Which Splits?—Certiorari in Conflicts Cases
[Editor’s Note: We recommend viewing this article in PDF form due to the number of figures and tables within the article.]
The Supreme Court is well-known to favor granting review in cases implicating circuit splits. When, for example, two federal appeals courts disagree over the meaning of a federal statute, the Supreme Court is likely to step in and resolve the confusion to ensure uniformity in federal law.
But the Court is also increasingly likely to let such splits languish for longer. It is taking fewer and fewer cases, year after year. And the Court dedicates much of the limited space on its docket to cases that do not involve circuit splits—cases that, say, present an opportunity to overrule precedent or that implicate patent matters.
These trends give rise to several questions: If the Supreme Court grants review to resolve only certain limited circuit splits, which splits is the Court likely to resolve? Conversely, which splits are likely to languish? And how, if at all, have these trends changed over time?
Answers to these questions—answers unearthed using computational text-analysis methods—have important implications not only for questions of uniformity (and associated rule-of-law values) but also for questions of Supreme Court reform. Specifically, a Court that is more focused on uniformity in federal law (and less focused on remaking our constitutional order) may be less politically volatile, and so we should perhaps require the Supreme Court to address more statutory splits through an invigorated certification process.
Table of Contents Show
Introduction
The Supreme Court is well-known to favor granting review in cases implicating a split in authority among, say, the federal courts of appeals.[1] This is because of a longstanding view that federal law should be uniform—that, for example, a federal statute should mean the same thing in both Kansas City, Kansas and Kansas City, Missouri.[2] Commentators and Justices, from Alexander Hamilton to Chief Justice Roberts, have described the Court’s “main job” as “to unite and assimilate the principles of national justice and the rules of national decisions.”[3] And many of the Court’s own decisions reflect a view that it is obligated to establish a uniform federal law.[4] Some Justices have traced this duty to the 1925 Judges’ Bill, which gave the Court effective control over much of its docket by way of the writ of certiorari, concluding that Congress there granted the Court greater docket discretion in exchange for an implicit promise to ensure uniformity in federal law.[5]
But even if we understand the Court to favor granting review in cases implicating conflicts among the state and federal appeals courts, several questions about the Court’s agenda-setting process remain unanswered. Deborah Beim and Kelly Rader are among the latest scholars to observe that the Supreme Court does not resolve all conflicts about the meaning of federal law.[6] Indeed, Margaret Meriwether Cordray and Richard Cordray once described a proliferation of circuit splits well beyond the size of the Court’s docket.[7] And so we might wonder which splits: which splits will be resolved?; which splits will languish?; and how have the Court’s priorities evolved over time?
Answering these questions is more than a mere academic exercise.[8] The shape of the Supreme Court’s docket can have profound legal, practical, and political implications.[9] An unresolved split means that similar parties are treated differently under federal law for arbitrary reasons, such as geography.[10] And it also matters whether the Court’s agenda involves resolving a series of circuit splits on federal tax and bankruptcy law, or whether it is preoccupied with resolving tensions on the interpretation of the Constitution’s First and Second Amendments.[11]
It is thus important to study and understand the Court’s conflicts docket—the slice of the Court’s entire agenda where review is granted to resolve splits or conflicts[12] among state and federal appellate courts and thereby advance uniformity interests. Such an understanding may help to confer (on both petitioners and lower court judges) greater transparency and predictability in the Court’s certiorari decisions and may even help to inform political reforms that address concerns about the Court’s legitimacy.
My analysis of the conflicts docket reveals two notable findings, with implications for matters of Supreme Court reform. First, the Court’s conflicts docket pays special attention to questions of statutory construction and interpretation. Hence, Congress exercises a surprising degree of indirect control over the Court’s conflicts docket: When Congress legislates, its enactments frequently pose interpretative questions that eventually wend their way to the Supreme Court. Second, while the Court’s agenda has focused on different statutes at different times (reacting, for example, to congressional action), the Court’s conflicts docket remains steady in view of changes to the Court’s personnel or expansions to the Court’s docket-setting discretion. Once an issue area secures the Court’s attention, it is wont to retain that focus. For example, even though the Justices might prefer to avoid tax matters (and such cases are, to be sure, less common than they once were), they continue to comprise a significant portion of the Court’s conflicts docket. By contrast, the Court’s important-questions docket—which focuses much more substantially on constitutional questions—is far more responsive to changes in the composition of the Court. That is, individual appointments matter a great deal to which constitutional questions the Court will address, or which precedents it will overturn, thereby causing the Court to appear more political (and perhaps less legitimate). But that is much less so for the questions of statutory meaning on the Court’s conflicts docket.
These conclusions draw upon the text-analysis (i.e., co-word analysis) methods that I have used in prior studies of the Supreme Court’s agenda-setting decisions, and that I reprise here. Specifically, the Court’s merits opinions frequently offer a brief and oft-overlooked account of its decision to grant review. These “certiorari paragraphs” can offer a meaningful peek into the priorities that inform the Court’s certiorari decisions.[13] And though these certiorari paragraphs can be brief and thinly reasoned, their sheer volume across nearly a century of certiorari decisions, together with advances in the computational analysis of legal texts, offer an opportunity to understand how the Court’s agenda-setting priorities have evolved over time.[14] Hence, viewed in aggregate, these paragraphs elaborate on which splits matter and (perhaps by implication) which may not.
Compared to my prior work, which employed this approach to study the Court’s certiorari decisions in its important-questions docket,[15] this Article examines those cases where uniformity is expressly at stake, i.e., its conflicts docket.[16] As noted, in some cases the Court has cited a special need for uniformity in tax cases, ensuring equal administration of the tax system;[17] in other cases, the Court has explained that conflicts about a statute’s extraterritorial application or the permissible scope of appellate review call out for review.[18] Hence, examining the Court’s certiorari decisions in its conflicts docket helps us understand when the Court seems to think uniformity matters most.[19] Moreover, comparing the findings in this study to those in my prior works may allow us to say with greater clarity which sorts of cases the Court is likely to hear only for the sake of ensuring uniformity, or only for their particular importance, i.e., for addressing an “important question.”[20]
This Article thus adds to, and builds upon, the existing literature on certiorari and the Supreme Court’s agenda-setting decisions. Michael Livermore, Allen Riddell, and Daniel Rockmore, for example, compared the federal appeals courts’ collective caseload to the Supreme Court’s docket to discern whether the Court grants review in an idiosyncratic set of cases. It does, suggesting that the splits that matter (for certiorari purposes) are not merely a representative sample of the cases before the courts of appeals and the concomitant conflicts that may arise as those courts carry out their responsibilities.[21] Deborah Beim and Kelly Rader likewise study a large sample of circuit splits to discern what factors are most likely to flag a conflict for the Court’s attention, finding (among other things) that those splits “that will be resolved are resolved within three years . . . and that splits are more likely to be resolved when they exhibit contemporaneous and growing disagreement.”[22] And these studies are only among the latest in a line of scholarship that considers the relationship between the Court’s agenda and uniformity values.[23]
This Article contributes to these literatures on certiorari and the Supreme Court’s agenda by advancing our understanding of the Court’s approach to uniformity across several dimensions, including across legal fields (e.g., tax law versus civil rights claims) and over time (e.g., the Rehnquist Court versus the Roberts Court). Indeed, because this Article studies the Court’s entire conflicts docket from 1925 to 2022, it offers insight into how the Court’s priorities have evolved since the Taft Court. Moreover, by extending the implementation of this novel method of analyzing the Court’s certiorari decisions from important-questions cases to the Court’s conflicts agenda, this Article adds to the growing literature on the data-driven analyses of legal materials (such as opinion text).[24]
This Article proceeds in three parts.
First, I describe my approach to understanding how the Court sets its conflicts docket. I begin by describing how the Court’s longstanding focus on uniformity has shaped its approach to agenda-setting. I then describe my method for evaluating the Court’s agenda.
Second, I present the results of my analysis. As in my earlier studies, these results are based upon measures of significance for a specialized dictionary of terms, such as tax*,[25] that appear in the Court’s descriptions of its certiorari grants. Specifically, I describe an “Importance Score” for each such term (indicating the strength of that term’s correlation to the Court’s certiorari decisions in its conflicts docket),[26] as well as a measure for changes (Deltas, or Δ) in such scores over time. And, as before, my analysis of these results relies on both these quantitative scores and a complementary close reading of the opinions that give rise to these scores. This analysis uncovers two notable findings, noted above and elaborated at length below. First, the Court’s conflicts docket pays special attention to questions of statutory meaning, with a particular emphasis on statutes that apply widely across the economy and population. Hence, Congress exercises a surprising degree of indirect control over the Court’s conflicts docket. Second, the Court’s conflicts docket is less responsive, as compared to its important-questions docket, to changes in the Court’s personnel.
Third, in view of these noteworthy outcomes, I consider possible reforms to the Court’s agenda-setting discretion. I conclude that the relative constancy in the Court’s conflicts docket is good: It reflects, in this slice of the Court’s agenda, stability and predictability on an important matter of Supreme Court practice and procedure. It is good for judicial legitimacy that the Court’s agenda—at least in this respect—does not vacillate more strongly in response to new appointments. I thus consider ways to make the Court’s certiorari procedures more transparent and to make the Court’s conflicts docket a bigger part of its agenda. Specifically, I reiterate my call for the Court to develop a common law of certiorari, like other procedural doctrines. And I call for the federal appeals courts to certify cases implicating statutory conflicts to the Supreme Court for mandatory review.
I. The Importance of Uniformity
A. One Hundred Years of Certiorari
1. Certiorari Jurisdiction
Until 1891, the Supreme Court was required to review every case arising under its appellate jurisdiction (including, e.g., a wide range of diversity-jurisdiction cases).[27] And though the scope of that appellate jurisdiction was somewhat narrower than the class of cases the modern Court may hear[28]—the Court’s workload was crushing. In 1890, the Court’s docket encompassed over 1,800 cases,[29] or about thirty times more than the Court’s current caseload.[30] But Members of the Court indicated that they could realistically handle only about one-quarter of that amount.[31]
In response, Congress enacted the Judiciary Act of 1891.[32] The 1891 Act created federal appellate courts, granting them jurisdiction over most appeals.[33] And while the Supreme Court retained mandatory appellate jurisdiction over subsequent appeals in most federal-questions cases, the Court was granted discretionary jurisdiction—exercised via writs of certiorari—to ensure uniformity in certain classes of cases, including diversity-jurisdiction cases, as well as various federal admiralty, criminal, patent, and tax cases.[34] Hence, Congress has designated the need for uniformity as an important factor in the construction of the Court’s tax docket (and its patent docket, and its admiralty docket, among others) for well over a century. Overall, this revised institutional design for the Federal Judiciary was meant to manage the Supreme Court’s workload while ensuring that the Supreme Court could provide uniformity in federal law.[35]
Congress’s plan was, at best, only partly successful, as the Court’s workload continued to grow.[36] In 1925, Congress consequently replaced most of the Court’s mandatory appellate jurisdiction with discretionary certiorari jurisdiction.[37] Uniformity remained at the fore of Congress’s concerns.[38] Indeed, some Justices understood Congress’s decision to grant the Court this docket discretion as premised on an implicit (or perhaps explicit[39]) promise to ensure uniformity in federal law.[40]
In the one hundred years since the Judges’ Bill of 1925, the scope of the Court’s discretionary jurisdiction has steadily expanded, so much so that, over thirty-five years ago, three leading experts on Supreme Court practice and procedure co-authored an “epitaph” for the Court’s mandatory jurisdiction.[41] But even as the Court’s discretion has expanded, uniformity has remained a guiding light for certiorari.[42] In all, while the Court’s docket is characterized by discretion, that discretion is not, in practice, entirely unbounded: For as long as the Court has had the ability to choose the cases it decides, that agenda-setting power has come with a tacit instruction to ensure uniformity in federal law.
I do not mean to overstate the case: The Court has—and has exercised—a wide amount of discretion in setting its agenda to, say, grant review for reasons other than uniformity. In prior work, for example, I found that the Roberts Court was uniquely inclined to grant review for the purposes of revisiting (and potentially overruling) the Court’s own precedents.[43] Such cases cannot implicate a split among the state and federal appellate courts, since only the Court may overrule its precedents.[44] The modern Supreme Court’s docket is thus characterized by discretion. It is essentially free to choose the cases it will decide.[45] The Supreme Court’s “exercise of this discretion is a matter of great practical consequence and scholarly interest.”[46] Former Justices, for example, have explained that case selection “is second to none in importance” because the decision to take a case often both reflects contemporary concerns and, significantly, helps “shap[e] the nation’s political, social, and economic agenda.”[47] Moreover, the Court’s agenda-setting decisions may help to reveal its institutional values, as each certiorari decision hints at the Court’s “subjective notions of what is important or appropriate for review.”[48]
Hence, the Court’s emphasis on splits highlights its view on the importance of uniformity.[49] The Court has repeatedly noted that its “principal responsibility” is generally to ensure uniformity in federal law.[50] This is because uniformity helps to protect the legitimacy of the law and of the federal courts by avoiding a public impression of arbitrariness; it facilitates predictability; it mitigates forum shopping; and it reduces costs for multistate actors.[51] About half[52] of the Court’s entire docket consists of cases presenting splits among the state and federal appellate courts,[53] and conflicts are consistently identified as the single most important determinant of the likelihood of Supreme Court review. Moreover, the Court’s selection of which splits to review may tell us more about when the Court is most concerned about disuniformity. As noted, Congress has emphasized uniformity in the construction of the Court’s tax docket (among other issue areas) for well over a century. And following Congress’s lead, the Court has repeatedly emphasized the importance of the “uniform application of the Internal Revenue laws.”[54]
The fact that conflict seems to be the single most significant input to the Court’s decision whether to grant review is only a starting point. While Supreme Court Rule 10 (among many other sources) identifies conflict as a primary criterion for the Court’s certiorari decisions, it does only a little to elaborate further (as in the tax example noted above).[55] And, as scholars have noted, the Court’s “shrinking docket” seems to leave an ever-growing number of circuit splits to fester.[56] Hence, the Court appears to choose some conflicts for review and leave others to languish.
2. Scholarly Studies
So what explains the Court’s decisions to grant review in some cases and deny review in others? Scholars have advanced an array of explanations encompassing ideology, timing, subject matter, and other considerations (e.g., the presence of a dissenting opinion, the number of circuits implicated by a split, and the willingness of the Executive Branch to weigh in on the matter).[57]
On ideology, several authors have suggested that a wide ideological gap between decisions in a split is more likely to precipitate Supreme Court review.[58] Some authors have also concluded that the Court is likely to grant review in cases that both present a conflict among the lower courts[59] and conflict with the Court’s ideological preferences.[60] Put otherwise, these scholars found that if one side of a conflict is ideologically far from the Court’s own views, the Court is likely to grant review to bring the law “closer to its own ideological preferences.”[61]
Scholars have also noted the importance of timing to the Court’s certiorari decisions. Arthur Hellman, for example, explained nearly thirty years ago that the Court is more likely to deny review when a respondent explains that an issue “is not yet ripe for definitive resolution”—that is, when the issue has not had enough time to percolate among the circuits and would benefit from more information through further legal analysis and experience.[62] Conversely, he explained that a conflict is likely to be ignored if it is “not one of continuing importance.”[63] More recent work by Deborah Beim and Kelly Rader found that the sweet spot—neither too early nor too late—is for a split to develop within three years on a given issue.[64] This is particularly so if the split consists of a “flurry of decisions . . . such that there is conflicting and simultaneous development of doctrine” that “spread[s] beyond” the two circuits to first disagree.[65] Hence, though the Court may occasionally agree to address a conflict that has lingered for twenty-eight years,[66] some conflicts might seem destined to remain unresolved forever.[67]
Some scholars have also noted differences across issue areas. Karen Gebbia, for example, found that if the Court is to grant review in a case involving the bankruptcy code, it will only do so in the context of a split (though that hardly indicates the Court’s willingness to do so even given a split).[68] Others found that the Court is especially likely to grant review in conflicts over constitutional questions[69] or civil liberties matters.[70] Matthew Fletcher found that the Court is likely to deny petitions for certiorari filed on behalf of tribal interests.[71] And Deborah Beim and Kelly Rader found that “splits over issues involving Economic activity are about 31% more likely to be resolved than those involving Criminal procedure[,]”[72] though cases arising out of prisons might have once stood a better chance of earning review.[73]
Several of these conclusions give rise to testable hypotheses[74]: Does other evidence corroborate the view that the Court disfavors criminal procedure conflicts? Or that a dissenting opinion can shine a spotlight on an important split?[75] Can we say for sure that prisoner or criminal cases were more important to the Court before than they are now?
The Court’s apparently disuniform interest (across issue areas, say) in ensuring uniformity (across geographies on matters of federal law) thus calls for closer scrutiny.[76] We need more complete and more current information on which splits matter and which do not. Such information is important, at minimum, for predictability. If we know only that splits matter, then we may be surprised when the Court declines to hear a split over, say, the application of the Fourth Amendment’s protections against unreasonable searches.[77] But if we know that tax splits matter because of a special interest in uniform tax administration[78] and that disuniformity in matters of criminal procedure is more tolerable to the Court (for whatever reason), then we may be able to make greater sense of the Court’s docket-setting decisions. Such information, moreover, can help guide the decisions of litigants, who may be more or less willing to invest in seeking Supreme Court review—a potentially expensive proposition—depending on how their own case fits into such understandings of the Court’s practices.
Better information about the Court’s agenda-setting practices would also help enable greater democratic oversight and accountability for the Judiciary. If, for example, Congress believes that uniformity is especially important in patent cases, but the Court has consistently declined to grant review in such cases, then legislators can amend the Judiciary’s structure accordingly. Congress might move patent cases back into the Court’s mandatory appellate jurisdiction; or it might create a specialized intermediate appellate court with jurisdiction over all patent appeals. Indeed, as some readers may know, that is precisely what happened: Congress created the U.S. Court of Appeals for the Federal Circuit to ensure uniformity in patent law and in response to frustration over the Supreme Court’s inability (or unwillingness) to do that job itself.[79]
3. Turning to Opinion Text
How, then, might we uncover this information, important for predictability and accountability purposes, regarding the Court’s agenda-setting practices?
As noted above, the Court frequently sets out the basis for its certiorari decisions in its final merits opinions, having done so since Congress enacted the 1925 Judges’ Bill.[80] In one case during the Taft Court, for example, the Court explained that “the writ of certiorari was granted” in view of a conflict between the Sixth and Third Circuits on the validity of “the same patent.”[81] During Chief Justice Hughes’s tenure, the Court agreed to review a set of cases “on writs of certiorari to resolve a conflict . . . with respect to the application of section 101 of the Revenue Act of 1928,” regarding the taxation of inherited property.[82] In Armour & Co. v. Wantock, a companion to the more famous Skidmore v. Swift & Co., the Stone Court explained that it agreed to grant review “to resolve the conflict” on the meaning of overtime pay under the Fair Labor Standards Act of 1938.[83] The Vinson Court explained that it granted review in one case to address a “conflict” on the interaction between the Federal Employees Compensation Act and the Public Vessels Act of 1925.[84] Chief Justice Warren explained that the Court agreed to review one case to resolve a conflict over the meaning and effect of Federal Rule of Criminal Procedure 11.[85] The Burger Court “granted certiorari to resolve [a] conflict” over when “a delay in instituting forfeiture proceedings violates due process.”[86] The Rehnquist Court “granted certiorari to resolve a conflict among the Circuits as to the scope of the Minturn decision and to consider whether Minturn should be overruled.”[87] And, in Thompson v. Clark, the Roberts Court granted certiorari to resolve a “split over how to apply the favorable termination requirement of the Fourth Amendment claim under § 1983 for malicious prosecution.”[88]
In short, the Court’s opinions contain some important details regarding the splits that have earned its attention. And these descriptions, insofar as they reflect the Court’s decisions to grant review, do more than simply explain that the Court resolves conflict to ensure uniformity: They help to describe which issue areas are more salient (and, in aggregate, how that salience may vary over time); and they may even offer insight into the Court’s other motivations—its willingness, say, to use the occasion of a conflict to overrule precedent.[89]
B. Method for Understanding Certiorari in Conflicts Cases
The Court’s opinions thus offer a window into its docket-setting priorities. And by treating the Court’s decisions as data[90]—by using computational tools to analyze the portions of the Court’s opinions describing its decision to grant certiorari—we can understand both how the Court has set its conflicts docket and how that docket has changed over time. I offer a brief summary of my approach for doing so, based in methods of co-word analysis, in the following sections. I begin with my data sources and subsequently turn to my methods for analysis. I also offer more detail both in the Methods Appendix to this Article and in the documentation to my similar prior work. Indeed, because this approach is substantially similar to that which I have employed in my previous studies of certiorari (focusing on the important-questions docket), the description of my methodology is abridged, both citing extensively to earlier, more detailed descriptions of the method and emphasizing the key differences between my present study of conflicts cases and my previous studies of important-questions cases.[91]
1. Dataset
The dataset begins on October 5, 1925, the first day of the first complete Term (October Term 1925) after the enactment of the 1925 Judges’ Bill.[92] I rely on the Supreme Court Database (developed and maintained at the Washington University School of Law) to identify every case arising under the Court’s certiorari jurisdiction since October 5, 1925. Moreover, the Supreme Court Database codes (in general terms) the reason for certiorari, if any, given in the Court’s opinion. I use these coded values to focus attention on those cases granted to resolve a “conflict,” a “putative conflict,” or “confusion or uncertainty” among the state and federal courts, as well as those cases coded as granted for no specified reason (but later limited to only those whose terms are suggestive of a place on the Court’s conflicts docket).[93]
This gave rise to a list of 6,673 cases through October Term 2021: 2,273 cases granted to resolve some form of a conflict or confusion (under the Database’s coding methodology) and 4,400 cases granted for no apparent reason (but, as noted, later limited to those whose text indicates that a conflict was associated with the decision to grant review). Figure 1 describes how these categories of cases—those with some explanation, and those with none—have evolved over time.
Figure 1. Reason for Certiorari (in Percent) by Chief Justice, 1925–2021 Terms
Next, with the help of research assistants, I obtained access to the full text of those 6,673 opinions from the Caselaw Access Project, CourtListener, and Justia (and, in one case, by manually converting a PDF file from the Supreme Court’s website to plain text).
Finally, I sharpened focus to the sections of those opinions that discussed the Court’s reason for granting certiorari—statements like those highlighted above—by selecting for discrete paragraphs that contain the term “certiorari.”[94] Here, I took advantage of the Court’s fairly consistent approach to opinion writing—an approach that, as noted, typically includes a summary of the case’s procedural history that concludes with the Court’s decision to grant certiorari.
As Justice Frankfurter once explained, this “helpful practice of explaining, in the ultimate opinion on the merits, the reasons which had moved it to grant the writ” helps to clarify “the canons which guide the Court’s exercise of [its docket-setting] discretion.”[95] Indeed, even cases coded in the Supreme Court Database as lacking an explicit rationale for the certiorari grant sometimes include a paragraph noting the Court’s decision to grant the certiorari petition—and those paragraphs occasionally imply some foundation for the Court’s review by, for example, gesturing at the case’s subject matter or identifying some other notable feature (like a circuit split).[96]
Moreover, these descriptions of the Court’s decisions to grant review can offer information—particularly when analyzed in the aggregate—that is not easily gleaned from other sources (such as datasets organizing cases by issue area or circuit court). To elaborate on a finding described above: My earlier work noted the Roberts Court’s inclination to grant review for the purpose of revisiting precedent (even, in some cases, where it ultimately decided to leave settled law alone).[97] Hence, a study that examined the Court’s cases only by issue area or number of precedents overruled alone might overlook some notable outcomes.
The decision to focus on paragraphs containing the term “certiorari” is not, of course, without trade-offs. As I have explained before,[98] this approach is both underinclusive and overinclusive. It is underinclusive because not all opinions include such paragraphs (though about 88 percent of the opinions examined here do).[99] And it is overinclusive because some decisions may contain multiple paragraphs containing the term certiorari, including, for example, a complaint from a dissent or concurrence about that certiorari grant or the manner in which that case was decided.[100] But these risks are tolerable, in part because this approach helps to address other concerns that alternate methods might introduce. For one, the overinclusivity introduced by this approach (incorporating all certiorari paragraphs, including those in separate opinions) may offer additional insights about the Court’s approach to certiorari at any given time. For example, a dissent’s discussion of a decision to grant review may provide some insight into the Court’s docket-setting priorities.[101] And the underinclusivity inherent to this approach helps avoid confounding the certiorari-centered analysis with content that is far afield from the Court’s decision to grant review (e.g., when the Court grants review to decide a question arising under the Freedom of Information Act but then decides the case on unrelated grounds of mootness).[102]
Hence, as detailed in my prior work, I ultimately concluded that the benefits of limiting focus to any paragraph containing the term “certiorari” were likely to outweigh the costs of this approach. And so this collection of paragraphs provides the dataset for further analysis.
2. Analysis
In order to discern which splits have earned the Court’s attention, I began with a list of 163 distinct terms that appeared most frequently in the set of paragraphs isolated from the Court’s opinions. I consolidated and distilled these terms into a Term Index of 88 distinct Index Terms.[103]
I then scored each Index Term according to two metrics: frequency (i.e., in how many cases did the term appear?) and proximity (i.e., how close was the term to opinion statements describing the Court’s decision to grant certiorari?). My emphasis on these metrics is informed by both domain-specific knowledge and general principles of natural language processing. Data scientists rely on both frequency and proximity metrics to approximate the meaning of terms in a corpus (or body of text).[104] And these general data science insights are reinforced by those that lawyers and scholars might intuit from reading a number of the Court’s opinions: A series of statements, across a number of opinions, explaining that uniformity is especially important in, say, tax law suggests that frequency matters. And the proximity of an Index Term (such as tax* or patent)[105] to words such as “conflict” or “certiorari” is suggestive of that term’s salience to the decision to grant review in that case.[106] Together, then, an Index Term’s frequency and proximity metrics form its “Importance Score.”[107]
To specify further, a term’s frequency measure is a scaled representation of how often that term appears in a given time period, with 0 representing the most frequent term and 1 representing the least frequent. Likewise, a term’s proximity measure is a scaled representation of how close that term is to certain key phrases (i.e., Focal Terms), such as we_granted_certiorari. Collectively, certiorari*, question*, and conflict* comprise these Focal Terms.[108] Hence, the origin point (0, 0) can be understood as representing certiorari* (since it is, by definition, in each case represented in the dataset, and no other term can be closer to it than itself) and every term’s distance from the origin is a representation of its significance to the Court’s decision to grant certiorari. Smaller distances from the origin thus suggest greater significance.[109]
3. Limitations
As I have explained in prior work, the method of analysis I employ here has some important limits. For one, this approach offers a perspective on the Court’s docket-setting decisions that is primarily internal to the Court: It relies upon—taking as given—the Court’s own stated approach to its docket discretion, rather than upon external signals, such as the content of petitions, effects on law and policy, or outside observers’ views of the Court.[110] I use this approach because I am primarily interested in understanding the principles that guide the Court’s own exercise of its docket-setting discretion and how those principles have evolved over time. In short, I am interested in the Court’s unstated common law of certiorari (to the extent one exists),[111] and I attempt to discern that hidden doctrine with an analysis of term frequency and proximity.[112] Moreover, the method described above is largely quantitative—though, as in my other prior works, I supplement my quantitative results with a close reading of the text that underlies the results I present below.[113]
I emphasize, again, that I do not mean for my methodological choices to diminish the importance of alternate approaches. The Court can significantly affect matters of public policy by granting certiorari in an area of law that it is otherwise prone to ignore (i.e., that would score poorly on my frequency metric).[114] Moreover, scholars of the Court have attempted to predict the Court’s docket-setting decisions through a close examination of petitions for certiorari.[115] While my approach may not capture these signals, it may identify other factors that shape the Court’s docket. Hence, studies that employ a range of methods advance our understanding of the Court and its public role, and I simply mean to acknowledge that the method employed here has a particular focus.
My method also has further limits inherent to its design. For one, there is an element of subjective selection in my construction of the Term Index, one which is unavoidable no matter the approach employed.[116]
Moreover, my study of the Court’s preferences is shaped by available petitions, excluding those cases where no petition for certiorari was filed. While this is a formal limit on my study, I do not believe that it is a particularly significant one. Though the Court is constrained by the petitions that are presented to it, the Court has a wide range of petitions from which to select, and it has a variety of mechanisms by which to signal its interest in receiving a particular petition.[117] Even if one side of a split may hesitate to bring a case to the Court (say, because it thinks it likely that the Court will rule against it and enshrine an unfavorable rule nationally), every split has at least two sides, and so the other side is likely to be motivated to bring the issue to the Court’s attention.
More notably, my study focuses on the Court’s selectively reasoned decisions to grant review—to the substantial exclusion of the Court’s almost-entirely unreasoned decisions to deny review, except in the comparatively rare opinions concurring in or dissenting from the Court’s denial of the writ of certiorari (or what the Court calls “Opinions Relating to Orders”).[118] The absence of reasoned opinions in most decisions to deny review limits my ability to analyze why some splits are left to fester for so long. I cannot compare grants to denials. This is an unavoidable consequence of the Court’s practice of declining to explain certiorari-stage decisions independent of merits-stage decisions. As noted, I attempt to mitigate this by including any available text explaining a decision to deny review, and this limitation—which would affect any such study of the Court’s docket-setting priorities—may suggest one further reason to urge the Court to better explain its certiorari-stage decisions.[119]
Finally, I emphasize that these results are descriptive of the Court’s behavior from 1925 to 2021, but they are not necessarily predictive. Indeed, the Court’s substantial docket discretion includes the power to significantly remake its certiorari practices in the future.
* * *
Though the Supreme Court does not independently explain its certiorari-stage decisions, it does often explain its decisions to grant review in its merits opinions. These explanations can be analyzed using modes of text analysis and language processing. Here, I have developed and employed a method generating a two-dimensional score (frequency, proximity) for each in a list of salient terms that may help to describe the Court’s decision to grant review where its docket-setting discretion is greatest. In short, this text analysis can help us better understand how the Court selects the splits that it is likely to address and the splits that are likely to fester.
II. Which Splits?
Justice Breyer once explained that “the Supreme Court is charged with providing a uniform rule of federal law in areas that require one.”[120] But which areas require one? And have these priorities changed over time? One might imagine that the Court’s energies are focused on immigration, bankruptcy, and tax cases, given that the Constitution expressly articulates a need for uniformity on “rule[s] of naturalization”; “subject[s] of bankruptcies”; and “duties, imposts, and excises.”[121] Or, looking to the history of certiorari, one might imagine that the Court focuses on admiralty, criminal, patent, and (again) tax cases, given that Congress has emphasized the need for uniformity in these aspects of the Court’s docket for well over a century.[122]
Fortunately, we need not imagine. The Court’s opinions, to the extent we can accurately analyze them, can help us address these questions.
A. Results
1. Importance Scores
Which splits are likely to attract the Court’s attention? As noted above (and in prior work), I cannot say definitively. I can only make inferences based on frequency and proximity and limited by the accuracy of the various filters applied to my indices. But acknowledging those limits, here is a map of the Importance Scores for terms correlated to the Court’s conflicts docket since 1925. Specifically, the two dimensions of the plot reflect the two dimensions (frequency, proximity) of each term’s Importance Score.
Figure 2. Terms (All) Plotted by Importance Score, 1925–2021
Figure 2 plots each of the 88 Index Terms, based on its overall two-dimensional score, on a two-dimensional plane. The outer and inner arcs drawn on Figure 2 represent the median distance from the origin minus 1 and 2 (respectively) median absolute deviation(s) (over all terms across all times). Hence, terms inside the inner arc are more than 2 median absolute deviations away from the median, and terms inside the outer arc are more than 1 median absolute deviation away. In short, the terms inside the arcs are significantly more important than those outside. Figure 2A, below, limits focus to these terms.
Figure 2A. Terms (Significant) Plotted by Importance Score, 1925–2021
We can also rank these terms by their linear distance from the origin (imagine a line from each term to the lower-left corner), thereby converting the two-dimensional score into a one-dimensional score (one that weighs both factors equally). As described above and in the Methods Appendix, terms closer to the origin point (0, 0)—i.e., smaller Distances from Origin (DFOs)—are more suggestive of the Court’s priorities.
Table 1 presents only the list of terms that are inside the arcs described in Figure 2 (i.e., only the terms displayed in Figure 2A). Hence, this is a list of only significant terms.
Table 1. Terms (Significant) Ranked by Importance Score, 1925–2021
Some of these results help confirm findings in other work. Deborah Beim and Kelly Rader, for example, describe the importance of a dissent to the Court’s decision to grant certiorari.[123] The standing of the term divided_lower_court* corroborates that finding. I clarify that the term divided_lower_court* refers not to division among courts (i.e., conflict), but rather to division among members of a single court of appeals or state supreme court (e.g., a single dissent or a dissent from the denial of a petition for rehearing en banc).[124] Hence, the noteworthy presence of a dissent (i.e., a dissent that the Court sees fits to describe in a certiorari paragraph) is correlated with the Court’s decisions to grant review, perhaps because the division between the majority and the dissent reflects an existing conflict,[125] or perhaps because such a dissent can serve to highlight and sharpen the contrast between conflicting approaches. In short, the lower DFO score for divided_lower_court* indicates that noteworthy dissents may help prompt the Court to grant review, perhaps validating a strategy discussed by many appellate judges.[126]
The importance of statute* and related terms may also help us to resolve some confusion over the scope of the Court’s conflicts docket. Several Justices, for example, have commented on the relationship between conflicts over the meaning of statutory text and the Court’s workload. Chief Justice Roberts, for example, has attributed the Court’s declining docket to congressional inactivity, and Justice Breyer has explained that the Court’s docket grew in the 1960s in response to a contemporaneous “flurry of legislation.”[127] But Justice Stevens once wrote that Congress, and not the Court, should own the task of resolving conflicts arising out of statutory text.[128]
Here, statute*’s strong correlation with the Court’s decision to grant certiorari suggests that, over time, Justice Stevens’s view has not carried the day. Instead, questions of statutory interpretation and construction (e.g., statute*, interpretation, and construction*) appear to dominate the Court’s conflicts docket. Importantly, these terms alone do not say which statutes have occupied the Court’s time—though other terms, like labor*, and employ*, or sentence*, hint that these cases might implicate labor and employment statutes or provisions about criminal sentencing.[129]
Still other results may seem surprising. For example, several Justices have described tax cases as uniquely undesirable—“deadly dull” or “‘dogs’” (among other pejoratives).[130] But, perhaps because Congress has long asked the Court to ensure uniformity in revenue-related matters, or perhaps because of the Constitution’s special mention of the importance of “uniform” “duties, imposts, and excises,”[131] tax* cases seem uniquely important to the Court’s conflicts docket. Indeed, tax* cases earned a place of prominence above even constitutional (constitution*) cases. Similarly surprising may be the Court’s attention to questions of property* law, given that much of property law arises under state law.[132]
Figures 2 and 2A, alongside Table 1, offer analyses of the Court’s conflicts docket across nearly a century, from the 1925 Term to the 2021 Term. But the size and content of the Court’s docket has changed quite a bit since 1925: Erie, for example, was decided in 1938, effectively sweeping entire bodies of law, including many state law property questions, out of the Court’s docket.[133] Moreover, the Court decided many more cases in those early years of certiorari jurisdiction than it does now, perhaps giving those early years a relative advantage vis-à-vis more recent decades (given this research design emphasizing frequency).[134] Hence, given these important changes over time, it can also be instructive to look at smaller time horizons. In previous work, consistent with other temporal analyses of the Supreme Court, I have defined and compared different eras of the Court by the tenure of Chief Justice.[135] I reprise that approach here.
Table 2. Terms Ranked by Importance Score by Chief Justice Tenure
Table 2 presents ranked lists of significant terms during the tenure of each Chief Justice since 1925 (e.g., the Taft Court or the Roberts Court). Notably, if unsurprisingly, many of the results are consistent with the summary results presented in Table 1. Each Court, for example, appears to emphasize questions of statutory interpretation: statute* is a top-three term during the tenure of each Chief Justice, and interpretation and construction* also feature prominently at various points. Likewise, tax* is an area of focus from the Hughes Court through the Warren Court, but while remaining significant, it has become decreasingly central to the Court’s conflicts docket since then. Patent cases, too, were a focus of the Taft Court’s conflicts docket, but practically disappeared from prominence thereafter—a decline that is perhaps reflected in later complaints about persistent splits in patent cases, concomitant forum shopping,[136] and the eventual creation, in 1982, of the U.S. Court of Appeals for the Federal Circuit.[137] Overall, such analysis of the prominent terms during, say, the tenures of each Chief Justice can reveal trends over time.
2. Change in Importance Scores
These shifting priorities highlight one further approach to studying the Court’s conflicts docket. We can also rank terms by their change in importance: How, for example, do terms change from one Court to the next? Table 3 offers information about these shifts in the Court’s priorities, highlighting the most significant changes towards and away from the origin. (I emphasize that negative numbers indicate a move towards the origin, and hence an increase in significance. Likewise, positive numbers represent a move away from the origin, thus indicating a decrease in significance.) Specifically, by organizing the information by Chief Justice tenure, we can examine how individual terms have changed in importance over time.
Table 3. Terms Ranked by Change in Importance Score by Chief Justice Tenure
As noted, Table 3 presents lists of significant terms, ranked by their change in position over the previous Court.[138] Building on some of the examples noted above, Table 3 suggests, for example, that the Hughes Court traded patent cases for tax* ones, with patent cases declining precipitously in importance, while tax* cases rocketed in importance during Chief Justice Hughes’s tenure specifically. Since then, tax* cases dropped in importance significantly during the Vinson and Burger Courts. Indeed, we can visualize tax*’s importance score over time.
Figure 3. tax* Plotted by Importance Score by Chief Justice
Figure 3 plots tax*’s progression over time, beginning with the Taft Court (T), then to Hughes (H), Stone (S), Vinson (V), Warren (W), Burger (B), Rehnquist (Re), and finally Roberts (Ro). We see, consistent with Tables 2 and 3, that tax* cases rise to prominence on the conflicts docket during the Hughes Court, fade out during the Vinson Court, enjoy a brief resurgence during the Warren Court, but have since been increasingly disfavored. Figure 4, likewise, plots patent’s change over the tenure of each Chief Justice.
Figure 4. patent Plotted by Importance Score by Chief Justice
Figure 4, consistent with Tables 2 and 3, suggests that patent cases were central to the Taft Court’s conflicts docket, in keeping with Congress’s early direction to ensure uniformity in patent matters.[139] But the Court seems, almost immediately, to lose interest in such cases, as the term plummets away from the origin beginning with the Hughes Court.[140] And so, while we might have expected to see patent’s importance fade significantly during the Rehnquist Court, when Congress created the U.S. Court of Appeals for the Federal Circuit to assure uniformity in patent cases,[141] Figure 4 helps to explain why Table 3 does not indicate a significant drop in patent conflicts during the Rehnquist Court: It is likely because patent cases were already so disfavored by that point that any further shifts were not especially significant.[142]
The results in Tables 2 and 3 might also help us better understand the evolving influence of certain aspects of judicial behavior. I noted above that divided_lower_court*’s salience corroborates other findings regarding the influence of dissenting and concurring opinions from the federal appeals courts and state supreme courts.[143] Tables 2 and 3 and Figure 5 help us understand how that influence has changed over time.
Figure 5. divided_lower_court* Plotted by Importance Score by Chief Justice
During the Roberts Court, for example, the term divided_lower_court* moved substantially closer to the origin, suggesting that the signals sent by dissenting or concurring opinions on the lower courts are more important now than before. We can also see that this trend began in earnest with the Vinson Court: During Chief Justice Vinson’s tenure, the divided_lower_court* term ranked among the most significant, in terms of both its importance score and its change in importance score.
Similarly, Table 3 suggests that individual Members of the Court are, beginning particularly with the Burger Court, more likely to write separately to describe their views about certiorari. During the Burger Court, first_person* terms (terms like i_would, which are generally part of phrases like “I would grant the petition” or “I would deny certiorari”[144]) grew in importance to the Court’s certiorari decisions. This result leaves open a question: Are the Court’s Justices writing separately more, or are their separate writings more focused on certiorari-related issues? A two-dimensional look can help offer an answer.
Figure 6. first_person* Plotted by Importance Score by Chief Justice
Figure 6 plots first_person*’s progression over time. During the Burger Court, the term’s importance increases along dimensions of both proximity and frequency. Overall, since the Taft Court, the term’s salience has grown predominantly along the proximity dimension, suggesting that the Justices’ separate opinions are, over time, increasingly focused on certiorari-related matters. However, such opinions (at least, such opinions mentioning certiorari*) are not becoming all that more common relative to the total number of opinions issued per Court.
This dimensionality analysis can help us better understand other results too. Several scholars have bemoaned the Court’s inattention to Fourth Amendment matters.[145] And yet Table 3 shows that probable_cause grows significantly in importance to the Roberts Court’s conflicts docket, and Table 2 suggests that fourth_amendment concerns are (perhaps barely) one significant consideration to the Roberts Court’s construction of its conflicts docket. What gives? Figure 7 suggests that, during the Roberts Court, probable_cause-related issues are more central to the Court’s decision to grant certiorari in those cases where they are implicated—but that such cases remain relatively rare.[146]
Figure 7. probable_cause Plotted by Importance Score by Chief Justice
Understood together, Tables 2 and 3 offer one look at the Court’s conflicts docket over time. In some respects, these results are expected: The Court’s conflicts docket has long been, and continues to be, dominated by questions of statutory interpretation (e.g., statute*, interpretation, and construction*). In other respects, these results may seem surprising: e.g., the comparatively low standing of terms like constitution*, or the Court’s longstanding attention to ensuring uniformity in tax* matters. And a closer review may help uncover some nuance in some of these results: As with patent cases, for example, the Court’s attention to tax* matters might be informed by Congress’s early instruction to exercise certiorari jurisdiction to ensure uniformity in these areas of law;[147] but, in the time since, the Court’s preference to avoid such “deadly dull” matters may have overtaken, if only partially, Congress’s direction.[148]
B. Interpretations
In this section, I take some of the results noted in isolation above and aim to make some greater sense of the data amassed from this large-scale analysis. In so doing, I emphasize two main themes.
First, I investigate the Court’s durable attention to conflicts over matters of statutory interpretation. I find that terms like statute* consistently seem especially salient to the Court’s decisions to grant certiorari along both frequency and proximity dimensions, spanning nearly every era of the Supreme Court (e.g., the Hughes Court, the Warren Court, the Burger Court, etc.). Moreover, many of the other shifts in the Court’s certiorari priorities can be traced back to congressional action, reflecting the Court’s emphasis on statutory splits. In short, Congress exercises a potentially surprising degree of indirect control over the Court’s conflicts docket: When Congress legislates, its enactments frequently pose interpretative questions that eventually wend their way to the Supreme Court.
Second, I compare the Court’s conflicts docket to its important-questions docket, the subject of my previous work employing this methodology. Here, I find that the Court’s focus on statutory interpretation questions in its conflicts docket contrasts with the Court’s focus on constitutional questions in its important-questions docket. Indeed, constitutional cases rank surprisingly low on the Court’s conflicts docket. This distinction also appears to have important implications for docket volatility and the influence of the Justices’ personal priorities: The Court’s conflicts docket is less responsive to changes in the Court’s personnel than its important-questions docket.
1. Statutory Splits and Congressional Control
As noted above, the Court’s conflicts docket, both over time and during the respective tenures of individual Chief Justices, seems to have paid special attention to questions of statutory interpretation and construction. Consistent with the Court’s longstanding role “to unite and assimilate . . . the rules of national decisions,”[149] the Court has dedicated much of its conflicts docket to ensuring that Congress’s enactments apply uniformly across the country. This finding also deepens this Article’s titular puzzle: Which statutes form the splits that garner the Court’s attention?
My answer to this question has two parts. First, and perhaps unsurprisingly, the Court’s conflicts docket seems to focus on broad and recent statutes that apply widely across the country and economy.[150] Second, the matters that arise on the Court’s conflicts docket can be sticky, persisting over time. Rather than definitively resolving a conflict in one case (and casting the matter behind it), the Court sometimes finds itself drawn into a cycle of refining its holdings and interpretations across a series of cases. Viewed together, these findings suggest that Congress wields substantial, if indirect, control over the Court’s agenda: Congress can, simply by legislating, nudge the Court to confront questions implicated by its enactments, and those questions may play a recurring role on the Court’s agenda.[151]
a. Broad and Recent Statutes
The results described above suggest that broad statutes with wide implications—such as revenue statutes with implications for taxpayers,[152] labor and employment protections that affect many businesses,[153] and sentencing provisions that attend to the vast federal criminal apparatus—are especially likely to give rise to conflicts that merit the Court’s attention.
Consider, again, some of the terms in Tables 2 and 3. As noted above, the Taft Court (following, perhaps, Congress’s directions from certiorari’s earliest days) seems to have paid special attention to patent cases, including cases that, say, considered the “statute of limitations . . . [for] vacat[ing] and annul[ling] . . . patent[s],”[154] or the “purpose and effect of the [Trade-Mark Act of February 20, 1905].”[155] Likewise, the Hughes Court’s special attention to tax* cases appears to be a direct product of Congress’s regular passage of Revenue Acts, alongside a special attention to assuring “a uniform application to a nation-wide scheme of taxation.”[156] Stated otherwise, tax* cases were both important to the Hughes Court’s conflicts docket and increased significantly in importance to the conflicts docket because of Congress’s various tax-related enactments. So too for the Stone Court and labor cases: Labor cases were both important to that Court’s conflicts docket and increased in importance during the Stone Court (over the Hughes Court), in part because Congress’s passage of the Fair Labor Standards Act in 1938,[157] among other provisions,[158] gave rise to various interpretative questions that required the Court’s attention. The Rehnquist Court evinced similar trends. During Chief Justice Rehnquist’s tenure, sentencing-related cases (sentence*) occupied a substantial portion of the Court’s conflicts docket.[159] And, again, this new focus for the Court’s conflicts docket seems to be the consequence of congressional action, namely, Congress’s passage of the federal sentencing guidelines[160] and the Armed Career Criminal Act.[161]
Figure 8. sentence* Plotted by Importance Score by Chief Justice
Looking beyond such recent[162] and broad statutes, we can see that some terms implicate trans-substantive questions—remedial questions, for example—that may arise out of provisions in new statutes. Questions of damages, for example, were of particular interest to the Rehnquist Court—and several of these cases required interpreting the remedial provisions of recently enacted statutes, such as the Americans with Disabilities Act.[163] Moreover, congressional enactments that pertain to the Judiciary itself also give rise to splits that affect the Court’s agenda. Changes in 1940 to the Federal Judiciary’s diversity jurisdiction seems to have motivated the Vinson Court’s attention to such cases.[164] In short, Congress exercises a fair degree of control over the Court’s agenda, through both its general legislative powers and its powers over the Federal Judiciary (e.g., jurisdiction and remedies).
b. Stickiness
The Court’s conflicts docket thus tends to be shaped by congressional enactments—particularly those of broad applicability, or those that pertain to the Judiciary itself. And while many of these conflicts arise out of recent congressional enactments, once an issue earns a spot on the Court’s agenda, it tends to stay there. Stated otherwise, Congress’s indirect control over the Court’s agenda is sticky.[165]
Recall, for example, that tax* cases earned a place of prominence on the Court’s docket because of Congress’s revenue-related legislative activity. And now such cases seem to have become difficult to dislodge, frequent laments about such cases from some of the Court’s Members notwithstanding. In short, even classes of “deadly dull” cases manage to persist on the Court’s conflicts docket.[166] I am careful not to overstate this finding: tax* cases have waned in their significance, as Figure 3 illustrates; and, after Taft, the Court seems to have lost interest in patent conflicts.[167] But even as tax* cases have waned in significance, they have so far retained some significance to the Roberts Court’s conflicts docket. Cases about sentence*-related issues also remain important to the Roberts Court after their rise to prominence during Chief Justice Rehnquist’s tenure.[168] So too did the significance of labor and employ* cases persist for decades after their ascension on the conflicts docket.[169] Overall, Tables 2 and 3 suggest several terms that persist on the Court’s agenda long after the passage of those statutes that might have first helped to deliver them to the conflicts docket.
What explains the apparent stickiness of the Court’s attention to such statutory splits? I cannot say for sure, but one theory—regarding the dialogic function of reasoned decisionmaking in the courts[170] and stemming from a close review of many of the cases underlying the data presented above alongside the associated literature on the Supreme Court’s docket—seems promising. In short, the Court’s opinions are one entry in an ongoing dialogue about the meaning of federal law. Congress and other courts carry that conversation forward until it is volleyed back to the Supreme Court.
Consider the phenomenon of “Supreme Court Repeaters”—that is, cases that “defy the odds and obtain certiorari more than once” (by, say, returning to the Court after a remand).[171] In their study of these cases, Jason Iuliano and Ya Sheng Lin suggest that the Court is drawn to some “repeaters” in order to help “clarify unsettled rules of law” by reinforcing or refining the holding of the first case in the series.[172] Stated otherwise, the first case in the series left some questions open, or provoked a new issue, that required a second intervention. Such a phenomenon can be generalized beyond a particular case and to a class of cases, such as those about the Fair Labor Standards Act. Indeed, Justice Frankfurter explained as much in 10 East 40th Street Building v. Callus, writing that “[t]he Act has produced a considerable volume of litigation and has inevitably given rise to judicial conflicts and divisions” and that “enforc[ing it] [has] involve[d] the courts in the empiric process of drawing lines from case to case.”[173] This has been true even for lines of cases with closely analogous facts, as the Court explained in Borden Co. v. Borella: “Once again, as in Kirschbaum v. Walling, 316 U.S. 517 . . . we are required to consider the application of the Fair Labor Standards Act of 1938 to employees engaged in activities relating to the maintenance and operation of a building.”[174] Put simply, because its opinions must be interpreted and applied in subsequent litigation, the Court’s conflicts cases beget more conflicts cases.
Just as the Court’s cases may begin a dialogue across the Judiciary as the appellate courts try to make sense of the Supreme Court’s rulings, so too might these cases instantiate a dialogue with Congress—one that also requires the Court to review and revisit responsive legislative enactments. As William Eskridge has written (both in solo-authored work and in co-authored work with Matthew Christiansen), Congress frequently overrides the Court’s statutory interpretation decisions,[175] though, perhaps, with less frequency in recent years.[176] And these subsequent enactments themselves raise interpretative questions that may wend their way back to the Court.[177] Consider, for example, one series of cases during the Roberts Court implicating employ* terms (which, recall, were significant to the Roberts Court’s conflicts docket). The Court’s decision in Ledbetter v. Goodyear Tire & Rubber Co., construing the statute of limitations for certain employment discrimination claims, was famously overridden by the Lilly Ledbetter Fair Pay Act of 2009.[178] But Congress’s new enactment gave rise to a new question: Does the Lilly Ledbetter Fair Pay Act mean that the Court should upturn its interpretative approach to all such statute of limitations questions—or does Congress’s decision to override this one discrete decision leave the rest of the Court’s precedents intact?[179] That question was at issue in one of the Court’s later conflicts cases, Green v. Brennan.[180]
This interbranch dialogue may also help to explain the Court’s longstanding attention to tax cases. As Nancy Staudt has written, Congress legislates in response to the Court’s tax decisions quite frequently, and the Court has heard a surprisingly large number of tax cases.[181] Some of these cases interpret legislative enactments or regulations that are themselves a reaction to the Court’s cases. For example, the Court granted review in United States Department of Treasury v. Fabe[182] to resolve a conflict on the meaning of the McCarran-Ferguson Act, which was itself “enacted in response to th[e] Court’s decision in United States v. South-Eastern Underwriters Ass’n.”[183] The Court’s conflicts docket might thus be understood to reflect a dialogue between Congress and the Judiciary.
2. Statutory Stability and Constitutional Volatility
The Court’s focus on statutory splits—and the stickiness of the Court’s attention to certain statutory questions—invites us to consider how the Court’s conflicts docket compares to the rest of its agenda—its important-questions docket. Is the Court’s important-questions docket similarly focused on questions of statutory construction and interpretation? Is the Court’s important-questions docket similarly stable? To both questions, the answer is no.
As I have detailed in my prior work, the Court’s important-questions docket appears to focus far more on constitutional questions (including questions on the constitutionality of federal statutes).[184] And, perhaps consequently, the Court’s important-questions docket responds to changes to the Court’s personnel with more volatility.[185]
As Tables 2 and 3 indicate, conflicts over the meaning or application of constitutional provisions are—perhaps surprisingly—not a particularly substantial part of the Court’s conflicts docket. While constitution* appears on Table 1’s list of significant terms, it is outranked by terms like tax*, labor*, and employ*. Moreover, constitution* cases do not even begin to occupy a significant spot on the Court’s conflicts docket until Chief Justice Burger’s tenure, as Tables 2 and 3, and Figure 9 (below), indicate.
Figure 9. constitution* Plotted by Importance Score by Chief Justice
Moreover, the sharp increase in attention to constitution* cases on the conflicts agenda is likely due—at least in part—to changes in the scope of the Court’s certiorari jurisdiction.[186] In 1976, Congress eliminated the Court’s mandatory jurisdiction over “the large class of suits challenging the constitutionality of state or federal statutes,” moving such cases to the Court’s discretionary certiorari jurisdiction.[187]
But even once these constitutional cases were freed from the Court’s mandatory docket, they continued, and continue, to rank surprisingly low on the Burger, Rehnquist, and Roberts Courts’ conflicts dockets. In their important-questions dockets, for example, constitution* was consistently the third most significant term—but, on the conflicts docket, the term never cracked the top five.[188] Other indicia also suggest that constitutional cases were (and are) a less significant component of the Court’s conflicts docket.[189] This finding is consistent with other scholars’ studies of the Court’s agenda. Richard Pacelle, for example, once described the Court’s agenda as “bifurcated” into a “volitional agenda” (analogous to the important-questions docket) and an “exigent agenda” (analogous to the conflicts docket). In his view, the Court stocks its important-questions docket—its volitional agenda—with cases that “fulfill the policy designs or goals of its members.”[190] By contrast, the Court’s exigent agenda—its conflicts docket—sounds in the Court’s longstanding duty to ensure uniformity in the federal courts and otherwise perform its supervisory role in the Judiciary.[191] And so the curation of the Court’s conflicts agenda is guided by the Court’s more traditional “institutional rules and norms.”[192]
Hence, we might expect the Court’s conflicts docket to be more stable than its important-questions docket—and we might expect it to be especially less responsive to changes in the Court’s personnel. As I explained in my prior study of the Court’s important-questions docket:
[A]s the Court has gained greater discretion to set its own docket, its exercise of that discretion has become more fluid. Moreover, changes in the Court’s composition seem to yield even more substantial changes to the shape of the important-questions docket. Stated similarly, the Court has—and has exercised—wide discretion to shape its own docket, and who is on the Court may matter to how the Court exercises that docket-setting discretion.[193]
However, the Court’s conflicts agenda does not seem to share this proclivity for volatility in view of personnel shifts. We can see this by plotting the volatility of the Court’s conflicts docket, as in Figure 10 (below). There, each “plotted value represents the average change, Supreme Court Term over Supreme Court Term, across all the Index Terms common to the compared subsequent Terms (e.g., October Term 2018 over October Term 2017).”[194] And Figure 10 separates those terms where the Court’s composition changed from those where it remained stable.
Figure 10. Overall Docket Volatility, Term over Term
Figure 10 suggests that, unlike in the important-questions context, the Court’s conflicts docket has not become more volatile as its docket discretion has grown. Moreover, changes in the Court’s personnel do not seem to correlate with more dramatic changes in the shape of the Court’s conflicts docket.
Again, I am careful not to overstate this finding. It is certainly possible to identify some shifts in the Court’s conflicts agenda that we might associate with conventional views of that Court’s ideology. The Warren Court, for example, is famous for its attention to the due process rights of criminal defendants,[195] priorities which Table 2 and Table 3 seem to confirm, given the stature of terms like criminal*[196] and due_process[197] during Chief Justice Warren’s tenure.
Likewise, previous studies of the Court have suggested that, “[b]eginning with the Rehnquist Court, justices have become more willing to reject precedents they think were badly reasoned, simply wrong, or inconsistent with their own senses of the constitutional framers’ intentions.”[198] And Table 3 suggests that the Rehnquist Court did indeed evince a growing interest in using the occasion of a conflict to overrule precedent.[199] But in the Rehnquist Court, the magnitude of that shift reflected the fact that previous courts only rarely used splits as an opportunity to overrule settled law—and so, in this instance, only one or two examples of overruling precedent on the conflicts docket (especially as the Court’s docket continued to shrink) caused a significant change. In all, then, even after accounting for these examples, changes in personnel do not appear to result in large changes in the Court’s conflicts agenda, as Table 4 (a distillation of Figure 10) indicates.
Table 4. Average Docket Volatility
Indeed, the Court’s conflicts docket has become more stable (i.e., less volatile) over time, even as it has gained greater docket discretion, as was the case in 1988 when Congress virtually eliminated the Court’s mandatory jurisdiction.[200] This contrast between the Court’s important-questions docket and its conflicts docket—with the latter more focused on statutory questions and less responsive to changes in the Court’s composition—may suggest some lessons for reforming the Supreme Court’s agenda.
III. Certiorari and Judicial Reform
As described above, the Court’s conflicts docket pays special attention to questions of statutory construction and interpretation. And while the Court’s agenda has focused on different statutes at different times, the Court’s conflicts docket has not become more volatile as the Court has gained greater docket discretion, nor as the Court’s personnel changes.
I turn now to consider how these findings may bear on questions of Supreme Court reform. In particular, I elaborate on two specific proposals.
First, I propose a mechanism to get more statutory splits onto the Court’s agenda. Specifically, I suggest that Congress buttress the federal courts of appeals’ existing powers to certify cases implicating statutory splits to the Supreme Court for mandatory review. I make this suggestion for several reasons. One, as the Court’s docket shrinks, many conflicts fester. But our legal system values uniformity—uniformity protects the legitimacy of the law by avoiding impressions of arbitrariness; it facilitates predictability; it mitigates forum shopping; and it reduces costs for multistate actors.[201] And, in that system, the Supreme Court is architected to provide uniformity. Two, a Supreme Court that is more focused on resolving new and festering conflicts on the meaning and application of federal statutes may find that it has less capacity to select and decide cases that remake our constitutional order. And it is good for the Judiciary—and for our democracy—when the Supreme Court’s agenda reflects the priorities of our democratic enactments rather than vacillates to mirror the individual ideological priorities of the Court’s current composition.
Second, I reiterate my call for the Court to describe its certiorari decisions—a critical procedural step in Supreme Court litigation—in more detail. By developing a more transparent common law of certiorari, the Court can help provide greater transparency and predictability for litigants, the public, and the political branches.
A. The Return of a Mandatory Docket
Although splits remain the most important determinant of certiorari, the Court is deciding fewer and fewer cases, leaving many splits unsettled. Scholars have, as noted above, decried the Court’s unwillingness to provide guidance and assure uniformity on matters ranging from securities litigation and ERISA[202] to Fourth Amendment concerns.[203] And the results presented above reinforce these complaints. Bankruptcy, for example, has never been a major priority on the Court’s conflicts docket—even though the Constitution itself emphasizes the importance of uniformity in bankruptcy contexts, and even though scholars have noted the Court’s decisions to decline to address some persistent splits in bankruptcy matters (giving rise to substantial forum shopping).[204] Indeed, the Court’s tendency to focus on a relatively stable set of concerns in its conflicts docket—i.e., the stickiness described above—may leave a significant number of splits entirely unresolved.
Fortunately, Congress has already provided a mechanism for addressing the problem of languishing circuit splits.[205] 28 U.S.C. § 1254(2) grants the federal courts of appeals the power to certify questions of law to the Supreme Court. As Amanda Tyler has explained, though “the current phrasing of the statute [is] in terms that easily could be read as permissive rather than mandatory, in form and history, this certified question jurisdiction is mandatory.”[206]
Hence, the federal appeals courts should certify questions implicating splits to the Supreme Court for its review and resolution. Consider, for example, Connecticut Fine Wine & Spirits v. Seagull, a case about the applicability of the Sherman Act to certain state provisions governing the sale of alcohol. While the details of the litigation are not important for present purposes,[207] the case features an opinion dissenting from the denial of petition for rehearing en banc, in which four members of the U.S. Court of Appeals for the Second Circuit noted that the court’s decision “perpetuates a circuit split” of “longstanding duration,” namely thirty-two years.[208] It is possible to imagine that two more members of the court, though perhaps unwilling to reconsider circuit precedent, would be willing to certify the case to the Supreme Court for its consideration to provide uniformity in the law and to ensure that alcohol distributors operating nationally do not face different rules in different states.[209] Indeed, Justice Sotomayor recently suggested that the federal courts of appeals certify cases implicating a split to the Supreme Court when they are likely to otherwise escape further review.[210]
The Court has recently dismissed such certificates, thereby treating § 1254(2) as permissive and preferring to retain control over its own agenda. But an increasing flow of certifications from the federal appeals courts may serve an important informational function, helping to alert both the Court and Congress to those splits that, in the view of the courts of appeals, need greater attention.[211] Indeed, the divided_lower_court* results described above indicate that this sort of signaling is both feasible and effective: The lower courts can implement this signal (just as they write separately from panel and en banc decisions), and the Supreme Court is increasingly attentive to it (as it seems to play a growing role in the Court’s certiorari decisions).
Moreover, precisely because the Court has recently treated its certification jurisdiction as merely permissive,[212] Congress should exercise its Article III powers over the Court’s appellate jurisdiction[213] and buttress § 1254(2) to specify that the Court must hear certified questions implicating splits over the meaning or interpretation of a federal statute. The first appendix to this Article offers one such possible amendment.
Some readers may wonder why we should even consider the possibility of an expanded conflicts docket. After all, the results described above seem to suggest that the conflicts docket is, so far, working relatively well: It is an example of the Court acting in an apparently less politicized manner,[214] consistent with its “institutional rules and norms,”[215] and carrying out its longstanding duty “to unite and assimilate the principles of national justice and the rules of national decisions.”[216] Nevertheless, there are at least three reasons to consider reviving the Court’s mandatory docket to encompass certified statutory splits.[217]
First, some splits seem to languish, including long after they have percolated. The split of “longstanding duration”—thirty-two years—noted above is one example. A copyright split that persisted for twenty-eight years is another.[218] Creating a mandatory conflicts docket would help address these lingering divisions. Moreover, as I imply above, doing so would reaffirm Congress’s longstanding view on the importance of uniformity, and the values underlying these branches’ shared commitment to addressing splits in federal law in order to advance the rule of law, avoid the spectre of arbitrariness, ensure predictability, and mitigate forum shopping concerns.[219]
Second, notwithstanding the apparent successes of the Court’s conflicts docket, the Court is increasingly viewed as a political institution.[220] And this perception appears to be, in large part, a consequence of the Court’s important-questions agenda, which, as noted, seems to increasingly reflect the individual constitutional ideologies of the Justices. The Court’s important-questions docket has a comparatively greater focus on constitutional cases, and the selection of those cases is shaped by the individual views of the Court’s individual Members.[221] But the Court’s conflicts docket is statutory and less responsive to swings in the Court’s composition.
One solution is, then, to “crowd out” the Court’s important-questions docket in favor of more statutory conflicts cases.[222] Stated simply, perhaps one way to address concerns about the Court’s legitimacy is to require the Court to do more of what is working well, and less of what is not. Cases ensuring uniformity in the application of the bankruptcy code are much less political than, say, cases about reproductive rights.[223] Analogously, Justice Barrett has suggested that greater unanimity is better for both the Court and the national political environment,[224] and the Court’s conflicts docket has, since Chief Justice Taft’s tenure, evinced greater consensus than its important-questions docket.[225] And so a larger statutory conflicts docket may help to rehabilitate the Court (and, perhaps, the Court’s public image).
Third, as described above, the Court’s conflicts docket often reflects Congress’s statutory priorities, subjecting the Court’s agenda to some indirect democratic control. That is, the results presented above suggest that the Court resolves conflicts in the context of statutes that Congress has recently considered and enacted. And those Court decisions may engender further responses from Congress (as in the Lilly Ledbetter example or in the ongoing feedback cycle of tax cases and tax legislation).[226] Indeed, the very fact of certification may motivate Congress to action;[227] in other cases, for example, Congress has legislated to preempt questions on the Court’s docket. Such democratic control is good for both our democracy and the Court. For one, it helps to ensure that the Court’s docket is focused on those matters that Congress has deemed important. It focuses the Court’s attention on the matters that our elected representatives have prioritized. And this “democratic control over the Court’s docket is legitimating, as the Constitution itself suggests”[228]: When political priorities cross the Court, it is far better that they come by way of reforms enacted in the political branches, rather than through the individual ideological preferences of the Court’s Members, or even in the Court’s unreasoned and opaque agenda-setting decisions.
I note two possible objections to a more robust certification process.
First, some readers might observe that I am extrapolating from prior results, with no guarantee that an expanded conflicts docket will display the same virtues as the Court’s present conflicts docket. Stated otherwise, perhaps a larger conflicts docket will prove to be more volatile. I concede that this is a possibility, especially at first when the docket sees the sudden arrival of cases implicating issue areas that the Court may have neglected. But, given the examination of the cases underlying the Court’s conflicts docket, I suspect that any such volatility will be short-lived. As noted above, there is a stickiness to the Court’s docket that is the consequence of the dialogue, across the branches and among the courts, that emerges from the Court’s decisions. So while new issues may arise, or some may lose sway, this seems a slow process as the Court continues to react to new splits and responsive legislation. Moreover, even if a reformed docket would evince some increased volatility, it is important to emphasize that those swings do not reflect the mere whims of the Court’s Justices; rather, they reflect changes in our democratic priorities, consistent with Article III’s text, which confers on Congress the power to set and modify the Supreme Court’s appellate jurisdiction.[229]
Second, some readers may object that such a pathway creates an opportunity for courts of appeals to “game” the Court’s docket[230] by intentionally creating a split and certifying a case for the sole purpose of putting it on the Court’s docket. I concede that is possible. But courts of appeals can already create splits to increase the likelihood of Supreme Court review.[231] And, as divided_lower_court* suggests, the state and federal appeals courts have several tools they can employ to increase the probability of certiorari. The approach suggested here addresses the problems of disuniformity and languishing splits in particular without creating very much incremental downside risk. Moreover, this approach, as noted, may help to confer greater legitimacy on the Supreme Court, both by giving Congress more control over the Court’s docket and by reducing the Court’s own capacity to pick and choose “important” cases that reflect the idiosyncratic priorities of the Court’s then-current membership. Instead, it focuses more of the Court’s attention on its core functions, such as resolving the interpretative, statutory questions that have perplexed the lower courts.
B. A Common Law of Certiorari
As noted above, it is true that Congress and the lower courts play a significant role in shaping the Court’s agenda. But so long as the Court retains some certiorari jurisdiction, it will continue to exercise agenda-setting discretion. And although this Article, among other studies, can help us “identify patterns and trends in the Court’s . . . docket,” the Court’s exercise of that discretion continues to lack “a critical substrate of judicial reasoning.”[232]
For example, we may know that the Court has been (and remains) especially drawn to tax* cases; and we seem to have found, as Tables 2 and 3 suggest, that splits over bankruptcy* matters rarely find a place on the Court’s docket. But we do not know why this is so. The Constitution emphasizes a need for uniformity in both contexts.[233] And, through the 1891 Evarts Act,[234] Congress moved both classes of cases to the Court’s discretionary docket,[235] assigning to the Court the role of assuring uniformity in both tax and bankruptcy contexts. So what explains their difference? We cannot say for sure. Likewise, we can discern that criminal* cases did not occupy a significant portion of the conflicts docket until the Warren Court and so may predict when the Court is more or less likely to hear cases presenting such splits. But we do not really understand why the Court was so averse to such cases for so long, particularly in view of Chief Justice Taft’s own statements, reflecting on the Judges’ Bill of 1925, emphasizing the need for “uniformity” in “securing a final construction of new criminal statutes.”[236]
We should demand that the Court offer greater transparency in the form of binding “certiorari doctrine, rather than mere certiorari text.”[237] In doing so, the Court can help to resolve the difference “between the what and the why” of its agenda-setting decisions. Other scholars, from Kathryn Watts to William Baude and Steve Vladeck, have all called for greater transparency in the Court’s certiorari processes.[238] My modest suggestion, which I have described in prior work and reiterate here, is for the Court to expand upon its “helpful practice of explaining, in the ultimate opinion on the merits, the reasons which had moved it to grant the writ”[239] by linking it more expressly to established modes of legal reasoning.[240] Indeed, the growth of first_person* over time suggests that the Justices themselves are—at least individually—more willing to elaborate on which cases do and do not merit review. They should do so collectively, as a judicial institution, too.
Such a common law approach “would require the Court to interpret and apply Supreme Court Rule 10 like other rules of procedure, giving its docket-setting canons more doctrinal content consistent with common law norms of reason-giving: transparency; predictability; bindingness (to a degree) in future deliberations (yet flexible to evolve as needed, as with other common law doctrines); and susceptibility to political scrutiny.”[241] Common law procedural doctrines are already familiar to our legal system, and a certiorari-specific doctrine would not require any drastic changes to the Court’s existing practices. For example, the Court in Gomery already explained that splits over the status of a single res—in that case, a patent—can be untenable.[242] Likewise, in the context of the important-questions docket, the Court explained that its supervisory powers could warrant granting certiorari to ensure the orderly review of cases, including by enforcing appropriate standards of review.[243] And, more recently, the Court suggested that the exigencies related to a national emergency, such as a pandemic, may require that the Court weigh in.[244] The Court should build upon these disconnected and sporadic hints to develop a more complete doctrine addressing what matters to the Court, and why, when granting certiorari.
The development of such a procedural doctrine could help to address concerns about transparency, predictability, and ultimately, legitimacy. The Court’s power and legitimacy are derived, at least in part, from norms of transparency.[245] The Court’s tradition of public reasoning binds the Judiciary in future decisions, thereby generating trust in the Court’s decisions. Thus, transparency also facilitates predictability.[246] But the Court’s current, opaque approach to “certiorari-stage transparency undermines certiorari-stage predictability” and may cause some counsel to appeal in vain to conflicts that are unlikely to attract the Court’s attention, without ever knowing why. A common law doctrine of certiorari would address these predictability-related concerns for litigants who are “considering whether to undertake the costly process of filing a petition for certiorari.”[247] Such a doctrine, moreover, “might aid the Court in its own docket-selection processes, as clearer certiorari standards might cause more litigants, whose cases obviously do not raise certworthy questions, to drop out, allowing the Court to focus on more meritorious petitions.”[248] In all, opacity in agenda setting can undermine the Court’s legitimacy by both hindering predictability for litigants and insulating these decisions from public and political[249] scrutiny. Hence, a common law of certiorari can offer both the bar and the academy better insights into the Court’s docket-setting processes and may thereby inspire greater confidence that the Court’s certiorari canons are consistently applied (or else inspire the Court to apply them more consistently).[250]
* * *
The Supreme Court’s conflicts docket is, perhaps, an example of the Court working well—or, at least, as intended. In its conflicts docket, the Court offers its opinions in dialogue with Congress and other courts nationwide. And those opinions are on issues that have mattered to those institutions, both because Congress has seen fit to legislate on them and because they have presented perplexing legal questions that require uniform resolution. In contrast, the Court’s important-questions docket is more likely to reflect the Court’s own priorities—priorities that seem profoundly shaped by the Court’s composition of personalities and that are otherwise hardly explained.
In view of this contrast, I offer two modest suggestions that expand on what appears to be working well in the Court’s conflicts docket: a larger, mandatory statutory conflicts docket; and a better elaborated common law of certiorari. Each of these proposals may improve the Supreme Court’s standing by focusing more of its attention on core judicial business. A common law of certiorari gives the Court the opportunity to engage in greater reasoned decision-making, facilitating transparency and predictability in the Court’s agenda. By addressing more statutory conflicts on certification, the Court can focus on deciding legalistic issues in contexts that have mattered both to Congress and in litigation nationwide. Together, they give the Court the opportunity to repudiate opaque procedures that yield constitutional lawmaking that reflects the priorities of individual Justices. Some Members of the Court, both today and in history, may have preferred the latter. But the Court, as an institution of our government, should prefer the former.
Conclusion
The Supreme Court is well-known to favor granting review in circuit splits. But which splits is the Court likely to review? And how have the Court’s priorities changed over time?
A co-word analysis—a type of computational text analysis—of the “certiorari paragraphs” contained within the Court’s opinions offers one view into the Court’s agenda. This Article’s analysis finds that the Court’s conflicts docket has been especially focused on matters of statutory construction and interpretation. And while the Court’s agenda has homed in on different statutes at different times, often reflecting Congress’s most recent priorities, the Court’s conflicts docket has not become more volatile in view of greater docket discretion or even changes to the Court’s personnel. Such findings contrast with my prior studies of the Court’s important-questions docket, which is comparatively more focused on constitutional cases and more responsive to changes in the Court’s personnel.
These patterns—and they are only patterns—may have some implications for reforming the Court’s certiorari processes. For one, the fact that these patterns are largely unexplained suggests that we might desire and demand a more reasoned and binding common law of certiorari from the Supreme Court. And given that the Court’s conflicts docket seems less political than its important-questions dockets, we might require that the Court hear more conflicts cases. Both our democracy and the Court itself are better served by an agenda that is focused on resolving splits over questions of statutory interpretation than by one that attempts to make and remake our constitutional order.
Appendix: Amendment to § 1254
(a) Cases in the courts of appeals may be reviewed by the Supreme Court by the following methods:
(1) By writ of certiorari granted upon the petition of any party to any civil or criminal case, before or after rendition of judgment or decree;
(2) By certification at any time by a court of appeals of any question of law in any civil or criminal case as to which instructions are desired, and upon such certification the Supreme Court may give binding instructions or require the entire record to be sent up for decision of the entire matter in controversy.
(b) Cases in the courts of appeals must be reviewed by the Supreme Court upon:
(1) Certification at any time, by a majority of the circuit judges of the circuit who are in regular active service, of any question of statutory law in any civil or criminal case as to which the courts of appeals are divided. Upon such certification the Supreme Court shall give binding instructions or require the entire record to be sent up for decision of the entire matter in controversy.
Methods Appendix
This Methods Appendix provides a more detailed account of the underlying datasets and the methods used to derive the results in the Article’s main text.
Constructing the Dataset
The dataset is built primarily on four sources: (1) the Supreme Court Database (Washington University in St. Louis School of Law); (2) the Caselaw Access Project (Harvard Law School); (3) CourtListener (Free Law Project, a non-profit organization); and (4) Justia (another internet resource for case law).
The Supreme Court Database codes “the reason, if any, that the Court gives for granting the petition for certiorari,”[251] with possible values ranging from 1 to 13, where 1 indicates that the case did not arise under the Court’s certiorari jurisdiction, and any value from 2 through 9 indicates that the Court granted review to resolve some conflict or confusion among the lower courts. Because this Article focuses only on cases that arise under the Court’s certiorari jurisdiction and that are granted to resolve some form of a conflict, I limited my analysis to those cases coded 2 through 9, as well as those cases coded 12 (i.e., cases granted for no specified reason).[252] There are a total of 6,673 cases coded as satisfying these criteria in the Supreme Court Database between October Term (OT) 1925 (the Term in which the Judges’ Bill of 1925 took effect) and October Term 2021 (inclusive of OT 2021).[253]
Though the Supreme Court Database contains many important, useful details about each Supreme Court opinion, it does not contain the full text of those opinions. Hence, the research team obtained the full opinion text of these 6,673 opinions from the Caselaw Access Project in HTML (4,117 opinions), from Justia in HTML (2,353 opinions); from CourtListener in plain text (202 opinions), or manually in plain text (1 opinion).
After cleaning and processing these text files,[254] the research team extracted from each opinion any paragraph containing the term certiorari.[255] 5,882 out of 6,673 opinions contained at least one such paragraph (roughly 88%), and those paragraphs formed the basic corpus for further study.
Constructing the Term Index
Next, I used this corpus to construct a dictionary of relevant terms.[256] As before, I began with a list of the 1,321 terms or phrases that appeared in over 1% of the cases in which certiorari appeared (i.e., terms that appeared in 59 or more cases).[257] Many of these 1,321 terms were clearly unhelpful—e.g., court or we granted certiorari, so I filtered this list to only those terms that I thought likely to be helpful in assessing the reasons for the Court’s decision to grant review. In prior work, I had already reviewed many of these terms (and I verified those inferences with two law school research assistants),[258] and so here, I focused my own review on the 58 terms that were new to this dataset of certiorari paragraphs. (For completeness, I also note that 94 terms were a part of the dictionary underlying that prior work’s Term Index but were removed here for failing to satisfy the frequency threshold.[259]) This gave rise to a list of 163 terms. Of those 163 terms, I consolidated some terms with shared roots and close synonyms—e.g., employee, employees, employment, and worker—under a single entry—e.g., employ*; some other terms, such as patent, were not so consolidated.[260] An asterisked term thus stands in for a collection of related terms (e.g., employ*) while any term without an asterisk represents only itself. As before, I relied on my prior efforts at consolidating terms and then integrated any terms that were specific to this project. This process gave rise to a final list of 88 unique terms—i.e., the Term Index. Cognizant that this approach can raise replicability and transparency concerns, I have reproduced the list of 163 retained original terms and the Term Index, at Appendix Table 1.
Measuring Importance (Importance Scores)
Finally, the tables of ranked terms and topics introduced in the Article’s main text derive from each term’s score along dimensions of both frequency and proximity.[261] In order to discern the Court’s approach to certiorari in important-question cases, the research team generated a two-dimensional score for each term.
One dimension represents proximity: On average, how close is each Index Term (i.e., each term in the Term Index) to any Focal Term (i.e., those terms in a study paragraph that serve as a locus for discerning the Court’s approach to deciding which questions are important enough to merit certiorari)? I used question*, certiorari*, and conflict* terms as Focal Terms.[262] During the Roberts Court, for example, habeas* appeared (on average) 14.1 terms away from any of question*, certiorari*, or conflict*. Moreover, during the Roberts Court, one of the most proximate terms was punishment (which appeared next to a Focal Term), and the least proximate term was robbery (which appeared 56.5 terms away from the closest Focal Term).[263] Each term was assigned a linearly scaled score, between 0 and 1. Hence, in this example, punishment was assigned a score of 0, robbery was assigned a score of 1, and habeas* was assigned a score of 0.236.[264]
The second dimension represents frequency: In how many cases does the Index Term appear? To continue with the same example, the term habeas* appeared in 16 cases during the Roberts Court. Moreover, during the Roberts Court, the most frequent such term was statute* (which appeared in 109 cases), and one of the least frequent terms was punishment (which appeared in 1 case). Again, each term was assigned a scaled score, between 0 and 1. Hence, in this example, statute* was assigned a score of 0, punishment was assigned a score of 1, and habeas* was assigned a score of 0.861.[265]
One complication bears mentioning. As noted, the dataset encompasses cases coded as having been granted for some conflict-related reason (i.e., coded 2 through 9 in the Supreme Court Database), as well as cases that were granted for no noted reason (i.e., coded 12 in the Supreme Court Database). But cases that were granted for no noted reason might, in fact, have been granted for reasons unrelated to conflicts among the lower courts. Thus, to sharpen focus on conflict-related cases among the cases coded as giving no rationale for the grant of review, I include only those cases that include a conflict* term.[266] Hence, Importance Scores are based on 2,341 cases (while, as described above, the Term Index draws from the larger set of all certiorari paragraphs).
In sum, each Index Term is assigned a two-dimensional score in the form of (Frequency Scaled Score, Proximity Scaled Score). To compare terms over time, a single Importance Score for each Index Term is calculated by plotting and measuring that Term’s distance to the origin (i.e., distance from (0, 0), which represents certiorari*).[267] The charts and tables presented in the Article’s main text are largely limited to those Index Terms whose Importance Score is less than the median of all Importance Scores minus one median absolute deviation.[268]
Measuring Change in Importance (Deltas)
Change in importance for a given Index Term from, say, the Rehnquist Court to the Roberts Court is measured straightforwardly by computing the difference (Delta, or ∆) between Importance Scores for the two time periods.[269] Where an Index Term does not appear in one of the time periods under comparison, it is treated as minimally important (i.e., assigned a position of (1, 1)). The charts and tables presented in the Article’s main text are limited to those Index Terms whose Deltas are greater than the mean of all Importance Scores plus one standard deviation (indicating a move away from the origin, or a drop in importance), or are less than the mean of all Importance Scores minus one standard deviation (indicating a move toward the origin, or an increase in importance).
Measuring Change in Change (Volatility)
Measuring change in change, as in Figure 10, is somewhat more complex. I begin with the absolute value of the Delta for each Index Term that is common across the compared time periods (i.e., Term), excluding terms that do not appear in both time periods. The approach here deviates from the approach taken above (where missing terms are assigned a score) because the studied time periods are much shorter, and so any inference about the meaning of an absent Index Term is much more tenuous. Overall change in change is represented by the average across all such measures for a given time period. This value offers a sense of overall volatility: Did priorities shift wildly, with previously less important terms moving up significantly and previously important terms falling? Or did terms stay roughly in place?
In Figure 10, Index Terms are compared Supreme Court Term over Supreme Court Term (year over year), using the Supreme Court Database’s “Term of Court” field. To identify “Stable Court Composition” Terms and “New Court Composition” Terms, I used the Supreme Court Database’s list of starting dates for each natural court. If a new natural court started during a given Term or during the summer immediately preceding that Term (i.e., after July 1), then that Term was noted as one with a new composition. I use this approach because though new Justices seated during, say, the summer of 1994 are formally confirmed during the 1993 Term, their effect on the Court’s docket is unlikely to be felt until the 1994 Term. Hence, this approach seems likely to be marginally more accurate—though an analogous approach using October 1 as the cutoff date yields substantially similar results.[270]
Selecting Terms for Study
I conclude with a word on how various Index Terms were selected for extended discussion in the Article’s main text. This is an admittedly subjective exercise, one that is susceptible to unavoidable critiques of cherry-picking and prior-confirming. But I limit focus to those terms that appeared in either the main text tables for Importance Scores and Deltas (i.e., to terms that were significant for some period, or that significantly increased or decreased in importance from one period to the next). Moreover, I emphasize terms that are in both tables. That is, if an Index Term was both significantly important during the Roberts Court and experienced a significant change (e.g., increase) in importance from the Rehnquist Court to the Roberts Court (or if an Index Term was both significantly important during the Rehnquist Court and experienced a significant decrease in importance from the Rehnquist Court to the Roberts Court), then I am more likely to emphasize such a term in the narrative descriptions above.
Appendix Tables and Figures
Appendix Table 1. Original Terms and Index Terms
Please see the PDF version of the Article for access to the Appendices.
Appendix Table 2. Stop Words
Please see the PDF version of the Article for access to the Appendices.
Appendix Figure 1. Average Majority Votes by Chief Justice, Important-Questions Docket vs. Conflicts Docket
Please see the PDF version of the Article for access to the Appendices.
Copyright © 2025 Tejas N. Narechania, Professor of Law, University of California, Berkeley, School of Law. I owe many thanks to Deborah Beim, Andrew Crespo, Dan Deacon, Erin Delaney, Judge Jeremy Fogel (ret.), Barry Friedman, Bert Huang, Justice Leondra Kruger, Alli Orr Larsen, Maggie Lemos, Marin Levy, Andrew Martin, Julian Davis Mortenson, Tom Schmidt, and Chas Tyler, as well as audiences at Duke Law School, George Washington University Law School, Northwestern University Law School, the University of California, Berkeley, School of Law, the University of Michigan Law School, and Washington University in St. Louis for their valuable comments and suggestions. I am grateful to Ilya Akdemir, Ankur Jain, and Mathew Cha, whose technical expertise were instrumental to the analyses conducted in this project. For excellent research assistance, I thank Connor Hughes and Rakhi Kundra. I also thank Jordan Barton and the editors of the California Law Review for their excellent suggestions and thoughtful edits.
[1]. Sup. Ct. R. 10(a)–(b); see, e.g., Stephen M. Shapiro, Kenneth S. Geller, Timothy S. Bishop, Edward A. Hartnett & Dan Himmelfarb, Supreme Court Practice §§ 4.3–4.4, (11th ed. 2019); Doris Marie Provine, Case Selection in the United States Supreme Court 39 (1980) (quoting Felix Frankfurter & Henry M. Hart, Jr., The Business of the Supreme Court at October Term, 1933, 48 Harv. L. Rev. 238, 267 (1934) (“[C]onflict ‘remains by far the most frequent ground for granting the writ.’”); H.W. Perry, Jr., Deciding to Decide: Agenda Setting in the United States Supreme Court 246 (1991) (“Without a doubt, the single most important generalizable factor in assessing certworthiness is the existence of a conflict or ‘split’ in the circuits.”); Amanda Frost, Overvaluing Uniformity, 94 Va. L. Rev. 1567, 1631–32 (2008) (“[T]he presence of a conflict remains by far the most important criteria in the Court’s case selection . . . .”).
[2]. See, e.g., Nichols v. United States, 578 U.S. 104, 108 (2016) (describing a circuit split that gave rise to one outcome for a defendant living in “the Kansas City area—on the Missouri side” and a different outcome for a defendant on the Kansas side); see also Tejas N. Narechania, Certiorari in Important Cases, 122 Colum. L. Rev. 923, 925 (2022) (making a similar observation) [hereinafter Narechania, Important Cases].
[3]. Supreme Court Chief Justice Roberts, C-SPAN 18:05–18:25 (June 19, 2009), https://www.c-span.org/video/?286078-1/supreme-court-chief-justice-roberts&start=1077 (explaining that, in his view, the Court’s “main job” is to ensure that “federal law is uniform across the country”); The Federalist No. 82, at 494 (Alexander Hamilton) (Clinton Rossiter ed., 1961). For other, similar comments, see, for example, Sandra Day O’Connor, The Majesty of the Law 211 (Craig Joyce ed., 2003); Fred M. Vinson, Work of the Federal Courts, Address Before the American Bar Association (Sept. 7, 1949) 69 S. Ct. v, vi (“The debates in the Constitutional Convention make clear that the purpose of the establishment of one supreme national tribunal was, in the words of John Rutledge of South Carolina, ‘to secure the national rights and uniformity of [Judgments].’”); Chi.-Kent Coll. of Law at Ill. Inst. of Tech., A Conversation with Justice Elena Kagan, Oct. 16, 2017, Chicago-Kent College of Law, YouTube 23:50–24:05 (Oct. 19, 2017), https://www.youtube.com/watch?time_continue=4&v=MWKA79-3vHA; Stephen G. Breyer, Reflections on the Role of Appellate Courts: A View from the Supreme Court, 8 J. App. Prac. & Proc. 91, 92 (2006) (explaining that “the Supreme Court is charged with providing a uniform rule of federal law in areas that require one”); William H. Rehnquist, The Changing Role of the Supreme Court, 14 Fla. St. U. L. Rev. 1, 11–12 (1986); Byron R. White, The Work of the Supreme Court: A Nuts and Bolts Description, 54 N.Y. St. B.J. 346, 349 (1982) (suggesting that the Court should focus on “provid[ing] some degree of coherence and uniformity in federal law”).
[4]. See, e.g., City & County. of San Francisco v. Sheehan, 575 U.S. 600, 610 (2015) (“[C]ertiorari jurisdiction exists to clarify the law . . . .”); Bullock v. BankChampaign, N.A., 569 U.S. 267, 273, 276 (2013); Thompson v. Keohane, 516 U.S. 99, 106 (1995); Key v. Doyle, 434 U.S. 59, 67–68 (1977); Tidewater Oil Co. v. United States, 409 U.S. 151, 170 (1972); Hanna v. Plumer, 380 U.S. 460, 463 (1965); Magnum Import Co. v. Coty, 262 U.S. 159, 163 (1923); Lau Ow Bew v. United States, 144 U.S. 47, 58 (1892); Martin v. Hunter’s Lessee, 14 U.S. (1 Wheat.) 304, 347–48 (1816) (Story, J.) (“[The Supreme Court’s appellate jurisdiction highlights] the importance, and even necessity of uniformity of decisions throughout the whole United States,” as dis-uniformity “would be truly deplorable.”); see also Kansas v. Marsh, 548 U.S. 163, 183 (2006) (Scalia, J., concurring) (“Our principal responsibility under current practice . . . is to ensure the integrity and uniformity of federal law.”).
[5]. See Wright v. North Carolina, 415 U.S. 936, 936–38 (1974) (Douglas, J., dissenting from denial of certiorari and citing the 1925 Bill); Perry, Jr., supra note 1, at 248; see also Margaret Meriwether Cordray & Richard Cordray, The Philosophy of Certiorari: Jurisprudential Considerations in Supreme Court Case Selection, 82 Wash. U. L.Q. 389, 436–37 (2004) (noting the “explicit commitment” the Justices made to Congress to protect the uniformity of federal law). But see Perry, Jr., supra note 1, at 248 (quoting a Justice as disagreeing with this understanding of the 1925 Judges’ Bill).
[6]. Deborah Beim & Kelly Rader, Legal Uniformity in American Courts, 16 J. Empirical Legal Stud. 448, 449 (2019) (finding that the Supreme Court only resolves about one-third of circuit splits, namely those that the Court finds to be “harmful and important”); see also Kenneth W. Starr, The Supreme Court and its Shrinking Docket: The Ghost of William Howard Taft, 90 Minn. L. Rev. 1363, 1366 (2006) (“[T]he Court is willing to allow conflicts in federal law to exist—and, even worse, to persist.”).
[7]. Margaret Meriwether Cordray & Richard Cordray, The Supreme Court’s Plenary Docket, 58 Wash. & Lee L. Rev. 737, 772 (2001). But see Shannon Duffy, Inside the Highest Court; Souter Describes Justices’ Relationship, Caseload Trend, Pa. L. Wkly., Apr. 17, 1995, at 10 (quoting Justice Souter’s view that greater uniformity in federal appellate decisions has led to fewer circuit splits worthy of the Court’s attention); see also David R. Stras, The Supreme Court’s Declining Plenary Docket: A Membership-Based Explanation, 27 Const. Comment. 151, 161 (2010) (similar).
[8]. In addition to the implication I highlight above—the difference between a Court that focuses on constitutional questions with broad social implications and one that focuses on comparatively narrow or technical questions of statutory interpretation—I should also note one more. The Supreme Court’s preferences may echo in the appeals courts’ own decisional rules. For example, the federal courts of appeals may, in certain cases, defer to its colleagues on another appellate bench if the Supreme Court declares uniformity to be especially important in such cases. Stated otherwise, when the Court appears to demand uniformity in one area of law, the appeals courts may be more likely to follow suit. Compare, e.g., Comm’r v. Bilder, 369 U.S. 499, 501 (1962) (granting certiorari in view of “the need for a uniform rule” on a matter of tax law), with First Charter Fin. Corp. v. United States, 669 U.S. 1342, 1345 (9th Cir. 1982) (“Uniformity among the circuits is especially important in tax cases to ensure equal and certain administration of the tax system. We would therefore hesitate to reject the view of another circuit.”), and N. Am. Life & Cas. Co. v. Comm’r, 533 F.2d 1046, 1051 (8th Cir. 1976) (“[U]niformity of decision among the circuits is vitally important on issues concerning the administration of the tax laws . . . . Thus, decisions of other courts of appeals in the area of taxation should be followed unless they are demonstrably erroneous . . . .”). And, likewise, perhaps where the Court is more willing to let splits linger, the circuits may be more willing to splinter. See Brief in Opposition to a Petition for Certiorari at 15–16, Renaissance Art Invs. v. AXA Art Ins. Corp. (No. 13-438), 2013 WL 5936536 (noting the impact of the Court’s determination that a uniform federal rule was unnecessary on the circuit courts); see also Brief in Opposition by Respondent E. Pihl in Bankruptcy at 4, Carswell v. E. Pihl & Son. (No. 21-675), 2022 WL 94483 (similar); cf. United States v. Kimbell Foods, 440 U.S. 715, 729 (1979) (concluding that “uniform . . . rules of priority” are not required for the “effective administration of [the federal government’s] lending programs”); Johnson v. U.S. Dep’t of Ag., 734 F.2d 774, 785 (11th Cir. 1984) (citing Kimbell Foods and similarly “perceiv[ing] no need for national uniformity in the method of foreclosure on [certain federal] housing loans” and so incorporating state rules of decisions). I develop this observation more fully in a forthcoming project, Uniformity Rules (manuscript on file with author).
[9]. See, e.g., Tejas N. Narechania, Certiorari in the Roberts Court, 67 St. Louis. L.J. 587, 589 (2023) (describing some of the legal, practical, institutional, and political ramifications of one of the Court’s certiorari decisions) [hereinafter Narechania, Roberts Court].
[10]. See, e.g., Mendoza v. Lumpkin, No. 23-1004, 2024 WL 4426541 (Oct. 7, 2024) (mem.) (denying certiorari in a case presenting an “open and acknowledged ‘circuit split’” over the meaning of the Antiterrorism and Effective Death Penalty Act).
[11]. See Daniel Epps & William Ortman, The Lottery Docket, 116 Mich. L. Rev. 705, 708 (2018) (advocating for a Supreme Court docket that includes a random sample of “the mine-run of cases”).
[12]. I refer to splits and conflicts interchangeably throughout this Article.
[13]. Narechania, Important Cases, supra note 2, at 934 (explaining that we can “learn a lot about the Supreme Court and its docket by simply looking to the Court’s own descriptions of its decisionmaking criteria”).
[14]. Id. at 932 (explaining that, although the Court’s certiorari canons are set out in “terse, citationless text” that “resist[s] traditional doctrinal evaluation,” we might nevertheless analyze this opinion text as data).
[15]. Id. at 934; see also Narechania, Roberts Court, supra note 9, at 592 (updating the previous study).
[16]. See Narechania, Important Cases, supra note 2, at 929 (introducing the distinction between these two dockets); see also Richard L. Pacelle, Jr., The Transformation of the Supreme Court’s Agenda: From the New Deal to the Reagan Administration 28 (1991) (describing the Court’s agenda as “bifurcated” into a “volitional agenda,” analogous to the important-questions docket, and an “exigent agenda,” analogous to the conflicts docket).
[17]. For tax cases, see, for example, Comm’r v. Bilder, 369 U.S. 499, 501 (1962) (granting certiorari in view of “the need for a uniform rule” on a matter of tax law); Turnbow v. Comm’r, 368 U.S. 337, 339 (1961) (similar). For Federal Tort Claims Act cases, see, for example, Richards v. United States, 369 U.S. 1, 5 (1962) (“It was to resolve the three-fold conflict and to enunciate a rule that can be applied uniformly in Tort Claims Act cases that we granted certiorari.”).
[18]. See, respectively, RJR Nabisco v. Eur. Cmty., 579 U.S. 325, 335 (2016); Nasrallah v. Barr, 590 U.S. 573, 578 (2020).
[19]. See supra note 1.
[20]. See Sup. Ct. R. 10(c).
[21]. Michael Livermore, Allen Riddell & Daniel Rockmore, The Supreme Court and the Judicial Genre, 59 Ariz. L. Rev. 837, 871 (2017).
[22]. Beim & Rader, supra note 6, at 448.
[23]. E.g., Arthur D. Hellman, The Shrunken Docket of the Rehnquist Court, 1996 Sup. Ct. Rev. 403, 415–16 (1996); S. Sidney Ulmer, The Supreme Court’s Certiorari Decisions: Conflict as a Predictive Variable, 78 Am. Pol. Sci. Rev. 901, 910 (1984); The Supreme Court’s Certiorari Jurisdiction: Cue Theory, in Jud. Decision-Making 111, 118–19 (Glendon Schubert ed., 1963); see also supra note 1.
[24]. See generally, e.g., Michael Livermore & Daniel Rockmore, eds., Law as Data (2019) (compiling legal scholarship that uses data-driven analysis).
[25]. I use an italicized font, like this, to refer to terms that appear in the results of the data analysis described and presented below. Moreover, as I explain below, I use an asterisk (*) to denote terms that encompass multiple related terms. This nomenclature is roughly derived from some computational contexts, where asterisks are used as wildcard characters (and so asterisks do not indicate levels of significance). But, as I elaborate below, such asterisks are not strictly used as wildcard characters here. See infra note 103 and accompanying text. For more on significance testing in these results, see the Methods Appendix.
[26]. I clarify that, here, the word “importance” in “Importance Score” refers to the importance of a term as to the Court’s conflicts docket. It does not, in this context, refer to the important-questions docket. However, in earlier studies, the “Importance Score” referred to the importance of a term as to the Court’s important-questions docket. See, e.g., Narechania, Important Cases, supra note 2, at 934.
[27]. See Cohens v. Virginia, 19 U.S. 264, 404 (1821) (“It is most true that this Court will not take jurisdiction if it should not: but it is equally true, that it must take jurisdiction if it should . . . . We have no more right to decline the exercise of jurisdiction which is given, than to usurp that which is not given.”).
[28]. See Tara Leigh Grove, The Exceptions Clause as Structural Safeguard, 113 Colum. L. Rev. 929, 958 (2013) (discussing the expansion of the Court’s appellate review power in 1891) [hereinafter Grove, Exceptions Clause].
[29]. Felix Frankfurter & James M. Landis, The Business of the Supreme Court: A Study in the Federal Judicial System 60 (1928).
[30]. See Statistics, SCOTUSblog, https://www.scotusblog.com/statistics/ [https://perma.cc/K4XX-HH9Q] (noting that the Court authored fifty-nine majority opinions in its 2023 Term).
[31]. H.R. Rep. No. 51-1295, at 3 (1890) (noting that, according to Justice Harlan, the Court addressed 451 appeals in 1886).
[32]. Act of Mar. 3, 1891, 26 Stat. 828.
[33]. Id.
[34]. Act of Mar. 3, 1891, ch. 517, § 6, 26 Stat. 828.
[35]. Grove, Exceptions Clause, supra note 28, at 958.
[36]. See Tara Leigh Grove, Tiers of Scrutiny in a Hierarchical Judiciary, 14 Geo. J.L. & Pub. Pol’y 475, 480 (2016) (explaining that the Court’s docket “continued to swell” after the 1891 Act) [hereinafter Grove, Scrutiny Tiers]. One partial explanation for the continued growth of the Court’s docket lies in the design of the 1891 Act, which, while giving the Court discretionary jurisdiction over several important classes of cases (such as diversity-jurisdiction cases), also expanded the Court’s mandatory appellate jurisdiction in a number of ways, including by “reducing the amount-in-controversy requirement [for federal-question cases] from $5,000 to $1,000” and by “significantly expanding the Court’s jurisdiction in criminal cases.” Grove, Exceptions Clause, supra note 28, at 958.
[37]. Judiciary Act of 1925, Pub. L. 68-415, 43 Stat. 936, 936-37.
[38]. See Grove, Exceptions Clause, supra note 28, at 966–68.
[39]. Grove, Scrutiny Tiers, supra note 36, at 481 n.27 (collecting statements from Members of the Court advocating for the statute and explaining that the Court would continue to ensure—and that the new statute would aid the cause of—uniformity).
[40]. Wright v. North Carolina, 415 U.S. 936, 1453 (1974) (Douglas, J., dissenting from denial of certiorari); see Cordray & Cordray, supra note 5, at 436–37 (2004) (“[S]ome believe that the legislation was based on an explicit commitment that the Justices made to Congress to protect the uniformity of federal law in return for Congress’ ceding the Court so much control over case selection.”).
[41]. Robert L. Stern, Eugene Gressman & Stephen M. Shapiro, Epitaph for Mandatory Jurisdiction, 74 A.B.A J. 66 (Dec. 1988) (describing the Supreme Court Case Selections Act of 1988, Pub. L. 100–352, 102 Stat. 662 (1988)). One notable exception from the Supreme Court’s discretionary jurisdiction is its continued mandatory jurisdiction over direct appeals from three-judge district courts. 28 U.S.C. § 1253 (2012). Such courts are now used primarily to address certain questions of electoral redistricting. See Act of Aug. 12, 1976, Pub. L. No. 94-381, 90 Stat. 1119 (codified in part at 28 U.S.C. § 2284); see also Shapiro et al., supra note 1, at 16; Michael E. Solimine & James L. Walker, The Strange Career of the Three-Judge District Court: Federalism and Civil Rights, 1954–1976, 72 Case W. Rsrv. L. Rev. 909, 955-59 (2022).
[42]. See supra note 1 (collecting sources making this point).
[43]. See Narechania, Roberts Court, supra note 9, at 604; Narechania, Important Cases, supra note 2, at 934.
[44]. Bosse v. Oklahoma, 580 U.S. 1, 2 (2016) (“It is this Court’s prerogative alone to overrule one of its precedents.”) (alterations omitted). But cf. Amy Coney Barrett, Precedent and Jurisprudential Disagreement, 91 Tex. L. Rev. 1711, 1731 (2013) (suggesting that conflicts among lower courts may help “put[] a challenge to precedent on the Court’s agenda”).
[45]. Narechania, Important Cases, supra note 2, at 924.
[46]. Tejas N. Narechania, Certiorari, Universality, and a Patent Puzzle, 118 Mich. L. Rev. 1345, 1357 (2018) [hereinafter Narechania, Patent Puzzle].
[47]. William J. Brennan, Jr., The National Court of Appeals: Another Dissent, 40 U. Chi. L. Rev. 473, 477, 483 (1973); Narechania, Patent Puzzle, supra note 46, at 1357.
[48]. Margaret Meriwether Cordray & Richard Cordray, Setting the Social Agenda: Deciding to Review High-Profile Cases at the Supreme Court, 57 U. Kan. L. Rev. 313, 313, 318 (2009) (quoting Eugene Gressman, The National Court of Appeals: A Dissent, 59 A.B.A. J. 253, 255 (1973)). At the very least, such decisions help reveal the priorities of at least four Justices (because granting review requires only four votes). Rogers v. Missouri Pac. R.R. Co., 352 U.S. 521, 529 (1957) (Frankfurter, J., dissenting) (describing the “rule of four”).
[49]. See, e.g., The Federalist No. 82 (Alexander Hamilton) (Clinton Rossiter, ed., 1961) (“[The Supreme Court] is destined to unite and assimilate the principles of national justice and the rules of national decisions.”). For a contrary take on the normative importance of uniformity, see Amanda Frost, Overvaluing Uniformity, 94 Va. L. Rev. 1567 (2008).
[50]. Kansas v. Marsh, 548 U.S. 163, 183 (2006) (Scalia, J., concurring); see also the sources cited supra notes 3–4 (collecting decisions of the Supreme Court, as well as statements by its individual Members, emphasizing the importance of uniformity).
[51]. See Narechania, Patent Puzzle, supra note 46, at 1362 & nn.81–84 (describing these values and collecting sources).
[52]. See Narechania, Important Cases, supra note 2, at 938 n.71.
[53]. See Shapiro et al., supra note 1, at §§ 4.3–4.4, 4.7, 4.9; see also, e.g., Maryland v. King, 567 U.S. 1301, 1302 (2012) (Roberts, C.J., in chambers) (noting the likelihood that the Court will grant certiorari in view of a conflict between the Virginia Supreme Court and the Maryland Court of Appeals, among other courts).
[54]. Turnbow v. Comm’r, 368 U.S. 337, 337 (1961); see also Comm’r v. Bilder, 369 U.S. 499, 501 (1962) (similar).
[55]. See supra note 54 and accompanying text.
[56]. Starr, supra note 6, at 1372 (suggesting that the Court lets too many circuit splits fester).
[57]. Beim & Rader, supra note 6, at 462. I note that these studies are all, to varying degrees, subject to the difficulty of objectively determining what counts as a circuit split. See, e.g., Aaron-Andrew P. Bruhl, Measuring Circuit Splits: A Cautionary Note, 4 J. Legal Metrics 361, 362–63, 365–66, 372–76 (2014).
[58]. Emily Grant, Scott A. Hendrickson & Michael S. Lynch, The Ideological Divide: Conflict and the Supreme Court’s Certiorari Decision, 60 Clev. St. L. Rev. 559, 575 (2012).
[59]. A note on terminology: I use the term “lower court,” despite having long ago (as a law student) been cautioned away from the phrase by Judge Sack (among other, similar admonitions at later points in my career) for its implicit pejorative connotation, both because of its ubiquity in the literature and for lack of a better term. See, e.g., U.S. Court of Appeals for the Federal Circuit, Internal Operating Procedure No. 11 (9) (“Respect for the tribunals from whose judgments and decisions appeals are taken to this court requires care in referring to those tribunals in our opinions. Reference should not be made to ‘the court below,’ ‘the lower court,’ ‘the lower tribunal,’ or ‘the judge below.’”). And so I clarify that I mean no disrespect by using this disfavored term. I do not mean to denigrate these courts by using it; rather, I use the term simply to describe these courts’ place in the Judiciary’s organizational chart.
[60]. Tom S. Clark & Jonathan P. Kastellac, The Supreme Court and Percolation in the Lower Courts, 75 J. Pol. 150, 162 (2013) (concluding that “conflicts . . . that run against the Supreme Court’s inherent ideological preferences will push the Court to intervene earlier” than it otherwise might have); see also Álvaro Bustos & Tonja Jacobi, Judicial Choice among Cases for Certiorari, 27 Sup. Ct. Econ. Rev. 117, 117 (2019) (concluding that the Supreme Court tends to “take[] the case that will most significantly shape future lower court case outcomes in the direction that the Court prefers”).
[61]. Grant et al., supra note 58, at 575.
[62]. Arthur D. Hellman, By Precedent Unbound: The Nature and Extent of Unresolved Intercircuit Conflicts, 56 U. Pitt. L. Rev. 693, 732-33 (1995); see also Clark & Kastellac, supra note 60, at 164.
[63]. Hellman, supra note 62, at 733.
[64]. See Beim & Rader, supra note 6, at 457, 462.
[65]. Id.
[66]. See Marin K. Levy & Tejas N. Narechania, Interbranch Information Sharing: Examining the Statutory Opinion Transmission Project, 108 Calif. L. Rev. 917, 935 n.91 (2020) (describing the conflict at issue in Fourth Estate Pub. Benefit Corp. v. Wall-Street.com, LLC, 586 U.S. 296 (2019)).
[67]. See Wayne A. Logan, Constitutional Cacophony: Federal Circuit Splits and the Fourth Amendment, 65 Vand. L. Rev. 1137, 1155–56, App’x A (noting a conflict that has persisted for nearly forty years).
[68]. Karen M. Gebbia, Certiorari and the Bankruptcy Code: The Statutory Interpretation Cases, 90 Am. Bankr. L.J. 503 (2016); see also Anthony J. Casey & Joshua C. Macey, Bankruptcy Shopping, 37 Emory Bankr. Devs. J. 463, 497 (2021) (“The Supreme Court has granted certiorari to resolve circuit splits in a number of bankruptcy cases in the past few years . . . . On the other hand, several substantive splits have remained unresolved for years and, in some cases, decades.”).
[69]. See, e.g., Commission on Revision of the Federal Court Appellate System, Inter-Circuit Conflicts, in A National Court of Appeals in Structure and Internal Procedures: Recommendations for Change at 17 (1975); Robert M. Lawless & Dylan Lager Murray, An Empirical Analysis of Bankruptcy Certiorari, 62 Mo. L. Rev. 101, 108 (1997).
[70]. Stuart H. Teger & Douglas Kosinski, The Cue Theory of Supreme Court Certiorari Jurisdiction: A Reconsideration, 42 J. Pol. 834, 839 (1980).
[71]. See generally Matthew L.M. Fletcher, Factbound and Splitless: The Certiorari Process as a Barrier to Justice for Indian Tribes, 51 Ariz. L. Rev. 933 (2009).
[72]. Beim & Rader, supra note 6, at 463; see also Wayne A. Logan, Constitutional Cacophony: Federal Circuit Splits and the Fourth Amendment, 65 Vand. L. Rev. 1137, 1194 (2012).
[73]. Teger & Kosinski, supra note 70, at 841.
[74]. Some of these observations, however, are more difficult to test using the methodology I describe below. We would not, for example, expect the Supreme Court to concede that it granted review in a given case because of an ideological disagreement with that case’s outcome (even if that were so)—and so such concerns are beyond my present scope.
[75]. See Beim & Rader, supra note 6, at 463.
[76]. Narechania, Patent Puzzle, supra note 46, at 1349 (distinguishing between geographic uniformity and legal universality).
[77]. See, e.g., Orin Kerr (@OrinKerr), x (Sept. 13, 2022), https://twitter.com/OrinKerr/status/1569758759379148802 [https://perma.cc/UXV4-QCDH] (“There are so many circuit splits on how the Fourth Amendment applies, and yet there were no Fourth Amendment cases on the Supreme Court docket last year and there are no such cases on the Court’s docket for next year.”).
[78]. See supra note 17 and accompanying text.
[79]. See H.R. Rep. No. 97-312, at 23 (1981); S. Rep. No. 97-275, at 5–6 (1981).
[80]. Narechania, Important Cases, supra note 2, at 939–40 & nn.76–83 (collecting similar examples from important-questions cases, rather than conflicts cases).
[81]. Concrete Appliances Co. v. Gomery, 269 U.S. 177, 178 (1925). Notably, the Court’s decision in Gomery cites a prior case, Thomson Co. v. Ford Motor Co., 265 U. S. 445 (1924), to support the proposition that such a conflict merits review. See Narechania, Important Cases, supra note 2, at 932 n.36 (citing United States v. Kebodeaux, 570 U.S. 387, 391 (2013) as one “exceedingly rare” example of the modern Court explaining, and perhaps developing, its certiorari standard in the mode of common law reasoning, whereas Gomery is one additional, if much older, example).
[82]. McFeely v. Comm’r., 296 U.S. 102, 105 (1935).
[83]. 323 U.S. 126, 126-27 (1944).
[84]. Johansen v. United States, 343 U.S. 427, 430 (1952).
[85]. McCarthy v. United States, 394 U.S. 459, 468-69 (1969).
[86]. United States v. Eight Thousand Eight Hundred & Fifty Dollars ($8,850) in U.S. Currency, 461 U.S. 555, 562 (1983).
[87]. Exxon Corp. v. Central Gulf Lines, 500 U.S. 603, 607-08 (1991); Burns v. United States, 501 U.S. 129, 131–33 (1993) (explaining that the Court “granted certiorari” to resolve a conflict over whether and when a court must provide notice to criminal defendants of “its intent sua sponte to depart upward from an applicable Guidelines sentencing range”).
[88]. 596 U.S. 36, 40 (2022); see also Hartman v. Moore, 547 U.S. 250, 255-56 (2006) (“The Courts of Appeals have divided on the issue of requiring evidence of a lack of probable cause in 42 U.S.C. § 1983 and Bivens retaliatory-prosecution suits . . . . We granted certiorari, to resolve the Circuit split . . . .”).
[89]. Exxon Corp., 500 U.S. 603; see Barrett, supra note 44, at 1731–32 (2013) (suggesting that conflicts among lower courts may help “put[] a challenge to precedent on the Court’s agenda”).
[90]. Many legal scholars, of course, are employing similar or related approaches to study and understand a range of terms in a range of contexts. See, e.g., Livermore et al., supra note 21; see also James Hicks, Informative Patents? Predicting Invalidity with the Text of Claims (2020) (working paper) (using text analysis methods to show that the content of patent claims can be a strong predictor of invalidity decisions); Jennifer Mascott, Who Are “Officers” of the United States, 70 Stan. L. Rev. 443 (2018) (using text analysis approaches to study the meaning of “officer” in the Founding Era); Stephen C. Mouritsen, The Dictionary Is Not a Fortress: Definitional Fallacies and a Corpus-Based Approach to Plain Meaning, 2010 B.Y.U. L. Rev. 1915 (2011) (proposing a method for determining the meaning of statutory terms grounded in corpus linguistics).
[91]. The description of my method in this section draws heavily from my prior work. See Narechania, Important Cases, supra note 2, at 940-53.
[92]. See 39 Stat. 726 (1916) (providing that the “Supreme Court shall hold . . . one term annually, commencing on the first Monday in October”); 43 Stat. 936 (1925) (Judges’ Bill of 1925).
[93]. I provide more detail on how I treated each of the thirteen coded values in the Methods Appendix.
[94]. The Methods Appendix offers more detail on how relevant text was extracted.
[95]. Felix Frankfurter & Henry M. Hart, Jr., The Business of the Supreme Court at October Term 1934, 49 Harv. L. Rev. 68, 82–83 (1935). I acknowledge that some readers may be skeptical that these paragraphs merit the attention that Felix Frankfurter gave them in the 1930s and that I give them here, contending that these paragraphs are barely given a second thought in the Justices’ own processes of drafting, reviewing, and revising their opinions—and thus are hardly scrutinized by the Court as a whole. I respectfully disagree, for reasons I have described in detail in earlier work. See, e.g., Narechania, Important Cases, supra note 2, at 946 n.101.
[96]. One example of a case coded as having offered no reason for the grant of review, but including terms suggestive of its place on the Court’s conflicts docket, is Graham v. John Deere Co., 383 U.S. 1, 4 (1966) (noting that the case “presents a conflict between two Circuits over the validity of a single patent”); see also O’Donnell v. Elgin J. & E. Ry. Co., 338 U.S. 384, 387 (1949) (noting a conflict over the application of the “Safety Appliance Act”).
[97]. See Narechania, Important Cases, supra note 2, at 966-68.
[98]. See id. at 947–48.
[99]. Notably, this percentage is almost identical to the figure from my prior study. See id. at 1009. Moreover, as before, most of the 792 cases that did not include the term “certiorari” and were therefore ultimately excluded from the final dataset were coded in the Supreme Court Database as providing no reason for the grant. Only forty-one of the excluded cases came from cases coded as offering some conflict- or confusion-related reason for review.
[100]. See, e.g., CSX Transp. Inc. v. Ala. Dep’t of Rev., 562 U.S. 277, 303 n.3 (2011) (Thomas, J., dissenting) (“The majority declines to reach the comparison class issue . . . . The lower courts have split over the proper scope of the comparison class, and the issue was presented in this case. I would decide it.”); see also United States v. Dalm, 494 U.S. 596, 614 n.2 (1990) (Stevens, J., dissenting) (disagreeing with the majority’s characterization of a conflict over a tax law matter).
[101]. See supra note 94 and accompanying text.
[102]. See U.S. Dep’t of Just. v. Provenzano, 469 U.S. 14, 14 (1984). In Provenzano, the Court’s opinion contains a certiorari paragraph noting a conflict on the Freedom of Information Act. Id. In subsequent paragraphs, it goes on to note that the question presented has been mooted by a change in statutory law. Id. at 15. A research design that looked beyond certiorari paragraphs might have mistakenly attributed the case’s discussion of mootness to the decision to grant review. But a research design that is limited to certiorari paragraphs may avoid making that mistake (though, perhaps, at the cost of missing other pertinent certiorari signals in other cases). For more, see Narechania, Important Cases, supra note 2, at 947–48 & n.106.
[103]. The Methods Appendix describes in greater detail how I generated the Term Index. In short, it is derived from the most common words in the corpus (while excluding those terms that do not seem to advance our understanding of the Court’s certiorari process, such as granted). And, as explained in the Methods Appendix, asterisked Index Terms represent multiple words (etymologically similar words or close synonyms). For example, employ* encompasses employee, employees, employment, and worker.
[104]. E.g., Tomas Mikolov, Kai Chen, Greg Corrado & Jeffrey Dean, Efficient Estimation of Word Representations in Vector Space, 3 (2013) (noting that the frequency of words is a useful metric for certain text analysis methods); Peter D. Turney & Patrick Pantel, From Frequency to Meaning: Vector Space Models of Semantics, 37 J. A.I. Rsch. 141, 166 (2010) (noting the use of frequency and proximity queries in natural language processing).
[105]. See supra notes 25, 103 (clarifying the meaning of asterisked terms).
[106]. Landers v. Nat’l R.R. Passengers Corp., 485 U.S. 652, 654 (1988) (“We granted certiorari to resolve the conflict between two Courts of Appeals over this question of federal railway labor law.”) (internal citation omitted); United States v. Union Cent. Life Ins., 368 U.S. 291, 293 (1961) (“In order to settle this conflict and because of the importance of the question in the administration of the revenue laws, we granted certiorari.”).
[107]. See supra note 26 (clarifying the term “Importance Score”).
[108]. See infra note 263 (describing these Focal Terms).
[109]. One could easily imagine extending this method to other terms and contexts of interest. For example, one might examine a body of state tort law cases and identify terms associated with reasonabl*, in order to study how that state’s tort law has changed over time.
[110]. My previous work has used an external approach, examining petitions for writs of certiorari to try to discern which petitions seem to attract the Court’s attention. See Narechania, Patent Puzzle, supra note 46, at 1348; see also Livermore et al., supra note 21, at 874 (noting other analyses based on external signals).
[111]. See Narechania, Important Cases, supra note 2, at 984-89 (advocating for a more explicit, and more binding, common law of certiorari).
[112]. See infra note 129 and accompanying text (discussing one further limitation with this focus on frequency and proximity and one way that the method employed addresses that limitation).
[113]. See Narechania, Important Cases, supra note 2, at 968-84 (employing a similar approach).
[114]. See, e.g., Petition for a Writ of Certiorari, United States v. Rahimi, 143 S. Ct. 2688 (2023) (No. 22-915) (mem.) (granting certiorari), 2023 WL 2600091 at *14 (noting conflict in a Second Amendment case). I emphasize that Rahimi is outside the scope of my dataset.
[115]. See generally Teger & Kosinski, supra note 70.
[116]. See Justin Grimmer & Brandon M. Stewart, Text as Data: The Promise and Pitfalls of Automatic Content Analysis Methods for Political Texts, 21 Pol. Analysis 267, 270 (2013) (“Automated content analysis methods . . . will not . . . eliminate the need for careful thought by researchers nor remove the necessity of reading texts.”).
Indeed, even the different choice to use automated tools to construct a Term Index, rather than the manual approach used here, would reflect some subjective selection—both the subjective judgment that automated tools are better than manual selection informed by domain expertise, as well as whatever (possibly unknowable) subjective biases that may inform the underlying automated process.
[117]. See Frederick Schauer, Is it Important to be Important?: Evaluating the Supreme Court’s Case-Selection Process, 119 Yale L.J.F. 77, 77 (2009) (explaining that the Court’s preferences might be shaped by a skewed perception of the universe of disputes).
[118]. See Opinions Relating to Orders—2023, Sup. Ct. of the U.S., https://www.supremecourt.gov/opinions/relatingtoorders [https://perma.cc/FN4U-5HLG] (“Opinions may be written by Justices to comment on the summary disposition of cases by orders, e.g., if a Justice wants to dissent from the denial of certiorari or concur in that denial.”); see generally Barry P. McDonald, SCOTUS’s Shadiest Shadow Docket, 56 Wake Forest L. Rev 1021 (2021) (studying these opinions dissenting from the denial of certiorari, and similarly adopting the Court’s nomenclature of “Opinions Relating to Orders”).
[119]. See, e.g., Kathryn A. Watts, Constraining Certiorari Using Administrative Law Principles, 160 U. Pa. L. Rev. 1, 17 (2011); Narechania, Important Cases, supra note 2; William Baude, Foreword: The Supreme Court’s Shadow Docket, 9 N.Y.U. J.L. & Liberty 1, 5–6, 16–18 (2015).
[120]. Breyer, supra note 3, at 92.
[121]. U.S. Const. art. I, § 8, cl. 1, 4.
[122]. Act of Mar. 3, 1891, ch. 517, § 6, 26 Stat. 828.
[123]. Beim & Rader, supra note 6, at 462.
[124]. See, e.g., Kennedy v. Bremerton Sch. Dist., 597 U.S. 507 (2022). Except as expressly noted, every Supreme Court case cited in this Part is drawn from my results.
[125]. Cf. Levy & Narechania, supra note 66, at 939 (“One judge’s dissenting opinion may become the majority view in another circuit.”).
[126]. See, e.g., Diane P. Wood, When to Hold, When to Fold, and When to Reshuffle: The Art of Decisionmaking on a Multi-Member Court, 100 Calif. L. Rev. 1445, 1465–66 (2012); Patricia M. Wald, The Rhetoric of Results and the Results of Rhetoric: Judicial Writings, 62 U. Chi. L. Rev. 1371, 1410–11 (1995); Bernice B. Donald, The Intrajudicial Factor in Judicial Independence: Reflections on Collegiality and Dissent in Multi-Member Courts, 47 U. Mem. L. Rev. 1123, 1130 (2017); Michael Boudin, Friendly, J., Dissenting, 61 Duke L.J. 881, 881 (2012); Alex Kozinski & James Burnham, I Say Dissental, You Say Concurral, 121 Yale L.J. F. 601, 602 (2011-2012); Kermit V. Lipez, Some Reflections on Dissenting, 57 Me. L. Rev. 313, 322 (2005); see also Antonin Scalia, The Dissenting Opinion, 1994 J. Sup. Ct. Hist. 33, 37 (1994) (“[A] dissent is also a warning flag to the Supreme Court . . . evidence that the legal issue is a difficult one worthy of the Court’s attention.”); cf. Leroy Rountree Hassell, Sr., Appellate Dissent: A Worthwhile Endeavor or an Exercise in Futility?, 47 How. L.J. 383, 388 (2004); Francis P. O’Connor, The Art of Collegiality: Creating Consensus and Coping with Dissent, 83 Mass. L. Rev. 93, 96 (1998) (“The Massachusetts [Supreme Judicial Court] looks to the published opinions of the Massachusetts Appeals Court and especially to the reasoning not only of that court but also of dissenting opinions.”); Eva M. Guzman & Ed Duffy, The (Multiple) Paths of Dissent: Roles of Dissenting Judges in the Judicial Process, 97 Judicature 105, 107 (2013); but see Joseph v. United States, 574 U.S. 1038 (2014) (Kagan, J., respecting denial of certiorari) (suggesting that intracircuit division is usually not enough to warrant certiorari). Together, these results suggest that, for a lower court judge looking to induce further review of a case or question, it may be more effective to write a separate opinion describing why circuit precedent is incorrect than to try to create an intracircuit split. (I clarify that Joseph does not derive from my dataset. See supra note 123.)
[127]. See, respectively, Marcia Coyle, Despite Docket Drop, Justices Face Packed Homestretch, Nat’l L.J. (June 6, 2007), https://www.law.com/almID/900005482945 [https://perma.cc/L9Q3-Y32L] (reporting on Chief Justice Roberts’s speech to the Alaska Bar Association); Andrew Hamm, Breyer on the Court’s Shrinking Docket, Foreign Law, and More, SCOTUSblog (Sept. 17, 2015), https://www.scotusblog.com/2015/09/breyer-on-the-courts-shrinking-docket-foreign-law-and-more [https://perma.cc/3YDF-49WP] (reporting on Justice Breyer’s book talk at Sixth & I Synagogue, in Washington, D.C.); see also Procedure in Federal Courts, Hearing Before A Subcommittee of the Committee on the Judiciary, 68th Cong. 1 (1924) (statement of McReynolds, J.) (“About 1904, . . . Congress provided for taxing adulterated butter. That statute was construed in the Sixth Circuit. Several years afterwards the Eighth Circuit . . . arrived at a different conclusion. It came to us upon certiorari . . . . Congress is passing many statutes, which must be construed and applied. The number of federal questions is constantly increasing. There are now so many . . . .”).
[128]. John Paul Stevens, Some Thoughts on Judicial Restraint, 66 Judicature 177, 183 (1982) (“If the source of the conflict is ambiguity resulting from an omission in a statute, it would seem to make good sense to assign Congress the task of performing the necessary corrective lawmaking.”).
[129]. See Grimmer & Stewart, supra note 116, at 270 (“Automated content analysis methods . . . will not . . . eliminate the need for careful thought by researchers nor remove the necessity of reading texts.”). Hence, we cannot base our conclusions on only the terms in this list, though my method does try to account for various concerns. For instance, corporate* will capture cases raising questions of corporate law, but it may also capture cases involving a corporate party (without respect to the nature of the question at issue). The method accounts for such concerns (which may extend to like terms, e.g., admin_agency*) through the proximity dimension of the measure employed: In cases where the corporate status of the parties is not germane to the question presented, we would generally not expect it to appear near the Court’s description of the question presented, and so it is de-emphasized along the proximity axis. See, e.g., Narechania, Important Cases, supra note 2, at 956 n.135 (making the same point, contrasting Texaco v. Dagher, 547 U.S. 1, 5 (2006) (“grant[ing] certiorari to determine the extent to which the per se rule against price fixing applies to an important and increasingly popular form of business organization, the joint venture,” suggesting that a business law question is central to the grant of certiorari) with Puerto Rico Dep’t of Consumer Affs. v. Isla Petrol., 485 U.S. 495, 499 (1988) (noting that “several oil companies, brought actions that were consolidated . . . alleging, inter alia, that [the challenged] orders were unconstitutional on pre-emption grounds . . . . [B]ecause of the importance of the issue presented we granted the petition for certiorari,” suggesting that constitutional, preemption questions were more central to the Court’s certiorari grant)). I clarify that the examples in the preceding explanatory parenthetical are drawn from that earlier work and not from the result set presented here. I investigate the clues from my present result set in more detail below.
[130]. See Stuart Taylor, Jr., Powell on His Approach: Doing Justice Case by Case, N.Y. Times (July 12, 1987), https://www.nytimes.com/1987/07/12/us/powell-on-his-approach-doing-justice-case-by-case.html [https://perma.cc/X636-P79N] (quoting Justice Powell); see also Robert A. Green, Justice Blackmun’s Federal Tax Jurisprudence, 26 Hastings Const. L.Q. 109, 109 n.1 (1998) (collecting sources).
[131]. U.S. Const. art. I, § 8, cl. 1, 4.
[132]. Cf. Narechania, Roberts Court, supra note 9, at 603-04 & n.66.
[133]. Erie R.R. Co. v. Tompkins, 304 U.S. 64 (1938). I note that Erie is not drawn from my result set.
[134]. I say perhaps because much depends on whether those early cases arose under the Court’s certiorari jurisdiction and if those cases were granted to address a conflict. Moreover, I note that more cases need not yield a greater diversity of terms—as Table 2 infra suggests, the Court’s selection of conflicts cases in these earlier eras appears to have concentrated in significant part on fewer issue areas (as compared to later eras).
[135]. See generally Narechania, Important Cases, supra note 2; see also, e.g., Harold Spaeth, Lee Epstein, Ted Ruger, Jeffrey Segal, Andrew D. Martin & Sara C. Benesh, 2020 Supreme Court Database Codebook (ver. 01, 2024), at http://www.supremecourtdatabase.org; Lee Epstein, William M. Landes & Richard Posner, When It Comes To Business, The Right And Left Sides of the Court Agree, 54 Wash. U. J.L. Pol. 33 (2017); see generally Tonja Jacobi & Matthew Sag, Taking Laughter Seriously at the Supreme Court, 72 Vand. L. Rev. 1423 (2019).
[136]. See, e.g., Roman L. Hruska et al., Commission On Revision Of The Federal Court Appellate System, Structure And Internal Procedures: Recommendations For Change 1, 162-68 (1975).
[137]. Federal Courts Improvement Act of 1982, Pub. L. No. 97-164, 96 Stat. 25.; see also Rep. No. 97-312, at 23 (1981); S. Rep. No. 97-275, at 5–6 (1981). Notably, the Hruska Commission’s report, supra note 136, advocated against a patent-specific court and for a national court of appeals, reasoning that such a court, charged with addressing the growing number of unresolved circuit splits, would adequately address forum shopping concerns across issue areas. See, e.g., Peter Menell & Ryan Vacca, Revisiting and Confronting the Federal Judiciary Capacity “Crisis”: Charting a Path for Federal Judiciary Reform, 108 Calif. L. Rev. 789, 816 (2020).
[138]. As I explain in the Methods Appendix, the terms highlighted for inclusion evince a change in position that is greater than 1 standard deviation (in either a positive or negative direction) from the mean of all the changes across all terms from one Court to the next.
[139]. See Act of Mar. 3, 1891, ch. 517, § 6, 26 Stat. 828.
[140]. Cf. Narechania, Important Cases, supra note 2, at 929-33, 943 (noting the rise of patent cases on the Roberts Court’s important-questions docket).
[141]. See Federal Courts Improvement Act of 1982, Pub. L. No. 97-164, 96 Stat. 25.; Rep. No. 97-312, at 23 (1981); S. Rep. No. 97-275, at 5–6 (1981).
[142]. Notably, the Rehnquist Court’s conflicts docket does not exclude patent cases entirely, in part because of disagreement among the federal appellate courts over the Federal Circuit’s jurisdiction. See generally Christianson v. Colt Indus. Operating Corp., 486 U.S. 800 (1988) (addressing a split of the Federal Circuit’s jurisdiction).
[143]. See supra notes 116–119.
[144]. See, e.g., Nestlé USA v. Doe, 593 U.S. 628, 657 (2021) (Alito, J., dissenting) (“The primary question presented in the two certiorari petitions filed in these cases is whether domestic corporations are immune from liability under the Alien Tort Statute . . . . I would decide that question.”); Hernandez v. Mesa, 582 U.S. 548, 555 (2017) (Thomas, J., dissenting) (“When we granted certiorari in this case, we directed the parties to address, in addition to the questions presented by petitioners, ‘[w]hether the claim in this case may be asserted under [Bivens].’ I would answer that question, rather than remand for the Court of Appeals to do so.”).
[145]. See e.g., Wayne A. Logan, Constitutional Cacophony: Federal Circuit Splits and the Fourth Amendment, 65 Vand. L. Rev. 1137 (2012); see also Orin Kerr (@OrinKerr), x (Sept. 13, 2022), https://twitter.com/OrinKerr/status/1569758759379148802 [https://perma.cc/UXV4-QCDH] (“There are so many circuit splits on how the Fourth Amendment applies, and yet there were no Fourth Amendment cases on the Supreme Court docket last year and there are no such cases on the Court’s docket for next year.”).
[146]. See, e.g., Byrd v. United States, 584 U.S. 395, 410 (2018) (noting that certiorari-stage briefing raised two issues: first, whether the petitioner “had a Fourth Amendment interest in [a] rental car” and, second, even if so, whether the law enforcement officers’ search of such a vehicle was justified by “probable cause to believe it contained evidence of a crime when they initiated their search”).
The plot of the term fourth_amendment, on file with the author, is similar to (but slightly less scrutable than) the probable_cause plot. The one exception is that, during the Roberts Court, the frequency score of the term ticked up and the proximity score of the term ticked down—but only slightly, especially as compared to the shifts for probable_cause.
[147]. See Act of Mar. 3, 1891, ch. 517, § 6, 26 Stat. 828.
[148]. See supra note 122 and accompanying text.
[149]. The Federalist No. 82 (Alexander Hamilton) (Clinton Rossiter ed., 1961).
[150]. See, e.g., Procedure in Federal Courts, Hearing Before a Subcommittee of the Committee on the Judiciary, 68th Cong. 1 (1924) (statement of McReynolds, J.) (“About 1904, . . . Congress provided for taxing adulterated butter. That statute was construed in the Sixth Circuit. Several years afterwards the Eighth Circuit . . . arrived at a different conclusion. It came to us upon certiorari . . . . Congress is passing many statutes, which must be construed and applied. The number of federal questions is constantly increasing. There are now so many . . . .”); Hamm, supra note 127(reporting on a talk where Justice Breyer explained that the Court’s docket grew in the 1960s in response to a contemporaneous “flurry of legislation”).
[151]. It does not matter for my purposes whether Congress leaves (or perhaps delegates) these interpretative questions to the courts intentionally. Sometimes Congress will intentionally leave some provisions vague in order to assure passage of an important bill. See, e.g., Victoria F. Nourse & Jane S. Schacter, The Politics of Legislative Drafting: A Congressional Case Study, 77 N.Y.U. L. Rev. 575, 596 (2002). But some ambiguities in statutory language may be surprising, unintentional, or result from changes over time. E.g., Levy & Narechania, supra note 66.
[152]. See, e.g., Burnet v. Harmel, 287 U.S. 103, 105, 110 (1932) (“This Court granted certiorari . . . to resolve a conflict . . . under corresponding provisions of the Revenue Act of 1921.”); see also, e.g., McFeely v. Comm’r, 296 U.S. 102, 105 (1935) (“These cases were brought here on writs of certiorari to resolve a conflict between Circuits with respect to the application of section 101 of the Revenue Act of 1928.”); White v. United States, 305 U.S. 281, 283 (1938) (“We granted certiorari . . . to resolve a conflict[,] . . . which arose under related sections of the 1932 Revenue Act.”); Helvering v. Leonard, 310 U.S. 80, 81 (1940) (“This case involves the question of the taxability to the grantor under the Revenue Act of 1928 . . . . We granted certiorari . . . because of [a] probable conflict . . . .”); Hartley v. Comm’r, 295 U.S. 216, 217 (1935) (“We granted certiorari to resolve a conflict . . . under section 202 of the 1918 Revenue Act.”); McDonnell v. United States, 288 U.S. 420, 421 (1933) (“Because of conflict . . . certiorari was granted, limited to the question of the validity of the waiver under section 278(e) of the Revenue Act of 1924.”).
[153]. See, e.g., 10 E. 40th St. Bldg. v. Callus, 325 U.S. 578, 579–80 (1945) (“The Fair Labor Standards Act of 1938 regulates wages and hours not only of employees who are ‘engaged in commerce’ but also those engaged ‘in the production of goods for commerce.’ . . . When these provisions first came here we made it abundantly clear that their enforcement would involve the courts in the empiric process of drawing lines from case to case . . . . The Act has produced a considerable volume of litigation and has inevitably given rise to judicial conflicts and divisions.”).
[154]. Indep. Coal & Coke Co. v. United States, 274 U.S. 640, 650 (1927). Interestingly, Independent Coal & Coke Co. regards a land patent rather than an invention patent.
[155]. U.S. Printing & Lithograph Co. v. Griggs, Cooper & Co., 279 U.S. 156, 158 (1929). Trademarks are, and were, registered with the “Patent Office,” as the case describes. See id. at 157.
[156]. See supra note 152.
[157]. See Southland Gasoline Co. v. Bayley, 319 U.S. 44, 45 (1943) (“By writs of certiorari these two cases were brought here to resolve the conflict between them over the proper interpretation of section 13(b)(1) of the Fair Labor Standards Act of 1938.”); Armour & Co. v. Wantock, 323 U.S. 126, 127 (1944) (“Armour and Company, petitioners, have been held liable to certain employees for overtime, liquidated damages, and attorneys’ fees under the Fair Labor Standards Act . . . . The Court of Appeals for the Fifth Circuit on facts of considerable similarity reached an opposite result . . . . To resolve the conflict we granted certiorari in both cases.”); Western Union Tel. v. Lenroot, 323 U.S. 490, 491 (1945) (granting certiorari to address a question arising under the Fair Labor Standards Act of 1938); A. H. Phillips, Inc. v. Walling, 324 U.S. 490, 492 (1945) (granting certiorari to address a conflict regarding the “overtime and record provisions of the Act”); 10 E. 40th St. Bldg. v. Callus, 325 U.S. 578, 579–80 (1945) (granting certiorari to address a question arising under the Fair Labor Standards Act of 1938); Borden Co. v. Borella, 325 U.S. 679, 680 (1945) (“Once again . . . we are required to consider the application of the Fair Labor Standards Act of 1938 . . . .”); Roland Elec. Co. v. Walling, 326 U.S. 657, 660 (1946) (granting certiorari to address a question arising under the Fair Labor Standards Act of 1938); Martino v. Michigan Window Cleaning Co., 327 U.S. 173, 174 (1946) (granting certiorari to address the “same [questions] as those in Roland” under the Fair Labor Standards Act of 1938); Okla. Press Pub. Co. v. Walling, 327 U.S. 186, 191–92 (1946) (“Because of the importance of the issues for administration of the Act and also on account of the differences in the grounds for the two decisions, as well as between them and decisions from other circuits, certiorari was granted in both cases.”); D.A. Schulte v. Gangi, 328 U.S. 108, 110 (1946) (“The primary issue presented by the petition for certiorari is whether the Fair Labor Standards Act precludes a bona fide settlement of a bona fide dispute over the coverage of the Act on a claim for overtime compensation and liquidated damages where the employees receive the overtime compensation in full.”).
[158]. See Congress of Indus. Orgs. v. McAdory, 325 U.S. 472, 474 (1945) (granting certiorari to review a question about the National Labor Relations Act and the Railway Labor Act, which was amended in 1934 and 1936).
[159]. Notably, this result contrasts with findings in other studies of certiorari at the Supreme Court. See Livermore et al., supra note 21, at 874, 876. This study compares the Supreme Court’s docket to the entire body of federal appellate decisions. And so it is possible that the mismatch they observe—a relatively lower number of sentencing-related cases at the Supreme Court––is a consequence of the Sentencing Guidelines causing an increase in the number of sentencing-related appeals in the appellate courts. See, e.g., Sara Sun Beale, Too Many and Yet Too Few: New Principles to Define the Proper Limits for Federal Criminal Jurisdiction, 46 Hastings L.J. 979, 986-87 (1995); Cynthia J. Thomas, Report on the Federal Sentencing Guidelines and Their Impact, 28 Rev. Juridica U. Interamericana P.R. 283, 293–94 (1994); see also Narechania, Important Cases, supra note 2, at 964 & n.147 (making a similar observation). Stated otherwise, the number of sentencing-related cases reviewed by the Court reflects only a small proportion of the total number of such federal appellate cases, even as these cases occupied a larger proportion of the Court’s conflicts docket (as compared both to other types of cases and to the previous Burger Court).
[160]. See, e.g., Mistretta v. United States, 488 U.S. 361, 362, 371 (1989) (noting “disarray among the Federal District Courts” and explaining that the Court “granted certiorari before judgment in the United States Court of Appeals for the Eighth Circuit in order to consider the constitutionality of the Sentencing Guidelines promulgated by the United States Sentencing Commission”); Williams v. United States, 503 U.S. 193, 198 (1992) (“We granted certiorari to resolve a conflict among the Circuits on whether a reviewing court may affirm a sentence in which a district court’s departure from the guideline range is based on both valid and invalid factors.”); Chapman v. United States, 500 U.S. 453, 456 (1991) (granting certiorari to address an interpretative question arising under a sentencing statute and concomitant guideline); Stinson v. United States, 508 U.S. 36, 40 (1993) (“The various Courts of Appeals have taken conflicting positions on the authoritative weight to be accorded to the commentary to the Sentencing Guidelines, so we granted certiorari.”); see also Braxton v. United States, 500 U.S. 344, 347–48 (1991) (addressing the relationship of the Court’s certiorari jurisdiction to the Sentencing Commission’s powers to review and revise the Sentencing Guidelines).
[161]. See, e.g., Taylor v. United States, 495 U.S. 575, 579-80 (1990) (noting that the Eighth Circuit affirmed a decision “using Taylor’s Missouri convictions for second-degree burglary to enhance his sentence” and explained that it “granted certiorari . . . to resolve a conflict among the Courts of Appeals concerning the definition of burglary for purposes of § 924(e)[,]” the Armed Career Criminal Act).
[162]. A statute’s recency is relative, and the meaning of recent seems to have narrowed over time as judicial review has accelerated. The Taft Court’s attention to patent matters, and to the activities of the Commissioner of Patents, seems, at least in part, to arise out of concerns regarding the construction, interpretation, and application of the “Trade-Mark Act of February 20, 1905.” See U.S. Printing & Lithograph Co. v. Griggs, Cooper & Co., 279 U.S. 156, 158 (1929). The Taft Court, that is, took fifteen years (or more) to address questions arising out of intellectual property statutes. The Hughes Court’s tax* cases seem to come anywhere from around seven to eleven years after the relevant underlying Revenue Act. See Burnet v. Harmel, 287 U.S. 103, 105 (1932) (“This Court granted certiorari . . . to resolve a conflict . . . under corresponding provisions of the Revenue Act of 1921.”); McDonnell v. United States, 288 U.S. 420, 421 (1933) (“Because of conflict . . . certiorari was granted, limited to the question of the validity of the waiver under section 278(e) of the Revenue Act of 1924.”); McFeely v. Comm’r, 296 U.S. 102, 105 (1935) (“These cases were brought here on writs of certiorari to resolve a conflict between Circuits with respect to the application of section 101 of the Revenue Act of 1928.”). But the Rehnquist Court addressed the “disarray among the Federal District Courts” over “the constitutionality of the Sentencing Guidelines” within four years. Mistretta v. United States, 488 U.S. 361, 362, 371 (1989). And while some of that speed might be attributed to the Court’s decision to grant a writ of certiorari before judgment in the Court of Appeals, it is consistent with other cases of the time. See id. at 362. For example, the Court granted certiorari to address a conflict stemming from the Armed Career Criminal Act within six years of its enactment. See Taylor, 495 U.S. at 581. Analogously, Deborah Beim and Kelly Rader find that the Court generally acts within three years to resolve a split—though, importantly, their findings are about the duration of the split, rather than the time difference between Congress’s enactment and the Court’s decision. See Beim & Rader, supra note 6 at 448. Nevertheless, overall, the Court seems to move more quickly now to address splits arising out of statutory language than before.
[163]. See, e.g., Bd. of Trs. of Univ. of Alabama v. Garrett, 531 U.S. 356, 360, 363 (2001) (holding that the Eleventh Amendment bars employees from recovering damages from a state’s failure to comply with Title I of the Americans with Disabilities Act); Pennsylvania v. Union Gas Co., 491 U.S. 1, 2 (1989) (finding that the Comprehensive Environmental Response, Compensation, and Liability Act of 1980 expresses an intent to hold States liable in damages in federal court); Kaiser Aluminum & Chem. Corp. v. Bonjorno, 494 U.S. 827, 834, 836 (1990) (Federal Courts Improvement Act of 1982); see also City of Ranchos Palos Verdes v. Abrams, 544 U.S. 113, 119, 127 (2005) (on the interaction between Section 1983 and the Telecommunications Act of 1996).
[164]. E.g., Nat’l Mut. Ins. Co. of D.C. v. Tidewater Transfer Co., 337 U.S. 582 (1949); see also Aetna Casualty & Co. v. Flowers, 330 U.S. 464, 466 (1947) (“We granted certiorari because of an apparent conflict . . . as to the jurisdictional minimum requirement.”). See also supra note 142 (citing Christianson v. Colt, 486 U.S. 800 (1988) and explaining how the Rehnquist Court addressed a split regarding the Federal Circuit’s jurisdiction).
[165]. Richard L. Pacelle, Jr., The Transformation of the Supreme Court’s Agenda: From the New Deal to the Reagan Administration 29 (1991) (“Once an issue has been on the volitional agenda, its life may be extended long after the Court’s interest in the issue has waned.”).
[166]. See supra note 130.
[167]. But see id.
[168]. See supra Table 2.
[169]. See id.
[170]. See James J. Brudney & Ethan J. Leib, Statutory Interpretation as “Interbranch Dialogue”?, 66 UCLA L. Rev. 346, 351 (2019) (“[I]nterbranch dialogue [over statutory interpretation] . . . is best understood as responsive communication between the two institutions, acting through their representatives, in which each institution listens, learns, takes seriously, and enhances its understanding of statutory issues over time.”).
[171]. Jason Iuliano & Ya Sheng Lin, Supreme Court Repeaters, 69 Vand. L. Rev. 1349, 1350 (2016).
[172]. Id. at 1370.
[173]. 325 U.S. 578, 579 (1945).
[174]. 325 U.S. 679, 680 (1945).
[175]. William Eskridge, Overriding Supreme Court Statutory Interpretation Decisions, 101 Yale L.J. 331, 332 (1991); Matthew Christiansen & William Eskridge, Congressional Overrides of Supreme Court Statutory Interpretation Decisions, 1967–2011, 92 Tex. L. Rev. 1317, 1317-18 (2014).
[176]. Christiansen & Eskridge, supra note 175, at 1319; see also Richard L. Hasen, End of the Dialogue? Political Polarization, the Supreme Court, and Congress, 86 S. Cal. L. Rev. 205, 209 (2013).
[177]. See Brudney & Leib, supra note 170, at 366.
[178]. Lilly Ledbetter Fair Pay Act of 2009, Pub. L. No. 111-2, 123 Stat. 5 (overriding Ledbetter v. Goodyear Tire & Rubber Co., 550 U.S. 618 (2007)); see also Ledbetter, 550 U.S. at 661 (Ginsburg, J., dissenting) (“[T]he Legislature may act to correct this Court’s parsimonious reading of Title VII.”).
[179]. Deborah A. Widiss, How Courts Do – and Don’t – Respond to Statutory Overrides, 104 Judicature 51, 55 (2020) (describing “situations in which judges could reasonably disagree about whether the override statute or the pre-existing precedent should apply”).
[180]. Compare Green v. Brennan, 578 U.S. 547, 560–61 (2016) (“Amica and the dissent also argue that their interpretation is more consistent with this Court’s prior precedent on when the limitations period begins to run for discrimination claims”) (internal citations omitted) with id., at 569 & n.3 (Alito, J., concurring) (“Congress has since abrogated Ledbetter’s precise holding . . . but it did not disturb the reasoning of the precedents on which Ledbetter was based.”).
[181]. Nancy C. Staudt, Rene Lindstadt & Jason O’Connor, Judicial Decisions as Legislation: Congressional Oversight of Supreme Court Tax Cases, 1954–2005, 82 N.Y.U. L. Rev. 1340, 1349 (2007).
[182]. 508 U.S. 491, 499 (1993).
[183]. See PPL Corp. v. Comm’r, 569 U.S. 329, 334 (2013) (granting review to resolve a conflict on the meaning and interpretation of a statute and concomitant regulation that “codifies longstanding doctrine dating back to Biddle v. Commissioner, 302 U.S. 573, 578–79 (1938)”). Both Fabe and PPL Corp. are drawn from my results. For two examples from outside of my result set, see Helvering v. Griffiths, 318 U.S. 371 (1943) (granting review to address provisions in various Revenue Acts that were enacted in response to the Court’s decision in Eisner v. Macomber, 252 U.S. 189 (1918)); South Dakota v. Wayfair, Inc., 585 U.S. 162, 168 (2018) (reviewing a South Dakota state tax enacted in response to the Court’s decision in Quill Corp. v. North Dakota, 504 U.S. 298 (1992)).
[184]. Narechania, Important Cases, supra note 2, at 975.
[185]. Id. at 979.
[186]. As I have explained before, some of the cases represented in Figure 9 during and after the Burger Court may be displaced from the Court’s mandatory docket. But the Court’s certiorari docket did not absorb all of the constitutional cases that were once on the mandatory docket, as the overall size of the Court’s docket decreased by almost one-fifth between the years immediately preceding this jurisdictional change to the years immediately after it. See Narechania, Important Cases, supra note 2, at 974-75.
[187]. Stern et al., supra note 41, at 66.
[188]. See Narechania, Important Cases, supra note 2, at 962.
[189]. See, e.g., infra note 261; see also Narechania, Roberts Court, supra note 9, at 602–03 (using dictionary differences to describe differences across sets in a similar manner).
[190]. Pacelle, supra note 16, at 28; Ryan C. Black & Ryan J. Owens, Agenda Setting in the Supreme Court: The Collision of Policy and Jurisprudence, 71 J. Pol. 1062, 1073 (2009) (“Justices have nearly total discretion to decide which cases the Court will hear, meaning they have freedom to pursue their raw policy goals . . . .”).
[191]. See Pacelle, supra note 16 at 28; see also Samuel Estreicher & John E. Sexton, A Managerial Theory of the Supreme Court’s Responsibilities: An Empirical Study, 59 N.Y.U. L. Rev. 681, 722 (1984) (“We regard the resolution of conflicts between two or more courts—whether state or federal and whether the issue is one of federal constitutional, statutory, or common law—as one of the Court’s principal tasks.”).
[192]. Pacelle, supra note 16, at 28.
[193]. Narechania, Important Cases, supra note 2, at 984 (emphasis in original).
[194]. Id. at 980.
[195]. See Raquel E. Aldana & Thomas O’Donnell, A Look Back at the Warren Court’s Due Process Revolution Through the Lens of Immigrants, 51 Univ. Pac. L. Rev. 633, 634 (2020) (“The Warren Court’s criminal due process revolution has generated significant reflections on its impact, both positive and negative, on the criminal justice system.”).
[196]. See, e.g., Opper v. United States, 348 U.S. 84, 86 (1954) (“Certiorari was granted because of asserted variance or conflict between the legal conclusion reached in this case—that an extrajudicial, exculpatory statement of an accused, subsequent to the alleged crime, needs no corroboration—and other cases to the contrary.”); Lott v. United States, 367 U.S. 421, 424 (1961) (“Because of a conflict between the circuits upon the question presented [regarding the time to appeal in criminal cases] and of its importance to the proper administration of the criminal Rules, we granted certiorari.”); Di Bella v. United States, 369 U.S. 121, 127 (1962) (“The precise question before us has been much canvassed in the lower courts. It has not only produced a conflict among the circuits, but has provoked practical difficulties in the administration of criminal justice and caused expressions of dissatisfaction even in courts that have sustained an appeal.”).
[197]. See, e.g., Shaughnessy v. Pedreiro, 349 U.S. 48, 49–50 (1955).
[198]. David Schultz, The Supreme Court Has Overturned Precedent Dozens of Times in the Past 60 Years, Including when It Struck Down Legal Segregation, Conversation (Sept. 20, 2021), https://theconversation.com/thesupreme-court-has-overturned-precedent-dozens-of-times-in-the-past-60-years-includingwhen-it-struck-down-legal-segregation-168052 [https://perma.cc/96QE-BPPD]; see also Narechania, Important Cases, supra note 2, at 981 (noting that the Court’s current proclivity to grant review on the important-questions docket for the purpose of considering whether to overrule precedent began with the Rehnquist Court and has accelerated during the Roberts Court).
[199]. See Exxon Corp. v. Cent. Gulf Lines, 500 U.S. 603, 607 (1991); see also Barrett, supra note 44, at 1731 (suggesting that conflicts among lower courts may help to “put[] a challenge to precedent on the Court’s agenda”).
[200]. As noted above, Stern et al. authored an “epitaph” for the Court’s mandatory jurisdiction, reflecting on changes made in 1988 to the Court’s jurisdiction. See generally Stern et al., supra note 41. The Court’s discretionary jurisdiction also grew in 1976. See Stern et al., supra note 41, at 66. Table 4 does not change substantially if we use 1976, instead of 1988, as the cutoff between columns, so I use 1988 for consistency with prior work. See Narechania, Important Cases, supra note 2, at 98 tbl.5.
[201]. See Narechania, Patent Puzzle, supra note 46, at 1362 & nn.81–84 and accompanying text.
[202]. See Amanda Tyler, Setting the Supreme Court’s Agenda: Is There a Place for Certification?, 78 Geo. Wash. L. Rev. 1310, 1316 (2010).
[203]. See Orin Kerr (@OrinKerr), x (Sept. 13, 2022), https://twitter.com/OrinKerr/status/1569758759379148802 [https://perma.cc/UXV4-QCDH] (“There are so many circuit splits on how the Fourth Amendment applies, and yet there were no Fourth Amendment cases on the Supreme Court docket last year and there are no such cases on the Court’s docket for next year.”).
[204]. See, e.g., Casey & Macey, supra note 68, at 497 (“The Supreme Court has granted certiorari to resolve circuit splits in a number of bankruptcy cases in the past few years . . . . On the other hand, several substantive splits have remained unresolved for years and, in some cases, decades.”).
[205]. As elaborated throughout the rest of this subsection, my focus here is on certification. This is not, of course, the only solution. When, for example, Congress faced growing concern regarding splits in patent cases, see Menell & Vacca, supra note 137, at 794 (describing the Hruska Commission Report, among other sources of concern), giving rise to increasing forum shopping, it reformed the structure of the federal courts of appeals to consolidate nearly all patent appeals in a single court. This change dramatically reduced the likelihood of a conflict. We might imagine similar reforms aimed at languishing splits in other areas of law: Congress, might, for example, create a specialized appeals court for bankruptcy cases. And while such an idea is potentially intriguing, I set this proposal aside for two main reasons. First, my focus is, as noted above, primarily on the splits that the Court has agreed to review—largely to the exclusion, by necessity, of the splits that the Court has ignored. Because my method focuses on granted cases, it is somewhat difficult for me to use these results to suggest specific areas—bankruptcy, securities, etc.—that might benefit the most from such a new judicial institution. And more importantly, while reviews of the Federal Circuit are mixed, see, e.g., Rochelle Cooper Dreyfuss, The Federal Circuit: A Continuing Experiment in Specialization, 54 Case W. Rsrv. L. Rev. 769 (2004); R. Polk Wagner & Lee Petherbridge, Is the Federal Circuit Succeeding? An Empirical Assessment of Judicial Performance, 152 U. Pa. L. Rev. 1105 (2004); Paul Gugliuzza, The Federal Circuit: A Failed Experiment in Specialization?, FedCircuitBlog (Feb. 28, 2022), https://fedcircuitblog.com/2022/02/28/online-symposium-the-federal-circuit-a-failed-experiment-in-specialization [https://perma.cc/96YB-3AG2], the pathologies associated with creating tribunals of such narrow jurisdiction are sufficiently serious that we should proceed cautiously before creating yet another Federal Circuit. For one early warning, see Simon Rifkind, A Special Court for Patent Litigation? The Danger of a Specialized Judiciary, 37 A.B.A. J. 425, 425 (1951).
Another possibility is to establish a national court of appeals. See, e.g., Commission on Revision of the Federal Court Appellate System, Inter-Circuit Conflicts, in A National Court of Appeals in Structure and Internal Procedures: Recommendations for Change 16 (1975). Setting aside the controversy that would surely attend to such a proposal, see, e.g., William J. Brennan, Jr., The National Court of Appeals: Another Dissent, 40 U. Chi. L. Rev. 473 (1973); Eugene Gressman, The National Court of Appeals: A Dissent, 59 A.B.A. J. 253 (1973), a national court of appeals would still need some agenda-setting mechanism (whether a mandatory right of appeal in all cases, absolute discretion, or a combination of mandatory and discretionary jurisdiction). And so the certification process I outline here is one solution that, for reasons described in this subsection, is best applied to the Supreme Court itself.
[206]. Tyler, supra note 202, at 1321 n.62 (citing 17 Charles Alan Wright Et Al., Federal Practice & Procedure § 4038, 62–64 & nn.2, 10 (3d ed. 2006)); see also Wheeler Lumber Bridge & Supply Co. of Des Moines, Iowa, v. United States, 281 U.S. 572, 577–78 (1930) (Van Devanter, J.) (describing certification jurisdiction as mandatory); Cohens v. Virginia, 19 U.S. 264, 404 (1821) (“It is most true that this Court will not take jurisdiction if it should not: but it is equally true, that it must take jurisdiction if it should . . . . We have no more right to decline the exercise of jurisdiction which is given, than to usurp that which is not given.”); cf. Old Colony Tr. Co. v. Comm’r, 279 U.S. 716, 728–29 (1929) (Taft, C.J.) (describing certification as “an invocation of the [Court’s] appellate jurisdiction”).
[207]. But I am happy to indulge those readers that are interested. Connecticut Fine Wine & Spirits v. Seagull is about “post-and-hold” statutes, in which alcohol wholesalers are required to share their prices with competitors on a monthly basis, are given a short period to adjust their posted prices (with the proviso that they cannot undercut the lowest posted price), and then must adhere to their adjusted prices for one month. See 936 F.3d 119, 123 (2nd Cir. 2019).
[208]. Id.
[209]. See, e.g., United States v. Seale, 577 F.3d 566, 567, 570 (5th Cir. 2009) (certifying question to the Supreme Court by a majority vote of en banc court, after the en banc court affirmed the district court’s decision by an equally divided vote).
[210]. See In re Bowe, 144 S. Ct. 1170, 1170-71 (2024) (Sotomayor, J., statement respecting the denial of the petition for a writ of habeas corpus) (expressing a “desire for this Court to resolve this split,” noting “considerable structural barriers to this Court’s ordinary review via certiorari petition,” and so echoing the Government’s suggestion that “a court of appeals seeking clarity could certify the question to this Court”).
[211]. Cf. Levy & Narechania, supra note 66, at 918 (suggesting that the federal courts of appeals use the “statutory opinion transmission project” to send opinions implicating splits to Congress).
[212]. See Tyler, supra note 202.
[213]. U.S. Const. art. III, § 2, cl. 2 (“[T]he supreme Court shall have appellate Jurisdiction, both as to Law and Fact, with such Exceptions, and under such Regulations as the Congress shall make.”).
[214]. Compare Narechania, Important Cases, supra note 2, at 983, tbl.5, with infra Table 4.
[215]. Pacelle, supra note 16, at 28.
[216]. The Federalist No. 82 (Alexander Hamilton) (Clinton Rossiter ed., 1961).
[217]. In addition to the reasons described here, it is possible that growing the Court’s mandatory doctrine would yield “other salutary benefits, including that the Court might be more constrained in its ability to craft special rules for cases with particular facts, while trying to duck the consequences of applying those rules to other circumstances.” Narechania, Important Cases, supra note 2, at 993 n.259 (citing Bush v. Gore, 531 U.S. 98 (2000) (explaining that the Court’s opinion “is limited to the present circumstances”); Nat’l Inst. of Fam. and Life Advocs. v. Becerra, 585 U.S. 755, 775 (2018) (suggesting that the Court’s opinion is limited to its specific context and not affecting “the legality of health and safety warnings long considered permissible”)).
[218]. See Levy & Narechania, supra note 66, at 935.
[219]. See Narechania, Patent Puzzle, supra note 46, at 1362 & nn.81–84 (describing these values and collecting sources).
[220]. See, e.g., Richard L. Hasen, Polarization and the Judiciary, 22 Ann. Rev. Pol. Sci. 261, 267 (2019) (“[N]o one doubts that the Supreme Court is growing more polarized in its decision making.”); Stephen I. Vladeck, The Solicitor General and the Shadow Docket, 133 Harv. L. Rev. 123, 157 (2019) (arguing that effecting doctrinal changes through the shadow docket “risks the perception” that the Court is ruling in a certain way “depending upon the identity . . . of the sitting President, or perhaps . . . the political or ideological valence of the particular . . . policy at issue”); Greg Stohr, Supreme Court Approval Drops as New Term Brings Divisive Issues, Bloomberg (Sept. 22, 2021), https://www.bloomberg.com/news/articles/2021-09-22/supreme-court-approval-drops-as-new-term-brings-divisive-issues [https://perma.cc/WUV5-SSAZ] (citing “a new poll . . . showing declining confidence in the [C]ourt and growing concern that it is driven by politics”).
[221]. See Narechania, Important Cases, supra note 2, at 976-84, subsection “Discretion and Personnel.”
[222]. But see J. Harvie Wilkinson III, If It Ain’t Broke, 119 Yale L.J. F. 67, 69 (2010) (citing sources contending “that the Court should hear more cases because the discipline and time required to do so would discourage . . . judicial activism,” but ultimately disagreeing with that conclusion).
[223]. See, e.g., Dan Epps & Will Baude, Small Victories, Divided Argument Podcast (June 19, 2024) (statement of Will Baude) [https://perma.cc/HHC8-LR3G] (contrasting the political and ideological nature of “abortion case[s]” with an apparently more apolitical “bankruptcy uniformity case”).
[224]. See Trump v. Anderson, 601 U.S. 100, 117-18 (2024) (Barrett, J., concurring in part and concurring in the judgment).
[225]. See Appendix Figure 1. Indeed, as that figure suggests, the gap between the important-questions docket and the conflicts docket (in terms of average votes in the majority) is now wider than ever before.
[226]. Likewise, the Court’s decisions can incite a dialogue with and among the state and federal appeals courts, which are charged with interpreting the reach of the Court’s decision.
[227]. See, e.g., United States v. Microsoft, 584 U.S. 236, 239-40 (2018) (explaining that the case became moot when it was superseded by Congress’s subsequent enactment of the Clarifying Lawful Overseas Use of Data (CLOUD) Act); see also Levy & Narechania, supra note 66, at 929-35 (detailing cases in which circuit court judges in their opinions alerted Congress to statutory interpretation issues that may need amending).
[228]. Narechania, Important Cases, supra note 2, at 992.
[229]. U.S. Const. art. III, § 2, cl. 2.
[230]. Cf. Aaron L. Nielson & Paul Stancil, Gaming Certiorari, 170 U. Pa. L. Rev. 1129, 1130 (2022) (describing how “lower court judges can reverse engineer their decisions to make certiorari more or less attractive for the Justices”).
[231]. See id.
[232]. Narechania, Important Cases, supra note 2, at 985. This subsection draws from this previous work.
[233]. U.S. Const. art. I, § 8, cl. 1, 4.
[234]. Act of Mar. 3, 1891, 26 Stat. 828.
[235]. See William Howard Taft, The Jurisdiction of the Supreme Court Under the Act of February 13, 1925, 35 Yale L.J. 1, 1–2 (1925).
[236]. See id. at 6.
[237]. Narechania, Important Cases, supra note 2, at 989.
[238]. Watts, supra note 119, at 55–61; Baude, supra note 119, at 16–18; Stephen I. Vladeck, The Supreme Court Needs to Show Its Work, The Atlantic (March 2021).
[239]. Frankfurter & Hart, Jr., supra note 95, at 82–83.
[240]. See, e.g., Concrete Appliances Co. v. Gomery, 269 U.S. 177, 178 (1925) (citing Thomson Spot Welder Co. v. Ford Motor Co., 265 U. S. 445 (1924), to support the proposition that such a conflict merits review); Narechania, Important Cases, supra note 2, at 932 n.36 (citing United States v. Kebodeaux, 570 U.S. 387, 391 (2013) as one “exceedingly rare” example of the modern Court explaining, and perhaps developing, its certiorari standard in the mode of common law reasoning). Both Gomery and Kebodeaux suggest that developing such a common law of certiorari is neither implausible nor impracticable. See also Frankfurter & Hart Jr., supra note 95, at 82 (explaining that this “needed elucidation” can come through the “less time-consuming means” of more thoroughly explaining the reasons for review in merits opinions).
[241]. Narechania, Important Cases, supra note 2, at 989–90.
[242]. Concrete Appliances Co., 269 U.S. at 178 (citing Thomson Co. v. Ford Motor Co., 265 U. S. 445 (1924)).
[243]. Teva v. Sandoz, 574 U.S. 318, 324 (2015) (“The Federal Circuit reviews the claim construction decisions of federal district courts throughout the Nation, and we consequently believe it important to clarify the standard of review that it must apply when doing so.”); cf. Amy Coney Barrett, The Supervisory Power of the Supreme Court, 106 Colum. L. Rev. 324 (2006).
[244]. Narechania, Important Cases, supra note 2, at 972 & n.165.
[245]. As Adam Unikowsky has elaborated:
I view the practice of judges providing reasoned explanations for decisions as a fundamental component of due process. It’s right up there with notice, a hearing, and an impartial adjudicator. First, providing reasoned explanations serves a disciplining function that improves the quality of judicial decisionmaking. Second, it engenders public confidence in the legal system—people may disagree with Supreme Court decisions, but at least they will know that the decisions are based on reason rather than caprice. Third, it ensures that litigants are treated with dignity by allowing them to understand why things are happening to them.
Adam Unikowsky, Trust the Border Patrol, Adam’s Legal Newsletter (Feb. 6, 2024), https://adamunikowsky.substack.com/p/trust-the-border-patrol [https://perma.cc/X8D2-AY8J].
[246]. Narechania, Important Cases, supra note 2, at 987 (citing Kent Greenawalt, Statutory and Common Law Interpretation 198–99 (2012)).
[247]. Id. at 991.
[248]. Id.
[249]. See, e.g., Deepak Gupta, Testimony before the Presidential Commission on the Supreme Court of the United States: Access to Justice and Transparency in the Operation of the Supreme Court 12 (June 30, 2021).
[250]. Compare Nestlé USA v. Doe, 593 U.S. 628, 640 (2021) (Gorsuch, J., concurring) (noting that although the Court granted review to address a specific question, “rather than resolve that question . . . the Court rests its decision on other grounds,” and concluding that “[t]hat is a good thing”), with Fulton v. City of Philadelphia, 593 U.S. 522, 618 (2021) (Gorsuch, J., concurring) (noting that the Court granted review to address a specific question but lamenting the Court’s decision “to sidestep th[at] question” because the Court’s approach would cause some “problems [to] emerge”).
[251]. See Spaeth et al., supra note 135.
[252]. Id.
[253]. Id.
[254]. Among other things, the text was edited to remove star page notation, to correct terms that were hyphenated across lines, to concatenate block quotes into single paragraphs, and to convert text to lowercase for consistency.
[255]. Paragraphs from files in HTML format were extracted using the Beautiful Soup package. Paragraphs from files in plain text were extracted using Python’s native string manipulation functions. Note that this extraction occurs before generating bigrams and trigrams, and so searches for certiorari here encompass all certiorari* terms.
[256]. See Laura K. Nelson, Computational Grounded Theory: A Methodological Framework, 49 Socio. Methods & Rsch. 3, 38 n.17 (2020) (“If a dictionary that is relevant to the patterns identified does not already exist, the researcher can create their own dictionary by identifying lists of words relevant to their categories. These manually created dictionaries, of course, should always be validated.”); see also David E. Pozen, Eric L. Talley & Julian Nyarko, A Computational Analysis of Constitutional Polarization, 105 Cornell L. Rev. 1, 22 (2019) (using the term “dictionary” to analogously refer to a “designated lexicon”).
[257]. This list of terms and phrases was generated by processing paragraph text with a tokenizer, to convert sentences into words, and with a phraser (twice), to capture bigrams and trigrams (i.e., common two- and three-word phrases, such as fourteenth_amendment or we_granted_certiorari). The list of terms and phrases excludes common stop words. See Livermore et al., supra note 21, at 867 (excluding stop words in a similar manner).
In previous work, in order to avoid losing terms that were unusually concentrated in one period of time, I also included any additional terms that appeared in more than 3% of cases for any given Court (defined by the tenure of a Chief Justice), presidential administration, or decade. Here, I ran a similar test but found that it added no additional terms. That is, any term that appeared in more than 3% of cases during any such salient era was already included among the list of terms that appeared in more than 1% of all cases in which certiorari* appeared.
[258]. See Narechania, Important Cases, supra note 2, at 1010–11 & n.277.
[259]. Consistent with my analysis in the Article’s main text, several of the terms that were dropped implicated constitutional concerns, such as speech, sixth_amendment, eighth_amendment, cruel, president, executive, political, confession, racial, and constitutional_rights.
[260]. As noted above, the asterisk is used here to loosely represent a wildcard character (as in a regular expression or other search query) and does not represent significance.
[261]. See, e.g., Tomas Mikolov, Kai Chen, Greg Corrado & Jeffrey Dean, Efficient Estimation of Word Representations in Vector Space (2013) (noting that the frequency of words is a useful metric for certain text analysis methods); Turney & Pantel, supra note 104, at 166 (noting the use of frequency and proximity queries in natural language processing).
[262]. Question* encompasses, for example: constitutional_question, question, question_presented, and question_whether. Certiorari* encompasses, for example: certiorari, court_granted_certiorari, petitions_for_certiorari, and we_granted_certiorari. Conflict* encompasses conflict, conflict_with, conflicts, and conflicts_among.
[263]. In some cases, both the Index Term and Focal Term may appear multiple times, giving rise to several possible proximity measures for such cases. In these instances, I decided to use the lowest value (i.e., the closest available proximity). Such an approach helps, for example, to protect constructions such as “granted certiorari to decide this important bankruptcy question” from being affected by other references to bankruptcy proceedings described as procedural history in the study paragraphs.
[264]. One edge case merits explanation: constitution* is an Index Term, question* is a Focal Term, and constitutional_question is one of the original terms in the corpus that is encompassed by both asterisked terms. In this example, constitution* is measured to be 0 terms away from any Focal Term, namely, this question* term.
[265]. For a discussion on alternate approaches, see Narechania, Important Cases, supra note 2, at 1013 n.285.
[266]. For two examples of a case coded 12, but including conflict* suggestive of its place on the Court’s conflicts docket, see supra note 97.
[267]. Indeed, as noted above, certiorari* represents (0, 0), since it appears in each case in which certiorari* appears and because it is most immediately proximate to itself. For a discussion on alternate approaches, see Narechania, Important Cases, supra note 2, at 1014 n.289.
[268]. Figure 1 is an exception, as it displays all terms.
[269]. For a discussion on alternate approaches, see Narechania, Important Cases, supra note 2, at 1015 n.291.
[270]. See 39 Stat. 726 (1916) (providing that the “Supreme Court shall hold . . . one term annually, commencing on the first Monday in October”).
