values into decision making. By contrast, preserving the fiction of scientific neutrality, and thus excluding scientists from full participation in ethical debate, may deprive society of an important avenue of moral criticism.

The difficulty, of course, is that the research community seems often not to be aware of the social and ethical judgments that are embedded in the practice of science or the design of technology. This is an area where feminist critiques of science have opened a discussion of the preconceptions of science (Keller, 1985; Haraway, 1989). For example, there is now general agreement that women's health issues have received disproportionately little attention in biomedical research and that study designs have often been insensitive to questions of gender and culture. One illustration is the case of clinical trials of contraceptives that assumed, inappropriately, that the characteristics of user populations in the developing countries would be the same as those in industrialized countries. Examples such as this suggest that, in order to play an effective part in public moral discourse, scientists may first need to develop a more reflective understanding of the ways in which they do science.

Social and political institutions continue in hard cases to delegate to science the power to make binding ethical judgments. Such delegation often takes place in the guise of seeking a consensus on facts and then allowing these facts to dictate rules of action. Thus, medical expertise has variously been called upon to define "viability" for purposes of distinguishing lawful from unlawful abortions; to define "death" so as to guide the use of life-prolonging technologies; and to define "pre-embryos" in order to establish the limits of permissible research with the human concepts. Building a scientific consensus around such morally loaded concepts can be a convenient way of ending disputes. Science then appears to draw a bright line that neatly separates permissible from impermissible social action. The facts provided by science constitute a seemingly objective basis for distinguishing right from wrong behavior.

Although this strategy of delegation may work well in terms of producing social harmony, we should keep in mind one or two caveats. First, as noted above, Americans are not in general given to unquestioning acceptance of the authority of science. Indeed, our legal system has been a powerful instrument for holding scientific experts accountable to lay members of the public. In part, the law accomplishes this result by making scientists speak in language that is accessible to nonexperts; this requirement has been particularly emphasized in the judicial review of agency decisions. Further, the adversarial processes of the law very effectively probe the basis for scientific claims and expose the assumptions and judg-