So it’s over. Seriously. It is hard to find a simple workaround to rescue affirmative action in college admission after the 6-3 Supreme Court ruling on Thursday that the programmes at Harvard and the University of North Carolina violate the 14th Amendment. Oh, given time, we’ll come up with something. But Chief Justice John Roberts’s majority opinion paints so broadly that those of us who hope to find a path around it will have to be exceedingly clever.

Let’s start with what the majority got right.

First, as Roberts writes, colleges and universities should not be allowed to “establish quotas for members of certain racial groups or put members of those groups on separate admissions tracks” — a quotation drawn from the court’s 2003 opinion in Grutter v Bollinger. I agree. Not only are quotas intrinsically bad, but as the court points out, they lead to confusion, such as casually lumping East Asians and South Asians together, and fights over who counts as Hispanic. (Also, although Roberts does not mention the point, over who counts as Black.)

Second, the majority is at least arguably correct that a school violates the Constitution when it acts as though “minority students always — or even consistently — express some characteristic minority viewpoint on any issue.” True! There’s no such thing as, for instance, the Black point of view, and to suggest there is — as we’ll see — is dangerous.

So those two points the court got right.

The rest is mushy mealy hasty pudding.

Let's unpack the ingredients.

Using race is subject to strict scrutiny — check. The court has hardly ever approved it, except as a narrow remedy for particular discriminatory acts — check.

After that, things get ... well, subjective. The central point seems to be that an admission programme that treats being a specified race as a plus ipso facto treats being a member of another race as a negative. Wrong, says the dissent. But as long as the number of places in the entering class is limited, the majority is right. It may not be pleasant to contemplate, but that does not make it untrue.

That also doesn’t make it unconstitutional. Even strict scrutiny is a balancing test. What matters is not whether the selection process downgrades some applicants, unfortunate though that may be; it’s the strength of the argument presented in the programme’s favour.

I support affirmative action, but I have long worried about some of the terms in which it is justified. For example, some say students of colour are needed on campus to bring viewpoint diversity. But I object, strongly, to essentialism of this kind, with its implication that there is, say, a correct way to “think Black” — not least because the further necessary inference is that those who fail to think the right way are, by virtue of their failure, less Black. Or, in the existing absurd argot, inauthentic. Such notions are dangerous, to say nothing of unacademic, and colleges and universities should not be endorsing them, even implicitly. Affirmative action ought to be a tool for enhancing, not restricting, the freedom of thought of those it purports to help.

The more persuasive case is the claim for democratisation — not in the sense of making sure every institution “looks like America”, but in the sense of giving students from diverse backgrounds opportunities to show what they can do. Not guarantees, mind — opportunities. We as a nation have come a long way, but we remain awash in racial stereotypes. We cannot wish them away. But giving those who suffer from stereotype a chance to shine is a strong constitutional rationale.

That is why the thumb on the scale for admission is more important than the thumb on the scale for, say, Phi Beta Kappa or law review. At application time, you get a plus. A small plus, but a plus. The rest has to be earned.

Some argue in response that we need broader measures of diversity. I agree. I’m in favour of taking into account not only, say, race or gender identity, but also poverty and geography and lots of other factors. Yes, schools often do this, but more idiosyncratically than systematically. And one could argue that they should expand their reach. For instance, I would be comfortable if being a White evangelical Christian put a positive thumb on the admission scale, too, because we don’t get many of those in the Ivies these days. I continue to believe the case for race to be by far the more urgent; but that doesn’t mean it’s the only case that can be made.

The difficulty with the court’s approach is that even a programme of broader reach would likely prove unsatisfactory. The majority’s effort to sketch one reads like a bad joke: “[A]s all parties agree, nothing in this opinion should be construed as prohibiting universities from considering an applicant’s discussion of how race affected his or her life, be it through discrimination, inspiration, or otherwise.”

Ha, ha, very funny! Because just a few pages earlier, Roberts is excoriating Harvard and UNC for not operating their affirmative-action programmes according to criteria that are “sufficiently measurable to permit judicial [review]”. Yet one is hard-pressed to see how this requirement can be met by a school that says, “We liked this essay better than that one.”

No, I think the broader programme, too, might be down the drain.

I’m not insensitive to the concerns of those who do not like racially conscious programmes, and I don’t think opposing them marks one as a tool of White supremacy. Every selection rule disfavours someone, and no matter how justified a programme may seem in the abstract, it is different when it is your children. So I get it. And I quite agree that some colleges have taken matters too far. But unlike the court majority, I don’t think we have reached that fabled moment when we really can put all consideration of race behind us. That’s why, for now, we still need this imperfect solution to a compelling problem.

I myself am a proud product of affirmative action. Absent a thumb on the scales, it is unlikely I would have been admitted to Yale Law School. I have done my best, ever since, to secure opportunities for others — not only in admission but in hiring research and teaching assistants. No quotas — but a thumb on the scale. In this I am far from alone. If six Supreme Court justices want to tell me I have been doing it wrong over the past four decades, they are welcome to their views. But if they want to persuade me, they will need a much better argument than presented.

• Stephen L. Carter is a Bloomberg Opinion columnist. A professor of law at Yale University, he is author, most recently, of Invisible: The Story of the Black Woman Lawyer Who Took Down America's Most Powerful Mobster