“In our adversarial system of adjudication, we follow the principle of party presentation.” The principle dictates that courts decide only the questions presented by the parties before them and pass upon only those arguments the parties make. Last Term, in Perttu v. Richards, the Supreme Court held 5–4 that in cases covered by the Prison Litigation Reform Act (PLRA), a party has a right to a jury trial on administrative exhaustion when the question of exhaustion is intertwined with the merits of their claim. Hanging over the Court’s split on the PLRA’s meaning was a secondary dispute about the rule of party presentation. Four dissenting Justices accused the majority of violating the rule by resolving the case on grounds the parties had never raised. But party presentation in the Supreme Court has long been a fairly pliant principle. And as party presentation violations go, Richards is rather trivial — making the decision a minor deviation from a concededly flexible rule.
Perttu v. Richards began when Kyle Brandon Richards and two other incarcerated plaintiffs sued prison manager Thomas Perttu over a year’s worth of “prolific and repetitive sexual abuse.” Their handwritten complaint, filed without counsel, alleged a litany of constitutional violations. According to Richards, Perttu approached him one morning and said, “Do what the fuck I tell you to do. Fuck who I tell you to fuck.” When Richards refused, Perttu threatened to kill him. On a separate occasion, Perttu warned that he could keep Richards in prison “forever,” so “[j]ust do what I ask you to.” Another time, Perttu demanded: “Come on bitch boy, take another piss for me. Piss for me boy, and Ill [sic] let you go home.”
Richards sought relief through the prison complaint system, submitting at least a dozen grievances under the Prison Rape Elimination Act. While some were processed, any grievances describing sexual abuse “never got processed and were destroyed.” Perttu reportedly said he was “not going to let [Richards] file any sexual assault grievances.” On several occasions, Perttu tore up the grievance forms and promised to kill the plaintiffs if they submitted any more.
So Richards and his co-plaintiffs brought a Section 1983 suit in federal court. In addition to describing Perttu’s interference with the prison grievance procedure, they claimed that the facility had retaliated against them by placing them in solitary confinement, thereby violating the First Amendment. Perttu moved for summary judgment on the basis “that the plaintiffs had failed to exhaust available grievance procedures as required by the PLRA.” The court sided with Perttu and dismissed the case for failure to exhaust. Richards, still acting pro se and now the sole plaintiff, appealed.
The Sixth Circuit appointed counsel and reversed. The panel recognized that the PLRA required incarcerated plaintiffs to exhaust available administrative remedies before filing suit. But remedies had to be available. If “prison administrators [had] thwart[ed] inmates from taking advantage of a grievance process through machination, misrepresentation, or intimidation,” the exhaustion requirement didn’t apply. Whether exhaustion precluded Richards’s suit was therefore bound up with the merits of his First Amendment retaliation claim, since both turned on the same factual question: Had Perttu blocked Richards from filing his grievances? The judge had erred by deciding that question himself. “[T]he Seventh Amendment requires a jury trial when the resolution of the exhaustion issue under the PLRA would also resolve a genuine dispute of material fact regarding the merits of the plaintiff’s substantive case.”
Five Justices voted to affirm, but not because the Seventh Amendment demanded it. Writing for the majority, Chief Justice Roberts ducked the constitutional question and opted for “a construction of the statute . . . by which the [constitutional] question may be avoided.” As a statutory matter, the Court held that “parties have a right to a jury trial on PLRA exhaustion when that issue is intertwined with the merits of a claim that falls under the Seventh Amendment.” A 2007 decision by the Chief Justice, Jones v. Bock, had taken up whether PLRA exhaustion was an affirmative defense or a mandatory element of a plaintiff’s complaint. Jones had landed on the former because the PLRA’s “silen[ce] on the issue” represented “strong evidence that the usual practice should be followed,” and exhaustion was usually an affirmative defense. That logic applied equally here. Because the PLRA was “‘silent on the issue’ whether judges or juries should resolve factual disputes related to exhaustion,” it incorporated the settled practice that when factual disputes are intertwined with the merits of a claim — thus triggering the Seventh Amendment — the whole bundle goes to the jury, including issues (like exhaustion) a judge would ordinarily decide on her own.
Justice Barrett dissented, joined by Justices Thomas, Alito, and Kavanaugh. Justice Barrett first recognized the difficulty of identifying a founding-era analogue to the defense of exhaustion given that the defense arose “long after the founding.” Happily, Richards did not require answering that particular question, since the case concerned the unique circumstance of a factual dispute about exhaustion intertwined with the merits of the claim. Unhappily, in avoiding the Seventh Amendment question, the majority relied upon a “statutory theory” “not present[ed]” by either party in the Supreme Court or any court below. Resolving the case on that basis violated “party-presentation principles,” under which the Court generally “refuse[d] to consider arguments that the parties failed to make before” it and that “the court of appeals did not analyze first.” To make matters worse, the statutory argument was incorrect. Whereas the majority guaranteed Richards a jury trial on account of the PLRA’s silence, Justice Barrett “understood silence to mean . . . that Congress did not affirmatively confer such a right.”
The disagreement between the Justices represents more than a difference in statutory interpretation. By accusing the majority of reaching out to decide a case on grounds that no party advocated, the dissent highlighted a vexed question of judicial practice: party presentation principles in the United States Supreme Court. Time and again, the Court has emphasized that litigants, through their counsel, “know what is best for them, and are responsible for advancing the facts and argument entitling them to relief.” As a result, the Court “rel[ies] on the parties to frame the issues for decision” while acting as a “neutral arbiter of matters the parties present.” Fair enough. But in practice, the Supreme Court has not exactly followed its own instruction — suggesting that party presentation is more of a guiding principle than an absolute rule.
Before Richards, the Court’s most extended treatment of the party presentation principle occurred in the 2020 decision United States v. Sineneng-Smith. There, the Court reversed a Ninth Circuit judgment on account of the circuit court’s “takeover of the appeal.” After briefing and oral argument, the Ninth Circuit had ordered further briefing by court-appointed amici on three additional questions the court wanted answers to. It then reversed the district court’s judgment on the basis of amici’s responses — not arguments made by the defendant herself. The Court found this insupportable. While “a court is not hidebound by the precise arguments of counsel, [] the Ninth Circuit’s radical transformation of this case [went] well beyond the pale.”
All that makes Sineneng-Smith seem like a straightforward confirmation of the fact that party presentation limits courts’ ability to look beyond the parties’ arguments. Yet Sineneng-Smith itself was a bit of a dodge. The Court had granted certiorari on the question of whether a federal statute criminalizing the encouragement of unlawful immigration for profit violated the First Amendment. Its resolution of that question was “hotly anticipated.” But the Court never reached the First Amendment issue. Instead, it disposed of the case in a relatively sparse opinion by resting entirely on the party presentation principle.
Do as I say, not as I do. As Judge Eric D. Miller observed, by deciding the case this way, the Court itself bent the rule. In its opening brief urging reversal, all the United States had said on the issue of party presentation boiled down to a short paragraph at the tail end of the document, noting that “[i]n reaching out to address this issue, the Ninth Circuit . . . deviated from . . . the normal course of party-driven litigation.” The government “did not elaborate on that objection, let alone advance it as a basis for reversal.” Both parties devoted their briefing to the First Amendment issue. While chastising the court of appeals for breaking from the principle of party presentation, the Court took a detour of its own.
Small wonder Sineneng-Smith described “[t]he party presentation principle [as] supple, not ironclad.” And in a footnote, the Court self-consciously appended an addendum of cases from 2015–20 where it had “called for supplemental briefing or appointed amicus curiae,” including “to present argument in support of the judgment below when a prevailing party has declined to defend the lower court’s decision or an aspect of it.” None of the cases, the Court insisted, bore “any resemblance to the redirection ordered by the Ninth Circuit panel in” Sineneng-Smith. But why not? “While the Court seems completely confident of its authority to appoint such amici, it provides no explanation for the practice.” Maybe the difference lies between appointing amici to defend the lower court’s decision versus arguments never offered by the parties themselves. Perhaps the former approach takes the issues as they arrive for review, while the latter veers too far into naked judicial activism. Less charitably, “perhaps the Court doth protest too much; the length of the list is enough to make one suspect that the Court has been less than scrupulous in its adherence to the principles it articulated.”
Turn the clock back a little further, and the list will only grow. Professor Amanda Frost has noted that “some of the Supreme Court’s landmark cases were decided on grounds that were never raised by the parties.” This includes Erie Railroad Co. v. Tompkins, where “neither the petitioner nor the respondent took issue with Swift v. Tyson’s holding that federal courts sitting in diversity jurisdiction could create federal common law”; Washington v. Davis, where the defendants agreed that plaintiffs could prove an equal protection violation without pleading discriminatory animus — only for the Court to disregard the concession and hold that plaintiffs could not; and Dickerson v. United States, where “both the Fourth Circuit and the Supreme Court questioned whether a federal statute . . . [could] displac[e] the Court’s ruling in Miranda v. Arizona, even though neither party relied on that statute.”
Recall the possibility that appointing amici to support an undefended judgment might be permissible, but not to brief arguments the parties never raised below. Dickerson complicates this distinction. The federal statute in question, 18 U.S.C. § 3501, sought to overturn Miranda’s reticulated warnings framework by allowing confessions to be admitted whenever voluntarily given. “Because no party to the underlying litigation argued in favor of § 3501’s constitutionality in this Court,” the Court explained in a footnote, it had “invited Professor Paul Cassell to assist [its] deliberations by arguing in support of the judgment below.” Yet the statute had gone undefended even in the court of appeals. The Fourth Circuit explicitly recognized that “the applicability of § 3501 was not briefed by the Government,” and that the Department of Justice had “actually prohibit[ed] the U.S. Attorney’s Office from briefing the issue.” It was the Washington Legal Foundation, a conservative public interest group, which had presented the § 3501 argument in an amicus brief. A panel majority saw fit to press on with the § 3501 analysis, over a dissent which strenuously disagreed. In the Supreme Court, Chief Justice Rehnquist’s majority opinion made no mention of the problem of potentially violating party presentation. All the airtime party presentation got was the single-sentence footnote.
By playing fast and loose with the party presentation principle, Dickerson resembles another Fourth Amendment decision rendered almost forty years earlier — one that hadn’t been litigated as a Fourth Amendment case at all. In Mapp v. Ohio, a woman convicted for possessing obscene materials challenged her conviction under the First Amendment. A majority of Justices voted to reverse her conviction, and in so doing issued a watershed decision incorporating the exclusionary rule against the states. No party had urged that outcome. And the Court had to overturn one of its prior decisions, Wolf v. Colorado, to get there. As the Court recognized in a footnote (by now a familiar move), Mapp herself “did not insist that Wolf be overruled”; indeed, Mapp’s counsel “expressly disavowed” that request in oral argument. But a brief by “amicus curiae [the ACLU] . . . did urge the Court to overrule Wolf” — in a single three-sentence paragraph. That was sufficient.1 Dissenting, Justice Harlan chastised the majority for “simply ‘reach[ing] out’ to overrule Wolf” when Mapp’s brief had “not even cite[d] the case.” More than six decades later, Chief Justice Roberts and Justice Barrett’s dueling opinions reprised in Richards this very old debate.
Given Dickerson, Mapp, and Sineneng-Smith, the majority’s move in Richards is not so radical. Yet even on its own terms, the Court’s statutory analysis was not a major departure from the arguments the parties had made. Chief Justice Roberts acknowledged Justice Barrett’s critique but contended that “the statutory question ha[d] been fully briefed by amici and involves the same precedents relied on by the parties.” The Court thus appeared to concede that the dispositive statutory argument had been made by amici (a group of law professors), not the parties themselves. It didn’t need to. Richards’s merits brief had mentioned the “usual practice” no fewer than seven times. For instance, he had argued that under Beacon Theaters, Inc. v. Westover, “juries decide all factual questions necessary to resolve the merits of a legal claim, even when some of those questions may overlap with issues a judge could otherwise decide,” “and there [was] no indication that Congress intended to depart from” this “usual practice” — meaning that the PLRA incorporated it.
It is true that Richards did not explicitly urge the Court to avoid the constitutional holding and rely on a statutory rationale. But in this context, the distance between a statutory and constitutional ruling is smaller than it first appears. The “usual practice” at the time of the PLRA’s enactment had the Seventh Amendment baked in: Courts tried legal claims before equitable ones, or declined to resolve jurisdictional issues that would effectively decide the merits, because that was what the Seventh Amendment demanded. In the end, the distinction that makes a difference might boil down to the Court’s decision to “express no view [] on whether Congress could have required otherwise in the PLRA without violating a party’s Seventh Amendment right.” Said another way, because Richards is not a constitutional decision, Congress may overrule it.
* * *
Why does the Court stray from the supposedly fundamental principle of party presentation more often than the Justices seem willing to admit?
At a high level, the debate over party presentation echoes the longstanding competition between two models of judicial review: (1) law declaration — which would favor a loose party presentation principle — and (2) dispute resolution — which would command a strict rule. Advocates for the former model warn that parties cannot be allowed to “completely control the judiciary’s statements of law . . . lest they undermine the federal courts’ role to independently ascertain the meaning of legal texts for the benefit of all.” Critics retort that since courts exist for the very purpose of dispute resolution, the parties alone should decide “both the legal and factual claims relevant to their dispute.” Yet in the end, most agree that the American judicial system contains features of both approaches; its “choice among these models is ‘all of the above.’”
Consistent with this blended approach, the Court’s departures from party presentation have varied in degree. Resolving a case on grounds not wildly dissimilar from the parties’ positions (as in Richards) tilts only weakly toward a law declaration model. By contrast, basing an opinion on an argument expressly disavowed by petitioner’s counsel (as in Mapp) reflects a more muscular exercise of judicial agenda control. The practice of appointing amici to argue an undefended judgment sits between the two poles. Doing so echoes law declaration in that the Court refuses to let one side’s concession constrain its ability to say what the law is. But appointing amici might also be understood as guardrail on freewheeling law declaration by forcing “adversary presentation as an aid to the accurate determination of factual and legal issues.” Finally, the Court’s habit of formulating entirely new questions presented indicates a certain fondness for law declaration. But at least the parties will themselves brief the additional question; at least they have notice of a potential basis for the Court’s eventual disposition.
Why should the Court be so free to deviate from party presentation when, as Sineneng-Smith indicates, lower federal courts are not? Nearly thirty years ago, Judge Laurence Silberman posited that the Justices did not wish “to restrict their own ability to reach out to issues not presented in cases brought to the Court, nor . . . justify that practice by openly acknowledging the Supreme Court as not subject to normal judicial constraints.” He did not think this was a good thing. But as a structural matter, the Supreme Court may face unique pressures to get the law “right” regardless of what the parties do. Unlike lower court decisions, which are precedential only within their respective jurisdictions, the Supreme Court’s holdings bind the nation, making it especially important for the Court to reach the right result. Moreover, a circuit court panel can choose to issue an unpublished opinion tying only the parties to the consequences of their litigation choices. The Supreme Court cannot.
The Court’s turn towards law declaration (and attendant departures from party presentation) may also reflect its increasingly outsized role in resolving national disputes. Perttu is a win for a relatively slim class of incarcerated litigants, but party presentation has taken a beating in more high-profile cases too. For instance, the Court recently ordered supplemental briefing on an issue raised by a law professor as amici in litigation challenging the National Guard’s deployment in Illinois — an issue to which the United States had “devote[d] only a cursory footnote,” and which the respondents had not briefed at all.
The upshot is that the Court has not, and likely never will, box itself in by enshrining a strong form of party presentation. Viewed in perspective, Richards is hardly a departure from a principle that has never been particularly strict. By resolving the case on grounds advanced chiefly (though not exclusively) by amici, the Richards Court followed the “usual practice” in more ways than one.