And on affirmative action, the U.S. Supreme Court's holding is . . . ticktock.
That's the sound of the countdown clock on race-based college admissions that Justice Sandra Day O'Connor started in 2003, when she wrote for the court that she expected such policies might be necessary (and therefore constitutional) until 2028.
In Fisher v. Texas, Justice Anthony Kennedy, writing for a 7-1 majority, declined to strike down the University of Texas's use of race as a factor in admissions. Yet he strengthened the requirement that the state prove its program was strictly necessary, signaling that the end might be approaching.
The decision gave something to everyone -- and signaled that Kennedy is, despite himself, influenced by O'Connor's frank pragmatism even though she's no longer on the court.
The fact that Texas's plan wasn't struck down counts as a win for affirmative-action advocates, who feared Kennedy might provide the deciding vote to require perfect race neutrality. It also explains why the liberal Justices Stephen Breyer and Sonia Sotomayor joined the decision.
When O'Connor retired and was replaced by Justice Samuel Alito, there was reason to think that the delicate compromise she had crafted -- in essence preserving affirmative action while noting that its days were numbered -- wouldn't last long. After all, Kennedy, who became the court's swing vote post- O'Connor, had joined the main dissent in that landmark 2003 case -- Grutter v. Bollinger -- in which then-Chief Justice William Rehnquist would have struck down the University of Michigan's "race-plus" admissions system.
In deciding to take the Fisher case in October, the court's conservatives were seen as signaling that they expected Kennedy would join them in overturning race-conscious admissions for good. That he didn't do so -- yet -- should give supporters of affirmative action some hope.
At the same time, Kennedy held that the lower courts had made a mistake by not closely scrutinizing whether the use of race as an admissions factor was truly necessary to achieve the legitimate goal of educational diversity. This opinion echoed his separate dissent in Grutter, in which he also complained that the lower courts hadn't properly engaged in the analysis known as "strict scrutiny" that is required when a government program takes race into account.
Now, in Fisher, Kennedy sent the case back down to the district court to redo its analysis. And he did it with special emphasis. The court should defer to the University of Texas's educational expertise in claiming the importance of diversity, Kennedy said. But when it came to the specific program adopted to achieve that diversity, the court should not defer at all: "The reviewing court must ultimately be satisfied that no workable race-neutral alternatives would produce the educational benefits of diversity."
That is a very high requirement, and a win for affirmative-action's opponents. It sets the stage for the lower courts potentially to find that Texas could achieve diversity without expressly considering race. Indeed, Texas has tried to do so through its Top 10 Percent Plan, which guarantees admission to the University of Texas to the top 1/10 of students from all Texas high schools -- which are so heavily racially segregated that some racial diversity is necessarily achieved.
In the next phase of the case, and thus of the affirmative action saga, the state will have to argue to the district court that there is no way it could achieve racial diversity without in some way taking account of race. Reading the tea leaves, the district court might well say there must be some other way to accomplish this.
But what way, exactly? Justice Ruth Bader Ginsburg, today's lone dissenter, pointed out that even the 10-percent plan isn't really racially neutral, because it was designed precisely to achieve racial diversity. This question Kennedy left unanswered.
And given that the 10-percent plan hasn't successfully produced enough diversity to satisfy the state, who gets to decide how much diversity is enough? Kennedy held that it is up to the state to say that diversity is valuable. Does this imply that it is up to the state to say how much diversity is needed? That question, too, remains to be answered.
The upshot of the Fisher holding is that if Kennedy does want to strike down affirmative action, he wants to do it gradually, stepwise, over several more years of litigation. That is exactly what O'Connor wanted with her highly unusual 25-year clock.
Kennedy tends to prefer principles to such explicit pragmatic compromise. But here, with the chance to make headlines by striking down affirmative action, he acted like O'Connor.
That gives us a clue for what Kennedy might do in the two same-sex marriage cases that still loom. If he is foisting gradualism on conservatives who want to end affirmative action, he may be preparing to foist gradualism on liberals who want full marriage rights for gay people right now. That would require Kennedy to eschew some of his grandly principled holdings in earlier fundamental rights cases -- but it would further consolidate his position as the most powerful justice since, well, Sandra Day O'Connor.
Noah Feldman, a law professor at Harvard University and the author of "Cool War: The Future of Global Competition," is a Bloomberg View columnist.