« Another day, another NFL protest | Main | Law Professor Letter to Flood and McGahn »

Tuesday, June 05, 2018

The Marks Rule’s Fate After Hughes 

Hughes v. United States is out, and it’s a disappointment for those of us following the law of fragmented decisions. Though the Hughes oral argument featured by far the Court’s most thorough discussion of the Marks rule, the justices ultimately chose not to address the precedential significance of 4-1-4 decisions like Freeman v. United States. Instead, the Court revisited the merits and set a new majority rule, much as the Court has done in some past decisions that posed Marks questions. Interestingly, however, the Court’s ruling still managed to undermine one of the main defenses of the Marks rule.

Regular readers know that I oppose the Marks rule, as explained in a forthcoming article and an amicus brief in Hughes. And there were several reasons to think that, if the Court was ever going to address the Marks rule’s problems, this case would be the time to do so. The parties and amici had identified numerous circuit splits on how to apply the rule. The Court had granted review on not one but two Marks questions. And the US Solicitor General had specifically requested that the Court issue a Marks ruling to end the confusion. Yet the Court said not a word to alleviate the problem. 

Oddly, the Court did not explain why it was free to avoid deciding the Marks question, even though a Marks holding in Freeman would arguably have bound the justices. In past cases, however, the Court had set Marks aside in part because its application had badly divided lower courts, and that factor was present in Hughes as well. The decision to take a hard pass on Marks probably also resulted in part from a combination of end-of-term time pressure and an inability to form a majority on just what to do with Marks. Needless to say, it would have been awkward if the Court had issued a fragmented decision on the meaning of fragmented decisions. 

Perhaps the justices plan to take some other corrective action now that they are so well aware of the confusion that the Marks rule is causing.

One option would be to seek out another, even better vehicle to provide clarity in this area. But, again, the Court has consistently declined to clarify the meaning of Marks, and lower courts have long been left to struggle and generate circuit splits. After Hughes, I am not optimistic that the justices will do better in the years ahead. As a result, the fate of the Marks rule could ultimately be worked out in the courts of appeals. Notably, some circuits, most particularly the DC Circuit and the Ninth Circuit, have already taken steps to narrow the Marks rule, gently ushering it off the stage. Hughes rewarded those courts. 

Another corrective option would be for the justices to work harder to avoid issuing fragmented decisions with unclear precedential implications. It will be interesting to see if we can glean evidence that the justices are taking action along those lines. For example, if Justice Sotomayor were again in the position of being the solo justice in a 4-1-4 decision, she might be more inclined to cast her lot with one side right away, rather than waiting several years for the Marks rule to generate a circuit split. In other words, Hughes might mean that we will see greater reliance on the “Screws rule” (discussed here) rather than the Marks rule.  

In the meantime, Hughes offers some new fodder for thinking about Marks. Perhaps most interestingly, Hughes is something of an embarrassment for one of the main defenses of Marks. Under that view, the Marks rule is desirable because it aligns with the “predictive model” of precedent, whereby lower courts strive to predict the decisions of their judicial superiors. At the Hughes oral argument, Chief Justice Roberts forcefully asserted this view: 

CHIEF JUSTICE ROBERTS:  The first question we posed was how to apply Marks in this situation, and I wonder if I'm a court of appeals judge, it seems to me the most important thing in deciding the case is to make sure that I'm not reversed.  And it seems to me the best way to do that is through the - whatever you want to call it, the walking through, sort of counting out what would happen if you count where the different votes are. And it seems to me if you take any other approach, you're -- you're subject to reversal because, by definition, a majority of the Court here would -- would reach a different result.  

The Chief’s attempt to defend the Marks rule was eyebrow-raising at the time, given the Court’s frequent efforts to discourage lower courts from engaging in predictions when construing Court precedent. But Hughes shows why the Chief’s predictive approach to Marks is problematic on its own terms.  

If a lower court had relied on Freeman to predict how the Court would rule, it would have concluded that a case like Hughes would come out in favor of the government by a vote of 5-4. But, in fact, the vote in Hughes was 6-3 in favor of the defendant. That’s because of two developments, each of which was independently sufficient to negate the predictive value of the Freeman opinions. First, Justice Sotomayor changed her vote from Freeman. Second, a Hughes justice (Gorsuch) voted differently from his predecessor in Freeman (Scalia). So even though the precedent at issue was only seven years old, its opinions turned out to offer a very poor basis for predicting how the justices would rule. 

In short, Hughes doesn’t decide the Marks rule’s fate, but it may play some part in contributing to the rule’s abandonment.

Posted by Richard M. Re on June 5, 2018 at 08:45 AM in 2018 End of Term | Permalink


I agree with Orin that if a majority of the Court actually does agree on how Marks applies in a particular case, it would be great for them to tell the rest of us.

For example, one great part of Richard's article is the set of tables of which cases have produced big lower court splits on what opinion is binding. (Freeman, Santos, Williams v. Illinois and Missouri v. Seibert are all big winners.) If it turns out that in any of those cases the Justices actually knew and agreed all along on which opinion was supposed to be binding -- wow! They could have avoided the split; or even now, maybe they could just resolve it in a summary order.

Posted by: William Baude | Jun 6, 2018 2:41:04 PM

Richard writes: "The main difference, I think, is that I don't believe that the justices would or should have to designate precedent by applying the Marks rule -- that is, by figuring out the 'narrowest' concurring opinion, whatever that might mean. Why would they? "

I think they already do, or at least that was my perhaps-idiosyncratic experience. My sense is that most if not all of the Justices are very much aware of the Marks rule, and they have a good sense of which fragmented opinions are binding on lower courts. They also have a sense of which set of fragmented opinions won't produce any binding decision under Marks. See, e.g., Arizona v. Gant (Scalia, J., concurring) (agreeing to join the majority opinion because his preferred rule would lead to there being no binding decision under Marks). My view is just that the Justices should be explicit about their already-occurring thinking in individual cases, as it turns out lower courts and others aren't as clear on the Marks implications of specific cases as the Justices are.

More broadly, I see Marks as a time-saving rule. I think it runs something like this, with apologies if what I am saying is not at all new in the world of Marks scholarship (which I confess I don't really follow). As I see it, many (if not most) Supreme Court cases raise an issue of whether to rule narrowly or broadly. You could announce a rule or approach that resolves a lot of cases, or you could announce a rule of approach that resolves just a few cases. Marks recognizes the reality that in a world where a majority opinion was needed to make binding law on lower courts, the Justices who wanted a broad approach very likely would eventually agree to join a narrow majority opinion and then just write a concurring opinion saying that they agree with the majority as far as it goes but would rather go beyond that. The difficulty with that approach is that requiring it to make the cert grant worthwhile would suck up time and energy, as there would be some game theory back-and-forth over exact language with Justices being able to deny the other opinion binding status unless they joined. They would get there, eventually. But it would take a lot of time. Marks is a time-saver in that it lets everybody write the opinions they want to, without subsequent back-and-forth, in a way that leads to substantially the same ultimate rule controlling on lower courts.

Posted by: Orin Kerr | Jun 6, 2018 11:32:00 AM

Thanks to all for the comments. Some thoughts below.

Marty - We are on the same page. The Sotomayor concurrence strikes me as quite praiseworthy, particularly for being candid in confessing error -- something the justices rarely do.

Asher - I think your second paragraph is worth underscoring: under the SG's view, Judge Kavanaugh's view, and your view, the choice of Marks theories was quite important when it comes to applying Freeman. And there are COA opinions that likewise basically adopt some version of the "all opinions" approach, so the choice of how to construe the Marks rule mattered as to Freeman. Also, deliberately construing the Freeman plurality (or other opinions) in a way that averts application of Marks is in itself a way of narrowing the practical scope of the Marks rule. In my article, I describe this as two levels of narrowing Marks: at the level of the rule itself and then at the level of application.

Orin - I think you are on to something here, but note that your proposal is already a very significant step toward my view, since you would have the justices themselves expressly assume the responsibility of identifying when precedent formation takes place. The main difference, I think, is that I don't believe that the justices would or should have to designate precedent by applying the Marks rule -- that is, by figuring out the "narrowest" concurring opinion, whatever that might mean. Why would they? They could just designate whatever binding precedent that five of them agree on, for whatever set of (legitimate) reasons. That is what I call the "Screws rule."

Also, Orin, you write that, when you clerked, "the Justices themselves were fully aware of the Marks implications of their fragmented opinions." But I'm not sure how that is possible, given the considerable disagreement among the justices and circuit courts about how Marks is to be applied. And I don't think there was ever a time period when all important Marks applications were easily predictable. Take the era of Bakke / Grutter, given the splits and divisions that Marks generated in that context. Could you say more about the cases that may have given you a different view?

Posted by: Richard | Jun 5, 2018 6:23:12 PM

A third option would be for the Justices to be explicit in their fragmented opinions about which of the opinions they think are binding under Marks. Maybe my experience is quirky, but when I was a clerk, the Justices themselves were fully aware of the Marks implications of their fragmented opinions. But somewhat surprisingly, the opinions wouldn't actually say what the Justices themselves thought. Instead of making the Marks implications explicit, it would be left for the reader to figure out what the Justices were thinking. It seems to me that one way out of the Marks puzzle is just for the Justices to say in their opinions what they think the Marks implications are of their opinions. For example, a plurality opinion could say, "Because our rule is broader than Justice X's concurrence, we acknowledge that Justice X's concurrence is now binding on lower courts under Marks." Or in a case where there is no binding opinion under Marks, they could be clear about that, too. I suspect that would go a long way toward clarifying the law if each fragmented decision expressly addressed its Marks implications.

Posted by: Orin Kerr | Jun 5, 2018 2:15:09 PM

In my enduring quest to persuade the world that confusing opinions, not Marks, are what make fragmented decisions confusing, I just want to point out the irony that Kennedy's opinion in Hughes inadvertently clarifies that the nine circuits that thought Freeman contained a binding holding were wrong, at least under their own reasoning. While the SG argued that Sotomayor's opinion was binding in this case because both she and the four dissenters would have denied relief, the nine circuits that found Sotomayor's opinion binding generally found it so on the theory that Kennedy's plurality would always provide for relief, that Sotomayor's opinion would sometimes provide for relief, and that Sotomayor's opinion was therefore a completely lesser included subset of Kennedy's. (Judge Pryor's opinion below in Hughes is a particularly lucid exposition.) The D.C. and Ninth Circuits disagreed, but not particularly because they had a different theory of Marks, as you suggest, but rather because they understood Kennedy's ambiguous opinion to deny relief in a few cases where Sotomayor's opinion would grant it. It turns out they were right; now that Kennedy was writing with the weight of judgment, he expanded on the cryptic remarks in his opinion that led the Ninth and D.C. Circuits to doubt that his opinion provided an exception-free rule, and described the few instances in which he would deny relief, which are indeed instances where Sotomayor's rule would have allowed for relief had the underlying plea agreement that the district court approved referenced the Guidelines. In sum, the circuit split over Freeman was purely a function of a confusing Kennedy opinion, which isn't, of course, a kind of circuit split that is unique to the context of fragmented decisions.

Now, to be sure, a circuit split theoretically could have developed on Freeman's precedential content even if Kennedy had written a clear opinion in the first place, as there are theories of Marks (including mine, and the SG's/Judge Kavanaugh's) that would make Sotomayor's opinion binding as to at least some sets of facts, even though it turns out not to be a completely lesser included subset of Kennedy's opinion. But I don't know that there are actually many circuits that subscribe to those theories.

Posted by: Asher Steinberg | Jun 5, 2018 2:13:24 PM

"Another corrective option would be for the justices to work harder to avoid issuing fragmented decisions with unclear precedential implications." Yup. What was most interesting to me was Justice Sotomayor's "mea culpa"--she virtually came out and said (see below) that from now on she'll work harder to join even opinions with which she does not agree in order to avoid lower-court confusion. (Easier to say now that she's the sixth, rather than the fifth, vote, but still.)


The integrity and legitimacy of our criminal justice system depends upon consistency, predictability, and
evenhandedness. Regrettably, the divided decisions in Freeman, and my concurrence in particular, have done
little to foster those foundational principles. Quite the opposite, my individual views, which “[n]o other Justice
. . . shares,” have contributed to ongoing discord among the lower courts, sown confusion among litigants, and
left “the governing rule uncertain.” Arizona v. Gant, 556 U. S. 332, 354 (2009) (Scalia, J., concurring); see
Brief for National Association of Criminal Defense Lawyers et al. as Amici Curiae 3–27 (arguing that the Freeman
concurrence leads to unpredictable and inconsistent results).

I therefore join the majority in full because doing so helps to ensure clarity and stability in the law and promotes
“uniformity in sentencing imposed by different federal courts for similar criminal conduct.”

Posted by: Marty Lederman | Jun 5, 2018 1:04:44 PM

The comments to this entry are closed.