1. Introduction

Janet Kourany speaks for many current critics of science when she claims that “science has done more to perpetuate and add to the problems women confront than to solve them” (Reference Kourany2001). Indeed, there are now in the literature hundreds of examples of the purported negative influence of race/class/gender stereotypes in the history of science. Sargent's (Reference Sargent2004) paper in this volume discusses in detail Elizabeth Potter's book about the role of gender in the work of Robert Boyle.

Kourany goes on to issue a feminist manifesto calling for a thorough-going reconceptualization of the proper aims, methods, and values of scientific inquiry (Reference Kourany, Pinnick, Koertge and Almeder2003). As Pinnick's (Reference Pinnick2002) paper illustrates, Kourany is joined by at least a dozen feminist epistemologists. Yet, concrete, detailed, positive proposals illustrating how scientists might actually incorporate progressive political values into the content of science are in much shorter supply. In this paper I will analyze one such proposal—Anne Fausto-Sterling's writings on sexual dimorphism—and then use it as a probe to compare the views of Longino and Kitcher. (Rolin's Reference Rolin2004 essay provides a fuller analysis of Longino's approach.) My paper concludes on a cautionary note.

2. Feminist Historiography of Science

In his ground-breaking 1963 monograph “Towards an Historiography of Science” Joseph Agassi argued that historians’ views about the nature of scientific method helped to shape the histories they wrote (Agassi [Reference Agassi1963] 1967). A survey of classic works in the history of science reveals inductivist histories that focus on data collecting, conventionalists who emphasize the economy of new conceptual schemes, and critical rationalists who document the role of refuted hypotheses. The influences that Agassi was talking about were comparatively subtle and there was still a considerable overlap between the stories of Galileo produced by an Alexandre Koyré and a Stillman Drake. Feminist historians, by contrast, very consciously set out to look for the influences of race/class/gender (which I'll abbreviate as RCG) on all aspects of the scientific enterprise. And as illustrated by Sargent's analysis of Potter's work, the resulting studies differ profoundly from earlier historical accounts.

The fundamental thesis of feminist historiography is that the old distinction between the contexts of discovery and justification is irrelevant to someone looking for the influence of RCG factors. Not only do background prejudices about RCG influence the space of problems that scientists investigate and the hypotheses that they find plausible, such prejudices also influence the tests scientists design and the thresholds they set in order to deem a theory scientifically acceptable. There is no sanitized context of justification in the feminist conception of scientific inquiry (see Koertge Reference Koertge, Pinnick, Koertge and Almeder2003a).

Guided by their RCG heuristic, feminist historians uncover cases that are intended to provide empirical support for their claims about the distorting effect of stereotyping on the content of science. Sometimes the influence is crude and direct, as when Aristotle assumes that the male contribution to reproduction is active and form-defining while that of the female is passive and material. In other cases the purported influence is subtle and indirect, as when Linnaeus's choice of mammary glands to define the class we call mammals is said to be influenced by his opposition to wet-nursing (Schiebinger Reference Schiebinger1993). The general RCG argument goes like this: Alternate hypotheses were available at the time. The data available were not enough to determine the choice. The theory chosen has sexist/racist/classist implications. Therefore, RCG prejudices influenced its selection.

Many of the case studies produced by the feminist historiographic research program can be faulted. Again, the counterarguments share some general features. Often the feminist analysis overlooks the role of broader theoretical considerations in the evaluation of scientific hypotheses. So in the case of Linnaeus, for example, Schiebinger notes that alternate defining characteristics of mammals, such as the properties of hirsuteness (Pilosa) or having hollow bones in the inner ear (Aurecaviga) were mooted about at the time. But she fails to note that there are clear scientific advantages to the choice of Mammae. Scientists prefer classification schemes that are not only easy to use but also of theoretical import. Any good comprehensive account of biological organisms will have to describe and explain modes of reproduction. For both practical and theoretical reasons, mammae are a better choice than hair or ear bones (Koertge Reference Koertge, Pinnick, Koertge and Almeder2003b). There is no need to bring in external factors.

But let us grant that there may well be episodes in the history of science where RCG-inspired misconceptions have remained intact for a disturbingly long time. We then need to ask how best to remedy the situation. There are a wide variety of feminist responses, ranging from suggestions for recruiting more women into science to a call for a total overhaul of traditional canons of scientific method (Koertge Reference Koertge and Koertge1998). But the most intriguing normative proposal is to put gender politics into the content of scientific theories. I formulate such a policy as follows:

RCG stereotypes in the past have distorted the content of science and had a negative impact on the most vulnerable members of society. It is impossible to remove political evaluations from the scientific process. Therefore, we must at all times be mindful of the political implications of our scientific practice. We should deliberately inject politics into all aspects of scientific inquiry but it should be the right kind of politics, a politics that incorporates progressive social values.

There has been much discussion among feminists about the philosophical details of such a proposal but little exploration of how it would work out in practice. People often talk about the kinds of research that wouldn't be done (e.g., anything smacking of sociobiology or genetic determinism), but concrete, positive suggestions are rare. There is, however, one intriguing feminist scientific research program that has been worked out in considerable detail: Fausto-Sterling's writings on concepts of biological sex. We will trace her attempts to introduce progressive political values into the very foundations of the science of human biology.

3. Fausto-Sterling and Concepts of Biological Sex

In a recent comprehensive work, Sexing the Body: Gender Politics and the Construction of Sexuality, Anne Fausto-Sterling (Reference Fausto-Sterling2000a) analyzes the influence of gender politics on past and present biological concepts of human sexuality and then proposes a new conceptual framework, one that she claims will be both scientifically and politically superior. Fausto-Sterling is exceptionally well qualified to undertake this task. A professor in both biology and women's studies at Brown, her Reference Fausto-Sterling1992 book, Myths of Gender, which criticized research on sex differences, was well received. And her Reference Fausto-Sterling2000a book, which we will be discussing, received the Robert K. Merton Award from the American Sociology Association.

Fausto-Sterling's work sits squarely within the feminist tradition of questioning dualisms, especially ones having to do with sexuality and gender. Feminists have consistently argued that gender and sexual preference are richly variegated and hence must obviously be decoupled from biological sex, which they took to be binary. Fausto-Sterling's work extends the traditional feminist program of deconstruction to include biological sex by advocating that the fixed, binary conception of male/female be replaced with a pluralistic, fluid, multidimensional array. I'll call this reconceptualization the F-S proposal.

An important part of the scientific and political motivation for the F-S proposal comes out of a concern with the problems of intersexed neonates. Recent books, such as As Nature Made Him: The Boy Who Was Raised as a Girl, John Colapinto's exposé of John Money's surgical approach, and Alice Dreger's scholarly Hermaphrodites and the Medical Invention of Sex have drawn attention to the physical and psychological mutilations imposed on babies that don't fit the two-sex mold. Not content with just calling for a more humane and enlightened approach to this admittedly horrible social issue, Fausto-Sterling uses this material to interrogate the very idea that Homo sapiens is a sexually dimorphic species.

If our basic conceptual scheme were to hold that there are not two, but a plurality of sexes, she argues, then it would make no sense to force what we now call intersexed neonates into Procrustean categories. Furthermore, she claims, when we say there are just two sexes, this invites the idea that one is superior to, or more highly valued than the other (Reference Fausto-Sterling1993). Pluralism is more congenial to egalitarianism than dualisms. She is quite explicit about the political rationale for reconstructing the concept of sex, but she also adduces scientific grounds for rejecting a strict binary notion. The existence of hermaphrodites and other configurations of gonads undermines the idea of sexual dimorphism. Chromosomal variants are well-known to Olympic committees faced with the task of trying to sort athletes into only two categories, male and female. Fausto-Sterling also looks at the history of the discovery of sex hormones, arguing that it was the power of binary concepts of gender that led to the labeling of certain steroids as male and female. Since traditional males and females each have both estrogen and testosterone, she claims that there is no scientific motivation for talking about male and female sex hormones, although she recognizes that sex hormones were first isolated from men's and women's urine respectively (Reference Fausto-Sterling2000a).

Fausto-Sterling concludes that the idea of two biological sexes is a socially pernicious myth, an arbitrary simplification of the variety that exists in nature. Thanks to the feminist movement we all have more freedom to create gender identities and sexual orientations. We should extend this flexibility to biological sex. For example, Fausto-Sterling proposes that people shouldn't be forced to identify as male or female when they apply for a passport or driver's license. But carrying out this revolution will require an overhaul of both our popular and scientific conceptual schemes. She sums up in the introduction to her book:

I have presented only a minimal account of Fausto-Sterling's proposal (see Koertge Reference Koertge, Patai and Koertge2002 and Reference Koertge, Pinnick, Koertge and Almeder2003b for more details) but that will be sufficient to illustrate how she integrates empirical and political factors into her justification of the new conceptual scheme. Feminist response to the F-S proposal has been overwhelmingly positive, especially now that she has dropped her original call for five sexes and has adopted instead the idea of a pluralistic array. Members of the Women's Studies Discussion List now talk about using Fausto-Sterling's work in their classes to correct the misconceptions that students have picked up in their biology classes (see Patai and Koertge Reference Patai and Koertge2002, 305–313). Yet, ironically, it is traditional biology that has provided the account of sexual variation of gonads, chromosomes, and hormones with which Fausto-Sterling supports her proposal. And it is modern biology's concept of species as variable populations that has refuted the folk practice of labeling as monsters any organism that deviates from the prototype.

Nevertheless, parts of the F-S account could serve as a useful reminder of the complexity and variability of biological sex (see Blackless et al. Reference Blackless, Charuvastra, Derryck, Fausto-Sterling, Lauzanne and Lee2000). Certainly there would be no scientific harm done if we were to stop talking about sex hormones and talk of steroid hormones instead. And we know from science itself that any adequate account of the origins and functions of these chemicals will certainly be very complex indeed. Ditto for the varieties of intersexed babies, and what, following Freud, we might call the polymorphous perversity of genders and sexualities. So why shouldn't biologists just gracefully admit that, strictly speaking, Homo sapiens is not a two-sexed species?

In many parts of science it would make little difference. After all, one of the hallmarks of modern biology is thinking in terms of populations instead of type specimens and social scientists almost always find they need to work with cluster concepts instead of simple conjunctions of necessary and sufficient conditions. Sexologists and psychologists have for some time been using multidimensional schemes to talk about both sex and gender (see surveys in the Kinsey Institute volumes edited by Reinisch et al. Reference Reinisch, Rosenblum and Sanders1987, Reference Reinisch, McWhirter and Sanders1990).

But evolutionary theory, a central part of modern biology, could not easily give up its talk of two sexes and still retain its explanatory power. There are major differences in the evolutionary paths of sexual and asexual species. In fact, producing mathematical models of just how much difference it makes in the rates of variation and adaptation when offspring share genetic materials from both parents has been one of the major achievements of biological theory. Species that reproduce sexually tend to be rather good at sorting out which of their conspecifics belong to which sex, and this often is associated with quite a marked sexual dimorphism.

One also wonders how to give scientific accounts of reproduction and genetics within the F-S scheme. If we refuse to “privilege” XX and XY pairs of sex chromosomes in any way we lose our ability to make generalizations about sexual dimorphism in various species and to classify certain diseases as sex-linked. New technologies may make it possible in the future to have reproduction without the involvement of people of two different sexes, but it seems undeniable that the evolution of our present bodies and minds happened within a species that had two clearly distinct and easily recognizable sexes. Furthermore, the F-S account of the interaction between the physical and social body is too vague to be of much help in either biological or social science. All we get by way of clarification are analogies to nested Russian dolls and the “two sides” of a Möbius strip (Fausto-Sterling Reference Fausto-Sterling2000a).

Since Fausto-Sterling holds that theories should be evaluated in political terms, one may also question how socially progressive her scheme is. Even if science were to take over her proposal lock, stock, and barrel, it seems highly unlikely that ordinary people would stop asking whether the new baby was a boy or a girl. Intersexed children will continue to live in a world in which two sexes are anything but a myth—even Fausto-Sterling finds herself having to speak of “conventional males and females.” They need to be protected from the enthusiastic surgeon's knife and social stigmatization as do children with six fingers or toes, but tampering with our biological characterization of Homo sapiens is a terribly ineffective way to accomplish this worthy goal.

4. Evaluation of the RCG Normative Proposal

We can use the F-S example to illuminate the attractions and pitfalls of a general methodology that encourages researchers to include political considerations into the content of their science. By dramatizing the plight of intersexed neonates, it encourages scientists, especially science teachers, to emphasize that dimension of diversity within Homo sapiens. But like all too many social constructionist reforms, it tends to suggest that if we simply change our terminology (for example, speaking of steroid hormones instead of sex hormones), social practice will change.

The F-S proposal does not minimize the importance of empirical adequacy—it does a good job of documenting the variation of gonads and chromosomes, gender identification, and sexual orientation, and the lack of correlation between these various dimensions of sexuality. Yet, like so many would-be revolutionary suggestions, it ignores the requirement of consilience with other existing successful scientific theories. Defenders of the new account neglect to make an inventory of all of the roles that the binary conception of sex plays in biology and to speculate how the F-S proposal might accommodate them. Even a crank theory can point to some empirical evidence in its favor—it is less easy to propose a new theory with broad explanatory power (compare Giere's Reference Giere1996 discussion of the “patchwork quilt” fallacy).

The primary motivation for the F-S proposal was political (there was no crisis in biology that prompted it), so it was natural that it would receive most attention from its political constituency. As a consequence, the primary feedback it received was political in nature. Her original proposal for five sexes, which postulated three varieties of hermaphroditism, was roundly criticized by intersex activists for its emphasis on genital configurations (Fausto-Sterling responds in Reference Fausto-Sterling2000b). Scientists, however, are interested in the scientific adequacy of the scheme. We thus arrive at a situation where the political constituency feels the F-S proposal has been thoroughly scrutinized and revised in the light of (political) criticism while the scientific community judges it to be scientifically inadequate (Hull Reference Hull2003).

The upshot is very likely to be an increased feeling of alienation by the political advocates who are either unaware of or unimpressed by the scientific objections. It need not always be so—many AIDS activists educated themselves on both the science of retroviruses and the rationale for FDA testing procedures and thus were in a good position to have an impact on scientific practice. However, feminists are so convinced by historiographic claims about the RCG bias in traditional science that they tend to assume that any scientific criticism of feminist proposals is inspired by reactionary politics! At least in the short run, the result of preliminary attempts to make science more feminist in content is likely to be an increased feeling of alienation from science amongst feminist academics and women's studies students (Patai and Koertge Reference Patai and Koertge2002, 305–311).

Such considerations lead me to question whether the RCG approach is even politically superior to the traditional model that gives ideological factors considerable spielraum in the contexts of discovery and application, but rules them out in the context of justification. By imposing a political filter on our hypothesis space, RCG rules out ab initio possible scientific solutions to current political problems. As the founders of the Royal Society knew full well, it is more difficult to get a working agreement on matters of religion and politics than on scientific matters. For example, if current intersex activists were to set the scientific agenda, they might rule out or set a low priority on studies of the interaction between any biological factors and patterns of social behavior, such as sexual preference or gender identity, on the grounds that such hypotheses smack of genetic determinism. On the other hand, current transsexual activists welcome such research because they are looking for more effective ways of altering their bodies in order to realize a more complete traditional male or female gender identity. It is not easy to adjust science so as to make all oppressed groups simultaneously happy!

In each case, it seems to me, what the activists really need are reforms in the choices that are available to people with unusual bodies or psyches, not a biological science that is somehow tweaked to suit each of the multitude of current political agendas. Science should be a resource for all, but not, as Feyerabend put it, a docile pussy cat (thus revealing, by the way, that PKF had little knowledge of cats!).

In my discussion so far, I have used a very general formulation of what I have called the RCG call for the introduction of social values into science. Let us now look briefly at two recent attempts by philosophers of science to articulate more carefully what a more politically sensitive methodology would be like and ask how the F-S example would fare under each of their models. We begin with Longino's feminist epistemological socialism (Reference Longino1990, Reference Longino2001).

The Fausto-Sterling proposal satisfies all of Longino's desiderata for a feminist theory (Longino Reference Longino1995). It introduces conceptual novelty and its ontology is heterogeneous. It is neither reductionist nor idealized, but instead, highlights complex interactive relationships. It demonstrates a fine sensitivity to human needs. As for my complaints that the F-S proposal does not mesh well with the rest of biology, Longino emphasizes that in a feminist science, the traditional norms of external consistency and scope may need to be downgraded. So Longino's account would give high marks to the F-S proposal. However, it might well give low marks to the reception of F-S by the scientific community. Longino calls for broader participation in scientific debates, a more active response to dissenting views, and more equality of intellectual authority.

Yet when we look more carefully, it seems to me that the scientific community has treated the F-S proposal quite appropriately. Fausto-Sterling's more popular writings have received positive recognition amongst social scientists. Her co-authored, technical metastudy in the American Journal of Human Biology on the incidence of intersexed neonates (Blackless et al. Reference Blackless, Charuvastra, Derryck, Fausto-Sterling, Lauzanne and Lee2000) has been cited by others in the field. One reply (Hull Reference Hull2003) criticizes in detail the statistical techniques employed and argues that the F-S estimate of the percentage of live births that are not dimorphic is too high by an order of magnitude. Carrie Hull's concluding sentence reflects the traditional scientific concern about politically inspired research: “While this data will undoubtedly transform over the next years, the numerous errors and omissions made by Blackless et al. suggest that they were too keen to find a relatively high incidence of sexual nondimorphism” (Hull Reference Hull2003, 114).

Fausto-Sterling's response is mild in tone (Reference Fausto-Sterling2003). She welcomes the search for better numbers, but points out that knowing the exact frequency is not needed in order to correct the practice of physicians who tell intersexed adults that their condition is so rare that it would be useless to hope to find other similar people to talk to. (Here the internet is now an invaluable resource.) Fausto-Sterling does reiterate her interest in the theoretical project of reconceptualizing sexual difference so as better to accommodate human sexual variation and expresses her hopes that it will foster discussion. However, if my evaluation above of the scientific merits of the F-S proposal are correct, then benign neglect is exactly the response we would expect from the scientific community. Perhaps the “more active response to dissenting views” that Longino calls for is best supplied by philosophers!

What does Kitcher's Reference Kitcher2001 normative account of well-ordered science say about the F-S proposal and its reception? Kitcher adopts Giere's metaphor that theories operate like maps, and he explicitly says that classificatory concepts are dependent on the interests of those who use them. This would seem to allow for the possibility that people interested in the plight of the intersexed might adopt one concept of biological sex while evolutionary biologists prefer another. However, Kitcher does hold that in the long run all scientific maps must be consistent. There may be sharp divergences within science and consensus may be slow to emerge, but as debate continues, views should be tempered,.

In Kitcher's conception of well-ordered science, scientific deliberations must avoid two hazards that tend to pull in opposite directions, which he calls Inadequate Representation and Tyranny of the Ignorant (Reference Kitcher2001, 129–131). But it's not clear where he would locate the F-S proposal between these two extremes. Longino would certainly be expected to argue that feminist concerns have been inadequately represented in science. And she claims that Kitcher's democratic ideal is too elitist, saying it espouses government for the people rather than by the people. However those who are dismayed by the failure of advocates of the F-S proposal to take account of the main corpus of biology would be concerned about the problem of widespread ignorance, especially within the area of women's studies.

It is probably unrealistic to expect a theory of methodology to give sharp criteria for acceptable vs. unacceptable scientific behavior. Yet it seems disappointing that neither Longino nor Kitcher supply a framework that helps us to make cogent criticisms of the F-S proposal.

5. Conclusion

For Bacon the great promise of science was to provide both the light of understanding and fruitful applications. Science has been remarkably successful on both scores. It has given us comprehensive accounts of the world on both the cosmological and the subatomic scale. It has prolonged life and relieved drudgery to a degree that Bacon could not have dreamed of, even in a New Atlantis. Today feminists and many others are asking science to play a larger role in the project of relieving prejudice and oppression. This is a legitimate request, and the very fact that it is being made shows that science is respected for its power to dispel ignorance and to better the human condition (Koertge Reference Koertge, Patai and Koertge2002).

But feminist epistemologists and others also believe that science cannot contribute to a liberationist agenda without incorporating a particular gender politics into the heart of its methodology. This essay has looked at one implementation of a gendered methodology. Our critical evaluation of the F-S proposal suggests that a thorough-going feminist overhaul of scientific methodology would be counterproductive. The attempt to evaluate scientific theories in part on the basis of their alleged political progressiveness narrows the hypothesis space that scientists may pursue in their attempts to solve problems. In the case of politically congenial theories, it becomes all too easy for political enthusiasts to overlook criticisms of the theory's scientific adequacy. With a politically unwelcome theory it becomes all too easy for activists to ignore the degree to which it is empirically supported. Furthermore, in a pluralistic society there will rarely be wide agreement on which theory would be most politically progressive if it were true. And political appraisals are likely to be unstable over time (see Koertge Reference Koertge2000).

Political passion can be an effective motivation for pursuing the arduous business of scientific research. Kinsey's burning desire to overcome ignorance and judgmentalism about sexual behavior undoubtedly helped keep him going on his exhausting treks around the country to collect sexual histories. A political sensitivity to the harm that can be done in applying theories that turn out to be seriously mistaken can reinforce the systematic attempts that scientists already make to insure a thorough and impartial evaluation of their results. Political activists can legitimately alert those who apply scientific theories in the arena of social engineering to a variety of unanticipated consequences of such interventions.

But reformers must strenuously resist the temptation of the fallacy of wish-fullfillment. To award a theory good marks because it fits our political agenda, or bad marks because it doesn't, is in blatant violation of the most basic canons of critical thinking. This is an important caveat to any call for the blending of social values into science.