The Functional View of Science: Nozick and the American Pragmatist Tradition

Z. Introduction When the Boston Globe, the New York Times and the Harvard University Gazette reported the death of Robert Nozick on January 23, 2002, few who knew his name would have doubted that a great philosopher had died. Precisely because his Anarchy, State, and Utopia (1974) had stirred so much controversy, some of his other work, such as in the philosophy of science, was far less weU known. However, Nozick did propose an important view in the philosophy of science that he caUed the. functional view of science. We see this view in his last work, Invariances (2001), and we also see there how "American"—that is, how pragmatic in a Peircean sense—his conception of science actually is. It is certain deep connections between Nozick's and Peirce's (pragmatic) conception of science that I would like to explore. Nozick had already reflected on the nature of science as a graduate student of Carl Hempel at Princeton, in his seminal Philosophical Explorations (1981), and in his later The Nature of Rationality (1993) and Socratic Puzzles (1997). However, he never seems to have made systematic his views on something like a "position" on the "nature of science." In his later works (1981,13; 2001, 3) he avoided "coercive" philosophy and did not try to give compeUing arguments and "proofs." Instead, he simply tried to lure us into a web of philosophy as explanation (not proof) and always succeeded. He did so also in the case of thinking about science, most explicitly in his contribution to a volume in remembrance of Carl Hempel (who died in 1997), in which he introduces Úte functional view of science (Nozick 2000). He gave the piece a natural place in his latest book which deals with the structure of the objective world.1 In the sections to follow I will expose and explore Nozick's functional view of science. The issues he treats—rationality, progress, The Functional View of Science Nozick and the American Pragmatist Tradition Herman C.D.G. de Regt TRANSACTIONS OFTHE CHARLES S. PEIRCE SOCIETY Vol. 41, No. 4 ©2005 H U 780 CO objectivity, and veridicality—are of course big topics and here I can only Z briefly pick out some aspects of his view of science. Ultimately, I will conO centrate on the issue of the underdetermination of scientific theories by empirical data. My main perspective throughout this article will be the question to what extent Nozick's functional view can and must he placed in the -i American tradition of pragmatism, especially in the version offered by CS. ¡y-j Peirce.2 There are two reasons for this: (I) there are striking similarities in *7 Peirce's and Nozick's functional approach to some crucial aspects of science γ standard model emerge. The radical answer accepts the fact that these factors « undermine the objectivity of science. The defensive answer "formulates a more ^ nuanced notion of objectivity, one according to which science is objective z despite the complications'^ 105). Given this latter answer one must argue ■how we can circumvent these biases and restore the objectivity of science. ·_ Nozick, however, offers a third alternative which he dubs the functional view of science: "science is rational and objective, not despite the complicating factors » but (in part) because of them." (106) Ç Starting off stating in a very general way that "we do not start from η scratch unless we have to" (106), Nozick shows how the list of possible complications to the standard model actuaUy point to factors that (in part) make science rational, objective, progressive, and veridical. With regard to the problem of testing theories with auxiliary hypotheses (the Duhem problem) Nozick adheres to the Kuhnian insight that it is rational to work within a paradigm. Only within a paradigm can we stand "upon our existing theories ", so that we "can reach to predictions yielded by theories still further from bare observations" (107). The theory-ladenness of observation is mitigated by a very Peircean looking strategy: although observations are always theory-laden since our senses are shaped by natural evolution, "the more evolution has laden our observations with accurate theory, the better" (108). In the same vein Nozick downgrades the problem...