To my eyes, the bitter struggle over "science" demonstrates that, for those who have advanced beyond religious irrationality, there is still a need for a hegemonic intellectual framework. Much of the appeal of classical Marxism, of course, was based on its supposed "scientific" character. Everyone else was just poking around; Marxists understood the "laws of motion" of capitalism, if not human history in general. A popular argument for capitalism, on the other hand, is that the combination of the market economy and the "marketplace of ideas" has given rise to a never-ending increase in productivity and living standards. The case for a division of knowledge, according to which no one can know enough to govern the system as a whole, but intellectual specialists can prosper amid general competition, mirrors the case for a division of labor. Of course, just as the hierarchies of the modern corporation contradict the contractual utopia of competitive capitalism, the hierarchies of science belie the individualist conception of knowledge. The pomos would play the role of atheists, denouncing science as a false god. Unfortunately, few of them know much real science (this is now beyond dispute), and they deny the obvious fact that modern science, for all its defects, is a uniquely successful human enterprise: it really does advance over time, in a sense in which no other form of human knowledge does. There are two levels to this problem. The easiest (but still difficult) is the critique of the application of science in the modern world, its subordination to profit and power. I think we have the tools we need to do this, and there is a steady stream of literature, in the environmental sciences, science and technology studies, and elsewhere, that shows how it can be done. (Think for instance of the last 25 years of research on the application of technology to the labor process.) The more difficult problem is dislodging science from its ideological role while continuing to recognize its distinctive character. Very generally, I think this requires a close look at the institutions through which science is organized, such as educational systems, journals, and the grant economy. Just as the explanation for science's progressivism is found not in an ideal methodology ("the scientific method") which is largely fiction, but in the distinctive social organization of science (as Kuhn and others since have shown), so its embeddedness in structures of wealth and power can be traced institutionally, and not in the critique of an abstract conception of "knowledge/power". Unfortunately, institutionalism, which might provide intellectual moorings for this project, is itself contaminated with false scientism. (Dewey's use of science in particular is atrocious, even though his insight concerning the potential relationship between scientific investigation and democratic discourse is still central. Veblen is also problematic here.) Traditional institutionalist analysis uses an idealized "scientific" model of knowledge and action as the counterfactual against which to measure actual social institutions. Institutional analysis of science seems to me to be the right way to go, but we will have to be much less doctrinaire about counterfactuals. The doctrinaire frame of mind, in religion, science, and social theory, is the problem. Peter Dorman