Will Wilkinson at The Economist takes down the "gay couples are bad—or less good—parents" justification for banning gay marriage:
Let's suppose, though, that there is a credible basis for the proposition "that gay people, on average, are less successful as parents", and that this has something to do with the gay-marriage debate we have been having here in Iowa. What then? Consider an analogy. There is evidence that people in dire poverty are, for a number of reasons, less "successful" as parents. Suppose some of us therefore proposed banning marriages between poor people. The first argument against this proposal is that the right to marry should not depend on membership in a class that is, on average, as successful at parenting as other classes. The second argument is that stripping poor people of the right to marry strips them of legal equality and what John Rawls, the great political philosopher, called "the social bases of self-respect". This harmful injustice would be suffered by the whole class marginalised by official discrimination, but it would be especially salient in the case of exemplary poor families clearly deserving of equal standing, recognition, and social esteem.
For the record: there is no credible evidence that gay people are, on average, "less successful as parents." Quite to the contrary: the evidence shows that gay parents are just as good... or better. (Wilkinson via Sullivan.)