The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent

Dobbs and the Holdings of Roe and Casey

Reactions to readers, and reflections on Dobbs.


I'd like to thank Eugene one more time for giving me the opportunity to blog this week at The Volokh Conspiracy. And I'd also like to thank TVC's readers. I came to see my project differently just in the course of writing my posts this week. And your comments have given me even more food for thought. (For those interested in the Georgetown Journal of Law and Public Policy article I'll publish later, I'll rewrite the draft I have now thoroughly on the basis of these posts and your comments. I expect to upload a revised draft by the beginning of February.)

Dobbs v. Jackson Women's Health Center is currently before the Supreme Court, and the litigants and most onlookers believe that the case presents the question whether Roe v. Wade (1973) and Planned Parenthood of Southeastern Pennsylvania v. Casey (1992) should be reaffirmed or overruled. During oral argument in Dobbs, Chief Justice Roberts explored whether those cases entitle women to fair opportunities to get abortions, but not necessarily up to the thresholds for viability. That interpretation goes against the cases' conventional readings, that pregnant women are presumptively entitled by federal substantive due process to get abortions up to the time of fetal viability.

Although one can certainly understand why the Chief Justice explored this interpretation, as Monday's post claimed it is not a narrow reading but a rewriting of Roe and Casey. As Tuesday's post showed, the judgment in Roe was a declaratory judgment declaring that the challenged Texas abortion statutes were unconstitutional on their faces. As Tuesday's post also showed, necessary to that judgment was some reason for decision like this: Prima facie, a state law violates federal substantive due process if it restrains pregnant women's opportunities to abort before viability.

As Wednesday's post showed, necessary to Roe's judgment was one other reason for decision: A state statute is facially unconstitutional if it violates the right protected by the first reason, and if the violations seem substantial in relation to the number of situations in which the statute could be enforced constitutionally. As yesterday's post showed, Casey and another 11 Court decisions applied those two reasons for decision to declare other state abortion restrictions unconstitutional on their faces. Since the Mississippi Gestational Age Act ("the MS GAA," the law challenged in Dobbs) prohibits most abortions after 15 weeks, unless Roe and the cases discussed yesterday are overruled, their holdings require that the GAA be declared unconstitutional on its face.

Today, I'll respond to feedback I've received this week. ReaderY wonders, because "Roe had ordinary standing," "it's not clear … that overbreadth was anything to anything the Roe court did." I agree with ReaderY that one effect of overbreadth is to relax ordinary standing rules. If ReaderY is claiming that standing relaxation is the only effect of overbreadth, though, we disagree. As Wednesday's post explained, overbreadth has two other significant effects. When a party (with standing) challenges the constitutionality of a statute, overbreadth doctrine entitles her to criticize the statute for the effects it has on third parties. And if a law turns out to be overbroad, it is unconstitutional not only as applied to the challenger but on its face. And that takes me back to the most important claim I made on Wednesday. To understand the holding of Roe, an inquiring lawyer needs to start with the judgment—a declaratory judgment, that the challenged Texas statutes were unconstitutional on their faces. To declare the statutes facially unconstitutional, in Roe the Court needed to find that they "swe[pt] too broadly."

Richard Re asked me some questions offline about my restatement of overbreadth doctrine on Wednesday, and his questions convince me that I should offer a correction and an observation. As I explained Wednesday, standard black-letter overbreadth doctrine has two requirements. A statute is unconstitutional on its face if its illegitimate applications seem substantial in relation to its constitutional applications, and if there is no readily-apparent way to construe the statute narrowly to avoid the constitutional problems. The correction: On Wednesday, I attributed both of these requirements to Broadrick v. Oklahoma (1973). I was wrong to attribute the second requirement to Broadrick. It comes from cases like Erzoznik v. City of Jacksonville (1975), and Dombrowski v. Pfister (1965).

The observation: Although I haven't reviewed overbreadth case law comprehensively, my impression is that Supreme Court abortion cases apply the "no readily-apparent narrowing construction" requirement much less often than other overbreadth cases do. The requirement is considered in the partial-birth abortion case Stenberg v. Carhart (2000), but not in Roe or most of the other cases. I'm not sure whether my impression is right. Or what it means if it is right. I'm curious what readers think.

Here is another possible objection. In these posts, I've focused on the Court cases (and passages of Court cases) declaring state abortion restrictions unconstitutional. What about the cases (and passages) in which state restrictions were declared constitutional? (For one example, picked only because it's relatively recent, Ayotte v. Planned Parenthood of Northern New England (2006), which upheld from challenge a state parental-notification statute.) I have two answers. First, the cases I've studied here create a conflict between the MS GAA and the holdings of 13 cases. Cases like Ayotte don't make the conflict any more worse, but they certainly don't get rid of it.

Second, all of those statute-upholding cases are distinguishable. As Tuesday's post showed, when a statute purports to regulate abortion pre-viability without banning it, Roe and subsequent cases require courts to apply three legal propositions. First, that the restriction is restraining a constitutional right prima facie. Second, if the statute could arguably be classified as a law regulating abortion without prohibiting it, the court must determine whether it passes muster as a maternal-health measure or some other justified regulation. (In Ayotte, the challenged law constituted a justified regulation because it secured parental rights without unduly burdening the rights of pregnant minors.) But if the statute does not pass muster as a regulation, then courts (third) need to inquire whether it is substantially overbroad, whether it restrains a substantial number of pre-viability protected abortions. The MS GAA could not possibly pass muster as such a regulation—because it prohibits most abortions after 15 weeks. The cases in which the Court takes the off-ramp at proposition (2) are distinguishable from the cases that stay on the road and go off the cliff past proposition (3).

That response takes me to an objection raised by Lee Moore. On Tuesday, I argued that Roe's discussion of second-trimester maternal-health regulations was dicta. Moore didn't follow my argument. Rereading what I wrote, I agree with Moore that I didn't lay my argument out clearly, and I'd like to try again. A dictum is a proposition in a case not necessary to support the judgment in the case. Roe's discussions of second-trimester maternal-health regulations were not necessary to support the judgment that the challenged statutes were facially unconstitutional. The Texas statutes prohibited all abortions except those necessary to preserve the pregnant woman's life. The passages Moore asks about indicated that abortion restrictions might be constitutional if they related reasonably to maternal health for second-trimester pregnancies. Since the challenged statutes prohibited most abortions, there was no way they could ever have been classified plausibly as regulations, laws that kept abortion safe while keeping it legal. So nothing the Court said in Roe about second-trimester maternal-health regulation was necessary to support its judgment that the statutes were facially unconstitutional. (For the record, much of that dicta became reasons for decision the very same day—in Roe's companion case, Doe v. Bolton (1973). I discussed these passages of Roe just to illustrate the distinction between a dictum and a reason for decision.)

Assume I have shown what I set out to show at the beginning of this week. Readers may wonder: What effect do my arguments have on Dobbs? My arguments confirm the conventional wisdom that held while Dobbs was being argued: The MS GAA conflicts with the holdings and the main reasons for decision from Roe and Casey, and the main options before the Court are to reaffirm or to overrule Roe and Casey. But I doubt very much that my arguments tilt the scales on whether Roe and Casey should be reaffirmed or overruled.

When high courts consider stare decisis questions, they balance many different factors. Two matter here (I get them from Janus v. AFSCME (2018)): the quality of a challenged decision's reasoning, and its consistency with other related decisions. For supporters of Roe and Casey, those decisions and 11 other cases have applied two reasons for decision consistently, and that consistency cuts in favor of standing by Roe and Casey. Critics believe that Roe and Casey are very poorly-reasoned, so much so that none of the other stare decisis factors can save them. But these posts show that overbreadth doctrine is an exception to more basic principles about standing and remedies. These posts also suggest that abortion cases apply a particularly hard-edged form of overbreadth analysis. So critics might come away thinking that one of the two main reasons for decision in Roe and later cases is not consistent—with principles that federal courts usually follow when they issue remedies and hand down judgments.

In short, I'm pessimistic that there's a satisfying third option in Dobbs. I'll close with one last objection by ReaderY (whom I thank for staying with me all week!): In the real world "[t]here almost always more than two options." A lot turns on what ReaderY and I mean by an "option." If the word means something like "a choice an actor might conceivably take," yes, Justices have many options and not just two or three. When I talk about "options," though, I probably assume that the word has an ideal meaning. (An "option" implies a "promising" choice, or a "satisfying" choice.) And in that sense I doubt there are more than two real options.

Readers were surely aware of some of these doubts before this week. Roe and Casey have really, really settled conventional readings. (Just go and read opinions by lower court federal judges applying them—like the opinions in the Fifth Circuit in Dobbs itself.) Any "third way" strategy bears a burden of showing why the conventional readings have been wrong for 30 years and more. Next, for better or worse, people tend to be familiar with the upsides and downsides of conventional options. By definition, unconventional options are less well-known. People can hope rashly that options they don't understand will work out better than options they know warts and all. The theory I've studied this week suffers from serious legal problems. But those problems are hard to grasp until one sees how different it is to spot a reason for decision in a successful overbreadth challenge than in a successful as-applied challenge.

To be sure, there still may be some fourth or fifth way to decide Dobbs without completely reaffirming or overruling Roe and Casey. But assume some such way surfaces between now and this summer. Inquiring minds should ask: Why wasn't it proposed earlier? If it's so promising, why didn't it grab anyone's interest before the rule of decision studied here was explored during oral argument in Dobbs? And since such a proposed rule of decision would probably come out of nowhere, what problems would it have and how difficult would they be to spot?

Obviously, I think that the post-argument conventional wisdom about Dobbs is right. The exploratory theory discussed here is in substance a rewrite of Roe and Casey. I doubt the Justices will find any more promising rule of decision if they give up on the rewrite and search for some as-yet-unarticulated fourth or fifth way to resolve Dobbs. Better to choose between the two obvious options—to overrule Roe and Casey, or to reaffirm them.