Doctrine and Meaning: Harder Cases

What I’ve said so far is that I think the legitimacy of doctrine should be evaluated by starting with what I’ve called constitutional meaning and asking whether the doctrine is a reasonable way to implement that meaning. An example of doctrine that’s easy to justify is heightened scrutiny for discrimination against women or racial minorities. I think that this heightened scrutiny overenforces the constitutional meaning—it will strike down some discriminatory acts that do not violate the constitutional meaning of the Equal Protection Clause—but overenforcement is justified because there’s a history of invidious discrimination against women and minorities, and both groups are underrepresented in legislative bodies.

Now I’m going to talk about some harder cases that I think are right: Lawrence v. Texas and Roe v. Wade. These are Due Process cases, so the first issue is what the meaning of the Due Process Clause is. The idea that the clause contains some substantive protections is very old. In its original form, it worked by setting limits on governmental power (rather than by identifying rights that defeated otherwise-valid exercises of government power). It did this by a priori deductions about the sorts of powers people would not want their government to exercise and would therefore not delegate to it: the power, for instance, to take property from A and give it to B. A government act purporting to accomplish such a thing might comply with formal requirements—it might be enacted in a procedurally valid manner—but it would not be a law, and therefore a constitutional provision prohibiting deprivations of liberty without due process of law could be used to strike it down.

The most general limit on the police power was the requirement that the government act in the public interest. At one point this was understood to mean that legislation had to confer benefits generally, rather than on particular groups, and the attempt to enforce this understanding is a large part of the story of the Lochner era, as Howard Gillman has shown. A more relaxed version would require only that the law produce net benefits, regardless of the distribution of benefits and burdens. That is the constitutional meaning I will be using.

This requirement, of course, is basically the requirement that laws be good policy. Since legislatures, for reasons of institutional competence and electoral accountability, are generally better at making policy judgments, the modern Court has generally adopted very deferential doctrine to enforce the public interest requirement. It will generally uphold legislative balancing of costs and benefits if the legislature might rationally have believed its choice to serve the public interest. So the key question for generating Due Process doctrine, I think, is whether there is some reason not to extend the ordinary deference in particular cases. This question is much like the question of when you should apply heightened scrutiny in Equal Protection cases.

Lawrence (which does not defer, though it does not explicitly apply heightened scrutiny either) is relatively easy from this perspective. More searching scrutiny for laws burdening the interests of gays and lesbians is justified for the same reasons explicitly heightened Equal Protection scrutiny is justified for discrimination against women and racial minorities. There are still significant portions of the population who think, for instance, that a criminal ban on sodomy is an effective means of protecting some important state interest, and not an expression of animus, but the Court seems to have decided that that view is going to lose in the long run. I think it’s correct in that estimation, which makes the decisions right, but I think these are somewhat harder cases than discrimination against women or minorities because the new consensus hasn’t formed yet.

Roe is substantially harder. What I’ve said about the meaning of the Due Process Clause should suggest that I don’t view it as a repository of unenumerated fundamental rights. I don’t think the search for unenumerated rights is necessarily illegitimate, but it would make more sense to characterize that jurisprudence as enforcement of the 9^th Amendment and the 14^th Amendment’s Privileges or Immunities Clause. Still, even granting (as I do, but many people don’t) that these provisions, taken together, suggest the existence of unenumerated constitutional rights against state governments, there remains the question of how we justify allocating decisionmaking authority to judges rather than legislatures. Heightened scrutiny in modern fundamental rights jurisprudence is justified by the “fundamental” nature of the right, which makes the crucial question the distinction between fundamental rights and mere “liberty interests.” But I don’t think anyone has come up with a particularly satisfying way of determining fundamentality, much less one at which judges are clearly better than legislatures.

The way I prefer to analyze these issues is to suppose that the constitutional requirement is that the legislature act in the public interest, in such a way that the benefits of a law exceed its burdens. Translating this requirement into doctrine is primarily a matter of determining when the legislature can be trusted. In the abortion context, the fact that the burden falls most heavily on a relatively politically weak group (not women generally, but younger women, and perhaps especially younger, single, lower-income women) suggests that some degree of judicial second-guessing might be justified. One way to do this would be conventional heightened scrutiny, which would get you something like Roe or Casey. A more finely-tuned approach to evaluating the legislative balancing of life vs. liberty might be to look at the way in which states generally handle the life/liberty tradeoff—do they have good Samaritan laws, mandatory blood or postmortem organ donations, etc.? Such an analysis suggests that when the liberty of all is involved, states generally decline to impose even fairly minimal burdens in order to promote life. And that in turn suggests that women’s liberty is being discounted in an impermissible way. (You could boil this down to the point that if men could get pregnant, abortion would be legal. Or you could expand it by reading Jack’s recent book, What Roe v. Wade Should Have Said..)

Roe and Lawrence are examples of cases where I think there are plausible reasons not to defer to the legislative judgment. You can start with a plausible constitutional meaning and tell a plausible story about why you’ve chosen a non-deferential doctrinal rule to enforce that meaning. I’m not going to discuss the Commerce Clause cases here, but I think it’s harder to do the same thing with Morrison and Lopez, which may be why that doctrinal experiment seems to be collapsing. And I think it’s impossible to do with my last example, strict scrutiny for race-based affirmative action.

I’ve said before that I take the core meaning of the Equal Protection clause to be a prohibition on discrimination inspired by animus—discrimination designed to stigmatize or inflict harm for its own sake. We could also read it as prohibiting unjustified discrimination, and operationalize that by demanding that the benefits to society generally exceed the costs inflicted on the burdened group. Having decided on that constitutional meaning, courts then face the question of when to leave this balancing question with the legislature (the default rule) and when to second-guess the legislative assessment. None of the ordinary reasons—ignorance about the characteristics or interests of the burdened group, indifference to their welfare, distribution of political power—will support heightened scrutiny for race-based affirmative action. The decision of a majority to burden itself in order to benefit an underrepresented majority is the least plausible decision for judges to second-guess.

My perspective on the creation of doctrine suggests that strict scrutiny for race-based affirmative is of dubious legitimacy. It could of course be defended by an alternate account of constitutional meaning—namely, that what the Equal Protection Clause means is that no race-based discrimination is permitted. That is how conservative Justices tend to write. But that is not what the text says, and I have not seen any serious attempt to argue that the drafters thought they were forbidding all differential treatment based on race. (There are interesting originalist things to say about the scope of the Equal Protection Clause, namely its application to civil rather than political or social rights, and to rights rather than benefits, but those do not support heightened scrutiny for affirmative action either.)

I think strict scrutiny for race-based affirmative action is a strong candidate for the least legitimate piece of current constitutional doctrine. I expand on that point in the book, and I also offer some thoughts about why we ended up where we have—namely that the Court has lost sight of the distinction between doctrine and meaning—but I don’t think I can present those points intelligibly in this space. Indeed, I have some doubts about the intelligibility of what I’ve presented thus far—it makes me think that perhaps I should actually finish the book before trying to summarize it. But, if you’ve read this far, thanks, and I hope you keep your eyes open for the book, which should be coming out next summer. The current working title is This Honorable Court.