NELSON LUND: A SOCIAL EXPERIMENT WITHOUT SCIENCE BEHIND IT

Advocates of same-sex marriages can’t back up claims about positive long-term effects.

By Nelson Lund (March 26, 2013)

The Supreme Court is hearing two cases this week that represent a challenge to one of the oldest and most fundamental institutions of our civilization. In Hollingsworth v. Perry and United States v. Windsor, the court is being asked to rule that constitutional equal protection requires the government to open marriage to same-sex couples.

The claimed right to same-sex marriage is not in the Constitution or in the court’s precedents, so the court must decide whether to impose a new law making marriage into a new and different institution. The justices are unlikely to take so momentous a step unless they are persuaded that granting this new right to same-sex couples will not harm children or ultimately undermine the health of our society.

A significant number of organizations representing social and behavioral scientists have filed briefs promising the court that there is nothing to worry about. These assurances have no scientific foundation. Same-sex marriage is brand new, and child rearing by same-sex couples remains rare. Even if both phenomena were far more common, large amounts of data collected over decades would be required before any responsible researcher could make meaningful scientific estimates of the long-term effects of redefining marriage.

The conclusions in the research literature typically amount at best to claims that a particular study found “no evidence” of bad effects from child rearing by same-sex couples. One could just as easily say that there is no reliable evidence that such child-rearing practices are beneficial or harmless. And that is the conclusion that should be relevant to the court.

Social-science advocacy organizations, however, have promoted the myth that a lack of evidence, so far, of bad effects implies the nonexistence of such effects. This myth is based on conjecture or faith, not science.

Nor is the leap of faith from “no evidence” to “don’t worry” an accident. The late Sen. Daniel Patrick Moynihan, himself a distinguished social scientist at Harvard, once observed: “Social science is rarely dispassionate, and social scientists are frequently caught up in the politics which their work necessarily involves . . . [T]he pronounced ‘liberal’ orientation of sociology, psychology, political science, and similar fields is well established.”

This orientation has been on rich display in the research on same-sex parenting, which is scientific primarily in the sense that it is typically conducted by people with postgraduate degrees. There are no scientifically reliable studies at all, nor could there be, given the available data. Yet the Supreme Court has been solemnly assured by many scientific organizations, such as the American Psychological Association and the American Academy of Pediatrics, that the overwhelming weight of evidence indicates that same-sex couples are every bit the equal of opposite-sex parents in every relevant respect. The number of studies may be overwhelming but the evidence assuredly is not.

The prominent National Longitudinal Lesbian Family Study, for instance, relied on a sample recruited entirely at lesbian events, in women’s bookstores and through lesbian newspapers. Other studies relied on samples as small as 18 or 33 or 44 cases. The effect of parenting by male homosexual couples remains in the realm of anecdotes. Most research has relied on reports by parents about their children’s well-being while the children were still under the care of those parents. Even a social scientist should be able to recognize that parents’ evaluations of their own success as parents might be a little skewed.

In 2012, sociologist Loren Marks conducted a detailed re-analysis of 59 studies of parenting by gays and lesbians that were cited by the American Psychological Association in a 2005 publication. Mr. Marks, who teaches at Louisiana State University, concluded that the association drew inferences that were not empirically warranted.

There has been only one study using a large randomized sample, objective measures of well-being, and reports of grown children rather than their parents. This research, by Mark Regnerus, a sociologist at the University of Texas Austin, found that children raised in a household where a parent was involved in a same-sex romantic relationship were at a significant disadvantage with respect to a number of indicators of well being—such as depression, educational attainment and criminal behavior—compared with children of intact biological families.

One might expect this work at least to raise a caution flag, but it has been vociferously attacked on methodological grounds by the same organizations that tout the value of politically congenial research that suffers from more severe methodological shortcomings. This is what one expects from activists, not scientists.

As everyone knows, some states have begun to experiment with legalizing same-sex marriage, and public opinion seems to be shifting in favor of the change. Perhaps this will work out well, and the overwhelming majority of states that have been more cautious will eventually catch up. But experiments are never guaranteed to succeed, and one advantage of democracy is that it allows failed experiments to be abandoned. If the Supreme Court constitutionalizes a right to same-sex marriage, however, there will be no going back. The court cannot possibly know that it is safe to take this irrevocable step.

Mr. Lund is a professor at George Mason University School of Law in Arlington, Va. This article is based on an amicus curiae brief in support of the petitioners in Hollingsworth v. Perry, filed on behalf of Leon R. Kass (University of Chicago), Harvey C. Mansfield (Harvard University), and the Institute for Marriage and Public Policy.