|
Article Excerpt
THERE ARE INSTANCES OF SOCIAL SCIENCE "TRUTHS" THAT HAVE BEEN subsequently and convincingly demonstrated to be false. Is it useful to label these now discarded truths as social science mistakes? Perhaps, but as argued in this essay, the mistake was more likely a political than a scientific one--that is, a mistake whose origin is to be found in the assumptions, preferences, and prejudices brought to the research question. If I persuade the reader of this point, it leads to a broader and more interesting assertion. Modern social science was established as two inseparable projects: a science project--deeper understanding of human behavior, relationships, institutions--and a political project to, among other things, improve the human condition, protect the homeland, and grow the economy. It is in the inseparable nature of these dual projects that we can best treat the issue before us.
AN EXAMPLE
I start with an example, one that, even though it predates the formal establishment of the social sciences, was a "theory" covering a wide range of human behavior. Samuel George Morton, who, were he writing today, would claim interdisciplinarity by virtue of working at the boundaries of anthropology and zoology, was one of many mid-nineteenth-century scientists intent on proving that there was a hierarchy of races. Morton measured cranial capacity, reporting in 1849 that, yes, Caucasians were blessed with the most capacious skulls, followed by Mongolian, Malay, Native American, and with Negroes (and selected aboriginal groups) on the lowest rung. That Morton's science was deeply flawed--self-selected samples, classification errors, dubious statistical treatment--is now well documented. But in his time his influence was substantial, and "his time" lingered deep into the twentieth century.
Morton gave early currency to the pseudo-race science of polygenetics--the races were separately created species, endowed with different and unequal capacities. In the mid-nineteenth century, Louis Agassiz became the best-known scientist who argued the Mortonian polygenetic line. A less famous believer, however, had the greater impact. Josiah Nott can be credited with no less an accomplishment than turning the US census into an instrument for determining whether the offspring of cross-race breeding were mentally defective and would die young. In service of this hypothesis the 1850 census introduced a Mulatto category (and later in census history, still working the same social science territory, the Octoroon and Quadroon categories). As one close scholar has noted: "The 1850 census proved to be a watershed, not only because (social) scientists were marshaled in its service, but because they brought with them, as scientists, their thinking about race. This census boldly ushered in the inextricable and enduring link between census categorization, racial scientific thought, and public policy in the United States" (Nobles, 2000: 20). This was applied social science at a scale not previously imagined and seldom produced since. It bolstered apartheid policies that lasted well into the twentieth century. For example, laws prohibiting miscegenation, echoing fears that racial intermarriage produced defective children, remained on the books in 16 states until 1967, when the Supreme Court (Loving v. Virginia) finally declared them unconstitutional.
The long unhappy story, including sociologist Herbert Spencer's survival of the fittest theory and the lengthy twentieth-century detour into eugenics and theories of racial purity, is too well known to require more than bringing it back to mind in the present context. There was a race-science. It was empirical, starting with Morton's crude cranial capacity measures and extending deep into psychometrics. It was theoretical, purporting to explain many features of social behavior and social structure. It was the central preoccupation of what passed for social science in the pre-Civil War era. "By any reasonable yardstick," writes an historian of science, "phrenology was a mid-century social science" (Cravens, 1985: 187). Measured against the criterion of social and political applicability, it was enormously successful.
The social science mistake was an elementary one. As noted by Stephen Jay Gould, it was the "claim that worth can be assigned to individuals and groups by measuring intelligence as a single quantity" (Gould, 1981: 20). On this premise was built a race-science using craniometry as the principle methodology in the late nineteenth century and (certain styles) of IQ testing as the principle methodology in the twentieth century. Gould records this history in persuasive detail, and is careful to distinguish documented instances of fraud (H. H. Goddard's altered photographs and Cyril Burt's fabricated twin data) from the more interesting, because they were honest, conceptual and methodological mistakes. The mistake continues well into the twentieth century; see H. J. Eysenck's argument for black inferiority, wherein correlation is mistaken for causation (Eysenck, 1971).
Our issue is how to characterize this social science mistake. It is obvious that neither the formulation of race-science nor its subsequent rejection can be understood solely in scientific terms--that is, by simply considering hypotheses, data, theory construction, better data, new hypotheses, theory modification, ad infinitum. Both its formulation and its rejection have to be understood as part of a larger political project: its formulation on behalf of defending slavery and sustaining racial separation; its rejection on behalf of educational programs to discredit racist thinking and government policies to compensate for past racial injustice. In this example, the inseparability of a social science theory and its political uses indicates how a science project and a political project were unfolding in tandem, resulting in a social...
|
