In the most recent issue of the Manhattan Institute's City Journal, Jim Manzi has a fascinating piece on the limits of social science. I posted my initial reaction to the piece on Marquette University Law School's faculty blog. His observation are particularly timely in light of the recent decision of a district court judge in Perry v. Schwarzenegger striking down California's constitutional amendment limiting marriage to one man and one woman.
My point relates not to same sex marriage itself, but Judge Walker's use of social science evidence and how that might relate to constitutional litigation generally. The key question is the extent to which courts ought to rely on social science evidence to overturn the laws enacted by voters and their elected representatives. This question extends well beyond the matter of same sex marriage.
According to Judge Walker, social science evidence has proven that same sex marriage will have no impact on the mores and vitality of marriage between heterosexuals and that it has proven "beyond any doubt that parents' genders are irrelevant to children's developmental outcomes."
It is certainly true that most academics who have devoted themselves to the study of gay and lesbian isses believe these assertions. But it is quite another matter to say that they have proven them. Our experience with same sex marriage is quite recent and limited. It is not clear that there is enough data to draw any conclusions about its long term impact.
As for the "genderless" nature of parenting, there are severe methodological difficulties in obtaining randomly selected comparable samples of gay and lesbian couples raising children. Comparision groups of heterosexual parents are often not limited to intact married couples raising their biological tradition and there are very few, if any, adequate longitundinal studies.
But beyond this, Manzi's article suggests that it may be difficult to ever say - at least by the standards of the physical science - what the impacts of same sex marriage are or will be. While his article focused on implications for policy making, there are lessons for judicial decisionmaking as well. A large part of Judge Walker's decision is given over to declaring the judgment of a majority of California's voters as irrational because social science purports to establish that what they believe to be true is not.
The question is not an easy one. Traditional equal protection analysis requires some scrutiny of the justificaton for legal distinctions. But it may well be that social science evidence needs to be viewed with greater skepticism than shown by Judge Walker in Perry. Constitutional litigation ought not to be seen as a battle of experts who attempt to claim the dispassion and robustness of the physical sciences to resolve contested issues of social policy.