Why the significant difference? Again, causation is difficult to determine, but there seem to be at least three possible explanations. On the one hand, perhaps when confronted with a statutory case, the Justices are looking to the interpretive canons and tools for guidance in the first instance. When the canons and tools point clearly to a particular interpretation, perhaps the Justices are reaching high degrees of consensus and engaging in very little dueling canon use because the interpretive tools are pointing in a clear direction. Conversely, when the interpretive tools are indeterminate, that could be driving the Justices to divide over the correct statutory construction--and to engage the canons in a dueling manner.
On the other hand, it could be the case that the Justices are starting with a preferred statutory construction--perhaps based on ideology, or considerations of practical consequences or equitable concerns, or some other intuition--and are invoking the interpretive tools to justify that construction. In other words, the Justices could be using the canons as "just window-dressing for results reached for other reasons." (135) If so, the Justices could be dueling more in the divided-vote cases (when they disagree over the correct statutory construction) because they are using the canons strategically, with the distinct purpose of neutralizing an opposing opinion's arguments.
A third possibility is that the Justices may be doing different things in different cases. That is, the close-vote-margin cases may be ones in which the Justices generally are starting with a preferred construction, perhaps based on ideology, and working backward to the interpretive tools--making them more prone to find competing applications for the same tool and to duel in such cases. In the unanimous cases, by contrast, the order of decisionmaking might be reversed; that is, these might be cases in which the Justices do not have strong ideological preferences and might be looking to the interpretive canons in the first instance for guidance.
Bottom line: the canons and interpretive tools do not seem to be constraining the Justices to vote against their ideological preferences. But it also does not seem to be the case that the interpretive canons and tools are being readily manipulated to support competing outcomes in the same case, in the manner suggested by Llewellyn's famous critique. That is, for most individual interpretive canons and tools, reliance by a majority opinion is not resulting in a dissenting opinion countering that same canon or tool most of the time (in roughly 75 percent of the cases). Only the plain meaning rule and Supreme Court precedent are being used in a dueling manner in a large percentage of the cases. And, surprisingly, many interpretive tools that seem highly susceptible to competing invocation--including legislative history, dictionary definitions, substantive canons, and statutory purpose--are generating only low levels of dueling (in roughly 25 percent of the relevant cases). (136) The practical consequences and other statutes tools exhibit slightly higher rates of dueling, at 30.9-34.5 percent, but these tools, like Supreme Court precedent, are traditional tools of legal analysis rather than statutory interpretation-specific maxims; indeed, the other statutes tool has a significant precedent component, as it involves analogies to words in other statutes that have been given a particular construction by courts in prior cases. Also, importantly for the foundational theories of statutory interpretation, no individual statutory interpretation-specific tool of interpretation--other than the plain meaning rule--seems more prone than others to trigger counter-references in the opposite direction. Some implications of these findings are discussed below in Part III.
III. DOCTRINAL FINDINGS
A. Inside Story: Patterns of Dueling Canon Use
Llewellyn observed that a court must "take the music of any statute as written by the legislature; it must take the text of the play as written by the legislature. But there are many ways to play that music, to play that play...." (137) This Section examines the many ways in which majority and dissenting opinions play the same interpretive resource to reach different readings of the same statute. The Section explores the dueling canon cases by individual canon or interpretive tool, uncovering some noteworthy patterns in how the Justices duel. The analysis in this Section is doctrinal rather than empirical, focusing on understanding how the Justices employ the same tools to reach opposing statutory constructions rather than on how frequently they duel over particular tools.
1. Text / Plain Meaning Rule. My analysis identified three primary ways that opposing opinions in the same case tend to duel over application of the plain meaning rule. The majority and dissenting opinions in the case might:
1. Focus on different words or phrases in the text of the same statute.
2. Focus on the text of different statutes, where more than one statute is implicated in the parties' dispute.
3. Focus on the same word or phrase in a statute, but ascribe different meaning to that text.
Further, when dueling in the third manner, one opinion might adopt the prototypical, or core meaning of the term at issue, while the opposing opinion adopts a more expansive reading of the same term.
Let us consider in turn each form of judicial dueling over the plain meaning rule.
a. Focus on different words or phrases in the text of the same statute. When one party alleges that particular conduct gives rise to a claim under a statutory provision, judges may disagree about which precise words or phrases in the statute govern the statute's reach. For example, in Holder v. Humanitarian Law Project,nh the Court construed a provision of the Patriot Act that makes it a federal crime to "knowingly provide material support or resources to a foreign terrorist organization...." (139) Pursuant to the statute, the Secretary of State designated the Kurdistan Workers' Party in Turkey and the Liberation Tigers of Tamil Eelam in Sri Lanka as "foreign terrorist organizations." (140) Some U.S. citizens and domestic organizations who wished to provide support for the lawful, nonterrorist activities of these organizations argued that the statute should be read to require proof that a defendant intended to further a foreign terrorist organization's illegal activities. (141) A majority of the Court disagreed, focusing on the word "knowingly" and arguing that the statute requires only knowledge about the organization's connection to terrorism, not a specific intent to further the organization's terrorist activities. (142) The dissenting opinion, by contrast, focused on the words "material support" and argued that the statute required the government to show that the defendant knew his acts were likely to further the organization's terrorist aims, not just its lawful ones. (143)
b. Focus on the text of two different statutes, where more than one statute is implicated in the parties' dispute. This form of dueling over application of the plain meaning rule does not seem to be very common, but it did occur in a few Roberts Court cases. In AT&T Corp. v. Hulteen, (144) for example, the majority opinion emphasized the text of Title VII, while the dissenting opinion focused on the text of the Pregnancy Discrimination Act. (145)
c. Focus on the same word or phrase in a statute, but ascribe different meaning to that text. This is the most common, and perhaps most interesting, form of judicial dueling over plain meaning. Roughly three-fourths of the cases in which the Justices dueled over the plain meaning rule involved a disagreement over the meaning of the same word or phrase in the statutory text. (146) In almost half of those cases, however, something more than simple disagreement was at work. (147) Specifically, the Justices seemed to divide over whether to adopt the "core" or "prototypical" meaning of the word at issue versus the "legalist" meaning.
The distinction between "prototypical" and "legalist" meaning is not new. Linguist Larry Solan brought the concept of "prototypical meaning" to the attention of statutory interpretation scholars at least a decade ago, defining it as the meaning that focuses on the "core example" that the statute was designed to reach, rather than a meaning that stretches to the conceptual or logical extension of the word at issue. (148) More recent work by Victoria Nourse has highlighted the distinction between "prototypical meaning" and the latter, logical extension kind of meaning, which she calls "legalist meaning." (149) Nourse's work also demonstrates the significance of the distinction, observing that "legalist meaning" is the meaning that New Textualists like Justice Scalia tend to employ, abstracting from the core meaning and considering all logical possibilities that fit within a statutory term. (150) Both Solan and Nourse urge judges to follow "prototypical" rather than "legalist" meaning when construing statutes. (151)
As noted, the Roberts Court's first five-plus terms contained a significant number of cases in which the majority and dissent reached opposing statutory constructions based on the statute's prototypical versus legalist meaning. Consider, for example, the dueling opinions in Ali v. Federal Bureau of Prisons. (152) Ali involved the Federal Tort Claims Act (FTCA), which waives the United States' sovereign immunity for claims based on torts committed by federal employees. (153) The FTCA contains several exceptions, including one for "[a]ny claim arising in respect of the assessment or collection of any tax or customs duty, or the detention of any ... property by any officer of customs or excise or any other law enforcement officer...," (154) Ali, a prisoner who was transferred between prisons, noticed that several items of his personal property were missing when he arrived at his new facility. (155) Ali brought an FTCA claim alleging that Federal Bureau of Prisons (BOP) officials lost his property in transit. The BOP argued that Ali's claim was barred under the above exemption.
A majority of the Roberts Court agreed with the BOP, relying heavily on the plain meaning of the phrase "any other law enforcement officer," which it read to confer immunity on all law enforcement officers. (156) The majority argued that "[t]he phrase 'any other law enforcement officer' suggests a broad meaning" and observed that "[w]e have previously noted that read naturally, the word 'any' has an expansive meaning, that is, 'one or some indiscriminately of whatever kind.'" (157) To support its construction, the majority cited both a dictionary definition and prior cases that gave expansive meaning to phrases such as "any other term of imprisonment" and "any other final action." (158) The majority opinion acknowledged that the statute's use of the terms "tax or customs duty" and "officer[s] of customs or excise" demonstrated Congress's focus on preserving immunity for claims arising from an officer's enforcement of tax and customs laws. But it emphasized that the text also explicitly mentions claims arising from the "detention of property" and insisted that there was no indication that Congress intended immunity to turn on the type of law being enforced. (159) This is a legalist approach to plain meaning--abstracting out from the statute's core application to customs officers, and relying on broadening interpretive resources such as the dictionary, which provides definitions detached from statutory context, to aid in this conceptual expansion. By contrast, Justice Kennedy's dissenting opinion found that "the plain words of the statute indicate that the exception is concerned only with customs and taxes" and protects only law enforcement officers who deal with customs and taxes. (160) Justice Kennedy's dissent emphasized that property damage resulting from customs and tax-related forfeitures was the core situation that the exemption was designed to reach, and faulted the majority for reading the phrase "detention of property" too broadly (161) and for focusing on the meaning of the word "any" in isolation, rather than in light of the statutory context. (162)
What is especially interesting about the core-versus-legalist-meaning divide is that it does not necessarily play out along predictable ideological or jurisprudential lines across cases. For example, at least in the cases decided during the Roberts Court's first five-and-a-half terms, arch-textualist Justice Scalia voted to follow the statute's prototypical meaning nearly as often as he did to follow its legalist meaning. (163) Similarly, purposivist jurist Justice Ginsburg voted nearly as often to adopt a statute's legalist meaning as she did to adopt its prototypical, or core, meaning. (164) Purposivist Justice Breyer and intentionalist Justice Stevens were more predictable, voting to adopt a statute's prototypical meaning in most, though not all, cases. Justice Thomas was similarly predictable, voting to adopt a statute's legalist meaning in most cases, while Justices Roberts, Alito, and Kennedy were more evenly divided in their use of legalist versus prototypical meaning. (165)
The above doctrinal patterns provide important insight into why the plain meaning rule does not seem to constrain jurists in the manner that textualists predict. Justices interpreting statutory text may find different words, phrases, or provisions within the same statute to be controlling or, more cynically, may emphasize different words, phrases, or provisions in order to justify their preferred statutory constructions. Even when the Justices agree on the precise text at issue, some may focus on the core situations that text was designed to cover, while others may seek the general, broad meaning of the word. None of this tells us directly whether the Justices are being disingenuous in their use of text/plain meaning in pursuit of predetermined outcomes, or are simply disagreeing honestly about the rule's application. They could be doing either, or both. Regardless, this doctrinal information teaches that there is ample room for judicial discretion in identifying the so-called "plain" import of statutory terms.
At the same time, however, the doctrinal insight about "prototypical" versus "legalist" meaning suggests a path by which at least some of the indeterminacy involved in application of the plain meaning rule might be eliminated, and judicial dueling over this interpretive canon reduced. Specifically, if jurists were to select either "legalist" or "core" meaning as the appropriate measuring device for a statute's plain meaning, and all Justices sought to identify only that form of plain meaning in all cases, then at least some uncertainties about a statute's meaning could be diminished and judicial dueling over this interpretive canon could be reduced appreciably--perhaps by as much as one-third.
2. Whole Act Rule. As with the plain meaning rule, there are a number of ways in which the members of the Roberts Court tend to duel over application of the whole act rule:
1. In some cases, the majority and dissenting opinions apply different subparts of the whole act rule to the same statutory provision.
2. In others, the majority and dissenting opinions focus on different subparts of the statute at issue.
3. In still other cases, opposing opinions focus on the same subpart of the whole act rule, but reach different end results.
a. Apply different subparts of the whole act rule to the same statutory provision. This form of dueling whole act rule use is perhaps most akin to the "thrusts and parries" in Llewellyn's famous list, as it involves one canonical maxim canceling out another. A majority opinion, for example, might invoke the "meaningful variation" subpart of the whole act rule, which holds that when Congress includes particular language in one section of a statute but omits it in another section of the same statute, courts should presume that Congress acts intentionally and purposefully in the disparate inclusion or exclusion; while the dissenting opinion might invoke the "consistent usage" rule, which presumes that statutory terms mean the same thing throughout all provisions of a statute. Consider the 2009 case, Carcieri v. Salazar. (166) Carcieri involved construction of the Indian Reorganization Act of 1934 (IRA), which authorizes the Secretary of Interior to acquire land and hold it in trust "for the purpose of providing land for Indians." (167) The Act defines "Indian" to "include all persons of Indian descent who are members of any recognized Indian tribe now under Federal jurisdiction." (166) At the time the IRA was enacted, federal authorities considered a tribe called the Narragansett to be under state, rather than federal, jurisdiction and refused to provide it with federal assistance. The Narragansett Tribe did not gain formal recognition from the federal government until 1983. (169) Several years later, a dispute arose regarding land that the Tribe had purchased. While litigation was pending, the Secretary notified the State of Rhode Island that he intended to accept the disputed land into trust. (170) Rhode Island challenged the Secretary's authority to accept the land into trust, arguing that the statutory term "now" refers to the time of the IRA's enactment, and permits the Secretary to take land into trust only for recognized tribes that were "under Federal jurisdiction" in 1934. (171)
A majority of the Court agreed with the State's construction. Among other canons, the majority opinion invoked the whole act rule "meaningful variation" principle, noting that other sections of the IRA refer to "now or hereafter," and arguing that this shows that Congress knows how to draw both contemporaneous and future events into a statute when it wants to, and that its failure to include "or hereafter" in the provision at issue means that only tribes recognized at the time of the IRA's enactment come under the provision. (172) The dissenting opinion, by contrast, invoked a different whole act rule subpart, the "consistent usage" rule, and different statutory text. It observed that the majority's construction works only if one reads the term "Indians" to refer to individuals, not an Indian tribe. (173) And it noted that Congress has used "Indians" and "Indian tribe" interchangeably in other parts of the IRA, and that the consistent usage rule dictates that Congress should be understood to have done the same in this provision of the IRA as well. (174)
b. The majority and dissenting opinions focus on different subparts of the statute at issue. This form of dueling over application of the whole act rule can involve application of the same whole act rule subpart to different statutory provisions, or can involve two different subparts of the whole act rule applied to two different statutory provisions. Begay v. United States (175) provides a good example of the first form of whole act rule dueling. Begay raised the question whether driving under the influence of alcohol is a "violent felony" under the Armed Career Criminal Act. The majority held that a DUI does not constitute a "violent felony," relying significantly on a whole act "rule against superfluity" argument that compared the first and second clauses of [section] 924(e)(2)(B) and noted that if "violent felony" meant all risky crimes, there would be no need for Congress to include specific examples of crimes such as burglary and arson in the second clause--the examples would be superfluous and would render the first clause, which covers the use of physical force, superfluous. (176) The dissent also made a whole act rule superfluity argument, but based on a different statutory provision--[section] 924(e)(2)(A), which includes offenses that "have as an element the use or threatened use of violence." (177) The dissent maintained that this subparagraph shows that [section] 924(e)(2)(B)(ii) is not limited to "violent" crimes because, if it were, it would be redundant, covering the exact same ground as [section] 924(e)(2)(A). (178)
c. Opposing opinions apply the same subpart of the whole act rule, but reach different end results. This form of dueling over application of the whole act rule is similar to the third form of plain meaning dueling, in which the majority and dissenting opinions take different views about what constitutes the "plain" meaning of the same statutory text. It suggests that judicial application of the canons suffers not merely from a "cherry-picking" problem, whereby judges have substantial discretion to decide which canons or tools to apply and which precise statutory terms and provisions to focus on, but that even when judges agree on the precise tool to apply and the precise terms or provisions of the statute at issue, there is substantial room for disagreement about the interpretive conclusion to be drawn. In other words, this form of dueling canon use suggests that it is not just the exceptions and countercanons that can render the tools of statutory interpretation indeterminate; but, rather, that the canons and interpretive tools themselves are fairly open to interpretive license.
The Court's dueling statutory-structure inferences in Dean v. United States (179) provide a good example. Dean involved a sentencing-enhancement statute that contained three subparts: the first imposes a five-year mandatory-minimum enhancement if a firearm is used or carried during and in relation to a violent crime, the second increases the enhancement to seven years "if the firearm is brandished," and the third increases the enhancement to ten years "if the firearm is discharged." (180) At issue was whether the ten-year mandatory enhancement applies if the gun was discharged accidentally, or whether there is an intent requirement for the ten-year enhancement. A majority of the Court interpreted the statute not to contain an intent requirement, relying significantly on the whole act rule "meaningful variation" and "holistic endeavor" subparts. The majority opinion pointed to the "is brandished" provision immediately preceding the "is discharged" provision and noted that it expressly includes an intent requirement, because the brandishing must be done "in order to intimidate." (181) It concluded that Congress's failure to include a similar intent requirement in the "is discharged" provision should be viewed as intentional and deliberate. (182) Justice Stevens' dissenting opinion also relied heavily on the whole act rule "holistic endeavor" subpart; but in contrast to the majority, it read the three enhancements clauses, taken as a whole, to provide escalating mandatory sentences for increasingly culpable conduct. (183) Based on this statutory structure, it drew a different conclusion from the intent requirement in the "is brandished" provision--reading it to mean that the "is discharged" clause also must contain an intent requirement because unintentional discharges are less culpable than intentional brandishing, and it would be nonsensical for clause (iii) to impose a higher sentence for conduct less culpable than that described in clause (ii). (184)
The multiple ways in which the Justices duel over application of the whole act rule are instructive. They reveal that majority-dissenting dueling is not necessarily a simple canon-for-countercanon affair. The Justices disagree not just over the applicable legal rule, but also over which parts of the statute are relevant to the interpretive question at issue. These doctrinal observations show that Justices seeking to counteract an opposing opinion's whole act rule argument will have many avenues available for crafting a dueling "parry." Indeed, given the evidence that federal statutes increasingly are drafted by multiple committees, with different sections pieced together in a hurried fashion--rather than carefully planned out by one drafter attentive to how different subparts interact--it is almost surprising that we do not see more judicial dueling over this interpretive tool.
3. Statutory Purpose. Again, there were a number of ways in which the members of the Roberts Court dueled over the proper application of statutory purpose in the majority and dissenting opinions in the same case. The primary patterns I observed were:
1. Majority and dissenting opinions focus on different, sometimes competing, statutory purposes.
2. Majority and dissenting opinions invoke the same statutory purpose, but draw different conclusions about how that purpose should impact their construction of the statute.
3. Related to the first, majority or dissenting opinion focuses on a broad, general statutory purpose, while the opposing opinion focuses on a narrow or specific purpose of the statute.
a. Majority and dissenting opinions focus on different, sometimes competing, statutory purposes. This form of dueling over statutory purpose is, I think, what Justice Scalia has in mind when he argues that interpreting statutes based on purpose leaves too much discretion to individual judges. Justice Scalia has, for example, noted that every statute has multiple purposes and that "[n]o legislation pursues its purposes at all costs" (185)--meaning that the limitations and compromises enacted into a statute are as much a part of its purpose as are its broad goals. (186)
The competing-purposes form of dueling occurred in fourteen of nineteen dueling purpose cases. (187) A good example is Conkright v. Frommert, (188) an ERISA case that raised the question whether courts must defer to subsequent interpretations made by a plan administrator whose initial interpretation of a benefits plan was deemed "unreasonable" by a court. (189) The majority opinion held that deference remained appropriate, relying in part on ERISA's statutory purpose of balancing the need to "ensur[e] fair and prompt enforcement of rights" with the need to "encourag[e] the creation of such plans" by employers. (190) The dissenting opinion, by contrast, construed the statute to allow more searching judicial review where a plan administrator's first interpretation was unfair to employees, arguing that this construction is most consistent with ERISA's "core purpose of promoting] the interests of employees and their beneficiaries in employee benefit plans." (191)
b. Majority and dissenting opinions invoke the same statutory purpose, but draw different conclusions about how that purpose should impact their construction of the statute. As with the "plain meaning" rule and the whole act rule, this form of dueling over the same, agreed-upon statutory purpose suggests an inherent looseness in translating statutory purpose into statutory construction, as opposed to mere judicial discretion in choosing among different versions of a canon or tool. That is, it suggests an "inherent indeterminacy," rather than merely a "cherry-picking" problem in the use of interpretive tools.
Edwards v. Aguillard, 482 U.S. 578, 636-37 (1987) (Scalia, J., dissenting) ("The number of possible motivations, to begin with, is not binary, or indeed even finite."); see also Barnhart v. Sigmon Coal Co., 534 U.S. 438, 461 (2002) (Thomas, J.) ("[The statute's] delicate crafting reflected a compromise amidst highly interested parties attempting to pull the provisions in different directions."); City of Joliet v. New West, L.P., 562 F.3d 830, 836-37 (7th Cir. 2009) (Easterbrook, C.J.) (stating that when courts look to purpose, "judges become effective lawmakers, bypassing the give-and-take of the legislative process").
The Court's opinions in Jerman v. Carlisle, McNellie, Rini, Kramer & Ulrich, L.P.A. (192) provide a good illustration. Jerman involved construction of the Fair Debt Collection Practices Act, which prohibits debt collectors from making false representations to debtors about a debt's character, amount, or legal status. (193) The Act contains an exception if a debt collector can show that its violation "was not intentional and resulted from a bona fide error." (194) At issue was whether a law firm that filed a lawsuit on behalf of a mortgage company mistakenly seeking to foreclose on property owned by Jerman qualified for the bona fide error exception. (195) A majority of the Court concluded that the bona fide error defense does not apply to violations that result from a debt collector's mistaken interpretation of the legal requirements of the FDCPA. (196) In so ruling, the Court cited the statute's purpose of "eliminat[ing] abusive debt collection practices by debt collectors [and] insur[ing] that those debt collectors who refrain from using abusive debt collection practices are not competitively disadvantaged" and argued "that immunizing debt collectors who adopt aggressive but mistaken interpretations of the law would be inconsistent with the statute's broadly worded prohibitions on debt collector misconduct." (197) The dissenting opinion quoted the same statutory purpose, but concluded that when referring to "abusive debt collection practices" Congress "surely did not contemplate attorneys who act based on reasonable, albeit ultimately mistaken legal interpretations" and noted that debt collectors do not gain a competitive advantage by making good-faith legal errors. (198)
c. Majority or dissenting opinion focuses on a broad, general statutory purpose, while the opposing opinion focuses on a narrow, or specific purpose of the statute. This form of dueling over statutory purpose also may be one that Justice Scalia has in mind when he criticizes purpose-based statutory interpretation. It is a standard textualist complaint that by framing the statutory purpose at whatever level of abstraction she wishes, a judge can reach the interpretive outcome that she prefers. (199)
Ledbetter v. Goodyear Tire & Rubber Co. (200) provides a good example of this form of dueling over statutory purpose. In Ledbetter, a female retiree sued her former employer under Title VII and the Equal Pay Act, alleging that poor performance evaluations she received as a result of sex-based discrimination early in her tenure resulted in her receiving lower pay than her male colleagues from the time of the evaluations through the end of her career. (201) Goodyear argued that Ledbetter's claims were time-barred under a Title VII provision requiring that individuals who wish to bring a discriminatory pay lawsuit must file a charge with the EEOC within 180 days "after the alleged unlawful employment practice occurred." (202) Ledbetter countered that her claims were timely because each unequal paycheck she received constituted a new Title VII violation and gave her an additional 180 days to file an EEOC charge. (203) A majority of the Court disagreed, concluding that the later effects of past discrimination do not restart the clock for filing an EEOC charge. (204) The majority opinion relied in part on the narrow, specific purpose of the time limitation, noting that statutes of limitations serve a "policy of repose." (205) The dissenting opinion, by contrast, focused on Title VII's broad, general purpose of "assuring] equality of employment opportunities" and "mak[ing] persons whole for injuries suffered on account of unlawful employment discrimination." (206) The dissent also criticized the majority's construction as "totally at odds with the robust protection against workplace discrimination Congress intended Title VII to secure." (207)
4. Practical Consequences. Finally, many of the Roberts Court's statutory cases contain dueling assessments of the practical consequences that are likely to flow from adopting a particular interpretation. Within this subset of dueling opinions, one pattern or divide is prominent: in 61.3 percent of the cases, one opinion makes an "administrability" argument, while the opposing opinion makes a "policy constancy" argument. (208) In an earlier empirical study of the Roberts Court's statutory interpretation practices, I described the "administrability" form of practical consequences as encompassing discussions about the practical difficulty of implementing the particular interpretation, the likely effect the interpretation will have on judicial or other public resources, the consistency or lack of consistency between federal and state laws created by the interpretation, and the clarity or predictability of the legal rule or landscape going forward in light of the interpretation. (209) The "policy constancy" form of practical consequences, by contrast, includes discussions about inconsistencies in statutory policy likely to result from a particular interpretation, the equity or justness of the interpretation, the likelihood that the interpretation will render the statutory provision "meaningless" or ineffective, and assertions that logical absurdities or statutory incoherence will result from the interpretation. (210)
There are many examples of majority-dissent division over administrability versus policy-constancy concerns. In Gonzalez v. United States, (211) the Court considered whether the Federal Magistrate
Act empowers magistrate judges to preside over the voir dire and selection of prospective jurors in a felony criminal trial based on the consent of defense counsel, or whether the Act requires the personal consent of the defendant for magistrate supervision. (212) A majority of the Court read the FMA to allow magistrate supervision based on defense counsel consent alone. (213) In so ruling, the majority emphasized the practical necessity of giving defense counsel control over trial-management matters--an "administrability" argument. (214) "The adversary process could not function effectively if every tactical decision required client approval." (215) The dissenting opinion countered with a "policy constancy" absurd results argument, noting that the statute requires defendant's express, informed consent in order for a magistrate judge to conduct a misdemeanor trial and calling it a "glaring" inconsistency to read the statute to require greater consent for defendant's waiver of rights in a misdemeanor case than in a felony case. (216)
In earlier work, I described how the "administrability" versus "policy constancy" divide reflects a deeper jurisprudential divide between the Justices over the kind of coherence that jurists should prioritize when construing statutes; (217) the fact that this divide shows up prominently in the dueling practical consequences cases is intriguing, and its implications are explored further in Part III of this Article.
B. Other Observations
Last, let us turn to judicial dueling over dictionary definitions and legislative history. No noteworthy patterns were found regarding how the Justices dueled over either of these interpretive tools. Nevertheless, it is worth discussing doctrinal trends in the cases involving dueling over these tools because, as noted earlier, the low level of dueling for these tools was so surprising.
1. Dictionary Definitions. Two things stand out when one examines the dueling dictionary cases. First, in most of the cases, majority and dissenting opinions dueled over different definitions of the same word, as one would expect. (218) However, in a nontrivial minority of the cases, opposing opinions dueled over the dictionary definitions for different words. (219) This parallels the manner in which the Justices sometimes dueled over the plain meaning of different statutory words and phrases. (220)
Second, when the Justices did duel over dictionary definitions, they rarely placed significant weight on such definitions as an interpretive aid. Of the more than thirty-two opinions in the sixteen cases that involved dueling dictionary definitions, the vast majority referenced such definitions in a corroborative manner, to note that they pointed in the same direction as other interpretive tools. Some opinions even referenced dictionary definitions only to note that they were not particularly helpful in construing the statute. Only six opinions seemed to use the dictionary in a manner that was dispositive, or that significantly directed the interpretive outcome--all but one of these was authored by Justice Scalia or Justice Alito. (221) These findings are consistent with a recent empirical study about the Supreme Court's dictionary use, which found that the bulk of the Court's dictionary references employed the dictionary in an "ornamental role"--rather than as a primary aid in identifying the best interpretation--and that when the Court did use dictionary definitions in a dispositive manner, it was almost exclusively in an opinion authored by one of the conservative Justices." (222)
These findings also suggest an explanation for the Court's low overall rate of dueling over dictionary definitions, despite the abundance of dictionaries and definitions within those dictionaries. Perhaps, when confronted with an opinion that employs a dictionary definition in an "ornamental" or corroborative fashion, the Justices writing in opposition do not feel compelled to invoke a countervailing dictionary definition to undermine the first opinion's reasoning. That is, perhaps dueling over dictionary definitions does not seem worth the effort to the Justices because the dictionary definition does not seem to add much to their opponent's statutory construction--so that providing a dueling definition conversely would not subtract much from the opposing construction.
2. Legislative History. With respect to legislative history, I examined the dueling cases for patterns such as one opinion that relied on legislative history countered by another that used legislative history merely to corroborate, or one opinion that focused on the evolution of the statute countered by another that invoked statements from the legislative record. But no such patterns emerged. Instead, most of the dueling legislative history cases involved majority and dissenting opinions that relied on different pieces of legislative history, such as a conference report versus a rejected proposal, or floor statements versus committee reports, and so on." (223) Only a handful of the cases even arguably involved majority and dissent dueling over the meaning of the same form of legislative history. (224) Thus, when the Justices dueled over legislative history, they typically did so in the "cocktail party" or "something for everyone" manner that textualists decry.
Further, in contrast to the Justices' largely "ornamental" use of dictionary definitions, the overwhelming majority of the dueling legislative history cases involved at least one opinion that relied on legislative history to reach its chosen construction, rather than merely to corroborate a construction arrived at through other interpretive tools. Half of the cases involved both majority and dissenting opinions that relied on legislative history. These data might lead one to hypothesize that the low overall rate of judicial dueling results from judicial disinclination to duel in cases where legislative history is used only to corroborate a statute's meaning; that is, perhaps most of the Roberts Court opinions that employ legislative history do so in a corroborative manner--and perhaps the Justices who author opposing opinions in those cases do not consider it necessary to counteract such secondary interpretive references. I explored this possibility, but the data were not consistent with such an explanation--most of the opinions in the study that referenced legislative history relied on such history, rather than used it in a corroborative manner, yet in the vast majority of those cases opposing opinions did not make competing legislative history references. (225)
I also reviewed the dueling legislative history cases to determine whether they involved "smoking-gun" pieces of legislative history that seemed to address the precise interpretive question at issue--perhaps explaining why these particular cases triggered a response of some kind in the opposing opinion. I found, however, that "smoking-gun" pieces of legislative history were rare and referenced in only a few of the dueling cases. (226)
Finally, in order to test the possibility that competing legislative history arguments may not have been available in many of the cases in which one opinion referenced legislative history, but the opposing opinion(s) did not, I examined the briefs in fifty-four of the fifty-six cases in the dataset in which only one opinion cited legislative history. (227) I found that in forty-eight of these cases ((88) . (9) percent), at least one brief supporting the reading opposed by the legislative-history-citing opinion contained competing legislative history references. (228) So the textualist claim that some argument from legislative history can be manufactured to support almost any reading does seem to hold true, although this does not seem to be translating into judicial manipulation of the legislative history in the manner that textualists predict.
Overall, the above patterns of dueling interpretive tool use show that there are numerous ways in which judges may disagree over the application of a particular interpretive resource or canon. There are not simply canons and countercanons, as Llewellyn pointed out, and there is not only a cherry-picking problem as Bill Eskridge has pointed out. (229) Rather, even when judges use the same interpretive tool, they may disagree about the statutory word, phrase, provision, or subpart to which that tool should be applied; they might draw competing inferences using the same canon or interpretive rule, or they might bring different focuses (general versus specific) when applying an interpretive tool. All of this amounts to significant judicial discretion in the application of particular interpretive tools and canons. Indeed, when viewed in light of all the possible ways the canons could be used in a dueling manner, it is surprising that the Court does not show higher rates of dueling use for most statutory-interpretation-specific tools.
IV. THEORETICAL IMPLICATIONS
The data reported in Part II have important implications for several prominent theories of statutory interpretation. The implications are greatest for interpretive theories that emphasize predictability or that seek to constrain judicial discretion, but theories that rely on other justifications also can benefit from--or be undermined by--this study's findings. This Part explores the theoretical upshot of the Roberts Court's dueling canon practices. It argues that the dueling canon data undermine several normative claims made by textualists, (230) calling into question key justifications that theory offers for privileging certain interpretive tools over others. At the same time, however, the data do not provide any particular support for textualism's chief competitor theory, purposivism. The data do lend some support, at least descriptively, to pragmatic theories of statutory interpretation--highlighting the significant role that practical reasoning plays in the Court's decisions and divisions. The Part concludes by discussing what the dueling canon data might instruct about the viability of methodological stare decisis--the suggestion that federal courts should adopt a binding statutory interpretation methodology.
Textualism is a formalist method of statutory interpretation that seeks answers primarily from the official language of the statute. It directs judges to identify the ordinary meaning, at the time of enactment, of the statutory term in question. (231) It prioritizes clarity and predictability (232) and is founded on a belief that there is one correct, definitive answer to every interpretive question. Textualism treats the interpretive process like a puzzle; (233) if the correct answer cannot be found through a plain reading of the text, then the dictionary, the statute's structure, and the Court's prior interpretations of the same word or phrase in other statutes should be consulted to decipher the statute's meaning. Textualists trust such bounded interpretive aids to lead courts to the proper statutory construction--and to restrict the opportunity for "strong-willed judges to substitute their own personal political views for those of the legislature." (234)
Related to its emphasis on predictability and judicial constraint, textualism fervently rejects judicial inquiry into legislative history, intent, and statutory purpose. (235) Textualists have articulated numerous objections to these interpretive sources, particularly legislative history. Formally, they point out that only the statutory text is law and that statements in the legislative record about statutory purpose or the meaning of particular words have not been enacted following the bicameralism and presentment requirements articulated in Article I, Section 7. (236) Textualists also challenge the concept of collective intent, noting that legislative history usually is created by congressional staffers or, at best, a few legislators--so it does not represent the views of the legislature as a whole. (237) Most relevantly for a study of dueling canons, textualists argue that judges use legislative history and statutory purpose selectively, picking and choosing statements or statutory objectives that support constructions that align with their personal policy preferences. (238) As Justice Scalia has commented, "In any major piece of legislation, the legislative history is extensive, and there is something for everybody." (239) In this vein, textualists are fond of invoking Judge Harold Leventhal's famous quip that using legislative history is like going to a cocktail party, and "look[ing] over the heads of the crowd to pick out your friends." (240)
The dueling canon data pose some serious problems for the textualist account. First, the data undermine textualism's central claim that text-based analysis constrains judicial decision-making and leads jurists to one correct result. (241) Most notably, the high rate of majority-dissent dueling over the plain meaning of statutory text-- over 40 percent--indicates that the Justices are not finding textual analysis particularly clarifying or determinative of one correct answer. Further, the data in Table 6f, reporting the ideological direction of the Justices' votes in the text/plain meaning dueling cases, suggest that the plain meaning rule is not constraining the Justices ideologically either. (242) Second, the data also undermine textualism's claim that interpretive tools such as legislative history, purpose, and intent are inherently more indeterminate or manipulable than textualism's favored tools. (243) There is no evidence, for example, that statutory purpose, legislative history, or intent is particularly susceptible to dueling use. Table (1) reports low overall levels of dueling for each of these interpretive tools--at or near 25 percent. (244) More importantly, these purposivist tools exhibited rates of dueling that were roughly equal to some of textualism's most favored tools--e.g., language canons/whole act rule and dictionary references--and exhibited lower rates of dueling than other textualism-favored tools, such as the plain meaning rule and other statutes. (245)
Textualists might counter that the low rates of dueling over purpose, legislative history, and intent reflect the fact that some Justices--particularly those who follow a textualist methodology-- refuse to reference these interpretive tools as a matter of principle. Thus, even where competing applications of these three interpretive tools exist, they may not be invoked in a dueling manner in cases authored by these Justices. If this is true, the argument might continue, then the dueling rates for these tools could be dampened--and therefore should be taken not as evidence that the purposivist tools are not manipulable or indeterminate, but merely as evidence that these tools are not popular, or widely accepted interpretive resources.
There is some surface appeal to this critique, as the overall reference rates for individual Justices show that Justices Scalia, Thomas, and Roberts rarely invoked purpose, legislative history, or intent in the statutory opinions they authored. (246) For a number of reasons, however, I do not believe that this factor can sufficiently explain the low rate of judicial dueling over the purposivist tools or reconcile the lack of more significant dueling with the traditional textualist critique of these interpretive tools. First, Justice Roberts has not expressed a philosophical opposition to citing legislative history, statutory purpose, or intent--in fact, he has expressly endorsed legislative history use when the statutory text is ambiguous; (247) so his low rates of legislative history citation could have many causes, including a failure to locate on-point legislative history, purpose, or intent in the cases he authored.
Second, although Justices Scalia, Thomas, and Roberts rarely invoked the purposivist interpretive tools in the opinions they authored, all of these Justices proved quite willing to join opinions that dueled with opposing opinions over statutory purpose or legislative history. (248) This is consistent with Brudney and Ditslear's pointed finding that there was no "Scalia Effect" for opinions authored by other conservative Justices. (249) Thus, it is difficult to tell whether Justices Scalia, Thomas, and Roberts's low rates of legislative history and purpose reliance are caused by philosophical objections--or by a lack of available legislative history, purpose, or intent arguments in the cases in which they authored opinions, or by some other factor.
Third, there is a difference between statements about statutory meaning made by legislators in the legislative record, on the one hand, and the objective statutory history that shows how a statute evolved over time, through multiple drafts and amendments, on the other. Although textualists object to the use of statements in the legislative record, even the most ardent among them is willing to consult evidence of how a particular statutory phrase or provision evolved over time. (250) The cases in my dataset were coded both for references to statements in the legislative record and for references to the evolution of the statute form of legislative history that textualists favor. Thus, philosophical objections to legislative history use cannot entirely explain textualist Justices' low rates of reference, or dueling over, this interpretive tool. Further, all of the Justices other than Justices Scalia, Thomas, and Roberts--including conservative Justices Alito and Kennedy--were willing to invoke statutory purpose and legislative history in a sizeable number of the cases they authored, which means that legislative history dueling was possible in all of the opposing opinions authored by six of the nine Justices on the Court at any given time. (251)
Finally, it is possible to take a crude measure of what the Justices' dueling over legislative history, purpose, and intent might look like without the skewing effect potentially created by Justices Scalia, Thomas, and Roberts--by examining the subset of cases in which none of these Justices authored an opinion on either side. (252) When we adjust the dataset in this manner, the results are surprising, and provide only mixed support for the "refusal to cite" hypothesis. Most surprisingly, the rate of judicial dueling for legislative history actually decreases, from 25.3 percent to 16.7 percent, when opinions authored by these three Justices are removed! (253) This means that the Court was actually more likely to duel over legislative history as an interpretive resource in cases where Justices Scalia, Thomas, or Roberts authored an opinion than in cases where they did not. Indeed, of the nineteen dueling legislative history cases in the full dataset, Justice Scalia authored one of the dueling opinions in five cases, and Justice Thomas did so in another four. (254) In other words, although Justices Scalia and Thomas may not prefer to reference legislative history of their own accord in the vast majority of opinions they author, these Justices seem willing to counter an opposing opinion's legislative history references with competing references of their own.
For purpose and intent, by contrast, when we eliminate opinions authored by Justices Scalia, Thomas, and Roberts from the dataset, the rates of judicial dueling increase, from 24.7 percent to 32.0 percent, and from 26.5 percent to 35.8 percent, respectively. (255) These latter rates do provide some crude evidence of a dampening effect on dueling in opinions authored by these three Justices. However, even with this adjustment, the rates of judicial dueling over purpose and intent remain comparable to the rates of dueling over many textualist-preferred tools--slightly higher than dictionary references (28.6 percent), at roughly the same level as the other statutes tool (34.5 percent), and below the rate for text/plain meaning rule (42.7 percent).
At bottom, the problem for textualism is that its claims about the ubiquity and malleability of legislative history, statutory purpose, and congressional intent have been so vigorous. In arguing that "there is something for everyone" and invoking Judge Leventhal's cocktail party quip, textualists have implied that competing purposes, congressional intent, and legislative history should be available--or manufacturable--in most cases. Against that rubric, a dueling rate of 25 percent seems quite low, even if one accepts some dampening effect based on Justices Scalia's and Thomas's disinclination to reference statutory purpose, intent, or the legislative record form of legislative history in the opinions that they author. Relatedly, textualists have made such strong claims about the ability of text-based interpretive tools to lead judges to one correct result that the rough equivalency between the rates of dueling exhibited for these interpretive tools and the purposivist tools is glaring. (256)
In short, the dueling canon data suggest that textualism's underlying assumptions about different interpretive tools' ability to constrain judges may be incorrect. If textualism wants to push interpreters to privilege text-based interpretive tools over tools such as purpose and legislative history, then, it may need a different theoretical rationale for this methodological prioritization. Perhaps that rationale is the formalist Article I, Section 7 argument that only the statutory text is law. Perhaps it is the landscape coherence idea that individual statutes should be interpreted in a manner that is consistent with the rest of the corpus juris. (257) Textualism already relies significantly on both of these justifications, so the lesson from the dueling canons study may simply be that the judicial restraint component of the textualist argument against legislative history and related purposivist tools of construction has been called into doubt. (258)
At the same time, the dueling canon data suggest that textualism may be able to find a way to render its preferred interpretive tool--the plain meaning rule--more constraining, or predictable. The doctrinal analysis in Section II.C showed that much of the dueling that occurs over the plain meaning rule involves one opinion that follows the "prototypical meaning" of a statute, pitted against an opposing opinion that employs a "legalist meaning" of the statute. This observation raises the possibility that courts could eliminate at least some judicial dueling that occurs over this interpretive tool by choosing decisively between prototypical and legalist meaning--that is, by declaring that plain meaning means the prototypical meaning, not the legalist meaning, or vice-versa--and by dictating the kinds of measures that jurists should look for when applying the plain meaning rule going forward (for example, the core example the legislature was seeking to cover or, conversely, the meaning given to a word or phrase in other parts of the corpus juris). (259) In other words, courts could adopt a sort of binding, mini-methodological stare decisis dictating how the plain meaning rule should be applied going forward. The data from the Roberts Court's first five-and-a-half terms suggest that such a rule could reduce the frequency of plain meaning dueling at least a little, and perhaps by as much as one-third.
Purposivism is an interpretive approach associated with the Legal Process movement. (260) In contrast to textualism, it advocates that jurists interpret the words of a statute by identifying the statute's purpose and selecting the meaning that best effectuates that purpose. (261) Purposive statutory interpretation typically involves inquiries into legislative history, the societal problem that prompted the legislature to enact the statute, legislative intent, and other sources that might illuminate a statute's objectives. It can entail guesswork and judicial discretion, but is often defended on the ground that reliance on legislative history and purpose helps restrict judicial discretion and fulfill congressional intent. (262)
Purposivism once was the dominant approach to statutory interpretation, but over the past few decades it has come under significant attack. (263) The criticism has come from both the New Textualism, which arose largely as a reaction to purposivism, and from public choice theorists, who have criticized purposivism as providing a rose-colored glasses view of the legislative process. Both sets of critics have emphasized the open-endedness of the search for statutory purpose and have argued that purpose-based interpretation enables judges to import their personal policy preferences into the statute. (264)
This study's dueling canon findings provide mostly good news for purposivism. First and foremost, purposivism can celebrate that the data show no evidence supporting textualism's attack on purposivist-preferred tools of interpretation. Indeed, for purposivists, the headline from the study should be that as interpretive tools, purpose, legislative history, and intent appear no more susceptible to judicial shaping to support competing interpretations than do textualist-favored tools such as the whole act rule, dictionary definitions, and the plain meaning rule. Moreover, during the period studied, the purposivist-interpretive tools were used to support competing readings of the same statute in only one-fourth of the cases in which they were invoked. (265) These data came as quite a surprise because most statutes have more than one motivating purpose and because legislative history (particularly statements made on the House and Senate floor) can cut in multiple directions--rendering these two purposivist interpretive tools ripe for dueling use in opposing opinions.
On the other hand, the data do not show any evidence supporting purposivism's affirmative claims that statutory purpose and legislative history constrain judges or ensure that jurists give statutes a meaning that is faithful to Congress's intent. In fact, the doctrinal analysis in the next section illuminates several ways in which judges can disagree over how to characterize a statute's purpose or over what the legislative history means. Moreover, Tables 6a and 6b demonstrate that when the Justices duel over statutory purpose and legislative history, they do so in a manner that is highly ideological. (The data for legislative intent are better, suggesting some constraining effect for Justices Ginsburg and Breyer). (266) Thus, while the frequency of judicial dueling over purpose, intent, and legislative history might be relatively low, the manner in which the Justices duel over these interpretive tools shows that there is substantial room for disagreement in application--and that these tools seem incapable of constraining judges to vote against their ideological preferences.
As with the plain meaning rule, however, the inside legislative history doctrinal analysis suggests a way that this purposivist tool might be made more determinate or predictable in application. Specifically, most of the Roberts Court's dueling over legislative history involved a majority opinion that cited one form of legislative history (e.g., a conference report) and a dissenting opinion that referenced a different form of legislative history (e.g., a rejected proposal). Thus, if courts were to adopt a binding hierarchy of legislative history, dictating a precise order in which judges should rely on specific forms of legislative history, at least some of the dueling over this interpretive tool potentially could be diminished. For example, the Supreme Court could adopt a rule that, say, conference reports trump all other forms of legislative history, followed by regular committee reports, statements by sponsors, rejected proposals, and so on down the line. If the Court enforced such a rule strictly, then the only forms of dueling over legislative history that should occur are disagreements about which interpretation a particular piece of legislative history favors and/or competing inferences drawn from inconsistent statements in the same piece of legislative history. This latter form of dueling was rare--occurring in only six cases in the dataset. (267) This suggests that a binding hierarchy of legislative history sources could provide clarity about which interpretation the legislative history points toward in a particular case, enabling litigants and their attorneys to know where to concentrate their resources when researching legislative history and perhaps making legislative history use more predictable.
Significantly, an informal hierarchy of legislative history sources does already exist. (268) Committee reports, for example, are widely considered to be the most authoritative form of legislative history, falling second only to conference reports. (269) And empirical studies have found that the majority of legislative history references made in Supreme Court opinions are to committee reports. (270) So adopting a formal hierarchy of legislative history sources would not require a sea change in the Court's interpretive practices or even great debate about the appropriate order of sources. What it would require is a judicial precommitment to employ the most authoritative forms of legislative history when they are available, and judicial discipline to avoid invoking less authoritative forms when the latter contradict the most authoritative forms. Of course, one cannot expect such a hierarchy to eliminate all judicial dueling over legislative history--the battleground could simply shift from competing forms of legislative history to arguments over the relevance that a particular piece of the legislative record bears for a particular interpretive question. (271) But a hierarchy could help to reduce the already low overall rate of dueling over legislative history by creating greater consistency in at least some cases. For example, where a conference committee report and a floor statement directly conflict on the same question, a hierarchy of sources would tell us which one ranks as more authoritative.
Pragmatism is not a jurisprudential approach that claims to ensure judicial constraint or to improve the predictability of statutory interpretation. Rather, it posits only that judges should construe statutes by focusing on the practical consequences that will result from an interpretation and seeking the best result. There is no single or uniform theory of pragmatism, but its various strands are united--and distinct from textualism and purposivism--in their open emphasis on outcomes and practical consequences and their frank recognition that statutory interpretation involves creative policymaking by judges. Judge Richard Posner, for example, has argued that the goal of statutory interpretation should be to produce the best results for society. (272) Judge Posner defines pragmatism, at its core, as "a disposition to base action on facts and consequences rather than on conceptualisms, generalities, pieties, and slogans." (273) Bill Eskridge similarly urges that statutory interpreters should take public values into account and construe statutes dynamically--to reflect current social, political, and legal contexts. (274)
Notably, pragmatism shares much in common with legal realism, which challenges the rule-bound formalist narrative and insists that judges decide cases based significantly on social or practical considerations. (275) In fact, Llewellyn himself was a pragmatist, albeit one who focused on disproving the formalist account as an impossible ideal and providing a more accurate description of how judges actually decide cases, rather than on articulating a normative theory for how judges should decide cases.
It is perhaps fitting, then, that the dueling canon data provide some descriptive support for pragmatism--and the legal realist account. As Table 1 shows, the Justices on the Roberts Court dueled over practical-consequences considerations slightly more often than they dueled over most traditional canons and tools of statutory construction--indeed, more often than they dueled over every other interpretive tool save the plain meaning rule and other statutes. (276) Moreover, all of the Justices other than Justice Thomas referenced practical consequences frequently in the statutory opinions they authored, as Table 9 reports. (277) Only the plain meaning rule and Supreme Court precedent exhibited higher rates of dueling consistently, across all Justices. In this sense, the data vindicate the larger point made in Llewellyn's article, that statutory cases are not very different from common-law cases in their susceptibility to judicial policymaking, despite the existence of neutral canons of construction.
What is perhaps most interesting about the Court's practical consequences dueling is that it occurs despite wide-spread concerns about judicial policymaking. Federal judges tend to resist the idea of making any kind of federal common law and their concerns are heightened in the statutory context, where legislative supremacy looms large. (278) Given this, the fact that the Roberts Court Justices openly dueled over practical consequences at a rate of 30.9 percent in their statutory cases suggests that the Justices believe that explaining an interpretation's practical effects will persuade readers of the interpretation's correctness or, at the least, that countering an opposing opinion's characterization of the consequences that would follow from an interpretation is necessary to persuade opinion readers. Otherwise, it would have been safer for the Justices to avoid mentioning that pragmatic considerations played any role in the Court's reasoning, for fear of being charged with judicial activism.
Finally, the doctrinal discovery of the Court's administrability-versus-policy-constancy dueling suggests that practical-consequences-based reasoning may be less ad hoc than the conventional wisdom suggests, and perhaps more predictable than pragmatists themselves realize. We now know that the Justices tend to focus their practical reasoning on concerns such as an interpretation's likely effect on judicial resources, the difficulty of implementing the interpretation, the clarity and predictability of the rule the interpretation creates or, conversely, on the fairness, arbitrariness, or consistency over time of an interpretation. Moreover, we know that these two sets of concerns sometimes point to different interpretations in the same case, and that different Justices prioritize these two sets of concerns differently. This understanding gives litigants a better sense of what to expect from judicial construction and scholars a better sense of what practical reasoning means in practice. It also opens the door to new debates within pragmatism, suggesting that it may no longer be enough for pragmatists to urge judges to choose the construction that achieves the "best results"--but may now also require pragmatists to provide more concrete accounts of what kinds of practical consequences judges should concern themselves with, and which consequences judges should seek to ensure or avoid when different consequences point in conflicting directions.
D. Methodological Stare Decisis
Recent scholarly work has illuminated efforts by state courts and state legislatures to dictate a binding hierarchy of interpretive tools that courts must follow, in a prescribed order--what Abbe Gluck has called "methodological stare decisis." (279) A good example is the Oregon Supreme Court's three-step interpretive framework, which the state court adopted on its own initiative and followed "religiously" for sixteen years. (280) Oregon's framework consisted of a three-tier hierarchy of interpretive resources that required courts to consult textual canons at step one, legislative history at step two if and only if the textual canons proved inconclusive, and substantive canons at step three, only as a last resort if both textual tools and legislative history proved indeterminate. (281) Proponents of methodological stare decisis have argued that this kind of binding interpretive framework renders statutory interpretation more predictable, promotes consensus, conserves litigant and judicial resources, and make courts seem less results-oriented. (282) Critics have argued that federal courts cannot agree on an interpretive framework, that diversity in interpretive methods improves the quality of deliberation about statutory meaning, and that different interpretive techniques may be appropriate for different courts. (283) This Article need not take sides in the debate over the normative value of methodological stare decisis. Rather, it is possible to bracket the question of the desirability of methodological consistency and focus instead on what the dueling canon data might teach us about the workability of methodological stare decisis in federal courts.
On the one hand, the dueling canon data provide some reason for optimism regarding the feasibility of methodological stare decisis. One significant concern about methodological stare decisis is that it simply will not work in federal courts, particularly at the Supreme Court level, where most interpretive questions are close and individual canons hypothetically are more likely to point in multiple directions. The Roberts Court's low overall rates of judicial dueling for most interpretive tools, however, contradict that gloomy prediction, showing that even in cases involving close statutory questions, the Justices did not tend to find multiple applications for most interpretive tools most of the time. (284) In so doing, the data demonstrate that it is quite possible that if the Justices were able to agree on a binding hierarchy of interpretive tools to use in a prescribed order, they could apply those tools in a nondueling manner much of the time.
On the other hand, two aspects of the dueling canon data caution against too much optimism about the prospects for methodological stare decisis. First, despite the low rates of judicial dueling reported for most interpretive tools, the Justices' rates of dueling over the plain meaning rule were rather high. This is significant because, as Oregon's framework illustrates, the plain meaning rule tends to place high in the hierarchy of interpretive tools that courts prescribe when they adopt some form of methodological stare decisis. (285) Thus, the dueling canon data suggest that the first interpretive tool in a preordained interpretive regime might prove inconclusive and have to be bypassed in many cases. Second, this Article's doctrinal analysis demonstrates that there are countless ways for interpreters to divide over the application of a particular interpretive tool, and countless ways for judges to invoke the same interpretive tool to reach different results. (286) Majority and dissenting opinions might focus on different statutory words, invoke different subparts of the whole act rule, argue about which subsection of a statute is relevant, disagree over whether to follow a statute's general versus its specific purpose, and so on. These multiple avenues for conflicting application of the same interpretive tool raise the possibility that even if the Justices could agree on a binding hierarchy of interpretive tools--e.g., committing to employ textual tools of interpretation before considering others--the result might be much higher rates of dueling over textualist tools of interpretation such as the whole act rule, dictionary definitions, and other statutes, rather than more judicial consensus. The same danger of heightened dueling exists for legislative history and other tools that are included lower in the interpretive hierarchy, as these tools become relevant when the textualist tools do not produce a clear answer. That is, in the tough cases with significant stakes, the judicial battleground might simply shift from disagreement over which tools to privilege to whatever terrain the new methodological regime requires--e.g., what the "prototypical meaning" of a particular word is.
Conversely, we might see Justices in opposing opinions claiming less often that statutory text or the whole act rule dictates an opposing construction, and more likely to characterize these interpretive tools as creating ambiguity--so that they legitimately can bring in helpful legislative history or other interpretive tools authorized at later steps. If either of these results ensues, methodological stare decisis might not, in practice, work to ensure the consensus, predictability, conservation of resources and the like in the federal context that it seems to have achieved in state courts.
Nevertheless, this study's findings might provide at least a modest boost for methodological stare decisis. As noted earlier, many of the doctrinal patterns identified within the Justices' dueling over individual interpretive tools--e.g., prototypical versus legalist meaning, administrability versus policy-constancy-based practical reasoning, different forms of legislative history--suggest small-scale opportunities for courts to adopt binding methodological rules governing the application of particular interpretive tools, in the name of greater predictability and clarity. That is, courts need not adopt an Oregon-like rigid hierarchy of interpretive tools but could, instead, bind themselves to consult only certain forms of legislative history (and to reject others or use them only if more authoritative forms are not available) or to seek the prototypical rather than the legalist plain meaning, or to privilege administrability--type practical concerns over policy-constancy concerns. Even if such small-scale binding rules could not ensure greater predictability or judicial restraint in high-stakes cases, they might do so for more run-of-the-mill, low-stakes cases.
More than sixty years ago, Karl Llewellyn inflicted serious injury upon the canons of construction, providing what many have regarded as anecdotal "proof" that the canons are indeterminate, easily manipulated facades used to give judicial decisionmaking a false aura of objectivity. Textualists long have argued that Llewellyn's critique was grossly exaggerated and that, in fact, it is the purposivist tools of construction that are indeterminate and easily manipulated by judges. Purposivists have countered that their preferred interpretive tools ensure fidelity to legislative intent, taking for granted that such tools point toward only one statutory construction. No one has bothered to test these relative claims empirically. This Article begins the effort to do so, seeking to measure the extent to which jurists serving on the nation's high Court employ individual interpretive tools to reach competing outcomes in the same case. The measure is imperfect, but it reveals a lot about how the Justices publicly defend their statutory constructions. Ultimately, the data suggest that some of the central claims made by textualism may be unfounded, and that pragmatism may benefit from more granular attention to the specific forms of practical reasoning the Court seems to employ, and to divide over. Throughout, this Article's aim has been to illuminate how the Court's actual practices align--or not--with the myths suggested both by Llewellyn's infamous list of canons and countercanons and by the most prominent theories of statutory interpretation.
Appendix: Dueling Canons by Case Legend: 1 = Text; 2 = Dictionary; 3 = Language/Grammar/Whole Act Rule; 4 = Other Statutes; 5 = Purpose; 6 = Legislative History; 7 = Substantive Canons; 8 = Common Law; 9 = Intent; 10 = Supreme Court Precedent; 11 = Practical Consequences. Canons 1 2 3 4 5 6 Case Rapanos v. United States Y Y (4-1-4) Hamdan v. Rumsfeld Y Y Y Y (5-4) AT&T Corp. v. Hulteen Y Y (7-2) Ledbetter v. Y Goodyear Tire & Rubber Co. (5-4) FCC v. Y Y Fox Television Stations, Inc. (5-4) Boumediene v. Bush Y Y (5-4) Massachusetts v. EPA Y (5-4) Altria Grp., Inc. v. Good Y Y (5-4) Bartlett v. Strickland Y (5-4) Harbison v. Bell Y Y Y (5-2-2) Zuni Public Sch. Dist. No. 89 v. Y U.S. Dep't of Educ. (5-4) Forest Grove Sch. Dist. v. T.A. Y Y Y (6-3) Riley v. Kennedy (7-2) Ricci v. DeStefano Y Y (5-4) Kasten v. Saint-Gobain Y Y Performance Plastics Corp. (6-2) Arlington Cent. Sch. Dist. Bd. of Y Educ. v. Murphy (5-1-3) Bloate v. United States Y Y (7-2) Bruesewitz v. Wyeth LLC Y Y Y Y (6-1-2) Wyeth v. Levine Y (5-2-3) PLIVA, Inc. v. Mensing (5-4) Marrama v. Y Citizens Bank of Mass. (5-4) Morgan Stanley Capital Grp., Y Inc. v. Publ. Util. Dist. No. 1 (5-2) Woodford v. Ngo (5-1-3) United States v. Santos Y Y (4-1-4) Gross v. FBL Fin. Servs., Inc. Y Y (5-4) Brown v. Plata Y Y (5-4) Gomez-Perez v. Potter Y (6-3) Lopez v. Gonzales Y Y Y (8-1) United States v. Rodriquez (6-3) Atl. Sounding Co. v. Townsend Y (5-4) Coeur Alaska, Inc. v. Y Y Se. Alaska Conservation Council (6-3) Riegel v. Medtronic, Inc. Y (8-1) Nken v. Holder Y Y Y (7-2) Kawasaki Kisen Kaisha Ltd. v. Y Y Regal-Beloit Corp. (6-3) Sykes v. United States Y (5-1-3) Carcieri v. Salazar Y Y Y (6-3) Graham Cty. Soil & Water Y Y Y Conservation Dist. v. United States ex rel. Wilson (7-2) Milner v. Dep't of Navy Y (8-1) Nat'l Ass'n of Home Builders v. Y Defs. of Wildlife (5-4) Winter v. Nat. Res. Def. Council Y (5-2-2) Powerex Corp. v. Y Reliant Energy Servs., Inc. (7-2) Tellabs, Inc. v. Y Makor Issues & Rights, Ltd. (6-2-1) Ali v. Fed. Bureau of Prisons Y Y Y (5-4) Begay v. United States Y Y Y (5-1-3) Howard Delivery Serv., Inc. v. Y Y Zurich Am. Ins. (6-3) Hall Street Assocs. v. Y Mattel, Inc. (6-3) Jerman v. Carlisle, McNellie, Y Y Y Y Rini. Kramer & Ulrich, L.P.A. (6-2-2) Conkright v. Frommert Y (5-3) CSX Transp., Inc. v. Y Y Ala. Dep't of Revenue (7-2) James v. United States Y (5-4) Gonzalez v. United States Y (7-1-1) Schindler Elevator Corp. v. Y United States ex rel. Kirk (5-3) Sossamon v. Texas Y (6-3) Perdue v. Kenny Y (5-4) Chamber of Commerce v. Brown (7-2) United States v. Ressam Y Y (6-2-1) Stoneridge Inv. Partners, LLC v. Y Scientific-Atlanta (5-3) AT&T Mobility LLC v. Y Y Y Concepcion (5-4) Entergy Corp. v. Y Riverkeeper, Inc. (5-1-3) Rockwell Int'l Corp. v. Y United States (6-2) Vaden v. Discover Bank Y Y (5-4) Negusie v. Holder Y Y Y (6-4-1) Cullen v. Pinholster (5-4) Winkelman v. Y Parma City Sch. Dist. (7-2) Chamber of Commerce v. Y Y Whiting (5-3) Dolan v. United States Y Y (5-4) Fowler v. United States Y (6-1-2) Dillon v. United States (7-1) Hamilton v. Lanning Y (8-1) Ransom v. Y FIA Card Servs., N.A. (8-1) Osborn v. Haley Y Y Y (7-2) Glob. Crossing Telecomm., Inc. Y Y v. Metrophones Telecomm., Inc. (7-2) Ministry of Def. & Support for Y Y the Armed Forces of the Islamic Republic of Iran v. Elahi (II) (6-3) Schwab v. Reilly Y (6-3) Holder v. Y Humanitarian Law Project (6-3) New Process Steel, L.P. v. NLRB Y Y (5-4) Barber v. Thomas Y Y (6-3) United States v. Y Tohono O'Odham Nation (5-2-1) Dean v. United States Y Y (7-2) Johnson v. United States (7-2) Boyle v. United States Y Y Y (7-2) United States v. Hayes Y (7-2) Janus Capital Grp., Inc. v. Y First Derivative Traders (5-4) Lawrence v. Florida Y (5-4) Corley v. United States Y (5-4) Cuomo v. Clearing House Ass'n Y Y (5-4) Bd. of Trs. of Leland Stanford Junior Univ. v. Roche Molecular Sys., Inc. (7-2) Bowles v. Russell (5~4) Anza v. Ideal Steel Supply Corp. (7-2) Maewood v. Patterson (5-4) Greenlaw v. United States (6-1-2) Watters v. Wachovia Bank, N.A. (5-3) Fed. Express Corp. v. Holowecki (7-2) United States v. Denedo (5-4) CSX Transp., Inc. v. McBride (5-4) Global-Tech Appliances, Inc. v. SEB S.A. (8-1) Rent-A-Center, W., Inc. v. Jackson (5-4) Hemi Grp. v. City of New York (5-3) 14 Penn Plaza LLC v. Pyett (5-4) CSX Transp., Inc. v. Hensley (7-2) Metro. Life Ins. v. Glenn (6-2-2) Preston v. Ferrer (8-1) John R. Sand & Gravel Co. v. United States (7-2) League of United Latin Am. Citizens v. Perry (5-4) Spears v. United States (5-3) CBOCS W., Inc. v. Humphries (7-2) Lance v. Dennis (8-1) Jefferson v. Upton (7-2) Holland v. Florida (6-1-2) Wood v. Allen (7-2) Roper v. Weaver (5-1-3) Panetti v. Quarterman (5-4) Canons 7 8 9 10 11 Case Rapanos v. United States Y Y (4-1-4) Hamdan v. Rumsfeld Y Y Y (5-4) AT&T Corp. v. Hulteen Y Y (7-2) Ledbetter v. Y Y Goodyear Tire & Rubber Co. (5-4) FCC v. Y Fox Television Stations, Inc. (5-4) Boumediene v. Bush (5-4) Massachusetts v. EPA (5-4) Altria Grp., Inc. v. Good Y (5-4) Bartlett v. Strickland Y Y (5-4) Harbison v. Bell Y (5-2-2) Zuni Public Sch. Dist. No. 89 v. Y U.S. Dep't of Educ. (5-4) Forest Grove Sch. Dist. v. T.A. Y (6-3) Riley v. Kennedy Y Y (7-2) Ricci v. DeStefano Y Y (5-4) Kasten v. Saint-Gobain Performance Plastics Corp. (6-2) Arlington Cent. Sch. Dist. Bd. of Educ. v. Murphy (5-1-3) Bloate v. United States Y Y (7-2) Bruesewitz v. Wyeth LLC (6-1-2) Wyeth v. Levine Y (5-2-3) PLIVA, Inc. v. Mensing Y (5-4) Marrama v. Citizens Bank of Mass. (5-4) Morgan Stanley Capital Grp., Y Y Inc. v. Publ. Util. Dist. No. 1 (5-2) Woodford v. Ngo Y (5-1-3) United States v. Santos Y Y (4-1-4) Gross v. FBL Fin. Servs., Inc. Y Y (5-4) Brown v. Plata Y (5-4) Gomez-Perez v. Potter Y Y (6-3) Lopez v. Gonzales (8-1) United States v. Rodriquez Y Y (6-3) Atl. Sounding Co. v. Townsend Y (5-4) Coeur Alaska, Inc. v. Se. Alaska Conservation Council (6-3) Riegel v. Medtronic, Inc. Y Y Y (8-1) Nken v. Holder Y (7-2) Kawasaki Kisen Kaisha Ltd. v. Y Regal-Beloit Corp. (6-3) Sykes v. United States Y (5-1-3) Carcieri v. Salazar (6-3) Graham Cty. Soil & Water Conservation Dist. v. United States ex rel. Wilson (7-2) Milner v. Dep't of Navy (8-1) Nat'l Ass'n of Home Builders v. Defs. of Wildlife (5-4) Winter v. Nat. Res. Def. Council (5-2-2) Powerex Corp. v. Reliant Energy Servs., Inc. (7-2) Tellabs, Inc. v. Y Makor Issues & Rights, Ltd. (6-2-1) Ali v. Fed. Bureau of Prisons Y (5-4) Begay v. United States (5-1-3) Howard Delivery Serv., Inc. v. Y Zurich Am. Ins. (6-3) Hall Street Assocs. v. Y Mattel, Inc. (6-3) Jerman v. Carlisle, McNellie, Y Y Rini. Kramer & Ulrich, L.P.A. (6-2-2) Conkright v. Frommert Y Y (5-3) CSX Transp., Inc. v. Y Ala. Dep't of Revenue (7-2) James v. United States (5-4) Gonzalez v. United States Y (7-1-1) Schindler Elevator Corp. v. United States ex rel. Kirk (5-3) Sossamon v. Texas Y (6-3) Perdue v. Kenny Y (5-4) Chamber of Commerce v. Brown Y (7-2) United States v. Ressam Y (6-2-1) Stoneridge Inv. Partners, LLC v. Y Scientific-Atlanta (5-3) AT&T Mobility LLC v. Y Y Concepcion (5-4) Entergy Corp. v. Riverkeeper, Inc. (5-1-3) Rockwell Int'l Corp. v. United States (6-2) Vaden v. Discover Bank Y Y (5-4) Negusie v. Holder (6-4-1) Cullen v. Pinholster Y Y (5-4) Winkelman v. Y Parma City Sch. Dist. (7-2) Chamber of Commerce v. Whiting (5-3) Dolan v. United States (5-4) Fowler v. United States Y (6-1-2) Dillon v. United States Y Y (7-1) Hamilton v. Lanning (8-1) Ransom v. FIA Card Servs., N.A. (8-1) Osborn v. Haley Y (7-2) Glob. Crossing Telecomm., Inc. Y v. Metrophones Telecomm., Inc. (7-2) Ministry of Def. & Support for the Armed Forces of the Islamic Republic of Iran v. Elahi (II) (6-3) Schwab v. Reilly (6-3) Holder v. Humanitarian Law Project (6-3) New Process Steel, L.P. v. NLRB (5-4) Barber v. Thomas (6-3) United States v. Y Tohono O'odham Nation (5-2-1) Dean v. United States Y Y (7-2) Johnson v. United States Y (7-2) Boyle v. United States Y (7-2) United States v. Hayes Y (7-2) Janus Capital Grp., Inc. v. Y First Derivative Traders (5-4) Lawrence v. Florida Y (5-4) Corley v. United States (5-4) Cuomo v. Clearing House Ass'n Y Y (5-4) Bd. of Trs. of Leland Stanford Y Junior Univ. v. Roche Molecular Sys., Inc. (7-2) Bowles v. Russell Y Y (5~4) Anza v. Y Y Ideal Steel Supply Corp. (7-2) Maewood v. Patterson Y Y (5-4) Greenlaw v. United States Y (6-1-2) Watters v. Wachovia Bank, N.A. Y (5-3) Fed. Express Corp. v. Holowecki Y (7-2) United States v. Denedo Y (5-4) CSX Transp., Inc. v. McBride Y (5-4) Global-Tech Appliances, Inc. v. Y SEB S.A. (8-1) Rent-A-Center, W., Inc. v. Y Jackson (5-4) Hemi Grp. v. City of New York Y (5-3) 14 Penn Plaza LLC v. Pyett Y (5-4) CSX Transp., Inc. v. Hensley Y (7-2) Metro. Life Ins. v. Glenn Y (6-2-2) Preston v. Ferrer Y (8-1) John R. Sand & Gravel Co. v. Y United States (7-2) League of United Latin Am. Y Citizens v. Perry (5-4) Spears v. United States Y (5-3) CBOCS W., Inc. v. Humphries Y (7-2) Lance v. Dennis Y (8-1) Jefferson v. Upton Y (7-2) Holland v. Florida Y (6-1-2) Wood v. Allen Y (7-2) Roper v. Weaver Y (5-1-3) Panetti v. Quarterman Y (5-4)
Anita S. Krishnakumar ([dagger])
([dagger]) Professor of Law, St. John's University School of Law. I owe deep thanks for valuable insights and conversations to John Q. Barrett, Aaron-Andrew Bruhl, Guy-Uriel Charles, Elaine Chiu, Marc DeGirolami, Vincent DiLorenzo, Abbe R. Gluck, Rebecca M. Kysar, Margaret H. Lemos, Yuxiang Liu, Janai Nelson, Margaret McGuinness, Mark Movsesian, Michael Perino, Rosemary Salomone, Kate Shaw, Deepa Varadarajan, David Zaring, and Adam Zimmerman. I am especially indebted to my husband, Ron Tucker, for his support and patience with this project. I am grateful to Dean Michael A. Simons and to St. John's University School of Law for generous research assistance; to participants at workshops and colloquia at Duke University School of Law, Seton Hall School of Law, and St. John's University School of Law for comments on earlier drafts of this article; and to the terrific editors at the Duke Law Journal. Christina Corcoran, Sade Forte, Ilya Mordukhaev, Jennifer Roseman, Samuel Sroka, Rita Wang, Kim Friedman, Lissa Yang, Peter Ryan, Vince Nibali, Christine Sammarco, and Thomas Combs provided excellent research assistance. All errors are my own.
(1.) The expressio unius canon is based on a Latin maxim, "expressio unius est exclusio alterius," which dictates that the express inclusion of one item implies the exclusion of other items not listed. See, e.g., William N. Eskridge, Jr., Philip Frickey & Elizabeth Garrett, Cases and Materials on Legislation: Statutes and The Creation of Public Policy 852-54 (4th ed. 2007).
(2.) William N. Eskridge, Jr. & Philip P. Frickey, Quasi-Constitutional Law: Clear Statement Rules as Constitutional Lawmaking, 45 VAND. L. REV. 593, 593 (1992).
(3.) Karl N. Llewellyn, Remarks on the Theory of Appellate Decision and the Rules or Canons About How Statutes Are to Be Construed, 3 VAND. L. REV. 395,401-06 (1950).
(4.) See, e.g., William N. Eskridge, Jr., Norms, Empiricism, And Canons In Statutory Interpretation, 66 U. CHI. L. REV. 671, 679-80 (1999); Daniel A. Farber, The Inevitability of Practical Reason: Statutes, Formalism, and the Rule of Law, 45 VAND. L. REV. 533, 547-48 (1992); John F. Manning, Continuity and Legislative Design, 79 NOTRE Dame L. Rev. 1863, 1863-64 (2004); John F. Manning, Legal Realism and the Canons' Revival, 5 GREEN BAG 2d 283, 283 (2001) ("Llewellyn largely persuaded two generations of academics that the canons of construction were not to be taken seriously."); Richard A. Posner, Statutory Interpretation--In the Classroom and in the Courtroom, 50 U. CHI. L. Rev. 800, 805-14 (1983); David L. Shapiro, Continuity and Change in Statutory Interpretation, 67 N.Y.U. L. REV. 921, 923 (1992) (noting that the canons have been "interred" by Llewellyn's essay); Cass R. Sunstein, Interpreting Statutes in the Regulatory State, 103 HARV. L. Rev. 405, 452 (1989). But see Geoffrey P. Miller, Pragmatics and the Maxims of Interpretation, 1990 Wis. L. Rev. 1179, 1190 ("Llewellyn's critique of the maxims appears overstated.").
(5.) See James J. Brudney & Corey Ditslear, Canons of Construction and the Elusive Quest for Neutral Reasoning, 58 Vand. L. Rev. 1, 4 (2005) [hereinafter Brudney & Ditslear, Canons of Construction]; David S. Law & David Zaring, Law Versus Ideology: The Supreme Court and the Use of Legislative History, 51 Wm. & Mary L. Rev. 1653,1654 (2010). Both of these studies measured judicial dueling by counting instances in which a majority opinion's reference to a particular interpretive resource (e.g., a language canon or legislative history) was countered by a dissenting opinion's reference to the same interpretive resource. That is, both of these studies defined "dueling" the same way that this Article does--rather than counting how often the majority opinion in a case invoked one interpretive tool (a "thrust") while the dissenting opinion countered with a different canon or tool (the "parry"), as suggested by Llewellyn's famous list.
(6.) Brudney & Ditslear, Canons of Construction, supra note 5, at 65 tbl.XII, 68 & n.222, 96-102.
(7.) Law & Zaring, supra note 5, at 1736-38,1739 tbl.9.
(8.) See, e.g., Eskridge & Frickey, supra note 2, at 595 & nn.2-3 (characterizing "Thrust But Parry" numbers 2-6, 8-10, and 13-14 as referential canons and numbers 11-12 and 15-28 as linguistic canons).
(9.) For a detailed explanation of this Article's methodology, see infra Part II.A.
(10.) See infra Table 1; discussion infra Part II.B.1.
(13.) See infra Table 3a; discussion infra Part II.B.2. Of the eighty-eight cases in the dataset that contained judicial dueling over one or more statutory interpretation-specific tools, 26.1 percent involved a criminal statute, 13.6 percent involved an antidiscrimination statute, and 6.8 percent involved an environmental statute.
(14.) See infra Part II.B.3; Tables 6a-6g, 8. In the subset of cases decided unanimously, by contrast, some of the interpretive tools consistently correlated with votes against individual Justices' ideology. See infra notes 127-36; Table 7.
(15.) "Other statutes" is an interpretive tool that directs interpreters to compare how courts in previous cases have interpreted identical or similar language in other statutes addressing issues similar to the statute at issue. See, e.g., Deborah A. Widiss, Undermining Congressional Overrides: The Hydra Problem in Statutory Interpretation, 90 TEX. L. REV. 859,871 (2012).
(16.) See William N. Eskridge, Jr., Textualism, the Unknown Ideal?, 96 MICH. L. Rev. 1509, 1548-49 (1998); John Paul Stevens, Judicial Predilections, 6 NEV. L.J. 1,1-2 (2005).
(17.) See infra Part III.A.4. This administrability versus policy-constancy divide is one that I describe in earlier empirical work regarding the Roberts Court. See Anita S. Krishnakumar, Statutory Interpretation in the Roberts Court's First Era: An Empirical and Doctrinal Analysis, 62 HASTINGS L.J. 221,244-45 (2010). Briefly, administrability concerns focus on an interpretation's effect on judicial resources, the difficulty of implementing the interpretation, or the clarity and predictability of the rule the interpretation creates. Policy-constancy concerns focus on whether an interpretation ensures consistent application of the statute over time, any arbitrariness created by the interpretation, or the justness of the interpretation.
(18.) See Victoria F. Nourse, Two Kinds of Plain Meaning, 76 Brook. L. Rev. 997, 997,1002 (2011); Lawrence M. Solan, The New Textualists' New Text, 38 LOY. L.A. L. REV. 2027,2027-28, 2060-61 (2005). The prototypical meaning of a statute focuses on the "core example" that the statute was designed to reach, whereas the legalist meaning looks broadly to the conceptual extension of the word at issue, often incorporating the specialized connotations and conventions the legal world has attached to the word. See Nourse, supra, at 1000-03. The differences between prototypical and legalist meaning are discussed in greater detail infra Part III.A.1.
(19.) Cf. Abbe R. Gluck, The States as Laboratories of Statutory Interpretation: Methodological Consensus and the New Modified Textualism, 119 Yale L.J. 1750, 1822-23 (2010) (describing state courts' adoption of a binding statutory interpretation methodology and suggesting that federal courts might benefit from adopting some form of methodological stare decisis as well).
(20.) Robert Weisberg, The Calabresian Judicial Artist: Statutes and the New Legal Process, 35 Stan. L. Rev. 213,213 (1983).
(21.) See Llewellyn, supra note 3, at 401.
(22.) See, e.g., Karl Llewellyn, The Bramble Bush 74 (1930); Karl Llewellyn, A Realist Jurisprudence--The Next Step, 30 COLUM. L. REV. 431, 443-44 (1930); Karl Llewellyn, Some Realism About Realism--Responding to Dean Pound, 44 HARV. L. REV. 1222,1241-42 (1931).
(23.) See Llewellyn, supra note 3, at 399 ("What we need to see now is that all of this is paralleled, in regard to statutes.").
(25.) See id. at 404 (describing "Thrust But Parry" number 17, which says "The same language used repeatedly in the same connection is presumed to bear the same meaning throughout the statute" versus "This presumption will be disregarded where it is necessary to assign different meanings to make the statute consistent").
(26.) Id. at 400.
(27.) See, e.g., Antonin Scalia, A Matter of Interpretation: Federal Courts and THE LAW 25-27 (1997); sources cited infra note 186.
(28.) See Brudney & Ditslear, Canons of Construction, supra note 5, at 15-69. The Brudney-Ditslear study coded for judicial reliance on two categories of interpretive tools: language canons, which the study defined to include grammar rules, Latin maxims, the whole act rule, and the in pari materia rule, and substantive canons, defined as judicial presumptions based on constitutional and common-law norms about how statutes should be interpreted. Id. at 12-13.
(29.) See CASS SUNSTEIN, AFTER THE RIGHTS REVOLUTION: RECONCEIVING THE REGULATORY State 147, 151-53 (1990); William N. Eskridge, Jr. & Philip P. Frickey, Foreword: Law as Equilibrium, 108 HARV. L. REV. 26, 66-67 (1994) ("One goal ... of the canons, is to lower the costs of drafting statutes.... The Court can perform a valuable coordinating function by generating 'off-the-rack,' gap-filling rules that are accessible ex ante to the drafters."); Shapiro, supra note 4, at 943.
(30.) Brudney & Ditslear, Canons of Construction, supra note 5, at 57-60.
(31.) Id. at 98.
(32.) See James J. Brudney & Corey Ditslear, Liberal Justices' Reliance on Legislative History: Principle, Strategy, and the Scalia Effect, 29 BERKELEY J. EMP. & Lab. L. 117,143-44 & tbl.6 (2008) [hereinafter Brudney & Ditslear, Legislative History],
(33.) Id. at 144 tbl.6.
(34.) Id. at 142-44; Brudney & Ditslear, Canons of Construction, supra note 5, at 6.
(35.) Brudney & Ditslear, Legislative History, supra note 32, at 160-69.
(36.) Id. at 169-70.
(37.) Brudney & Ditslear, Canons of Construction, supra note 5, at 26-27.
(38.) See Jane S. Schacter, The Confounding Common Law Originalism in Recent Supreme Court Statutory Interpretation: Implications for the Legislative History Debate and Beyond, 51 Stan. L. Rev. 1, 5, 12, 18, 21 (1998); Nicholas S. Zeppos, The Use of Authority in Statutory Interpretation: An Empirical Analysis, 70 TEX. L. REV. 1073, 1107-08 (1992); see also Krishnakumar, supra note 17, at 236-37 & tbl.1 (reporting that practical consequences were the third most frequently cited interpretive resource cited by the Roberts Court during its first three-and-a-half terms, showing up in 51.8 percent of all statutory cases decided during that time period).
(39.) Law & Zaring, supra note 5.
(40.) Additional variables for which Law and Zaring coded were the complexity and obscurity of the statute, as well as the number of times the statute was amended. See id. at 168998.
(41.) See id. at 1738.
(44.) This is the date that Justice Alito joined the Court. Biographies of Current Justices of the Supreme Court, SUPREME COURT OF THE U.S., http://www.supremecourt.gov/about/ biographies.aspx [http://perma.cc/8HLT-QVKN].
(45.) See Opinions, SUPREME COURT OF THE U.S., http://www.supremecourt.gov/opinions/ opinions.aspx [http://perma.cc/P5SJ-ENZP]. The Court's online database currently only goes back to the 2009 Term, id., but when I began coding cases in 2007, it went back to the 2005 Term.
(46.) I made this judgment call because the Federal Rules of Civil Procedure (FRCP) are created in a manner that differs significantly from federal statutes--they are drafted by judges rather than Congress and do not require the President's approval. Accordingly, several of the interpretive tools available when construing statutes either are not available with respect to the FRCP or provide a very different kind of context, from a very different perspective, when used to construe the FRCP--e.g., legislative history, intent, the whole act rule, and other statutes.
(47.) In such cases, the opinion was coded as unanimous, close margin, or wide margin based on the Justices' votes regarding the statutory interpretation question only; thus, if the Justices agreed unanimously that the statute should be read to mean X, but then split regarding the constitutional validity of the statute, the opinion was coded as unanimous. See, e.g., Nw. Austin Mun. Util. Dist. No. 1 v. Holder, 557 U.S. 193, 211 (2009).
(48.) For a list of the cases examined in the study, see infra Appendix.
(49.) In order to reduce the risk of inconsistency, I and at least one research assistant separately read and analyzed each opinion and separately recorded the use of each interpretive resource. In the event of a disagreement, I reviewed and reconsidered the case and made the final determination as to how a particular interpretive resource should be coded. For a detailed explanation of how the cases were coded, see Krishnakumar, supra note 17, Codebook at 291-96.
(50.) See, e.g., Schacter, supra note 38, at 11-12; Zeppos, supra note 38, at 1089; see generally Frank B. Cross, The Theory and Practice of Statutory Interpretation (2009) (surveying the tools of statutory interpretation and presenting empirical findings on the Rehnquist Court's use of interpretive tools).
(51.) The term "grammar canons" refers to interpretive maxims that are based on basic conventions of grammar and syntax, such as "'[o]r' means in the alternative" or "[s]hall' is mandatory, while 'may' is precatory." See Eskridge et al., supra note 1, at 856-60. The term "language canons" refers to canons that are based on Latin maxims, such as "expressio unius est exclusio alterius" (the inclusion of one item implies the exclusion of others not listed) or "noscitur a sociis" (terms in a statutory list should be interpreted in light of the other items in the list). See id. at 852-54. The whole act rule refers to a series of inferences that courts may draw about the meaning of one section of a statute based on how other sections of the statute are structured; it includes the rule that differences in similar or parallel statutory provisions should be deemed deliberate and intentional (the meaningful variation rule), rules dictating that the title and preamble of a statute are relevant but not dispositive in determining statutory meaning, and the rule that one section of a statute should not be construed in a manner that renders another section superfluous (the rule against superfluity). See id. at 862-65.
(52.) See infra Part III.A.4.
(53.) See id.-, see also Krishnakumar, supra note 17, Codebook at 294 (explaining parameters for coding for references to administrability-type practical consequences versus policy-constancy-type practical consequences).
(54.) An example may help illustrate. In Barber v. Thomas, 560 U.S. 474 (2010), the majority opinion relied on the plain meaning rule, statutory purpose, and practical consequences to conclude that the phrase "term of imprisonment" in a statute allowing good time credit for good behavior by prisoners applies to the time actually served by the prisoner, rather than the time the prisoner was sentenced to serve. Id. at 479-84. In so ruling, the majority rejected a legislative history argument advanced by petitioner, finding that the cited portions of the legislative record did not address the precise interpretive question at issue. See id. at 485. It also rejected rule of lenity and whole act rule arguments relied on by petitioner and the dissent, stating that the rule of lenity was inapplicable because the statute was not ambiguous and that interpreting "term of imprisonment" consistently throughout the statute would contradict the statute's text (a texttrumps-whole-act-rule argument). Id. at 486-89. The opinion was coded for reliance on text/plain meaning, purpose, and practical consequences, but not legislative history, substantive canons, or the whole act rule.
(55.) For example, in Schwab v. Reilly, 560 U.S. 770 (2010), the Court held that when a bankruptcy debtor underestimates the value of "property claimed as exempt" on her Schedule C filing, the bankruptcy trustee is entitled to retain in the estate any actual value above that listed, even if he did not initially object to the "property claimed as exempt." Id. at 774. The majority opinion relied primarily on the "clear" meaning of the Bankruptcy Code's text, combined with a whole act argument. Id. at 782. The opinion also noted that this interpretation was consistent with the historical treatment of bankruptcy exemptions (common-law precedent), disagreed with the majority's reading of Supreme Court precedent, and cited other Supreme Court precedent supporting its interpretation. Id. at 786-91. The opinion was coded for references to text/plain meaning, the whole act rule, common law, and Supreme Court precedent.
(56.) Cases decided by a vote margin of 5-41, 5-3 (with only eight justices participating), 5-2 (with only seven justices participating), and 4-1-4 (by a plurality) were coded as "close margin" cases. If the vote margin was 6-3, either with six justices in the majority or with only five justices in the majority and one concurring and three dissenting, the case was coded as a wide margin case. A quick perusal of the full dataset (not just dueling canon cases) reveals only six cases like this (6-3 with five in the majority plus one concurring).
(57.) See Krishnakumar, supra note 17, at 231-33.
(58.) For example, in Dolan v. United States, 560 U.S. 605 (2010), the Court held that a district court retained the power to order restitution to crime victims--even though the court missed the Mandatory Victims Restitution Act's 90-day deadline for establishing the amount of the victim's losses--because the court had made clear, prior to the deadline, that it would order restitution in some amount. Id. at 608. The majority opinion referenced several interpretive tools, including the statute's purpose of helping victims by imposing restitution on convicts. Id. at 613-14. The dissenting opinion argued that the statute's text did not permit such a reading, particularly given other criminal statutes that "only make sense against a background rule that trial courts cannot change sentences at will." Id. at 625-28 (Roberts, C.J., dissenting). The dissent rejected the majority's purpose argument, reasoning that the statute's broad purpose did not justify a reading that contradicts the statutory text. See id. at 625. The dissenting opinion was coded for references to text/plain meaning and other statutes, but not statutory purpose.
(59.) See SCALIA, supra note 27, at 27 (arguing that Llewellyn's "parries" do not contradict the corresponding canon but merely show that it is not absolute).
(60.) In this sense, my definition of "dueling" interpretive tool use departs from Llewellyn's, as Llewellyn's thrusts and parries certainly include opposing pairs of this kind.
(61.) See, e.g., SCALIA, supra note 27, at 27 ("Every canon is simply one indication of meaning; and if there are more contrary indications (perhaps supported by other canons), it must yield. But that does not render the entire enterprise a fraud ....").
(62.) See infra Table 1.
(63.) Table 1 uses the number of cases in which at least one opinion relied on an interpretive tool, rather than the total number of cases reviewed in the study (255) as a denominator because
(64.) See supra Table 1.
(65.) I counted as substantive canon "duels" those cases in which a majority opinion cited one substantive canon (e.g., the rule of lenity) and an opposing opinion cited another substantive canon (e.g., the constitutional avoidance canon). I did not count as "dueling" cases in which a majority opinion invoked the rule of lenity and the dissenting opinion argued that the rule was not applicable. See supra Part II. A.
(66.) Specifically, for statutes deemed in pari materia (in the same matter), "courts will apply prior judicial interpretations" of a particular term both "to subsequent cases that arise under the statute actually interpreted [and] also to identical or similar language in other statutes addressing similar issues." See Widiss, supra note 15, at 871.
(67.) James J. Brudney & Lawrence Baum, Oasis or Mirage: The Supreme Court's Thirst for Dictionaries in the Rehnquist and Roberts Eras, 55 WM. & MARY L. Rev. 483, 537 (2013).
(68.) Id. at 548.
(69.) This finding is consistent with the 28.8 percent rate of dueling dictionary use found in Brudney and Baum's study of employment law, business and commercial law, and criminal-law cases from 1986 to 2010. See id. at 526.
(70.) My earlier empirical study of the Roberts Court and Brudney & Baum's recent study of the Rehnquist and Roberts Courts found that the Justices cited the following five dictionaries most frequently: Webster's Third New International Dictionary, the Oxford English Dictionary, Webster's Second New International Dictionary, the American Heritage Dictionary, and Black's Law Dictionary. See Brudney & Baum, supra note 67, at 529; Krishnakumar, supra note 17, at 240 n.85.
(71.) Taniguchi v. Kan Pac. Saipan, Ltd., 132 S. Ct. 1997 (2012).
(72.) See id. at 2002-03, 2003 n.2.
(73.) See Brudney & Baum, supra note 67, at 530-31; Krishnakumar, supra note 17, at 240 n.85.
(74.) Briefs were not available online for the other two cases.
(75.) See, e.g., Brief for the State of Cal. at 44-46, Burlington N. & Santa Fe Ry. Co. v. United States, 556 U.S. 599 (2009) (Nos. 07-1601 & 07-1607); Brief for the United States as Amicus Curiae Supporting Respondents at 7,15, Fed. Express Corp. v. Holowecki, 552 U.S. 389 (2008) (No. 06-1322); Brief for the Petitioners at 16-17, Zuni Pub. Sch. Dist. No. 89 v. U.S. Dep't of Educ., 550 U.S. 81 (2007) (No. 05-1508).
(76.) See infra Part III.B.1.
(77.) See, e.g., Scalia, supra note 27, at 36 ("Legislative history provides, moreover, a uniquely broad playing field. In any major piece of legislation, the legislative history is extensive, and there is something for everybody."); In re Sinclair, 870 F.2d 1340, 1343 (7th Cir. 1989) (Easterbrook, J.) ("Often there is so much legislative history that a court can manipulate the meaning of a law by choosing which snippets to emphasize....").
(78.) See, e.g., WILLIAM N. ESKRIDGE, JR., DYNAMIC STATUTORY INTERPRETATION 276 (1994); Brudney & Ditslear, Canons of Construction, supra note 5, at 13.
(79.) Llewellyn, supra note 3, at 401.
(80.) See supra Table 1.
(81.) One-hundred and thirteen of the cases involved some kind of dueling, but in nineteen of these the Court dueled only over the application of its own precedent, and in another six it dueled only over the application of practical consequences reasoning (three) or over precedent and practical reasoning (three). In my view, cases in which the Court dueled only over precedent and/or practical consequences should not count as "dueling canon" cases because precedent and practical consequences are general tools of legal analysis, not canons or tools specific to statutory construction. If we count these precedent-only or practical-only cases, then 44.3 percent of the cases involved "dueling."
(82.) See infra Appendix (listing each "dueling canon" case and all tools over which majority and dissenting opinions dueled by case). If we include cases that involve dueling only over precedent or practical consequences, then thirty-seven of one-hundred and thirteen or 32.7 percent of the dueling cases involved dueling over only one interpretive tool.
(83.) See Brudney & Ditslear, Canons of Construction, supra note 5, at 96; Law & Zaring, supra note 5, at 1738.
(84.) This is also why I believe that Table 1--which measures not only the extent to which dissenting opinions respond to a majority or concurring opinion's reliance on a particular interpretive tool, but also the extent to which majority or concurring opinions respond to a dissenting opinion's reliance on the tool--provides a better gauge of judicial dueling than does either the Brudney-Ditslear or the Law-Zaring study.
(85.) See infra Table 3a.
(86.) See id.
(87.) Of the 255 statutory cases decided during the period studied, 32.9 percent involved criminal (21.1 percent) or antidiscrimination (11.8 percent) statutes. If we add in environmental law cases, the percentage of the Court's docket that involved these types of statutes was 37.3 percent. So the proportion of dueling cases accounted for by these statutory subject areas (46.5 percent) was a bit higher than the proportion of overall statutory cases that involved these subject areas.
(88.) See, e.g., Jones v. United States, 526 U.S. 227, 234 (1999) (referencing a robbery statute that served as a model for the statute at issue).
(89.) See, e.g., Lorillard v. Pons, 434 U.S. 575, 578-79 (1978) (noting that the ADEA was modeled on both the FLSA and Title VII); H.R.
Rep. No. 102-40, pt. 2, at 4 (1991), as reprinted in 1991 U.S.C.C.A.N. 694, 696-97 ("A number of other laws banning discrimination, including [the ADEA] are modeled after, and have been interpreted in a manner consistent with, Title VII.").
(90.) See, e.g., Jones, 526 U.S. at 234-36, 259-61 (discussing which of three model robbery statutes was most analogous to statute at issue). Compare Kasten v. Saint-Gobain Performance Plastics Corp., 131 S. Ct. 1325 (2011) (citing numerous state statutes and federal regulations that permit complaints to be filed orally), with Brief in Opposition at 6-8, Kasten v. Saint-Gobain Performance Plastics Corp., 131 S. Ct. 1325 (2011) (No. 09-834) (citing numerous different analogous statutes to argue that complaints must be filed in writing).
(91.) See infra Table 3b.
(92.) See infra Table 3b.
(93.) That is, government v. criminal defendant, employer v. employee, disability claimant v. state, minority voters v. local authorities, or environmental interests v. private-property interests.
(94.) Cf. Law & Zaring, supra note 5, at 1740 (concluding that the level of ideological alignment between authoring justice and enacting Congress is a statistically significant predictor of probability that a statutory interpretation opinion will reference legislative history).
(95.) Babbitt v. Sweet Home Chapter of Cmtys. for a Great Or., 515 U.S. 687 (1995).
(96.) See id. at 690; 16 U.S.C. [section] 1538(a)(1)(B) (2012).
(97.) 16 U.S.C. [section] 1532(19).
(98.) 50 C.F.R. [section] 17.3 (1994).
(99.) Babbitt, 515 U.S. at 692.
(100.) For example, Justice Scalia's dissenting opinion argued that the plain meaning of "take" encompasses only deliberate, intentional action directed toward individual animals, not incidental injuries to habitat. Justice Stevens' majority opinion countered that the ordinary meaning of "harm" in the definition of "take" covers incidental injury of the kind caused by habitat destruction. Compare Babbitt, 515 U.S. at 687, 697, 709 (stating that the dictionary definition of "harm" includes "injure" and habitat modification which kills individual animals of an endangered species causes "injury"), with id. at 717-18 (Scalia, J., dissenting) ("It is obvious that 'take' in this sense ... describes a class of acts (not omissions) done directly and intentionally....").
(101.) See id. at 714 (Scalia, J., dissenting) ("The Court's holding that the hunting and killing prohibition incidentally preserves habitat on private lands imposes unfairness to the point of financial ruin--not just upon the rich, but upon the simplest farmer who finds his land conscripted to national zoological use.").
(102.) See id. at 698 (majority opinion) ("[T]he broad purpose of the ESA supports the Secretary's decision to extend protection against activities that cause the precise harms Congress enacted the statute to avoid.").
(103.) Rapanos v. United States, 547 U.S. 715 (2006).
(104.) Id. at 721.
(105.) Id. at 787, 798-99 (Stevens, J., dissenting).
(106.) United Steelworkers v. Weber, 443 U.S. 193 (1979).
(107.) Id. at 197.
(108.) Id. at 206,220,230.
(109.) See id. at 207,207 n.7,229, 230 n.ll, 232,232 n.12 (Rehnquist, J., dissenting).
(110.) Id. at 202-04 (majority opinion).
(111.) See id. at 254 (Rehnquist, J., dissenting) ("There is perhaps no device more destructive to the notion of equality than the numerus clausus--the quota."). A recent Roberts Court case, Ricci v. DeStefano, 557 U.S. 557 (2009), involved a similar ideological clash over Title VII. See id. at 563 (discussing the legality under Title VII of a municipality's race-based promotion practices). The majority and dissenting opinions in Ricci dueled over the whole act rule, statutory purpose, and practical consequences, see id. at 583-84, 579-80, but again, the underlying division was over the fairness of abandoning the results of a firefighter promotion examination on which white firefighters significantly outperformed black firefighters, compare id. at 593 ("Many of the candidates had studied for months, at considerable personal and financial expense, and thus the injury caused by the City's reliance on raw racial statistics at the end of the process was all the more severe."), with id. at 608-11 (Ginsburg, J., dissenting) (noting that though "[t]he white firefighters who scored high on New Haven's promotional exams understandably attract this Court's sympathy ... they had no vested right to promotion" and that "[f]irefighting is a profession in which the legacy of racial discrimination casts an especially long shadow").
(112.) See, e.g., Ricci, 557 U.S. at 583-84, 579-80, 608-11; Bartlett v. Strickland, 556 U.S. 1, 6, 14-16, 27-29 (2009); AT&T Corp. v. Hulteen, 556 U.S. 701, 704, 709 n.3, 720-721 (2008); Ledbetter v. Goodyear Tire & Rubber Co., 550 U.S. 618,621,628,646-47 (2006).
(113.) Dean v. United States, 556 U.S. 568 (2009).
(114.) See id. at 570.
(115.) Compare id. at 572-73, 578, with id. at 575, 579 (Stevens, J., dissenting) (disagreeing over comparisons to common-law crimes).
(116.) Compare id. at 576 (defendant was "guilty of unlawful conduct twice over"), with id. at 585 (Breyer, J., dissenting) ("Congress would not have intended to punish so harshly.").
(117.) Compare, e.g., Lawrence v. Florida, 549 U.S. 327,336 (2007) (worrying about providing "incentives for state prisoners to file certiorari petitions as a delay tactic"), with id. at 345 (Ginsburg, J., dissenting) (emphasizing that the majority's reading "deprive[s] unwitting litigants of the opportunity to pursue their constitutional claims").
(118.) Agencies are not all created equal. Congress grants some agencies expansive powers and authority, while limiting the authority granted to others. The EPA is widely considered a powerful agency because Congress has granted it broad powers under several statutes, including the Clean Water Act, the Clean Air Act, and the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA). See, e.g., Steel Co. v. Citizens for a Better Env't, 523 U.S. 83, 87 (1998) ("The [EPA] has the most powerful enforcement arsenal: it may seek criminal, civil, or administrative penalties."); Stephen Breyer, The Executive Branch, Administrative Action, and Comparative Expertise, 32 CARDOZO L. Rev. 2189, 2191 ("Government again expanded the scope of regulation significantly in the 1970s with the creation of powerful but not necessarily organizationally 'independent' regulatory authorities, such as the Environmental Protection Agency, the Occupational Safety and Health Administration, and the National Highway Traffic Safety Administration."); Louis Jaffe, The Disorder of Politics: Lowi Takes the High Road (Book Review), 24 STAN. L. REV. 587, 590 (1972) (book review) ("[EPA] by a continuing series of statutes has been given very powerful anti-pollution tools allowing it to formulate and enforce air and water quality standards.").
(119.) The Equal Employment Opportunity Commission is a notoriously weak agency, to which Congress has delegated limited power, and traditionally receives little deference from courts. See, e.g., Trans World Airlines, Inc. v. Hardison, 432 U.S. 63, 76 n.11 (1977); Gen. Elec. Co. v. Gilbert, 429 U.S. 125,141 (1976); Skidmore v. Swift & Co., 323 U.S. 134, 139-40 (1944).
(120.) See supra Table 3a.
(121.) Legacy Datasets: 2014 Release 01, The SUPREME COURT DATABASE, http:// supremecourtdatabase.org/data.php?s=2 [http://perma.cc/RG9G-HM78].
(122.) See infra Table 4.
(123.) Figures are not reported for dueling over substantive canons or the common-law precedent interpretive tool because there were only 5-6 cases involving dueling over each of these interpretive tools.
(124.) In the remaining cases, Justice Scalia failed to join an opinion that referenced legislative history.
(125.) This calculation does not include Justice Souter, who retired from the Court in 2009, and did not participate in cases decided during two out of the five terms studied in the dataset (including nearly half of the dueling legislative history cases). It also does not include Justices Sotomayor or Kagan, who joined the Court in the last two years of the period studied, and did not participate in the overwhelming majority of the cases in the dataset.
(126.) See supra Tables 6b-6g. The one noteworthy exception was the conservative Justices' dueling uses of dictionary definitions; Justices Scalia, Roberts, and Kennedy employed this interpretive tool very even-handedly, showing almost no ideological slanting.
(127.) Brudney & Ditslear, Cannons of Construction, supra note 5, at 57-60.
(128.) Brudney & Ditslear, Legislative History, supra note 32, at 144 tbl.6.
(130.) Id. at 142-44; Brudney & Ditslear, Cannons of Construction, supra note 5, at 6.
(131.) Nine of twenty-one liberal dictionary referencing unanimous opinions were authored by conservative Justices; four of the eight conservative ones were--so not just explained by liberal Justices authoring the unanimous opinions that used dictionary to reach liberal results.
(132.) Seven of the seventeen liberal legislative history unanimous opinions were authored by conservative Justices; two of the nine conservative ones were--so not just explained by liberal Justices authoring the unanimous opinions that used legislative history to reach liberal results.
(133.) As noted earlier, Justice Kennedy's rate of invoking legislative history to support conservative outcomes was lower, at 42.1 percent. See supra Table 6a.
(134.) These figures do not include Justices Souter, Sotomayor, or Kagan, who did not participate in a substantial number of the cases, and whose rates of reference therefore are not comparable to those of the other Justices.
(135.) Eskridge, supra note 4, at 679.
(136.) See supra Table 1.
(137.) Llewellyn, supra note 3, at 399.
(138.) Holder v. Humanitarian Law Project, 561 U.S. 1 (2010).
(139.) 18 U.S.C. [section] 2339B(a)(l) (2012).
(140.) Humanitarian Law Project, 561 U.S. at 9.
(141.) Id. at 16.
(142.) Id. at 16-17.
(143.) See id. at 57-58 (Breyer, J., dissenting). Additional examples from outside the dataset include Babbitt v. Sweet Home Chapter of Communities for a Great Oregon, 515 U.S. 687, 691-95, 727-30 (1995) (focusing on the statutory term "harm," while the dissent focuses on the term "taking"), and Muscarello v. United States, 524 U.S. 125, 128-132, 143-48 (1998) (focusing on the plain meaning of statutory term "carried," while the dissent focuses on the phrase "carried a firearm").
(144.) AT&T Corp. v. Hulteen, 556 U.S. 701 (2009).
(145.) Hulteen concerned whether an employer violated the Pregnancy Discrimination Act (PDA) by calculating pension benefits using an accrual rule that gave less retirement credit for pregnancy leave than for other kinds of medical leave. Id. at 704. The PDA was enacted in 1978 to override a 1976 Supreme Court decision, General Electric Co. v. Gilbert, which held that discrimination on the basis of pregnancy is not sex discrimination. Gilbert, 429 U.S. 125 (1976), superseded by statute, Pregnancy Discrimination Act of 1978, 92 Stat. 2076,42 U.S.C. [section] 2000e(k) (1982). The Hulteen majority concluded that the employer's pension policy did not violate the PDA, relying on a Title VII provision that defines "unlawful employment practice" with respect to seniority systems. Hulteen, 556 U.S. at 707. The majority held that a seniority system violates Title VII only if it was "adopted for an intentionally discriminatory purpose" and found that there could be no "intentional" discrimination in a policy adopted before the PDA was enacted, while Gilbert supplied the governing legal rule. Id. at 711-14. Justices Ginsburg and Breyer dissented, arguing that the PDA's plain text unambiguously expressed Congress's disapproval of both the holding and reasoning of Gilbert--and made clear that pregnancy-based discrimination always has constituted sex discrimination. Id. at 720-21 (Ginsburg, J., dissenting).
(146.) There were forty-three cases in the dataset in which the Court dueled over plain meaning; thirty-four involved majority and dissenting opinions that ascribed different meaning to the same statutory word or phrase.
(147.) Of the thirty-four cases that dueled over the meaning of the same statutory term, fourteen (41.2 percent) pitted prototypical meaning against legalist meaning.
(148.) Solan, supra note 18, at 2061.
(149.) Nourse, supra note 18, at 1000.
(150.) See id.; see also Victoria F. Nourse, A Decision Theory of Statutory Interpretation: Legislative History by the Rules, 122 Yale L.J. 70, 124-25 (2012) (discussing prototypical vs. legalist meaning in the context of Professors Vermeule and Chomsky's debate over legislative history in Church of the Holy Trinity v. United States, 143 U.S. 457 (1892)).
(151.) Nourse, supra note 18, at 1001-04; Solan, supra note 18, at 2060-62.
(152.) Ali v. Fed. Bureau of Prisons, 552 U.S. 214 (2008).
(153.) 28 U.S.C. [section] 1346 (2012).
(154.) Id. [section] 2680(c).
(155.) Ali, 552 U.S. at 216.
(156.) Id. at 218-21.
(157.) Id. at 218-19 (emphasis omitted) (alterations in original) (quoting United States v. Gonzales, 520 U.S. 1,5 (1997)).
(158.) Mat 219.
(159.) Id. at 220-21.
(160.) Id. at 230 (Kennedy, J., dissenting).
(161.) See id. at 235-36 ("[T]he majority of the nine federal statutes other than [section] 2680(c) containing a reference to the detention of goods, merchandise, or other property are specific to customs and excise.").
(162.) Id. at 233-35.
(163.) There were fourteen dueling plain meaning rule cases that divided over prototypical-versus-legalist meaning; in six of these, Justice Scalia voted to adopt the prototypical, core meaning rather than the legalist meaning. See, e.g., Janus Capital Grp., Inc. v. First Derivative Traders, 131 S. Ct. 2296, 2304 (2011) (Thomas, J.) (finding that the prototypical conduct covered by SEC's Rule 10b-5 prohibition against false statements in prospectus is statements made by entities who have final authority over statements); United States v. Tohono O'Odham Nation, 131 S. Ct. 1723, 1728 (2011) (Kennedy, J.) (finding that the core conduct covered by statute's jurisdictional bar is Civil War-era cotton claimants who sued the United States in the Court of Claims, while concurrently suing federal officials in other courts under state tort law); Kawasaki Kisen Kaisha Ltd. v. Regal-Beloit Corp., 561 U.S. 89, 102-03, 111 (2010) (Kennedy, J.) (finding that the core meaning of the Carmack Amendment subjects only "receiving rail carriers]" and "delivering rail carrier[s]" to liability and, therefore, property received overseas and later transported to inland location by rail carriers was not covered); Hamilton v. Lanning, 560 U.S. 505, 524 (2010) (Scalia, J., dissenting) (finding that the core method of projecting debtor's income for Bankruptcy Code purposes is to follow Code's formula defining "current monthly income" as "the average monthly income from all sources that the debtor receives" during the past six months); Cuomo v. Clearing House Ass'n, 557 U.S. 519, 525 (2009) (Scalia, J.) (finding that the core meaning of National Bank Act is that "visitorial powers" protect national banks against supervisory powers and actions by states, not against all ordinary enforcement by states); Begay v. United States, 553 U.S. 137, 142 (2008) (Scalia, J., concurring) (finding that a DUI offense is not a "violent felony" because the residual clause of the Armed Career Criminal Act covers only offenses similar to the core offenses enumerated in the statute).
(164.) There were fourteen dueling plain meaning rule cases that divided over prototypical-versus-legalist meaning; in six of these, Justice Ginsburg voted to adopt the legalist meaning rather than the prototypical meaning. See Janus Capital Grp., 131 S. Ct. at 2307 (Breyer, J., dissenting) (reasoning that the "English language" does not impose boundaries on the word 'make' and relying on examples external to the securities context); Tohono O'Odham, 131 S. Ct. at 1739 (Ginsburg, J., dissenting) (identifying legalist focus on the general, detached meaning of the terms "claim" and "cause of action," and citing generalist treatise to conclude that entitlement to relief is an essential component of both terms); Kawasaki, 561 U.S. at 114-15 (Sotomayor, J., dissenting) (employing a legalist reading of the Carmack Amendment as providing liability regime for all rail carriage of property within United States, and citing generalist maritime treatises); Hamilton, 130 S. Ct. 2464, 2472 (Alito, J.) (referencing broad, general meaning that "projections" are based on likely future events, not an assumption that the past will repeat itself); Boyle v. United States, 556 U.S. 938, 944 (2009) (Alito, J.) (relying on a legalist reading of the term "any" to construe the RICO term "enterprise" to cover any group of individuals associated in fact); Ali, 552 U.S. at 220 (Thomas, J.) (voting with the majority and finding that the statutory mention of "any" law enforcement covers all law enforcement).
(165.) Justices Roberts, Kennedy, and Alito voted to adopt the prototypical meaning over the legalist meaning in five to six cases each, while Justice Thomas did so in only four of fourteen cases.
(166.) Carcieri v. Salazar, 555 U.S. 379 (2009).
(167.) Indian Reorganization Act, Pub. L. No. 73-383, 48 Stat. 985 (1934) (codified at 25 U.S.C. [section] 465 (2012)).
(168.) 25 U.S.C. [section] 479 (2012).
(169.) Carcieri, 555 U.S. at 384.
(170.) Id. at 382.
(172.) Id. at 389-90 (construing 25 U.S.C. [section] 472).
(173.) Id. at 409-10 (Stevens, J., dissenting).
(175.) Begay v. United States, 553 U.S. 137 (2008).
(176.) Id. at 147-48; see 18 U.S.C. [section] 924(e)(2)(B)(ii) (2012) (defining violent felony to mean "any crime punishable by imprisonment for a term exceeding one year" that "(i) has as an element the use, attempted use, or threatened use of physical force against the person of another or (ii) is burglary, arson, or extortion, involves use of explosives, or otherwise involves conduct that presents a serious potential risk of physical injury to another").
(177.) Begay, 553 U.S. at 160 (Alito, J., dissenting) (discussing 18 U.S.C. [section] 924(e)(2)(A)).
(178.) Id. at 159-60.
(179.) Dean v. United States, 556 U.S. 568 (2009).
(180.) See id. at 570 (discussing 18 U.S.C. [section] 924(c)(1)(A)).
(181.) Id. at 572-73 (emphasis omitted) (discussing 18 U.S.C. [section] 924(c)(4)).
(182.) Id. at 573.
(183.) Id. at 578-79 (Stevens, J., dissenting).
(184.) Id. at 579.
(185.) Freeman v. Quicken Loans, Inc., 132 S. Ct. 2034, 2044 (2012) (quoting Rodriguez v. United States, 480 U.S. 522, 525-26 (1987) (per curiam)).
(186.) See, e.g.. Barnhart v. Peabody Coal Co., 537 U.S. 149, 183-84 (2003) (Scalia, J., dissenting) (noting that " [i]t is naive for the Court to rely on guesses as to what Congress would have wanted in legislation as complicated as this, the culmination of a long, drawn-out legislative battle in which" interested parties attempted to pull the provisions in different directions); Babbitt v. Sweet Home Chapter of Cmtys. for a Great Or., 515 U.S. 687, 726 (1995) (Scalia, J., dissenting) ("Deduction from the 'broad purpose' of a statute begs the question if it is used to decide by what means (and hence to what length) Congress pursued that purpose.");
(187.) See infra Appendix for dueling purpose cases. The five cases in which the Court dueled without invoking different statutory purposes were: AT&T Mobility LLC v. Concepcion, 131 S. Ct. 1740 (2011); Jerman v. Carlisle, 559 U.S. 573 (2010); Altria Grp., Inc. v. Good, 555 U.S. 70 (2008); Hall Street Assocs. v. Mattel, Inc., 552 U.S. 576 (2008); Howard Delivery Serv., Inc. v. Zurich Am. Ins., 547 U.S. 651 (2006).
(188.) Conkright v. Frommert, 559 U.S. 506 (2010).
(189.) Id. at 509-11.
(190.) Id. at 517 (citations omitted). The majority noted several ways in which deference would benefit employers. Id.
(191.) Id. at 535-36 (Breyer, J., dissenting) (citations omitted).
(192.) Jerman v. Carlisle, McNellie, Rini, Kramer & Ulrich, L.P.A., 559 U.S. 573 (2010).
(193.) Id. at 576; see Fair Debt Collection Practices Act, Pub. L. 95-109, 91 Stat. 874 (codified as amended at 15 U.S.C. [section] 1692 etseq. (2012)).
(194.) 15 U.S.C. [section] 1692k(c) (2012).
(195.) Jerman, 559 U.S. at 579. The debt had already been paid, but the complaint contained a notice stating that the debt would be assumed valid unless Jerman disputed it in writing. Id.
(196.) Id. at 581-82.
(197.) Id. at 602.
(198.) Id. at 619 (Kennedy, J., dissenting).
(199.) See, e.g., John F. Manning, Competing Presumptions About Statutory Coherence, 74 Fordham L. Rev. 2009, 2010-11 (2006) (arguing that "legislators ... rely on semantic detail to express the level of generality at which a proposed legislative policy is acceptable to them" and that judges ignore the "untidy compromise[s]" designed to check a statute's reach when they engage in purposivist analysis); Mark Tushnet, Theory and Practice in Statutory Interpretation, 43 TEX. TECH L. REV. 1185,1196-97 (2011) (noting that Justice Scalia has made this point).
(200.) Ledbetter v. Goodyear Tire & Rubber Co., 550 U.S. 618 (2007), superseded by statute, Lilly Ledbetter Fair Pay Act of 2009, Pub. L. No. 111-2,123 Stat. 5.
(201.) Id. at 621-22.
(202.) Id. at 623-624 (discussing Civil Rights Act of 1964, Pub. L. 88-352, 78 Stat. 241 (codified at 42 U.S.C. [section] 2000e-5(f)(1) (2012))).
(203.) Id. at 624.
(204.) Id. at 628-29.
(205.) Id. at 630.
(206.) Id. at 660-61 (Ginsburg, J., dissenting) (quoting Teamsters v. United States, 431 U.S. 324, 328 (1977); Albemarle Paper Co. v. Moody, 422 U.S. 405, 418 (1975)).
(207.) Ledbetter, 550 U.S. at 661. Another good example of majority-dissent dueling over general versus specific purpose can be found in CSX Transportation, Inc. v. Alabama Department of Revenue, 131 S. Ct. 1101 (2011), where the majority opinion relied on a tax statute's broad purpose of restoring financial stability to railroads, id. at 280, while the dissent referenced the more specific statutory purpose of preventing states from discriminating against railroads, id. at 298 (Thomas, J., dissenting).
(208.) This was true in nineteen of thirty-one cases. See Fowler v. United States, 131 S. Ct. 2045, 2049, 2053 (2011) (majority argues that rejected interpretation would undermine another section of the statute [policy constancy] while dissent argues that majority's reading adopts a standard that will confuse judges and juries [administrability]); Magwood v. Patterson, 561 U.S. 320, 332-33, 355 (2010) (majority makes an absurd results argument [policy constancy], while dissent argues waste of prosecutorial resources [administrability]); Kawasaki Kisen Kaisha Ltd. v. Regal-Beloit Corp., 561 U.S. 89, 109-10, 137 (2010) (majority argues that rejected interpretation would be unworkable, leading to two different sets of rules depending on where shipping damage occurred [administrability], while dissent emphasizes unfair advantage to some carriers created by majority's reading [policy constancy]); Gross v. FBL Fin. Servs., Inc., 557 U.S. 167, 179, 191-92 (2009) (majority argues that rejected burden-shifting framework has proved difficult to apply in practice [administrability], while Justice Breyer's dissent argues that burden-shifting framework makes more sense because it would otherwise be practically impossible for plaintiffs to prove statutory violations [policy constancy]); Harbison v. Bell, 556 U.S. 180, 193-94, 204 (2009) (majority argues that rejected interpretation would lead to unworkable waste of resources, forcing defendants to get both state and federal counsel [administrability], while dissent argues that majority's interpretation will lead to absurd results and gamesmanship [policy constancy]); Vaden v. Discover Bank, 556 U.S. 49, 65, 76 (2009) (majority argues that rejected reading would create procedural mess [administrability], while dissent argues that majority's reading unjustly produces different results for parties with identical claims [policy constancy]); Bartlett v. Strickland, 556 U.S. 1, 17, 40 (2009) (plurality argues unworkability [administrability], while dissent argues consistency with statute's goals and avoidance of absurd results [policy constancy]); United States v. Hayes, 555 U.S. 415, 425-26, 436 (2009) (majority argues that rejected interpretation would make provision at issue inapplicable in most cases [policy constancy], while dissent argues that majority's reading makes statute difficult to administer, requiring elaborate fact-finding [administrability]); Greenlaw v. United States, 554 U.S. 237, 251-52, 256 (2008) (majority makes an absurd results argument [policy constancy], while dissent makes a judicial resources argument [administrability]); United States v. Santos, 553 U.S. 507, 528, 530 (2008) (concurrence argues that rejected interpretation would lead to perverse, unthinkable results [policy constancy], while dissent argues that proof problems will plague implementation of concurrence / majority reading [administrability]); United States v. Rodriquez, 553 U.S. 377, 383, 398-99 (2008) (majority claims rejected interpretation could lead to defendants receiving more than the maximum federal sentence [policy constancy], while dissent argues that majority's construction leads to lack of uniformity across states [administrability]); Gonzalez v. United States, 553 U.S. 242, 249-50, 265 (2008) (majority argues that adversary process could not function effectively under rejected construction [administrability], while dissent makes an absurd results argument [policy constancy]); Fed. Express Corp. v. Holowecki, 552 U.S. 389, 404, 420 (2008) (majority argues that rejected interpretation is "illogical and impractical" [policy constancy], while dissent argues that majority's reading leaves legal rule vague and unsettled [administrability]); Bowles v. Russell, 551 U.S. 205, 211-12, 216-17 (2007) (majority references Congress's institutional competence to decide statute of limitations parameters [administrability], while dissent references harshness of majority's result [policy constancy]); Winkelman v. Parma City Sch. Dist., 550 U.S. 516, 531, 542-43 (2007) (majority argues unfairness of treating procedural and substantive claims differently [policy constancy], while dissent argues that its construction conserves judicial resources and limits frivolous claims [administrability]); Watters v. Wachovia Bank, 550 U.S. 1,11,31-32 (2007) (majority emphasizes structural concerns - history of national bank immunity to state regulation [administrability], while dissent argues fairness [policy constancy]); Lawrence v. Florida, 549 U.S. 327, 336, 342-43 (2007) (majority cautions against interpretation that would encourage certiorari filings as a delay tactic [administrability], while dissent counters that majority's reading will produce absurd results [policy constancy]); Woodford v. Ngo, 548 U.S. 81, 89, 115-16 (2006) (majority emphasizes burden that prisoner suits place on judicial system [administrability], while dissent criticizes majority's reading as too absolute [policy constancy]); Rapanos v. United States, 547 U.S. 715, 721-22, 772, 806 (2006) (plurality emphasizes administrative costs and delays associated with permit process [administrability], while concurrence and dissent argue that plurality's reading produces inconsistent consequences [policy constancy]).
(209.) Krishnakumar, supra note 17, at 244-45.
(210.) Id. at 245-46.
(211.) Gonzalez v. United States, 553 U.S. 242 (2008).
(212.) Id. at 243.
(213.) Id. at 245.
(214.) Id. at 249.
(215.) Id. (quoting Taylor v. Illinois, 484 U.S. 400, 418 (1988)).
(216.) Id. at 262-63 (Thomas, J., dissenting); see also Winkelman, 550 U.S. at 531, 542-43 (majority making policy-constancy injustice argument about allowing parents to sue pro se to enforce IDEA; part-concurring, part-dissenting opinion counters with administrability argument that pro se cases impose unique burdens on lower courts, schools, and school districts).
(217.) See Krishnakumar, supra note 17, at 244-45.
(218.) This was true in eleven of the sixteen cases in which the Justices dueled over dictionary definitions. See Sossamon v. Texas, 131 S. Ct. 1651, 1659-60, 1668 (2011); New Process Steel, L.P. v. NLRB, 560 U.S. 674, 683, 692 (2010); Jerman v. Carlisle, McNellie, Rini, Kramer & Ulrich, L.P.A., 559 U.S. 573, 587, 613 (2010); Graham Cty. Soil & Water Conservation Dist. v. U.S. ex rel. Wilson, 559 U.S. 280, 287, 304 (2010); Cuomo v. Clearing House Assoc., 557 U.S. 519, 526, 539-40 (2009); Boyle v. United States, 556 U.S. 938, 946, 952-53 (2009); Nken v. Holder, 556 U.S. 418, 428, 441 (2009) (counting "injunction" and "enjoin" as the same word); United States v. Santos, 553 U.S. 507, 511, 531-32 (2008); Begay v. United States, 553 U.S. 137, 144, 151, 156 (2008); Nat'l Ass'n of Home Builders v. Defs. of Wildlife, 551 U.S. 644, 667-68, 692 n.12 (2007); Rapanos v. United States, 547 U.S. 715, 732-33,769-70,801 (2006).
(219.) See Kasten v. Saint-Gobain Performance Plastics, 131 S. Ct. 1325,1331-32,1337 (2011) (dueling over "filed" v. "complaint"); Bruesewitz v. Wyeth LLC, 562 U.S. 223, 231-32, 251 (2011) ("even though" v. "unavoidable"); Ali v. Fed. Bureau of Prisons, 552 U.S. 214, 228, 237 (2008) ("any" v. "detention" and "bailment"); James v. United States, 550 U.S. 192, 207, 213-14 (2007) ("potential" v. "otherwise"); Hamdan v. Rumsfeld, 548 U.S. 557, 632, 639, 711 (2006) ("constituted" v. "special").
(220.) See supra Part III.A.1 (noting a pattern that majority and dissenting opinions sometimes "[fjocus on different words or phrases in the text of the same statute").
(221.) See Boyle, 556 U.S. at 946 (Alito, J.); Nken, 556 U.S. at 441, 444 (Alito., J., dissenting); Begay, 553 U.S. at 156 (Alito, J., dissenting); James, 550 U.S. at 218 (Scalia, J., dissenting); Nat'l Ass'n of Home Builders, 551 U.S. at 667-68 (Alito, J.); Rapanos, 547 U.S. at 732-33 (Scalia, J.).
(222.) See Brudney & Baum, supra note 67, at 493,540,548 n.229,554 n.273,569-73.
(223.) Thirteen of nineteen cases containing dueling references to legislative history involved majority and dissenting references to different forms of legislative history.
(224.) Only six of nineteen cases could be said to fall into this category. See AT&T Mobility LLC v. Concepcion, 131 S. Ct. 1740, 1749, 1757 (2011) (H. Rep. v. H. Rep., floor and hearing statements); Bruesewitz, 562 U.S. at 240-42, 254-55 (H. Rep. v. H. Rep.); Carr v. U.S., 130 S. Ct. 2229, 2241, 2249 (2010) (H. Rep and evolution v. H. Rep.); Graham Cty. Soil & Water Conservation Dist. v. U.S. ex rel. Wilson, 559 U.S. 280, 296, 308 (2010) (evolution and H. Rep., S. Rep. v. evolution); AT&T Corp. v. Hulteen, 556 U.S. 701, 713, 718 (2008) (H. Rep v. H. Rep. and statutory evolution); Hamdan, 548 U.S. at 578-80, 665 (evolution, rejected proposal, floor statements v. floor statements).
(225.) Of the 528 opinions in the dataset, 137 referenced legislative history. Eighty-four of these relied on that legislative history, while fifty-three used legislative history to corroborate an interpretation reached primarily through other tools. Only nineteen of the eighty-four reliance cases, however, involved judicial dueling over this interpretive tool. See supra Table 1; infra Appendix.
(226.) In my view, the following are the only cases in the dataset in which the legislative history referenced by the Court could be considered of the "smoking-gun" variety: Gross v. FBL Fin. Servs., Inc., 557 U.S. 167, 174-75 (2009); Corley v. United States, 556 U.S. 303, 318 (2009); Hamdan, 548 U.S. at 578-80.
(227.) Briefs were not available online for two of these cases.
(228.) Most of these references were contained in the main or reply brief for the party; some were contained in an amicus brief.
(229.) See William N. Eskridge, Jr., The New Textualism and Normative Canons, 113 COLUM. L. REV. 531,534-35 (2013) (book review).
(230.) See, e.g., SCALIA, supra note 27, at 17-22, 132; ANTONIN SCALIA & BRYAN A. Garner, Reading Law: The Interpretation of Legal Texts xxviii-xxix, 18-19 (2012) (arguing that the textualist approach will produce greater certainty and less variation in the law--"The most destructive (and most alluring) feature of purposivism is its manipulability"--as well as that purposivism is unpredictable and five different judges are likely to have five different ideas about a particular statute's purpose); Frank H. Easterbrook, The Role of Original Intent in Statutory Construction, 11 Harv. J.L. & PUB. POL'Y 59, 65 (1988) (arguing that intent obfuscates the law and is less clear than textual meaning); Antonin Scalia, The Rule of Law as a Law of Rules, 56 U. Chi. L. Rev. 1175,1179 (1989) (arguing that textualism leads to predictability and clarity of statutory meaning).
(231.) See, e.g., Cipollone v. Liggett Grp., Inc., 505 U.S. 504, 548 (1992) (Scalia, J., concurring in part and dissenting in part); SCALIA, supra note 27, at 17; John F. Manning, The Absurdity Doctrine, 116 HARV. L. REV. 2387, 2397-98 (2003); W. David Slawson, Legislative History and the Need to Bring Statutory Interpretation Under the Rule of Law, 44 STAN. L. REV. 383, 416 (1992); David A. Strauss, Why Plain Meaning?, 72 NOTRE DAME L. REV. 1565,1565 (1997).
(232.) SCALIA & GARNER, supra note 230, at xxix, xxii-xxvi.
(233.) See Thomas W. Merrill, Textualism and the Future of the Chevron Doctrine, 72 WASH. U. L.Q. 351, 354, 372 (1994) (describing textualists' puzzle-solving approach); Richard J. Pierce, Jr., The Supreme Court's New Hypertextualism: An Invitation to Cacophony and Incoherence in the Administrative State, 95 COLUM. L. Rev. 749, 779 (1995) (observing that a textualist judge's interpretive process is analogous to solving a puzzle); see also Bradford C. Mank, Is a Textualist Approach to Statutory Interpretation Pro-Environmentalist?: Why Pragmatic Agency Decisionmaking is Better than Judicial Literalism, 53 WASH. & LEE L. REV. 1231, 1257 (1996) (noting textualists' conviction and certainty about their method).
(234.) Robert S. Summers, Judge Richard Posner's Jurisprudence, 89 Mich. L. Rev. 1302, 1320 (1991) (book review).
(235.) See, e.g., Scalia, supra note 27, at 16-18, 23, 25 ("[O]f course it's formalistic! The rule of law is about form."(emphases omitted)).
(236.) See U.S. CONST, art. I, [section] 7; SCALIA, supra note 27, at 35.
(237.) See, e.g., Hirschey v. FERC, 111 F.2d 1, 7-8 (D.C. Cir. 1985) (Scalia, J., concurring); SCALIA, supra note 27, at 16-23.
(238.) See Pierce, supra note 233, at 751; see also Patricia M. Wald, Some Observations on the Use of Legislative History in the 1981 Supreme Court Term, 68 IOWA L. Rev. 195, 205-06 (1983).
(239.) See SCALIA, supra note 27, at 36.
(240.) See, e.g., id.; Frank H. Easterbrook, Judicial Discretion in Statutory Interpretation, 57 OKLA. L. REV. 1,18 (2004).
(241.) Textualists may argue that the "dueling text" cases in the dataset contain one opinion that correctly construes the statute paired with an opposing opinion that incorrectly construes the statute--that is, the "dueling text" cases reveal corrupt applications of the textualist approach, rather than two correct, competing textualist constructions. Even if we accept this argument at face value, however, it means that the textualist-preferred interpretive tools are highly susceptible to judicial manipulation and can be made to appear to support the construction favored by the judge in a large percentage of the cases, precisely the charge textualists have leveled at the purposivist interpretive tools.
(242.) See supra Table 6f.
(243.) See supra Table 6f.
(244.) See supra Table 1.
(245.) See supra Table 1.
(246.) See infra Table 9.
(247.) See Confirmation Hearing on the Nomination of John G. Roberts, Jr. to Be Chief Justice of the United States Supreme Court: Hearing Before the S. Comm, on the Judiciary, 109th Cong. 318-20 (2005).
(248.) See AT&T Mobility LLC v. Concepcion, 563 U.S. 333,342 (2011); CSX Transp., Inc., v. Ala. Dep't of Revenue, 562 U.S. 277, 301 (2011) (Thomas, J., dissenting); Bruesewitz v. Wyeth LLC, 562 U.S. 223, 225, 240-42 (2011); Bloate v. United States, 562 U.S. 223, 234-35 (2010); Gross v. FBL Fin. Servs., Inc., 557 U.S. 167, 174-75 (2009); FCC v. Fox, 556 U.S. 502, 525 n.5 (2009); Harbison v. Bell, 556 U.S. 180, 198 (2009); Powerex v. Reliant Energy Servs., Inc., 551 U.S. 224,231 (2007); Hamdan v. Rumsfeld, 548 U.S. 557, 680 (2006).
(249.) Brudney & Ditslear, Legislative History, supra note 32, at 169-70.
(250.) See, e.g., Hubbard v. United States, 514 U.S. 695, 702-03 (1995) ("[T]he historical evolution of a statute--based on decisions by the entire Congress--should not be discounted for the reasons that may undermine confidence in the significance of excerpts from congressional debates and committee reports."); Morales v. Trans World Airlines, Inc., 504 U.S. 374, 385 n.2 (1992) (Scalia, J.) (comparing the enacted language of the Airline Deregulation Act with an earlier version that passed the Senate); Cont'l Can Co. v. Chi. Truck Drivers, Helpers & Warehouse Workers Union (Indep.) Pension Fund, 916 F.2d 1154, 1157 (7th Cir. 1990) ("[L]egislative history of a bill is valuable only to the extent it shows genesis and evolution....").
(251.) See infra Table 9.
(252.) Such a measure is overinclusive, excluding cases in which dueling may have occurred between a majority and dissenting opinion authored by other Justices, but it provides at least a crude control set of opinions for examining whether judicial dueling over purpose, legislative history, and intent looks substantially different when the authors of both opposing opinions are regular users of these interpretive tools.
(253.) There were only twenty-four cases in this reduced dataset in which at least one opinion referenced legislative history (compared to seventy-six such cases when cases containing opinions authored by Justices Scalia, Thomas, and Roberts are included). Of these, only four involved dueling over legislative history, compared to nineteen dueling legislative history cases in the full dataset.
(254.) See sources cited supra note 248.
(255.) For purpose, the no-Scalia-Thomas-Roberts dueling figure was seven of twenty-four cases; for intent, it was five of fourteen cases.
(256.) Moreover, the dampening argument made regarding purposivist tools is matched on the other end by dictionary definitions: Justices Souter, Ginsburg, Breyer, and Stevens--who are philosophically disinclined, though not opposed, to relying on dictionary definitions--all exhibited very low overall reference rates for dictionary reliance. See infra Table 9. If one accepts the argument that legislative history dueling would be higher if all Justices on the Court were willing to use this interpretive tool, then one also must accept that dictionary dueling would be higher if the liberal Justices were more willing to use this interpretive tool.
(257.) See SCALIA, supra note 27, at 17; Krishnakumar, supra note 17, at 270-71.
(258.) Or further into doubt, as other empirical studies have shown that the plain meaning rule does not constrain the Justices ideologically. See CROSS, supra note 50, at 176; Brudney & Ditslear, Canons of Construction, supra note 5, at 57-60.
(259.) Victoria Nourse, who along with Larry Solan, popularized the prototypical-legalist meaning distinction, argues that statutory interpreters should seek a statute's prototypical meaning. See Nourse, supra note 18, at 1000-04. For our purposes, it matters not which form of plain meaning rule courts choose, so long as they choose one and stick with it.
(260.) See, e.g., William N. Eskridge, Jr. & Philip P. Frickey, An Historical and Critical Introduction, in Henry M. Hart, Jr. & Albert M. Sacks, The Legal Process: Basic Problems in the Making and Application of Law, at li, lii (William N. Eskridge, Jr. & Philip P. Frickey eds., 1994) (noting the centrality of The Legal Process materials to purpose-based views of statutory interpretation).
(261.) Id. at xcii.
(262.) See Eskridge, supra note 16, at 1548-49; Stevens, supra note 16, at 1-2.
(263.) See Roger Colinvaux, What Is Law? A Search For Legal Meaning And Good Judging Under A Textualist Lens, 72 IND. L.J. 1133, 1139 (1997); William N. Eskridge, Jr. & Philip P. Frickey, Statutory Interpretation as Practical Reasoning, 42 STAN. L. REV. 321, 334-35 (1990); Albert C. Lin, Erosive Interpretation of Environmental Law in the Supreme Court's 2003-04 Term, 42 HOUS. L. REV. 565, 574 (2005); Manning, supra note 231, at 2416-17; see also John F. Manning, What Divides Textualists from Purposivists?, 106 COLUM. L. REV. 70, 103 (2006) (questioning "whether the construct of imputed semantic meaning reflects a more defensible conception of legislative supremacy than does the construct of imputed policy coherence").
(264.) See, e.g., ESKRIDGE, supra note 78, at 26-29 (commenting that the application of statutory purpose is dependent on the perspective of the interpreter); Philip P. Frickey, From the Big Sleep to the Big Heat: The Revival of Theory in Statutory Interpretation, 77 Minn. L. Rev. 241, 251 (1992); Adrien Vermeule, Legislative History and the Limits of Judicial Competence: The Untold Story of Holy Trinity Church, 50 Stan. L. Rev. 1833, 1885 (1998) (criticizing the "malleability of purposive interpretation").
(265.) See supra Table 1.
(266.) See supra Table 6g.
(267.) See cases cited supra note 219.
(268.) See, e.g., Otto J. Hetzel, Michael E. Libonati & Robert F. Williams, Legislative Law and Process 589 (3d ed. 2001); Eskridge et al., supra note 1, at 972 n.d (providing an ordered list of legislative history materials from OTTO Hetzel, Michael Libonati & Robert Williams, Legislative Law and Process 589 (3d ed. 2001)). For a persuasive argument that the most relevant legislative history is that which corresponds to the last decisionmaking point by Congress, see Nourse, supra note 149, at 98.
(269.) See In re Silicon Graphics Inc. Sec. Litig., 183 F.3d 970, 977 (9th Cir. 1999); accord NW. Forest Res. Council v. Glickman, 82 F.3d 825, 835 (9th Cir. 1996) ("[A] congressional conference report is recognized as the most reliable evidence of congressional intent because it 'represents the final statement of the terms agreed to by both houses.'" (quoting Dep't of Health & Welfare v. Block, 784 F.2d 895, 901 (9th Cir. 1986))).
(270.) See Jorge L. Carro & Andrew R. Brann, The U.S. Supreme Court and the Use of Legislative Histories: A Statistical Analysis, 22 JURIMETRICS J. 294, 299 (1982); Michael H. Koby, The Supreme Court's Declining Reliance on Legislative History: The Impact of Justice Scalia's Critique, 36 HARV. J. ON LEGIS. 369, 390 (1999); Krishnakumar, supra note 17, at 237.
(271.) For example, we often see different sources that address the interpretive question at issue with varying degrees of precision, such that an on-point line of questioning from a hearing arguably could trump, say, a vague or silent committee report.
(272.) See, e.g., RICHARD A. POSNER, The PROBLEMATICS OF MORAL & LEGAL THEORY 227 (1999); Richard A. Posner, The Problems of Jurisprudence 73-74 (1990).
(273.) Richard A. Posner, Law, Pragmatism, and Democracy 3 (2003).
(274.) See, e.g., Eskridge & Frickey, supra note 260, at 321; William N. Eskridge, Jr., Dynamic Statutory Interpretation, 135 U. Pa. L. Rev. 1479 (1987); see also T. Alexander Aleinikoff, Updating Statutory Interpretation, 87 MICH. L. REV. 20, 46 (1988) ("[S]tatutes ought to be responsive to today's world. They ought to be made to fit, as best they can, into the current legal landscape.").
(275.) See, e.g., E.W. THOMAS, THE JUDICIAL PROCESS: REALISM, PRAGMATISM, Practical Reasoning and Principles 4-6 (2005); Mark Tushnet, Red, White, and Blue: A Critical Analysis of Constitutional Law 193-94 (1988); Zipporah Batshaw Wiseman, The Limits of Vision: Karl Llewellyn and the Merchant Rules, 100 Harv. L. Rev. 465, 470-71 (1987).
(276.) See supra Table 1.
(277.) This is true even for those Justices who openly subscribe to other interpretive approaches and even for textualists, who object in principle to pragmatic reasoning.
(278.) See Lisa Schultz Bressman & Abbe R. Gluck, Statutory Interpretation from the Inside--An Empirical Study of Congressional Drafting, Delegation, and the Canons: Part II, 66 STAN. L. REV. 725,788 (2014).
(279.) Gluck, supra note 19, at 1754.
(280.) Portland Gen. Elec. Co. v. Bureau of Labor & Indus., 859 P.2d 1143, 1145-47 (Or. 1993); Jack. L. Landau, The Intended Meaning of "Legislative Intent" and Its Implications for Statutory Construction in Oregon, 76 OR. L. Rev. 47, 50 (1997).
(281.) Portland Gen. Elec. Co., 859 P.2d at 1146.
(282.) See Sydney Foster, Should Courts Give Stare Decisis Effect to Statutory Interpretation Methodology?, 96 Geo. L.J. 1863, 1885 (2008); Gluck, supra note 19, at 1767; Nicholas Quinn Rosenkranz, Federal Rules of Statutory Interpretation, 115 Harv. L. Rev. 2085, 2142 (2002).
(283.) See Aaron-Andrew P. Bruhl, Hierarchy and Heterogeneity: How to Read a Statute in a Lower Court, 97 CORNELL L. REV. 433, 443-44 (2012); Ethan J. Leib & Michael Serota, The Costs of Consensus in Statutory Construction, 120 Yale L.J. ONLINE 47 (2010).
(284.) See supra Table 1. Recall that most interpretive tools coded for showed dueling rates in the range of 21.8-28.6 percent (whole act rule, dictionary definitions, common law, purpose, intent, legislative history) or lower, at 10.9-13.3 percent (substantive canons and grammar canons). Two other interpretive tools, practical consequences and other statutes showed rates of dueling in slightly more than three out of ten cases that invoked these tools (30.9 percent and 34.5 percent, respectively). Only one interpretive tool--text/plain meaning--gave rise to judicial dueling at even close to half of the cases that invoked the tool (42.7 percent).
(285.) See Gluck, supra note 19 at 1754.
(286.) For a doctrinal analysis and discussion, see supra Part III.
Table 1: Dueling Canon Use 2005-2010 Terms Unanimous Majority Dissenting Cases Opinions Opinions (n=115) * (n=255) (n=155) Supreme Court 69 157 75 Precedent Text or Plain 68 145 64 Meaning Dictionary Rule 28 75 25 Other Statutes 80 44 33 Language 54 117 40 Canons + Whole Act Rule Whole Act Rule 56 103 35 Grammar or 19 38 14 Linguistic Canons Substantive 17 37 26 Canons Common Law 25 40 12 Purpose 30 75 47 Practical 45 93 69 Consequences Intent 22 40 39 Legislative 32 74 47 History Dueling Percent Opinions Dueling In (Majority or Divided- Percent Concurrence Vote Cases Dueling + Dissent) ([double In All ([dagger]) dagger]) Cases # Supreme Court 58 63.7% 32.6% Precedent (n=91) (n=178) Text or Plain 44 42.7% 25.3% Meaning (n=103) (n=174) Dictionary Rule 16 28.6% 18.2% (n=56) (n=88) Other Statutes 20 34.5% 21.1% (n=58) (n=95) Language 21 24.7% 14.1% Canons + Whole (n=85) (n=149) Act Rule Whole Act Rule 17 21.8% 12.7% (n=78) (n=134) Grammar or 4 13.3% 8.2% Linguistic (n=30) (n=49) Canons Substantive 3 6.8% 4.9% Canons (n=44) (n=61) Common Law 6 22.2% 11.5% (n=27) (n=52) Purpose 20 24.7% 19.6% (n=77) (n=107) Practical 29 30.9% 20.9% Consequences (n=94) (n=139) Intent 13 26.5% 18.3% (n=49) (n=71) Legislative 19 25.3% 17.6% History (n=75) (n=108) * This column reports the number unanimous cases in which the majority opinion invoked an interpretive tool to support its chosen construction. ** This column includes a small number of double-counts for cases in which more than one dissenting opinion referenced the interpretive tool. ([dagger]) This column counts the number of cases in which both the majority and one or more dissenting opinions relied on the interpretive tool to support its construction and a few cases in which a concurring opinion and at least one dissenting opinion relied on the interpretive tool. As explained in the article's methodology section, the figures reported in this column do not include cases in which one opinion relied on the interpretive tool and an opposing opinion rejected the tool as inapplicable or criticized the opposing opinion's reliance on the tool. ([double dagger]) Percentages in this column were calculated by dividing the number of cases that dueled over this interpretive tool (reported in previous column) by the number of divided-vote cases in the dataset in which at least one opinion relied on the tool to support its statutory construction (reflected in the figure n=90 for Supreme Court precedent, n=104 for Text/Plain meaning, and so on). # Percentages in this column were calculated by dividing the number of cases that dueled over this interpretive tool (reported in the fourth column) by the total number of cases in the dataset in which at least one opinion relied on the tool to support its statutory construction, including unanimous cases (reflected in the figure n=177 for Supreme Court precedent, n=175 for Text/Plain meaning, and so on). Table 2a: Dueling Canon Use by Vote Margin * Percent Dueling Over All Legal Analysis Tools Percent Dueling (Including Over Statutory Precedent-Only Interpretation- and Practical- Specific Tools (n) Only Dueling) (n) All Cases (Including 34.5% (88) 44.3% (113) Unanimous Cases) (n=255) Divided-Vote Cases 62.9% (88) 80.7% (113) (n=140) Close-Margin Cases 64.2% (34) 84.9% (45) (5-4 / 5-3) (n=53) Wide-Margin Cases 62.1% (54) 78.2% (68) (n=87) * A chi-squared test revealed no statistical difference between rates of dueling for close-margin versus wide-margin cases. Table 2b: Dissenting Opinion Reliance on Canons and Other Interpretative Tools Dissenting Opinion Dissenting Opinion Reliance When Reliance Absent Majority Opinion Majority Opinion Invokes the References Canon/Tool Text / Plain Meaning 32.3% (20 out of 62) 56.4% (44 out of 78) Dictionary 10.6% (10 out of 94) 34.8% (16 out of 46) Other Statutes 13.5% (13 out of 96) 45.5% (20 out of 44) Language Canons/ 24.7% (19 out of 77) 33.3% (21 out of 63) Grammar / Whole Act Rule Substantive Canons 19.5% (23 out of 118) 13.6% (3 out of 22) Common Law 5.0% (6 out of 121) 31.6% (6 out of 19) Purpose 30.8% (28 out of 91) 38.8% (19 out of 49) Intent 22.0% (26 out of 118) 59.1% (13 out of 22) Legislative History 28.9% (28 out of 97) 44.2% (19 out of 43) Practical Consequences 49.4% (40 out of 81) 49.2% (29 out of 59) Supreme Court 32.7% (17 out of 52) 65.9% (58 out of 88) Precedent Table 3a: Dueling Canon Use by Subject Area Percentage of All Dueling Cases That Involve Statutes in this Subject Area (n=88) Criminal 26.1% (23) (incl. AEDPA) Discrimination 13.6% (12) (incl. IDEA) Environmental 6.8% (6) Jurisdictional 4.5% (4) (not incl. criminal) Preemption 5.7% (5) FAA 3.4% (3) Securities 3.4% (3) Bankruptcy 5.7% (5) Immigration 2.3% (2) Communications Act 2.3% (2) False Claims Act 2.3% (2) NLRA / Employment 2.3% (2) Other * 21.6% (19) * Fourteen statutes/subject areas generated only one dueling canon case: Terrorism Risk Insurance Act; Civil Racketeer Influenced and Corrupt Organizations Act; Employee Retirement Income Security Act; Railroads; Fair Debt Collection Act; Religion; Federal Tort Claims Act; Freedom of Information Act; Tax; Indian law; Attorneys' Fees; Federal Power Act; National Banking Act; Jones Act. Three others generated two dueling canon cases each: the Prison Litigation Reform Act, statutes dealing with Procedure, and Intellectual Property. Table 3b: Subject Area Dueling Criminal Antidiscrimination Environmental Plain Meaning 43.5% (10) 41.7% (5) 33.3% (2) Dictionary 13.0% (3) 8.3% (1) 33.3% (2) Language/Whole Act 26.1% (6) 16.7% (2) 33.3% (2) Rule Other Statutes 21.7% (5) 25.0% (3) 0.0% (0) Purpose 4.3% (1) 16.7% (2) 33.3% (2) Legislative History 21.7% (5) 25.0% (3) 0.0% (0) Substantive Canons 0.0% (0) 0.0% (0) 0.0% (0) Table 4: Ideology by Case Type Conservative Neither (n) Liberal (n) (n) All Statutory Cases 49.8% (127) 46.7% (119) 3.5% (9) (n=255) Dueling Cases 62.5% (55) 37.1% (33) 0.0% (0) (n=88) Unanimous Cases 40.0% 53.0% 7.0% (n=115) Divided-Vote Cases 57.9% (81) 41.4% (58) 0.7% (1) (n=140) Table 5: Dueling by Vote Margin Percent of Cases Within Vote Percent of Dueling Margin Canon Cases Involving Decided By This Dueling Canons# Vote Margin 5 or 4 Justice Majority 64.2% 38.6% (5-4/5-3/4-1-4) (n=53) 5,4, or 6 Justice 68.4% * 72.7% Majority Cases (n=95) 7 or More Justice 51.1% 27.3% Majority Cases (n=45) Percent of Cases Within Vote Margin Involving Any Dueling (Including Precedent-Only and Practical-Only Dueling) 5 or 4 Justice Majority 84.9% (5-4/5-3/4-1-4) (n=53) 5,4, or 6 Justice 84.2% Majority Cases (n=95) 7 or More Justice 73.3% Majority Cases (n=45) # For purposes of this Table, "dueling canon use" refers to cases in which majority and dissenting opinions (or, in a few cases, a concurrence and at least one dissenting opinion) both employed the same statutory-interpretation-specific interpretive tools to reach different statutory readings and case outcomes. This column does not include cases in which majority and dissenting opinions dueled solely over the application of Supreme Court precedent, practical reasoning, or both precedent and practical reasoning. * A chi-squared test showed that the difference between rates of dueling for opinions with six or fewer Justices in the majority versus opinions with seven or more Justices in the majority approached significance, at p=.084. Table 6a: Judicial Ideology in Dueling Legislative History Cases (n=19) Justice Percent Percent Justice Ideology Conservative * Liberal ** Scalia Conservative 68.4% (13) 10.5% (2) Thomas Conservative 63.2% (12) 15.8% (3) Roberts Conservative 52.6% (10) 21.1% (4) Alito Conservative 73.7% (14) 15.8% (3) Kennedy Conservative 42.1% (8) 42.1% (8) Souter Liberal 21.1% (4) 36.8% (7) Ginsburg Liberal 31.6% (6) 57.9% (11) Breyer Liberal 15.8% (3) 68.4% (13) Stevens Liberal 15.8% (3) 63.2% (12) Sotomayor Liberal 10.5% (2) 21.1% (4) Kagan Liberal 5.2% (1) 10.5% (2) Percent Not Conservative Participating -Liberal Justice / Neither Differential Scalia 21.1% (4) 58.1% Thomas 21.1% (4) 47.4% Roberts 26.3% (5) 31.5% Alito 10.5% (2) 57.9% Kennedy 15.8% (3) 0.0% Souter 42.1% (8) -15.7% Ginsburg 10.5% (2) -26.3% Breyer 15.8% (3) -52.6% Stevens 21.1% (4) -47.4% Sotomayor 68.4% (13) -- Kagan 84.2% (16) -- * Percentage of cases involving majority-dissenting dueling over legislative intent in which the Justice authored or joined an opinion that referenced intent to reach a conservative outcome. ** Percentage of cases involving majority-dissenting dueling over text or plain meaning in which the Justice authored or joined an opinion that referenced text or plain meaning to reach a liberal outcome. Table 6b: Judicial Ideology in Dueling Purpose Cases (n = 19) Justice Percent Percent Justice Ideology Conservative * Liberal ** Scalia Conservative 63.2% (12) 26.3% (5) Thomas Conservative 68.4% (13) 26.3% (5) Roberts Conservative 68.4% (13) 26.3% (5) Alito Conservative 78.9% (15) 15.8% (3) Kennedy Conservative 73.7% (14) 26.3% (5) Souter Liberal 26.3% (5) 31.6% (6) Ginsburg Liberal 21.1% (4) 78.9% (15) Breyer Liberal 36.8% (7) 63.2% (12) Stevens Liberal 15.8% (3) 68.4% (13) Sotomayor Liberal 10.5% (2) 26.3% (5) Kagan Liberal 5.26% (1) 5.26% (1) Percent Not Conservative Participating -Liberal Justice / Neither Differential Scalia 10.5% (2) 36.9% Thomas 5.26% (1) 42.1% Roberts 5.26% (1) 42.1% Alito 5.26% (1) 63.1% Kennedy 0.0% (0) 47.4% Souter 42.1% (8) -5.3% Ginsburg 0.0% (0) -57.8% Breyer 0.0.% (0) -26.5% Stevens 15.8% (3) -52.6% Sotomayor 63.2% (12) -- Kagan 89.4% (17) -- * Percentage of cases involving majority-dissenting dueling over legislative intent in which the Justice authored or joined an opinion that referenced intent to reach a conservative outcome. ** Percentage of cases involving majority-dissenting dueling over text or plain meaning in which the Justice authored or joined an opinion that referenced text or plain meaning to reach a liberal outcome. Table 6c: Judicial Ideology in Dueling Dictionary Cases (n=16) Justice Percent Percent Justice Ideology Conservative * Liberal ** Scalia Conservative 50.0% (8) 50.0% (8) Thomas Conservative 62.5% (10) 37.5% (6) Roberts Conservative 50.0% (8) 43.75% (7) Alito Conservative 75.0% (12) 25.0% (4) Kennedy Conservative 56.25% (9) 43.75% (7) Souter Liberal 18.75% (3) 43.75% (7) Ginsburg Liberal 37.5% (6) 62.5% (10) Breyer Liberal 25.0% (4) 75.0% (12) Stevens Liberal 18.75% (3) 62.5% (10) Sotomayor Liberal 12.5% (2) 37.5% (4) Kagan Liberal 0.0% (0) 0.0% (0) Percent Not Conservative Participating -Liberal Justice / Neither Differential Scalia 0.0% (0) 0.0% Thomas 0.0% (0) 25.0% Roberts 6.25% (1) 6.25% Alito 0.0% (0) 50.0% Kennedy 0.0% (0) 12.5% Souter 37.5% (6) -25.0% Ginsburg 0.0% (0) -25.0% Breyer 0.0% (0) -50.0% Stevens 18.75% (3) -50.0% Sotomayor 62.5% (10) -- Kagan 100.0% (16) -- * Percentage of cases involving majority-dissenting dueling over legislative intent in which the Justice authored or joined an opinion that referenced intent to reach a conservative outcome. ** Percentage of cases involving majority-dissenting dueling over text or plain meaning in which the Justice authored or joined an opinion that referenced text or plain meaning to reach a liberal outcome. Table 6d: Judicial Ideology in Dueling Language/Grammar/Whole Act Cases (n=21) Justice Percent Percent Justice Ideology Conservative * Liberal ** Scalia Conservative 71.4% (15) 28.6% (6) Thomas Conservative 85.7% (18) 14.3% (3) Roberts Conservative 57.1% (12) 38.1% (8) Alito Conservative 71.4% (15) 28.6% (6) Kennedy Conservative 57.1% (12) 38.1% (8) Souter Liberal 28.6% (6) 57.1% (12) Ginsburg Liberal 28.6% (6) 61.9% (13) Breyer Liberal 9.5% (2) 66.7% (14) Stevens Liberal 0.0% (0) 76.2% (16) Sotomayor Liberal 0.0% (0) 9.5% (2) Kagan Liberal 0.0% (0) 0.0 (0) Percent Not Conservative Participating -Liberal Justice / Neither Differential Scalia 0.0% (0) 42.8% Thomas 0.0% (0) 71.4% Roberts 4.8% (1) 19.0% Alito 0.0% (0) 42.8% Kennedy 4.8% (1) 19.0% Souter 14.3% (3) -28.5% Ginsburg 9.5% (2) -33.3% Breyer 23.8% (5) -57.2% Stevens 23.8% (5) -76.2% Sotomayor 90.5% (19) -- Kagan 100.0% (21) -- * Percentage of cases involving majority-dissenting dueling over legislative intent in which the Justice authored or joined an opinion that referenced intent to reach a conservative outcome. ** Percentage of cases involving majority-dissenting dueling over text or plain meaning in which the Justice authored or joined an opinion that referenced text or plain meaning to reach a liberal outcome. Table 6e: Judicial Ideology in Dueling Other Statutes Cases (n=20) Justice Percent Percent Justice Ideology Conservative * Liberal ** Scalia Conservative 65.0% (13) 30.0% (6) Thomas Conservative 65.0% (13) 35.0% (7) Roberts Conservative 55.0% (11) 40.0% (8) Alito Conservative 55.0% (11) 40.0% (8) Kennedy Conservative 55.0% (11) 40.0% (8) Souter Liberal 30.0% (6) 45.0% (9) Ginsburg Liberal 30.0% (6) 65.0% (13) Breyer Liberal 15.0% (3) 85.0% (17) Stevens Liberal 15.0% (3) 75.0% (15) Sotomayor Liberal 0.0% (0) 20.0% (4) Kagan Liberal 0.0% (0) 0.0% (0) Percent Not Conservative Participating -Liberal Justice / Neither Differential Scalia 5.0% (1) 35.0% Thomas 0.0% (0) 30.0% Roberts 5.0% (1) 15.0% Alito 5.0% (1) 15.0% Kennedy 5.0% (1) 15.0% Souter 25.0% (5) -5.0% Ginsburg 5.0% (1) -35.0% Breyer 0.0% (0) -70.0% Stevens 10.0% (2) -60.0% Sotomayor 80.0% (16) -- Kagan 100.0% (20) -- * Percentage of cases involving majority-dissenting dueling over legislative intent in which the Justice authored or joined an opinion that referenced intent to reach a conservative outcome. ** Percentage of cases involving majority-dissenting dueling over text or plain meaning in which the Justice authored or joined an opinion that referenced text or plain meaning to reach a liberal outcome. Table 6f: Judicial Ideology in Dueling Plain Meaning Cases (n=43) Justice Percent Percent Justice Ideology Conservative * Liberal ** Scalia Conservative 61.4% (27) 34.1% (15) Thomas Conservative 72.7% (32) 22.7% (10) Roberts Conservative 50.0% (22) 45.5% (20) Alito Conservative 63.6% (28) 31.8% (14) Kennedy Conservative 50.0% (22) 50.0% (22) Souter Liberal 15.9% (7) 43.2% (19) Ginsburg Liberal 36.3% (16) 56.8% (25) Breyer Liberal 29.5% (13) 61.4% (27) Stevens Liberal 13.6% (6) 65.9% (29) Sotomayor Liberal 11.4% (5) 15.9% (7) Kagan Liberal 4.5% (2) 11.4% (5) Percent Not Conservative Participating -Liberal Justice / Neither Differential Scalia 4.5% (2) 27.3% Thomas 4.5% (2) 50.0% Roberts 4.5% (2) 4.5% Alito 4.5% (2) 31.8% Kennedy 0.0% (0) 0.0% Souter 40.9% (18) -27.3% Ginsburg 6.8% (3) -20.5% Breyer 9.1% (4) -31.9% Stevens 20.5% (9) -52.3% Sotomayor 72.7% (32) -- Kagan 84.1% (37) -- * Percentage of cases involving majority-dissenting dueling over legislative intent in which the Justice authored or joined an opinion that referenced intent to reach a conservative outcome. ** Percentage of cases involving majority-dissenting dueling over text or plain meaning in which the Justice authored or joined an opinion that referenced text or plain meaning to reach a liberal outcome. Table 6g: Judicial Ideology in Dueling Legislative Intent Cases (n=13) Justice Percent Percent Justice Ideology Conservative * Liberal ** Scalia Conservative 76.9% (10) 23.1% (3) Thomas Conservative 76.9% (10) 23.1% (3) Roberts Conservative 76.9% (10) 15.4% (2) Alito Conservative 61.5% (8) 30.8% (4) Kennedy Conservative 69.2% (9) 30.8% (4) Souter Liberal 30.8% (4) 69.2% (8) Ginsburg Liberal 46.2% (6) 53.8% (7) Breyer Liberal 46.2% (6) 46.2% (6) Stevens Liberal 0.0% 92.3% (12) Sotomayor Liberal 0.0% (0) 7.7% (1) Kagan Liberal 0.0% (0) 7.7% (1) Percent Not Conservative Participating -Liberal Justice / Neither Differential Scalia 0.0% (0) 53.8% Thomas 0.0% (0) 53.8% Roberts 7.7% (1) 62.5% Alito 7.7% (1) 30.7% Kennedy 0.0% (0) 38.4% Souter 7.7% (1) -38.4% Ginsburg 0.0% (0) -7.6% Breyer 7.7% (1) 0.0% Stevens 7.7% (1) 92.3% Sotomayor 92.3% (12) -- Kagan 92.3% (12) -- * Percentage of cases involving majority-dissenting dueling over legislative intent in which the Justice authored or joined an opinion that referenced intent to reach a conservative outcome. ** Percentage of cases involving majority-dissenting dueling over legislative intent in which the Justice authored or joined an opinion that referenced intent to reach a liberal outcome. Table 7: Ideology by Canon/Tool in Unanimous Cases Conservative Percent Percent Percent -Liberal Conservative Liberal Neither Differential Ideology in 40% 53% 7.0% -13.0% Unanimous Cases (n=115) Text/Plain 51.47% 41.17% 7.35% 10.3% Meaning ([dagger]) (n=68) Dictionaries (131) 28.57% 60.7% 10.71% -32.13% (n=28) Language / 38.89% 53.7% 74.07% -14.81% Grammar / Whole Act Rule (n=54) Other Statutes 44.4% 50.0% 5.55% -5.6% (n=36) Legislative 30.0% 56.7% 13.33% -26.7% History (132) (n=30) Purpose ([dagger]) 50.0% 38.46% 11.54% 11.54% (n=26) Intent 40.9% 50.0% 9.1% -9.1% (n=22) Practical 52.94% 38.24% 8.82% 14.7% Consequences ([dagger]) (n=34) Common 57.14% 28.57% 14.29% 28.57% Law * ([dagger]) (n=21) Substantive * 41.2% 47.05% 11.8% -5.85% ([dagger]) Canons (n=17) ([dagger]) Denotes that this canon or tool skewed in the opposite ideological direction from the overall ideological breakdown for unanimous cases (canon/tool tended to lead to conservative case outcomes, while a majority of unanimous cases had liberal case outcomes). * Indicates chi-squared test reveals a significant difference between the rates at which this tool was used to support liberal and conservative outcomes at p<.05. (Except that for substantive canons, the difference approaches significance at p=.057). Table 8: Ideology by Opinion Author Conservative Liberal Indeterminate Per Curiam (n=15) 26.7% 73.3% 0.0% Scalia (n=68) 63.2% 35.3% 1.5% Thomas (n=68) 58.8% 38.2% 3.0% Roberts (n=38) 52.6% 44.7% 2.6% Alito (n=52) 50.0% 46.2% 3.8% Kennedy (n=44) 43.2% 52.3% 4.5% Souter (n=34) 35.3% 64.7% 0.0% Ginsburg (n=46) 37.0% 63.0% 0.0% Breyer (n=73) 31.5% 63.0% 5.5% Stevens (n=60) 25.0% 70.0% 5.0% Sotomayor (n=23) 52.2% 43.5% 4.35% Kagan (n=7) 42.85% 42.85% 14.29% Table 9: Rates of Reliance on Interpretative Canons and Tools by Opinion Author Scalia Thomas Alito Roberts Kennedy (n=68) (n=67) (n=51) (n=38) (n=44) Supreme Court 36.8% 53.7% 41.2% 57.9% 61.4% Precedent Text or Plain Meaning 61.8% 64.2% 54.9% 50.0% 50.0% Rule * Dictionary Rule 22.1% 26.9% 29.4% 15.8% 27.2% Language Canons + Whole 27.9% 44.1% 37.3% 55.3% 29.5% Act Rule (Combined) Other Statutes 25.0% 17.9% 33.3% 28.9% 22.7% Common Law 16.2% 9.0% 13.7% 18.4% 4.5% Substantive Canons 8.8% 11.9% 9.6% 21.1% 13.6% Whole Act Rule 25.0% 37.3% 31.4% 47.4% 22.7% Practical 30.9% 17.9% 41.2% 31.6% 47.7% Consequences * Purpose * 8.8% 14.9% 25.5% 13.2% 47.7% Intent * 2.9% 6.0% 15.7% 7.9% 6.8% Legislative History * 10.3% 9.0% 23.1% 13.2% 22.7% Souter Ginsburg Breyer Stevens (n=35) (n=46) (n=73) (n=60) Supreme Court 54.3% 47.8% 49.3% 55.0% Precedent Text or Plain Meaning 45.7% 30.4% 27.4% 43.3% Rule * Dictionary Rule 17.1% 13.0% 15.1% 16.7% Language Canons + Whole 31.4% 30.4% 20.5% 25.0% Act Rule (Combined) Other Statutes 22.9% 23.9% 20.5% 16.7% Common Law 14.3% 6.5% 11.0% 16.7% Substantive Canons 11.4% 13.0% 8.2% 25.0% Whole Act Rule 31.4% 28.3% 19.2% 21.7% Practical 31.4% 50.0% 42.5% 30.0% Consequences * Purpose * 17.1% 34.8% 41.1% 30.0% Intent * 22.9% 21.7% 21.9% 41.7% Legislative History * 28.6% 32.6% 41.1% 38.3% * Indicates that one-way ANOVA test, using Bonferroni multiple comparison test, reveals a significant difference between rates of reliance by different Justices in the opinions they authored at p<.05. (For Text / Plain Meaning p=.0016; Purpose p=.0001; for Intent p<.00001; and for Legislative History p=.0001). For Substantive Canons, the differences in rates of reliance approached significance at p=.0597.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Supreme Court justices' different uses of legal canons in statutory interpretation; Continuation of II. Findings B. Dueling Statistics 3. Dueling Canons and Ideology through Conclusion, with footnotes and appendices, p. 957-1006|
|Author:||Krishnakumar, Anita S.|
|Publication:||Duke Law Journal|
|Date:||Feb 1, 2016|
|Previous Article:||Dueling canons.|
|Next Article:||An unreasonable ban on reasonable competition: the legal profession's protectionist stance against noncompete agreements binding in-house counsel.|