The [Kass/Mansfield] brief is an attempt at intellectual hygiene. Among the many annoying tics of contemporary liberalism is its insistence that liberal social policies are always and everywhere determined by the latest findings of social science. Redistribution, affirmative action, tighter economic regulation—name the policy and you’re sure to find some associate professor of some social science or another beavering away with a labful of undergraduates to discover its benefits. Such are the claims made for gay marriage. “More than thirty years of social science,” as one piece of NPR agitprop declared on Morning Edition last week, have demonstrated that children raised by homosexual couples show “no difference” in social outcomes from children reared in heterosexual households. And more recent cutting-edge data show the salubrious effects of gay marriage in general. We are told.
It is the aim of Kass and Mansfield to wave the Supreme Court away from “scientific findings” that are produced by culture warriors, as the findings in the field of “gay studies” nearly always are. “The social and behavioral sciences,” they write, “have a long history of being shaped and driven by politics and ideology.” They note pointedly that two generations ago, the “scientific consensus,” as represented by the American Psychiatric Association, was that homosexuality was a “mental disorder.” The consensus was publicly reversed in 1973, and science, to paraphrase Mae West, had nothing to do with it: Both positions, before and after, were determined by political and cultural considerations.
Now, of course, the American Psychological Association, which waited until 1975 to “depathologize” homosexuality, tries to lend its shaky intellectual credibility to the cause of gay marriage in general and gay parenting in particular. In 2005, it issued a bull declaring the “no difference” finding a matter of settled science. Kass and Mansfield point to a recent paper by Loren Marks of LSU, who had the temerity (and professional death wish) to go back and actually read the 59 studies the APA cited in its decree. They were shot through with conceptual and methodological flaws: small, nonrandom “convenience” samples, a recurring lack of control groups, shifting and poorly defined outcomes, and a steady pattern of comparing apples to oranges—for example, placing the children of intact, well-to-do lesbian households up against children reared by single heterosexual parents.
In all aspects of gay marriage, Kass and Mansfield write, the “body of research . . . is radically inconclusive.” There’s good reason for this, aside from the suspect motives and methods of the researchers themselves. Same-sex marriage and child rearing by self-defined same-sex couples are recent innovations. Whatever effects may flow from these unprecedented arrangements, good or bad or neutral, they are scientifically unknowable until gay marriage and child rearing are widespread enough to yield large samples that can be studied according to a rigorous methodology. “Large amounts of data collected over decades,” write Kass and Mansfield, “would be required before any responsible researcher could make meaningful scientific estimates of the effects.” And on these issues disinterested researchers are hard to come by.