Printer Friendly

Questioning marks: plurality decisions and precedential constraint.

Table of Contents

Introduction

I. Marks v. United States and Its Discontents
   A. Marks and the "Narrowest Grounds" Rule
   B. Marks in the Lower Courts: Three Different Ways of Counting to
      Five
      1. The "implicit consensus" approach
      2. The "fifth vote" approach
      3. The "issue-by-issue" approach
   C. The Supreme Court's Indifference to the Practical Problems of
      Marks

II. A Fourth Way of Counting to Five: Plurality Precedent and Shared
Agreement

   A. Stare Decisis and the Relationship Between Results, Rules, and
      Reasons
   B. What It Means to "Follow the Result" of a Plurality Decision
      1. Plurality decisions as incompletely theorized agreements
      2. Distinguishing depth from width
      3. Connecting results to reasons

III. The Case for the Shared Agreement Approach to Plurality Precedent

   A. Consistency with Marks
   B. Majoritarianism and Judgment-Supportiveness as Criteria of
      Precedential Legitimacy
   C. Implications for Plurality Precedent
   D. But "What About Guidance?

IV. Implementing the Shared Agreement Approach

Conclusion


Introduction

Justice William Brennan famously quipped that a critical skill for a Justice of the Supreme Court was the ability to count to five. (1) But with due respect to Justice Brennan, knowing how to "count to five" is at least as important for lower court judges forced to grapple with the Supreme Court's decisions. After all, if a majority of the Court's members cannot agree on what governing law requires, they are free to issue separate opinions expressing their own personal views of the law. Where such disagreements prevent the Court from converging on a single majority rationale for the outcome in a case-notwithstanding majority agreement on what that outcome should be--the result is a plurality decision. (2)

Judges in the lower courts do not possess such interpretive freedom. According to longstanding and widely accepted legal norms, lower court judges are strictly bound by controlling Supreme Court precedent regardless of their own views of the merits. (3) But the Supreme Court has provided such courts with frustratingly little guidance regarding how they should understand the precedential effect of the Court's own plurality decisions. The Court's clearest directive on the subject was set forth in a single sentence of a decision handed down more than four decades ago--Marks v. United States. (4) The so-called "narrowest grounds rule" that derives from Marks instructs that when the Justices fail to converge on a single majority rationale for a decision, "the holding of the Court may be viewed as that position taken by those Members who concurred in the judgments on the narrowest grounds." (5)

But this cryptic directive leaves many questions regarding the precedential force of Supreme Court plurality decisions unanswered. For example, does Marks require lower courts to search for a single "narrowest" opinion issued in the precedent-setting case and accord that opinion full stare decisis effect? Many lower court judges believe that it does--even where the putatively "narrowest" opinion reflects the reasoning of only one of the Court's nine members. (6) But others disagree, finding it inappropriate to accord binding effect to portions of a putatively "narrowest" opinion in which a majority of Justices did not explicitly or implicitly acquiesce. (7) Even if lower court judges could agree on an answer to this first question, there would remain a further unanswered question regarding what criteria of "narrowness" they should use to identify the "narrowest grounds" of decision in the precedent case. (8) Yet another unanswered question involves what role, if any, dissenting opinions should play in the Marks analysis. (9) These and other questions regarding the proper application of the Marks framework have long bedeviled lower courts' efforts to identify the controlling portions of Supreme Court plurality decisions.

The conceptual confusion surrounding Marks presents an important practical challenge for lower courts. Although Supreme Court plurality decisions were historically rare, they have grown more frequent since the mid-twentieth century and are now a familiar feature of the Court's decisionmaking. (10) As many Court watchers have observed, plurality decisions often occur in cases involving especially difficult and highly salient legal issues on which public opinion is sharply divided. (11) Some of the most significant and divisive Supreme Court cases in recent history--involving such issues as abortion, (12) gun control, (13) voting rights, (14) affirmative action, (15) capital punishment, (16) and the scope of congressional authority under the Commerce Clause (17)--have been decided by plurality decision. At the same time, the effects of plurality decisions extend well beyond such high-profile contexts. The proper interpretation of plurality precedent also matters for a variety of less prominent legal issues that nonetheless carry substantial importance to the workaday business of the federal courts, such as criminal procedure, (18) sentencing, (19) personal jurisdiction, (20) class certification, (21) and federal preemption of state law. (22)

By their nature, plurality decisions implicate a tension between two distinct but closely related forms of precedential obligation. The first of these obligations requires precedent-following courts to reconcile their decisions with the precedent court's specific judgment--that is, its specification of which party won, which party lost, and the nature of the relief awarded, if any. The second obligation requires that such courts also conform their decision-making to the more generally applicable rules and rationales on which the precedent-setting court relied in reaching that judgment. (23)

These two forms of precedential obligation usually support and reinforce one another. The traditional common law conception of a decision's binding effect--reflected in the notion of the ratio decidendi (literally, the court's "reason for deciding") (24)--presupposes such a connection by limiting a decision's precedential force to those portions of the deciding court's reasoning that were necessary to its judgment. (25) In cases that result in a plurality decision, this presupposed connection between result and rationale is lacking. Because the Justices whose votes were collectively necessary to the judgment do not agree with each other on the appropriate rationale, it is not possible to identify a single opinion from the precedent case as reflecting the controlling ratio.

But the inability to pick out a single opinion from the plurality case as reflecting the controlling ratio does not render the ratio decidendi approach to precedent unworkable. This Article proposes a revised approach to plurality precedent that builds from the ratio decidendi model's core premise: namely, that the precedential effect of a judicial decision should be determined by reference to the deciding court's reasons for judgment.

This revised approach starts from the recognition that plurality decisions involve a special kind of incompletely theorized agreement. (26) Specifically, the defining feature of a plurality decision is agreement among a majority of the Justices on the case judgment without any comprehensive agreement on the reasons why that judgment was correct. But the scope of the Justices' agreement will rarely be limited to the specific factual details of the particular dispute. Rather, the majority Justices will typically support their conclusions by reference to more generally applicable reasons for why the case should be decided in a particular way.

Though no single opinion in such a case provides a comprehensive account of why the winning party won and the losing party lost, the collective sei of the opinions that were together necessary to the judgment can supply such information. Where a case results in a plurality decision, the winning party won and the losing party lost because each of the judgment-supportive opinions compelled that outcome. By looking to this shared agreement among the majority Justices, lower courts can identify the universe of cases that are sufficiently "like" the precedent case to demand consistent treatment. This universe consists of those cases in which the reasons provided by each of the Justices whose vote was necessary to the judgment in the precedent case would compel the same result.

Of course, this approach leaves lower courts without binding precedential guidance in an important category of cases, namely those in which resolution requires choosing from among the various judgment-supportive rationales from the precedent case. But the resulting discretion available to lower courts is a direct result of the Supreme Court's own inability to converge on a single rationale. The Supreme Court's failure to reach majority consensus on the controlling rationale can thus be seen as an implicit decision not to resolve that issue for the lower courts and an implicit delegation of authority to those courts to continue addressing the issue in the manner they did before the Supreme Court intervened. But even in this category of cases, the plurality decision may still exert some meaningful constraining force by closing off certain rationales that may otherwise have been available to the lower courts, including any rationale that would have led to a different result in the precedent case itself.

This Article proceeds in four Parts. Part I briefly outlines the origins of the Marks narrowest grounds rule and describes three distinct approaches to that rule that are discernible in the lower courts' decisions: (1) the "implicit consensus" approach, which views the narrowest grounds rule as applicable to only a limited subset of plurality decisions involving logically "nested" or "telescoping" rationales; (2) the "fifth vote" approach, which treats as binding the opinion reflecting the views of the median or "swing" Justice in the precedent case; and (3) the "issue-by-issue" approach, which involves looking across all of the opinions in the case--including the plurality, concurrences, and dissents--to find points of agreement on discrete legal issues that were endorsed by a majority of the participating Justices. Part I also briefly discusses the inconsistent and haphazard treatment the Supreme Court has given its own plurality precedents in the decades following Marks.

Part II introduces and describes a revised approach that focuses on identifying the overlapping agreement among the majority Justices regarding why the particular judgment in the precedent case was correct. Under this "shared agreement" approach to plurality precedent, lower courts are tasked with identifying, from the various judgment-supportive opinions in the precedent case, the universe of future cases in which each of the judgment-supportive rationales from the plurality decision would compel the same result. This set of mutually agreed-upon results forms the basis for the lower courts' precedential obligation. Any result that would have been reached under every one of the judgment-supportive rationales constitutes a result that the lower court itself is similarly bound to reach in the later case. Where, however, the result would depend on a choice between two or more conflicting rationales from the precedent case, lower courts will typically enjoy a limited domain of interpretive discretion allowing them to choose which of the competing rationales reflects the best view of governing law.

Part III contends that this shared agreement approach to plurality precedent reflects a better way of understanding lower courts' precedential obligations than any of the three current approaches to the narrowest grounds rule. The shared agreement approach is consistent with the language and holding of Marks and fits comfortably with the handful of post-Manb-decisions in which the Supreme Court has explicitly invoked the narrowest grounds doctrine. The shared agreement approach also best accords with traditional conceptions of precedential legitimacy, which limit precedential effect to statements that are both supported by a majority of the Justices and necessary to the judgment in the precedent-setting case. Finally, by allowing lower courts a limited degree of interpretive freedom, the shared agreement approach holds out the possibility that the Supreme Court might benefit from the lower courts' continued engagement with the complex legal questions that divided the Justices in the precedent case.

Part IV illustrates how the shared agreement approach would operate in practice by considering its application to a recent Supreme Court plurality decision--Shady Grove Orthopedic Associates v. Allstate Insurance Co. (27)

I. Marks v. United States and Its Discontents

A. Marks and the "Narrowest Grounds" Rule

A full understanding of the Supreme Court's canonical guidance to lower courts regarding the precedential effect of its plurality decisions requires a brief review of the case in which that guidance was delivered: Marks v. United States. (28) Marks involved a due process challenge to the petitioners' criminal obscenity convictions. The petitioners contended that their convictions were based on the retroactive application of a restrictive First Amendment standard the Supreme Court had articulated only after their allegedly criminal conduct had occurred. (29)

To understand the basis of the petitioners' challenge, it is necessary to have a basic understanding of the three First Amendment precedents that were central to their argument. In the earliest of the three cases, Roth v. United States, (30) the Supreme Court articulated a standard to determine the constitutional permissibility of obscenity prosecutions. This standard asked "whether to the average person, applying contemporary community standards, the dominant theme of the material taken as a whole appeals to prurient interest." (31)

Nine years later, the Court revisited Roth in A Book Named "John Cleland's Memoirs of a Woman of Pleasure" v. Attorney General, (32) a plurality decision in which the Justices divided over the appropriate First Amendment standard. Justices Black and Douglas joined an opinion endorsing a categorical rule prohibiting all "governmental action aimed at suppressing obscenity." (33) Justice Stewart, writing only for himself, would have confined obscenity prosecutions to a narrow category of cases involving "hardcore pornography." (34) The plurality opinion, authored by Justice Brennan and joined by Chief Justice Warren and Justice Fortas, articulated a new three-part test for assessing whether the Roth standard, "as elaborated in subsequent cases," had been met. (35) This new test added two further requirements to the Roth standard: that the allegedly obscene material be (1) "patently offensive because it affronts contemporary community standards relating to the description or representation of sexual matters" and (2) "utterly without redeeming social value." (36) Three Justices dissented in separate opinions, urging application of either the Roth standard or an even less speech-protective standard. (37)

Seven years after Memoirs, in Miller v. California, a clear majority of the Court rejected each of the proposed standards endorsed by the Justices concurring in Memoirs and instead endorsed a new standard that was substantially similar to the earlier articulated Roth standard. (38)

This shifting First Amendment background set the stage for Marks. The petitioners in Marks were charged with obscenity based on conduct that occurred after Memoirs but before Miller, and they were convicted after Miller had been decided. (39) The trial court instructed the jury under the newly articulated Miller standard, and the jury convicted. (40) The court of appeals affirmed the conviction, rejecting the petitioners' retroactivity challenge and noting that the standard on which the petitioners' challenge relied--the three-part test endorsed by the Memoirs plurality--"had never been approved by a plurality of more than three Justices at any one time." (41)

The Supreme Court reversed, concluding that Memoirs had, in fact, changed the applicable First Amendment standard and that applying the later-announced Miller standard to the defendants' conduct was therefore impermissible. In so deciding, the Marks Court instructed that "[w]hen a fragmented court decides a case and no single rationale explaining the result enjoys the assent of five Justices, 'the holding of the Court may be viewed as that position taken by those Members who concurred in the judgments on the narrowest grounds."' (42) Applying this narrowest grounds test, the Marks Court concluded that Justice Brennan's plurality opinion constituted the holding of Memoirs. (43)

B. Marks in the Lower Courts: Three Different Ways of Counting to Five

Marks foreclosed what may have been the easiest way for lower courts to deal with Supreme Court plurality decisions--that is, by simply denying their precedential force and treating the various opinions issued in those decisions as mere persuasive authorities. (44) But while Marks made clear that lower courts were required to accord at least some precedential effect to plurality decisions, the Court provided little guidance regarding precisely how lower courts should go about identifying which aspects of those decisions are binding. The Court's cryptic instruction to treat as binding the "position taken by" the Justices "who concurred ... on the narrowest grounds" (45) has left a great many questions regarding the proper interpretation and application of Marks unanswered.

This Part surveys three distinct approaches that lower courts have employed in seeking to discern the precedential effect of a Supreme Court plurality decision. The first of these approaches interprets Marks as limited to a narrow subset of plurality decisions reflecting a clearly discernible "implicit consensus" or "common denominator" among the Justices. The second approach understands Marks as an instruction to lower courts to identify the opinion in a plurality decision that reflects the judgment-critical vote-typically the fifth concurring vote--and treat that opinion as the Court's holding. The third and final approach looks for points of majority consensus among different factions of concurring and dissenting Justices on distinct legal issues raised by the plurality decision.

Though commentators have recognized each of these three approaches as a distinct way of understanding what the narrowest grounds rule requires, (46) the distinction is often blurred or glossed over in lower court opinions applying the Marks doctrine. (47) And most lower courts have not been particularly consistent in applying a single approach 48 Thus, for example, different panels within the same circuit court of appeals will often adopt different approaches depending on the particular plurality decision that is being considered. (49) This doctrinal confusion among lower courts regarding the proper application of Marks has produced a series of longstanding circuit splits that have resulted from lower courts' disagreements regarding how the narrowest grounds rule should apply to particular Supreme Court plurality decisions. (50)

1. The "implicit consensus" approach

The "implicit consensus" approach interprets the Marks Court's instruction to seek the "narrowest grounds" of a decision as workable only with respect to a limited subset of plurality decisions. As the D.C. Circuit explained in a seminal decision establishing the implicit consensus approach as its official understanding of Marks.

   Marks is workable--one opinion can be meaningfully regarded as
   "narrower" than another--only when one opinion is a logical subset
   of other, broader opinions. In essence, the narrowest opinion must
   represent a common denominator of the Court's reasoning; it must
   embody a position implicitly approved by at least five Justices who
   support the judgment. (51)


As understood by the D.C. Circuit and other courts that have embraced a similar understanding of Marks, the narrowest grounds rule applies only in those circumstances "where it is clear that one opinion would apply in a subset of cases encompassed by a broader opinion" (52) such that "the rationales for the majority outcome are nested, fitting within each other like Russian dolls." (53) This particular alignment of opinions can be represented graphically using the Venn diagram depicted in Figure 1 below.

Figure 1 reflects a simple continuum in which the respective legal rules endorsed by the two "broader" opinions (Rules A and B) wholly subsume the rule endorsed by the "narrowest" opinion (Rule C). Consider, for example, the alignment of the three judgment-supportive opinions in Memoirs--the decision at issue in Marks itself. (54) The broadest rationale supporting the judgment in that case--the proposed categorical ban on obscenity prosecutions endorsed by Justices Black and Douglas--corresponds to the circle labeled Rule A. (55) The circles labeled Rule B and Rule C correspond, respectively, to the somewhat less speech-protective "hardcore pornography" rationale endorsed by Justice Stewart's sole concurrence and the seemingly still-more-permissive three-part test endorsed by Justice Brennan's plurality opinion. (56)

Where the opinions in the precedent case align in this way, the narrowest opinion can be seen as reflecting the core of an implicit consensus among the concurring Justices regarding the proper application of their respective rationales. (57) For example, any obscenity prosecution deemed impermissible under Justice Brennan's plurality opinion in Memoirs would necessarily be considered impermissible under the more speech-protective rationales endorsed by Justices Stewart, Black, and Douglas. Cases involving prosecutions that would be deemed impermissible under Justice Brennan's three-part test thus represent the universe of cases in which the rationales of each Justice whose vote was necessary to the judgment in Memoirs would point to the same result.

But not every plurality decision involves this particular alignment of judgment-supportive rationales. Many--if not most--plurality decisions involve situations in which the rationales of the various judgment-supportive opinions overlap in some respects but diverge in others. Figure 2 below illustrates this phenomenon.

The circles labeled Rule A, Rule B, and Rule C in Figure 2 have the same areas as the three identically labeled circles in Figure l. (58) But the set of results Rule C encompasses is no longer a subset of the results produced by either Rule A or Rule B. Instead, Rule C now produces some results that would not have been endorsed by the Justices who endorsed either of the two alternative rationales. Thus, the implicit consensus justification for identifying Rule C as the "narrowest grounds" of the Court's decision in the circumstances depicted in Figure 1 no longer applies. (59)

Consider, for example, the Supreme Court's plurality decision in Rapanos v. United States. (60) Rapanos addressed a challenge to the scope of the regulatory authority conferred on the U.S. Army Corps of Engineers (Corps) by the Clean Water Act (CWA). (61) The CWA prohibits discharge of certain pollutants into "navigable waters" but authorizes the Corps to issue permits allowing discharge of "dredged or fill material" into such waters. (62) The CWA defines the phrase "navigable waters" as "the waters of the United States, including the territorial seas." (63) The petitioners in Rapanos sought to develop certain wetlands that were not directly adjacent to any navigable waters but that lay "near ditches or man-made drains that eventually empty into traditional navigable waters." (64) The Corps asserted that such development was impermissible under the CWA because the wetlands at issue constituted "'waters of the United States' that could not be filled without a permit." (65)

In Rapanos, a four-Justice plurality opinion authored by Justice Scalia narrowly construed the phrase "waters of the United States," as used in the CWA, to exclude "channels through which water flows intermittently or ephemerally, or channels that periodically provide drainage for rainfall." (66) Applying this definition, the plurality concluded that the Corps's regulatory jurisdiction extended to only those wetlands having "a continuous surface connection to bodies that are 'waters of the United States' in their own right, so that there is no clear demarcation between 'waters' and wetlands." (67) Because the lower court had applied a different jurisdictional test in upholding the Corps's exercise of jurisdiction and because of the "paucity of the record," the plurality Justices believed the case should be remanded to determine whether jurisdiction existed. (68)

In a sole concurrence, Justice Kennedy joined the plurality in rejecting the expansive interpretation of the statute that had been urged by the Corps but adopted a different jurisdictional test for determining the Corps's regulatory authority. (69) Under Justice Kennedy's interpretation, wetlands would be subject to regulation if they possessed a "significant nexus" with navigable "waters of the United States," meaning that the wetlands, "either alone or in combination with similarly situated lands in the region, significantly affect the chemical, physical, and biological integrity of other covered waters more readily understood as navigable."' (70) Like the plurality, Justice Kennedy believed the case should be remanded to the lower court for further proceedings. (71)

The dissent, authored by Justice Stevens and joined by three additional Justices, endorsed a broad interpretation of the CWA that would have authorized a much broader scope for regulation than either the plurality's approach or Justice Kennedy's approach. (72) On the basis of that interpretation, the dissenters argued that the decision below should have been affirmed. (73)

Although some lower courts have concluded that Justice Kennedy's interpretation of the CWA would likely result in the invalidation of fewer federal regulations than would the plurality's, (74) the jurisdictional test he endorsed was not wholly subsumed within the plurality's test and could potentially result in a finding of jurisdiction in at least some cases where the plurality's test would not. (75) For example, the plurality's "continuous surface connection" test might allow for regulation of a very small stream flowing into an extremely large body of water, whereas the "significant nexus" test endorsed by Justice Kennedy might not. (76) As a result, the implicit consensus approach would not single out any particular opinion from the case as the "narrowest." (77)

The implicit consensus approach thus offers limited guidance to lower courts regarding the controlling force of Supreme Court plurality decisions. Unless the rationales offered by the participating Justices happen to line up with one another along a simple continuum in the manner suggested by Figure 1 above, it is impossible to identify any single opinion as reflecting an implicit consensus among the Justices. Plurality decisions that fail to conform to this paradigm are either treated as wholly nonprecedential (78) or, at most, limited to their "specific results." (79) The approach thus threatens to leave lower courts without meaningful precedential guidance with respect to many--perhaps most--plurality decisions.

2. The "fifth vote" approach

A different understanding of the Marks Court's instruction is reflected in the Third Circuit's decision in Planned Parenthood of Southeastern Pennsylvania v. Casey. (80) In Casey, the Third Circuit considered whether two then-recent Supreme Court plurality decisions had altered the standard of review applicable to laws regulating abortion. (81) In describing the precedential significance of those decisions, the Third Circuit interpreted the Marks narrowest grounds rule to require that "whenever possible, there be a single legal standard for the lower courts to apply in similar cases and that this standard, when properly applied, produce results with which a majority of the Justices in the case articulating the standard would agree." (82) To achieve this objective, the Casey court instructed that lower courts should treat as controlling "the opinion of the Justice or Justices who concurred on the narrowest grounds necessary to secure a majority," even if the opinion reflects the views of only one Justice. (83)

The Third Circuit's approach in Casey essentially views Marks as an instruction to search for the opinion reflecting the views of the Court's median or "swing" Justice--typically, the fifth vote (84)--and accord that decision full precedential effect. (85) The rationale for this "fifth vote" approach is usually premised on one of two alternative grounds: predictive power or constructive consent. The predictive rationale for the fifth vote approach is premised on the intuition that in any future case decided by the same group of Justices and involving the same set of issues, the median Justice's view is likely to control. (86) The constructive consent rationale, on the other hand, is premised on the assumption that the median Justice's preferred rationale is likely to reflect the position that the Court as a whole would have adopted if the Justices had been "forced to choose" a single rationale. (87) Drawing on insights from social choice theory, Maxwell Stearns argues that the fifth vote approach provides a decision rule that will correctly identify the "Condorcet winner"--that is, the option that would prevail over all other options in a set of pairwise comparisons--with respect to most plurality decisions. (88)

To illustrate the concept, Stearns uses the example of Memoirs itself. Stearns notes that the plurality's preferred precedential rule occupied a middle point between the more speech-protective standards endorsed by the other concurring Justices and the less speech-protective standard endorsed by the dissenters. (89) Stearns reasons that "if forced to choose among each of the remaining opinions, those writing or joining the opinions at the outer edge" of this continuum "would most prefer the one closest to them and least prefer the one farthest from them." (90) In other words, the dissenters would presumably prefer the plurality's rationale to either of the more speech-protective concurrences, and the Justices who wrote or joined those concurrences would presumably prefer the plurality's rationale to that of the dissenters. Based on this analysis, Stearns concludes that the Memoirs plurality represented the Condorcet winner in the case and infers that the narrowest grounds rule of Marks "is best understood as an application of the Condorcet criterion to fractured panel Supreme Court decisions." (91)

The fifth vote approach promises guidance with respect to a broader range of plurality decisions than does the implicit consensus approach. But it does so in a way that strikes many jurists and commentators as problematic. Perhaps most controversially, the fifth vote approach treats as binding all aspects of the opinion reflecting the median Justice's views, including propositions that no other participating Justice explicitly or implicitly assented to. (92) The fifth vote approach also implicitly accords weight to the views of dissenting Justices by allowing their views to influence the identification of the median Justice's opinion. (93)

But the fifth vote approach also sometimes fails to deliver clear guidance regarding the precedential effect of a plurality decision. Consider, for example, a hypothetical civil case brought against a foreign defendant in federal court where the defendant argues that the court lacks both subject matter jurisdiction to hear the case and personal jurisdiction over the defendant. Imagine that three Justices conclude that subject matter jurisdiction was lacking but refuse to reach the personal jurisdiction question. Suppose further that three other Justices believe personal jurisdiction is lacking and vote to dismiss on that ground without reaching the subject matter jurisdiction question. Finally, suppose that the three remaining Justices find both subject matter jurisdiction and personal jurisdiction and would thus allow the case to proceed in federal court. The fifth vote approach is unworkable in this context because the information provided by the various opinions in the case is insufficient to determine whether there is a dominant second-choice option that can be taken to reflect the views of the Court's "median" or "swing" Justice. (94)

The fifth vote approach may also fail to deliver clear guidance in cases that result in a "voting paradox" where one party "receives the votes of a majority of the Justices ... on every relevant issue yet loses the case anyway." (95) Because the Court generally resolves cases by voting on the ultimate outcome rather than on distinct legal issues, (96) it sometimes happens that the issue-level votes that can be inferred by looking to the Justices' individual opinions conflict with the Court's ultimate disposition. (97) Such dispositions are possible whenever "two or more issues are presented to the Court, when no one way of resolving both issues gets majority support, and when there is also a dissent." (98) In such cases, it is not possible to identify a single opinion reflecting the Justices' median position because there are, in effect, multiple distinct "majorities" composed of different Justices who do not share a common ranking for the available options. (99)

3. The "issue-by-issue" approach

A third approach to identifying the "narrowest grounds" of decision under Marks involves "pars[ing]" each of the various opinions in the plurality case-including the plurality opinion, concurrences, and dissents--"to determine each proposition where five or more Justices agree." (100) Like the fifth vote approach, this issue-by-issue approach purports to provide definitive guidance in at least some cases that do not conform to the "nested rationale" paradigm envisioned by the implicit consensus approach. (101) But the issue-by-issue approach avoids the fifth vote approach's uncomfortable conclusion that the views of a single Justice can establish binding precedent for the Court. Instead, the issue-by-issue approach looks for specific propositions that have actually been explicitly or implicitly assented to by a majority of Justices, though perhaps not the same majority whose votes were necessary to the judgment in the precedent case. (102)

Consider again the Supreme Court's Rapanos decision. As discussed above in Part I.B.I, the plurality Justices and Justice Kennedy would have limited the regulatory jurisdiction of the Corps under the CWA but could not agree on a single standard for assessing such jurisdiction. (103) The four dissenters in Rapanos joined an opinion authored by Justice Stevens that endorsed a much more permissive test that would have allowed for jurisdiction in each instance where such jurisdiction would be allowed under either the plurality's proposed test or the alternative test proposed by Justice Kennedy. (104) Because the interpretation of the CWA that Justice Stevens and his three fellow dissenters preferred was broad enough to encompass all instances of regulatory authority that would have been allowed under either test, Justice Stevens asserted that the lower court on remand should find jurisdiction "if either of those tests is met." (105)

Justice Stevens's proposal implicitly invited lower courts to view the various opinions in Rapanos as reflecting a set of implicit issue-by-issue votes on two distinct propositions--namely, (1) whether regulatory jurisdiction was permissible under the plurality's proposed test and (2) whether such jurisdiction was permissible under Justice Kennedy's proposed test. Eight Justices--the four plurality Justices and the four dissenters--would have answered the first question in the affirmative, while five Justices--the dissenters and Justice Kennedy--would have provided the same answer to the second question. Thus, there appears to have been clear majority support among the Justices who participated in Rapanos for finding jurisdiction under each jurisdictional test. (106) Some lower courts have endorsed Justice Stevens's reasoning, concluding that jurisdiction should be found in any instance where either test is satisfied. (107)

Nonetheless, the issue-by-issue approach remains controversial because it requires that dissenters' views be taken into account. (108) Critics contend that the approach conflicts with the Court's instruction that the holding of a case should be determined by looking to the "position taken by those Members who concurred in the judgments on the narrowest grounds." (109) Taking dissenters' views into account also conflicts with the longstanding view that only statements in judicial opinions that are in some way "necessary" to the judgment in the precedent case are entitled to precedential effect. (110) Because dissents, by definition, are not necessary to the judgment in the precedent case, they stand in a position similar to dicta and are thus, arguably, not entitled to precedential effect. (111)

Moreover, the approach is only workable in cases involving a so-called "dual majority" alignment where "there are in effect two majorities: the plurality and concurrence agreeing on the result, and the concurrence and dissent agreeing on the fundamental legal principles involved." (112) Such an alignment will not be present in every plurality decision. Consider again the hypothetical plurality decision discussed above in Part I.B.2, where the majority of Justices agreed jurisdiction was lacking but could not agree on whether personal jurisdiction or subject matter jurisdiction provided the appropriate grounds for dismissal. (113) The issue-by-issue approach would be unhelpful in extracting precedential guidance from such a decision because there are no points of consensus between the three dissenters--who believe the court possesses both subject matter and personal jurisdiction--and either faction of concurring Justices.

C. The Supreme Court's Indifference to the Practical Problems of Marks

Given the confusion that has plagued lower courts' efforts to apply the Marks narrowest grounds rule, it might be natural to expect the Supreme Court to step in with further clarifying guidance regarding how the rule should apply. Multiple individual Justices have expressed concern that the Court's plurality decisions leave lower courts and litigants with insufficient guidance. (114) And similar reservations regarding the workability of the Marks framework itself have found their way into the Court's own opinions. (115)

But despite the Justices' awareness of the difficulties lower courts and litigants routinely encounter in attempting to apply the narrowest grounds rule, the Court has repeatedly passed on opportunities to provide lower courts with clearer guidance regarding the rule's proper application. (116) In the four decades since Marks was handed down, the Court has applied or considered the "narrowest grounds" framework in only seven majority opinions. (117) In two of those opinions, the Court deemed the Marks analysis unhelpful, concluding that it did not seem "useful to pursue the Marks inquiry to the utmost logical possibility when it has so obviously baffled and divided the lower courts that have considered it." (118) In the remaining five cases, the majority did single out either the plurality opinion or one of the concurrences from the earlier case as being the "narrowest" and therefore controlling opinion under Marks. But the Court did not explain in any case why the particular opinion so designated was "narrower" than the others or offer any guidance to lower courts regarding how the Marks analysis should be carried out. And three of those five decisions resulted in a dissenting opinion that accused the majority of misapplying the Marks framework. (119)

Multiple factors feed into the Court's failure to clarify the proper application of Marks. Unlike lower courts, the Supreme Court itself possesses a significant degree of interpretive discretion in determining how much weight to give its own prior precedents. (120) This means that where a clear majority of the Court in a subsequent case favors a rationale endorsed by one of the opinions issued in connection with an earlier plurality decision, that majority can simply dispense with the Marks analysis and endorse its preferred rationale directly. (121) Moreover, even on the relatively rare occasions when the Court has explicitly engaged in the Marks analysis, the Court has sometimes concluded that the analysis is sufficiently complicated to render Marks unhelpful as an analytic tool. (122)

At the same time, the narrowest grounds rule does point to a definitive resolution in a nontrivial subset of plurality decisions: those involving opinions reflecting logically "nested" or "telescoped" rationales. And the Court has sometimes found Marks at least rhetorically useful in singling out a particular holding of a plurality decision as the "narrowest" rationale. (123) The Court has not, however, approached the Marks framework with anything like the theoretical rigor or consistency necessary to provide clear guidance to lower courts. (124)

Finally, even if the Justices of the Supreme Court were inclined to give further guidance regarding the proper application of Marks, the Justices--much like the judges of the lower courts--seem to lack a coherent theory of precisely what the doctrine should aim to accomplish. The proper meaning of the Marks narrowest grounds rule has divided the Supreme Court in much the same way that it has divided the courts below. (125)

II. A Fourth Way of Counting to Five: Plurality Precedent and Shared Agreement

Given the confusion that has long surrounded plurality precedent in general and the Marks narrowest grounds rule in particular, some commentators have despaired of ever constructing a workable and coherent framework to guide lower court decisionmaking. (126) Even commentators who have proposed reforms to the Marks doctrine have sometimes characterized their efforts as "damage control," viewing the task for lower courts as making the best of a bad situation the Supreme Court thrust upon them with its abdication of its institutional responsibility. (127) The foregoing survey of the necessary tradeoffs and incompleteness of the doctrinal solutions lower courts have constructed to make sense of Marks certainly lends some support to such pessimistic diagnoses.

This Part, however, provides a more optimistic perspective on the problem of plurality precedent. Much of the confusion that has surrounded the interpretation of plurality decisions derives from a failure of lower courts and commentators to accept such decisions for what they are--namely, incompletely theorized judicial agreements on particular judgments. The judgment in the precedent case provides the focal point of the majority of Justices' agreement--indeed, the only aspect of the case on which a majority of the Court need agree. The judgment thus provides the natural starting point for assessing the decision's precedential significance. This Part shows how the traditional common law notion of ratio decidendi provides both a useful framework within which to situate the problem of plurality precedent and a set of tools that lower courts can use to discern the binding precedential effect of such decisions.

A. Stare Decisis and the Relationship Between Results, Rules, and Reasons

The traditional starting point for assessing the scope of a given decision's precedential constraint has long been the deciding court's judgment--that is, the court's specification of which party won, which party lost, and the relief awarded, if any. (128) Some distinguished commentators on the common law have urged that only the result of the precedent-setting case should be treated as binding. (129) Under this approach, the task of a precedent-following court is merely "to identify a theory that can explain the results of previous cases, regardless of whether the precedent-setting courts themselves adopted the superimposed theory." (130) The precedent court's own explanation of the legal reasoning on which it based its decision--though perhaps entitled to some persuasive force--would not, on this view, be considered binding on precedent-following courts. (131)

But as numerous judges and scholars have recognized, this "results only" approach to stare decisis is unworkable. (132) One key difficulty with this approach is its inability to specify precisely which facts in the precedent case should matter to the precedent-following court in determining the binding effect of an earlier precedent. No two cases are ever precisely the same, and it will thus always be possible to identify some factual distinction between a present case and some earlier case asserted as a potential precedent. This difficulty cannot be obviously remedied by disregarding "immaterial" or "unessential" factual differences between the two cases. (133) As Henry Monaghan observes, "[t]he relevant facts" of a precedent case "do not identify and classify themselves; criteria are needed to determine what the legally relevant facts are and at what level of generality they are to be specified." (134) Looking to the precedent court's own explanation of the reasoning through which it reached its result significantly eases this difficulty by providing a set of criteria through which to assess the materiality of any discernible factual similarities and differences between the two cases.

Though traces of the facts-plus-outcome approach to precedent can still be glimpsed at times, (135) the more common practice among modern courts is to give precedential effect to at least some aspects of the reasoning through which the precedent-setting court arrived at its decision. (136) This approach binds lower courts by both the specific result in the precedent case and the broader rule or rationale that the precedent court articulated in explaining that result. (137) Ideally, these two forms of precedential obligation--sometimes referred to as "result stare decisis" and "rule stare decisis," respectively (138)--should work together to support and reinforce one another. Knowing the specific judgment the court reached in the precedent case helps concretize the more abstract rule or rationale on which the court relied by illustrating its application to a particular real-world dispute. At the same time, knowing the broader rule or rationale on which the precedent court relied helps lower courts generalize its result to other cases. (139)

But this mutually reinforcing relationship between rule stare decisis and result stare decisis will not be present in every case. For example, a precedentsetting court may issue an opinion that fails to clearly explain its judgment or may even issue a judgment unaccompanied by any explanation at all. (140) Where such a decision is rendered, the conventional approach courts have taken has been to posit a precedential rule to explain the judgment, relying on either their "sense of what the earlier judges probably thought" or their "own judgment about what should have mattered." (141) Such a decision imposes substantially fewer constraints on lower courts' decisionmaking because it allows them to posit a broad range of potential rationales that could account for the precedent court's judgment. But even an unexplained judgment may exert some weak constraining force on later courts by necessitating that any newly developed rationale be capable of accounting for the specific result in the precedent case itself. (142)

Rule stare decisis and result stare decisis may also diverge from one another where a court attempts to set forth a putatively binding precedential rule for future cases without connecting that rule to its specific judgment in the precedent case. (143) Such cases implicate the famously blurry and contested boundary separating holdings from dicta. (144) Despite its often elusive nature, the holding-dicta distinction is central to the common law theory of precedent. The distinction takes on added significance when applied to decisions issued by Article III courts, which by constitutional design are limited to issuing judgments in connection with particularized "[c]ases" or "[cjontroversies" and foreclosed from offering authoritative declarations of law outside that specific context. (145)

The most prominent modern formulation of the holding-dicta distinction focuses on the concept of "necessity" and regards only those aspects of a prior opinion that were in some way "necessary" to the deciding court's judgment as part of the case's holding. (146) Despite the prominence of this definition, it is subject to some well-known criticisms. For example, the necessity-based definition has trouble accounting for cases in which the deciding court relies on multiple alternative rationales. Such decisions are problematic under the necessity-based definition because the precedent-setting court could, by hypothesis, have premised its decision solely on only one of its alternative rationales. (147) Thus neither of the alternative rationales can be said to have been truly "necessary" to the disposition. (148) As a matter of practice, courts routinely treat both prongs of an alternative rationale as part of the holding. (149)

Recognizing the limits of the necessity-based definition, scholars have proposed alternative formulations of the holding-dicta distinction that do not depend on strict logical necessity. (150) But even these alternative accounts generally recognize the centrality of the precedent-setting court's judgmentits ultimate disposition in favor of either the plaintiff or the defendant--in setting the parameters that define the line that separates holding from dicta. (151)

B. What It Means to "Follow the Result" of a Plurality Decision

1. Plurality decisions as incompletely theorized agreements

The distinctive feature of a plurality decision is, as noted above, the existence of a majority agreement on the judgment in a particular case without a corresponding agreement on the underlying rationale that supports that judgment. (152) Such decisions thus reflect a particular species of "incompletely theorized agreement"--a phrase coined by Cass Sunstein to describe instances where decisionmakers are able to reach agreement on a particular proposition or result without reaching comprehensive agreement on the reasons that explain the agreed-upon resolution. (153)

Incompletely theorized agreements come in many forms. Some involve incomplete specification, reflecting general agreement on some relatively abstract principle--such as equality or freedom of speech--without corresponding agreement on how such principles apply to particular cases. (154) Others involve agreement on some midlevel principle without agreement on either the more general theory that justifies that principle or the proper application of that principle to the particular facts. (155) A third variety, and the one that is most centrally relevant to the present inquiry, involves "incompletely theorized agreements on particular outcomes, accompanied by agreements on the low-level principles that account for them." (156) Such agreements avoid the need to reach consensus on abstract theories or principles--a point at which agreement may be difficult or even impossible to achieve--by lowering the level of generality. (157) By focusing more particularly on the bottom-line outcome (that is, "who wins and who loses a case") as well as "a set of reasons" for that outcome "that typically do not venture far from the case at hand," judges may be able to more easily reach consensus. (158)

Plurality decisions involve a similar dynamic. But whereas the types of incompletely theorized judgments Sunstein takes as his focus involve agreement on both case-level outcomes and the "low-level principles" supporting those outcomes, plurality decisions involve agreement on outcomes alone. (159) Such decisions are not, however, equivalent to unexplained rulings, which leave lower courts wholly at sea in determining the basis for the Supreme Court's decision. (160) Rather, plurality decisions typically come accompanied by opinions setting forth the reasons given by the particular Justices who joined in the judgment as to why they believed that judgment was appropriate. These reasons necessarily diverge from one another in some particulars--such divergence being the defining feature of a plurality decision. But they will rarely be narrowly circumscribed by every particular factual detail that happens to have been present in the particular dispute that the Court is called upon to decide.

Reasons are by nature more general than the particular judgments they are offered as reasons for. (161) In offering a reason for a decision, decision-makers not only explain their decision to the affected individuals but also identify the decision as a member of a broader class of decisions supported by the same reason. Thus, to take a nonlegal example, a parent who tells a child that she can stay up late to watch a particular television program because the program is particularly educational implicitly signals a commitment, at least presumptively, to make similar exceptions to the ordinary bedtime for other programs of equivalent or higher educational value. (162)

Similarly, a precedent-setting court that articulates a reason for its decision commits itself (and other courts bound by its precedents) to render decisions consistent with that reason in future cases, including cases that differ in significant respects from the precedent case. Thus, for example, if a precedent-setting court explains its decision that a particular automobile manufacturer is liable to a remote purchaser for injuries caused by a negligently manufactured wheel because manufacturers of potentially dangerous products owe a duty of care to all persons who might foreseeably be harmed by their negligence, (163) the court thereby commits itself (and other courts bound by its precedents) to render decisions consistent with that rationale in all future cases in which that rationale applies. A later case involving different factual circumstances--for example, a negligently manufactured soda bottle or coffee urn (164)--can nonetheless be recognized as materially "similar" to the precedent case because the rule identified by the precedent court as the basis for its decision would compel the same result in both cases. In this way, the practice of reason-giving is intimately connected with--and perhaps even essential to--the practice of precedent following. (165)

Where multiple decision-makers are involved, the matter becomes somewhat more complex, particularly where the decision-makers provide different reasons for their agreed-upon judgment. But the basic structure of the analysis is the same. Consider a variation on the above-described bedtime hypothetical in which the child approaches each of her two parents separately to seek permission to stay up late. The response of the first parent is identical to the one described above, emphasizing the educational value of the program as the reason. The second parent likewise agrees to the request but instead offers as a reason the fact that there is no school the next day. Though both parents agree to the request, their respective reasons diverge in important respects.

But despite this disagreement, the child is not left in the same position she would have been in had her parents each simply nodded their heads without explanation. At a minimum, the parents have collectively committed themselves, at least presumptively, to granting the child's request in any future circumstance where both of their respective reasons are satisfied: that is, where the program in question exhibits sufficient educational value and does not fall on a school night.

Of course, this convergence of reasons does not provide complete guidance. For example, if the child wishes to watch a future program with clear educational value on the night before a school day, the prior decision and the reasons given for that decision will not suffice to determine whether the request should be granted. (166) But incomplete guidance is not the same as no guidance. Despite their incomplete overlap, the reasons given for the prior decision nonetheless provide meaningful guidance regarding an important category of future cases that do not involve either the specific program in question or the particular night in question.

2. Distinguishing depth from width

The possibility of extracting meaningful guidance from an incompletely theorized agreement that lacks a single, mutually agreed-upon rationale illustrates the importance of a distinction that Sunstein draws between the "depth" of judicial decisions and their "width." (167) In Sunstein's terminology, "depth" refers to the comprehensiveness of the decision's theoretical foundations. (168) Incompletely theorized agreements are, by their nature, "shallow rather than deep," reflecting surface-level agreements on outcomes without comprehensive agreement on their underlying foundations. (169)

But such shallowness need not entail narrowness. "Narrowness," in Sunstein's terminology, signals the range of outcomes or situations covered by a particular decision. (170) As Sunstein recognizes, it is possible for a decision to be both shallow and wide, lacking a deep theoretical grounding but nonetheless extending to a potentially broad range of future cases. (171) This distinction is significant because while the depth of a Supreme Court decision may have some effect on the scope of precedential constraint, understanding its width will typically be much more important to assessing the degree of precedential constraint it will exert on lower courts. (172)

The distinction between depth and width is well illustrated by the Supreme Court's decision in McDonald v. City of Chicago. (170) McDonald involved a constitutional challenge to a municipal ordinance that "effectively bann[ed] handgun possession by almost all private citizens who reside[d] in" Chicago. (174) Two years earlier, the Court had invalidated a substantially similar District of Columbia ordinance as a violation of the individual right to keep and bear arms under the Second Amendment. (175) The question presented in McDonald was whether the Second Amendment should be deemed "incorporated" against the states through the Fourteenth Amendment. (176)

Five Justices agreed that the Second Amendment should be incorporated against the states, but they divided as to the appropriate rationale for that decision. Justice Alito authored a plurality opinion, joined by three other Justices, which concluded that the Second Amendment was incorporated through the Fourteenth Amendment's Due Process Clause. (177) Justice Thomas, who provided the fifth vote in favor of incorporation, explicitly rejected the plurality's due process rationale and concluded instead that incorporation should be accomplished through the Fourteenth Amendment's Privileges or Immunities Clause. (178) Four dissenters would have rejected both of the asserted rationales for incorporation, finding the asserted Second Amendment rights not applicable to the states under either the Due Process Clause or the Privileges or Immunities Clause. (179)

Tabulating the votes cast by the Justices on each of the relevant issues in McDonald reveals the existence of a voting paradox. (180) Eight Justices--the four plurality Justices and the four dissenters--rejected the Privileges or Immunities Clause as a basis for incorporation, while five Justices--the four dissenters and Justice Thomas--would have refused to incorporate the Second Amendment under the Fourteenth Amendment's Due Process Clause. But despite the negative majority votes on each of these two asserted rationales for incorporation, a majority of the Court nonetheless concluded that incorporation was appropriate and, on that basis, held the challenged ordinance invalid. (181)

In theory, these features of the McDonald decision would seem to present a challenging task for lower courts seeking guidance from the Court's decision. Neither Justice Alito's plurality opinion nor Justice Thomas's concurrence reflected a logically entailed subset of the other. The plurality's due process rationale was arguably broader in that it would extend protection to all "persons"--not just citizens--while the alternative Privileges or Immunities Clause rationale favored by Justice Thomas might open the door to protecting additional rights beyond those previously recognized under the Court's incorporation jurisprudence. (182) The incomplete overlap between the judgment-supportive rationales would thus seem to render unworkable the implicit consensus approach to discerning the narrowest grounds of the Court's decision. The existence of a revealed voting paradox in the case likewise renders the fifth vote approach to Marks similarly unworkable. (183) And while the issue-by-issue approach might, in principle, provide some guidance to the lower courts, that guidance points to a set of legal conclusions that conflict with the Court's specific judgment--namely, that neither the Due Process Clause nor the Privileges or Immunities Clause supports incorporation.

But despite these seeming theoretical difficulties, lower courts have not struggled to extract guidance from the decision. To the contrary, lower courts have given little outward sign of even recognizing McDonald as a case calling for analysis under the Marks framework. (184) What accounts for the striking dearth of lower court confusion over McDonald while other, similarly "paradoxical" decisions have so vexed those courts?

The most plausible explanation derives from the substantial domain of conceptual overlap between the due process rationale endorsed by the plurality and the Privileges or Immunities Clause rationale endorsed by Justice Thomas. As noted above, the degree of overlap between the two rationales is not complete. But the areas of nonoverlap exist only at the margins--in cases involving either noncitizens or rights claims not previously endorsed under the Court's incorporation doctrine--leaving in place a substantial core of agreement among the judgment-supportive rationales.

McDonald is thus an example of a plurality decision that is both shallow and wide. Although Justice Alito's plurality opinion and Justice Thomas's sole concurrence reflect important points of theoretical divergence, they will converge on the same practical result in the vast majority of future cases to which McDonald might conceivably apply. In particular, in any future case involving a claim of Second Amendment rights (or other previously incorporated rights) asserted by a U.S. citizen, the two rationales will point to the same result, leaving lower courts free to apply the decision without having to choose between the two. (185)

McDonald is not unique in this regard. Consider another well-known Supreme Court decision involving a voting paradox--National Mutual Insurance Co. v. Tidewater Transfer Co. (186) Tidewater involved the constitutionality of a Statute conferring diversity jurisdiction on federal district courts over disputes between residents of the District of Columbia and residents of a state. (187) Five Justices concluded that the statute was constitutional but divided between two separate rationales. Three Justices concluded that the statute exceeded the scope of the diversity jurisdiction contemplated by Article III, which makes no mention of the District of Columbia, but nonetheless held that the statute was a permissible exercise of Congress's Article I power to legislate for the District. (188) The two remaining Justices who voted to uphold the statute specifically rejected this Article I rationale but nonetheless concluded that the District could be considered a "State" for purposes of the Article III Diversity Clause. (189) The four dissenting Justices rejected both of these rationales and would have found the statute unconstitutional. (190) Thus, as in McDonald, the Court awarded a judgment based on an outcome-level vote that was inconsistent with the way a majority of the Justices had voted on each of the two alternative legal theories that were necessary to that judgment. (191)

The scope of agreement between the two judgment-supportive opinions in Tidewater is somewhat narrower than the agreement between the Justices in McDonald. The novel Article I rationale embraced by the three-Justice Tidewater plurality might open the courts to a multitude of purely state law claims that would be barred by the more conventional understanding of Article III embraced by the remaining Justices in that case (including the two who joined in the judgment on other grounds). (192) But though many commentators have despaired of extracting any meaningful precedential guidance from the deeply fractured opinions in Tidewater, (193) lower courts have found surprisingly little practical difficulty in applying Tidewater to the actual cases that have come before them.

In a case decided shortly after Tidewater, the Ninth Circuit easily concluded that the Supreme Court's decision supported upholding the same federal diversity statute with respect to cases involving residents of Hawaii, which was then still a federal territory. (194) The Ninth Circuit determined that "[t]he reasons assigned by the two groups of Justices who concurred in the result" in Tidewater both applied with equal force to residents of the federal territories as they did to District of Columbia residents. (195) The court thus concluded that jurisdiction was appropriate. (196) The Seventh Circuit in a later case reached the same conclusion with regard to residents of Puerto Rico, another territory. (197)

Other lower courts have used a similar method--that is, avoiding the need to choose between competing opinions from the plurality case by determining that each of the judgment-supportive rationales would point to the same result in the particular case before them--to determine the binding effect of other seemingly confounding Supreme Court decisions. (198)

3. Connecting results to reasons

As discussed above, the traditional conception of the holding-dicta distinction conceives of "holdings" as limited to the judgment of the precedent-setting court and those portions of the court's opinion that were necessary to that judgment. (199) Thus, in a paradigmatic Supreme Court opinion--with majority agreement on both outcome and rationale--the "holding" or ratio should be susceptible to distillation into a statement of the following sort: "We hold for the prevailing party because the governing rule is Rule Xand this case is an instance in which Rule X demands such judgment." A later court bound by that holding would thus be bound to follow not only the specific result in the case (that is, the judgment in favor of the prevailing party) but also the reason for that judgment (that is, each of the propositions following the word "because" in the preceding sentence).

The absence of majority agreement on a judgment-supportive rationale in cases resulting in a plurality decision precludes such a straightforward distillation of a controlling ratio. But the fact that no single opinion, by itself, suffices to explain the Court's judgment does not mean that lower courts are therefore incapable of answering the question why the Court chose to award the judgment it did. Rather, in answering that question, lower courts must look to the set of opinions in the precedent case that were collectively necessary for the Court to reach its judgment. Such a survey of opinions will reveal multiple alternative rationales that necessarily diverge from one another in some respects. But such judgment-supportive rationales will also, by necessity, converge with one another in some respects, at least insofar as was necessary to determine the specific judgment in the precedent case.

A lower court is not, therefore, left in the same position it would have been in had the Supreme Court simply issued its judgment without explanation. In that circumstance, the lower court would be left largely unconstrained in selecting an imagined rationale on which the precedent Court might have based its ruling. (200) In a plurality decision, by contrast, the lower court has available to it information from which to determine the rationale--or, more accurately, the set of rationales--on which the precedent Court actually did base its ruling.

By looking to the convergent reasoning of the opinions that were collectively necessary to the precedent case judgment, a lower court seeking to understand why the winning party won and the losing party lost should usually be able to answer that question. For example, where the precedent case involves a plurality decision in which the Justices concurring in the judgment split their votes between two competing precedential rules--Rule X and Rule Y--the Court's reason for decision should be susceptible to a formulation of the following sort: "The prevailing party won because the governing rule was either Rule X or Rule Y and the precedent case was an instance in which both Rule X and Rule Y demanded such judgment." Thus, for example, in the McDonald case discussed above in Part II.B.2, the petitioner was able to prevail because his case happened to be one in which both the plurality's due process theory and Justice Thomas's competing privileges or immunities theory pointed to a judgment in his favor. (201)

Where a subsequent case falls within an area of overlap that would be resolved the same way under each of the alternative judgment-supportive rationales, the task for the precedent-following court is relatively straight-for ward. As the examples discussed above in Part II.B.2 show, where a subsequent case involves facts that would call for the same resolution regardless of which rationale is chosen, the lower court can avoid the need to choose between them.

Of course, this reconciliation strategy will not be available in every case. Sometimes, the choice between two competing rationales may be critical to determining how a later case should be resolved. At this point, the guidance offered by the plurality decision runs out. Because the judgment in the precedent case would have been equally valid if either of the rationales asserted by the concurring Justices were correct, the traditional common law notion of the ratio decidendi supplies no criteria by which to determine which of the competing rationales the lower court should accept.

But even in this category of cases, the plurality decision would continue to exert some meaningful precedential constraint on the lower courts. Though the lower court judge would be largely unconstrained in choosing between Rule Xand Rule Y, she would not be free to rationalize her decision in a manner that would produce results that neither Rule X nor Rule 7 could support. In other words, in rationalizing her decisions, the lower court judge must account for the domain of shared agreement on results defined by the respective rationales that were necessary to the precedent case's judgment. (202)

Thus, in a subsequent case where both Rule X and Rule Y would compel the lower court to reach the same result, the court would not be free to choose a rationale that would lead to a different result. Nor would the lower court be free to rationalize the approach in some other way that might preserve consistency with the specific result in the precedent case while ignoring the actual reasons that were provided for that decision. Thus, even in cases where the plurality precedent does not compel a specific result, the decision's precedential force can nonetheless narrow the "feasible choice set" available to lower courts and the resulting discretionary space within which their decisionmaking may proceed. (203)

The precedential obligation of lower courts under this approach is closely related to the often-invoked idea that such courts are bound by the "specific result" in the precedent case. (204) But for reasons that have already been discussed, the notion of a "specific result," in and of itself, lacks sufficient content to guide decisionmaking absent further specification of how broadly or narrowly the relevant "result" should be characterized. (205) Focusing on the domain of shared agreement between the judgment-supportive rationales responds to this concern by connecting the Court's judgment to the broader set of reasons that were collectively necessary to that judgment, thereby supplying precedent-following courts with criteria through which to assess the materiality of asserted factual similarities and differences between a later case and the earlier precedent-setting case.

III. The Case for the Shared Agreement Approach to Plurality Precedent

The foregoing Parts have described the three most prominent existing approaches to the Marks doctrine in the lower courts and laid out the basic framework of the alternative shared agreement approach. With a clear picture of these four alternatives in view, it is now possible to engage in a comparative assessment of which approach is best. This Part develops an affirmative case for the shared agreement approach as the appropriate conceptual framework for lower courts to use in addressing the problem of plurality precedent.

Part II1.A lays the groundwork for this argument by demonstrating the shared agreement approach's consistency with the Supreme Court's very limited existing case law discussing the narrowest grounds doctrine. Part III.B then highlights two traditional criteria of precedential legitimacy that the Supreme Court has emphasized in other contexts to limit its capacity to prescribe binding precedential rules: the requirement that precedential statements be supported by a majority of the Court's members and the requirement that such statements directly support the judgment that was rendered in the precedent-setting case. Part III.C demonstrates that only the shared agreement approach provides a solution to the problem of plurality precedent that conforms to these two requirements and provides a workable rule of decision for virtually every plurality decision that lower courts are likely to face. Finally, Part III.D responds to the concern that the shared agreement approach provides insufficient guidance to lower courts regarding the content of governing law. Though such guidance concerns are not wholly without merit, they do not provide a convincing argument for rejecting the shared agreement approach in favor of any of the alternative approaches to the narrowest grounds rule--at least not in the absence of clear and specific direction from the Supreme Court requiring that result.

A. Consistency with Marks

An important threshold question for any lower court considering how to approach the problem of plurality precedent is whether a proposed approach can be reconciled with the language and holding of the Marks decision itself. Even if a lower court concludes that the Marks Court adopted an unwise, or even unworkable, theory of plurality precedent, that view would not relieve the lower court of its obligation to follow the theory that the Court itself prescribed. (206) Any theory of plurality precedent that conflicts with the explicit language of Marks is thus practically unavailable to the judges of the lower courts--at least until the Supreme Court itself chooses to reconsider the precedential status of that decision.

Fortunately, the shared agreement approach to plurality precedent fits comfortably with the Court's decision in Marks. The Marks Court instructed lower courts to identify the "holding of the Court" by looking to the "position taken by those Qustices] who concurred in the judgments on the narrowest grounds." (207) Many lower courts have interpreted this language as requiring them to identify a single narrowest opinion from the precedent case and treat that opinion as fully binding. (208) But nothing in the Marks decision itself compels that interpretation. An alternative, and equally plausible, understanding of Marks is that precedential effect is reserved for the "reasoning within the concurring opinions that a majority of the concurring Justices support." (209) On this latter understanding, a purportedly "narrowest" opinion from a plurality case is binding on the lower courts only insofar as it reflects views to which a majority of the concurring Justices have logically committed themselves by virtue of their respective opinions in the precedent case.

Of course, in Marks itself the Court did single out a particular opinion--namely, Justice Brennan's Memoirs plurality--as the "controlling opinion" from that case. (210) But given the context in which that case arose, this conclusion was fully consistent with the shared agreement understanding of the narrowest grounds rule. Recall that the three judgment-supportive opinions in Memoirs aligned in a simple continuum from narrowest to broadest. (211) For the Marks defendants' due process claim to succeed, all they needed to establish was that the Memoirs decision had effected a change in the governing First Amendment standard that was at least as speech-protective as the standard articulated in the plurality opinion. (212) But the three other concurring Justices in Memoirs--Justices Black, Douglas, and Stewart--would necessarily have found the petitioners' conviction unconstitutional even if the prosecution had been able to satisfy Justice Brennan's standard. In the circumstances in which the Court considered the issue, therefore, treating Justice Brennan's opinion as "controlling" was fully consistent with requiring a majority consensus among the concurring Justices for binding precedential force to attach.

The Supreme Court's subsequent cases addressing the Marks framework are consistent with this conclusion. As noted above, the Court has addressed the narrowest grounds doctrine in only a handful of cases since Marks was handed down. (213) And though the Court has, in most of those cases, identified a particular opinion from the precedent case as controlling, each such decision is fully reconcilable with the majority-consensus understanding of Marks.

Consider, for example, the Court's recent opinion in Glossip v. Gross, (214) which concluded that the three-Justice plurality opinion in Baze v. Rees (215) reflected the narrowest, and therefore controlling, grounds of the Baze decision. (216) In Baze, the plurality had concluded that a challenged method of execution would be deemed cruel and unusual only if the challenger could prove the existence of a feasible alternative method that would "significantly reduce a substantial risk of severe pain." (217) Justices Scalia and Thomas concurred in that judgment but said they would only uphold a method-of-execution challenge in those cases where the challenged method had been "deliberately designed to inflict pain." (218) Granting the seemingly reasonable assumption that a method of execution "deliberately designed" to inflict unnecessary pain would also pose a "substantial risk" of inflicting such pain, the standard endorsed by the Baze plurality and the alternative standard endorsed by Justices Scalia and Thomas would necessarily point to the same result in any future case where the plurality would deem an execution method constitutionally permissible. (219)

The Court's decision in Panetti v. Quarterman (220) provides a further illustration. Panetti addressed the constitutional adequacy of procedures through which a prisoner could challenge his mental competence to face execution. (221) In a prior plurality decision, Ford v. Wainwright, a majority of the Justices had rejected one procedural regime for making that determination but adopted different standards in explaining why that procedure was deficient. (222) A four-Justice plurality opinion concluded that "the ascertainment of a prisoner's sanity as a predicate to lawful execution calls for no less stringent standards than those demanded in any other aspect of a capital proceeding." (223) Concurring in part and in the judgment, Justice Powell endorsed a less stringent procedural framework, requiring only that a prisoner contesting his competence be given a fair hearing that accorded with the requirements of "basic fairness." (224) The Panetti Court concluded that Justice Powell's opinion set "the minimum procedures a State must provide to a prisoner raising a Ford-based competency claim" and that, by failing to provide such procedures, the state defendant in that case had violated "clearly established" federal law. (225) But as in Glossip and Marks, the Panetti Court's decision to treat Justice Powell's concurrence as binding was fully consistent with the majority consensus understanding of the narrowest grounds rule. Because the plurality Justices would necessarily agree that a procedure failing Justice Powell's relaxed due process test would also fail their own preferred standard, the Court had no occasion to consider what the precedential effect of the decision would be in a case where the plurality and concurrence pointed to different results.

The remaining three cases in which a majority of the Court invoked the narrowest grounds rule to single out a particular opinion as "controlling" are all to similar effect. (226) In no case has the Court directly held that lower courts are obliged to treat as binding an opinion reflecting the views of a minority of Justices in those circumstances where majority consensus was absent. To the contrary, in the only two cases invoking Marks that might have required it to rule in this way, the Court instead concluded the narrowest grounds analysis was not "useful" and dispensed with any attempt to extract a binding precedential rule from the original decision. (227) In other words, each of the Supreme Court's prior decisions singling out a particular opinion as controlling has involved a nested rationale scenario where the identification of a single "narrowest" opinion is relatively straightforward. (228)

Thus, rather than standing in tension with the Marks Court's command, the shared agreement approach fits comfortably with both the language and specific holding of that case as well as with the Court's other post-Marks decisions engaging the narrowest grounds inquiry. But the same might arguably be said of other ways of understanding the Court's various statements regarding the precedential significance of plurality decisions. (229) It will therefore be useful to consider which of the various approaches to the Marks rule best accords with broadly accepted conventions of precedential legitimacy that apply outside the plurality context.

B. Majoritarianism and Judgment-Supportiveness as Criteria of Precedential Legitimacy

In the domain of federal law, the Supreme Court has long claimed for itself the ultimate authority to "say what the law is." (230) Because the stare decisis effect of its rulings binds lower court judges absolutely, (231) a decision of the Supreme Court has the practical effect of "making" law that is every bit as authoritative for a lower court judge as a statute or constitutional provision. (232) But knowing that the Supreme Court possesses institutional authority to "make" law in this way tells us relatively little about how such law is to be made. The Supreme Court, "like Congress, is a 'they,' not an 'it'" (233)--a multimember institution whose expression of collective intent depends on a system of technical rules for aggregating the views of its nine individual members. (234) Understanding the precedential effect of the Court's pronouncements--particularly in the plurality decision context--thus requires a basic understanding of the background rules and norms through which the Court exercises its law-declaring authority.

Unfortunately, the precise content of these background rules is surprisingly difficult to pin down. (235) There is no judicial equivalent to the constitutionally prescribed lawmaking procedures established for Congress in Article I, Section 7. (236) Instead, the mechanisms through which the Supreme Court exercises its power to establish binding precedent are governed by a set of largely unwritten conventions that have developed over time. (237) The unwritten nature of such conventions complicates efforts to discern their content, as does the traditional reluctance of common law courts to acknowledge their own lawmaking function. (238) There are, however, at least two broadly accepted conventions of Supreme Court decisionmaking that are highly relevant to assessing the precedential significance of the Court's pronouncements: the Court's commitment to majority decisionmaking and its limitation of binding precedential force to propositions that directly support its judgments in adjudicated cases.

Though nothing in the Constitution explicitly requires the Supreme Court to decide cases by majority vote, the commitment to majority decisionmaking is among the most deeply rooted features of the Court's institutional practices. (239) This commitment to majority decisionmaking means that precedential status is reserved for opinions that have received the assent of at least a majority of the Court's members. While pronouncements by individual Justices or groups of Justices comprising less than a majority may be looked to for persuasive value, they do not bind either the Supreme Court itself in subsequent cases or the judges of the lower courts. (240)

The requirement that precedential effect be limited to propositions that support the Court's judgments finds its roots in two of the Court's longstanding institutional practices--namely, its prohibition on advisory opinions and its adherence to the traditional common law distinction between holdings and dicta. The Court's prohibition on advisory opinions is almost as old as the Court itself. (241) The Court has grounded its refusal to issue such opinions in the text of Article III. (242) And though arguably not an "inevitable" reading of that provision's language, (243) the prohibition on advisory opinions is among the most deeply rooted of the Court's institutional practices. (244)

Only slightly less deeply rooted is the Supreme Court's institutional commitment to the common law distinction between holdings and dicta. As discussed above, the holding-dicta distinction presupposes a distinction between statements that lead the Court to its judgment and those that do not, reserving binding precedential force for the former and treating the latter as merely persuasive. (245) In practice, the significance of this distinction is somewhat muted by the notorious blurriness of the line separating holdings from dicta (246) as well as the willingness of many lower courts to follow even acknowledged Supreme Court dicta. (247) But the Court has repeatedly emphasized the distinction as a meaningful constraint on its own authority to establish binding precedent--both for itself (248) and for lower courts. (249) Indeed, some version of the distinction seems essential to the practical efficacy of the Court's ban on issuing advisory opinions. Because at least some litigated cases in virtually every area of law are bound to be brought before the Court in any given year, a Court that could issue binding pronouncements on matters nonessential to resolving the parties' dispute would enjoy the same practical authority as a Court explicitly empowered to issue advisory opinions. (250) The holding-dicta distinction can thus be seen as an important corollary of the limited decisional authority that the Court has long understood Article III to require. (251)

C. Implications for Plurality Precedent

The Court's institutional commitment to majority decisionmaking and its refusal to extend binding effect to statements that do not directly support its judgments suggest a particular conception of precedential authority. Under this conception, the power to establish binding precedent resides with the majority of Justices whose votes were collectively necessary to the Court's judgment. (252) The Court's written opinions, which explain its judgments, of course generate stare decisis effects that bind the judges of the lower courts. But "[a] judgment is no less a judgment, and no less final, if it is unaccompanied by a statement of reasons." (253) The Court has instructed that even its unexplained dispositions create binding precedential obligations for lower court judges. (254) Moreover, the Court has emphasized time and again that its authority to declare binding law for other legal actors, including the judges of the lower courts, is merely a byproduct of its primary responsibility of issuing judgments in contested cases. (255)

In most cases, this conception of precedent-setting authority is not particularly controversial. Since the time of Chief Justice Marshall, the Court has explained the large majority of its decisions through an official "opinion of the Court" that both declares the majority's judgment in the case and provides a set of reasons that each Justice in the majority assents to by joining the opinion. (256) But absent a clear statement by the majority Justices regarding the rationale or rationales they mutually relied upon in reaching their collective judgment, lower courts must fall back on some default rule for determining the precedential effect of the Court's ruling. Marks itself can be seen as an initial attempt by the Court to establish such a default rule. But for the reasons discussed above, Marks is fairly read to prescribe, at most, a narrow rule for dealing with a particular type of plurality decision involving logically "nested" rationales. (257)

But the fact that Marks itself did not prescribe a clear default rule for determining the precedential significance of the Court's fractured majority decisions does not mean that no such default rule exists. The presumption that the Supreme Court--like most multimember institutions--only possesses authority to act through a majority vote of its members underlies a longstanding default rule for dealing with cases where a majority decision cannot be reached. Where the Court divides equally on the proper resolution of a particular case, the long-settled default rule requires the affirmance of the judgment below but does not establish any new binding precedential rule for future cases. (258) "Because a tie vote demonstrates that no majority favors a particular disposition, the resulting disposition is the one that leaves the preexisting legal landscape intact." (259) Thus, under ordinary circumstances, the Supreme Court's inability to decide on the resolution of a particular case is practically equivalent to a decision to not create binding precedent at all. (260)

Of course, in cases that result in a plurality decision, there is a majority decision for action by the Court, and the Marks Court was thus correct to intuit that such decisions should produce at least some change in background legal rules. But the necessary majority agreement in such cases is limited to agreement on the appropriate judgment, not the broader judgment-supportive rationale.

Of the existing approaches to Marks reflected in the lower courts' jurisprudence, only the implicit consensus approach conforms to this particular conception of precedential authority. This approach insists on both a genuine agreement among a numerical majority of the Justices and requires that the particular majority whose votes are considered "concurred in the judgment." (261) But for reasons already discussed, this approach is capable of providing useful guidance to lower courts only with respect to a limited subset of plurality decisions where the concurring Justices' opinions happen to align with one another in a particular way. (262) The shared agreement approach, by contrast, provides a framework for determining the precedential significance of decisions that do not conform to the "nested" rationale paradigm that the implicit consensus approach envisions. By demonstrating how the concurring majority's shared agreement on the judgment can generate precedential consequences even in the absence of a comprehensive, mutually agreed-upon majority rationale, the shared agreement approach allows lower courts to extract meaningful precedential guidance from virtually all plurality decisions, regardless of how deeply fractured the Court's decision might be.

By contrast, both the fifth vote approach and the issue-by-issue approach transfer the power to establish precedent away from the judgment-supportive majority and toward some other faction of the Court. The fifth vote approach effectively places the power to establish precedent in the hands of the median Justice, presumably on the theory that her opinion reflects the position that a majority of the Justices would most likely have settled on had they been "forced to choose" a single rationale. (263) But plurality decisions are only possible because the Justices are not "forced to choose" in this way and have chosen not to choose a single opinion as the authoritative position of the Court. (264) The deciding majority, of course, has the capacity to converge on a single majority rationale, and history furnishes examples of Justices voting contrary to their sincerely held substantive positions in order to produce a majority opinion for the Court. (265) But acknowledging that the Justices could converge on the median Justice's position if they were forced to do so does not justify the conclusion that lower courts must act as if the majority actually did so. (266)

The issue-by-issue approach likewise transfers power away from the deciding majority by allowing a differently constituted majority--consisting of the dissenters and some subset of the concurring Justices--to establish precedent on discrete propositions of law. Unlike the fifth vote approach, the issue-by-issue approach at least ensures the existence of a majority agreement on the propositions to be treated as binding. But it does not insist on agreement among the particular majority whose votes were collectively necessary to the Court's judgment. This distinction is significant because the specification of a majority decision rule is necessarily ambiguous and incomplete unless one is prepared to answer the equally important question: "A majority of whom?" (267)

The implicit answer suggested by the issue-by-issue approach is that the relevant "majority" consists of a simple majority of all Justices participating in the precedent-setting case, irrespective of how those Justices cast their votes. (268) But this approach conflicts with the requirement that precedential statements must also be offered in support of the Court's judgment in the precedent case. Majority-supported dicta are not considered binding on the judges of the lower courts merely because they have been assented to by a majority. To the contrary, it is only with respect to majority-supported statements that the distinction between holdings and dicta can do any meaningful work.269 And like dicta, statements in dissenting opinions are neither "necessary to" nor even supportive of the judgment in the precedent-setting case.270 The issue-by-issue approach thus stands in sharp tension with the traditional notion that binding precedential force attaches only to those statements that lead the precedent-setting court to its judgment.

As noted above, the decision procedures employed by the Supreme Court are largely governed by informal custom and convention. Nothing in the text of the Constitution obviously and unambiguously requires the Supreme Court to adhere to a majority decision procedure or establishes the judgment-supportive criterion as a limit on the Court's precedential authority. Were a majority of the Court to clearly and unambiguously specify an alternative decision rule for addressing the problem of plurality precedent, there might well be a plausible argument that lower courts would be bound to follow that rule in their own decisions. (271)

But in the absence of such clear specification, the ordinary default rules for assessing the precedential significance of the Court's decisions apply. Those default rules both require majority agreement for action by the Supreme Court and limit the precedential effect of the Court's pronouncements to those statements offered in support of its judgments. Failure to satisfy either of these conditions leaves the preexisting state of the law unchanged. The shared agreement approach accords with this default rule by framing a theory of precedential obligation that maps directly onto the deciding majority's actual shared agreement regarding why the precedent case's judgment was correct.

D. But What About Guidance?

A persistent criticism of plurality decisions is that such decisions "create confusion" in the law "by failing to provide clear guidance to the lower courts." (272) One prominent way of understanding the Marks Court's instruction is as an effort to respond to such guidance concerns by allowing lower courts to identify a single precedential opinion in as many cases as possible. (273) And while this is hardly the only--or even the most plausible--way to read the Marks Court's instruction, (274) it is certainly understandable that some might strive to interpret the Court's directive to maximize precedential guidance and constraint.

Both the fifth vote approach and the issue-by-issue approach respond to this guidance concern by attempting to single out a particular rationale from the precedent case as controlling. Under the shared agreement approach, by contrast, the judges of the inferior courts will--at least in most cases--possess a limited degree of discretion in choosing between two or more judgment-supportive rationales. (275) The shared agreement approach's acknowledgment of a limited domain of discretion for lower court judges might thus raise concerns among those who highly value the determinacy and uniformity of federal law. But such uniformity concerns do not provide a particularly convincing argument against the shared agreement approach for several reasons.

First, as a purely descriptive matter, the Supreme Court--and the federal judicial system more broadly--places less of a value on national uniformity than is often assumed. Many of the Court's institutional practices--including the limited size of its discretionary docket; its toleration of persistent, unresolved circuit splits; and its frequent practice of issuing narrow, fact-specific majority opinions that fail to clearly guide the lower courts--have been criticized as contributing to the disuniformity of national law. (276) Other institutional features of the federal judicial system--such as the geographical division of the federal courts of appeals and the absence of intercircuit stare decisis--suggest a broader, systemic acceptance of disuniformity in federal law. (277)

Second, and relatedly, if the Supreme Court were inclined to maximize its precedential guidance, there are far clearer and more effective ways for it to achieve that objective than through Marks' s cryptic narrowest grounds doctrine. Perhaps most obviously, the Court could simply refrain from issuing plurality decisions completely. (278) Though the types of substantive jurisprudential divisions that drive such fractured decisions are likely impossible to eradicate entirely, a Court that placed a sufficiently high value on clarity and guidance might develop an informal norm requiring some Justices to switch their votes in order to produce a majority-supporting opinion. (279)

Alternatively, the Court could itself engage in the precedential analysis and provide "a brief per curiam notice at the beginning of [its] decision" instructing lower courts "which opinion is controlling." (280) Of course, this latter approach would require the Justices themselves to agree upon some decision procedure for selecting the authoritative opinion. The Justices might, for example, adopt the fifth vote interpretation as their official position. (281) But they might instead choose to simply follow the concurring opinion that received the most votes (282) or the opinion reflecting the views of the Court's most senior Justice. (283) They might even conceivably choose a controlling opinion by drawing lots. (284) Any of these methods would respond to the concern regarding precedential guidance. The fact that the Court has instead chosen to continue issuing plurality decisions in the absence of a clear rule for determining their precedential effect suggests the Court's willingness to entrust lower court judges with a significant degree of practical discretion in extracting precedential guidance from such decisions.

Third, the nature of lower court discretion contemplated by the shared agreement approach is not meaningfully different from the nature of the discretion lower courts already exercise in myriad decisionmaking contexts. Consider, for example, a Supreme Court majority opinion that resolves a case on narrow and highly fact-specific grounds. (285) Such "minimalist" decisions inevitably bestow upon lower courts a substantial degree of discretion in determining whether and how they should apply as precedents in a future case. (286) Such discretion is, of course, cabined by the particular language and holding of the Court's narrow decision, as well as by other relevant legal materials, such as the broader universe of Supreme Court case law, potentially relevant statutes and regulations, and the lower court's own earlier precedents. (287) But even after all those sources have been taken into account, lower courts will often possess some (limited) capacity to choose between two or more plausible understandings of governing law. (288) In other words, the judge is "not bound to decide the question one way rather than another" because there simply is "no wrong answer to the questions posed" or "at least, there is no officially wrong answer" that is clearly discernible at the time the judge must make her decision. (289)

Likewise, under the shared agreement approach, a lower court judge would not be left wholly unconstrained in choosing which of the judgment-supportive opinions from the precedential decision to follow. Such a judge presumably would not, for example, feel herself free to select one opinion over others based on the flip of a coin or the alphabetical order of the opinion authors' names. (290) Rather, she would likely feel herself compelled to offer an explanation for why she believed one of the opinions reflected a better view of the law or fit more comfortably within the surrounding universe of precedent and other governing legal sources. (291) In explaining why the understanding of the law reflected in one of the articulated rationales is best, a lower court judge employing the shared agreement approach can thus be expected to act in much the same way as she would in other contexts where the legal system affords the judge discretion to reach more than one plausible result. And, as Pauline Kim observes, "when discretion in this sense exists, it 'is quintessentially associated with variability of result.'" (292)

Fourth, it is far from clear that binding precedential guidance from the Supreme Court necessarily represents the type of unqualified good that many critics of plurality decisions seem to imagine. While precedential guidance can undoubtedly offer important societal benefits, these benefits do not come without cost. (293) Among the most obvious potentially undesirable consequences of precedent is that later decisionmakers may be constrained to reach substantively incorrect or undesirable results. (294) A practice of following precedent thus raises both the importance of identifying the "right" legal principle and the consequences of legal error because the precedent--setting court's decision will control not only the present dispute but also some unknowable number of future disputes.

Recognizing this danger of locking in potentially undesirable legal rules, the Justices have occasionally acknowledged that disuniformity at the lower court level may sometimes allow for desirable "percolation." (295) On this view, the Supreme Court's own decisionmaking is thought to benefit from a temporary period of deliberation and experimentation in the lower courts. Such percolation may, for example, allow the Justices to see how lower courts address the relevant questions in a range of different factual settings (296) and how the rules formulated by different circuit courts have worked in practice over time. (297) The Justices' decisionmaking may also benefit from the substantive "deliberation that occurs among lower courts as they respond to one another's decisions." (298)

Such percolation might be viewed as particularly useful in the plurality decision context. By their nature, plurality decisions tend to involve legal issues as to which the governing legal standards are unclear and elite legal opinion is closely divided. The failure of a majority to coalesce on any single rationale is a powerful signal of the unsettled state of the governing doctrine. It is precisely in this context that arguments for a continued period of percolation in the lower courts seem most persuasive. (299) By allowing lower courts to continue working out for themselves the difficult legal issues that divided the Justices in the original plurality decision, the Supreme Court may gain important insights and information that can help crystallize a consensus regarding the appropriate rule of law when the Court ultimately decides to revisit the issue.

The shared agreement approach facilitates the type of deliberation and experimentation that the percolation theory envisions. By encouraging (indeed, requiring) lower court judges to choose among the various judgment-supportive rationales based on their own best understanding of background legal principles, the shared agreement approach focuses lower courts' attention on the types of questions that are likely to be foremost in the Justices' minds if and when they decide to revisit the underlying issue. The fifth vote approach and the issue-by-issue approach, by contrast, each direct lower courts' attention away from the substantive merits and toward essentially arbitrary tests for selecting a particular opinion as controlling--tests that the Supreme Court itself has shown remarkably little interest in following or clarifying. (300) To the extent percolation provides any meaningful practical benefits to the Supreme Court's decisionmaking, the shared agreement approach thus holds out the greatest promise of allowing the judicial system to harness those benefits.

Finally, even if one concludes that a more guidance-enhancing plurality precedent rule is desirable, an important institutional design question would remain regarding how best to implement that rule. For example, even if scholars and judges agreed that the judicial system as a whole would be better off if all lower court judges were to converge on the fifth vote approach, it would not necessarily follow that any particular lower court judge should use that approach in her own decisionmaking. (301) If the chief benefit sought to be gained by adopting such an approach is national uniformity, that benefit will only be achieved if all lower court judges, or at least a substantial majority of them, converge on the same approach. But no individual judge acting on his or her own can ensure such national convergence.

It is, of course, conceivable that the lower courts might converge on a particular approach to the narrowest grounds rule organically. But four decades of experience with Marks suggests that such convergence is unlikely. In practice, the narrowest grounds rule has been at least as prone to generating circuit splits as it has to avoiding or resolving them. (302) This experience suggests that a "bottom-up" approach to clarifying the meta-doctrine governing the interpretation of plurality decisions is likely to meet with little success. If the goal is to enhance the clarity and determinacy of resulting precedential rules, a "top-down" approach of precedential guidance delivered directly from the Supreme Court seems far more likely to achieve its objectives.

In short, concerns about national uniformity and clear guidance to lower courts do not provide a convincing reason for lower courts to reject the shared agreement approach to plurality precedent. The shared agreement approach delivers clear precedential guidance to lower court judges regarding certain results they cannot reach and certain results they must reach. And though the approach does contemplate a limited domain of bounded discretion on the part of lower court judges in choosing between two or more judgment-supportive rationales, the nature of lower court discretion thus contemplated is not meaningfully different from that exercised by such courts in other contexts. Moreover, even if one concludes that the disuniformity costs of the shared agreement approach would be too great, both the institutional responsibility and the most plausible chance of practical success resides with the Supreme Court itself, not the judges of the lower courts.

IV. Implementing the Shared Agreement Approach

The foregoing Parts have focused on building the theoretical case for the shared agreement approach as the appropriate conceptual framework for determining the binding precedential force of Supreme Court plurality decisions. This Part demonstrates how the approach would work in practice by reference to a recent Supreme Court plurality decision--Shady Grove Orthopedic Associates v. Allstate Insurance Co. (303) Shady Grove provides a useful case study, as it involves both a fairly straightforward alignment of opinions and a single relevant question that the Court was asked to decide. The case also illustrates the nature of the discretion conferred on lower courts by the shared agreement approach and the role of other authoritative legal sources in defining the parameters of such discretion.

Shady Grove involved the question whether Federal Rule of Civil Procedure 23, which specifies the conditions under which class actions "may be maintained" in federal court, allowed for certification of a class action asserting State law claims that would not have been maintainable as a class action in the state's own courts. (304) This question, in turn, implicated a set of questions regarding the relationship between the federal courts and state substantive and procedural law that are generally classed together under the rubric of the "Erie doctrine." (305) To address such federal-state conflicts, the Court's earlier Erie doctrine precedents had established a framework that required the Court to first determine whether Rule 23 "answers the question in dispute" and, if so, to then determine whether that Rule "exceed[ed]" the scope of the rulemaking authority delegated by Congress through the Rules Enabling Act. (306) But the Justices in Shady Grove divided into three competing factions regarding the proper application of this framework to the particular case before them.

The plurality opinion, authored by Justice Scalia and joined in full by three others, answered the first question in the affirmative, concluding that Rule 23 spoke directly to the question whether a class action asserting the state law claims at issue was maintainable in federal court. (307) Given this conclusion, the agreed-upon framework called for looking exclusively to Rule 23 to determine whether a class could be certified unless that Rule was invalid. Justice Scalia's plurality opinion had little difficulty disposing of this latter point, concluding that, under prior case law--particularly the Court's 1941 decision in Sibbach v. Wilson & Co. (308)--a federal rule should be treated as per se valid so long as it "really regulat[es] procedure." (309)

In a sole opinion concurring in part and concurring in the judgment, Justice Stevens agreed with the plurality's analysis on the first step of the twostep framework and its conclusion regarding the validity and applicability of Rule 23 given the facts of the particular case. (310) But unlike the plurality, Justice

Stevens believed that the validity analysis required by the second step of the governing framework could not end with a conclusion that the federal rule "really regulated" procedure. Because the Rules Enabling Act explicitly provides that the federal rules "shall not abridge, enlarge or modify any substantive right," (311) Justice Stevens concluded that even an overtly procedural federal rule would not govern a particular case in which the rule would displace a state law that is so intertwined with a substantive state right or remedy that it functions to define the scope of the state-created right. (312) In effect, Justice Stevens advocated an approach under which the validity of particular federal procedural rules would be assessed on an "as-applied" basis, with the as-applied validity depending on the particular rule's potential impact on particular state-created substantive rights. (313)

Justice Ginsburg's dissent, joined by the three remaining Justices, disagreed with both the plurality and Justice Stevens regarding the proper application of the first step of the two-part governing framework--that is, the determination whether Rule 23 answers the question in dispute. (314) Unlike both Justice Stevens and the plurality, the dissenters saw no "inevitable collision" between the explicit terms of Rule 23 and the state law policy prohibiting class certification for certain categories of claims. (315) Based on this conclusion, Justice Ginsburg and her fellow dissenters declined to address the step-two validity question that divided Justice Stevens and the plurality. Instead, the dissent applied a separate test used to determine whether federal or state law should control where there is no direct conflict with any federal rule and concluded that this analysis supported adhering to the New York restriction on class certification. (316)

Shady Grove thus reflected a clear 4-1-4 split with Justice Stevens seeming to occupy the median point between the two more extreme factions: Justice Scalia's plurality opinion, on the one hand, and Justice Ginsburg's dissent, on the other. (317) Most lower courts that have considered the issue have concluded that Justice Stevens's concurrence reflects the controlling "narrowest grounds" opinion under Marks--a conclusion consistent with the fifth vote approach to the narrowest grounds rule. (318) But this conclusion highlights the legitimacy concerns inherent in the fifth vote approach. The invalidity analysis in Justice Stevens's opinion was not endorsed by any other member of the Court. (319)

The legitimacy concerns inherent in the fifth vote approach are exacerbated in Shady Grove by the role that preexisting precedent played in the Justices' analysis. The disagreement between the plurality Justices and Justice Stevens was itself driven by conflicting understandings of the Court's prior Erie doctrine case law, particularly Sibbach. The plurality Justices understood Sibbach as compelling their adoption of their wholesale approach to assessing the validity of a federal rule. (320) Indeed, Justice Scalia's opinion went so far as to accuse Justice Stevens of seeking to "overrule" Sibbach "or, what is the same, to rewrite it." (321) If the plurality's understanding of Sibbach were correct, then the practical effect of recognizing Justice Stevens's sole concurrence as the controlling opinion in Shady Grove would be to empower one Justice to overrule (or at least substantially narrow) a controlling Supreme Court precedent that was itself assented to by a majority of the Court.

Under the shared agreement approach, by contrast, no lower court would have been bound to follow either Justice Stevens's analysis or the plurality's view. Rather, each of those rationales would be part of the feasible choice set from which lower courts could choose without violating their precedential obligation. But the lower courts' discretion in this regard would not be wholly unfettered. As noted above, the disagreement that divided Justice Stevens from the plurality was itself driven in large part by conflicting understandings of what preexisting law--particularly Sibbach--required. A lower court forced to choose between the plurality's wholesale approach to the rule validity analysis and the alternative as-applied approach endorsed by Justice Stevens would thus be forced to confront this same question regarding the proper understanding of Sibbach. (322)

Troublingly, however, those lower courts that have identified Justice Stevens's Shady Grove opinion as controlling for Marks purposes have largely done so without giving any serious consideration to what Sibbach or the Court's other Erie doctrine precedents had to say about the choice between the wholesale approach and the as-applied approach to the rule validity analysis. (323) Instead, those courts have tended to assume that Justice Stevens's opinion controls regardless of that opinion's consistency with the Court's earlier Erie decisions. (324)

The lower courts' failure to give serious consideration to these questions is doubly unfortunate, as they implicate a set of pragmatic concerns that lower courts might be particularly well suited to decide. Lurking just below the surface of the interpretive disagreements that divided Justice Stevens from the plurality Justices was a set of more overtly pragmatic concerns regarding the likely practical impact of allowing as-applied validity challenges to particular federal rules. Justice Scalia's opinion for the plurality Justices expressed concern about the difficulty that lower courts would encounter were they required to apply Justice Stevens's proposed validity standard to the potentially "hundreds of state rules" to which that standard might conceivably apply. (325) Justice Stevens, on the other hand, denied that his approach would unduly burden the lower courts and contended that his approach would be no more burdensome than the alternative wholesale validity analysis endorsed by the plurality. (326) But neither faction of Justices offered any empirical or even anecdotal support for their conflicting intuitions.

The lower courts stand in a much better position to assess the respective burdens of the two approaches. As the frontline responders tasked with implementing the Court's procedural doctrines, lower courts--and particularly district courts--would bring to the task a wealth of hands-on experience that most members of the Court presently lack. (327) If lower courts were to engage with this question directly--without any preconceived notion of being bound to follow one or the other of the Shady Grove opinions--their reactions might provide the Court with much clearer and more concrete evidence of the actual administrative burdens involved in the respective approaches.

Unfortunately, the strength of any such signal that might be perceived from the lower courts' actual reactions to Shady Grove is significantly muted by the fact that those courts have, for the most part, not understood themselves as enjoying the freedom to engage with such questions directly. Instead, most courts have focused on answering a different set of questions regarding the proper application of the Marks narrowest grounds rule. This myopic focus on the proper application of Ma rks drains their post-Shady Grove decisions of most of the practical epistemic value that they might otherwise have contributed to the Supreme Court's future decisionmaking. (328)

Conclusion

Trying to understand the precedential effect of Supreme Court plurality decisions is a task that has confounded the lower courts for decades. The Court's decision in Marks, though designed to clarify the precedential significance of such decisions, has instead exacerbated the confusion. But once plurality decisions are recognized and appreciated for what they are--namely, incompletely theorized judicial agreements on results unsupported by more comprehensive majority agreement on the governing rationale--much of this confusion melts away. Because only the result in the precedent case enjoys the support of the deciding majority, the result provides the natural focus of the lower courts' precedential inquiry.

Of course, a mere result in a single case provides minimal practical guidance to lower courts. But by looking to the majority's partially overlapping and partially diverging reasons for that judgment, lower courts can identify a domain of shared agreement among the majority Justices regarding why that result was correct. Looking to this domain of shared agreement enables lower courts to identify a universe of results in future cases that they are required to reach as well as a corresponding universe of results they are forbidden from reaching. Understanding the domain of shared agreement can also help lower courts to identify the boundaries of their discretion to continue working through for themselves the challenging legal questions that divided the Court in the precedent case. As such, the shared agreement approach strikes the appropriate balance between precedential constraint and continued doctrinal growth and development.

(1.) See, e.g., Anthony Lewis, In Memoriam, William J. Brennan, Jr., 111 HARV. L.REV. 29, 32 (1997) ("Justice Brennan used to joke that a critical talent for a Supreme Court Justice was the ability to count to five."); Abner Mikva, The Scope of Equal Protection, 2002 U. Chi. LEGAL F. 1, 8 ("[A]s the late Justice Brennan used to say, the first rule of the Supreme Court is that you have to be able to count to five.").

(2.) See James F. Spriggs II & David R. Stras, Explaining Plurality Decisions, 99 GEO. L.J. 515, 517 (2011) ("Plurality decisions occur when a majority of Justices agree upon the result or judgment in a case but fail to agree upon a single rationale in support of the judgment.").

(3.) See Evan H. Caminker, Why Must Inferior Courts Obey Superior Court Precedents?, 46 STAN. L.REV. 817, 818 (1994) ("[Longstanding doctrine dictates that a court is always bound to follow a precedent established by a court 'superior' to it.").

(4.) 430 U.S. 188(1977).

(5.) Id. at 193 (quoting Gregg v. Georgia, 428 U.S. 153,169 n.15 (1976) (plurality opinion)).

(6.) See, e.g., United States v. Dixon, 687 F.3d 356, 359 (7th Cir. 2012) (holding that "[e]ven though eight Justices disagreed with Justice Sotomayor's approach [in Freeman v. United States, 564 U.S. 522 (2011),] and believed it would produce arbitrary and unworkable results, her reasoning" was nonetheless controlling under Marks (citation omitted)); Smith v. Univ. of Wash. Law Sch., 233 F.3d 1188, 1198, 1200 (9th Cir. 2000) (concluding that Justice Powell's opinion controlled in Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265 (1978), even though "none of the other Justices fully agreed with Justice Powell's opinion").

(7.) See, e.g., United States v. Epps, 707 F.3d 337, 348 (D.C. Cir. 2013) (concluding that the "narrowest opinion" under Marks "must represent a common denominator of the Court's reasoning" and "must embody a position implicitly approved by at least five Justices who support the judgment" (emphasis omitted) (quoting King v. Palmer, 950 F.2d 771, 781 (D.C. Cir. 1991) (en banc))).

(8.) See, e.g., United States v. Martino, 664 F.2d 860, 872 (2d Cir. 1981) (observing that the Supreme Court has not "elaborated on what was meant by 'narrowest grounds'"); cf. Lisk v. Lumber One Wood Preserving, LLC, 792 F.3d 1331, 1337 (11th Cir. 2015) ("For some issues, asking which of two opinions is narrower is akin to asking, 'Which is taller, left or right?'").

(9.) Compare, e.g., United States v. Donovan, 661 F.3d 174, 182-83 (3d Cir. 2011) (interpreting Marks and subsequent Supreme Court opinions to require that lower courts "examine the dissenting Justices' views to see if there is common ground" among a majority as to a rationale), with, e.g., King, 950 F.2d at 783 ("[W]e do not think we are free to combine a dissent with a concurrence to form a Marks majority.").

(10.) SeeSpriggs & Stras, supra note 2, at 519 (reporting that the Supreme Court issued only 45 plurality decisions between 1801 and 1955 but issued 195 plurality decisions between 1953 and 2006).

(11.) See, e.g., Pamela C. Corley, Uncertain Precedent Circuit Court Responses to Supreme Court Plurality Opinions, 37 AM. POL. RES. 30, 32 (2009) ("[P]lurality decisions are important to study because they tend to occur in highly salient issue areas such as civil liberties and civil rights." (citation omitted)); Spriggs & Stras, supra note 2, at 527 ("[P]lurality decisions tend to occur in difficult and highly salient cases....").

(12.) See Planned Parenthood of Se. Pa. v. Casey, 505 U.S. 833 (1992); Webster v. Reprod. Health Ser vs., 492 U.S. 490 (1989).

(13.) See McDonald v. City of Chicago, 561 U.S. 742 (2010) (concerning incorporation of the Second Amendment against state governments).

(14.) See, e.g., Crawford v. Marion Cty. Election Bd., 553 U.S. 181 (2008) (challenging Indiana's voter identification law).

(15.) See, e.g., City of Richmond v. J.A. Croson Co., 488 U.S. 469 (1989) (challenging the constitutionality of racial preferences in public contracting); Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265 (1978) (challenging racial preferences in higher education).

(16.) See, e.g., Baze v. Rees, 553 U.S. 35 (2008) (addressing the permissible methods of capital punishment); Gregg v. Georgia, 428 U.S. 153 (1976) (reaffirming capital punishment's constitutionality).

(17.) See Nat'l Fed'n of Indep. Bus. v. Sebelius, 132 S. Ct. 2566 (2012) (challenging Congress's authority to require that certain individuals either acquire health insurance or pay a penalty).

(18.) See, e.g., United States v. Duron-Caldera, 737 F.3d 988, 994 & n.4 (5th Cir. 2013) (noting the difficulty of applying the narrowest grounds rule to discern the Supreme Court's holding in Williams v. Illinois, 132 S. Ct. 2221 (2012), regarding whether particular statements prepared in the course of an investigation were "testimonial" for purposes of the Sixth Amendment's Confrontation Clause); United States v. James, 712 F.3d 79, 95-96 (2d Cir. 2013) (noting the same difficulty).

(19.) See, e.g., In re Sealed Case, 722 F.3d 361, 365 (D.C. Cir. 2013) (noting the divergence of lower court opinion regarding proper interpretation of the federal sentencing guidelines resulting from differing understandings of the Supreme Court's plurality decision in Freeman v. United States, 564 U.S. 522 (2011)).

(20.) See, e.g., Ainsworth v. Moffett Eng'g, Ltd., 716 F.3d 174, 178 (5th Cir. 2013) (applying Marks analysis to the Supreme Court's plurality decision in J. Mclntyre Machinery, Ltd. v. Nicastro, 564 U.S. 873 (2011)); AFTG-TG, LLC v. Nuvoton Tech. Corp., 689 F.3d 1358, 1363 (Fed. Cir. 2012) (per curiam) (engaging in the same inquiry).

(21.) See, e.g., In re Packaged Ice Antitrust Litig., 779 F. Supp. 2d 642, 659-60 (E.D. Mich. 2011) (applying Marks to determine the holding of the Supreme Court's plurality decision in Shady Grove Orthopedic Assocs. v. Allstate Ins. Co., 559 U.S. 393 (2010)); McKinney v. Bayer Corp., 744 F. Supp. 2d 733, 746-47 (N.D. Ohio 2010) (engaging in the same analysis).

(22.) See, e.g., Kemp v. Medtronic, Inc., 231 F.3d 216, 218, 224 & n.l (6th Cir. 2000) (applying Marks to determine the precedential effect of the Supreme Court's plurality decision in Medtronic, Inc. v. Lohr, 518 U.S. 470 (1996)); In re Medtronic, Inc. Sprint Fidelis Leads Prods. Liab. Litig., 592 F. Supp. 2d 1147, 1151 n.4 (D. Minn. 2009) (parsing the same opinion).

(23.) See James Hardisty, Reflections on Stare Decisis, 55 IND. L.J. 41, 52-57 (1979) (discussing the relationship between "rule" stare decisis and "result" stare decisis); cf. Caminker, supra note 3, at 865-66 (discussing justifications for lower court obedience to Supreme Court precedent more generally).

(24.) Ratio Decidendi, Black's LAW DICTIONARY (9th ed. 2009).

(25.) See, e.g., Lawrence B. Solum, How NFIB v. Sebelius Affects the Constitutional Gestalt, 91 WASH. U. L. REV. 1, 22 (2013) (describing ratio decidendi as the idea that "the holding of a case is the rule that is logically implied by the stated reasons necessary to the resolution of the case on the facts before the appellate court and the legal arguments presented by the parties"); see also, e.g., JOSEPH RAZ, THE AUTHORITY OF LAW: ESSAYS ON LAW AND MORALITY 184 (2d ed. 2009) ("Essentially the ratio is the reason(s) by which the court justifies its decision.").

(26.) See, e.g., CASS R. SUNSTEIN, ONE CASE AT A TIME: JUDICIAL MINIMALISM ON THE SUPREME COURT 11-14 (1999) (describing various kinds of "incompletely theorized agreements," including agreements about "concrete particulars" reached in the face of "disagreements or uncertainty about the basis for those concrete particulars" (emphasis omitted)).

(27.) 559 U.S. 393 (2010).

(28.) 430 U.S. 188(1977).

(29.) Id. at 189-91.

(30.) 354 U.S. 476 (1957).

(31.) Id. at 489.

(32.) 383 U.S. 413 (1966).

(33.) Marks, 430 U.S. at 193; see also Memoirs, 383 U.S. at 421 (Black & Stewart, JJ., concurring in the judgment) (noting that Justice Black "concur[s] in the reversal for the reasons [he] stated in" Ginzburg v. United States, 383 U.S. 463, 476 (1966) (Black, J., dissenting); and Mishkin v. New York, 383 U.S. 502, 515 (1966) (Black, J., dissenting)); id. at 428 (Douglas, J., concurring); cf. Ginzburg, 383 U.S. at 476 (Black, J., dissenting) ("[T]he Federal Government is without any power ... to put any type of burden on speech and expression of ideas of any kind [including obscenity].").

(34.) Marks, 430 U.S. at 193; see also Memoirs, 383 U.S. at 421 (Black & Stewart, JJ., concurring in the judgment) (noting that Justice Stewart "concur[s] in the reversal for the reasons [he] stated" in Ginzburg, 383 U.S. at 497 (Stewart, J., dissenting); and Mishkin, 383 U.S. at 518 (Stewart, J., dissenting)).

(35.) Memoirs, 383 U.S. at 418 (plurality opinion).

(36.) Id.

(37.) See id. at 441 (Clark, J., dissenting); id. at 456 (Harlan, J., dissenting); id. at 461 (White, J., dissenting).

(38.) 413 U.S. 15, 27-29 (1973).

(39.) Marks, 430 U.S. at 189-90.

(40.) Id. at 190-91.

(41.) United States v. Marks, 520 F.2d 913,919-20 (6th Cir. 1975), rev'd, 430 U.S. 188 (1977).

(42.) Marks, 430 U.S. at 193 (quoting Gregg v. Georgia, 428 U.S. 153,169 n.15 (1976) (plurality opinion)); see also Gregg, 428 U.S. at 169 n.15 (plurality opinion) ("Since five Justices wrote separately in support of the judgments in [Furman v. Georgia, 408 U.S. 238 (1972)], the holding of the Court may be viewed as that position taken by those Members who concurred in the judgments on the narrowest grounds....").

(43.) Marks, 410 U.S. at 193-94.

(44.) Prior to Marks, several lower courts had embraced this understanding of plurality decisions' precedential effect. See, e.g., Wiesenfeld v. Sec'y of Health, Educ. & Welfare, 367 F. Supp. 981, 988 (D.N.J. 1973); Mark Alan Thurmon, Note, When the Court Divides: Reconsidering the Precedential Value of Supreme Court Plurality Decisions, 42 DUKE L.J. 419, 420 (1992). Some state courts still adhere to this view regarding the precedential significance of their own plurality decisions. See, e.g., Rowland v. Washtenaw Cty. Rd. Comm'n, 731 N.W.2d 41, 47 n.7 (Mich. 2007) ("[Decisions in which no majority of the justices participating agree with regard to the reasoning are not an authoritative interpretation under the doctrine of stare decisis.").

(45.) Marks, 430 U.S. at 193 (quoting Gregg, 428 U.S. at 169 n.15).

(46.) See, e.g., John P. Neuenkirchen, Plurality Decisions, Implicit Consensuses, and the Fifth-Vote Rule Under Marks v. United States, 19 WIDENER L.REV. 387, 408 (2013) (contrasting the "implicit consensus," or common denominator, approach with the "fifth vote" approach); Michael L. Eber, Comment, When the Dissent Creates the Law: Cross-Cutting Majorities and the Prediction Model of Precedent, 58 EMORY L.J. 207, 239-41 (2008) (contrasting the common denominator, or "command model," approach with the issue-by-issue, or "prediction model," approach).

(47.) See Justin F. Marceau, Lifting the Haze of Baze: Lethal Injection, the Eighth Amendment, and Plurality Opinions, 41 Ariz. St. L.J. 159, 170 n.49 (2009) (noting the tendency of some lower courts to treat mutually inconsistent understandings of the Marks narrowest grounds rule "as if they are but two prongs of the same analysis").

(48.) See, e.g., United States v. Davis, 825 F.3d 1014, 1021 (9th Cir. 2016) (en banc) ("Our cases interpreting Marks have not been a model of clarity.").

(49.) Compare, e.g., Green v. Haskell Cty. Bd. of Comm'rs, 568 F.3d 784, 807 n.17 (10th Cir. 2009) ("Given that [Van Orden v. Perry, 545 U.S. 677 (2005),] was decided by a plurality, the separate opinion of Justice Breyer, who supplied the 'decisive fifth vote,' is controlling under the rule of Marks." (citations omitted) (quoting Heideman v. S. Salt Lake City, 348 F.3d 1182, 1198 (10th Cir. 2003))), with, e.g., United States v. Carrizales-Toledo, 454 F.3d 1142, 1151 (10th Cir. 2006) ("In practice, ... the Marks rule produces a determinate holding 'only when one opinion is a logical subset of other, broader opinions.'" (quoting King v. Palmer, 950 F.2d 771, 781 (D.C. Cir. 1991) (en banc))). Some circuit courts have even commented on this internal inconsistency. See, e.g., Lisk v. Lumber One Wood Preserving, LLC, 792 F.3d 1331, 1337 (11th Cir. 2015) ("[W]e apparently have taken as many as three different approaches [to the narrowest grounds rule]--or we at least have articulated our approach three different ways--when confronting other fragmented Supreme Court decisions.").

(50.) See, e.g., Garland v. Roy, 615 F.3d 391, 402-03 (5th Cir. 2010) (identifying a four-way circuit split regarding application of Marks to United States v. Santos, 553 U.S. 507 (2008), and rejecting all four in favor of a fifth distinct approach); Berkolow, Much Ado About Pluralities: Pride and Precedent Amidst the Cacophony of Concurrences, and Re-Percolation After Rapanos, 15 Va. J. SOC. POL'Y & L. 299, 334-44 (2008) (describing a circuit split regarding the proper application of Marks to the opinions in Rapanos v. United States, 547 U.S. 715 (2006)); B. Andrew Bednark, Note, Preferential Treatment The Varying Constitutionality of Private Scholarship Preferences at Public Universities, 85 Minn. L. REV. 1391, 1398-99 (2001) (identifying a three-court split as to the controlling opinion in Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265 (1978)).

(51.) King, 950 F.2d at 781; see also, e.g., United States v. Alean Aluminum Corp., 315 F.3d 179, 189 (2d Cir. 2003).

(52.) United States v. Robison, 521 F.3d 1319, 1323 (11th Cir. 2008) (Wilson, J., dissenting from the denial of rehearing en banc); see also King, 950 F.2d at 781 ("Marks is workable ... only when one opinion is a logical subset of other, broader opinions.").

(53.) Lewis A. Kornhauser & Lawrence G. Sager, The One and the Many: Adjudication in Collegial Courts, 81 CALIF. L. REV. 1,46 (1993).

(54.) See supra notes 32-38 and accompanying text (describing the various opinions in Memoirs).

(55.) See supra note 33 and accompanying text (discussing Justice Black's concurring opinion).

(56.) See supra notes 34, 36 and accompanying text (discussing Justice Stewart's concurring opinion and Justice Brennan's plurality opinion). Though the precise scope of Justice Stewart's "hardcore pornography" standard is somewhat murky, most commentators have viewed Stewart's proposed test as somewhat more speech-protective than the plurality's standard. See, e.g., Grutter v. Bollinger, 288 F.3d 732, 739-40 (6th Cir. 2002) (en banc) (characterizing the Brennan plurality as providing "the most limited First Amendment protection" of the three concurring opinions in Memoirs), affd, 539 U.S. 306 (2003); Maxwell L. Stearns, The Case for Including Marks v. United States in the Canon of Constitutional Law, 17 CONST. COMMENT. 321, 326-27 (2000) (assuming that Justice Stewart's standard would protect a broader class of sexually explicit material than the plurality opinion's).

(57.) See, e.g., Thurmon, supra note 44, at 428-32 (describing the "implicit consensus" rationale for the Marks rule).

(58.) In practice, there may be substantial practical measurement problems in comparing the scope of incompletely overlapping rationales to determine which rationale is "narrower" than the others. See United States v. Duvall, 740 F.3d 604, 619 (D.C. Cir. 2013) (Williams, J., concurring in the denial of rehearing en banc).

(59.) Cf. id. at 618 ("Without [the] requirement that one [rationale] be a subset of the other, the idea of 'narrowness' is inherently confusing and in fact indeterminate.").

(60.) 547 U.S. 715 (2006).

(61.) Id. at 720-22 (plurality opinion).

(62.) Id. at 723 (quoting 33 U.S.C. [section] 1344(a), (d)).

(63.) 33 U.S.C. [section] 1362(7) (2015).

(64.) Rapanos, 547 U.S. at 729 (plurality opinion).

(65.) Id. at 720-21 (citation omitted) (quoting 33 U.S.C. [section] 1362(7)).

(66.) Id. at 739.

(67.) Id. at 742.

(68.) Id. at 757.

(69.) Id. at 781 (Kennedy, J., concurring in the judgment).

(70.) Id. at 779-80.

(71.) Id. at 787.

(72.) Id. at 787-88 (Stevens, J., dissenting).

(73.) Id. at 810.

(74.) See, e.g., United States v. Gerke Excavating, Inc., 464 F.3d 723, 724 (7th Cir. 2006) (per curiam) (concluding that Justice Kennedy's rule would do less to "rein[] in federal authority"); cf. Rapanos, 547 U.S. at 810 n.14 (Stevens, J., dissenting) ("I assume that Justice Kennedy's approach will be controlling in most cases because it treats more of the Nation's waters as within the Corps' jurisdiction...." (formatting altered)).

(75.) See, e.g., United States v.Johnson, 467 F.3d 56, 64 (1st Cir. 2006) (acknowledging that the plurality's test might support jurisdiction in some cases where Justice Kennedy's test would not and vice versa); Gerke, 464 F.3d at 724-25 (per curiam) (coming to the same conclusion).

(76.) See Gerke, 464 F.3d at 724-25 (per curiam) (acknowledging this possibility); cf. Rapanos, 547 U.S. at 776-77 (Kennedy, J., concurring in the judgment) ("[B]y saying the Act covers wetlands (however remote) possessing a surface-water connection with a continuously flowing stream (however small), the plurality's reading would permit applications of the statute as far from traditional federal authority as are the waters it deems beyond the statute's reach.").

(77.) See supra notes 58-59 and accompanying text (noting the inability of the "implicit consensus" approach to identify a controlling opinion in cases involving partially overlapping rationales).

(78.) See, e.g., Anker Energy Corp. v. Consolidation Coal Co., 177 F.3d 161, 170 (3d Cir. 1999) ("[I]n cases where approaches differ, no particular standard is binding on an inferior court because none has received the support of a majority of the Supreme Court."); United States v. Robles-Sandoval, 637 F.2d 692, 693 n.l (9th Cir. 1981) ("The court in [Baldasar v. Illinois, 446 U.S. 222 (1980),] divided in such a way that no rule can be said to have resulted.").

(79.) See, e.g., Lair v. Bullock, 697 F.3d 1200, 1205 (9th Cir. 2012) ("If there is no such narrow opinion [reflecting a common denominator of the deciding court's reasoning], 'the only binding aspect of a splintered decision is its specific result."' (quoting United States v. Rodriguez-Preciado, 399 F.3d 1118,1140 (9th Cir. 2005))).

(80.) 947 F.2d 682 (3d Cir. 1991), affd in part, rev'd in part, and remanded, 505 U.S. 833 (1992).

(81.) See id. at 687-88 (identifying the "threshold question" as "whether the standard of review of abortion regulations promulgated by the [Supreme] Court" in earlier decisions had "survived" two more recent plurality decisions).

(82.) Id. at 693.

(83.) Id. at 694 n.7 (emphasis omitted). The language quoted in the text addressed what the Casey court described as a "slightly more complex" circumstance in which "six or more Justices join in the judgment and they issue three or more opinions." Id. But the court advanced a similar methodology for determining the controlling opinion in cases involving a more conventional split where only two opinions agree on the judgment. See id. at 693-94 ("In a constitutional case where (l) there is a 5-4 decision or where there are only two opinions in the majority and (2) the majority votes to uphold a law as constitutional, the 'narrowest grounds' principle will identify as authoritative the standard articulated by a Justice or Justices that would uphold the fewest laws as constitutional.").

(84.) Where a case is heard by fewer than nine Justices, it is conceivable that the median vote necessary to produce a judgment could be provided by the fourth vote rather than the fifth. Cf. 28 U.S.C. [section] 1 (2015) (providing that six Justices are sufficient to establish a quorum).

(85.) See Casey, 947 F.2d at 694 (emphasizing that the identified opinion "is as authoritative for lower courts as a nine-Justice opinion," even if it reflects the views of only one Justice); see also, e.g., United States v. Duvall, 740 F.3d 604, 610 (D.C. Cir. 2013) (Kavanaugh, J., concurring in the denial of rehearing en banc) ("In most cases, the commonsense way to apply Marks is to identify and follow the opinion that occupies the middle ground between (i) the broader opinion supporting the judgment and (ii) the dissenting opinion."); Annex Books, Inc. v. City of Indianapolis, 581 F.3d 460, 465 (7th Cir. 2009) ("Because the other Justices [in City of Los Angeles v. Alameda Books, Inc., 535 U.S. 425 (2002),] divided 4 to 4, and Justice Kennedy was in the middle, his views establish the holding.").

(86.) See, e.g., Frank B. Cross, The Justices of Strategy, 48 Duke L.J. 511,548 (1998) (book review) (contending that a "rational lower court" attempting to use a prior Supreme Court plurality decision to predict the Court's future decisions "will simply find the position of the fifth Justice and treat this as the law").

(87.) See, e.g., United States v. Gerke Excavating, Inc., 464 F.3d 723, 724 (7th Cir. 2006) (per curiam) (interpreting Marks to require "lower-court judges ... to follow the narrowest ground to which a majority of the Justices would have assented if forced to choose").

(88.) See generally MAXWELL L. STEARNS, CONSTITUTIONAL PROCESS: A SOCIAL CHOICE ANALYSIS OF SUPREME COURT DECISION MAKING 45, 124-41 (2000) (arguing that a fifth vote approach to the Afarfesrule--that is, one that accords controlling significance to the views of the Court's median Justice--will identify a Condorcet winner in most plurality decisions).

(89.) Id. at 128 tbl.3.5.

(90.) Id. at 128.

(91.) Id. at 129.

(92.) See, e.g., King v. Palmer, 950 F.2d 771,782 (D.C. Cir. 1991) (en banc) ("When eight of nine Justices do not subscribe to a given approach to a legal question, it surely cannot be proper to endow that approach with controlling force, no matter how persuasive it may be."); Dague v. City of Burlington, 935 F.2d 1343, 1360 (2d Cir. 1991) (noting "the anomaly of the views of one justice, with whom no one concurs, being the law of the land, where the Court is so divided on an issue and where there is no majority opinion at all").

(93.) See Neuenkirchen, supra note 46, at 404; see also, e.g., United States v. Duvall, 740 F.3d 604, 617 n.8 (D.C. Cir. 2013) (Kavanaugh, J., concurring in the denial of rehearing en banc) (emphasizing the importance of looking to all opinions in the precedent case, including dissents, in order to identify the controlling fifth vote opinion).

(94.) See, e.g., David G. Post & Steven C. Salop, Issues and Outcomes, Guidance, and Indeterminacy: A Reply to Professor John Rogers and Others, 49 Vand. l. rev. 1069, 1072 (1996) (arguing that because "judges need only declare their views on the issues that are dispositive for their own individual reasoning," neither the public nor the judges themselves may know "whether or not the paradoxical situation pertains").

(95.) Michael I. Meyerson, The Irrational Supreme Court, 84 NEB. L. REV. 895,901 (2006).

(96.) David Post & Steven C. Salop, Rowing Against the Tidewater: A Theory of Voting by Multijudge Panels, 80 GEO. L.J. 743, 750 (1992).

(97.) Two such voting paradox cases--McDonald v. City of Chicago, 561 U.S. 742 (2010), and National Mutual Insurance Co. v. Tidewater Transfer Co., 337 U.S. 582 (1949)--are discussed in detail in Part II below. See infra notes 173-85 (discussing McDonald), 186-98 (discussing Tidewater).

(98.) David S. Cohen, The Precedent-Based Voting Paradox, 90 B.U. L. REV. 183,185 (2010).

(99.) Consider, for example, Eastern Enterprises v. Apfel, which involved a corporation's constitutional challenge to a federal law requiring it to fund lifetime health benefits for certain of its former employees. 524 U.S. 498, 514, 517 (1998) (plurality opinion). A four-Justice plurality opinion authored by Justice O'Connor concluded that the law was an unconstitutional taking of private property, id. at 537, but expressed skepticism regarding the company's claim that the law also violated its substantive due process rights, id. at 537-38. Justice Kennedy, writing only for himself, concluded that the company's Takings Clause claim should be rejected but would have held the law's application to the company a violation of substantive due process. Id. at 539 (Kennedy, J., concurring in the judgment and dissenting in part). Finally, the four dissenting Justices would have rejected both the company's Takings Clause claim as well as its substantive due process claim. Id. at 554, 558-59 (Breyer, J., dissenting). Combining the votes from the various opinions in the case reveals a majority against each of the company's two constitutional theories--takings and substantive due process--but the Court nonetheless concluded that the law was unconstitutional. See Cohen, supra note 98, at 188-91 (using Eastern Enterprises to illustrate the phenomenon of a multiple-issue voting paradox). For a more detailed description of the necessary and sufficient conditions for a voting paradox to occur, see Maxwell L. Stearns, Should Justices Ever Switch Votes?: Miller v. Albright in Social Choice Perspective, 7 SUP. CT. ECON. REV. 87, 12327 (1999).

(100.) Eber, supra note 46, at 210; see also, e.g., United States v. Donovan, 661 F.3d 174, 182 (3d Cir. 2011) ("[W]e have looked to the votes of dissenting Justices if they, combined with votes from plurality or concurring opinions, establish a majority view on the relevant issue.").

(101.) See, e.g., Unity Real Estate Co. v. Hudson, 178 F.3d 649, 659 (3d Cir. 1999) (concluding that lower courts are "bound to follow the five-four vote against the takings claim" in Eastern Enterprises where four of those five votes were provided by the dissenters); cf. supra note 99 (describing the Eastern Enterprises opinions).

(102.) See Eber, supra note 46, at 223-25.

(103.) See supra notes 66-71 and accompanying text.

(104.) Rapanos v. United States, 547 U.S. 715,810 (2006) (Stevens, J., dissenting).

(105.) Id. (emphasis omitted).

(106.) See Eber, supra note 46, at 209 (observing that Justice Stevens's dissent "creatfed] at least an eight- or five-vote majority in favor of federal CWA jurisdiction" in all instances covered by either test).

(107.) See, e.g., United States v. Donovan, 661 F.3d 174, 183-84 (3d Cir. 2011) (adopting Justice Stevens's proposed approach); United States v. Bailey, 571 F.3d 791, 799 (8th Cir. 2009) (adopting this same approach); United States v. Johnson, 467 F.3d 56, 64-66 (1st Cir. 2006) (adopting this same approach).

(108.) See, e.g., Jonathan H. Adler, Reckoning with Rapanos: Revisiting "Waters of the United States" and the Limits of Federal Wetland Regulation, 14 MO. ENVTL. L. & POL'Y REV. 1, 14 (2006) (contending that "[njothing in the dissent" in a plurality decision "constitutes a portion of the judgment of the Court, so nothing in the dissent" can be "part of the actual holding of the case" under Marks).

(109.) Marks v. United States, 430 U.S. 188, 193 (1977) (emphasis added) (quoting Gregg v. Georgia, 428 U.S. 153, 169 n.15 (1976) (plurality opinion)); see also, e.g., United States v. Robison, 505 F.3d 1208, 1221 (11th Cir. 2007) ("In our view, Marks does not direct lower courts interpreting fractured Supreme Court decisions to consider the positions of those who dissented."); King v. Palmer, 950 F.2d 771, 783 (D.C. Cir. 1991) (en banc) ("[W]e do not think we are free to combine a dissent with a concurrence to form a Marks majority."). But see Johnson, 467 F.3d at 65 (arguing that Marks does not preclude consideration of dissenters' views).

(110.) See infra notes 146-51 (describing the necessity test for precedential significance).

(111.) See, e.g., In re Aggrenox Antitrust Litig., No. 3:14-md-2516 (SRU), 2016 WL 4204478, at *6 (D. Conn. Aug. 9, 2016) ("[T]he common denominator of a concurrence and a dissent does not support the judgment. It is, in effect, Marks-doctrine dicta rather than Marksdoctrine holding."); see also infra note 270 (collecting additional sources expressing a similar view).

(112.) Linda Novak, Note, The Precedential Value of Supreme Court Plurality Decisions, 80 COLUM. L. Rev. 756, 767-69 (1980) (describing such "dual majority" cases).

(113.) See supra notes 93-94 and accompanying text (describing the hypothetical).

(114.) See, e.g., William h. Rehnquist, Remarks on the Process of Judging, 49 WASH. & LEE L. REV. 263, 270 (1992) ("There must be an effort to get an opinion for at least a majority of the Court in every case where that is possible, in order that lower court judges and the profession as a whole may know what the law is without having to go through an elaborate head-counting process."); Judge Ruth Bader Ginsburg, Remarks on Writing Separately, Jurisprudential Lecture at the University of Washington School of Law (May 11, 1989), in 65 WASH. L. REV. 133, 148 (1990) (describing the "proliferation" of decisions without a clear majority as "unsettling"); Justice Lewis F. Powell, Jr., Stare Decisis and Judicial Restraint, Leslie h. Arps Lecture at the Association of the Bar of the City of New York (Oct. 17, 1989), in 47 WASH. & LEE L. REV. 281, 289 (1990) ("Splintered decisions provide insufficient guidance for lower courts ... [and] promote disrespect for the Court as a whole .... ").

(115.) E.g., Nichols v. United States, 511 U.S. 738, 745-46 (1994) (noting difficulties lower courts had encountered in seeking to apply Marks to the Court's fractured opinion in Baldasar v. Illinois, 446 U.S. 222 (1980), and concluding that "[w]e think it not useful to pursue the Marks inquiry to the utmost logical possibility when it has so obviously baffled and divided the lower courts which have considered it"); accord Grutter v. Bollinger, 539 U.S. 306, 325 (2003) (reaching a similar conclusion regarding lower courts' interpretation of Regents of the Univ. of Cal. v. Bakke, 438 U.S. 265 (1978)).

(116.) See, e.g., Justin Marceau, Plurality Decisions: Upward-Flowing Precedent and Acoustic Separation, 45 CONN. L. Rev. 933,975 n.183 (2013) ("My review of cases applying Marks suggests that the Court does not intervene to prevent lower courts from fumbling around with plurality precedent."). For example, notwithstanding significant disagreement among lower courts regarding the precedential effect of its decision in Rapanos v. United States, 547 U.S. 715 (2006), the Supreme Court has "declined to clarify the issue by granting certiorari," Michael Farrell, A Wetland Functions Approach to Applying the Opinions from the Rapanos Decision, 19 U. DENV. WATER L. REV. 61, 62 (2015).

(117.) Glossip v. Gross, 135 S. Ct. 2726, 2738 n.2 (2015); Panetti v. Quarterman, 551 U.S. 930, 949 (2007); Grutter, 539 U.S. at 325; O'Dell v. Netherland, 521 U.S. 151, 160 (1997); Romano v. Oklahoma, 512 U.S. 1, 9 (1994); Nichols, 511 U.S. at 745; City of Lakewood v. Plain Dealer Publ'g Co., 486 U.S. 750, 764 n.9 (1988).

(118.) Grutier, 539 U.S. at 325 (quoting Nichols, 511 U.S. at 745-46); Nichols, 511 U.S. at 745-46.

(119.) Glossip, 135 S. Ct. at 2793 (Sotomayor, J., dissenting); Panetti, 551 U.S. at 969 n.5 (Thomas, J., dissenting); Romano, 512 U.S. at 22-23 (Ginsburg, J., dissenting).

(120.) See generally Richard M. Re, Narrowing Precedent in the Supreme Court, 114 COLUM. L. REV. 1861, 1865 (2014) (noting the frequency with which the Court "narrow[s]" its own prior precedents without overtly overruling them).

(121.) See Marceau, supra note 116, at 966 (observing that the Supreme Court itself often ignores Marks in dealing with its own prior plurality decisions).

(122.) See Grutter, 539 U.S. at 325 ("It does not seem 'useful to pursue the Marks inquiry to the utmost logical possibility when it has so obviously baffled and divided the lower courts that have considered it.'" (quoting Nichols, 511 U.S. at 745-46)).

(123.) See supra notes 117-19 (discussing the handful of cases in which the Court specifically invoked the narrowest grounds doctrine).

(124.) See Marceau, supra note 116, at 987 (noting the Supreme Court's "seeming indifference to its own Marks rule" and contending that the Court itself does not view the rule as, "in any ordinary sense, a binding precedent"); cf. State v. Kikuta, 253 P.3d 639, 658 n.14 (Haw. 2011) (observing that the Marks doctrine "has been applied very rarely and inconsistently by the Supreme Court").

(125.) See City of Lakewood v. Plain Dealer Publ'g Co., 486 U.S. 750, 764 n.9 (1988) (questioning the dissenting opinion's reliance on combined views of dissenting and concurring Justices in an earlier plurality decision as inconsistent with Marks). Compare, e.g., Panetti v. Quarterman, 551 U.S. 930, 949 (2007) (identifying Justice Powell's concurring opinion in Ford v. Wainwright, 477 U.S. 399 (1986), as the narrowest grounds for the Court's decision under Marks), with, e.g., id. at 969 n.5 (Thomas, J., dissenting) (questioning the majority's application of Marks and contending that "it is difficult to say that Justice Powell's opinion" in Ford was "merely a narrower version of the plurality's view").

(126.) See, e.g., Marceau, supra note 116, at 986-87 (dismissing the Marks narrowest grounds rule as "a pretend precedent ... that has very little practical effect on the way that plurality precedent ought to be discerned by lower courts"); Adam S. Hochschild, Note, The Modern Problem of Supreme Court Plurality Decision: Interpretation in Historical Perspective, 4 WASH. U. J.L. & POL'Y 261, 286 (2000) ("No 'test'--recommended by the Supreme Court or otherwise--can dispel the confusion surrounding these plurality decisions.").

(127.) See, e.g., Thurmon, supra note 44, at 447 (characterizing the author's proposed revision to the Marks rule as "a form of damage control").

(128.) See, e.g., Solum, supra note 25, at 22 (describing the ratio decidendi conception of vertical stare decisis); Peter M. Tiersma, The Textualization of Precedent, 82 notre Dame L. Rev. 1187, 1247 (2007) (observing that in the common law tradition, "the precedential value of a case lay in its outcome, or judgment").

(129.) See infra note 131 (collecting sources).

(130.) Michael Abramowicz & Maxwell Stearns, Defining Dicta, 57 Stan. L. Rev. 953, 1045 (2005).

(131.) See, e.g., EDWARD H. LEVI, An Introduction to Legal Reasoning 2 (1949) ("[The common law judge] is not bound by the statement of the rule of law made by the prior judge even in the controlling case. The statement is mere dictum .... It is not what the prior judge intended that is of any importance; rather it is what the present judge, attempting to see the law as a fairly consistent whole, thinks should be the determining classification."); Arthur L. Goodhart, Determining the Ratio Decidendi of a Case, 40 Yale L.J. 161, 162 (1930) ("[T]he reason which the judge gives for his decision is never the binding part of the precedent."); Max Radin, Case Law and Stare Decisis: Concerning Prajudizienrecht in Amerika, 33 COLUM. L. REV. 199, 210 (1933) ("[I]t is the decision itself which must be followed and not the opinion ....").

(132.) See, e.g., MELVIN ARON EISENBERG, THE NATURE OF THE COMMON LAW 52-55 (1988) (criticizing the facts-plus-outcome approach to precedent); Abramowicz & Stearns, supra note 130, at 1045-46 (arguing that the facts-plus-outcome approach "simply fails to provide a positive explanation of judicial practice"); Larry Alexander, Constrained by Precedent, 63 S. CAL. L. REV. 1, 34-44 (1989) (criticizing the facts-plus-outcome approach and concluding that it "faces either indeterminacy or incoherence"); Michael C. Dorf, Dicta and Article III, 142 U. PA. L. REV. 1997, 2009-40 (1994) (showcasing the limits of the facts-plus-outcome approach with the example of cases concerning the President's removal power).

(133.) Cf. EUGENE WAMBAUGH, THE STUDY OF CASES: A COURSE OF INSTRUCTION IN READING AND STATING REPORTED CASES, COMPOSING HEAD-NOTES AND BRIEFS, CRITICISING AND COMPARING AUTHORITIES, AND COMPILING DIGESTS 14 (Boston, Little, Brown & Co. 2d ed. 1894) (acknowledging the inevitability of factual dissimilarities between cases but observing that "the obvious suggestion" to this difficulty "is that the differences may be immaterial").

(134.) Henry Paul Monaghan, Stare Decisis and Constitutional Adjudication, 88 colum. L. Rev. 723, 764 (1988); see also, e.g., Abramowicz & Stearns, supra note 130, at 1055 (observing that "facts, material or otherwise, do not speak for themselves" and that a "satisfactory definition of holding and dicta must therefore examine the reasoning that connects the material facts to the result").

(135.) See generally Dorf, supra note 132, at 2009-24 (surveying the manner in which the Supreme Court has applied this method to distinguish earlier precedents addressing the scope of the President's removal authority).

(136.) See, e.g., Seminole Tribe v. Florida, 517 U.S. 44, 67 (1996) ("When an opinion issues for the Court, it is not only the result but also those portions of the opinion necessary to that result by which we are bound."); Dorf, supra note 132, at 2037 ("[W]hen they are not busy circumventing precedent by abusing the holding/dictum distinction, judges typically pay a great deal of attention to the words as well as the results of judicial decisions.").

(137.) Cf., e.g., Planned Parenthood of Se. Pa. v. Casey, 947 F.2d 682, 692 n.6 (3d Cir. 1991) ("The lower courts in constitutional matters universally follow both the Supreme Court's choice of legal standard and the specific results the Court has reached by applying that legal standard."), affd in part, rev'd in part, and remanded, 505 U.S. 833 (1992).

(138.) See Hardisty, supra note 23, at 52-57; Novak, supra note 112, at 758 n.10.

(139.) See KENT GREEN A WALT, STATUTORY AND COMMON LAW INTERPRETATION 205-06 (2013) ("One vital guide to a court's view of material facts is its stated rules of law; but it is also true that its rule of law can best be understood in terms of the facts before it.").

(140.) See infra notes 253-54 and accompanying text (discussing the precedential status of unexplained dispositions by the Supreme Court).

(141.) GREENAWALT, supra note 139, at 187; see also, e.g., RUPERT CROSS & J.W. HARRIS, PRECEDENT IN ENGLISH LAW 47-48 (4th ed. 1991) (describing the reasoning process through which a lower court might "infer[]" such a rationale).

(142.) For example, in the aftermath of Brown v. Board of Education, 347 U.S. 483, 495 (1954), which declared racial segregation in public schools unconstitutional, the Supreme Court issued several unexplained per curiam decisions invalidating segregation in other types of public accommodations, see, e.g., Gayle v. Browder, 352 U.S. 903 (per curiam) (desegregating the bus system of Montgomery, Alabama), affg 142 F. Supp. 707 (M.D. Ala. 1956); Mayor of Bait. City v. Dawson, 350 U.S. 877 (per curiam) (desegregating public beaches), affg 220 F.2d 386 (4th Cir. 1955). Although the Court's opinion in Brown could have been read to suggest a narrower holding confined to the public schooling context, such a reading was no longer available to lower courts after the per curiam decisions were handed down. See Paul G. Kauper, The Supreme Court and the Rule of Law, 59 MICH. L. REV. 531, 549 (1961) (observing that Brown "rested squarely and peculiarly on the finding that segregation in public schools resulted in harmful, discriminatory effects on Negro children" but noting that "in view of' the "later per curiam decisions, it must now be inferred that the school desegregation decision really was grounded on a broader principle, namely that all segregation legislation is invalid" (footnote omitted)).

(143.) See, e.g., Abramowicz & Stearns, supra note 130, at 1037 (noting the risk that courts might attempt to set forth law for future cases based on hypotheticals unconnected to the facts of the present case).

(144.) See id. at 957-58 (noting the absence of any accepted definition or test for distinguishing holdings from dicta).

(145.) U.S. CONST, art. Ill, [section] 2, cl. 1; see also infra notes 241-44 and accompanying text (discussing the prohibition on advisory opinions).

(146.) See Abramowicz & Stearns, supra note 130, at 1056 (characterizing the necessity-focused definition of "dicta" as "[t]he most influential"); see also, e.g., Schwab v. Crosby, 451 F.3d 1308, 1327 (11th Cir. 2006) ("[T]hat which is not necessary to the decision of a case is dicta.").

(147.) See, e.g., Abramowicz & Stearns, supra note 130, at 1056-58 (characterizing the necessitybased definition of dicta as "the easiest to falsify" and using the problem of alternative rationales as the principal illustration of its inadequacy).

(148.) Id.

(149.) See, e.g., Pruitt v. Levi Strauss & Co., 932 F.2d 458, 465 (5th Cir. 1991) (per curiam) ("[Alternative holdings are binding precedent and not obiter dictum.").

(150.) See, e.g., GrEENAWALT, supra note 139, at 185 ("Although the standard formulation [of the holding/dicta distinction] is in terms of 'necessary to the resolution of the case,' in the United States at least 'important in' is substantially more accurate." (footnote omitted) (quoting Obiter Dictum, BLACK'S LAW DICTIONARY (8th ed. 2004))); Abramowicz & Stearns, supra note 130, at 1065 (defending an alternative formulation under which classification as a "holding" would be reserved for "those propositions along the chosen decisional path or paths of reasoning that (1) are actually decided, (2) are based upon the facts of the case, and (3) lead to the judgment").

(151.) See, e.g., Abramowicz & Stearns, supra note 130, at 1026 ("[UJnless the judicial system is willing to invite upon itself claims of complete disingenuousness, a subsequent court will need to reconcile its ruling with the earlier case."); Adam N. Steinman, To Say What the Law Is: Rules, Results and the Dangers of Inferential Stare Decisis, 99 VA. L. REV. 1737, 1783-84 (2013) ("Even those who argue in favor of rule-based stare decisis ... typically argue that there should also be a duty to reconcile results." (emphasis omitted)).

(152.) See supra note 2.

(153.) See Cass R. Sunstein, Commentary, Incompletely Theorized Agreements, 108 Harv. L. Rev. 1733, 1739-40 (1995) [hereinafter Sunstein, Incompletely Theorized Agreements]; see also SUNSTEIN, supra note 26, at 11-14; F. Andrew Hessick & Jathan P. McLaughlin, Judicial Logrolling, 65 FLA. L. REV. 443, 468-69 (2013) (noting that plurality decisions involve incompletely theorized agreements).

(154.) See Sunstein, Incompletely Theorized Agreements, supra note 153, at 1739.

(155.) Id.

(156.) Id. at 1740.

(157.) Id. at 1740-41.

(158.) Id.

(159.) Id. at 1740; see also Sunstein, supra note 26, at 13 (emphasizing that "[!Incompletely theorized agreements are by no means unaccompanied by reasons").

(160.) See supra notes 140-42 and accompanying text (discussing difficulties with extracting precedential guidance from unexplained rulings).

(161.) See FREDERICK SCHAUER, THINKING LIKE A LAWYER: A NEW INTRODUCTION TO LEGAL REASONING 78 n.35 (2009) ("[R]easons are always more general than the outcomes that they are reasons for ...."); see also Frederick Schauer, Giving Reasons, 47 Stan. L. Rev. 633, 638 (1995) [hereinafter Schauer, Giving Reasons] (discussing the relationship between reason-giving and generality); Sunstein, Incompletely Theorized Agreements, supra note 153, at 1740 (engaging in a similar discussion).

(162.) See Schauer, Giving Reasons, supra note 161, at 649. The commitments created in this way provide an independent reason to act in a particular manner in the future but may be outweighed by other competing reasons. Id. at 643 n.25 ("What I have a reason to do is not necessarily what I should do, all things considered.").

(163.) E.g., MacPherson v. Buick Motor Co., 111 N.E. 1050,1053 (N.Y. 1916).

(164.) E.g., Hoenig v. Cent. Stamping Co., 6 N.E.2d 415, 415-16 (N.Y. 1936) (per curiam) (extending MacPherson to a case involving a negligently manufactured coffee urn); Smith v. Peerless Glass Co., 181 N.E. 576, 577-78 (N.Y. 1932) (applying MacPherson to a case involving a soda bottle).

(165.) See, e.g., Jeremy Waldron, Stare Decisis and the Rule of Law: A Layered Approach, 111 Mich. L. Rev. 1, 11 (2012) (suggesting that the process of precedent following "cannot really get underway unless the precedent judge ... does something to present her decision in an articulate light that allows subsequent judges to go to work on it").

(166.) There might be some meta-rule in place that would resolve the uncertainty. For example, there might be an agreed-upon default rule that both parents' assent must be obtained to exceed the bedtime, resulting in the request's denial. Alternatively, a meta-rule might specify that either parent can grant permission unilaterally. But such a default rule may not exist, perhaps because the parents failed to consider the question or considered such a rule but could not agree on what the appropriate meta-rule should be. Cf. infra notes 257-71 and accompanying text (considering the question of default rules applicable to plurality precedent).

(167.) See SUNSTEIN, supra note 26, at 16 (emphasizing "depth" and "width" as distinct dimensions along which decisions may be analyzed).

(168.) Id. at 13.

(169.) Id. at 11-14 (formatting altered).

(170.) Id. at 10.

(171.) Id. at 18.

(172.) See, e.g., Tara Leigh Grove, The Structural Case for Vertical Maximalism, 95 cornell L. Rev. 1, 11 (2009) (observing that the width (or narrowness) of a Supreme Court decision, rather than its depth, "appears to have the most immediate impact on the lower federal and state courts").

(173.) 561 U.S. 742 (2010).

(174.) Id. at 750.

(175.) District of Columbia v. Heller, 554 U.S. 570, 573,635 (2008).

(176.) McDonald, 561 U.S. at 767.

(177.) See id. at 758-59 (plurality opinion); id. at 791 (Scalia, J., concurring).

(178.) Id. at 805-06 (Thomas, J., concurring in part and concurring in the judgment).

(179.) Id. at 858-60 (Stevens, J., dissenting); id. at 912-13 (Breyer, J., dissenting).

(180.) See, e.g., David S. Cohen, McDonald's Paradoxical Legacy: State Restrictions of Non-Citizens' Gun Rights, 71 MD. L. REV. 1219, 1220-22 (2012) (describing the voting paradox in McDonald).

(181.) Id. at 1222.

(182.) Id. at 1224-25.

(183.) See supra note 94 and accompanying text.

(184.) A Westlaw search for decisions citing both McDonald and Marks returned only a single result. See Binderup v. Att'y Gen. U.S., 836 F.3d 336 (3d Cir. 2016), petition for cert, filed, No. 16-847 (Jan. 5, 2017). Though the opinion announcing the Third Circuit's en banc decision in Binderup cited both McDonald and Marks, that opinion did not apply the narrowest grounds doctrine to determine the holding of McDonald. Instead, the opinion referred to Marks and the narrowest grounds doctrine in the course of describing the precedential significance of the en banc court's own fractured decision in Binderup. Id. at 356.

(185.) See, e.g., Commonwealth v. McGowan, 982 N.E.2d 495, 498 n.5 (Mass. 2013) ("[A]ny distinction between the plurality's and Justice Thomas's approach to incorporation [in McDonald] is immaterial where, as here, the defendant is both a person and a citizen of the United States.").

(186.) 337 U.S. 582 (1949); see also Post & Salop, supra note 96, at 748 (describing Tidewater as "[p]erhaps the most famous example" of a case involving a revealed voting paradox).

(187.) 337 U.S. at 583-84 (plurality opinion).

(188.) Id. at 588, 603-04.

(189.) Id. at 604, 607-08, 625-26 (Rutledge, J., concurring).

(190.) See id. at 646 (Vinson, C.J., dissenting); id. at 646-47 (Frankfurter, J., dissenting).

(191.) See Post & Salop, supra note 96, at 748-49.

(192.) See, e.g., James E. Pfander, The Tidewater Problem: Article III and Constitutional Change, 79 Notre Dame L. Rev. 1925, 1926-27 (2004) (characterizing the plurality's position as "com[ing] close to outright jurisdictional apostasy" and as "a challenge to the accepted portrait of federal courts as courts of limited jurisdiction").

(193.) See, e.g., Frank H. Easterbrook, Ways of Criticizing the Court, 95 Harv. L. Rev. 802, 820 n.41 (1982) ("What does stare decisis compel the Court to do the next time District residents file such a 'diversity' suit and the same problem is presented?"); Novak, supra note 112, at 771 ("There seems to be no easy way to extract any coherent precedential principles from a case like Tidewater.").

(194.) Siegmund v. Gen. Commodities Corp., 175 F.2d 952,953-54 (9th Cir. 1949).

(195.) Id. at 953.

(196.) Id. at 953-54.

(197.) Detres v. Lions Bldg. Corp., 234 F.2d 596, 601-03 (7th Cir. 1956); see also, e.g., Greene v. Teffeteller, 90 F. Supp. 387, 388 (E.D. Tenn. 1950) (concluding that Tidewater supported upholding jurisdiction over suits involving D.C. residents under a separate diversity statute).

(198.) See, e.g., United States v. Cundiff, 555 F.3d 200, 210-11 (6th Cir. 2009) (applying Rapanos); Banco Nacional de Cuba v. Chase Manhattan Bank, 658 F.2d 875, 881, 883-84 (2d Cir. 1981) (applying First Nat'l City Bank v. Banco Nacional de Cuba, 406 U.S. 759 (1972)); Serv. Oil, Inc. v. State, 479 N.W.2d 815, 819-21 (N.D. 1992) (applying Am. Trucking Ass'ns v. Smith, 496 U.S. 167 (1990)).

(199.) See supra notes 146-47 and accompanying text; see also Judge Pierre N. Levai, Judging Under the Constitution: Dicta About Dicta, James Madison Lecture at the New York University School of Law (Oct. 18, 2005), in 81 N.Y.U. L. Rev. 1249,1256-57 (2006).

(200.) See infra note 240 and accompanying text (discussing the precedential effect of unexplained Supreme Court decisions); see also, e.g., FREDERICK SCHAUER, PLAYING BY THE RULES: A PHILOSOPHICAL EXAMINATION OF RULE-BASED DECISION-MAKING IN LAW AND IN LIFE 184 (1991) (emphasizing that decisionmaking under such conditions is "substantially non-constraining").

(201.) See supra notes 173-85 and accompanying text (discussing McDonald).

(202.) As Jeremy Sheff has recently observed, legal rules can be "understood to define categories" of results that can be analyzed using the tools of mathematical set theory. Jeremy Sheff, Legal Sets 4-5 (St. John's Sch. of Law Legal Studies Research Paper Series, No. 16-0019, 2016) (emphasis omitted), http://ssrn.com/abstract=2830918. The domain of shared agreement covered by two competing rationales for a judgment--Rationale A and Rationale B--consists of the intersection of the set of results produced by Rationale A and the set of results produced by Rationale B. See id. at 18 ("The intersection of two sets A and B ... is defined as the set ... containing all objects that are elements of both A and B.")

(203.) See, e.g., Pauline T. Kim, Lower Court Discretion, 82 N.Y.U. L. Rev. 383, 414-15 (2007) (discussing similar constraints using the metaphor of a "zone of discretion"); Jack Knight, Are Empiricists Asking the Right Questions About Judicial Decisionmaking?, 58 Duke L.J. 1531, 1548-49 (2009) (discussing precedent and professional norms of justification as constraints on the "feasible choice set" available to judges).

(204.) See, e.g., United States v. Rodriguez-Preciado, 399 F.3d 1118,1140 (9th Cir. 2005) (Berzon, J., dissenting in part) (asserting, without elaboration, that where no single unambiguously "narrowest" opinion exists, "the only binding aspect of a splintered decision is its specific result" (quoting Anker Energy Corp. v. Consolidation Coal Co., 177 F.3d 161, 170 (3d Cir. 1999))).

(205.) See supra notes 132-34 and accompanying text (discussing critiques of the "facts-plus-outcome" conception of precedent).

(206.) The methodological nature of the Marks Court's comments regarding plurality decisions' precedential status might raise questions regarding whether such comments were part of the case's holding or more properly categorized as dicta. See, e.g., Randy J. Kozel, The Scope of Precedent, 113 Mich. L. Rev. 179, 193 (2014) (discussing the precedential status of "doctrinal frameworks" that "sweep far beyond the facts at hand to address other situations not concurrently before the court"). But this argument is easily disposed of. As discussed above, virtually any decision supported by a reason will sweep at least somewhat more broadly than the particular decision itself. See supra notes 16165 and accompanying text. Nor does the methodological nature of the guidance offered by the Court make any obvious difference to categorizing the Court's statements as either holdings or dicta. See, e.g., Michael Coenen, Rules Against Rulification, 124 Yale L.J. 644, 705-07 (2014) (considering and rejecting the argument that stare decisis norms do not extend to methodological commitments).

(207.) Marks v. United States, 430 U.S. 188,193 (1977) (quoting Gregg v. Georgia, 428 U.S. 153, 169 n.15 (1976) (plurality opinion)).

(208.) See, e.g., United States v. Cundiff, 555 F.3d 200, 208 (6th Cir. 2009) (interpreting Marks ils "instructing] lower courts to choose the 'narrowest' concurring opinion" in the precedent-setting case); Hall v. Kelso, 892 F.2d 1541,1543 n.l (11th Cir. 1990) (referring to the "narrowest concurrence rule" prescribed by Marks).

(209.) James A. Bloom, Plurality and Precedence: Judicial Reasoning, Lower Courts, and the Meaning of United States v. Winstar Corp., 85 WASH. U. L. REV. 1373, 1379 (2008) (noting this ambiguity in the Marks decision).

(210.) See Marks, 430 U.S. at 193.

(211.) See supra notes 32-38 and accompanying text (describing the opinions in Memoirs).

(212.) See supra notes 41-43 and accompanying text.

(213.) See supra notes 116-19 and accompanying text.

(214.) 135 S. Ct. 2726 (2015).

(215.) 553 U.S. 35 (2008).

(216.) Glossip, 135 S. Ct. at 2738 n.2.

(217.) 553 U.S. at 51-52 (plurality opinion).

(218.) Id. at 94 (Thomas, J., concurring in the judgment).

(219.) See Jackson v. Danberg, 594 F.3d 210, 222 (3d Cir. 2010) (observing that "any lethal injection protocol constitutionally acceptable to the plurality would invariably pass" the standard endorsed by Justices Scalia and Thomas). Justices Scalia and Thomas made clear that the only punishments they believed would violate the Eighth Amendment were those that were "designed to inflict torture as a way of enhancing a death sentence" and that were "intended to produce a penalty worse than death." 553 U.S. at 102 (Thomas, J., concurring in the judgment). Thus, even if a petitioner could establish that a challenged punishment had been designed for deliberately cruel purposes, the absence of a feasible alternative method of execution--the focus of the plurality's rationale-would almost certainly be determinative under their standard as well.

(220.) 551 U.S. 930 (2007).

(221.) Id. at 935.

(222.) 477 U.S. 399, 413-17 (1986) (plurality opinion); id. at 424-25 (1986) (Powell, J., concurring in part and concurring in the judgment).

(223.) Id. at 411-12 (plurality opinion).

(224.) Id. at 427 (Powell, J., concurring in part and concurring in the judgment).

(225.) 551 U.S. at 949 (quoting 28 U.S.C. [section] 2254(d)(1)).

(226.) See O'Dell v. Netherland, 521 U.S. 151, 160-62 (1997) (interpreting Gardner v. Florida, 430 U.S. 349 (1977)); Romano v. Oklahoma, 512 U.S. 1, 9 (1994) (interpreting Caldwell v. Mississippi, 472 U.S. 320 (1985)); City of Lakewood v. Plain Dealer Publ'g Co., 486 U.S. 750, 764 n.9 (1988) (interpreting Kovacs v. Cooper, 336 U.S. 77 (1949)).

In Gardner, the case interpreted in O'Dell, six Justices had agreed that judicial consideration of evidence that had not been made available to the defendant during the sentencing phase of a capital trial violated the defendant's constitutional rights. 430 U.S. at 362 (plurality opinion). The three-Justice plurality opinion concluded that the defendant "was denied due process of law when the death sentence was imposed, at least in part, on the basis of information which he had no opportunity to deny or explain." Id. But Justice White's sole concurrence differed from the three-Justice plurality opinion by emphasizing the specific nature of the evidence at issue--evidence relating to the "character and record of the individual offender," id. at 364 (White, J., concurring in the judgment) (quoting Woodson v. North Carolina, 428 U.S. 280, 304 (1976))--and the centrality of the Eighth Amendment to his analysis, rather than the plurality's more general due process reasoning, id. at 363-64. Given the more limited scope of Justice White's concurrence in Gardner, the O'Dell Court held that the subsequent extension of the plurality's reasoning in a later case had not "followed ineluctably" from Gardner's holding and that retrospective application of the later case was thus not required. O'Dell, 521 U.S. at 160-62; cf. United States v. Duvall, 740 F.3d 604, 607 n.2 (D.C. Cir. 2013) (Rogers, J., concurring in the denial of rehearing en banc) (describing O'Dell as "consistent with" a "'subset' interpretation of the Marks rule").

Although the O'Dell Court characterized Justice White's concurrence as having "provid[ed] the narrowest grounds of decision among the Justices whose votes were necessary to the judgment," 521 U.S. at 160, Justice White's opinion and the plurality opinion in Gardner collectively accounted for only four votes. The O'Dell Court did not mention the remaining two Justices who joined in the majority--Justice Blackmun and Chief Justice Burger. Justice Blackmun concurred in the judgment separately in a cursory opinion suggesting invalidation of the defendant's sentence was compelled by two of the Court's earlier Eighth Amendment decisions. Gardner, 430 U.S. at 364 (Blackmun, J., concurring in the judgment). The plurality opinion in Gardner did not mention either of the decisions referred to by Justice Blackmun, but one of those cases--Woodson v. North Carolina, 428 U.S. 280 (1976)--was the principal authority relied on in Justice White's concurrence, 430 U.S. at 364 (White, J., concurring in the judgment). It is therefore conceivable that the O'Dell majority viewed Justice Blackmun's opinion as expressing a view substantially similar to Justice White's. Cf. Neuenkirchen, supra note 46, at 415 (concluding that the basis for Justice Blackmun's Garner concurrence "must be similar to the position expressed by Justice White, who also relied on Woodson"). Chief Justice Burger concurred in the judgment without providing any explanation at all. Gardner, 430 U.S. at 362 (Burger, C.J., concurring in the judgment). But given the way the other opinions in the case aligned with one another, his vote could not have affected the outcome as the Court's decision would have been the same even if Chief Justice Burger had dissented or abstained.

In Romano, the Court singled out Justice O'Connor's sole concurrence from Caldwell as the narrowest and therefore controlling opinion. Romano, 512 U.S. at 9. Justice O'Connor would have concluded that evidence could not be admitted at the sentencing phase of a capital trial where the evidence both diminished the jurors' sense of responsibility and was inaccurate and misleading. Caldwell, 472 U.S. at 342 (O'Connor, J., concurring in part and concurring in the judgment). A four-Justice plurality would have dispensed with the latter requirement by mandating the exclusion of any evidence that diminished jurors' sense of responsibility, even if such evidence were accurate and were not misleading. See id. at 335-36 (plurality opinion).

In City of Lakewood, the Court identified the Kovacs plurality opinion as the controlling opinion. 486 U.S. at 764 n.9. Only one other Justice concurred in the Kovacs judgment--Justice Jackson, who argued that the Court should have repudiated its holding in an earlier case, Saia v. New York, 334 U.S. 558 (1948), rather than adopt the plurality's more speech-protective standard, 336 U.S. at 97-98 (Jackson, J., concurring).

(227.) See Grutter v. Bollinger, 539 U.S. 306, 325 (2003) (declining to conduct the Marks analysis to determine whether a rationale "set forth in part of [a concurring] opinion joined by no other Justice" was binding on later courts); Nichols v. United States, 511 U.S. 738, 745-46 (1994) (dispensing with the Marks analysis after observing that "[a] number of Courts of Appeals have decided that there is no lowest common denominator or 'narrowest grounds' that represents the Court's holding" in the prior plurality case).

(228.) See supra notes 58-59 and accompanying text (describing the distinction between opinions involving nested rationales and those involving non-nested rationales).

(229.) See, e.g., STEARNS, supra note 88, at 124-29 (arguing that Marks supports the fifth vote approach to plurality precedent).

(230.) Marbury v. Madison, 5 U.S. (1 Cranch) 137,177 (1803).

(231.) See, e.g., Rivers v. Roadway Express, Inc., 511 U.S. 298, 312 (1994) ("It is this Court's responsibility to say what a statute means, and once the Court has spoken, it is the duty of other courts to respect that understanding of the governing rule of law.").

(232.) See Justice Antonin Scalia, The Rule of Law as a Law of Rules, Oliver Wendell Holmes, Jr. Lecture at Harvard University (Feb. 14, 1989), in 56 U. CHI. L. REV. 1175, 1176-77 (1989) ("In a judicial system such as ours, in which judges are bound, not only by the text of code or Constitution, but also by the prior decisions of superior courts, ... courts have the capacity to 'make' law.").

(233.) Adrian Vermeule, The Judiciary Is a They, Not an It: Interpretive Theory and the Fallacy of Division, 14 J. CONTEMP. LEGAL ISSUES 549, 550 (2005).

(234.) See AKHIL REED AMAR, AMERICA'S UNWRITTEN CONSTITUTION: THE PRECEDENTS AND PRINCIPLES WE LIVE By 356 (2012) ("Multi-member institutions ... can do nothing--nothing at all!--unless certain basic social-choice rules are in place within these institutions." (formatting altered)).

(235.) See, e.g., Steinman, supra note 151, at 1738 ("[T]he ground rules for discerning the lawgenerating content and scope of a judicial decision remain remarkably murky.").

(236.) Compare U.S. CONST, art. I, [section] 7 (setting out basic lawmaking procedures for Congress), with id. art. Ill (lacking any comparable procedures).

(237.) Certain of the Court's institutional practices, such as its life-tenured judges and its limited jurisdiction, are specified by the Constitution. See id. art. Ill, [section][section] 1-2. Others, such as the number of Justices who compose the Court and the minimum number necessary for a quorum, are prescribed by federal statute. See 28 U.S.C. [section] 1 (2015). But many others, including those surveyed in this Subpart, are governed by "unwritten" legal conventions. See Stephen E. Sachs, The "Unwritten Constitution" and Unwritten Law, 2013 U. ILL. L. REV. 1797, 1801 (describing "unwritten" legal rules as those that "have no single and authoritative textual source, no pedigree tracing their validity back to a written ancestor" (emphasis omitted)).

(238.) See, e.g., Jeremy Waldron, Who Needs Rules of Recognition?, in THE RULE OF RECOGNITION AND THE U.S. CONSTITUTION 327, 335 (Matthew D. Adler & Kenneth Einar Himma eds., 2009) (attributing lack of "clear, shared, and public rules of change, so far as changes in the law wrought by the judiciary are concerned" partly to "profound!]" societal ambivalence "about judges' making and changing law").

(239.) See AMAR, supra note 234, at 357-61 ("From its first day to the present day, the [Supreme] Court has routinely followed the majority-rule principle without even appearing to give the matter much thought.").

(240.) See, e.g., CTS Corp. v. Dynamics Corp. of Am., 481 U.S. 69, 81 (1987) ("As the plurality opinion in [Edgar v. MITE Corp., 457 U.S. 624 (1982),] did not represent the views of a majority of the Court, we are not bound by its reasoning." (footnote omitted)).

(241.) See William R. Casto, The Early Supreme Court Justices' Most Significant Opinion, 29 OHIO N.U. L. REV. 173, 173 (2002) (noting that the prohibition is conventionally traced to a 1793 letter from the Justices to President Washington refusing a request for legal advice regarding certain foreign affairs controversies).

(242.) See, e.g., Flast v. Cohen, 392 U.S. 83, 96 (1968) ("[T]he implicit policies embodied in Article III, and not history alone, impose the rule against advisory opinions on federal courts.").

(243.) See, e.g., Antonin Scalia, The Doctrine of Standing as an Essential Element of the Separation of Powers, 17 SUFFOLK U. L. REV. 881, 882 (1983) (acknowledging that litigant standing requirements are "not a linguistically inevitable conclusion" from the text of Article III); Note, Advisory Opinions and the Influence of the Supreme Court over American Policymaking, 124 HARV. L. REV. 2064, 2067 (2011) ("This construction of the federal judicial power was not inevitable.... English judges had a longstanding practice of issuing advisory opinions upon the monarch's request.").

(244.) See CHARLES ALAN WRIGHT & MARY KAY KANE, LAW OF FEDERAL COURTS 65-66 (7th ed. 2011) (calling the ban on advisory opinions "the oldest and most consistent thread in the federal law of justiciability").

(245.) See, e.g., Kozel, supra note 206, at 187 (describing the "stark dichotomy" between holdings and dicta as key to the "classic account of precedential scope"); see also supra notes 143-46 and accompanying text (discussing the distinction).

(246.) See supra notes 147, 151 and accompanying text (noting the absence of any universally agreed-upon test or definition for distinguishing holdings from dicta).

(247.) See, e.g., Doughty v. Underwriters at Lloyd's, 6 F.3d 856, 861 n.3 (1st Cir. 1993) (arguing that circuit courts should treat "[c]arefully considered language of the Supreme Court" as authoritative even if such language is "technically dictum"); see also United States v. Oakar, 111 F.3d 146, 153 (D.C. Cir. 1997) (quoting Doughty, 6 F.3d at 861 n.3).

(248.) See, e.g., Kirtsaeng v.John Wiley & Sons, Inc., 133 S. Ct. 1351, 1368 (2013) ("[W]e are not necessarily bound by dicta should more complete argument demonstrate that the dicta is not correct."); Cent. Va. Cmty. Coll. v. Katz, 546 U.S. 356, 363 (2006) ("[W]e are not bound to follow our dicta in a prior case in which the point now at issue was not fully debated."); Humphrey's Ex'r v. United States, 295 U.S. 602,627 (1935) ("[D]icta ... may be followed if sufficiently persuasive but ... are not controlling...." (italics omitted)).

(249.) See, e.g., Kennedy v. Louisiana, 554 U.S. 407, 426-31 (2008) (observing that the majority of lower courts had correctly determined that they were not bound by overbroad dicta in a prior Supreme Court opinion); cf. Williams v. Taylor, 529 U.S. 362, 412 (2000) (declaring that only "the holdings, as opposed to the dicta, of th[e Supreme] Court's decisions" can constitute "clearly established" federal law for purposes of federal habeas relief).

(250.) See, e.g., Abramowicz & Stearns, supra note 130, at 1019-20 (observing that a judicial system lacking any norms or rules requiring judges to "limit their case dispositions to resolving issues implicated by material case facts" would "be akin to a legislature in which any single legislator could pass a bill").

(251.) See, e.g., Dorf, supra note 132, at 1997-98 (contending that the distinction between holdings and dicta carries Article III implications); Levai, supra note 199, at 1259 (contending that judicial "lawmaking through proclamation of dicta" is constitutionally illegitimate).

(252.) See, e.g., Edward A. Hartnett, A Matter of Judgment, Not a Matter of Opinion, 74 N.Y.U. L. REV. 123, 126 (1999) (observing that "[t]he operative legal act performed by a court is the entry of a judgment" and that written opinions merely provide "an explanation of reasons for that judgment").

(253.) Gary Lawson & Christopher D. Moore, The Executive Power of Constitutional Interpretation, 81 IOWA L. REV. 1267,1328 (1996).

(254.) See Hicks v. Miranda, 422 U.S. 332, 344-45 (1975) ("[L]ower courts are bound by summary decisions by this Court ...."). While the Court has made clear that lower courts must accord binding effect to its summary dispositions, it has warned that such dispositions should not be read as adopting the reasoning of the judicial decision below, Mandel v. Bradley, 432 U.S. 173, 176 (1977) (per curiam), and has cautioned that "[ajscertaining the reach and content of summary actions may itself present issues of real substance," Hicks, 422 U.S. at 345 n.14.

(255.) The Court has explained:

   The duty of this court, as of every judicial tribunal, is limited
   to determining rights of persons or of property, which are actually
   controverted in the particular case before it. When, in determining
   such rights, it becomes necessary to give an opinion upon a
   question of law, that opinion may have weight as a precedent for
   future decisions. But the court is not empowered to decide moot
   questions or abstract propositions, or to declare, for the
   government of future cases, principles or rules of law which cannot
   affect the result as to the thing in issue in the case before it.


California v. San Pablo & Tulare R.R. Co., 149 U.S. 308, 314 (1893); see also, e.g., DaimlerChrysler Corp. v. Cuno, 547 U.S. 332, 341 (2006) ("If a dispute is not a proper case or controversy, the courts have no business deciding it, or expounding the law in the course of doing so."); Muskrat v. United States, 219 U.S. 346, 361 (1911) (explaining that the power of judicial review exists only to the extent necessary to determine "the rights of the litigants in justiciable controversies").

(256.) See generally G. Edward White, The Internal Powers of the Chief Justice: The Nineteenth-Century Legacy, 154 U. PA. L. REV. 1463, 1467-68 (2006) (describing the emergence of the "opinion of the Court" practice under Chief Justice Ellsworth and its later entrenchment under Chief Justice Marshall).

(257.) See supra Parts I.B.I (describing the "implicit consensus" understanding of Marks), III.A (describing the Supreme Court's application of the narrowest grounds doctrine).

(258.) See, e.g., Durant v. Essex Co., 74 U.S. (7 Wall.) 107, 110 (1868) ("It has long been the doctrine in this country and in England, where courts consist of several members, that no affirmative action can be had in a cause where the judges are equally divided in opinion as to the judgment to be rendered or order to be made.").

(259.) Michael Coenen, Comment, Original Jurisdiction Deadlocks, 118 YALE L.J. 1003, 1008 (2009); see also, e.g., Edward A. Hart nett, Ties in the Supreme Court of the United States, 44 WM. & MARY L. REV. 643, 652 & n.38 (2002) (describing the rule as an application of the Latin maxim "semper praesumitur pro negante, that is, it is always to be presumed in favor of the negative").

(260.) For the now-obligatory citation for nondecision as a form of decision, see Rush, Freewill, on PERMANENT WAVES (Moon Records 1980): "If you choose not to decide / You still have made a choice."

(261.) Cf. Marks v. United States, 430 U.S. 188, 193 (1977) (indicating that the "narrowest grounds" of a decision should be determined by looking to the views of Justices who "concurred in the judgments" (quoting Gregg v. Georgia, 428 U.S. 153, 169 n.15 (1976) (plurality opinion))).

(262.) See supra Part I.B.I (describing the "implicit consensus" approach to Marks).

(263.) See, e.g., United States v. Gerke Excavating, Inc., 464 F.3d 723, 724 (7th Cir. 2006) (per curiam) (interpreting Marks to require "lower-court judges ... to follow the narrowest ground to which a majority of the Justices would have assented if forced to choose" (emphasis added)).

(264.) By contrast, a longstanding, informal convention of the Court demands that the Justices' collective votes always lead to a conclusive outcome in the case--for example, an affirmance, a reversal, or remand: where an initial vote reveals a three-way split between such options, some Justices have switched their votes to produce a majority-supported result. See, e.g., Rapanos v. United States, 547 U.S. 715, 810 (2006) (Stevens, J., dissenting) ("It has been our practice in a case coming to us from a lower federal court to enter a judgment commanding that court to conduct any further proceedings pursuant to a specific mandate. That prior practice has, on occasion, made it necessary for Justices to join a judgment that did not conform to their own views."); Hamdi v. Rumsfeld, 542 U.S. 507, 553 (2004) (Souter, J., concurring in part, dissenting in part, and concurring in the judgment); Screws v. United States, 325 U.S. 91, 134 (1945) (Rutledge, J., concurring in the result).

(265.) See Evan H. Caminker, Sincere and Strategic Voting Norms on Multimember Courts, 97 MICH. L. REV. 2297, 2317-19 (1999) (discussing formation of such "majority-opinion coalition[s]"). Such convergence tends to occur around the median Justice's position, and social choice theory suggests reasons for predicting that particular outcome. Id. at 2320. But such "[convergence on a center position ... is not guaranteed" because "the effects of small-group dynamics" might influence individual Justices' voting behavior in ways that are difficult to predict. Id.

(266.) As one perceptive commenter observed in critiquing Stearns's social choice argument for the fifth vote approach to Marks:

   [T]he problem with Stearns's social choice justification for the
   [fifth vote approach] is that it assumes that it is necessary, or
   at least desirable, to select one opinion as a "winner" in Supreme
   Court plurality decisions. In other words, defending the Marks
   doctrine as a Condorcetproducing rule does justify the "winners"
   that Marks selects, but does not justify the act of selecting a
   "winner" in the first place.


Joseph M. Cacace, Note, Plurality Decisions in the Supreme Court of the United States: A Reexamination of the Marks Doctrine After Rapanos v. United States, 41 SUFFOLK U. L. REV. 97, 127-28 (2007) (footnote omitted).

(267.) See, e.g., Adrian Vermeule, Absolute Majority Rules, 37 BRIT.J. POL. SCI. 643, 643-47 (2007) (observing that "[a] fully-specified voting rule must state both a multiplier and a multiplicand" and that "[t]he choice of multiplicand is at least as important as the choice of a multiplier").

(268.) See, e.g., Novak, supra note 112, at 768 (asserting that in the dual majority context "the technical alignment of the Justices is irrelevant" because "what is important is the presence of agreement by an actual majority of the Court").

(269.) See United States v. Johnson, 256 F.3d 895, 921 (9th Cir. 2001) (Tashima, J., concurring) ("By definition, dictum is an unnecessary statement made by the majority; unless a statement is made by a majority, there is no need to engage in an analysis of whether that particular statement is dictum or a holding.").

(270.) See, e.g., Jonathan H. Adler, Once More, with Feeling: Reaffirming the Limits of Clean Water Act Jurisdiction, in THE SUPREME COURT AND THE CLEAN WATER ACT: FIVE ESSAYS 81,94 (L. Kinvin Wroth ed., 2007) (arguing that a "dissent, like dicta from a majority opinion, ... does not--indeed cannot--form part of the holding of the Court"); A.M. Honore, Note, Ratio Decidendi: Judge and Court, 71 Law Q. Rev. 196, 198 (1955) ("[O]pinions of [dissenting] judges cannot form part of the ratio decidendi of a case [because] they are not reasons for the order made by the court ....").

(271.) See supra note 206 and accompanying text (discussing the precedential status of the Supreme Court's methodological guidance).

(272.) Corley, supra note 11, at 40; see also, e.g., John F. Davis & William L. Reynolds, Juridical Cripples: Plurality Opinions in the Supreme Court, 1974 Duke L.J. 59, 62 (suggesting that some plurality decisions do "more to confuse the current state of the law than to clarify it").

(273.) See, e.g., Planned Parenthood of Se. Pa. v. Casey, 947 F.2d 682,693 (3d Cir. 1991) (asserting that the "principal objective of the Marks rule, "promoting] predictability" in the law, "requires that, whenever possible, there be a single legal standard for the lower courts to apply in similar cases"), affd in part, rev'd in part, and remanded, 505 U.S. 833 (1992); STEARNS, supra note 88, at 135 (asserting that one of the "overriding objectives" of the Marks Court's directive was to "ensur[e] an identifiable holding in the maximum number of fractured ... decisions").

(274.) See supra Part III.A.

(275.) But see infra notes 306-08 and accompanying text (discussing the effect of prior precedent as an additional constraint on lower courts' discretion).

(276.) On the Court's shrinking appellate docket, see, for example, Ryan J. Owens & David A. Simon, Explaining the Supreme Court's Shrinking Docket, 53 WM. & MARY L. REV. 1219 (2012). On its willingness to tolerate longstanding circuit splits, see, for example, Wayne A. Logan, Constitutional Cacophony: Federal Circuit Splits and the Fourth Amendment, 65 VAND. L. REV. 1137, 1139-40 (2012). On its practice of issuing narrow, fact-specific rulings that provide limited guidance for future cases, see, for example, Grove, supra note 172, at 57-58; and Frederick Schauer, Abandoning the Guidance Function: Morse v Frederick, 2007 SUP. CT. REV. 205, 207.

(277.) See Amanda Frost, Overvaluing Uniformity, 94 VA. L. REV. 1567, 1610 (2008) (noting steps Congress could take to enhance uniformity were it so inclined).

(278.) See, e.g., Davis & Reynolds, supra note 272, at 81-86 (urging the Court to refrain from issuing plurality decisions whenever possible).

(279.) Cf. Arizona v. Gant, 556 U.S. 332, 354 (2009) (Scalia, J., concurring) (explaining the Justice's decision to concur in a majority rationale with which he substantively disagreed because of the need to avoid an "unacceptable" level of uncertainty in the governing legal rule). By contrast, the Court has developed such an informal vote-switching practice for cases where vote-switching is necessary to produce a majority-supported dispositional ruling in a case. See supra note 264 (describing this convention).

(280.) Jerome I. Braun, Five Minus Four Equals Nine: Understanding Fractured Supreme Court Decisions, and Some Ways the Court Could Make Them Less Vexing, 246 F.R.D. 265, 272 (2008) (suggesting such a reform and how it could operate).

(281.) See supra Part I.B.2 (describing the fifth vote approach to Marks).

(282.) See, e.g., Douglas J. Whaley, Comment, A Suggestion for the Prevention of No-Clear-Majority Judicial Decisions, 46 TEX. L. REV. 370, 376 (1968) (arguing for a rule requiring that "the opinion that the most nondissenting judges vote for" be treated as precedential to promote judicial compromise).

(283.) Cf. Saul Levmore, Ruling Majorities and Reasoning Pluralities, 3 THEORETICAL INQUIRIES L. 87, 102 n.29 (2002) (describing an Israeli law providing that "[w]here there is no majority for any one opinion in a civil matter, the view of the senior judge shall prevail" (quoting Courts Law (Consolidated Version), 5744-1984, [section] 80, 38 LSI 271 (1983-84))).

(284.) See Adam M. Samaha, Randomization in Adjudication, 51 WM. & MARY L. REV. 1, 67-70 (2009) (discussing certain costs and benefits of randomizing merits determinations by courts).

(285.) See, e.g., Schauer, supra note 276, at 207-08 (identifying the Court's decision in Morse v. Frederick, 551 U.S. 393 (2007), as "exemplify[ing] th[e] disturbing trend" of "narrow and fact-specific" Supreme Court rulings "that may in theory produce the right outcome for the particular case before the Court, but ... provid[e] virtually no assistance for lower courts").

(286.) See, e.g., Grove, supra note 172, at 28-29 ("[I]n our current judiciary, a minimalist Supreme Court opinion serves to delegate substantial decision-making responsibility to the Court's judicial inferiors.").

(287.) See, e.g., Kim, supra note 203, at 409 ("When judges decide cases, the body of applicable legal rules--statutes, regulations, and prior precedents--constitute the relevant constraints on their decisionmaking.").

(288.) Id. at 410 ("[E]ven judges with a strong legal preference for following superior court precedent will encounter cases in which 'the guidance of authoritative legal rules runs out' and therefore find themselves with discretion to decide." (footnote omitted) (quoting Charles M. Yablon, Justifying the Judge's Hunch An Essay on Discretion, 41 HASTINGS L.J. 231, 239 (1990))).

(289.) Maurice Rosenberg, Judicial Discretion of the Trial Court, Viewed from Above, 22 SYRACUSE L. Rev. 635, 637 (1971); see also, e.g., Kent Greenawalt, Discretion and Judicial Decision: The Elusive Quest for the Fetters That Bind Judges, 75 COLUM. L. REV. 359, 378 (1975) ("[W]hen more than one result will widely be regarded as a satisfactory fulfillment of his judicial responsibilities then it does not make good sense to say that a judge is under a duty to reach one result rather than another; as far as the law is concerned, he has discretion to decide between them.").

(290.) See, e.g., Kim, supra note 203, at 408 ("[Discretion ... implies something more than mere choice. It suggests that a decision should be made not randomly or arbitrarily, but by exercising judgment in light of some applicable set of standards, guidelines, or values.").

(291.) See, e.g., Frank B. Cross, Political Science and the New Legal Realism A Case of Unfortunate Interdisciplinary Ignorance, 92 NW. U. L. REV. 251, 260-61 (1997) (discussing the importance of professional norms of explanation and justification as a constraint on judicial decisionmaking).

(292.) Kim, supra note 203, at 410 (quoting George C. Christie, An Essay on Discretion, 1986 DUKE L.J. 747, 748).

(293.) See, e.g., Hart v. Massanari, 266 F.3d 1155, 1175 (9th Cir. 2001) ("While many consider the principle of binding authority indispensable[,] ... it is important to note that it is not an unalloyed good.").

(294.) See, e.g., Kozel, supra note 206, at 207-08 ("In both vertical and horizontal operation, precedent creates the risk of entrenching erroneous rules. When today's court is compelled to accept yesterday's unsound decision, society incurs a loss from the perpetuation of the incorrect rule.").

(295.) See, e.g., Arizona v. Evans, 514 U.S. 1, 23 n.l (1995) (Ginsburg, J., dissenting) ("We have in many instances recognized that when frontier legal problems are presented, periods of 'percolation' in, and diverse opinions from, state and federal appellate courts may yield a better informed and more enduring final pronouncement by this Court."); United States v. Mendoza, 464 U.S. 154, 160 (1984) ("Allowing only one final adjudication [on a specified legal issue] would deprive this Court of the benefit it receives from permitting several courts of appeals to explore a difficult question before this Court grants certiorari."); Doni Gewirtzman, Lower Court Constitutionalism: Circuit Court Discretion in a Complex Adaptive System, 61 Am. U. L. Rev. 457, 482-83 (2012) (discussing potential benefits of percolation).

(296.) See, e.g., RICHARD A. POSNER, THE FEDERAL COURTS: CRISIS AND REFORM 163 (1985) ("[A] difficult question is more likely to be answered correctly if it is allowed to engage the attention of different sets of judges deciding factually different cases than if it is answered finally by the first panel to consider it.").

(297.) See Michael C. Dorf, Foreword: The Limits of Socratic Deliberation, 112 HARV. L. REV. 4, 65 (1998) ("[T]he passage of time during which there is a circuit split [may] create[] a record of the consequences of different legal regimes.").

(298.) Gewirtzman, supra note 295, at 482-83.

(299.) See, e.g., Kornhauser & Sager, supra note 53, at 45 ("Given the doctrinal disarray that leads to plurality opinions ..., lower court judgment, experience, and argument are especially useful to the high court.").

(300.) See supra Parts I.B.2 (describing the fifth vote approach), I.B.3 (describing the issue-by-issue approach), III.A (describing the failure of the narrowest grounds rule to constrain the Court's own decisionmaking).

(301.) See Adrian Vermeule, Foreword: System Effects and the Constitution, 123 HARV. L. REV. 4, 6-8 (2009) (describing the fallacy of division, which involves the assumption that what is true of a group must also be true of its members).

(302.) See supra note 50 (collecting sources identifying circuit splits produced by different understandings of the Marks doctrine).

(303.) 559 U.S. 393 (2010).

(304.) Id. at 396 & n.2 (quoting Fed. R. Civ. P. 23(b)).

(305.) The Erie doctrine requires that federal courts sitting in diversity apply state substantive law and federal procedural law. Gasperini v. Ctr. for Humanities, Inc., 518 U.S. 415, 427 (1996); see also Erie R.R. Co. v. Tompkins, 304 U.S. 64,78-80(1938).

(306.) Shady Grove, 559 U.S. at 398 (plurality opinion) (first citing Burlington N.R.R. Co. v. Woods, 480 U.S. 1, 4-5 (1987); and then citing Hanna v. Plumer, 380 U.S. 460, 463-64 (1965)); see also id. at 417 (Stevens, J., concurring in part and concurring in the judgment) (adopting the same framework); id. at 437-38 (Ginsburg, J., dissenting) (acknowledging this framework as controlling).

(307.) Shady Grove, 559 U.S. at 398-99. Justice Stevens explicitly joined in the portion of the plurality opinion reflecting this discussion, rendering that portion of the opinion the official position of the Court. See id. at 416 (Stevens, J., concurring in part and concurring in the judgment).

(308.) 312 U.S. 1(1941).

(309.) Shady Grove, 559 U.S. at 407 (plurality opinion) (quoting Sibbach, 312 U.S. at 14).

(310.) Id. at 416 (Stevens, J., concurring in part and concurring in the judgment). Critical to Justice Stevens's conclusion that Rule 23 could be properly applied to the facts of the parties' dispute was his view that New York's restriction on class certification had been adopted for procedural, rather than substantive, reasons. Id. at 436.

(311.) 28 U.S.C.[section] 2072(b) (2015).

(312.) Shady Grove, 559 U.S. at 428 (Stevens, J., concurring in part and concurring in the judgment).

(313.) See Catherine T. Struve, Institutional Practice, Procedural Uniformity, and As-Applied Challenges Under the Rules Enabling Act, 86 NOTRE DAME L. REV. 1181,1185-90 (2011).

(314.) Shady Grove, 559 U.S. at 446-47 (Ginsburg, J., dissenting).

(315.) Id. at 449.

(316.) Id. at 452-58.

(317.) Justice Ginsburg's dissent arguably signaled that Justice Stevens's position was closer to her own than was Justice Scalia's by noting that Justice Stevens had "stake[d] out common ground" with the dissenters regarding the need to read the federal rules with sensitivity to state interests. Id. at 442 n.2.

(318.) See, e.g., James River Ins. Co. v. Rapid Funding, LLC, 658 F.3d 1207,1217 (10th Cir. 2011) (concluding that Justice Stevens's concurrence was the controlling narrowest grounds opinion under Marks); Davenport v. Charter Commc'ns, LLC, 35 F. Supp. 3d 1040, 1050 (E.D. Mo. 2014) (stating that although the Eighth Circuit had not yet adopted Justice Stevens's opinion, "the majority of federal courts to consider the issue have found that Justice Stevens' opinion controls"). But see Abbas v. Foreign Policy Grp., 783 F.3d 1328, 1336-37 (D.C. Cir. 2015) (concluding that none of the opinions in Shady Grove is controlling for Marks purposes).

(319.) Though Justice Ginsburg's dissent did not explicitly address the validity question that divided Justice Stevens from the plurality, at least one commenter has understood the dissenters' refusal to join in Justice Stevens's opinion as an implicit rejection of his validity analysis and an endorsement of the plurality's more "conventional" approach. See Kevin M. Clermont, The Repressible Myth of Shady Grove, 86 NOTRE DAME L. REV. 987, 1015 (2011) (interpreting the dissent to have "implicitly assented to the plurality's understanding" of Sibbach).

(320.) Shady Grove, 559 U.S. at 411-12 (plurality opinion).

(321.) Id. at 412.

(322.) Compare, e.g., Clermont, supra note 319, at 1014 n.135 (concluding that "the Sibbach Court" had effectively "rejected" Justice Stevens's view that a rule "deemed procedural at the federal level still must not impinge on matters deemed substantive at the state level"), with, e.g., Struve, supra note 313, at 1190 (contending that Sibbach does not foreclose as-applied review of rule validity).

(323.) See, e.g., James River Ins., 658 F.3d at 1217-18 (following Justice Stevens's Shady Grove opinion without mentioning Sibbach); Davenport, 35 F. Supp. 3d at 1050 (doing the same); McKinney v. Bayer Corp., 744 F. Supp. 2d 733, 747 (N.D. Ohio 2010) (doing the same); Bearden v. Honeywell Int'l Inc., No. 3:09-1035, 2010 WL 3239285, at *10 (M.D. Tenn. Aug. 16,2010) (doing the same).

(324.) The few post-Shady Grove decisions that have specifically addressed the continuing significance of Sibbach have reached differing conclusions. Compare, e.g., Abbas v. Foreign Policy Grp., 783 F.3d 1328, 1337 (D.C. Cir. 2015) (concluding that Sibbach requires wholesale assessment of rule validity), with, e.g., Phillips v. Philip Morris Cos., 290 F.R.D. 476, 481 n.4 (N.D. Ohio 2013) (denying the applicability of Sibbach by following Justice Stevens's more limited reading of Sibbach).

(325.) Shady Grove, 559 U.S. at 415 n.14 (plurality opinion).

(326.) Id. at 426 n.10 (Stevens, J., concurring in part and concurring in the judgment).

(327.) See, e.g., Aaron-Andrew P. Bruhl, Following Lower-Court Precedent, 81 U. Chi. L. Rev. 851, 868 (2014) ("One would expect that... the average federal or state trial judge probably understands the consequences of various procedural and evidentiary rules better than the justices, most of whom have little or no trial experience.").

(328.) See id. at 873 (demonstrating how differing understandings of the relevant question to be answered at the Supreme Court level and the lower court levels can impair the epistemic value the Court might glean from the lower courts' decisions).

Ryan C. William, Assistant Professor, Boston College Law School. My thanks to William Baude, Vincent Blasi, Pamela Bookman, Samuel Bray, Richard Briffault, Bernard Harcourt, Rebecca Ingber, Olatunde Johnson, Henry Monaghan, James Pfander, David Pozen, Richard Re, Alan Trammell, and Maggie Wittlin. My thanks as well to participants in workshops at Boston College Law School, Columbia Law School, the University of Richmond School of Law, and St. John's University School of Law, as well as the 2015 Junior Faculty Federal Courts Workshop.

Caption: Figure 1 Decision Involving Logically Nested Rationales

Caption: Figure 2 Decision Involving Partially Overlapping Rationales
COPYRIGHT 2017 Stanford Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Williams, Ryan C.
Publication:Stanford Law Review
Date:Mar 1, 2017
Words:33318
Previous Article:The downstream consequences of misdemeanor pretrial detention.
Next Article:Lost in the cloud: cloud storage, privacy, and suggestions for protecting users' data.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |