settled versus right: A Theory of precedent. By Randy J. Kozel. Cambridge University Press. 2017. PP. x + 180. $99.99 (hardcover), $34.99 (paper).
Not all precedent is equal. Some precedents are given more weight, are harder to overturn, than are others. Super-precedents, even super-duper-precedents, have special status. Legal academics consider them canonized; judges cite them with reverence; and judicial nominees, mum on whether they agree or disagree with the outcomes in lesser cases, announce their commitment to our super-precedents. If, compared to regular old precedents, some precedents count for more, might other precedents count for less? This essay makes the case for subprecedents: precedents that have only weak value, such that compared to other precedents, they are more easily ignored and easier to overturn. In his book, Settled Versus Right: A Theory of Precedent, Randy Kozel depicts a world in which judges (and particularly Supreme Court Justices) jockey to render their own rulings impervious to change and seek to elevate their own supporting rationales, hypotheticals, commentary, and even stray statements to the level of binding law, and in which the winners in high-stakes disputes over interpretive methodology and normative considerations get to shape and solidify legal rules for the nation. This essay, prepared for a symposium on Kozel’s book, tells a different story: of efforts to minimize the significance of judicial decisions, to limit their reach and power, and to curtail the work they may do in future cases. Kozel’s account is one of aggressive judicial ambition; mine, of deliberate timidity. Yet a full understanding of precedent requires attention to cases that start out or end up having limited precedential value. Subprecedents, this essay shows, serve some important, and at times surprising, functions in a system of stare decisis. Consideration of subprecedents thus sheds light on the questions that motivate Kozel’s own study and on his noteworthy efforts to ground stare decisis in a “commitment to the abiding continuity of constitutional law” (p. 135) by shifting the operations of precedent away from the predilections of individual judges.
The essay proceeds in seven parts. Each part identifies, in field-guide fashion, a specific kind of subprecedent (illustrated with a case or two), describes its characteristics, and discusses its roles and significance. The focus throughout is primarily (but not exclusively) on constitutional cases, though it bears mentioning that a similar category of subprecedents could be developed for statutory decisions; likewise, while the focus here is on rulings by the Supreme Court of the United States, the approach could be extended to the lower federal courts and to the state courts, all of which decide far more cases than does the U.S. Supreme Court. A short conclusion draws some general lessons and discusses their implications.
- Golden Tickets
Every now and then, in the style of a papal dispensation or a royal pardon to a prisoner whose neck is noosed, the U.S. Supreme Court sets aside the normal rules to help a criminal defendant screwed by the system. Typically, in such cases, the defendant has suffered, through no fault of his own, some terrible legal fate. His plight cries out for a remedy but the law is against him. The Supreme Court is sympathetic but it is also disinclined to announce a new legal rule that will have broad effect. The Court’s solution is to issue the defendant what is in essence a golden ticket: a decision tailored to the precise circumstances of the case, designed to benefit the defendant but nobody else. Indeed, in such cases, the Court may specifically warn against anybody else applying the ruling in other circumstances to benefit different defendants (for only popes and kings get to exercise this sort of power). The resulting Supreme Court ruling, in favor of the single defendant, represents a subprecedent.
Maples v. Thomas, decided in 2012, is a prime example. In that case, petitioner Cory Maples was convicted of murder and sentenced to death in 1997 in Alabama. Maples sought post-conviction relief in state court, alleging ineffective assistance of counsel and other trial and penalty-phase errors. His petition was prepared on a pro bono basis by two associates of the New York law firm, Sullivan & Cromwell. While the petition was pending in Alabama, the two associates left the firm for other positions that prevented them from continuing to represent Maples, but without advising him they were no longer his lawyers, filing a motion to withdraw, arranging for new counsel, or even notifying the Alabama courts of the change in their status. When the Alabama trial court denied Maples’ petition, it sent notices of the decision to the associates at Sullivan & Cromwell, but because they had left the firm, the mailroom there returned the notices unopened to the court. The time ran out for Maples to appeal the trial court denial without Maples even knowing he no longer actually had a lawyer. Subsequently, after learning from the Alabama prosecutor’s office that he had lost in trial court and that the time to appeal had expired, Maples, represented by new counsel, filed a federal habeas petition. The district court dismissed the petition on the ground that Maples had procedurally defaulted—by failing timely to appeal the Alabama trial court’s order—and there was no cause to excuse the default. The 11th Circuit affirmed.
Ordinarily, the circuit court ruling would be the end of the road for Cory Maples because even if it wrongly applied the law, that is not normally a basis for the U.S. Supreme Court to hear a case. Yet the Court, in an opinion by Justice Ginsburg, reversed. Her opinion favored Maples—but was deliberately and precisely tailored to him alone. “The sole question this Court has taken up for review,” Ginsburg wrote, “is whether, on the extraordinary facts of Maples’ case, there is ‘cause’ to excuse the default.” In holding that such cause did exist, the ruling tracked the exact circumstances of Maples’ case in a way that precluded (virtually) any other defendant from benefiting from the decision. Here is the meat of the ruling:
In the unusual circumstances of this case, principles of agency law and fundamental fairness point to the same conclusion: There was indeed cause to excuse Maples’ procedural default. Through no fault of his own, Maples lacked the assistance of any authorized attorney during the 42 days Alabama allows for noticing an appeal from a trial court’s denial of postconviction relief. . . . [H]e had no reason to suspect that, in reality, he had been reduced to pro se status. Maples was disarmed by extraordinary circumstances quite beyond his control. He has shown ample cause, we hold, to excuse the procedural default into which he was trapped when counsel of record abandoned him without a word of warning.
Notice all of the elements (“the extraordinary facts,” the “unusual circumstances”) that in the Court’s view aligned to generate an outcome in Maples’ favor: the lack of fault on his part that meant he had no lawyer representing him during the entire critical period in which an appeal could be filed; the lack of notice to Maples so that he had “no reason” even to “suspect” he lacked representation; the “extraordinary circumstances” beyond Maples’ own control that “disarmed” him from acting; and his being “trapped” because he was “abandoned,” without a single “word of warning.” Lest there be any mistake that this doctrine was baked for Maples alone, Justice Alito, in a short concurring opinion, underscored the one-off nature of the decision. After identifying “eight unfortunate events” that “combined” to leave Maples without a lawyer, Alito concluded: “What occurred here was not a predictable consequence of the Alabama system but a veritable perfect storm of misfortune, a most unlikely combination of events that, without notice, effectively deprived petitioner of legal representation. Under these unique circumstances . . . petitioner’s procedural default is overcome.” In sum, the Court produced a ruling for Cory Maples and him alone. While scores of other criminal defendants not named Cory Maples have cited the case, courts have had no trouble limiting it to the single beneficiary of the Supreme Court’s decision.
Four observations about golden tickets bear mention. First, the Supreme Court can only intervene to provide relief to an individual defendant because it has the ability to limit the scope of a ruling all the way down to a single case. Without that option as an important element of the power to craft and control precedent, there would be no golden tickets. Without the ability to curtail precedential effect, the Justices would not grant relief to a Cory Maples, because doing so would open the door to other criminal defendants making similar claims upon the Court; and, worse—from the perspective of a Court that disfavors federal judges interfering with state criminal convictions—to lower federal courts using the ruling to grant relief to other habeas petitioners at a clip that would outpace the Court’s limited capacity for review. The ability to limit a ruling at the outset is essential if the Court is to intervene to help the single petitioner like Cory Maples.
Second, however, golden tickets, while beneficial to the recipient, are of uncertain systemic value. There is no particular reason Cory Maples should be helped while thousands of other convicted criminal defendants—whose own cases could be spun into a story of misfortune and injustice—are denied the same remedy. The power to create a subprecedent allows the Court to perform a one-time act of mercy. But the price of issuing occasional golden tickets may be high. Random acts of kindness serve to highlight the absence of systemic fairness—the foundational principle of the rule of law. Golden tickets also highlight the Court’s skepticism of lower federal courts intervening in state criminal cases. The Supreme Court itself doesn’t have the capacity to decide whether every individual habeas petitioner deserves help but it also doesn’t quite trust the lower courts to make those decisions with adequate respect for state processes. The end result, an occasional intervention by the Justices, looks more regal than judicial.
A third observation is that just as Powerball winners discover that money doesn’t buy happiness, criminal defendants issued golden tickets may nonetheless find freedom beyond reach. For a win at the Supreme Court does not necessarily unlock the prison door. In 2015, on remand in Cory Maples’ case, the federal district court held that while, in accordance with the Supreme Court’s ruling, cause had been established, Maples had not shown prejudice from the procedural default. The district court therefore dismissed Maples’ habeas petition. In the spring of 2018, the circuit court vacated that ruling, on the ground that in assessing prejudice, the district court had applied the wrong legal standard, and it remanded the case for a new hearing and ruling. Cory Maples, who for now remains convicted and incarcerated on death row, might ultimately prevail, but his doing so will require more than what the Supreme Court gave him in 2012.
A fourth observation is that while the Court tries very hard to limit golden tickets to an individual case, sometimes they turn out to have more general application. One example here is the Court’s 1932 decision in Powell v. Alabama, which held that, under the factual circumstances of the capital case presented, denying the defendants access to counsel worked a violation of the Due Process Clause of the Fourteenth Amendment. Four decades later, the Court invoked Powell when it held, in Gideon v. Wainwright, that the Sixth Amendment conferred a more general obligation upon the states to provide criminal defendants with an attorney. Yet broad applications of golden tickets do not represent a failure. The Powell Court did not intend a general rule, but Gideon does not reveal a failed effort to limit the reach of a decision. By any measure, limiting a criminal procedural right for four decades (during which time states conducted millions of prosecutions) is a significant achievement.
- You Never Call, You Never Write
A second category of subprecedents also involves rulings that the Supreme Court confines to an individual case. In contrast to the golden tickets issued to criminal defendants and tailored closely to the facts of their cases, in this second category, the Court decides a legal question with potentially enormous significance. Yet the Court never (or very rarely) cites or otherwise relies upon the ruling in later cases, even those that present very similar legal issues. As with golden tickets, the Court may signal in the ruling itself that it should be understood as confined to the particular case. But the more important feature is the Court’s own refusal to rely upon the ruling again.
An obvious example here is Bush v. Gore. In that case, in a per curiam decision, the Supreme Court held that the state-wide manual recount ordered by the Florida Supreme Court during the 2000 presidential election violated the Equal Protection Clause of the Fourteenth Amendment because of a lack of uniform recount standards. Seven justices supported that holding but they did not agree on the remedy. Five held that there was no possibility of fashioning a constitutional recount within the time remaining for Florida to take advantage of the safe harbor provision of 3 U.S.C. section 5—effectively ending Al Gore’s challenge to Florida’s certification of George W. Bush as the winner of the electoral votes in the state. Writing separately, however, Justices Breyer and Souter, while agreeing that there was an Equal Protection violation, took the position that a remedy could be implemented that satisfied the Constitution’s requirements and the case should be remanded for that purpose. Chief Justice Rehnquist, joined by Justices Scalia and Thomas, issued a concurring opinion contending that the recount was also unconstitutional because the Florida Supreme Court had made new election law in violation of Article II Section 1. Justices Stevens and Ginsburg dissented in full.
In its per curiam decision, the Bush Court invoked general principles of equality in voting and cited landmark election cases, but it tied its holding to the 2000 election dispute. “Our consideration is limited,” the Court explained, “to the present circumstances, for the problem of equal protection in election processes generally presents many complexities.” In other words, the ruling might not be one on which anyone in a different case in the future can rely. That is indeed how the Court later understood the case. Bush v. Gore has been the subject of many books and law review articles. The decision has been invoked in hundreds of election (and other) cases in the lower federal courts and the state courts. It has also been cited and discussed in many petitions and briefs to the U.S. Supreme Court. However, since 2000, the decision has been cited just one time at the Supreme Court itself: by Justice Thomas, in a footnote in a dissent not joined by any other member of the Court and for the proposition that Article II permits the state legislatures to select presidential electors itself if it so chooses, not for the equal protection holding of the case. The Court, of course, has since decided many cases involving issues of equal protection and voting. Yet it has repeatedly done so without any mention of the equal protection ruling in Bush v. Gore. The Justices appear to have agreed that Bush v. Gore will not be spoken of again. Accordingly, the case has become limited, at least at the high court, to disputes where the parties are named Bush and Gore.
Golden tickets start out as limited. Cases in this second category depend upon an ongoing commitment—a pact—on the part of the Justices to keep a ruling confined to the case in which it was made. Maintaining this consensus over an extended period—nearly two decades now for Bush v. Gore—is not an easy accomplishment (though perhaps as the years tick by, acting against the consensus becomes more costly for an individual Justice). One wonders, for example, how many times individual Justices have been tempted to cite the case but then changed their minds; how often a law clerk (likely in diapers during the 2000 election) has innocently slipped in a reference to the case in a draft opinion, only to have it excised by the Justice. Where, though, the pact can be created and maintained, it provides the Court with a powerful tool to intervene in a single case without having to confront the broader implications of the ruling. In essence, while Kozel emphasizes competition among the Justices to specify the scope of precedent, in this category of subprecedent, we have, instead, convergence among the Court’s members on not giving a case precedential authority. That, too, is part of our system of stare decisis.
III. Ticking Time Bombs
A third category of subprecedents involves decisions that the Supreme Court designates as limited in time. One example is Grutter v. Bollinger, decided in 2003. In that case, the Court held that the Equal Protection Clause did not prohibit the University of Michigan Law School from considering race, as part of an individualized review of applicants taking into account all factors that contribute to diversity, in order to obtain the educational benefits of a diverse student body. However, the ruling contained a built-in expiration date. Writing that because “a core purpose of the Fourteenth Amendment was to do away with all governmentally imposed discrimination based on race,” Justice O’Connor for the Court concluded that “race-conscious admissions policies must be limited in time.” Because racial classifications are “potentially . . . dangerous,” she explained, “they may be employed no more broadly than the interest demands” and “[e]nshrining a permanent justification for racial preferences would offend this fundamental equal protection principle.” Therefore, the Court held, race-conscious admissions programs “must have a logical end point.” This meant two things. First, specific programs must be limited: the “durational requirement” could be met with “sunset provisions . . . and periodic reviews to determine whether racial preferences are still necessary to achieve student body diversity.” Second, there was a more general outer limit to race-based affirmative action programs: twenty-five years. Why that period? “It has been 25 years,” O’Connor explained in 2003, “since Justice Powell [in Regents of the University of California v. Bakke] first approved the use of race to further an interest in student body diversity in the context of public higher education.” Because, in that intervening period, “the number of minority applicants with high grades and test scores has . . . increased,” an end to lawful consideration of race was on the horizon: “We expect that 25 years from now, the use of racial preferences will no longer be necessary to further the interest approved today.”
Temporal limits can serve significant purposes. They can be an effective mechanism for bringing on board members of the Court who might not approve of the bottom line but nonetheless can live with a decision that, by design, is not forever. Indeed, a member of the Court might be inclined to join a decision just in order to ensure it contains an expiration date. Temporal limits can also promote agreement by persuading members of the Court who, again, do not support the bottom line, to sign on to a future case making use of the expiring precedent. Consider, in this regard, the Court’s most recent affirmative action case, Fisher v. University of Texas. In that case, the Court upheld a supplemental undergraduate admissions program at the University of Texas that took account of the race of applicants. Justice Kennedy, who had dissented in Grutter, wrote the majority opinion in Fisher to uphold the challenged program and in so doing he reiterated the durational limits Grutter had announced.
A second feature of this form of subprecedent is that when the expiration date arrives the question of overruling the decision does not (or arguably does not) even arise. Instead, the case can just die its natural death. And even if a future court wants to keep the patient alive, when it confronts the question whether to overrule, one key factor, “whether the rule is subject to a kind of reliance that would lend a special hardship to the consequences of overruling and add inequity to the cost of repudiation,” will support overruling the past decision. After all, nobody could have reasonably relied upon a decision that, by its own stated terms, is set to expire. Indeed, any reliance interests would be based upon the expectation that the decision is only good for a stated period. Nobody, in other words, should arrange their affairs in expectation that in 2028 race-based affirmative action programs in university admissions will be constitutional: the sensible approach would be to plan for their end. In short, subprecedents with expiration dates allow members of the Court to rig at the outset the stare decisis analysis that will occur in the future, perhaps long after those members themselves have left the Court.
- Slip Slidin’ Away
The above three categories of subprecedent all involve cases limited by the Court at the time they are decided. A fourth category of subprecedents consists of decisions that start out as having ordinary precedential effect but later end up limited. In this category are cases that, while not overturned, are undercut or reworked by subsequent decisions by a Court that aims deliberately to weaken the significance of the earlier ruling, in other words to turn it into a subprecedent.
One example here is Katzenbach v. Morgan. That case involved the constitutionality of Section 4(e) of the 1965 Voting Rights Act, which prohibited the use of literacy tests to deny the franchise to non-English speakers who had completed a sixth-grade education in Puerto Rico. Congress had enacted Section 4(e) using its Fourteenth Amendment Section 5 enforcement powers, but the VRA provision seemed inconsistent with the Court’s own 1959 decision in Lassiter v. Northampton Board of Elections, that literacy requirements were constitutionally valid. The Morgan case asked whether, under those circumstances, the federal statutory provision validly trumped a provision of New York’s own election laws requiring, as a condition of voting, ability to read and write English.
In upholding congressional power to enact Section 4(e) (and thus displace the New York requirement), the Supreme Court, in an opinion by Justice Brennan, provided two rationales. First, he reasoned, prohibiting literacy tests could be viewed as ordinary remedial legislation protecting judicially-recognized rights to non-discriminatory treatment by the government in provision of its services. Second, Brennan found, the law was also sustainable as directly targeting unconstitutional state laws. In that regard, the Court allowed Congress leeway to determine the reach of section 1 of the Fourteenth Amendment and hence of its own powers under section 5. In reaching this outcome, the Court specifically rejected the New York Attorney General’s argument that “§ 4(e) cannot be sustained as appropriate legislation to enforce the Equal Protection Clause unless the judiciary decides . . . that the application of the English literacy requirement prohibited by § 4(e) is forbidden by the Equal Protection Clause itself.” Brennan reasoned instead that “[a] construction of § 5 [of the Fourteenth Amendment] that would require a judicial determination that the enforcement of the state law precluded by Congress violated the Amendment, as a condition of sustaining the congressional enactment, would depreciate both congressional resourcefulness and congressional responsibility for implementing the Amendment.” Because Congress was given enforcement powers under section 5 of the Fourteenth Amendment, the Court’s task was merely to decide whether, “without regard to whether the judiciary would find that the Equal Protection Clause itself nullifies New York’s English literacy requirement as so applied . . . Congress [could] prohibit the enforcement of the state law by legislating under § 5 of the Fourteenth Amendment.” The answer to that question, Brennan explained, was “limited to determining whether such legislation is, as required by § 5, appropriate legislation to enforce the Equal Protection Clause.” Under the announced standard, the provision was easily sustained. Congress was entitled to determine that literacy requirements violated the Equal Protection Clause even if the Court itself would not necessarily reach that conclusion. In response to Justice Harlan’s criticism in his dissenting opinion that the majority “reads § 5 of the Fourteenth Amendment as giving Congress the power to define the substantive scope of the Amendment” such that Congress could also “dilute equal protection and due process decisions of this Court,” Brennan included his famous footnote specifying that Congress’s power to define rights ran in only one direction.
Three decades after Morgan, the Court decided City of Boerne v. Flores. The Boerne Court held, in an opinion by Justice Kennedy, that Congress lacked power under Section 5 of the Fourteenth Amendment to impose upon the states the Religious Freedom Restoration Act (RFRA). That statute, enacted in response to the Supreme Court’s 1990 decision in Employment Division v. Smith, prohibited the government from substantially burdening a person’s exercise of religion, even if the burden results from a rule of general applicability, unless the government can demonstrate the burden furthers a compelling governmental interest and is the least restrictive means of furthering that interest. Writing for the Court, Justice Kennedy held that RFRA, as applied to the states, was invalid because Congress’s power is limited to “enforcing” the substantive protections of the Fourteenth Amendment (here, including the incorporated protections of the First). Given the Court’s ruling in Smith rejecting the application of a compelling interest test with respect to generally applicable laws, RFRA exceeded Congress’s authority. “Legislation which alters the meaning of the Free Exercise Clause,” Kennedy explained, “cannot be said to be enforcing the Clause. Congress does not enforce a constitutional right by changing what the right is.” In particular, RFRA interfered with the power of the judiciary: “When the Court has interpreted the Constitution, it has acted within the province of the Judicial Branch, which embraces the duty to say what the law is. . . . Broad as the power of Congress is under the Enforcement Clause of the Fourteenth Amendment, RFRA contradicts vital principles necessary to maintain separation of powers and the federal balance.” What about Morgan’s recognition of congressional power to ratchet up rights beyond a judicial floor? Kennedy wrote that while “[t]here is language in our [Morgan] opinion . . . which could be interpreted as acknowledging a power in Congress to enact legislation that expands the rights contained in § 1 of the Fourteenth Amendment,” such a reading was not “a necessary interpretation” of the case, “or even the best one.” Instead, Kennedy explained, the Morgan Court had “perceived a factual basis” that justified Congress’s intervention so that the decision to uphold section 4(e) of the VRA “rested on unconstitutional discrimination by New York and Congress’ reasonable attempt to combat it.” In other words, according to Kennedy, the Morgan Court agreed that there existed an equal protection violation, so that the condition for use of the section 5 power was met. The Morgan holding was thus recast, not cast aside; turned from precedent into subprecedent.
Many commentators take the position that, in light of Boerne, reliance upon an expansive reading of Morgan—in which Congress can ratchet up rights—is no longer defensible. Understanding Morgan as a subprecedent suggests a different conclusion. Morgan (expansively understood) has not been overruled. It thus still lingers: perhaps not worth much, but nonetheless still alive. Its continued existence as a subprecedent means it might, one day, return to the stage in a starring role. It might, in other words, be turned back into ordinary precedent and guide the resolution of a future case. For one thing, the Supreme Court might itself reinvigorate Morgan. A shift in the composition of the Court or a reassessment of notions of congressional versus judicial power on some particular issue might result in the majority of the Justices doing to Boerne what Boerne did to Morgan: announcing that those of us who imagined that Boerne stood for the proposition that Congress lacks power to define substantive rights were interpreting Justice Kennedy’s opinion in a way neither “necessary” nor “best.” The Court could then well march forward with a generous understanding of congressional power grounded in Morgan and citing extensively Brennan’s opinion without any need to account for a doctrinal change. Likewise, Morgan’s continued existence might lead a creative lower court or Congress itself to rely upon the case in support of expansive congressional power. After all, there exist multiple easy ways to limit Boerne: as addressing an improper legislative effort to overturn a recent Supreme Court decision, for instance, or as addressing Congress’s power in the sole context of religious free exercise (and not, as in Morgan, equal protection or voting). In sum, making earlier precedent a subprecedent—rather than just overturning it—creates opportunities that might be seized by a later Court or by other actors.
Nonetheless, opportunity carries risk. A court that recasts rather than overturns a prior decision may not be able to control future uses of the case—whether by a future Court or by other players. Hence, a telltale signal of cases within this fourth category is that members of the Court majority perceive the risks and object that the prior decision has been recast rather than overturned. Schuette v. Coalition to Defend Affirmative Action provides an example. In rejecting an equal protection challenge to a state constitutional amendment barring (among other things) public universities from considering race in admissions, Schuette recast earlier political process cases including Hunter v. Erickson and Washington v. Seattle School District No. 1. Justice Scalia (joined by Justice Thomas) criticized the Schuette plurality for “reinterpreting” these two earlier cases “beyond recognition” and called instead for them simply to be overruled. That call reflects an understanding that limiting a case, even severely so, does not necessarily spell its end.
- Stairways to Heaven
A fifth category of subprecedents involves cases that, at the time they are issued, are presented as limited but then later turn out to be building blocks of broader rulings. In this manner, subprecedents serve, sometimes by deliberate design, as a staircase or bridge to a momentous decision in the future.
The Court’s recent Voting Rights Act decisions provide a good example. In 2009, in Northwest Austin Municipal Utility District No. One v. Holder, Chief Justice Roberts, writing for a unanimous Court, held that as a statutory matter all political subdivisions were eligible to seek a bailout from the preclearance provisions of Section 5 of the Voting Rights Act. Because the utility district that brought the case was therefore entitled to seek bailout as a statutory matter, the Court did not rule on the district’s additional argument that, in extending Section 5 for an additional 25 years in 2006 based upon the decades-old coverage formula of section 4(b), Congress had exceeded its powers under Section 2 of the Fifteenth Amendment. In the course of his opinion, Chief Justice Roberts highlighted “federalism concerns” the 2006 VRA extension presented. Nonetheless, in light of the statutory ruling, Roberts concluded: “Whether conditions continue to justify such legislation is a difficult constitutional question we do not answer today.” Justice Thomas, concurring in the statutory holding, would have reached the constitutional issue: in his opinion, “the lack of current evidence of intentional discrimination with respect to voting renders §5 unconstitutional.” However, all of the other members of the Court joined Roberts’ opinion in full.
Four years later, Roberts also authored the Court’s opinion in Shelby County v. Holder, ruling that the section 4(b) coverage formula was unconstitutional and thus did not support the Section 5 pre-clearance requirements. This time, the Court’s ruling was 5-4. In the course of his opinion, Roberts repeatedly invoked and quoted from Northwest Austin, which he presented as foretelling the outcome. “[W]e expressed serious doubts about the Act’s continued constitutionality,” Roberts wrote of Northwest Austin; “[e]ight Members of the Court subscribed to these views, and the remaining Member would have held the Act unconstitutional.” Northwest Austin, then, paved the way to invalidating in Shelby County the contested provision of the VRA.
Deploying subprecedents to build bridges (or staircases) has an obvious benefit. The Court can act as though it is not breaking new ground but merely completing an edifice already underway. This explains Roberts’ repeated invocation, in Shelby County, of Northwest Austin and his reminder that that decision was unanimous. In Roberts’ presentation, invalidating the coverage formula involved nothing more than putting the next foot forward. The approach obviously depends upon persuasion that a prior, seemingly weak case actually has strong legs. Therein lies the risk. Other members of the Court may well object to the transformative role of a subprecedent, particularly one they signed onto in its modest form. Thus, in Shelby County, the four dissenters strongly objected to Roberts’ use of the opinion in Northwest Austin that they had joined. Writing for the dissent, Justice Ginsburg took the position that the relevant precedent was not Northwest Austin but South Carolina v. Katzenbach, in which the Court had rejected federalism challenges to the VRA’s coverage formula and pre-clearance provisions. Ginsburg did not remotely share Roberts’ view that Northwest Austin had pre-ordained holding those VRA provisions now invalid. Among other points, Ginsburg objected particularly to Roberts’ invocation of Northwest Austin to support his application, in Shelby County, of a constitutional principle of equal state sovereignty. She wrote:
If the Court is suggesting that dictum in Northwest Austin silently overruled Katzenbach’s limitation of the equal sovereignty doctrine to “the admission of new States,” the suggestion is untenable. Northwest Austin cited Katzenbach’s holding in the course of declining to decide whether the VRA was constitutional or even what standard of review applied to the question. In today’s decision, the Court ratchets up what was pure dictum in Northwest Austin, attributing breadth to the equal sovereignty principle in flat contradiction of Katzenbach.
Of course, these are the words of the dissent. Roberts himself was able to keep a majority to deploy Northwest Austin to invalidate key components of the VRA.
The possibility of the Court creating subprecedents as a step towards or as the basis for a more momentous ruling in the future has some implications for Kozel’s project to forge a commitment to stare decisis in a world of “interpretive pluralism” (p. 99). If the Justices indeed overcome interpretive divides and sign onto a strong form of stare decisis but, at the same time, they can strategically craft limited rulings for later deployment towards larger goals, there is likely to quickly emerge a significant feedback effect upon how cases are decided. Justices who commit to stare decisis, and especially to a doctrine that “no precedent may be jettisoned unless a supermajority of justices votes to overrule it” (p. 157), but fear another Northwest Austin/Shelby County episode will be disinclined to join rulings holding, stating, or suggesting anything beyond precisely what they individually support in full. As a result, getting to five, let alone nine, may well require decisions and opinions limited in the extreme and with language specifying the shared understood scope of a ruling. In other words, many cases (and perhaps all) might end up being resolved as narrow, fact-specific subprecedents that carry warnings against any broader use. For instance, the Court’s opinion in Northwest Austin might have looked very different had Justice Ginsburg predicted its later use. But the problem is broader than four Justices sometimes being hoodwinked. After Shelby County, all of the Justices would be wise to wonder whether something they sign onto or do not object to today will be deployed against them tomorrow. The final outcome, where there exists both stare decisis and suspicion, could be the Court regularly issuing a per curiam decision that simply announces the judgment—accompanied by nine separate opinions in which each Justice sets out his or her own view of the case and its proper disposition.
- Out of Steam
So far the examples of subprecedent have involved cases where the Justices themselves exercise control over and manipulate the reach and limits of their rulings. We turn next to two additional categories of subprecedent in which the Court is not in control. Both categories involve instances where efforts on the part of the Justices to create and manage precedent flounder so that subprecedents result.
One category involves cases in which the Court acts ambitiously to move doctrine in a new direction but where the effort does not succeed. Whereas super precedents serve as “foundations for subsequent lines of judicial decisions,” the subprecedents in this category involve doctrinal innovations that run out of steam. United States v. Lopez, the Court’s 1995 decision invalidating the federal Gun-Free School Zones Act as beyond Congress’s Commerce Clause power, is a good example. When Lopez was decided, commentators spoke of the beginning of a federalism revolution. As the Court issued additional decisions limiting national power, including United States v. Morrison, debate centered on just how far this federalism revolution would go.
However, the statute in Lopez itself was promptly and easily corrected with minimal congressional effort and the revolution, if ever there was, soon seemed to come to a halt. In 2005, in Gonzales v. Raich, the Court upheld the application of the federal Controlled Substances Act to marijuana grown at home for personal use that was never bought or sold and never moved across any state line. Over a strong dissent by three members of the Lopez majority, the Court explained that Congress could reach intrastate non-commercial activities as part of a broader regulation of the interstate economy—thus providing Congress with a green light for renewed broad regulation under the Commerce Clause.
At his confirmation hearing, John G. Roberts described Lopez and Morrison as merely “two decisions in the more than 200-year sweep of decisions in which the Supreme Court has . . . recognized extremely broad authority on Congress’s part, going all the way back to Gibbons v. Ogden and Chief Justice John Marshall, when those Commerce Clause decisions were important in binding the Nation together as a single commercial unit” and he observed that Raich had demonstrated that Lopez and Morrison did not “junk all the cases that came before” them. Responding to Senator Feinstein’s question (about Lopez), “[A]t what point does crime influence commerce?,” Roberts stated: “I think it does. . . . [The Act] didn’t have a requirement that the firearm be transported in interstate commerce . . . . [If] the Act had required that, which I think . . . it’s fairly easy to show in almost every case . . . then that would have been within the Congress’s power under the Commerce Clause.”
The views of somebody hoping to persuade the Senate to confirm him to the Supreme Court may be a poor guide as to how the judiciary understands (or should understand) the scope of congressional power. As to guns near schools, though, we also have some numbers. In 1994, there were three cases filed under the Gun Free School Zones Act of 1990 (the statute at issue in Lopez). In 1995, seven new cases were filed under the statute in the months prior to the Court’s invalidation of the law (in April of that year). In response to Lopez, in September of 1996 Congress reenacted the invalidated statute with a new Commerce Clause hook that constrained the law’s application to any firearm “that has moved in or that otherwise affects interstate or foreign commerce.” Prosecutors wasted no time in putting the revised statute to use: by the close of 1996, 42 cases had been filed in federal district court.
Lopez—recall the revolution—had aspirations of super-precedence, but within a short period it had plainly failed to achieve the goal. Today, because Lopez does not meaningfully constrain congressional power (not even to regulate guns in or near schools) it is best understood as a subprecedent.
In sum, Lopez demonstrates that efforts to create a strong line of doctrine might not succeed. Courts that trade on commitments to stare decisis to launch doctrinal innovations run the risk that those efforts will come up short. Subprecedent, because it might be unplanned, captures that phenomenon.
Unplanned subprecedents also constitute a seventh and the final category. In this category, subprecedents emerge when case law is not settled. Despite efforts by the Court to issue a solid doctrinal line that will guide future decisionmaking, the Court issues contrary and confusing decisions in short periods of time. The result is that, like a television screen that never focuses, reliable doctrinal rules fail ever to come into view. We are left with a muddle of cases that lack general reliability.
A prime example is the Court’s shifting and confusing rulings on the constitutionality of the death penalty. The story is well known and requires only brief explication. In 1972, following growing public concern with how the death penalty was administered, in Furman v. Georgia the Court issued a single-paragraph per curiam order holding that in the cases before it the state capital punishment statutes, by giving too much discretion to prosecutors and juries, violated the Eighth Amendment.  The per curiam order was accompanied, bizarrely, by nine separate opinions: one from each Justice, and with no Justice joining in any other. The effect of Furman was to halt the use of the death penalty until, four years later, in Gregg v. Georgia, the Court ruled that intervening reforms in state sentencing processes had resolved the constitutional infirmities.
The transition from Furman to Georgia itself produced uncertainty: would the Court in another four years again reverse course? But that is just one chapter in the story. Since 1976, the Court has issued, without discernable rhyme or reason, a string of decisions invalidating the use of the death penalty in particular circumstances: for rape of an adult woman where the victim is not killed; for murder absent specific aggravating factors; for a minor participant in a felony who does not kill, attempt to kill, or intend to kill; if the defendant is insane; for crimes committed at age fifteen or younger; and for child rape and other crimes that did not result in death of a victim.
Yet the trajectory of the death penalty cases has not been in a single direction. The Court held in 1989, in Stanford v. Kentucky, that the death penalty could be imposed for crimes committed at age 16 or 17, but then overruled that decision in 2005 in Roper v. Simmons. In 1989, in Penry v. Lynaugh, the Court held that there was no bar to executing a mentally-retarded defendant, but overruled that decision thirteen years later in Atkins v. Virginia. In 1990, in Walton v. Arizona, the Court held that a judge could find the requisite aggravating factors for imposition of the death penalty; in 2002, in Ring v. Arizona, the Court held these factors must be found by a jury.
We are also clearly not at the end of the story. Recently, in a case involving the validity of Oklahoma’s lethal-injection protocol, Justice Breyer (joined by Justice Ginsburg) wrote a forty-page dissent calling for consideration of whether, as it is administered today, the death penalty has become unconstitutional. Justices often reiterate their view that a case has been wrongly decided and should be overturned. But Breyer’s point was different. He explained that since Gregg, “[t]he circumstances and the evidence of the death penalty’s application have changed radically” and that “[g]iven those changes, I believe that it is now time to reopen the question.” Breyer argued that four decades of experience showed that efforts on the part of the states to develop adequate processes for administering capital punishment “ha[d] failed” and that “[t]oday’s administration of the death penalty involves three fundamental constitutional defects: (1) serious unreliability, (2) arbitrariness in application, and (3) unconscionably long delays that undermine the death penalty’s penological purpose.” Citing dozens of empirical studies and other evidence, Breyer concluded:
I recognize that in 1972 this Court, in a sense, turned to Congress and the state legislatures in its search for standards that would increase the fairness and reliability of imposing a death penalty. The legislatures responded. But, in the last four decades, considerable evidence has accumulated that those responses have not worked. Thus we are left with a judicial responsibility. The Eighth Amendment sets forth the relevant law, and we must interpret that law. . . . I believe it highly likely that the death penalty violates the Eighth Amendment. At the very least, the Court should call for full briefing on the basic question.
Breyer has since pressed the same argument in other cases.
The Court’s shifting case law on capital punishment and the death penalty poses a basic question: why should anybody adhere to any particular decision when, to state the obvious, lives are at stake? No sensible observer would confidently conclude that if, in reliance upon the patchwork of current Supreme Court decisions, we execute a particular defendant today, ten years from now we will still consider the execution consistent with the Constitution. Subprecedents, including those that are unplanned, simply do not provide reliable guidance to lower courts and other actors.
Here, a comparative law lens brings some additional clarity. Recently, the Constitutional Court of Italy has taken the position that domestic courts should only defer to rulings of the European Court of Human Rights on the requirements of the European Convention on Human Rights if those judgments represent “consolidated . . . case law,” that is, “case-law which has been consistently applied” by the European Court. If, instead, a decision of the European Court involves a new interpretation or ruling, or the court, over time, has not been consistent in its interpretations and applications of Convention provisions, then domestic courts are free to ignore the rulings that the European Court issues. On this view, it makes little sense for domestic courts, though bound by the Convention (and the European Court’s interpretations of it) to follow a ruling that reflects “a position that has . . . [not yet] become final,” such that there is not a “definitive choice in favor of one approach [to the Convention provision at issue] as opposed to another.” This, in essence, is the recognition that subprecedents may result from confused lines of decision and that their force in a hierarchical system is weaker—and, at least in the view of the Italian court—justifiably so.
Subprecedents are an important component of a legal system that values stare decisis. Any effort to understand how stare decisis operates, how it shapes and constrains judicial decisionmaking, and how it might usefully be reoriented or reformed requires attention also to the nature and purposes of subprecedents. As this essay demonstrates, subprecedents can take different forms and serve different roles. Sometimes, subprecedents empower courts. They allow judges to issue rulings limited to individual cases and circumstances and thus to head off broader uses and applications of a decision or its supporting rationale. Subprecedents permit judges also to control the future by building into rulings temporal limits and expiration dates and by manipulating in advance any later stare decisis analysis. In addition, subprecedents provide a powerful tool for courts to reach back and curtail the scope of a case that has fallen out of favor rather than having to overrule it. Subprecedents can also quietly pave the way to and provide later justification for a momentous decision down the road. Sometimes, however, subprecedents thwart judicial control and ambition. They can stop a doctrinal revolution in its tracks. They can also render an area of law shifting and confused in ways that free up and empower lower court judges and other legal actors who are normally constrained by vertical stare decisis.
Recently, in confirmation hearings and in the academic literature, super-precedents—cases that seem immune to overruling—have occupied a good deal of spotlight. That is not hard to understand. For those who, like Kozel, favor a strong commitment to stare decisis, particularly in constitutional cases, as well as for those who worry that such a commitment may be of exaggerated importance or even misguided, designation of some cases as having stronger precedential effect than others requires close attention. Subprecedents merit equal consideration. While super-precedents take center stage, subprecedents tend to operate with quiet persistence in the background. Nonetheless, in important and often unnoticed ways, subprecedents shape opportunities to seek and obtain legal remedies, facilitate and constrain judicial decision-making, and, ultimately, guide the direction of the law.
. Professor of Law, Notre Dame Law School.
. Albert E. Jenner, Jr. Professor of Law and Director, Program in Constitutional Theory, History and Law, University of Illinois at Urbana-Champaign.
. See Michael J. Gerhardt, Super Precedent, 90 Minn. L. Rev. 1204, 1205 (2006) (“[A]t least as a descriptive matter, there may be something akin to ‘super precedent’ in constitutional law.”); Michael Sinclair, Precedent, Super-Precedent, 14 Geo. Mason L. Rev. 363, 364 (discussing origins of the term “super-precedent”); Jeffrey Rosen, So, Do You Believe in ‘Superprecedent’?, N.Y, Times, Oct. 30, 2005, at C1.
. Senator Arlen Spector used the term “super-duper precedent” at the 2005 confirmation hearing of John G. Roberts. See Confirmation Hearing on the Nomination of John G. Roberts, Jr. To Be Chief Justice of the United States: Hearing Before the S. Comm. on the Judiciary, 109th Cong. 144–45 (2005) (asking Roberts whether Roe v. Wade was a “super-duper precedent in light . . . of 38 occasions to overrule it”).
. Not everyone agrees that the term, super-precedent, is a useful way to understand cases. See, e.g., Jack Balkin, Don’t Talk to Me About Superprecedents, https://balkin.blogspot.com/2005/10/dont-talk-to-me-about-superprecedents.html (Oct. 30, 2005) (last visited May 3, 2018) (describing the term “super-precedent” as “highly misleading”). There is also room for disagreement about what counts as super-precedent, though some cases make many lists. See Amy Coney Barrett, Precedent and Jurisprudential Disagreement, 91 Tex. L. Rev. 1711, 1734–35 (2013) (“The following cases are included on most hit lists of superprecedent: Marbury v. Madison, Martin v. Hunter’s Lessee, Helvering v. Davis, the Legal Tender Cases, Mapp v. Ohio, Brown v. Board of Education, and the Civil Rights Cases.”) (footnotes omitted).
. See, e.g., Bruce Ackerman, The Living Constitution, 120 Harv. L. Rev. 1737, 1752 (2007) (“The Supreme Court has an institutional obligation to recognize that superprecedents crystallize fixed points in our constitutional tradition, and should not be overruled or ignored in the course of doctrinal development. . . . [S]uperprecedents resemble formal amendments, . . . in the operational canon.”).
. One example comes from the competing claims by the Justices in the majority and in the dissent to the legacy of Brown v. Board of Education (a decision virtually everyone considers a superprecedent) in Parents Involved in Community Schools v. Seattle School District No. 1, 551 U.S. 701 (2007).
. Compare Confirmation Hearing on the Nomination of Samuel A. Alito, Jr. To Be an Associate Justice of the United States: Hearing Before the S. Comm. on the Judiciary, 109th Cong. 452 (2006) (describing Brown as “one of the greatest, if not the single greatest thing that the Supreme Court of the United States has ever done”), with id. at 432 (stating, in response to a question about whether he believed the Constitution protects a right to abortion, that “if I am confirmed and if this issue were to come up, the first question that would have to be addressed is the question of stare decisis. . . . And then if I were to get beyond that . . . then I would have to go through the whole judicial decisionmaking process before reaching a conclusion”).
. See Sup. Ct. R. 11; Chief Justice Fred M. Vinson, Address to the American Bar Association: Work of the Federal Courts (Sept. 7, 1949), in 69 S. Ct. v, vi (1949) (explaining that the Court “is not, and has never been, primarily concerned with the correction of errors in lower court decisions”).
. Maples, 565 U.S. at 290.
. Id. at 290 (Alito, J., concurring).
. See, e.g., Young v. Westbrooks, 702 F. App’x. 255, 265 (6th Cir. 2017) (“The Maples Court was careful to cabin its abandonment finding to the ‘extraordinary circumstances’ and ‘uncommon facts’ presented by that case.”); Raplee v. United States, 842 F.3d 328, 334 (4th Cir. 2016) (“Although the facts of this case bear some similarity to those in Maples, they differ in a crucial respect: abandonment by his attorneys did not cause Raplee to miss the filing deadline.”).
. See Jason Mazzone, When the Supreme Court is Not Supreme, 104 Nw. U. L. Rev. 979, 1015–28 (2010) (describing how federal legislation and Supreme Court decisions have curtailed federal habeas review).
. Maples v. Dunn, No. 5:03–CV–2399–SLB, 2015 WL 5444096, at *14 (N.D. Ala. Sept. 14, 2015).
. Maples v. Commissioner, No. 15-14586, 2018 WL 1640132, at *9 (11th Cir. Apr. 5, 2018).
. See, e.g., id. at 343 (“While the Court, at the close of its Powell opinion, did, by its language, as this Court frequently does, limit its holding to the particular facts and circumstances of that case, its conclusions about the fundamental nature of the right to counsel are unmistakable.”).
. 531 U.S. 98 (2000) (per curiam).
. Id. at 134–35 (Souter, J., dissenting); id. at 146–47 (Breyer, J., dissenting).
. Id. at 119 (Rehnquist, C.J., concurring).
. Id. at 123 (Stevens, J., dissenting); id. at 137 (Ginsburg, J., dissenting).
. See id. at 104–05 (first quoting Harper v. Virginia Bd. of Elections, 383 U.S. 663, 665 (1966); and then quoting Reynolds v. Sims, 377 U.S. 533 (1964)).
. See, e.g., Bush v. Gore: The Question of Legitimacy (Bruce Ackerman ed., 2002); Alan M. Dershowitz, Supreme Injustice: How the High Court Hijacked Election 2000 (2001); Richard A. Posner, Breaking the Deadlock: The 2000 Election, the Constitution, and the Courts (2001).
. See, e.g., Hunter v. Hamilton County Bd. of Elections, 635 F.3d 219, 234–36, 241–42 (6th Cir. 2011) (invoking Bush in reviewing a preliminary injunction in a case raising equal protection and due process challenges to a county board’s treatment of provisional ballots); Ariz. Minority Coal. for Fair Redistricting v. Ariz. Indep. Redistricting Comm’n, 121 P.3d 843, 856–57 (Ariz. Ct. App. 2005) (rejecting plaintiff’s reliance upon Bush in a case challenging redistricting measures).
. See, e.g., Petition for Writ of Certiorari, Segovia v. United States, 2018 WL 1920389, at *23 (2018) (No. 17-1463); Petition for Writ of Certiorari, Bergdoll v. Torres, 2018 WL 1891485, at *15 (2018) (No. 17-1450); Petition for Writ of Certiorari, Pickens v. Gannett Co., Inc., 2016 WL 7494804, at *16 (2016) (No. 16-826).
. See Arizona v. Inter Tribal Council of Ariz., Inc., 570 U.S. 1, 36 n.2 (Thomas, J., dissenting) (“This Court has recognized, however, that ‘the state legislature’s power to select the manner for appointing [presidential] electors is plenary; it may, if it chooses, select the electors itself.’”) (quoting Bush v. Gore, 531 U.S. 98, 104 (2000) (per curiam)).
. See, e.g., Harris v. Ariz. Indep. Redistricting Comm’n, 136 S. Ct. 1301 (2016) (rejecting equal protection challenge to state legislative district map and providing extensive discussion of equal protecting and voting); Crawford v. Marion Cy. Election Bd., 553 U.S. 181 (2008) (rejecting Fourteenth Amendment challenges to state voter identification law).
. Id. at 341–43 (quotation omitted).
. Id. (citing Bakke, 438 U.S. 265 (1978)).
. Planned Parenthood of Se. Pa. v. Casey, 505 U.S. 833, 855 (1992) (plurality opinion).
. Lassiter, 360 U.S. 45 (1959).
. Morgan, 384 U.S. at 646–47.
. Id. at 668 (Harlan, J., dissenting).
. Id. at 651, n.10 (“We emphasize that Congress’ power under § 5 is limited to adopting measures to enforce the guarantees of the Amendment; § 5 grants Congress no power to restrict, abrogate, or dilute these guarantees.”).
. Boerne, 521 U.S. at 512–15.
. 134 S. Ct. at 1642 (Scalia, J., concurring).
. 557 U.S. 193, 204 (2009).
. Id. at 216 (Thomas, J., concurring in part and dissenting in part).
. See, e.g., id. at 542 (“These basic principles [of Northwest Austin] guide our review of the question before us.”); id. at 544, 553.
. Id. at 336; see Shelby County, 570 U.S. at 568 (Ginsburg, J., dissenting) (“Katzenbach supplies the standard of review.”).
. Roberts thus complained that the dissenting Justices had missed the significance of the 2009 ruling. Id. at 556.
. Id. at 588 (Ginsburg, J., dissenting) (citations omitted).
. This is not unheard of. See, e.g., Furman v. Georgia, 408 U.S. 238 (1972).
. Gerhardt, supra note 3, at 1213.
. United States v. Bailey, 115 F.3d 1222, 1233 (5th Cir. 1997) (Smith, J., dissenting) (“Lopez is a landmark, signaling the revival of federalism as a constitutional principle, and it must be acknowledged as a watershed decision in the history of the Commerce Clause.”); Larry D. Kramer, The Supreme Court 2000 Term: Foreword: We the Court, 115 Harv. L. Rev. 4, 129 (2001) (citing the “revolution” in federalism doctrine).
. See, e.g., Mark Tushnet, Permissible Gun Regulations After Heller: Speculations About Method and Outcomes, 56 UCLA L. Rev. 1425, 1441 (2009) (“Even without any relevant change in the Court’s composition, the Federalism Revolution inaugurated by Lopez sputtered out.”); Rachel E. Barkow, Federalism and Criminal Law: What the Feds Can Learn from the States, 109 Mich. L. Rev. 519, 529 (2011) (“The Court’s federalism revolution seems to have gone out with a whimper, leaving little room for constitutional regulation of the issue.”).
. Id. at 45 (O’Connor, J., dissenting).
. Confirmation Hearing on the Nomination of John G. Roberts, Jr. to Be Chief Justice of the United States: Hearing Before the S. Comm. On the Judiciary, 109th Cong. 225, 271–72 (2005).
. Jason Mazzone & Carl Emery Woock, Federalism as Docket Control, 94 N.C. L. Rev. 7, 82 (2015) (citing Bureau of Justice Statistics).
. 18 U.S.C. § 922(q)(2)(A) (Supp. IV 1998). The amendments were included in the Omnibus Consolidated Appropriations Act of 1997, 110 Stat. 3009, 3009-369-71 (1996). While the Supreme Court has not ruled on whether the statute, as amended, is constitutional, lower courts have upheld it. See, e.g., United States v. Danks, 221 F.3d 1037 (8th Cir. 1999) (holding the revised statute constitutional); United States v. Tait, 202 F.3d 1320 (11th Cir. 2000) (same).
. Mazzone & Woock, supra note 121.
. Coker v. Georgia, 433 U.S. 584 (1977).
. Godfrey v. Georgia, 446 U.S. 420 (1980).
. Enmund v. Florida, 458 U.S. 782 (1982). Still, five years later, the Court held imposition of the death penalty is valid for a felony-murder defendant who was a major participant in the underlying felony and exhibits extreme indifference to human life. Tison v. Arizona, 481 U.S. 137 (1987).
. Ford v. Wainwright, 477 U.S. 399 (1986).
. Thompson v. Oklahoma, 487 U.S. 815 (1988).
. Kennedy v. Louisiana, 554 U.S. 407 (2008).
. Glossip v. Gross, 135 S. Ct. 2726 (2015).
. Id. at 2755–56 (Breyer, J., dissenting).
. See, e.g., Dunn v. Madison, 138 S. Ct. 9, 13 (2017) (Breyer, J., concurring).
. Corte Cost., 14 gennaio 2015 (It.) translated in Judgment No. 49 Year 2015 Corte Costituzionale, https://www.cortecostituzionale.it/documenti/download/doc/recent_judgments/S49_2015_en.pdf.