Sex & Objectivity: "Bad" Science

Guided by a feminist perspective (but certainly not limited to it), some contend that knowledge obtained through science can be "wrong" as a result of bias introduced by sex and by gender. They claim that women are specially situated to identify "bad science" and root out error, and that they are critical to avoiding similar mistakes in the future. Behind this critique are two related claims:

First, science is not "objective" because the sex of the researcher can affect what plausible hypotheses s/he introduces, what s/he observes or how s/he interprets data. Each can lead to unjustified support for certain theories.

Second, sex can bias the basic questions asked. Where questions are not asked, knowledge is not found. Because omissions can seriously distort knowledge, scientific knowledge from males is incomplete and therefore not "objective."

These claims may be particularly evident to women while relatively invisible to men, because the implied lapses of objectivity typically favor men--in continuing to legitimize relationships where men dominate, maintain power, or establish their way of doing things as normal. In this view, we must critique scientific knowledge, not merely in terms of the quality of the experimental evidence, but also in terms of the ways in which it may sustain power.

The clearest examples of sex bias historically involve theories related to reproduction and to sex differences. Hence, we may be sensitized to the simple but far-reaching assumptions in defining the egg as the passive gamete, the sperm as active. This characterization likely originates in a parallel perception about the male and female organisms in human social terms. At the very least, (male) scientists holding this view of male/female social roles are less likely to question the active/passive distinction, once introduced. --And they are less likely to search "actively" for or notice ways in which the egg might participate in the interaction--for instance, in admitting sperm through the vitelline membrane (perhaps selectively). Students may, in turn, extrapolate inappropriately from the active/passive characterization of the gametes to a social level, and see male and female roles in humnan society as biologically based [see also "A Philosophy of Sex?", page 10].

Readers may have also noticed in the news recently the introduction of a new theory on the evolutionary role of menstruation. Profet challenges the conventional notion that the endometrial lining is sloughed off as useless--as a failed chance at reproduction. (Here, once alerted to bias, one can see the strong overtones for a woman's self-image). Rather, Profet suggests, the process may actively eliminate germs carried in by sperm; menstruation would then be a more active defense against infection.

Students these days are well attuned to discrimination in the past. But they are less inclined to see how science supported or maintained that bias. In the mid-1800s, for example, women's nervous systems were viewed biologically as finer, more irritable and prone to overstimulation. The uterus was seen as the "central" female organ and connected to the nervous system (which carried a finite amount of energy). The implication--derived from scientific and medical "fact"--was that women ought not to be highly educated. Their bodies were not suited for challenging mental activity. Any excess work would divert energy away from their primary biological function, reproduction. We can criticize such science today only because some women (or men adopting a woman's perspective) challenged the assumptions that shaped the "objective" observations. The science was real, but it was "bad" because women's views of the same "evidence" were excluded.

Case Study: Craniology and Sex Differences

One of the most dramatic cases of the sex-biased interpretation of data is in nineteenth-century craniology, superbly documented by Elizabeth Fee (see References). In the 1860s, in the shadow of a growing women's rights movement, many men sought to protect women from new threats to their perceived biological nature. Anthropologists in Britain recognized that "the subordinate position of women had for too long rested on easy assumptions about female inferiority; this inferiority must now be scientifically investigated." One could compare the intellectual properties of men and women, they postulated, by examining their brains--or indirectly, their skulls. The conclusion was perhaps already known, but measurement would reveal an authentic, scientific answer. Ironically, perhaps, a strong ideological bias was coupled to a commitment to be rational and objective.

Anthropologists measured the skulls in numerous ways. They combined some lengths or angles into computed ratios to capture numerically what they suspected were significant features of shape or form. Ultimately, over 600 instruments were used. One researcher claimed that over 5,000 separate skull measurements were required. There was, as Fee notes, "a Baconian orgy of quantification." (Here is a fascinating case for those who see measurement and quantification alone as the hallmarks of science!)

Researchers first based their conclusions on the most obvious feature, cranial volume. Using this criteria, one could rank women's skulls as smaller and as similar to infants and lower races: that is, as inferior on both anatomical and evolutionary scales.

However, as Fee notes, "this logic had one fatal flaw, which surfaced as the `elephant problem.' If either the absolute size of the brain or cranial volume was to be taken as a measure of intelligence, then the elephant and the whale must be the lords of creation. They possessed brains much larger than that of man. Some craniologists tried to finesee the elephant problem by asserting that the relative weight of the brain to the body was the true measure of intelligence. Napoleon's brain, for example, was small on an absolute scale, but large in relation to his body weight. Disturbingly though, some measurements suggested that in relation to body weight, the female brain was larger than the male."

Rather than rest on that conclusion, however (which violated their intuitions), craniologists pursued other measurements. The great paleontologist Georges Cuvier proposed that the relative proportion of the cranial bones to the bones of the face was an appropriate measure of the dominance of the brain over other senses. According to his criteria, however, birds, anteaters and bear-rats would rank above humans. Cranial height, as indicative of expanded brains, was also discussed, since women had smaller foreheads. But John Cleland noted that Kaffirs, Negroes and Australians had large cranial heights and, therefore, one could not adopt it as an index of intelligence. Both measures (though later rejected as standards) were clearly proposed because they followed basic sexual stereotypes.

Hermann Schaaffhausen in Germany focused on certain cranial features. The "projection of the parietal protuberances, the lesser elevation of the frontal bone, the shorter and narrower cranial base, and . . . the more elliptical dental arch and the inclination to prognathism [or small facial angle]" were all signs, he suggested, of incomplete development. In the same year, Alexander Ecker found that the facial angle of women was, in fact, large (orthognathic). He did not thereby conclude, according to Schaaffhausen's criteria, that women were superior. Instead, he claimed that the large facial angle was a sure mark of inferiority, confirmed by its presence in children. Women with large brains or large facial bones--those who contradicted his rule--Ecker noted, were intelligent or ugly. They were thus exceptions, and for that very reason could be excluded as unrepresentative of their sex(!). Ecker's unchallenged assumptions about the beauty and nature of women lay beneath both his scientific proposals and evaluations.

By today's standards, the theoretical blindness of the male 19th-century craniologists seems outrageous. We find it easy to relegate their case to the ignorance of the past: we know better now. Indeed, we may want to pride ourselves that objective, scientific thinking was itself the very agent responsible for exposing the truth hidden by sex-laden ideology. But this way of thinking represents a different (and equally dangerous) form of short-sightedness on our part. The central and much more disturbing question is how scientists at the time, in their own context could be convinced by these claims--while having no clue that they were "wrong," misleading, or sexually biased. That is, one must withhold one's disbelief for a moment and reconstruct how the craniology of sex differences could once have been considered serious, "good" science.

We need only remind ourselves of the fundamental biological relationship between structure and function. Indeed, it was in the early 1860s that Paul Broca localized speech function to a small area in the left cerebral cortex--a discovery we frequently celebrate as illustrating this relationship. It was not unreasonable for Broca himself, and others, to hypothesize that the structure of the skull might reflect something of the function of the brain inside, and that there might be meaningful differences across the sexes. We can only say now that we can demonstrate no such correlations. The error was in not having female-centered hypotheses equally promoted and guiding alternative interpretations of the data. Objectivity was not necessarily subverted by the enterprise itself, but by the political orientation of the researchers, here determined by their sex. The male scientists at the time, caught within their own cultural perspective, comfortable in (and possibly unaware of) their power, did not see the assumptions that distorted their science and so perceived no need to probe any deeper.

What ultimately was necessary to expose the bias and to force the scientific community to revise its canonical view in this case? Most of the "subversive" work was done by the first two women in the field, Alice Lee and Marie Lewenz. Lee's first study demonstrated that skull capacity was not correlated to intellectual power. She published data for individuals, showing that specific women had larger cranial capacity than even some scientists in the field. She then argued that sexual (and/or racial) comparisons had no significance, since they obviously could not be used to make conclusions about any one person or pair of persons.

Lee also showed that every cranial volume measurement contained a margin of error as high as 3%, due to the variation in packing seeds or sand inside the skulls. Because the proposed sex differences were well within 3%, they could easily be accounted for by the imprecision of measurement. As students of Karl Pearson, Lee and Lewenz strengthened their work with statistical rigor, but they clearly had a motivation for designing their studies in ways that previous males had not. They, and others in Pearson's group, systematically found the flaws in the work of others.

The craniology case shows how the influence of sex may be both unintentional and indirect. Nothing suggests that the craniologists deliberately misrepresented data. And though their research was often motivated by political concerns, they did not explicitly justify specific theories (say, about facial angle or cranial index) by the need to repress women. We can see their biases retrospectively, of course, but it is much harder to characterize the error within the researchers' context. Nor were male craniologists so overwhelmed by their perspective that they could not recognize anomalous data when they encountered it (though they revised their theories in a way consistent with the male perspective). The entry of female scientists into the field was critical, however. The women's perspective was perhaps no less biased or ideologically motivated, but it was effective here in baring certain assumptions and inadequate techniques. It both brought certain data to the surface (where it had to be addressed) and exposed the males' context as limited. Women in science may thus be necessary to illuminate "bad" science. They can likely show where unrecognized "male" assumptions exist--especially in explanations of sex, reproduction or "natural" female roles.

Asking Different Questions

The craniology case indicates that scientific conclusions may vary given interpretations by different sexes. But science may also vary on an even more basic level: due to the problems posed or the questions asked. Again, perspective may determine what is relevant--and hence what is included or excluded from a given theoretical account.

Men and women may notice different things and thus may find different subjects or different aspects of the same subject worth investigating. Perhaps the most blatant example is Martha McClintock's published study on synchronous reproductive cycles in female dormitories. For many women, the notion that women living together for an extended time come to share menstrual periods was common knowledge, not "science." Yet when McClintock completed a simple formal study (originally for a college science class), her conclusions were deemed revolutionary by the mostly male scientific community (and not accepted immediately). Extant theories were inadequate--and failed the test of objectivity--because they had blinded themselves from noticing the fact and from considering it as plausible.

Topics related to sex or to female concerns may obviously be omitted from the male canon, but the sphere of differential "noticing" may extend more broadly. For example, the chemistry of cellular interactions does not seem have any explicit sexual connotations, yet Penny Gilmer, a chemist at Florida State University, finds that her work differs markedly from her male colleagues. She explains: "In my research I am interested in how the chemical and physical properties of one mammalian cell can modulate its recognition by another cell. Specifically, my research group is determining how a tumor target cell can modulate its cell surface properties to inhibit or block recognition by an immune cell. Typically, men in this same research area focus on the immune cell rather than the tumor target cell as I do. Perhaps this is because the immune cell is the one delivering the damage or lethal hit to the tumor cell. However, a woman like myself considers the tumor cell as an equal partner in the cellular recognition event and investigates how the tumor cell can modulate its recognition by the immune system." The male establishment has lately been considering Dr. Gilmer's approach more seriously.

Few if any would claim that the variation in problems or questions in science is directly caused biologically by the sex of the researcher. Rather, the differences "en-gendered" by our culture places men and women in different roles. Gilmer thus explains her work as distinctive because "women as a group typically have different life experiences than do men, so we may offer a different perspective on science." The critical element here was what some call the female `standpoint'. The `standpoint' can reveal interpretations from: (a) the viewpoint of a biological female (say, about their own bodies); (b) the viewpoint of a group whose social role is affected (adversely) by scientific conclusions (as in the craniology case) ; and (c) the viewpoint of someone whose enculturation and thus way of thinking differs from the professed cultural norms. A women's and/or feminist standpoint is thus not exclusively limited to females, though one might suppose from historical evidence that it is less likely to emerge spontaneously in men without some deliberate effort.

Physical sciences are thus as open to reinterpretation by women as are biological sciences. One example may be the recent boom in studies of chaos (non-linear dynamic systems). Stephen Kellert has noted that many "chaotic" phenomena were observed early in the century but were neglected--even though the mathematics was sophisticated enough to pursue them. He suggests that male ways of seeing the world led scientists to disregard or suppress this area of research. The recent explosion of interest, by contrast, accompanies a richer appreciation of complexity and the role of context in how systems behave--factors introduced perhaps indirectly by stronger feminist ways of thinking in the past few decades. The feminist standpoint, here, is interpreted quite broadly or loosely.

A woman's perspective, ultimately, may not challenge the reliability of the results in science. But it does surely question the authority or objectivity of scientific theories that may be incomplete. A feminist approach thus highlights how completeness is another way of defining what is sometimes called "good-versus-bad science." "Good science" does not omit relevant questions or exclude them as irrelevant because they have been asked from a particular perspective.

Implications

A male-dominated science, then, needs women's perspectives to rescue itself from male-bias. That is, to maintain objectivity, we must go deeper than we have traditionally. Only through criticism from the feminist standpoint can we reach this new ideal. Sandra Harding calls this new standard "strong objectivity." Conventional or weak objectivity merely considers the quality of the evidence. Strong objectivity also considers the standpoint from which questions are asked and data are interpreted.

When women's views are accepted as crucial, the analysis of sex and objectivity has important further implications for women in science. Plainly, the issue is not merely equity. Women's participation in science may have very real effects on the content of science. To achieve Harding's "strong objectivity," scientific disciplines must aim to incorporate women and their specific perspective. Science, as well as women, benefits when females are well represented in its various fields.

Finally, the importance of the feminist critique of sex and science can be extended, once again, politically and ethically. The significance of the female standpoint is typically not in introducing merely one more perspective, but in introducing the perspective of the minority or oppressed. Largely, it is only the politically disempowered whose voice is excluded from science--and who need to be addressed in ensuring objectivity. Feminist approaches thus intersect and overlap strongly with racial, ethnic and international political perspectives on science. Critiques from less developed countries, for example, may highlight how the "objective" science of the West (say, about high-tech agriculture or medicine) only serves the West. Science or knowledge may not lay claim to "universality" if it does not apply to the real problems of less affluent nations. Framed in this way, the lessons of sex, gender, "bad science" and `strong objectivity' are fundamentally lessons about power and science.

Classroom Strategy

To convey the potential for bias to students, I use a simple general phrase: "Beware of knowledge that supports elitism!" I inscribed the caveat in capital letters on a yellow diamond fashioned in the style of traffic warning signs. It hangs in a prominent place, as a constant reminder that knowledge that "translates" into a power relationship is especially vulnerable to bias and deserves our scrutiny. Harding synopsizes the same view with a halting rhetorical question. She asks simply and provocatively: "Whose science? Whose knowledge?"

References/Reading


The SHiPS Teachers' Network helps teachers share experiences and resources for integrating history, philosophy and sociology of science in the the science classroom.