Orin Kerr keeps asking how much the factual record matters in Perry v Schwarzenegger:

Appellate judges will naturally discount, if not entirely ignore, efforts to create false certainty out of unknowns by stating that they are facts. To pick an extreme example, imagine Judge Walker made a factual finding that “the defendant’s position that Prop 8 is unconstitutional is incorrect.” If appellate courts have to defer to everything a District Judge labels a fact, then would that mean the appellate courts have to defer to this judgment? If so, that would make the constitutionality of same-sex marriage entirely up to the discretion of the District Court Judge. The judge could make his decision unreviewable either way by presenting his legal conclusion as a fact. And what if two District Court judges disagree on the factual findings? What is the Supreme Court supposed to do to reconcile opposite conclusions? For such reasons, the fact section of Judge Walker’s opinion is likely to matter a lot less than it would normally matter in appellate litigation.

Is it an "unknown" that there are no procreative conditions for a marriage license in California? Is it an "unknown" that gay couples are as capable of committing to one another and taking care of one another as heterosexuals? Is it an unknown that civil marriage is not a religious institution and requires no religious test and is indeed available to atheists? Those are the facts Walker is pointing out. I don't doubt that they can be reviewed; but how could they be denied? Kerr points to this Kennedy ruling:

A State . . . has no obligation to produce evidence to sustain the rationality of a statutory classification. “[A] legislative choice is not subject to courtroom factfinding and may be based on rational speculation unsupported by evidence or empirical data.” A statute is presumed constitutional, , and “[t]he burden is on the one attacking the legislative arrangement to negative every conceivable basis which might support it,” whether or not the basis has a foundation in the record. Finally, courts are compelled under rational-basis review to accept a legislature’s generalizations even when there is an imperfect fit between means and ends. A classification does not fail rational-basis review because it “ ‘is not made with mathematical nicety or because in practice it results in some inequality.’ “The problems of government are practical ones and may justify, if they do not require, rough accommodations-illogical, it may be, and unscientific.”

But in the same decision, Kennedy also writes

the standard of rationality as we so often have defined it must find some footing in the realities of the subject addressed by the legislation.

That's the point Walker is making. And the defenders of Prop 8 could not provide any "footing in the realities of the subject". (I'm assuming no difference here between a law and a Proposition and I may be wrong. For the record, the California legislature passed a marriage bill twice, only to be vetoed by the governor.) Kerr's strongest point is that a prediction of no social harm cannot be seen as a fact. But that merely brings us back to the same pure-tradition argument, which argues that a mere intuition of social damage, even if based in no historical or sociological reality, is enough to justify a proposition that ensured a minority be denied equal status with a majority. Maybe Kennedy believes that. But he didn't seem to in Lawrence or Romer.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.