; in epistemology, What is it for one to know what one is thinking of?; and in metaphysics, What makes for the objective samenesses in the world that one's thoughts are of? Her interlocking answers form a picture of human cognition that challenges tradition on several scores, even as she seeks to defend tradition in the form of scientific realism and the correspondence theory of truth.

Thought

What one's thoughts are of is, according to Millikan, determined by their historically selected function. All intentional items (bee dances, linguistic forms, perceptions, desires, fears, and so forth) have such proper functions, and what any particular intentional item is about, its content, is determined by such functions. (That individual words or token mental states have proper functions and that their content owes to proper functions are claims that have encountered vigorous opposition.) Specifically, a proper function of a feature F of an organism O is a task whose performance by earlier instances of F in other organisms of O 's kind in O 's lineage accounts for the proliferation of F in O 's kind here and now. Importantly, there are nonbiological cases of proper function—for example, customs, hammers, and nails—so the relevant notions of task and lineage must be understood broadly. The content of a representation type R is given by the connection between instances of R and worldly circumstances, recurrent exploitation of which by consumers of R has contributed to their proliferation over time.

What makes mental representations, such as thoughts, beliefs, and desires, distinct from other information-bearing items, such as bee dances? Mental representations are representations that "when they perform their proper function, their referents are identified" (Millikan 1984, p. 13). By identified, Millikan means that the referent is represented as being the same thing again. For example, Clarence's visual perception that a spider is crawling up his leg is an intentional state with a job to do, and such states exist in us because historically selected for performance of that job. The function of his thoughts is to coordinate information he already has about the spider with new information he is acquiring as well as with his subsequent action, trying to brush the self-same spider off his leg. For Clarence's thoughts to be of the spider, then, they must meet the additional requirement of functioning to create this sort coordination of information. The capacity to think of the same as being the same, or "coidentifying" (Millikan 2000), is an important accomplishment, distinctive of advanced cognition.

Millikan here joins company with P. F. Strawson and Gareth Evans in claiming that some form of reidentificatory capacity is necessary for thought about the objective world. Unlike Strawson or Evans, Millikan takes her insight about coidentification to have dramatic consequences for self-knowledge.

Self-Knowledge

What sort of access do we have to our own thoughts? Millikan is a content externalist—just as the meanings of one's words are not settled by one's intentions, the content of one's thoughts are also determined by facts outside one's ken. To know what one is thinking of, then, is not an a priori matter. Some find this consequence troubling and seek to reconcile content externalism with first-person authority. But Millikan (1993) embraces this result, and argues that a still more radical conclusion follows from her functional account of cognition, namely, that nothing is epistemically "given" to thinkers. In particular, "meaning rationalism"—the doctrine that sameness and difference of meaning, univocity, and meaningfulness are all a priori accessible—is false. (It is a good question just who qualifies as a meaning rationalist —some argue, pace Millikan, that even Gottlob Frege not.) Millikan's rejection of meaning rationalism has several startling consequences: We can have no a priori access to logical possibility; there is nothing rationally wrong with believing contradictions; the validity of inferences is not an a priori property; and the very idea of a Fregean mode of presentation must be discarded. In short, like meaning, rationality ain't in the head.

Millikan's radical anti-individualism about meaning and rationality might be opposed by more moderate externalisms. And her attack on the very idea of modes of presentation meets with resistance from those who see a genuine explanatory role for modes, even within naturalistic accounts of the mind.

Ontology

Coidentification is the heart of thought because the goals of organisms are served by coidentifying. But the goals of organisms would only be thus served if there were genuine coidentifiables in the objective world. Millikan's ontology is decidedly realist. Her functional take on concepts has her carving nature at different joints than others might, however. For Millikan, empirical concepts are of substances, that is, coidentifiables. The category substance includes real kinds (e.g., mouse ), individuals (e.g., Mama ), event types (e.g., breakfast again ), and numerous other stuffs and types (e.g., ice, Starbucks Coffee House ). At an important level of abstraction, there is no genuine ontological distinction to be made among these things.

Summary

In briefest summary: Millikan's program for understanding the nature of representation—which is to say, for understanding ourselves—is impressive for its combination of detail and scope.

Bibliography

works by millikan

Language, Thought and Other Biological Categories: New Foundations for Realism. Cambridge, MA: MIT Press, 1984.

White Queen Psychology: And Other Essays for Alice. Cambridge, MA: MIT Press, 1993.

"On Unclear and Indistinct Ideas." In Philosophical Perspectives, Vol. 8: Logic and Language. Edited by J. Tomberlin, pp. 75–110. Ascadero, CA: Ridgeview, 1994.

"Images of Identity." Mind 106 (423) (1997): 499–519.

On Clear and Confused Ideas: An Essay about Substance Concepts. Cambridge, U.K.: Cambridge University Press, 2000.

Varieties of Meaning. Cambridge, MA: MIT Press, 2002.