Printer Friendly

Against coherence in statutory interpretation.

ABSTRACT

A long tradition in legal theory views the judicial role as centrally including the duty to make the entire body of law "speak with one voice. " This coherence ideal permeates much of the law of statutory interpretation, but one body of doctrine that it has particularly influenced is the set of standards that federal courts use to determine when a newly enacted statute overrides preexisting legal rules. Determining whether Congress implicitly intends to preempt state law, repeal previous legislation, or displace federal common law is an increasingly important part of the "ordinary diet of the law. " And although, this Article maintains, modern preemption doctrine is largely consistent with the presumptive judicial role in statutory interpretation--that of Congress's faithful agent--the desire for coherence has motivated the Court to develop standards governing repeal and displacement that deviate from the preemption framework.

This Article argues that courts should abandon the quest for coherence in statutory interpretation. In a reasonably pluralistic society like ours, widespread agreement on a coherent ranking of basic values is unlikely. Against this backdrop of deep disagreement, collective social action is purchased only by hammering out specific compromises, and the overall pattern of compromises is unlikely to be coherent. Imposing coherence on the body of law accordingly unravels the very compromises that allowed the legislature to act and, in doing so, both disrespects the process of mutual compromise that made collective action possible and impedes future legislative action. Recognizing the importance of compromise to modern legislation should lead to the rejection of normative coherence as an ideal in statutory interpretation. And, absent some other justification for their current divergence, the doctrinal standards for repeal and displacement should be unified with the current preemption framework.

INTRODUCTION

In unpublished teaching materials that would come to represent, for later generations, a near-canonical encapsulation of mid-twentieth-century American thought about public law and legal theory, Henry Hart, Jr., and Albert Sacks articulated a vision of the American legal system as "a single system of law: it speaks, in relation to any particular question with only one ultimately authoritative voice." (1) On one understanding, the metaphor of speaking with one voice is banal. Amy system of law that applies across a large, diverse geographical area and purports to be a single, comprehensive system must include mechanisms for avoiding logical inconsistency. Were people routinely governed by two inconsistent yet equally authoritative legal rules--directions to [phi] and not to [phi]--the legal system would surely "like Pavlov's dogs ... suffer a nervous breakdown." (2)

But elsewhere in the legal process materials, Hart and Sacks indicate that they have in mind a more robust understanding of the single-voice metaphor. They repeatedly press the reader to consider which outcome in a given case "harmonizes best with the general body of statutory law." (3) Near the beginning of their materials, they suggestively ask: "Does not a legal system require some means for rationalizing the fabric of its law as a whole?" (4) By the time one emerges from the other end of the materials, the particular means that Hart and Sacks have in mind is no mystery: it falls to courts, through a process of "reasoned elaboration," (5) to interpret statutes " 'if possible, so as to harmonize them with more general principles and policies' of law." (6) For Hart and Sacks, then, the single-voice metaphor involves some larger consistency of basic purpose, a vision of law as a seamless web formed of coherent and mutually reinforcing strands.

Hart and Sacks are not alone in urging courts to consider, in interpreting statutes, what outcome fits most coherently into the fabric of law as a whole. Ronald Dworkin, almost certainly the most influential legal philosopher of the late twentieth century and one of many intellectual heirs to Hart and Sacks's thought, (7) famously urged courts to "identify legal rights and duties, so far as possible, on the assumption that they were all created by a single author--the community personified--expressing a coherent conception of justice and fairness." (8) However, the sheer intellectual dominance of the coherence ideal is perhaps best demonstrated by Justice Antonin Scalia. Scalia--the fiercest modern proponent of a formalist alternative to Legal Process Theory (9)--has embraced the judicial duty to "make sense rather than nonsense out of the corpus juris" by construing ambiguous statutory language so as to fit "most logically and comfortably into the body of both previously and subsequently enacted law." (10) The impulse toward coherence runs deep.

One domain that has been profoundly influenced by the coherence ideal is the set of doctrines that determine when legislatively enacted rules are taken to have superseded previously existing legal rules. (11) When Congress enacts a new set of legal rules but remains silent regarding the effect of its handiwork on preexisting bodies of law, it falls to courts to determine the extent to which the new rules trump the old. And courts have articulated a set of standards to aid them in interpreting Congress's silence on this point. To determine implied congressional preemption of state law, the Court has developed a set of legal standards that, this Article argues, are reasonably well adapted to faithfully determining congressional intent, as expressed by text in context. But the doctrinal tests that the Court has developed to govern the other two primary areas of supersession diverge from the preemption paradigm. The standards for determining whether Congress has impliedly repealed previous federal legislation are significantly more demanding than preemption doctrine, and the standards governing displacement of federal common law are significantly less so.

This divergence in the Court's supersession doctrine demands explanation for intrinsic, practical reasons; questions concerning congressional supersession of preexisting legal norms make up an increasingly important part of the federal docket. However, the divergence deserves attention for another, deeper reason: it raises fundamental questions about the appropriate role of the coherence ideal. Below, I argue that the divergence of the Court's three supersession (12) doctrines is in large part explained by the tension between two models of the judicial role: the fidelity model and the coherence model. While the current preemption framework is largely consistent with the judiciary's usual role in cases of statutory interpretation--that of Congress's faithful agent--courts have been led both to undervalue Congress's intent to impliedly repeal legislation and to place a thumb on the scale in favor of displacement of federal common law out of a desire to make the total body of law fit together as harmoniously as possible.

This Article argues that the coherence ideal fails to justify the courts' departure from their presumptive duty to faithfully carry out Congress's will. Coherence fails as a justification because it is inconsistent with the central feature of lawmaking in a modern democracy: the importance of compromise. As citizens of the United States, we disagree persistently and sometimes stridently over the appropriate priority among basic values, and we act together when our value systems come apart only by striking particular compromises, one by one. In a society marked by what John Rawls termed "reasonable pluralism," (13) this process of mutual compromise is what enables collective social action; but the overall pattern of compromises that results is unlikely to exhibit global normative coherence precisely because it is made up of the bargains struck in the face of disagreement between basic value systems. Accordingly, when they impose coherence on the body of law, judges unravel the very compromises that allowed Congress to act.

Such unraveling is pernicious in two ways. Looking back, disrespecting the compromises embedded in the law disrespects the citizens and representatives who were willing to set aside some of their own views in the name of working together as a society; looking forward, refusal to honor such compromises gums up the process that enables legislation to move forward in a pluralistic society, making future collective action harder to achieve. Coherence, then, is both unrealistic and unattractive when imposed on the corpus juris of a large, diverse, federal republic like the United States. Courts should abandon the quest for coherence in statutory interpretation. And rejection of the coherence ideal shifts the burden to those who would defend the Court's current supersession doctrine to put forward some persuasive justification not sounding in coherence. Absent such a justification, the divergence should be abandoned, and the two estranged types of supersession--repeal and displacement--should be governed by the doctrinal standard currently governing preemption.

The Article unfolds in five Parts. Part I unpacks the idea of coherence and fleshes out the coherence and fidelity models of supersession. Part II examines preemption. It briefly recounts how early preemption doctrine exhibited the desire for coherence, but argues that, as currently articulated, the doctrine does a reasonably good job of discerning and carrying out congressional intent. Part III turns to repeal, examining the ways in which it diverges from preemption and the role that coherence has played in this divergence. Part IV undertakes similar analysis of displacement. Part V turns to the theoretical case against the coherence model. An eclectic set of powerful theoretical considerations suggests that coherence is both inconsistent with fidelity and insufficiently attractive to justify a departure from fidelity. Accordingly, unless some other justification is forthcoming, all three types of congressional supersession should be governed by the set of doctrinal rules that govern preemption.

I. COHERENCE AND FIDELITY: TWO MODELS OF SUPERSESSION

The three Parts that follow this one will examine the Court's supersession doctrine in an effort to show that the divergence between the Court's three supersession standards is payable in large part to the tension between two models of how supersession should be determined: the coherence model and the fidelity model. But before diving into the cases, it will be helpful to have a clearer understanding of the two approaches I suggest we will find there. I begin by clarifying the precise conception of coherence relevant to this area of legal theory before describing both the model of supersession it inspires and its competitor, the fidelity model.

A. Understanding Coherence

One frequently encounters assertions that the law is coherent, or should be. Often these come in the repetition of familiar phrases, like "the law is a seamless web" (14) that should "speak with one voice." (15) While all recognize that the law as it is falls short of the ideal, the ideal itself is attractive to many. But what precisely is coherence in this sense?

One thing that coherence is not bears emphasis at the outset: coherence is not equivalent to logical consistency. (16) The ideal of a corpus juris that is free from logical inconsistencies is not merely attractive; some close-as-possible approximation of the ideal is demanded by the conceptual structure of law itself. This can be seen in two basic steps: first, much of the law is concerned with providing reasons for action. (17) And second, a logically inconsistent set of legal commands is incapable of guiding human action. It is no surprise, then, that Lon Fuller lists avoiding contradictions of this type as part of "the morality that makes law possible." (18) The word "coherence" is indeed sometimes used in something like this way, to import nothing more than intelligibility. But in the theoretically interesting sense, the ideal of coherence entails something more robust. But what more, precisely?

To get at this something more, it is helpful to distinguish two types of coherence. The first is epistemic coherence. (19) With this type of coherence, the claim is that the coherence of our system of beliefs is part (or all) of what justifies those beliefs. (20) With epistemic coherence, the "something more" than logical consistency is closely related to ideals of holism, (21) theoretical simplicity, (22) and mutual support. (23) Epistemic coherence has not figured prominently in legal theory, (24) so I mention it only to set it aside.

In legal theory, the type of coherence that is normally bandied about is normative coherence. (25) Here, coherence is not understood as part of what makes beliefs about the law true or justified, but rather as a desirable property that the body of legal rules should exhibit. Normative coherence is itself a value that consists in a certain relationship between other values. (26) But what relationship, precisely, besides logical consistency? Like epistemic coherence, normative coherence sees the body of law in a strongly holistic

way: the basic unit of justification is the corpus juris as a whole. (27) Unpacking the further content of normative coherence, however, proves complex.

First, we need to distinguish between means and ends. The law rarely sets forth naked ends: "Be just," or "conform all your actions to the rule that leads to the highest aggregate level of utility." Instead, the law in the main sets forth rules designed to guide behavior. (28) These rules, if justified, are justified as means to some valued end. This alone is enough to get us part of what normative coherence means: insofar as a set of legal rules is understood as a means of advancing a particular value, the set of rules as a whole should advance that value effectively and harmoniously, without working at cross-purposes. If tort law, for example, were understood as single-mindedly pursuing the end of economic efficiency, it would be incoherent in this sense for a jurisdiction to define its tort causes of action in terms that most effectively encourage efficient behavior but then adopt a bungled set of remedies that ultimately encourage inefficiency.

One understanding of coherence would stop here. It would deem the entire body of legal rules coherent only if the rules all worked together effectively and harmoniously to advance one master-value. (29) This would be to define normative coherence in such a way that it entails value monism, or the meta-ethical position that everything of value in the human experience is ultimately reducible to a single master-value ("utility" would seem to be a popular candidate). Some normative coherence theories in law are indeed monistic in this way; (30) but, on reflection, requiring value monism as a definitional matter is far too restrictive, since it risks excluding the most prominent coherence-based legal theory of the twentieth century: Dworkin's theory of Law as Integrity. (31)

The trajectory of Dworkin's thought arcs toward value monism, and his penultimate book reveals that he did indeed come to accept this position. (32) But certain crucial passages in Law's Empire suggest that Dworkin, at this stage, recognized that the human experience is comprised of irreducibly plural values, and that these values at times compete with one another. (33) We might conclude that the apparent value pluralism of Law's Empire is inconsistent with coherence, properly understood; or we might, instead, revise our conception of coherence to accommodate it. On this thinner but more plausible understanding of coherence, a set of legal rules is coherent either if they all function harmoniously as means to a single end or if they harmoniously answer to a set of irreducible but consistently ranked ends. (34)

This understanding of coherence helps illuminate assertions that the law should "speak with one voice" (35) or that judges should "make sense rather than nonsense out of the corpus juris" (36) If, say, a legal system's rules of contract law are best understood as advancing the value of economic efficiency at great expense to the value of equality in the distribution of material resources, but its system of tort rules can only be understood as embodying the opposite ranking, the corpus juris as a whole fails to "make sense" by schizophrenically speaking with multiple voices at once. If the national legislature passes a statute placing industries operating in the western half of the country under expensive regulations in furtherance of environmental values, but then goes on to pass legislation allowing industries in the eastern half of the country to pollute without limit, there seems no way to make the second statute "compatible with the surrounding body of law into which the provision must be integrated," (37) even if the two statutes are not logically inconsistent.

Coherence, so defined, seems a particularly attractive ideal for judges when we imagine the common law as their natural habitat; here, the "law as a seamless web" metaphor limps the least. But the coherence ideal has influenced the interpretation of positive law, as well. In particular, the ideal proved powerfully appealing to the judges who crafted our supersession doctrine. The following Section discusses the particulars of this model of supersession.

B. The Coherence Model

The norm-conflicts that give rise to the problem of supersession provide a great deal of grist for coherence's mill. A doctrine of supersession is necessary precisely because of the existence of bodies of law enacted by separate, hierarchically ranked authorities; but the very existence of these diverse bodies of law poses a stark threat to the coherence ideal, since the authorities who have promulgated them may act without any overall consistency of basic purpose. Understanding how courts might cope with this threat to coherence helps us to see why it is plausible that both the repeal and displacement frameworks are based on the coherence model despite the fact that they have diverged in opposite directions.

Conceptually, we can imagine at least three possible judicial strategies for defending coherence against the threat posed by separate authorities. First we might demand that judges assume that the two authorities are acting coherently, when interpreting their final, collective product. When the facts inconveniently get in the way of this assumption, this first approach would leave it to judges to catch as catch can, nudging the corpus juris toward coherence, law-by-law, to the extent that the limits of plausible interpretation allow. As we will see, this "benevolent fiction" strategy is an only slightly stylized version of the approach courts have taken for centuries in discerning one legislature's implied repeal of an earlier, jarring statute. (38)

Second, judges might attempt to preserve coherence through the use of sharp domain restrictions. Acknowledging that incoherence would inevitably result if more than one authority governed the same policy domain, this strategy would endeavor to ensure that such convergence never took place. This "separate spheres" strategy, or something like it, is the paradigm that influenced generations of judges attempting to erect a regime of "dual federalism" to govern federal-state relations. (39)

Finally, judges might allow that there exists some policy space where both authorities act concurrently but insist that where the superior authority has acted on some part of a well-defined subset of this space, the inferior one is automatically precluded from acting within the entire subset. This third, "automatic supersession" strategy has also been influential. Preemption doctrine pursued the strategy for a short time upon the failure of the second, separate spheres strategy, (40) and it is this third strategy that has come to govern displacement of federal common law. (41)

Though the pursuit of coherence has led the doctrine governing repeal and displacement in remarkably divergent directions, this divergence, I shall argue, is best understood as the result of the Court's tinkering with these different strategies. Before digging into the details of this argument, however, we should turn our attention to the other supersession model on offer: fidelity.

C. The Fidelity Model

The fidelity model begins by noticing what type of question supersession is: a question of statutory interpretation. This is clearest in cases of express supersession: where Congress says, "State (or common-law, or previous legislative) rule 'x' is hereby superseded," what Congress says goes. Moreover the courts' task is clearly one of statutory interpretation. (42) Similarly, where Congress, in express terms, communicates its intent to preserve a preexisting rule, the courts' role is merely the ordinary one of carrying out Congress's statutory instructions. But the water is muddier when Congress, infuriatingly, has remained entirely silent about the question.

Consider two approaches our jurisprudence might have taken to these silences, but ultimately did not. First, the Court might have held that Congress can only exercise its admitted authority to supersede inconsistent rules if it says so expressly. Silence means nothing. Second, courts might have read Congress's silence as an unfettered grant of discretion to courts to supersede inconsistent rules as they deem necessary. Silence means anything goes. Rather than reading Congress's silence in either of these ways, the Court has opted to interpret congressional silence according to the rules and practices of ordinary statutory interpretation. (43) This points the way toward the fidelity model, for it is widely, though not universally, believed that the proper judicial role in matters of statutory interpretation is that of Congress's faithful agent. To be sure, debate has long raged over the methodology by which Congress's will is most faithfully carried out. (44) But most parties to the debate share much the same conception of institutional role: Congress is the principal, courts the agents, and within constitutional bounds, the function of the courts is generally limited to carrying out the instructions of the principal. (45)

Viewing the judicial standards governing preemption, repeal, and displacement through the lens of fidelity rather than coherence reveals a different model of their interrelationship. All three doctrines seek to answer the same question--did Congress intend to supersede a prior legal rule--so they should seek to answer it in the same way--by faithfully interpreting congressional intent as expressed by text in context. Although it is not the purpose of this Article to wade into the battle over the correct methodology of statutory interpretation, the main contenders in the statutory interpretation wars should be able to unite around at least two relevant principles. First, Congress's silence can be meaningful. Just as in ordinary conversation we frequently communicate a great deal more than we explicitly say, (46) so too Congress often conveys just as much by what it does not say as by what it does. (47) Second, faithfully discerning congressional intent requires attentiveness to the particularity of what Congress has decided, as expressed by text in context. In Part V, I will adjudicate the dispute between the fidelity and coherence models by relying on an eclectic but robust set of theoretical insights--insights that have become central to the three dominant methodologies of statutory interpretation: intentionalism, textualism, and purposivism. (48) These insights indicate that Congress does not adopt whole schemes of coherent basic purpose in the abstract. Instead, it achieves specific, negotiated compromises between groups of our fellow citizens who have very different ideas of how best to collectively govern ourselves. True fidelity to congressional will entails discerning and applying these particular compromises, even if awkward or incoherent.

Commitment to the faithful agent theory is not universal. (49) And regardless of whether fidelity or its competitors has the better of the argument, even those holding up the fidelity end would likely concede that the Court's duty of fidelity to Congress is not entirely exclusive; in at least two important areas, our practice does seem to allow for judicial disregard of congressional will in favor of non-fidelity-based reasons. (50) The power of judicial review entails that courts will decline to faithfully enforce statutes that contravene their own sense of the limits placed by the Constitution on Congress's power. And the doctrine of stare decisis contemplates that, occasionally, a court should follow a previously settled interpretation of a statute even if that interpretation no longer accords with the court's best understanding of Congress's will.

But those faithful to the faithful agent theory might point out that even in these two areas, the tug of fidelity has been sufficiently strong that courts have worked to conceptualize both apparent departures as actually consistent with fidelity. Since Hamilton, the dominant defense of judicial review has argued that the Constitution reflects limits placed on congressional power by a higher principal: We the People. (51) And while some of the considerations associated with stare decisis, such as stability and predictability, do seem to be non-fidelity-based, courts often justify adherence to statutory precedents in fidelity-based terms, based on the idea that a judicial change of course would invade the legislative function, (52) or on notions of implied congressional consent. (53)

Moreover, those who would reject the faithful agent theory also seem to owe an important concession. While fidelity-based reasons may not be the only legitimate reasons for interpreting a statute one way or another, they are presumptively legitimate in a way that non-fidelity-based reasons are not. (54) The faithful agent theory is sufficiently entrenched in our constitutional structure and practice that adopting a particular interpretation of a statute because it accords best with congressional will requires no special justification; interpretation grounded in non-fidelity-based reasons does. It may well be that such justification is available, that important values--such as stability, (55) protecting deeply entrenched legal principles, (56) or keeping the positive law in line with modern sensibilities (57)--justify judicial creativity unconstrained by congressional will, or even contrary to it. The question as between the fidelity and coherence models of supersession, then, is twofold: first, whether coherence is actually consistent with fidelity; if not, second, whether coherence is sufficiently valuable to justify a departure from fidelity. In Part V, I will argue that coherence loses on both scores. But first, I must begin to redeem my assertion, repeated often in what has come so far, that the tension between these two models is actually responsible for the current, muddled state of modern supersession doctrine. The following three Parts take up that task.

II. PREEMPTION

A. The Development of Preemption
   The Constitution's Supremacy Clause provides that:

   This Constitution, and the Laws of the United States which shall be
   made in Pursuance thereof ... shall be the supreme Law of the Land;
   and the Judges in every State shall be bound thereby, any Thing in
   the constitution or Laws of any State to the Contrary
   notwithstanding. (58)


The Supremacy Clause uncontroversially--at least to most modern ears--gives Congress the power to trump or "preempt" state law, when acting within its constitutionally prescribed sphere. (59) But while the Constitution clearly provides for the supremacy of national law, it does little to clarify how precisely this skeletal relationship between federal and state authority should be fleshed out. Eventually, I will argue, the Court arrived at an approach largely consistent with the fidelity model, but the road to this destination was far from smooth. And while a variety of considerations influenced this area of law's development, it is important not to ignore the role of the coherence ideal.

Recall the three strategies we noted in Section I.B for defending coherence from the threat posed by the existence of multiple sovereigns with potentially overlapping authority. First, the Court could adopt the "benevolent fiction" that Congress and the states actually do work coherently together and attempt to nudge the state and federal bodies of law toward this ideal when those bodies fall short of coherence in practice. (60) Second, the Court might try to cordon the state and federal governments into "separate spheres," to prevent any incoherent intermingling of the two. (61) Third, the Court could pursue the "automatic supersession" strategy by withdrawing all state authority over a field of law as soon as Congress enters that field. (62)

As a method of structuring federal-state relations, the first strategy never got off the ground. (63) But the second "separate spheres" strategy was certainly attractive to the Marshall Court; (64) indeed, "dual federalism"--the conceptual paradigm that grew to govern the judiciary's understanding of the federal-state structure during the nineteenth century--is centrally about drawing such lines. (65) And one of the chief motivations behind this attempt to build a structure of dual federalism was undeniably the quest for coherence. (66) In case after case, early judges struggled to determine whether the constitutional scheme of federal-state relations could be structured against a baseline of exclusivity, with the zone of concurrent power narrowly confined. (67) The "battle royal" over the exclusivity of federal authority played out over the commerce power. (68) The exclusivity vel non of the commerce power remained up in the air until 1852, when Cooley v. Board of Warden (69) held that federal and state power over commerce was concurrent, except over those subjects "in their nature national" that "admit only of one uniform system, or plan of regulation." (70) Cooley thus marked the end of the Court's flirtation with the second approach to preserving coherence in the face of federalism; the attempt to strictly divide state and federal power into separate spheres had failed. But while Cooley's structure seemed conceptually tidy, further difficulties lurked beneath the surface. Within the zone of concurrent power, what relationship between federal and state authority prevails?

The Court avoided definitively answering this question for the next sixty years, but shortly after the turn of the century, the call of the coherence ideal proved too seductive for the Court to resist any longer. The penny dropped in the 1912 case Southern Railway Co. v. Reid,. (71) In a brief opinion, Justice McKenna held for the Court that a North Carolina statute requiring railroads to accept freight for transportation whenever tendered (72) was preempted by the Interstate Commerce Act, (73) which evinced a desire by Congress to "take[ ] control of the subject of rate making and charging," (74) imposing "affirmative duties upon the carriers which the State cannot even supplement." (75) In a series of cases over the next few years, (76) the Court consolidated this doctrine of "latent exclusivity," (77) effectively opting for the third approach to resolving the tension between coherence and federalism: automatic preemption of any state regulation touching a topic on which Congress has already acted. (78)

Latent exclusivity's flame burned brightly, but it was short-lived. Between 1912 and 1933, the Court struck down state law after state law on the basis that they "cover [ed] the same field" as congressional legislation, and "that which is not supreme must yield to that which is." (79) But in the 1933 case Mintz v. Baldwin, (80) the Court signaled a volte-face, declining to hold New York cattle inspection laws preempted by federal legislation on the same subject because "[t]he purpose of Congress to supersede or exclude state action against the ravages of the disease is not lightly to be inferred. The intention so to do must definitely and clearly appear." (81) A decade later, the Court consolidated this new approach in Rice v. Santa Fe Elevator Corp, (82) Key to the modern approach articulated in Rice is an emphasis on Congress's intent to preempt, in which intent is to be inferred from its silence by applying a series of rules of thumb that direct attention to the comprehensiveness of the federal law, the nature of the subject matter, and the degree of conflict. (83) Whether the modern framework is successful in its attempt to accurately capture Congress's intent is the question to which we now turn.

B. The Structure of Preemption: Faithfully Interpreting Congress's Silence

The basic framework of modern preemption doctrine has changed little since Rice. In the absence of an express preemption clause, the Court will infer preemption in two broad categories of cases: "field preemption," where "federal law so thoroughly occupies a legislative field 'as to make reasonable the inference that Congress left no room for the States to supplement it,"' (84) and "conflict preemption," where state law "actually conflicts with federal law." (85) The "conflict preemption" category is itself divided into two subcategories: "impossibility preemption" where "'compliance with both federal and state regulations is a physical impossibility,'" (86) and "obstacle preemption" where "state law 'stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress.'" (87) Finally, the Court occasionally rounds out its preemption analysis with a "presumption against preemption," based on the premise that "'the historic police powers of the States [are] not to be superseded by [a] Federal Act unless that was the clear and manifest purpose of Congress.'" (88)

Because implied preemption is a matter of interpreting what Congress has not said, the adequacy of the modern framework depends on the appropriate principles for interpreting silence. In ordinary conversation, we communicate information through our silences all the time. For example, imagine that my daughter asks me "can I have a cookie and a glass of milk?" To which I reply, "you may have a glass of milk." Though my response is entirely silent on the topic of cookies, if my daughter goes on to help herself to both the milk and the cookie, I would certainly take her to have done something that I impliedly forbade. Or, to give a classic example, imagine that X has written a letter of recommendation on behalf of one of her students, T for admission into a graduate program. The letter reads: "Fs attendance in my class was punctual, and his handwriting is exemplary." Would not Xs silence with respect to any further, more relevant qualifications be taken to imply their absence, so far as X knows? (89)

As modern language theory has shown, our everyday use of language allows us to communicate content through silence because speakers and listeners (or writers and readers) understand that, under ordinary circumstances, conversational partners follow certain cooperative principles that help them to economize on the information they explicitly convey. (90) Scholars in the field of pragmatics, the branch of linguistics which studies the impact of context on meaning, (91) have taught us a great deal about these principles over the last forty or fifty years. (92) Without getting too far into the weeds, (93) let me introduce one principle that is widely accepted: under ordinary conditions, we assume that speakers' contributions to a conversation are

as informative as the purpose of the conversation requires. (94) When a speaker appears to have violated this "informativeness principle" (95) on the level of what he has explicitly said, (96) we see if we can infer information from the speaker's silence that would make up (or account for) the information deficit. (97)

So, to return to my opening examples, the explicit content of both my response to my daughter ("you may have milk") and Xs recommendation (" Fis punctual and an elegant penman") fail the test of relevant informativeness, since the context called for something more (a response about the cookie, in the first case, and information more pertinent to Fs abilities as a student, in the second). Rather than make the unhappy assumption that the speaker is simply being uncooperative, an average listener will try to preserve the assumption that the speaker is following the informativeness principle by casting about for an alternative explanation. In both cases, the listener is likely to infer that the speaker declined to give an answer that would have been more completely informative ("... and you can have the cookie, too," or "Y, moreover, is also an incredibly bright, promising student") because the speaker was unwilling to do so.

Interpreting silence, then--at least in ordinary speech--is not a "venture into speculative unrealities;" (98) it is thoroughly commonplace. To be sure, applying Grice's insights to the legislative context is a complex task, and some have questioned whether the insights apply at all; (99) but a brief glance at legal interpretative practice shows that inferring meaning from congressional silence is also common and, often, uncontroversial. I will limit myself to just two examples. First, it is a well-established rule of legal interpretation that each word or clause in an authoritative legal text is to be given meaning, if possible; no part of the text should be rendered "mere surplusage." (100) This sensible rule of interpretation seems quite clearly to be based on something like the informativeness principle: if a potential interpretation of a statute would imply that Congress included some words or clauses out of pointless redundancy, that interpretation should be pro tanto disfavored. (101) A second venerable rule of construction is the canon expressio unius est exclusio alterius, (102) or "the expression of one thing implies the exclusion of other related things that were not expressed." Here again, courts appear to have felt their way to an application of the informativeness principle. Just as my daughter would sensibly take my inclusion of the milk but exclusion of the cookie as an implied prohibition, courts regularly apply the same rule of thumb to Congress. (103) These examples are sufficient to illustrate that inferring meaning from Congress's silence, based on some maxim along the lines of the informativeness principle that guides our everyday conversations, is a widespread--even mundane--part of statutory interpretation.

With the informativeness principle in hand as a guide to reliable inference from congressional silence, we can turn to an evaluation of the adequacy of the Court's modern test for implied preemption. The "impossibility" branch of conflict preemption proves easiest. Consider a hypothetical example of impossibility preemption: federal law requires individuals to maintain minimum levels of health insurance coverage; (104) imagine that Texas passes a state law forbidding any Texas resident from purchasing health coverage at the federally mandated level. Here, Texas would have effectively trumped the federal rule, and the Constitution itself gives us the default rule for this situation: the "Laws of the United States" apply "any Thing in the Constitution or Laws of any State to the Contrary notwithstanding." (105) While it is certainly possible to think that Congress could effectively reverse the rule of priority if it so desired, the intention to do so is sufficiently peculiar that we would expect a reasonably informative Congress to articulate it explicitly.

Obstacle preemption, the other branch of the Court's "conflict preemption" tree, is also largely consistent with appropriate principles of inference from silence, though the argument here is a bit more complex. One specific, well-known application of the informativeness principle relies on salient "scales of informativeness" to impliedly communicate information. (106) Suppose you ask me how many children I have, and I reply: "I have two children." You will generally take me to have implied that I have two and no more than two children, even though my response is, technically, consistent with having three children or more. (107) You will infer this because, if I do indeed have three children, the most relevantly informative response would have been for me to say so. By responding with a value lower down on the relevant scale (two children, rather than three or four), I imply--by way of the principle of informativeness--that higher values on the scale do not hold.

I have elsewhere argued at length that this type of inference can justify the bulk of the Court's obstacle preemption cases; (108) here, let me sketch two examples. Gade v. National Solid Wastes Management Ass'n (109) dealt with a conflict between an Occupational Safety and Health Administration (OSHA) regulation requiring hazardous waste workers to have received "a minimum of three days actual field experience," (110) and an Illinois requirement that a partially overlapping category of workers must have received the equivalent of 500 days' field experience before state licensure. (111) The Court held that the Illinois requirements were "impliedly pre-empted as in conflict with the full purposes and objectives" of the federal requirement. (112) Gade fits nicely with the principles for interpreting silence that we have been discussing. A requirement of three days' field experience implies that workers need not have more than three days' experience, since if Congress had wanted a higher requirement of, say, 500 days, it would have said so. Its use of the less stringent "three days" strongly implies that the more stringent "500 days" is not required, an implication that ran contrary to Illinois's law.

The conclusion that state law is preempted is not as mechanical as this example might seem to suggest, of course. Take Florida Lime & Avocado Growers, Inc. v. Paul, (113) a classic case where the Court declined to find preemption in a way that interestingly contrasts with Gade (114) In Florida Lime, the purported conflict was between a state law which prohibited the sale of avocados containing less than eight percent oil (115) and federal regulations which certified for sale some avocados that did not meet this requirement. (116) Structurally, this conflict is similar to the one in Gade, but Florida Lime declined to find preemption, and the main consideration cited by the Court is instructive: "[T]he maturity of avocados is a subject matter of the kind this Court has traditionally regarded as properly within the scope of state superintendence." (117) This traditionally local nature of the subject matter blocked any inference that the federal standards were meant to exclude parallel action by the states.

Florida Lime reminds us that the question whether a federal law's particular choices are meant to preempt parallel state choices is, like the interpretation of silence more generally, an irreducibly contextual inquiry that depends on an assortment of circumstantial cues. (118) These cues include the finely calibrated or exhaustive nature of the federal scheme and the legal subject matter on which it touches. (119) And the importance of these contextual factors shows us where to fit the two remaining pieces of the Court's preemption framework. As Florida Lime recognized, consideration of the national or local nature of the subject matter is an important guide to inferring preemptive intent from Congress's silence. The Court accounts for this consideration by applying "the assumption that the historic police powers of the States [are] not to be superseded ... unless that [is] the clear and manifest purpose of Congress." (120)

Moreover, we are already well on our way to seeing how the Court's category of field preemption fits within this approach, as well. Another reason we might have to infer that a federal scheme excludes parallel state action is the exhaustiveness of the scheme; if federal legislation constitutes a comprehensive, detailed, finely balanced treatment of a subject, the broad scope of the federal system might justify the inference that any additional layer of regulation at the state level would impermissibly derange the delicate array of choices made by Congress. The Court's modern preemption framework elevates this consideration of scope into a separate branch of implied preemption that it calls "field-preemption." (121)

Note the difference between the modern doctrinal framework--including what the Court calls field preemption--and the coherence-based latentexclusivity approach employed by the Court in the early twentieth century. (122) Under that earlier approach, all preemption was essentially field preemption, and it was automatic, prompted by the Court's desire to preserve as much coherence in the total body of rules as possible. Under the latent exclusivity approach, the federal regulations governing avocado sales in Florida Lime would almost certainly have resulted in preemption. Under the modern approach, the Court takes a potential congressional desire for coherence into account, but in a way that is particularistic, not across the board. While I will argue in Part V that there is cause for significant skepticism about Congress's interest in global coherence, it remains possible that the uniformity and exclusivity of a discrete federal scheme of regulation was itself part of the complex balance struck under the Capitol dome. When the best understanding of text in context indicates that this is so, courts can take Congress's affirmative desire for some measure of coherence into account, in this particularistic, nested way, while remaining within the confines of the fidelity model.

Accordingly, the Court's modern preemption jurisprudence is largely consistent with the fidelity model. When certain considerations--that, in the main, track the categories and presumptions the Court applies--are present, we are justified in inferring that Congress's scheme, though it didn't say so in so many words, preempts parallel regulation by the states.

III. REPEAL

Just as the Constitution makes federal legislation superior to the states, it is well established that Congress has plenary authority to supersede legislation passed by a previous Congress: leges posteriores priores contrarias abrogant. (123) Although repeal is conceptually similar to preemption in this respect, a look at the Court's implied repeal doctrine shows that the two standards are in fact importandy different. A typical formulation of the modern doctrine is as follows:
   It is, of course, a cardinal principle of statutory construction
   that repeals by implication are not favored. There are, however,
   two well-settled categories of repeals by implication--(1) where
   provisions in the two acts are in irreconcilable conflict, the
   later act to the extent of the conflict constitutes an implied
   repeal of the earlier one; and (2) if the later act covers the
   whole subject of the earlier one and is clearly intended as a
   substitute, it will operate similarly as a repeal of the earlier
   act. But, in either case, the intention of the legislature to
   repeal must be clear and manifest. (124)


It is easy to spot four initial similarities with preemption doctrine. First, both modern preemption and repeal are matters of discerning "the intention of the legislature." (125) Second, the Court's "irreconcilable conflict" category of implied repeal bears a striking kinship to the "impossibility" branch of its conflict preemption doctrine. (126) Next, the Court's second category of implied repeal calls to mind the category of field preemption. (127) And fourth, in both preemption and repeal cases, the Court purportedly applies a presumption disfavoring supersession.

But this final similarity brings us to the first of two important differences. While the Court often cites the "presumption against preemption" as a sort of rhetorical flourish, a look at the Court's preemption outcomes gives a different picture. A series of empirical studies stretching back over the last two decades indicate that the Burger, Rehnquist, and Roberts Courts have found preemption in roughly half of the cases to raise the issue. (128) And an important study of preemption in the lower federal courts found a "strong predominance of decisions preempting state or local regulations." (129) In other words, in the context of preemption the data bear out the impressionistic observation by most commentators that "[t]he Supreme Court's devotion to its presumptions [against preemption] ... can only be described as fickle." (130) Contrast the pattern of outcomes in the Court's implied repeal cases. According to one student of the doctrine, "[bjetween 1986 and 2003, the Court addressed the question of a potential implied repeal in at least twenty-five cases" but "concluded that an implied repeal had occurred in at most two of these cases." (131) One Justice's own count is even stingier. "We have not found any implied repeal of a statute since 1975" wrote Justice O'Connor in 2003, "[a]nd outside the antitrust context, we appear not to have found an implied repeal of a statute since 1917." (132) In the area of repeal, then, the presumption "seems to have evolved into a virtual rule against implied repeals." (133)

Moreover, the evident difference in strength between the presumptions in these two areas is reinforced by a second, doctrinal difference: repeal doctrine has no equivalent to obstacle preemption. Above, I argued that this category plays an important role in guiding inference of preemptive intent from congressional silence. (134) But repeal doctrine lacks any mechanism for detecting this type of implied intent to repeal. (135)

The Court's repeal doctrine, then, is considerably more robust than its preemption doctrine, screening out a dramatically higher percentage of claims that a provision has been superseded. And a look at the development of the doctrine indicates that the most prominent defense of the bias against implied repeals is grounded in considerations of coherence. The first recorded articulation of the presumption is found in Sir Edward Coke's report of Dr. Foster's Case, (136) decided in 1614. In discussing the best way to reconcile two statutory provisions, Coke supported a presumption against implied repeal by insisting that "Acts of Parliaments are established with ... gravity, wisdom and universal consent of the whole realm, for the advancement of the commonwealth." (137) Accordingly, they ought to be read as harmoniously as possible and be "maintained and supported with a benign and favourable construction." (138)

The coherence-based justification for the presumption against implied repeal remained prominent throughout the centuries (139) and is still the predominant justification advanced today. In their recent treatise on statutory interpretation, Justice Scalia and Professor Garner urge that later enactments should often be interpreted to "change the meaning that would otherwise be given to an earlier provision that is ambiguous," rather than effect a repeal of the prior provision, since "a law is to be construed as a whole." (140) Along the same lines, Professor David Shapiro grounds the canon against implied repeals in "the importance of reading a new statute against the legal landscape," thereby "recognizing the value of minimal disruption of existing arrangements." (141) And Professor Amanda Tyler, in an influential article, celebrates the presumption because of its potential to "aid the courts in building a coherent statutory web" by equipping "the judiciary with the tools required to fit a statute into 'the general fabric of the law.'" (142)

But perhaps the difference between the Court's preempdon and repeal doctrine, and the grounding of the latter in the coherence ideal, is best illustrated by looking at two exemplary cases. In Radzanower v. Touche Ross & Co., (143) the Court dealt with a clash between two venue provisions. (144) A special venue rule contained in the 1954 Securities Exchange Act provided that "[a]ny suit or action to enforce any liability or duty created by this [Act] ... may be brought in ... the district wherein the defendant is found or is an inhabitant or transacts business." (145) But a less generous venue statute, dating from 1875, applied specifically to national banks and provided that venue in a suit against such a defendant would lie only in "the district in which such association may be established." (146) In Radzanower, the plaintiff brought a securities suit against a national bank in a venue appropriate under the Securities Exchange Act's venue provision but inappropriate under the earlier rule.

The Court was thus faced with a conflict between an earlier, restrictive venue provision governing this type of party, and a later, expansive venue provision governing this type of case. It seems reasonably clear that this conflict is sufficiendy direct to result in supersession under an approach equivalent to the Court's modern preemption doctrine--if not as a case of "physical impossibility," than certainly as a case of "obstacle" repeal. In Radzanower, however, the Court invoked the "cardinal principle" disfavoring implied repeals (147) and found that "it is possible for the statutes to coexist." (148) It found the conflict insufficient "to indicate 'that Congress consciously abandoned its [prior] policy,'" (149) and held the earlier, narrower venue provision valid and controlling. (150) As the Court noted, "when two statutes are capable of co-existence, it is the duty of the courts ... to regard each as effective." (151)

As a second example, consider the deeply fractured decision in Branch v. Smith, (152) In that case, the Court struggled to come to terms with the interaction between two federal laws governing redistricting procedures. Due to the 2000 census, Mississippi lost one of its congressional seats, necessitating redistricting. (153) The Mississippi legislature, however, failed to pass a new districting plan, and, after several tangled legal machinations in state and federal court, it fell to the United States District Court for the Southern District of Mississippi to design a temporary redistricting plan to govern the 2002 congressional elections. (154) In doing so, however, the court faced seemingly contrary commands from two provisions of federal law. The older provision, 2 U.S.C. [section] 2a(c), provides that, when a state's congressional delegation is decreased in size, the entire delegation "shall be elected from the State at large," but only "[u]ntil a State is redistricted in the manner provided by the law thereof." (155) However, the more recent provision, 2 U.S.C. [section] 2c, provides in seemingly categorical terms that
   [i]n each State entitled ... to more than one Representative ...
   there shall be established by law a number of districts equal to
   the number of Representatives to which such State is so entided,
   and Representatives shall be elected only from districts so
   established, no district to elect more than one Representative.
   (156)


Section 2a(c), in other words, apparendy required that Mississippi's congressional delegation be elected at large in 2002, at least unless the state had been "redistricted in the manner provided by the law thereof," (157) while [section] 2c seemed to categorically bar any at-large election from taking place.

The district court opted to follow the dictates of [section] 2c, promulgating a redistricting plan that included single-member districts. (158) On appeal, the Supreme Court affirmed by a vote of 7-2, though no single rationale for affirmance garnered a majority. Writing for a plurality of four Justices, Justice Scalia acknowledged the force of the argument that the two provisions were in such irreconcilable conflict that the latter should be read as repealing the former. (159) However, he found the strength of the presumption against implied repeal too strong to resist and opted instead to cast about for a way of "harmonizing" the two provisions. (160) His solution was to read the phrase "[u]ntil [the] State is redistricted," which marks the end of [section] 2a(c)'s interim requirement of at-large elections, as applying to judicial, no less than legislative, redistricting. (161) On this reading, [section] 2a(c)'s requirement of at-large elections subsides once a redistricting plan--either legislatively or judicially crafted--is in place, and both the legislature and the courts are bound by [section] 2c's mandate of single-member districts when they are doing the crafting. While hardly the most natural interpretation, Justice Scalia found the awkward fit justified by "the most rudimentary rule of statutory construction ... that courts do not interpret statutes in isolation, but in the context of the corpus juris of which they are a part." (162)

Branch is again instructive because it seems hard not to believe that, had [section] 2a(c)'s requirement been found in a provision of Mississippi law, the Court would have concluded that [section] 2c preempts it, likely under the onerous and rarely-met impossibility standard. Interpreted honestly, [section] 2a(c) requires that anyone authoritatively determining how a state's congressional delegation "shall be elected" be governed by its requirement of at-large elections. Similarly, [section] 2c sweepingly requires that single-member districts "shall be established." The candid conclusion seems to be that the district court faced two contradictory legal commands--order an at-large election, but do not do that--and was physically incapable of complying with both. Under the Court's current preemption jurisprudence, a state version of [section] 2a (c) would almost certainly have to give way. (163) And accordingly, this situation, shorn of any "mystic overtones" (164) about the law "speak[ing] with one voice," (165) calls for implied repeal as well under any fidelity-grounded understanding of the doctrine.

IV. DISPLACEMENT

As with preemption and repeal, the authority of Congress to override judge-made federal common law is well established, (166) at least insofar as the judicially crafted norms are not constitutionally grounded. (167) While the "general common law" of Swift v. Tyson (168) fame was an assumed part of our legal system from the get-go, (169) the existence of a truly "federal" common law--authoritative because traceable to the sovereignty of the United States and superior over state law under the Supremacy Clause--sparked debate from the beginning (170) and was rejected, with respect to the common law of crimes, quite early on. (171) And once the general common law was dispatched by Erie Railroad Co. v. Tompkins, (172) the tension between the patches of truly "federal" common law that survived and our constitutional structure was even more obvious. (173) Recognizing these structural concerns, the Court has emphasized that the proper instances for exercise of federal common-law power are "few and restricted." (174) Moreover, this tension would become truly intolerable were Congress deprived of the ability "to set matters right" should the courts "go too far off the desired beam." (175) Accordingly, the Court has "always recognized that federal common law is 'subject to the paramount authority of Congress.'" (176)

Once again, it is important to note the structural similarity between displacement of federal common law, on the one hand, and preemption and repeal, on the other. In all three areas of supersession, we can now see, the structural authority relationship is the same: Congress has broad authority, within constitutional bounds, to supersede preexisting legal norms. And once again, it is also important to note that the standard that has come to govern displacement is different from the other two areas of supersession.

Courts will find that Congress has displaced preexisting federal common law whenever a federal statute "'speak[s] directly' to the question addressed by the common law." (177) This standard is indeterminate, depending on how the phrase "speaks directly" is cashed out. (178) In some cases, the Court has suggested that the standard is demanding, emphasizing the word " directly" by demanding that the statute in question has "spoken to a particular issue." (179) Statements to this effect are isolated and scattered, however, and in its most recent pronouncement on the issue, American Electric Power Co. v. Connecticut, (180) the Court suggested a much more forgiving standard. Because federal common law is structurally disfavored, " [l]egislative displacement of federal common law does not require the 'same sort of evidence of a clear and manifest [congressional] purpose' demanded for preemption of state law." (181) Indeed, in an important earlier case, the Court noted that "[s]ince the States are represented in Congress but not in the federal courts, the very concerns about displacing ... state law in the absence of clear intent actually suggest a willingness to find congressional displacement of federal common law." (182)

Read in this dominant way, the "speaks directly" test looks strikingly similar to a test for supersession we have already encountered: the early-nineteenth-century latent exclusivity version of preemption. (183) While the Court's motivation in crafting this watered-down test for displacement is no doubt based at least in part on its nagging doubts about the very legitimacy of federal common law, (184) the test's similarity to latent-exclusivity preemption suggests that the coherence ideal is also at work. Indeed, a look at two illustrative cases reveals that the standards are alike not only in demanding much less than modern preemption before finding supersession, but also in grounding this relaxed standard on coherence concerns. In the area of federal common law, in other words, Courts have settled on the third strategy described in Part I: attempting to preserve coherence in a domain of concurrent power by assuming that any intervention by Congress in the domain automatically excludes all attempts by the subordinate authority (here, federal courts) to take parallel, even complementary, action. (185)

In 2011, the Supreme Court made headlines by reversing the Second Circuit's effort to propel federal courts into the business of regulating greenhouse gas emissions through the common law of nuisance. (186) In American Electric Power Co. v. Connecticut, the Court held that Congress had taken the issue in hand by passing the Clean Air Act (CAA), leaving no room for a regime of judge-made public nuisance law regulating the same issue. (187) The Court thus closed off a method of regulating interstate pollution that dated back to 1906 (188) and that advocates of increased environmental regulation had increasingly come to embrace. (189) In determining that the CAA displaced the federal common law of nuisance governing air pollution, the Court focused on the CAA's delegation of authority over air pollution to the Environmental Protection Agency (EPA). (190) The EPA had yet to issue regulations governing greenhouse gas emissions, and Connecticut relied on that fact to argue against displacement. (191) Indeed, if the Court took an approach to displacement similar to its modern preemption jurisprudence, it is hard to see how, in the absence of any specific legislative or administrative treatment of the issue, the conflict would be direct enough to justify supersession. But the Court insisted that "the relevant question for purposes of displacement is 'whether the field has been occupied, not whether it has been occupied in a particular manner.'" (192) Indeed, the "critical point is that Congress delegated to EPA the decision whether and how to regulate carbon-dioxide emissions," and this "delegation is what displaces federal common law." (193) By entrusting the "complex balancing" involved in determining "[t]he appropriate amount of regulation in any particular greenhouse gas-producing sector" through "informed assessment of competing interests" to an "expert agency," (194) Congress had simply left no room for the judicial branch to jigger with the finely balanced congressional scheme by creating "a parallel track" (195) and "issuing ad hoc, case-by-case injunctions." (196)

American Electric Power's reasoning is suggestive of the close kinship between the Court's "speaks directly" test for displacement and the latent exclusivity form of preemption it flirted with during the early twentieth century. With both standards, supersession is nearly automatic, triggered at the lightest congressional touch. And both standards are based on a desire for coherence. A case from admiralty law, perhaps the hoariest enclave of federal common law, further illustrates the point.

The law governing wrongful death actions for the estates of those who die at sea is a baroquely intricate tapestry that interweaves strands of both statutory and common law. In general, the individuals who might die at sea fall into two categories: true seamen, who are "member[s] of a ship's company," (197) and others--predominandy longshoreman--who, though not part of any ship's crew, work at sea. The law regarding true seamen is what concerns us here. Several different sources give seamen a wrongful death action in a variety of circumstances. First, the Death on the High Seas Act (DOHSA) (198) provides a wrongful death action to any person whose death is caused by an action occurring on the "high seas," defined as "beyond a marine league from the shore of any State," (199) in two circumstances: death caused by negligence (200) and death caused by the "unseaworthiness" of the vessel (201)--a strict-liability-like duty owed by shipowners to true seamen and longshoremen alike. (202) For seamen who instead die within the territorial waters of a state, the Jones Act provides a wrongful death action, but only if the death was a result of the defendant's negligence. (203) This statutory patchwork thus fails to provide any death action sounding in unseaworthiness for a seaman whose death occurs within a state's territorial waters. In Miles v. Apex Marine Corp., the Court filled this gap by creating, as a matter of federal common law, a wrongful death action for true seamen who die within territorial waters as a result of the vessel's unseaworthiness. (204) But in Miles, the Court also faced the question of what remedy would be appropriate for this judge-made cause of action, and it is this part of the holding that is most interesting for our purposes. (205)

For actions under the Jones Act and DOHSA, the only available remedy is pecuniary damages--the traditional remedy in death actions at the time those statutes were passed. (206) Many more recent wrongful death actions granted by state law, however, allow recovery of nonpecuniary damages, (207) and the Court in Miles had to decide whether its common-law death action for unseaworthiness should adopt this modern, more liberal remedy. (208) Based on coherence-grounded concerns that should by now sound familiar, the Court declined to do so. "We no longer live in an era when seamen and their loved ones must look primarily to the courts as a source of substantive legal protection," Justice O'Connor wrote for the Court, and "an admiralty court must be vigilant not to overstep the well-considered boundaries imposed by federal legislation." (209) Though the Jones Act only "applies when a seaman has been killed as a result of negligence," the Court nevertheless felt that " [i]t would be inconsistent with our place in the constitutional scheme were we to sanction more expansive remedies in a judicially created cause of action in which liability is without fault than Congress has allowed in cases of death resulting from negligence." (210)

Like American Electric Power, Miles is instructive for two reasons. First, it is unlikely that the Court would have found supersession had it used an approach parallel to its modern preemption framework. The unseaworthiness-based death action at issue in that case was entirely a creature of federal common law, and it is hard to see how providing more generous damages for this action either directly conflicts or poses an obstacle to Congress's decision to limit the remedy for DOHSA and Jones Act actions to pecuniary damages. Second, Miles demonstrates the centrality of coherence-based concerns to the Court's displacement doctrine. In stark parallel to the latent exclusivity preemption cases from the early twentieth century, the Court in Miles declined to intrude on the "uniform plan of maritime tort law Congress created in DOHSA and the Jones Act." (211) Perhaps the only difference is that the Court traded a military metaphor for a nautical one. The "latent exclusivity" cases found automatic preemption where Congress "has occupied the field." In Miles, the Court noted: "We sail in occupied waters." (212)

V. COHERENCE AND FIDELITY IN STATUTORY INTERPRETATION

In Part I, I described two models that might govern the question of supersession: the coherence model and the fidelity model. The balance of the Article, up until now, has worked to establish two things: first, that the Court's preemption, repeal, and displacement doctrines are importantly different from one another, and second, that this divergence is largely traceable to the tension between these two models. In this final Part, I adjudicate the dispute between coherence and fidelity; I ultimately declare fidelity the winner.

In Part I's discussion of the fidelity model, (213) I came to the following conclusion, important enough to bear repetition here. While the claim that fidelity-based reasons are the exclusively legitimate reasons for interpreting a statute one way rather than another is a strong claim which I do not mean to make, the faithful agent theory does seem to have one important advantage over alternative theories of the judicial role. The legitimacy of fidelity-based interpretation is deeply rooted enough to be safely assumed; the legitimacy of non-fidelity-based interpretation must be argued for. In light of this presumptive validity of fidelity-based interpretation, the question whether coherence is a sufficient justification for the divergence of repeal and displacement from the (fidelity-based) preemption framework is best answered in two steps.

First, it may be that we can reasonably understand the judicial commitment to coherence as consistent with--perhaps even the best way of furthering--fidelity, as Dworkin, for one, argued with great feeling. (214) We can imagine two arguments along these lines: a direct argument and an indirect argument. Most directly, if all congressionally enacted law were in fact coherent, courts would accurately track congressional will by interpreting the law as coherent. (215) Alternatively, and indirectly, even if the body of congressionally enacted law is not in fact coherent, we might think that Congress possesses a second-order desire for coherence.

Finally, if we cannot understand coherence and fidelity as consistent, we must face a more nakedly normative question: is coherence sufficiently attractive to justify non-fidelity-based interpretation? I consider each of these arguments in turn.

A. Coherence as Fidelity

1. The Direct Argument

Whatever the attractiveness of normative coherence in the abstract, several diverse strands of modern legal theory unite in indicating that we should not expect the output of our political branches to be coherent. Congress has the right to be incoherent, and we should expect the right to be frequently exercised. Moreover, while at least one of the reasons we should expect legislative incoherence has, like the word itself, a negative connotation, other reasons are traceable to the deliberate choices embedded in our constitutional structure, choices which themselves further important and attractive values.

All of the reasons to doubt that the legislature's total output will be coherent are ultimately bottomed on what John Rawls called "the fact of reasonable pluralism," that a diversity of "conflicting and irreconcilable," yet still "reasonable comprehensive religious, philosophical, and moral doctrines ... is a permanent feature of the public culture of democracy." (216) This fundamental disagreement over correct value choices is part of what Jeremy Waldron calls the circumstances of politics: "[W]e find ourselves living and acting alongside those with whom we do not share a view about justice, rights or political morality." (217) Nevertheless, citizens also frequently desire to "act together in some common concern even though they disagree among themselves what exactly is to be done." (218) In our constitutional system, the most prominent way of resolving such disagreement is to elect representatives who then assemble and choose an authoritative course of action by majority vote. (219) The body of legislation deserves our respect because of "its emergence, under specified procedures, as a 'unum' out of a plurality of ideas, concerns, and proposals." (220) But a closer look at this process by which our legislation is crafted provides cause for skepticism about the coherence ideal.

The most discouraging form of skepticism finds voice in Arrovian social choice theory. According to Kenneth Arrow's impossibility theorem, any method of collective choice that meets certain plausible constraints (221) will be unable to guarantee a transitive collective ranking of preferences. (222) A simple example helps to illustrate: imagine that three legislators, A, B, and C, are tasked with choosing between three options, x, y, and z, by majority vote. A prefers x to y to z, B prefers y to z to x, and C prefers z to x to y. The legislature first votes x against y, and x prevails by a 2-1 vote. Next, they vote the victor, x, against z; z defeats x 2-1. But if the legislature then decides to resurrect y and see how it fares against z, y beats z 2-1. So while the body as a whole prefers z to x and x to y, it also prefers y to z. The collective ranking of preferences is intransitive.

The implications of this intransitivity for the coherence ideal are fatal. Normative coherence demands that the body of legal rules to which it applies function harmoniously to advance a single, consistent ranking of values. (223) The ranking, to be a ranking, must be transitive. (224) Arrow's theorem shows that a legislature is perpetually at risk of failing to meet this constraint. In our simplified example above, the legislature's choice between x, y, and z will be determined by the agenda: whoever dictates the order in which the votes occur and the point at which voting stops in fact dictates the legislature's choice. (225) This means that a series of legislative choices between x, y, and z made by the same legislators, but following different procedural pathways, will result in different, inconsistent rankings of the three values. Where these conditions hold, coherence is conceptually impossible.

Do the conditions contemplated by Arrow's theorem hold often? There is some debate, here, prompted by the fact that we simply do not see legislatures exhibiting this type of irrationality all that frequently. (226) One condition necessary for Arrow's theorem to apply--the existence of different legislators holding divergent, inconsistent value rankings--we can be confident holds, given the circumstances of politics. (227) But another condition--that the legislators' rankings line up in the particular array exhibited by our simplified example--may hold less frequently. (228) And the opaque, reticular legislative structures which legislation must navigate might work fairly well at inducing equilibrium. (229) So while Arrow's theorem should certainly put us on our guard for pockets of incoherence, perhaps there is no reason to expect the widespread irrationality and chaos that Arrovian social choice theory at first might lead us to fear.

Arrow's impossibility theorem poses such a stark challenge to the coherence ideal because it suggests that even where legislators holding different fundamental value rankings are deeply committed to collective action in pursuit of whatever values are selected by majority vote, the conceptual nature of collective choice makes impossible any guarantee that the resulting value choices will be coherent. Arrovian social choice theory gives off a disheartening aura, or at least a certain "'eat your spinach' quality." (230) If the widespread occurrence of Arrovian cycling were our sole reason for concluding that fidelity to the legislature is inconsistent with the coherence ideal, this might do no more than incline us to re-think fidelity. (231) But even where the conditions contemplated by Arrow's theorem do not hold, and transitive collective choices are accordingly possible, there are far more benign reasons for concluding that the legislative output will not exhibit coherence.

To see why, return to the fact of pluralism: our society is composed of many diverse citizens and groups with divergent rankings of basic values. If the representative system is doing its job, this implies that the legislature, too, will be comprised of members with differing and inconsistent value rankings. Even where Arrovian disequilibrium does not hold and these inconsistent preferences are arrayed in such a way as to permit a stable, all-or-nothing majority vote on a controversial proposal, the only possible or sensible way to proceed will probably not be in this winner-take-all fashion. Instead, one of the most important contributions of positive political theory--a line of political science scholarship concerned with the strategic interaction of political institutions and actors--is to remind us that the proponents of legislation will very often have to make considerable concessions to the other side in order to assemble a coalition sufficiently strong to pass the legislation. (232) These concessions may simply take the form of a diminution in the strength of the proposed legislation (lighter restrictions on the emission of pollutants than the ardent environmentalists would prefer, say). In that case, the final compromise bill may retain a significant degree of coherence. However, compromise often takes alternative forms far more threatening to the coherence ideal.

In order to attract sufficient legislators into the coalition or to get the proposal through certain veto-gates, the proponents of the legislation might have to include provisions that are quite jarring. (233) For example, Dean Rodriguez and Professor Weingast, in their insightful account of the Civil Rights Act of 1964, chronicle how the necessity of gaining the support of northern congressional Republicans led proponents of the civil rights legislation to include amendments that, inter alia, exempted de facto school segregation in the North from the scope of the bill, exempted union seniority systems, and exempted employers of seasonal agricultural workers. (234) Similarly, ardent supporters of legislation might have to fold unrelated goodies for influential special interest groups into the final package to attract sufficient support (235) (the famed "'Cornhusker Kickback'" proposed during the debate over the Affordable Care Act leaps to mind). (236) And the inclusion of exceptions or kickbacks is likely to render even (perhaps especially) the most ambitious and public-regarding legislation ultimately incoherent, in the sense discussed above, in manifold ways. (237)

An alternative way of achieving compromise in the face of dissensus is through the use of what Cass Sunstein has called "incompletely theorized agreements." (238) Since " [l]egal institutions are composed of many people" who must choose a collective course of action in the face of "sharp and often intractable disagreements on basic principle," (239)--because, in other words, of the circumstances of politics and the fact of reasonable pluralism (240)--those institutions will frequently reach agreements that are "incompletely theorized" in one of three senses. First, parties to an agreement may accept some general principle but differ on its application in concrete cases. (241) Second, groups may agree on a "mid-level principle" but differ on both the "general theory that accounts for it" and the "outcomes in particular cases." (242) And third, the parties may agree on "particular outcomes," but reach no agreement on higher-level principles or theories. (243)

Sunstein's incompletely theorized agreements are tightly related to a strategy that legislative observers identified long ago: choosing to employ vague terminology or "essentially contested concepts" (244) to reach a bargain at a situated, particular level of generality while avoiding the need to form a comprehensive agreement that runs all the way up to a definitive ranking of basic values and all the way down to a specification of all particular outcomes. (245) To faithfully understand a series of legislative compromises that may well reflect under-determinate or incompletely theorized agreements we must attend to the particularity of the agreements that were reached, without attempting to superimpose on the set of agreements a coherent, collective ranking of values that simply did not exist. (246)

Moreover, our discussion of all of these ways in which legislative compromise undercuts the coherence ideal has been operating on the simplifying assumption that the relevant legal rules were all passed by a single, static legislature. But, of course, the corpus juris as a whole is not the product of a single legislature. The work product of any single group of lawmakers makes up only a tiny percentage of the body of legal rules as a whole. The bulk of this larger body was passed by political actors who long ago "folded up [their] papers and joined [their] friends at the country club or in the cemetery." (247) The body of federal law is composed largely of sediment. How realistic is it to expect to find harmony of basic purpose among the legal rules crafted by this temporally diverse group of lawmakers? (248)

Finally, while the discussion so far has assumed that the only obstacle to global normative coherence is the existence of divergent rankings of basic values, an even more fundamental challenge to the coherence ideal questions the intelligibility of ranking basic values to begin with. Recent scholarship in meta-ethics has raised the possibility that our most basic values may be not only irreducibly plural, but incommensurable as well. Two values are incommensurable if neither is "better than the other" and they are not of equal value. (249) Many of the choices we make in life involve ranking two different options against each other or calculating which option is best overall. But with incommensurable values, the concepts of ranking and calculating do not seem to intelligibly apply. What sense is there in trying "to sum together the size of this page, the number six, and the mass of this book," (250) or endeavoring to determine "whether a particular line is longer than a particular rock is heavy." (251) The incommensurability thesis holds that the same failure of ranking and calculation applies to the most basic human values, such as friendship, beauty, and truth. (252)

If the thesis is true, the coherence ideal fails on an even deeper level, because it mistakes the most fundamental nature of our moral experience. The coherence ideal requires the existence of one master value or one consistent ranking of irreducibly plural values. (253) But if the incommensurability thesis holds, this consistent ranking is conceptually impossible. Indeed, looked at in the aggregate, a series of choices between incommensurable values will almost certainly be intransitive. (254) Accordingly, the incommensurability thesis, if true, is deeply inconsistent with the coherence ideal. Imposing coherence on the total set of our particular choices of one legal rule over another--which is to say, imposing a single, consistent ranking on our basic values--does great violence to the richness and diversity of our moral experience. (255) For all of these reasons, the direct argument for coherence fails.

2. The Indirect Argument

We have seen that a commitment to coherence is not grounded in fidelity to congressional will in the sense of accurately tracking a normative coherence that the body of legal rules actually exhibits. But while reasons grounded in the circumstances of politics might prevent Congress from enacting a coherent body of law, it is still possible that Congress holds and acts on a second-order desire that the corpus juris be as coherent as possible. If Congress indeed possesses such a desire for coherence, this might underwrite an indirect--but still fidelity-based--argument in favor of coherence, along the lines that Congress's silence on the question of how far its enacted rules are meant to supersede existing legal rules ought to be interpreted consistently with its second-order desire for coherence.

One way of cashing out this argument relies on considerations of knowledge. We might expect, this line goes, to find a more resolute desire by Congress for coherence in the body of positive federal law than any similar taste for coherence in the combined federal-and-fifty-state corpus juris, explaining at least the divergence between preemption and repeal doctrine. One might suspect this disparity for three reasons having to do with Congress's knowledge. First, when Congress enacts new law we might expect it to have federal law more intimately in view than state law for the simple reason that there is only one body of federal law, as opposed to fifty separate bodies of state law. Second, we might hope that Congress would have sharper knowledge of federal statutory law because Congress--as an institution--is, after all, responsible for creating the body of federal statutes. Finally, the universe of laws that might be impliedly repealed is temporally closed:, only those valid laws enacted prior to the current one are possible targets of repeal. But preemption is temporally symmetric, state laws enacted after a federal law can be preempted just as surely as state laws that predate that federal provision, and there is no way Congress can know the content of these future state laws. So there are a variety of reasons to suspect that Congress knows a great deal more about the details of federal law than state law, and if Congress knows the content of existing federal laws when it enacts new law, this tack concludes, we might appropriately conclude from its failure to affirmatively supersede them that it desires that they remain on the books--neatly explaining both the strong presumption against implied repeals and the much weaker showing required to establish preemption. (256)

Whether Congress possesses more intricate knowledge of federal than state law is largely an empirical question, (257) but the reasons just given for suspecting that this is the case strike me as certainly plausible. But this likely disparity in Congress's knowledge is not itself sufficient to redeem the coherence ideal as a justification for current supersession doctrine. Recall that--since we are here trying to interpret coherence as consistent with fidelity--the question is whether Congress's silence on the question of repeal is best interpreted, in light of its presumed knowledge of preexisting federal law, as a decision not to repeal any provision of that law. But the fact that Congress knows of the existence of some prior, potentially clashing provision of law, but does not express a desire to supersede it, does not tell us that Congress in fact meant to save it. After all, if Congress was aware of this earlier provision and meant to save it, why didn't it affirmatively express that desire? (258) What the argument needs, in other words, is some reason to think Congress was operating against one baseline rather than the other. A second-order desire by Congress for coherence would supply the missing premise. But this just shows that the success of the argument hangs on showing that Congress in fact desires coherence in this way; the argument from disparate knowledge of federal and state law turns out to be makeweight.

What, then, about this suggested second-order desire for coherence? We should not attribute such a desire to Congress, I think, for reasons related to those already canvassed. Legislators surely understand the give-and-take nature of politics: that progress comes through compromise and that today's loyal opposition might well be tomorrow's governing coalition. But, as we saw above, by pursuing the unitary ranking of values thought to be embodied in the prevailing thrust of a piece of legislation at the expense of the competing values held by those who acquiesced in the legislation only after receiving critical concessions, the coherence ideal threatens to undo the very compromises that make legislation possible. It is implausible that senators and representatives, after fighting out the awkward details of a compromise piece of legislation, would consciously embrace an ideal that threatens to undue all past compromises of this nature (as well as the similar compromises hammered out by other, subordinate legislatures; and as well as this very set of compromises, in light of future, yet to be envisioned enactments). The coherence ideal, moreover, is wildly unpredictable: who knows what corner of the law will have to be revised to enhance global coherence; and who can guess when today's compromises will be undone by some future, seemingly unrelated enactment? (259)

Moreover, it is worth bearing in mind that the costs of a coherence-based approach are not just marginal. (260) At issue is not just a power by courts to read the breathtaking generalities in Congress's statutes as implicit delegations of law-elaborating authority and to exercise that authority in a way that pursues coherence. Nothing in this Article denies that courts have significant creative authority to flesh out the blurrier contours of congressional legislation and that they can appropriately pursue coherence when doing so. What is at issue here, instead, is a power to rely on a purported congressional taste for coherence as a reason to forsake what would otherwise be the best reading of congressional intent. In Radzanower, for example, this view would use Congress's purported desire for coherence to read Congress's failure to specify in the Securities Exchange Act that it meant to supersede the earlier, narrow venue provision to conclude that it meant to preserve that earlier law, despite the fact that the earlier provision quite clearly frustrated the operation of the Securities Act. It simply does not seem plausible that Congress would find attractive an ideal that threatened to frustrate its specific value choices in this way.

Accordingly, the indirect argument for coherence falls for much the same reasons as the direct argument. Both arguments in favor of the coherence ideal fail to take account of the particularity of our legal settlements. One of the most important insights of modern legal theory--an insight that, I have argued, emerges from several distinct strands of thought, including social choice theory, positive political theory, and meta-ethics--is that our legislatures do not adopt policies in the abstract. Instead, they reach particular, situated, negotiated settlements; and the lines these settlements draw are frequently awkward, complex, and highly incoherent. Above, these considerations indicated that the corpus juris is unlikely to be globally coherent; below, I will draw on the same considerations to urge that the coherence ideal is normatively unattractive. These considerations also suggest that Congress is unlikely to possess a second-order desire for coherence, once its destabilizing implications are appreciated.

B. Coherence Against Fidelity

Coherence, then, is ultimately irreconcilable with fidelity. Before we can nail the coffin shut, however, we must face one last argument. In Part I, I noted that courts do occasionally adopt interpretations that are openly grounded in non-fidelity-based reasons. Rather than laying out a case against all such non-fidelity-based interpretations, I settled, there, for a weaker claim: fidelity-based reasons are favored in a way that non-fidelity-based reasons are not, for an interpretation grounded in fidelity is presumptively legitimate, while an interpretation grounded in non-fidelity-based reasons requires special justification. We have seen in the previous Section that coherence is indeed a non-fidelity-based reason. The last question we must face is whether it is a sufficiently attractive one to justify a limited departure from fidelity.

This final question is an important one, and the considerations already discussed in this Part serve to sharpen it. If the legislative process is as messy as I have suggested, that might seem to underscore the wisdom of relying on judges to tidy up the corpus juris by imposing coherence on the back end. (261) Drawing this conclusion from the role of compromise in the legislative process, however, fails to appreciate the value of such compromises and the importance of protecting them. The coherence ideal profoundly fails to describe our legal system for reasons that, potential Arrovian cycling aside, are traceable to some of the most attractive features of that system. The existence of divergent value systems competing for prominence in the legislature is a result of our desire to act together in the face of reasonable disagreement. We do so by adopting discrete compromises that allow us to proceed partially without agreeing wholly. While some compromises are covenants with death or agreements with hell, (262) the phenomenon of compromise, more generally and mundanely, is deeply valuable for the same reasons that acting together is valuable. Indeed, one of the most important ways our constitutional design ensures domestic tranquility, promotes the general welfare, and secures the blessings of liberty is by encouraging compromise in the face of disagreement. (263)

But the tools of compromise frequently and ineluctably lead to incoherence--for the very reasons that they work. When we meditate on the compromises necessary to achieve the accomplishments of our past, we often do not see further than the warts. But we focus on what we could not accomplish at the risk of forgetting what we did; the 1964 Civil Rights Act, and all the good that it made possible, was passed only because, on some issues, its ardent supporters were willing to accept half of a loaf. (264) The same could be said for any of our historic legislative achievements. As Waldron puts it, "[i]n the United States, in Western Europe, and in all other democracies, every single step that has been taken by legislatures towards making society safer, more civilized, and more just has been taken against a background of disagreement...." (265) Indeed, to the extent that the incommensurability thesis is true, our inability as citizens to converge on a single value ranking is not to be lamented, but rather celebrated as a necessary part of our complex moral experience.

Global normative coherence is unattractive because it disserves this value of compromise, in two ways. First, imposing global coherence on a corpus juris that is actually comprised of many discrete compromises between values simply undoes many of those compromises, giving one side a whole loaf when it was forced to settle for half. But "respect for legislation is in part the tribute we should pay to the achievement of concerted, co-operative, co-ordinated, or collective action in the circumstances of modem life," (266) and by undoing past compromises in this way, the coherence ideal becomes unattractive for the very reason that compromise is attractive.

Second, by disrespecting past compromises, coherence makes future compromise--and, therefore, future collective action in the face of disagreement--more difficult. Consider the Court's super-strong canon against implied repeals. By straining to "harmonize" each new enactment with extant, inconsistent statutory provisions, the Court effectively gives the supporters of legislation a far less effective statute than they bargained for, making future, similar bargains less valuable. And though displacement doctrine strains in the opposite direction, it has the same ultimate effect. Because a statute reflecting a particularistic compromise between competing values is given a ripple effect on surrounding common-law rules that neither side bargained to disturb, the limitations in statutory scope that moderates may have received as the price of their support suddenly become much less valuable, making them less likely to return to the bargaining table. The coherence ideal, then, not only disrespects past compromises; it gums up the wheels of our political system by making future compromises less likely. (267)

CONCLUSION

Coherence is a value of undeniable merit in many contexts. It is important to our judgments of aesthetic merit in literature and of explanatory success in science; it is central as well to our vision of the appropriate judicial role in the interpretation and growth of the common law. But this Article has argued that it is an inappropriate ideal in the context of modern statutory interpretation. Pursuit of coherence cannot be understood as consistent with the courts' presumptive role, when interpreting statutes, as Congress's faithful agents, and it is insufficiently valuable to justify a departure from fidelity. The citizens of a modern pluralistic society like ours face the challenge of acting together in the face of deep disagreement over basic value choices; they surmount this challenge by adopting a series of specific compromises, compromises which are incoherent precisely because they are compromises. By unraveling these compromises in the name of coherence, judges fail to pay the respect they owe to their fellow citizens for finding a way to act together when separated by so much; and in so doing, judges make successful social action less likely downstream by signaling to future parties at the negotiating table that their compromises may not be honored. Because it in this way fails to appreciate the central importance of compromise to modern legislation, coherence should be rejected as an ideal in statutory interpretation.

I have focused on spinning out one implication of the failure of the coherence ideal. Preemption of state law, repeal of prior legislation, and displacement of federal common law all govern circumstances in which a higher source of law might be understood to have superseded an inferior one; but while modern preemption doctrine fits rather well with the Court's role as Congress's faithful agent, pursuit of coherence has motivated the Court to articulate standards for repeal and displacement that depart from the preemption paradigm. Given the failure of coherence, unless some justification other than coherence is forthcoming, the standards governing displacement and repeal should be realigned according to the fidelity model, along the lines of our current preemption doctrine.

The rejection of coherence as an appropriate value in statutory interpretation promises to reveal other important implications. I note two such implications, here, though I cannot explore them in any depth. First, Justice Scalia has been the strongest modern proponent of the idea that, in interpreting statutory language, courts have a duty to "construe [an ambiguous term] to contain that permissible meaning which fits most logically and comfortably into the body of both previously and subsequently enacted law." (268) This modern incarnation of the classic in pari materia canon (269) is clearly based on the coherence ideal, and would seem to founder for the same reason as the Court's muddled supersession doctrine unless it is justified by other sufficient reasons. Second, the inconsistency of coherence with fidelity severely undermines an old but famous and surprisingly resilient idea: Dean Pound's suggestion that courts should "receive [a statute] fully into the body of the law to be reasoned from by analogy the same as any other rule of law." (270) Though I cannot pursue the point to its conclusion here, this approach to statutory interpretation would also seem to pay too little attention to the particularity of the compromises forged by Congress.

In an ideal world, perhaps--or at least one vision of an ideal world--we would all agree on the one true ranking of basic values, and, having seen the "common good" face to face, we might sensibly expect our total set of legal rules to harmoniously advance that common good. But that world is not our world. We perceive the common good darkly, as through a glass, and we disagree with one another over its true shape. Our legislatures proceed in the face of this disagreement only by hammering out particular compromises, one at a time, and true fidelity to congressional will means taking Congress as we find it.

(1) Henry M. Hart, Jr. & Albert M. Sacks, The Legal Process 159 (William N. Eskridge, Jr. & Philip P. Frickey eds., 1994) (quoting Henry M. Hart, Jr., The Relations Between State and Federal Law, 54 Colum. L. Rev. 489, 489 (1954)). On Hart and Sacks's Legal Process materials as encapsulating postwar American legal theory, see William D. Popkin, Statutes in Court 147-49 (1999); William N. Eskridge, Jr. & Philip P. Frickey, Legislation Scholarship and Pedagogy in the Post-Legal Process Era, 48 U. Pitt. L. Rev. 691, 694-95 (1987).

(2) Hart & Sacks, supra note 1, at 159.

(3) Id. at 467; see also id. at 515 (criticizing a decision as "destroy[ing] the harmony of the underlying... law in this general area"); id. at 777 ("To what extent should [a legislator] assume an obligation to try to make the provisions [she does] enact rationally consistent with established doctrines in related fields of law with which [she does] not deal?").

(4) Id. at 97.

(5) Id. at 147.

(6) Id. at 1209 (quoting id. at 148).

(7) William N. Eskridge, Jr. & Philip P. Frickey, An Historical and Critical Introduction, in Hart & Sacks, supra note 1, at cxvii; see also Laura Kalman, The Strange Career of Legal Liberalism 66-67, 91-93 (1996); Vincent A. Wellman, Dworkin and the Legal Process Tradition: The Legacy of Hart & Sacks, 29 Ariz. L. Rev. 413 (1987).

(8) Ronald Dworkin, Law's Empire 225 (1986).

(9) The view in the text--that Scalia's textualism constitutes a root-and-branch rejection of legal process theory--is the conventional one. See John F. Manning, What Divides Textualists from Purposivistsf, 106 Colum. L. Rev. 70, 73 (2006) (describing textualism as "challeng[ing] the prevailing judicial orthodoxy" that largely tracked Hart and Sacks); Jonathan T. Molot, The Rise and Fall of Textualism, 106 Colum. L. Rev. 1, 25-26 (2006) (describing textualism as a reaction against legal process theory). In many ways, however, the textualist theory articulated by Justice Scalia has remained within the basic presuppositions of legal process theory. See Eskridge & Frickey, supra note 7, at cxxviii n.332 (noting similarities).

(10) W. Va. Univ. Hosps., Inc. v. Casey, 499 U.S. 83, 100-01 (1991).

(11) This domain is comprised of rules for resolving conflicts between two lawmakers (or, in the case of repeal, two instantiations of the notionally same lawmaker), one of which is more authoritative than the other. Another body of rules, for resolving conflicts between equally authoritative lawmakers, comprises the complex set of doctrines known as conflicts of laws. This Article will not be concerned with this second set of rules.

(12) A note on terminology--as is probably, by now, apparent, terms to describe congressional override of other legal norms abound. While "repeal" is quite consistendy used to refer only to legislative overrides of previous legislation, the Court--and, I hasten to add, scholars--have traditionally used "preempt" and "displace" interchangeably, with profligate abandon. Quite recently, the practice of reserving "preempt" for the federal override of state law and "displace" for the congressional override of federal common law has begun to gel. I enthusiastically follow this practice, in the cause of terminological clarity, and use "repeal," "preempt," and "displace" to refer to the three distinct types of override just indicated. When referring to the phenomenon of override generally, I join the ranks of the profligate: "override," "supersede," "abrogate," and "trump" are used interchangeably throughout.

(13) John Rawls, Political Liberalism 36 (1993).

(14) See, e.g., Fourth Amendment--Consent Search Doctrine--Co-occupant Refusal to Consent--Leading Cases, 120 Harv. L. Rev. 125, 173 (2006). The phrase's origins go back to F.W. Maitland, A Prologue to a History of English Law, 14 L.Q. Rev. 13, 13 (1898) ("Such is the unity of all history that any one who endeavours to tell a piece of it must feel that his first sentence tears a seamless web.").

(15) See, e.g., Dworkin, supra note 8, at 165; Hart & Sacks, supra note 1, at 159.

(16) See Ken Kress, Coherence and Formalism, 16 Harv. J.L. & Pub. Pol'y 639, 662 (1993) ("Consistency at a time is necessary for normative coherence, but is not sufficient for it." (emphasis omitted)).

(17) See Lon L. Fuller, The Morality of Law 74 (1964) (defining the law as "the enterprise of subjecting human conduct to the governance of rules" (internal quotation marks omitted)); H.L.A. Hart, The Concept of Law 38-42 (2d ed. 1994) (describing law's function as "a means of social control").

(18) Fuller, supra note 17, at 65-70.

(19) See Joseph Raz, Ethics in the Public Domain 277-79 (1994).

(20) Id. Alternatively and relatedly, coherence might be thought part of what makes a set of propositions true, because truth itself is defined in terms of coherence. We might, therefore, distinguish further between a coherence theory of knowledge and a coherence theory of truth. Sec Nicholas Rescher, The Coherence Theory of Truth 23 (1973).

(21) John L. Pollock & Joseph Cruz, Contemporary Theories of Knowledge 71-72 (2d ed. 1999); Andrei Marmor, Coherence, Holism, and Interpretation: The Epistemic Foundations of Dworkin's Legal Theory, 10 L. & Phil. 383, 405-06 (1991). On holism generally, see W.V. Quine & J.S. Ullian, The Web of Belief (1970) (describing holism).

(22) Keith Lehrer, Theory of Knowledge 108-10 (1990).

(23) Id. at 94-97; Pollock & Cruz, supra note 21, at 66-75.

(24) Raz, supra note 19, at 286.

(25) Kress, supra note 16, at 650-59.

(26) Dworkin, supra note 8, at 177-78.

(27) See id. at 245 ("Law as integrity, then, requires a judge to test his interpretation of any part of the great network of political structures and decisions of his community by asking whether it could form part of a coherent theory justifying the network as a whole."); Rolf Sartorius, The Justification of the Judicial Decision, 78 Ethics 171, 178 (1968) ("Any particular decision will ... be justified in terms of its coherence with the system which at any given time achieves a maximum of correspondence in this sense to the total set of relevant existing judicial obligations.").

(28) See supra text accompanying note 18.

(29) See Raz, supra note 19, at 274 ("The more unified the set of principles, the more coherent it is. If all the principles follow from one of their number, then we have strong monistic coherence.").

(30) E.g., Ernest J. Weinrib, Legal Formalism: On the Immanent Rationality of Law, 97 Yale L.J. 949, 949 (1988).

(31) Most commentators have understood Dworkin's theory of Law as Integrity as based on normative coherence. E.g., Kress, supra note 16, at 653-54; Marmor, supra note 21. For a notable dissent, see Raz, supra note 19, at 319-25.

(32) Ronald Dworkin, Justice for Hedgehogs 1, 10-11, 47-48, 93-94, 113-20, 184-88 (2011) (arguing in favor of "a large and old philosophical thesis: the unity of value").

(33) Dworkin, supra note 8, at 176-78 (describing integrity as "an independent ideal" which "can conflict with ... other ideals"); id. at 271-75 (acknowledging the possibility that "different ideologies produced different parts of the law" resulting in "fundamental contradictions of principle"); id. at 441-44 n.20 (suggesting that "any decent response to the world's complexity" will involve the construction of multiple principles which, though "not contradictory," are "competitive"). Commentators during this period widely understood Dworkin as embracing pluralism. E.g., Raz, supra note 19, at 320-22; Kress, supra note 16, at 653, 660.

(34) See Dworkin, supra note 8, at 184 ("Integrity is flouted ... whenever a community enacts and enforces different laws each of which is coherent in itself, but which cannot be defended together as expressing a coherent ranking of different principles of justice or fairness or procedural due process." (emphasis added)); S.L. Hurley, Natural Reasons 260-70 (1989); Raz, supra note 19, at 290 (discussing the possibility of ranking). I leave open whether the ranking need be cardinal (allowing for measurable increments of difference between the ranked values; as 70 degrees is ranked warmer than 60 degrees by precisely 10 degrees of difference) or merely ordinal (naked, lexical ranking; as one film is given the Oscar for best picture and the other nominees are not, with no specification that the winner was "n points better" than the losers).

(35) Dworkin, supra note 8, at 165; Hart & Sacks, supra note 1, at 159.

(36) W. Va. Univ. Hosps., Inc., v. Casey, 499 U.S. 83, 101 (1991).

(37) Green v. Bock Laundry Mach. Co., 490 U.S. 504, 528 (1989) (Scalia, J., concurring).

(38) See infra text accompanying notes 123-35.

(39) See infra text accompanying notes 64-70.

(40) See infra text accompanying notes 71-79.

(41) See infra text accompanying notes 177-85.

(42) See, e.g., Viet D. Dinh, Reassessing the Law of Preemption, 88 Geo. L.J. 2085, 2092 (2000) ("[T]he task for the Court is to discern what Congress has legislated and whether such legislation displaces concurrent state law--in short, the task of statutory construction.").

(43) See, e.g., Cal. Fed. Sav. & Loan Ass'n v. Guerra, 479 U.S. 272, 280 (1987) ("In determining whether a state statute is pre-empted by federal law and therefore invalid under the Supremacy Clause of the Constitution, our sole task is to ascertain the intent of Congress.").

(44) See William N. Eskridge, Jr. et al., Legislation and Statutory Interpretation 219-56 (2d ed. 2006) (describing the debate).

(45) See Hart & Sacks, supra note 1, at 1374 ("In trying to discharge th[e] function [of interpretation] the court should ... [r]espect the position of the legislature as the chief policy-determining agency of the society, subject only to the limitations of the constitution under which it exercises its powers...."); Abner J. Mikva & Eric Lane, An Introduction to Statutory Interpretation and the Legislative Process 4 (1997) ("All judicial approaches to statutory interpretation are framed by the constitutional truism that the judicial will must bend to the legislative command."); Lawrence M. Solan, The Language of Statutes 51 (2010) ("Once a statute is enacted, the common-law court's authority is curtailed, and its obligation is to effect the decision that the legislative body has made.").

(46) See infra text accompanying notes 89-97.

(47) See infra text accompanying notes 98-103.

(48) One of the primary strands of thought that I draw upon in Part V is positive political theory, which is taken by many academics to be the most promising and sophisticated form of intentionalism. See Victoria Nourse, Misunderstanding Congress: Statutory Interpretation, the Supermajoritarian Difficulty, and the Separation of Powers, 99 Geo. L.J. 1119, 1158 (2011) ("[Positive political theory] is more finely attuned to congressional action than is purposivism or textualism."). Both "first" and "second generation" textualists have also embraced many of the considerations discussed in Part V. Frank H. Easterbrook, Statutes' Domains, 50 U. Chi. L. Rev. 533, 547 (1983) (applying Arrow's theorem to statutory interpretation); John F. Manning, Second-Generation Textualism, 98 Calif. L. Rev. 1287, 1290-1317 (2010) [hereinafter Manning, Second-Generation Textualism] (discussing the differences between first and second generation textualism). Finally, many purposivists have apparently begun to move towards recognition of the particularity of congressional purpose. Sec John F. Manning, The New Purposivism, 2011 Sup. Ct. Rev. 113.

(49) See, e.g., William N. Eskridge, Jr., Dynamic Statutory Interpretation (1994) (articulating a "critical pragmatist" theory of statutory interpretation); T. Alexander Aleinikoff, Updating Statutory Interpretation, 87 Mich. L. Rev. 20, 21 (1988) (defending a "nautical" model of statutory interpretation as "an on-going process (a voyage) in which both the shipbuilder and subsequent navigators play a role" (internal quotation marks omitted)).

(50) When discussing the interpretation and application of statutes, I will refer to reasons in favor of or against adopting a particular rule of decision within the statute's domain as "fidelity-based" reasons to the extent that those reasons are grounded in faithfully discerning and carrying out the best interpretation of what the text means, according to whatever suitable theory of interpretation the interpreter adopts. I refer to a reason in favor of or against a rule of decision as "non-fidelity-based" just in case that reason is grounded in considerations that are understood as coming from outside the text at issue. One standard example of a non-fidelity-based reason would be adjudicative efficiency.

(51) See The Federalist No. 78, at 499 (Alexander Hamilton) (Modern Library ed. 2000) ("Nor does [judicial review] by any means suppose a superiority of the judicial to the legislative power. It only supposes that the power of the people is superior to both...."). While judicial review is arguably consistent with the faithful agent theory, it nonetheless constitutes a non-fidelity-based reason for interpreting a statute in a particular way, on my definition, see supra note 50, since it is not grounded in fidelity to the statute. In what follows, I will elide these complexities and generally speak as though fidelity-based reasons are the only reasons consistent with the faithful agent theory.

(52) See Square D Co. v. Niagara Frontier Tariff Bureau, Inc., 476 U.S. 409, 424 (1986) ("If there is to be an overruling of the Keogh rule, it must come from Congress, rather than from this Court.").

(53) See Johnson v. Transp. Agency, 480 U.S. 616, 629 n.7 (1987) ("Congress has not amended the statute to reject our construction, nor have any such amendments even been proposed, and we therefore may assume that our interpretation was correct.").

(54) Eskridge, supra note 49, at 14 ("Intentionalism is in some respects a natural way to view statutory interpretation in a representative democracy."); Aleinikoff, supra note 49, at 56 ("Despite the established nonoriginalist elements in common law and constitutional adjudication, something rubs us the wrong way about [updating] models of statutory interpretation.").

(55) See Amanda L. Tyler, Continuity, Coherence, and the Canons, 99 Nw. U. L. Rev. 1389, 1428 (2005) ("Judges are, at root, charged with accommodating new legislation within the greater statutory framework....").

(56) Such as the values safeguarded by the "substantive" canons of construction, see generally Eskridge, supra note 49, at 275-306 (discussing the substantive canons), or at least those substantive canons that cannot plausibly be justified on fidelity-based grounds, see Amy Coney Barrett, Substantive Canons and Faithful Agency, 90 B.U. L. Rev. 109, 163-81 (2010) (urging that at least some use of substantive canons is consistent with faithful agent theory).

(57) See Aleinikoff, supra note 49, at 58 ("Law is a tool for arranging today's social relations and expressing today's social values; and we fully expect our laws, no matter when enacted, to speak to us today.").

(58) U.S. Const, art. VI, cl. 2.

(59) See, e.g., Cipollone v. Liggett Grp., Inc., 505 U.S. 504, 516 (1992) ("[S]ince our decision in McCulloch v. Maryland, it has been settled that state law that conflicts with federal law is 'without effect.'" (internal citations omitted)).

(60) See supra text accompanying note 38.

(61) See supra text accompanying note 39.

(62) See supra text accompanying notes 40-41.

(63) Since there are no cases considering and affirmatively rejecting this approach, we can only guess as to why the approach did not take hold. It seems likely that at least two considerations played a role. First, the earliest decisions structuring the federal-state relationship were made during the Marshall Court, when the experience of unrestrained, divergent state legislation under the Articles of Confederation was a recent memory and would hardly have sparked hope for salvaging coherence out of state-federal interaction. See Brown v. Maryland, 25 U.S. (12 Wheat.) 419, 445-46 (1827) ("The oppressed and degraded state of commerce previous to the adoption of the constitution can scarcely be forgotten."). Second, as Professor Nelson has persuasively argued, the language of the Supremacy Clause bears a striking resemblance to what was known, at the time, as a "non obstante" clause. Caleb Nelson, Preemption, 86 Va. L. Rev. 225, 254-60 (2000). Such clauses were a well-known technique of directing courts to interpret an enactment fairly rather than straining to harmonize it with previous statutes by applying the traditional presumption against implied repeals. Id. at 235-44.

(64) See Richard A. Epstein & Michael S. Greve, Introduction: Preemption in Context, in Federal Preemption 1, 10 (Richard A. Epstein & Michael S. Greve eds., 2007) ("For much of the nineteenth and early twentieth centuries, the Supreme Court thought of state and federal regulation as separate and mutually exclusive spheres...."); Stephen A. Gardbaum, The Nature of Preemption, 79 Cornell L. Rev. 767, 787-95 (1994) (similar).

(65) See Barry Cushman, Rethinking the New Deal Court 142 (1998) (describing dual federalism). For the origin of the term, see Edward S. Corwin, The Passing of Dual Federalism, 36 Va. L. Rev. 1 (1950).

(66) The role of coherence in this early attempt to preserve the exclusivity of federal and state power is nicely illustrated by the infamous case Priggv. Pennsylvania, 41 U.S. (16 Pet.) 539 (1842). In concluding that Congress's power over fugitive slaves was exclusive, Justice Story reasoned that the subject should be "controlled by one and the same will, and act uniformly by the same system of regulations throughout the Union." Id. at 623. Were power over fugitive slaves concurrent, regulation of the subject "would have no unity of purpose, or uniformity of operation." Id. at 623-24. While Prigg's pursuit of coherence is hardly the greatest of its sins, it is instructive because it reveals the importance of the coherence ideal to the Court's early attempts to articulate clear lines between the federal and state spheres of authority.

(67) See id. (holding that Congress's power over the return of fugitive slaves is exclusive); Houston v. Moore, 18 U.S. (5 Wheat.) 1, 24 (1820) (holding that Congress's power over the militia is not entirely exclusive); Sturges v. Crowninshield, 17 U.S. (4 Wheat.) 122, 196 (1819) (holding that Congress's power over bankruptcy law is not entirely exclusive).

(68) In Gibbons v. Ogden, 22 U.S. 1 (9 Wheat.) (1824), for example, Chief justice Marshall hinted at the exclusive interpretation, see id. at 209; the fractured opinions two decades later in Thurlow v. Massachusetts (The License Cases), 46 U.S. (5 How.) 504 (1847), and Smith v. Turner (The Passenger Cases), 48 U.S. (7 How.) 283 (1849), indicate the degree to which the Justices were divided over the issue.

(69) 53 U.S. (12 How.) 299 (1851).

(70) Id. at 319.

(71) 222 U.S. 424 (1912).

(72) 1879 N.C. Sess. Laws 340.

(73) Interstate Commerce Act of 1887, ch. 104, 24 Stat. 379.

(74) Reid, 222 U.S. at 438.

(75) Id. at 437.

(76) E.g., Chi., Rock Island & Pac. Ry. Co. v. Hardwick Farmers Elevator Co., 226 U.S. 426, 435 (1913); Mondou v. N.Y., New Haven & Hartford R.R., 223 U.S. 1, 55 (1912).

(77) See Gardbaum, supra note 64, at 801-02 (describing the doctrine of "latent exclusivity").

(78) The centrality of coherence to the latent exclusivity approach to preemption is nicely illustrated by the holding in New York Central Railroad Company v. Winfield, 244 U.S. 147 (1917), that a state no-fault compensation scheme for workplace-related injuries was preempted by the Federal Employers' Liability Act, which provided a cause of action to interstate railroad employees injured on the job, but only if their injury was the result of negligence. Id. at 149-50. The Court reasoned that "[o]nly by disturbing the uniformity which the act is designed to secure and by departing from the principle which it is intended to enforce can the several States require such carriers to compensate their employees for injuries in interstate commerce occurring without negligence." Id. at 153.

(79) Mondou, 223 U.S. at 55; see also Lindgren v. United States, 281 U.S. 38, 46-47 (1930) (holding that an act of Congress "is paramount and exclusive, and supersedes the operation of all state statutes dealing with that subject"); Erie R.R. v. Winfield, 244 U.S. 170, 174 (1917) (stating that "[i]t is beyond the power of any State to interfere with the operation of" an act of Congress); Chi., Rock Island & Pac. Ry. Co., 226 U.S. at 435 (1913) (holding that, while a state does have "a right to exert its authority in the absence of legislation by Congress ... the power of the State over [a] subject-matter cease[s] to exist from the moment that Congress exert[s] its paramount and all embracing authority over [that] subject").

(80) 289 U.S. 346 (1933).

(81) Id. at 350.

(82) 331 U.S. 218 (1947).

(83) See Gardbaum, supra note 64, at 805-07 (discussing the "[c]entrality of [i]ntent" to the modern approach).

(84) Cipollone v. Liggett Grp., 505 U.S. 504, 516 (1992) (quoting Fid. Fed. Sav. & Loan Ass'n v. de la Cuesta, 458 U.S. 141, 153 (1982)).

(85) California v. ARC Am. Corp., 490 U.S. 93, 100 (1989).

(86) Gade v. Nat'l Solid Wastes Mgmt. Ass'n, 505 U.S. 88, 98 (1992) (quoting Fla. Lime & Avocado Growers, Inc. v. Paul, 373 U.S. 132, 142-43 (1963)).

(87) Id. at 98 (quoting Hines v. Davidowitz, 312 U.S. 52, 67 (1941)).

(88) Altria Grp., Inc. v. Good, 555 U.S. 70, 77 (2008) (first alteration in original) (quoting Rice v. Santa Fe Elevator Corp., 331 U.S. 218, 230 (1947)).

(89) For the source from which these sources were adapted, see Paul Grice, Studies in the Way of Words 22, 33 (1989).

(90) See id. at 26-31 (arguing that conversations are "characteristically ... cooperative efforts; and each participant recognizes in them ... a common purpose or set of purposes" and accordingly attempts to make each contribution to the conversation "such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange"); Stephen C. Levinson, Presumptive Meanings 27--35 (2000) (accounting for implicature as a mechanism that makes communication more efficient).

(91) For overviews of the field, see Geoffrey N. Leech, Principles of Pragmatics (1983); Stephen C. Levinson, Pragmatics (1983); George Yule, Pragmatics (1996).

(92) Grice, whose Logic and Conversation is the foundational work in the study of "conversational implicature," as this type of inference is known, articulated four such principles or "maxims": (1) The maxim of Quantity, which enjoins a speaker to "make [her] contribution as informative as is required" but no more so; (2) The maxim of Quality, which provides that a speaker should "not say what [she] believe [s] to be false ... [or] that for which [she] lackfs] adequate evidence"; (3) The maxim of Relation: "Be Relevant"; and (4) The maxim of Manner, which instructs speakers to "[a]void obscurity of expression ... [and] ambiguity," and to "[b]e brief ... [and] orderly." Grice, supra note 89, at 26-27. Scholarship on conversational implicature after Grice generally falls into one of two camps: the work of the Relevance Theorists, who have attempted to build a theory of pragmatics and implicature on Grice's maxim of Relevance alone, see Dan Sperber & Deirdre Wilson, Relevance: Communication and Cognition 155-63, 176-202 (1986); Robyn Carston, Informativeness, Relevance and Scalar Implicature, in Relevance Theory 179, 212-26 (Robyn Carston & Seiji Uchida eds., 1998), and the work of the Neo-Griceans, who have worked to salvage as much of Grice's framework as possible, see Laurence R. Horn, A Natural History of Negation 192-203 (1989). Though both groups of scholars diverge from his original four maxims in important respects, Grice's work remains foundational.

(93) For a discussion of the topic of conversational implicature and its relevance for legal theory that is somewhat more technical and that canvasses the philosophical and linguistic literature more thoroughly, see John David Ohlendorf, Textualism and Obstacle Preemption, 47 Ga. L. Rev. 369 (2013).

(94) See Grice, supra note 89, at 26-27 (discussing the maxim of Quantity); Sperber & Wilson, supra note 92, at 172-254 (discussing the role of implicature in Relevance theory); Laurence R. Horn, Implicature, in The Handbook of Pragmatics 3, 6-17, 24-25 (Laurence R. Horn & Gregory Ward eds., 2006) (discussing Neo-Gricean principles of quantity implicature).

(95) Some scholars have articulated their own "informativeness principles," which are meant to occupy a technical space in the spectrum of conversational implicature. E.g., Levinson, supra note 90, at 114. I mean my discussion of the "informativeness principle" to refer, as a sort of nontechnical shorthand, to the larger phenomenon of Gricean implicature, not as a reference to the technical maxims developed in the literature under the same name.

(96) Since Grice, pragmatics has relied on a central distinction between "what is said" and "what is implicated." See Grice, supra note 89, at 24-26. "What is said" is, roughly, determined by the semantic and syntactic conventions that govern the relevant language. Id. at 25. "What is implicated" is not conventionally determined but rather is worked out through an inferential process that reasons from what is said, the context of utterance, and the conversational maxims. Id. at 24-26, 30-31, 39-40. For some complications, see Sperber & Wilson, supra note 92, at 182 (discussing explicatures); Kent Bach, Conversational Impliciture, 9 Mind & Language 124, 124-33 (1994) (discussing implicitures); 1 Scott Soames, Drawing the Line Between Meaning and Implicature--and Relating Both to Assertion, in Philosophical Essays: Natural Language: What It Means and How We Use It 298, 316-21 (2009) (discussing pragmatic enrichment).

(97) See Grice, supra note 89, at 30-31.

(98) Helvering v. Hallock, 309 U.S. 106, 120 (1940).

(99) E.g., John F. Manning, The Absurdity Doctrine, 116 Harv. L. Rev. 2387, 2462 n.274 (2003); Andrei Marmor, The Pragmatics of Legal Language, 21 Ratio Juris 423, 435-40 (2008). Elsewhere, I have argued at some length that Grice's Quantity maxim, at least, is very sensibly applied to the legal context, and that we should be quite grateful that this is so, since Gricean-style reasoning is extremely widespread in statutory interpretation and unjustifiable on any other plausible grounds. Ohlendorf, supra note 93, at 422-38. The extent to which the rest of Grice's framework applies in the statutory context is an interesting question that lies, unfortunately, beyond the scope of this Article.

(100) Williams v. Taylor, 529 U.S. 362, 404 (2000) ("It is ... a cardinal principle of statutory construction that we must 'give effect, if possible, to every clause and word of a statute.'" (quoting United States v. Menasche, 348 U.S. 528, 538-39 (1955))).

(101) See M.B.W. Sinclair, Law and Language: The Role of Pragmatics in Statutory Interpretation, 46 U. Pitt. L. Rev. 373, 394 (1985) (explaining the rule against surplusage along Gricean lines).

(102) Arizona v. United States, 132 S. Ct. 2492, 2520 (2012) (discussing "the canon expressio unius est exclusio alterius" as reflecting "[c]ommon sense").

(103) See Geoffrey P. Miller, Pragmatics and the Maxims of Interpretation, 1990 Wis. L. Rev. 1179, 1195-1202 (describing expressio unius as a Gricean quantity implicature); Sinclair, supra note 101, at 414-20 (same).

(104) 26 U.S.C. [section] 5000A (2012).

(105) U.S. Const, art. VI, cl. 2.

(106) This type of inference is known in the pragmatics literature as "scalar implicature." See generally Julia Hirschberg, A Theory of Scalar Implicature 1-5 (1991) discussing the concept of scalar implicature). The literature on scalar implicature is vast, and I explore much of it in Ohlendorf, supra note 93, at 388-401.

(107) Every person who has three children also has two children (plus one additional one).

(108) Ohlendorf, supra note 93.

(109) 505 U.S. 88 (1992).

(110) 29 C.F.R. [section] 1910.120(e) (3) (i) (2011).

(111) 1988 111. Legis. Serv. P.A. 85-1195 (West) (repealed 2009).

(112) Gade, 505 U.S. at 99 (citing Hines v. Davidowitz, 312 U.S. 52, 67 (1941)).

(113) 373 U.S. 132 (1963).

(114) Id. at 157-59.

(115) California Fruit and Vegetable Standardization Act of 1925, ch. 350, [section] 10, 1925 Cal. Stat. 625, 633.

(116) 7 C.F.R. [section][section] 51.3050-51.3069 (2014).

(117) Fla. Lime, 373 U.S. at 144.

(118) See Ohlendorf, supra note 93, at 397-401, 413-22.

(119) See id. at 419-21.

(120) Wis. Pub. Intervenor v. Mortier, 501 U.S. 597, 605 (1991) (quoting Rice v. Santa Fe Elevator Corp., 331 U.S. 218, 230 (1947)). The Court sometimes articulates the presumption in such a way as to suggest that it applies the presumption across the board, not only in cases implicating traditionally local topics. See, e.g., Exxon Corp. v. Governor of Md., 437 U.S. 117, 132 (1978) ("This Court is generally reluctant to infer preemption .... "). However, the results in the Court's preemption cases from the past several decades indicate that the Court is not using the presumption in this strong way. See infra text accompanying notes 128-30.

(121) See Gade v. Nat'l Solid Wastes Mgmt. Ass'n, 505 U.S. 88, 98 (1992) (noting that courts will find "field pre-emption" if "the scheme of federal regulation is so pervasive as to make reasonable the inference that Congress left no room for the States to supplement it .... " (quoting Rice, 331 U.S. at 230) (internal quotation marks omitted)).

(122) See supra text accompanying notes 71-83.

(123) At the Founding, it was a well understood principle of English political theory that Parliament could not bind its successors. See 1 William Blackstone, Commentaries on the Laws of England *90 ("Acts of parliament derogatory from the power of subsequent parliaments bind not."). A line of cases stretching back to Marbury reaffirm the principle.

See, e.g., United States v. Winstar Corp., 518 U.S. 839, 872 (1996) (plurality opinion) (citing Blackstone for the "centuries-old concept that one legislature may not bind the legislative authority of its successors," though conceding that the principle is in some tension with "the theory that legislative power may be limited"); Manigault v. Springs, 199 U.S. 473, 487 (1905) ("As this is not a constitutional provision, but a general law enacted by the legislature, it may be repealed, amended!,] or disregarded by the legislature which enacted it."); Marbury v. Madison, 5 U.S. (1 Cranch) 137, 177 (1803) (declaring that an act of the legislature is "alterable when the legislature shall please to alter it").

(124) Radzanower v. Touche Ross &: Co., 426 U.S. 148, 154 (1976) (citations omitted) (quoting United States v. United Cont'l Tuna Corp., 425 U.S. 164, 168 (1976); Posadas v. National City Bank, 296 U.S. 497, 503 (1936)) (internal quotation marks omitted).

(125) Id.; see also supra notes 79-83 and accompanying text (describing the role of legislative intent in modern preemption doctrine).

(126) Language from other cases indicating that the conflict must be one of "plain repugnancy," see Otter Tail Power Co. v. United States, 410 U.S. 366, 372 (1973) (quoting United States v. Phila. Nat'l Bank, 374 U.S. 321, 350-51 (1963)), supports a reading of the two standards as essentially the same.

(127) See Karen Petroski, Retheorizing the Presumption Against Implied Repeals, 92 Calif. L. Rev. 487, 519 (2004) (noting that the field preemption doctrine "recalls the comprehensive revision theory of implied repeal").

(128) See Gregory M. Dickinson, An Empirical Study of Obstacle Preemption in the Supreme Court, 89 Neb. L. Rev. 682, 694 (2011) (analyzing obstacle preemption cases); Michael S. Greve & Jonathan Klick, Preemption in the Rehnquist Court: A Preliminary Empirical Assessment, 14 Sup. Ct. Econ. Rev. 43, 57 (2006) (analyzing the Rehnquist Court); David M. O'Brien, The Rehnquist Court and Federal Preemption: In Search of a Theory, 23 Publius 15, 22 (1993) (analyzing the Rehnquist and Burger Courts).

(129) David B. Spence & Paula Murray, The Law, Economics, and Politics of Federal Preemption Jurisprudence: A Quantitative Analysis, 87 Calif. L. Rev. 1125, 1159 (1999).

(130) S. Candice Hoke, Preemption Pathologies and Civic Republican Values, 71 B.U. L. Rev. 685, 733 (1991).

(131) Petroski, supra note 127, at 511.

(132) Branch v. Smith, 538 U.S. 254, 293 (2003) (O'Connor, J., concurring in part and dissenting in part).

(133) Petroski, supra note 127, at 511.

(134) See supra text accompanying notes 106-20.

(135) Compare United States v. One Ford Coupe Auto., 272 U.S. 321, 331 (1926) ("Direct conflict is not established by showing merely differences in details of procedure."), with Crosby v. Nat'l Foreign Trade Council, 530 U.S. 363, 379 (2000) ("The fact of a common end hardly neutralizes conflicting means .... ").

(136) (1614) 77 Eng. Rep. 1222 (K.B.).

(137) Id. at 1232; see also 9 Matthew Bacon, A New Abridgment of the Law 228 (Henry Gwyllim 8c Charles Edward Dodd eds., Philadelphia, T. & J.W. Johnson & Co. 1876) (1759) (suggesting that an implied repeal "carries with it a reflection upon the wisdom of the former parliament").

(138) 77 Eng. Rep. at 1232.

(139) See, in chronological order, Warder v. Arell, 2 Va. 282, 299 (1796) (Carrington, J.) ("Whatever apparent [incon]sistencies may appear in the declarations of the legislative [will,] yet it is not decent to presume that they would change [their] mind upon the subject, without saying so in express terms."); Theodore Sedgwick, A Treatise on the Rules Which Govern the Interpretation and Construction of Statutory and Constitutional Law 106 (New York, Baker, Voorhis 8c Co. 2d ed. 1874) ("[L]aws are presumed to be passed with deliberation, and with full knowledge of all existing ones on the same subject .... "); Peter Benson Maxwell, On the Interpretation of Statutes 133 (London, William Maxwell & Son 1875) ("An author must be supposed to be consistent with himself ... [and] if in one place he has expressed his mind clearly, it ought to be presumed that he is still of the same mind in another place, unless it expressly appears that he has changed it .... ").

(140) Antonin Scalia & Bryan A. Garner, Reading Law 330 (2012).

(141) David L. Shapiro, Continuity and Change in Statutory Interpretation, 67 N.Y.U. L. Rev. 921, 937 (1992).

(142) Tyler, supra note 55, at 1426-28 (quoting Hart & Sacks, supra note 1, at 1416).

(143) 426 U.S. 148 (1976).

(144) Id. at 149.

(145) Securities Exchange Act of 1934, Pub. L. No. 73-291, [section] 27, 48 Stat. 881, 902-03 (codified as amended at 15 U.S.C. [section] 78aa(a) (2012)).

(146) 12 U.S.C. [section] 94 (1970). The statutory language is actually that an action "may be had" in such a district, but the Court had earlier interpreted this language to be exclusive. Mercantile Nat'l Bank v. Langdeau, 371 U.S. 555, 560 (1962).

(147) Radzanower, 426 U.S. at 154 (quoting United States v. United Cont'l Tuna Corp., 425 U.S. 164, 168 (1976)).

(148) Id. at 156.

(149) Id. at 158 (alteration in original) (quoting Morton v. Mancari, 417 U.S. 535, 551 (1974)).

(150) Id. at 156.

(151) Id. at 155 (alteration in original) (quoting Mancari, 417 U.S. at 551) (internal quotation marks omitted).

(152) 538 U.S. 254 (2003).

(153) Id. at 258.

(154) Id. at 258-61.

(155) 2 U.S.C. [section] 2a(c) (2012).

(156) 2 U.S.C. [section] 2c.

(157) 2 U.S.C. [section] 2a(c).

(158) Smith v. Clark, 189 F. Supp. 2d 548, 549 (S.D. Miss. 2002) (ordering the use of the court's previously proposed plan); Smith v. Clark, 189 F. Supp. 2d 512, 512-25 (S.D. Miss. 2002) (setting forth the court's proposed plan).

(159) Branch, 538 U.S. at 273 (plurality opinion).

(160) Id. at 273-74.

(161) Id. at 274 (internal quotation marks omitted).

(162) Id. at 281. Tellingly, Justices O'Connor and Thomas, in dissent, faulted the plurality for paying insufficient concern to the coherence ideal by adopting an interpretation of [section] 2a(c) that essentially repealed it "sub silentio." Id. at 308 (O'Connor, J., concurring in part and dissenting in part).

(163) It is revealing that Justice Stevens's concurrence, which was the only opinion to openly find implied repeal, also articulated an understanding of repeal doctrine much closer to preemption doctrine than it actually is. See id. at 285 & n.1 (Stevens, J., concurring).

(164) Alexander M. Bickel, The Least Dangerous Branch 17 (1962).

(165) Dworkin, supra note 8, at 165.

(166) City of Milwaukee v. Illinois (Milwaukee II), 451 U.S. 304, 313 (1981) ("We have always recognized that federal common law is 'subject to the paramount authority of Congress.'" (quoting New Jersey v. New York, 283 U.S. 336, 348 (1931))).

(167) It is widely recognized that many of the judicially crafted rules of decision that make up constitutional law are only loosely tethered to the text of the Constitution and can be called "interpretations" of the Constitution in only the most attenuated sense. One influential strand of the literature discusses these norms as "implementing" or "prophylactic" rules. See, e.g., Richard H. Fallon, Jr., Implementing the Constitution (2001); David A. Strauss, The Ubiquity of Prophylactic Rules, 55 U. Chi. L. Rev. 190 (1988). Another group of scholars, writing from an originalist perspective, refers to roughly the same phenomenon as "construction," as distinguished from "interpretation." See, e.g., Randy E. Barnett, Interpretation and Construction, 34 Harv. J.L. & Pub. Pol'y 65 (2011); Lawrence B. Solum, The Interpretation-Construction Distinction, 27 Const. Comment. 95 (2010). Importantly, some scholars have suggested that this set of implementing rules or constructions is best conceptualized as a form of federal common law. See Thomas W. Merrill, The Common Law Powers of Federal Courts, 52 U. Chi. L. Rev. 1, 6 (1985); Henry P. Monaghan, The Supreme Court, 1974 Term--Foreword: Constitutional Common Law, 89 Harv. L. Rev. 1 (1975). And in a line of cases, the Court has indicated that these constitutionally grounded, judicially crafted rules of decision are not revisable by Congress. See Dickerson v. United States, 530 U.S. 428 (2000) (Miranda warnings); United States v. Morrison, 529 U.S. 598 (2000) (Equal Protection Clause doctrine); City of Boerne v. Flores, 521 U.S. 507 (1997) (Establishment Clause doctrine).

(168) 41 U.S. (16 Pet.) 1 (1842).

(169) See William A. Fletcher, The General Common Law and Section 34 of the Judiciary Act of 1789: The Example of Marine Insurance, 97 Harv. L. Rev. 1513, 1515-21 (1984) ("[L]ong before Swift, federal courts employed the general common law as an important part of their working jurisprudence.").

(170) Stewart Jay, Origins of Federal Common Law: Part One, 133 U. Pa. L. Rev. 1003, 1039 (1985) ("Hostility to the invocation of common-law criminal jurisdiction by federal courts became increasingly acute as the decade of the 1790's unfolded.").

(171) United States v. Hudson & Goodwin, 11 U.S. (7 Cranch) 32, 33-34 (1812) (holding that federal courts do not have the power to create common-law crimes based on structural considerations).

(172) 304 U.S. 64, 78 (1938).

(173) See Bradford R. Clark, Federal Common Law: A Structural Reinterpretation, 144 U. Pa. L. Rev. 1245, 1255 (1996) ("Erie established the general rule--based on mutually reinforcing principles of federalism and separation of powers--that federal courts must apply the substantive law of the state in which they sit in the absence of positive federal law to the contrary. Federal common law appears to represent a departure from this rule."); Martha A. Field, Sources of Law: The Scope of Federal Common Law, 99 Harv. L. Rev. 881, 885 (1986) ("The appropriate bounds for federal common law have always been unclear. Its very legitimacy seemed questioned in 1938 when Justice Brandeis announced for the Court in Erie Railroad v. Tompkins that '[t]here is no federal general common law.'" (alteration in original) (footnotes omitted) (quoting Erie, 304 U.S. at 78)).

(174) Milwaukee II, 451 U.S. 304, 313 (1981) (quoting Wheeldin v. Wheeler, 373 U.S. 647, 651 (1963)).

(175) Henry J. Friendly, In Praise o/Erie--And of the New Federal Common Law, 39 N.Y.U. L. Rev. 383, 419 (1964). One caveat is in order: some cases have suggested that, given the constitutionally prescribed role of Congress in approving interstate compacts, see U.S. Const, art. I, [section] 10, cl. 3, Congress is without power to pass ordinary legislation on the subject of interstate disputes. See Kansas v. Colorado, 206 U.S. 46, 97 (1907) (raising this possibility). I am unconvinced that the Compact Clause carries this implication; and Milwaukee II, 451 U.S. at 304, stands at least for the sensible proposition that legislation pursuant to Congress's enumerated powers can displace a body of federal common law grounded, in part, in the Court's jurisdiction over interstate disputes. See Illinois v. City of Milwaukee (Milwaukee I), 406 U.S. 91, 105 (1972) (grounding federal common law of interstate air pollution in part on Article Ill's grant of original jurisdiction over disputes between states). Nonetheless, the sum of Congress's enumerated powers may be less extensive than the Court's jurisdiction over interstate disputes, entailing that some judge-made rules of decision would be beyond ordinary congressional control.

(176) Milwaukee II, 451 U.S. at 313 (quoting New Jersey v. New York, 283 U.S. 336, 348 (1931)).

(177) United States v. Texas, 507 U.S. 529, 534 (1993) (quoting Mobil Oil Corp. v. Higginbotham, 436 U.S. 618, 625 (1978)).

(178) See Thomas W. Merrill, Global Warming as a Public Nuisance, 30 Colum. J. Envtl. L. 293, 311-12 (2005) (discussing the "field displacement" theory and the "conflict displacement" theory).

(179) Cnty. of Oneida v. Oneida Indian Nation, 470 U.S. 226, 237 (1985) (alteration in original) (quoting Milwaukee II, 451 U.S. at 313, 315) (internal quotation marks omitted). Moreover, the Court occasionally references a "presumption favoring the retention of long-established and familiar principles" against "[s]tatutes which invade the common law" unless "a statutory purpose to the contrary is evident." Texas, 507 U.S. at 534 (quoting Isbrandtsen Co. v.Johnson, 343 U.S. 779, 783 (1952)) (internal quotation marks omitted).

(180) 131 S. Ct. 2527 (2011).

(181) Id. at 2537 (alteration in original) (quoting Milwaukee II, 451 U.S. at 317).

(182) Milwaukee II, 451 U.S. at 317 n.9.

(183) See supra text accompanying notes 71-78.

(184) See supra text accompanying notes 166-76.

(185) See supra text accompanying notes 40-41.

(186) See, e.g., Adam Liptak, Justices Rebuff States on Utilities ' Gas Emissions, N.Y. Times, June 21, 2011, http://www.nytimes.com/2011/06/21/science/earth/21warming.html.

(187) Am. Elec. Power Co. v. Connecticut, 131 S. Ct. 2527, 2540 (2011); see Restatement (Second) of Torts [section] 821B (1979) ("A public nuisance is an unreasonable interference with a right common to the general public.").

(188) See Missouri v. Illinois, 200 U.S. 496 (1906) (deciding a water pollution dispute between St. Louis and Chicago); see also Georgia v. Tenn. Copper Co., 206 U.S. 230, 239 (1907) (holding that the discharge of "noxious gas" by a copper company located in Tennessee across the border into Georgia constituted a public nuisance).

(189) See, e.g., David A. Grossman, Warming Up to a Not-So-Radical Idea: Tort-Based Climate Change Litigation, 28 Colum.J. Envtl. L. 1, 3 (2003) ("Given the increasing evidence connecting human use of fossil fuels to climate change, and perceiving a lack of meaningful political action to address global warming, environmental lawyers have begun exploring litigation strategies.").

(190) A delegation the Court had found in the CAA only four years earlier. See Massachusetts v. EPA, 549 U.S. 497, 532 (2007).

(191) Am. Elec. Power Co., 131 S. Ct. at 2538.

(192) Id. (quoting Milwaukee II, 451 U.S. 304, 324 (1981)).

(193) Id.

(194) Id. at 2539.

(195) Id. at 2538.

(196) Id. at 2539.

(197) Moragne v. States Marine Lines, Inc., 398 U.S. 375, 395 (1970).

(198) Pub. L. No. 66-165, 41 Stat. 537 (1920) (codified as amended at 46 U.S.C. [section][section] 30301-30308 (2012)).

(199) Id. [section] 1.

(200) See id. (providing a cause of action for "death ... caused by ... neglect").

(201) See Kernan v. Am. Dredging Co., 355 U.S. 426, 430 n.4 (1958) (noting that a DOHSA action will lie for death caused by unseaworthiness).

(202) See Seas Shipping Co. v. Sieracki, 328 U.S. 85, 96 (1946) (extending the duty of seaworthiness to longshoreman working onboard).

(203) Merchant Marine Act of 1920, ch. 250, [section] 33, 41 Stat. 988, 1007 (codified as amended at 46 U.S.C. [section] 30104). The Jones Act's cause of action applies to the death of a seaman that occurs on the high seas as well. Id.

(204) Miles v. Apex Marine Corp., 498 U.S. 19, 27-30, 37 (1990).

(205) See id. at 32-37.

(206) DOHSA's death action is limited on its face to recovery of "a fair compensation for the pecuniary loss sustained," 46 U.S.C. [section] 30303, which has been interpreted to exclude loss of society. See, e.g., Middleton v. Luckenbach S.S. Co., 70 F.2d 326, 330 (2d Cir. 1934) ("There may be no allowance for loss of society and companionship."). The Jones Act wrongful death action has been interpreted as limited in the same way. See, e.g., Miles, 498 U.S. at 32 ("There is no recovery for loss of society in a Jones Act wrongful death action.").

(207) See Stuart M. Speiser & James E. Rooks, Jr., Recovery for Wrongful Death [section] 6:9 (4th ed. 2005) (cataloging statutes).

(208) Indeed, in an earlier case dealing with a federal-common-law death action, sounding in unseaworthiness, for longshoremen injured within territorial waters, the Court had allowed nonpecuniary damages. See Sea-Land Servs., Inc. v. Gaudet, 414 U.S. 573, 587-88 (1974). But this particular death action was superseded by the 1972 amendments to the Longshoremen's and Harbor Workers' Compensation Act, Pub. L. No. 92-576, [section] 18(a), 86 Stat. 1251, 1263 (codified as amended at 33 U.S.C. [section] 905(b)).

(209) Miles, 498 U.S. at 27.

(210) Id. at 32-33.

(211) Id. at 37; cf. supra note 78 (discussing N.Y. Cent. R.R. v. Winfield, 244 U.S. 147, 153 (1917)).

(212) Miles, 498 U.S. at 36.

(213) See supra Section I.C.

(214) Dworkin, supra note 8, at 313-54; Ronald Dworkin, The Arduous Virtue of Fidelity: Originalism, Scalia, Tribe, and Nerve, 65 Fordham L. Rev. 1249 (1997).

(215) I confine the scope of the direct argument to congressionally enacted law (therefore limiting its relevance for this Article to the area of repeal); but if the argument fails with respect to this subset of legal rules, it clearly fails with respect to the broader set (including state-enacted and federal judge-made rules) as well.

(216) Rawls, supra note 13, at 36.

(217) Jeremy Waldron, Law and Disagreement 105 (1999). Waldron thus extends Rawls's fact of reasonable pluralism to encompass reasonable disagreement over the "right," and justice, as well as disagreement over the "good." See id. at 105-06, 149-63.

(218) Id. at 102.

(219) This description is of course highly stylized, but the additional complexities do nothing to take away from the basic thrust of the argument.

(220) Waldron, supra note 217, at 144.

(221) See Kenneth J. Arrow, Social Choice and Individual Values 22-33 (2d ed. 1963). For an influential summary of the constraints, see Amartya K. Sen, Collective Choice and Social Welfare 167-69 (1970).

(222) Arrow, supra note 221.

(223) See supra text accompanying notes 32-37.

(224) Arrow, supra note 221, at 11-14; Hurley, supra note 34, at 256-60.

(225) Daniel A. Farber & Philip P. Frickey, Law and Public Choice 39-40 (1991); Richard H. Pildes & Elizabeth S. Anderson, Slinging Arrows at Democracy: Social Choice Theory, Value Pluralism, and Democratic Politics, 90 Colum. L. Rev. 2121, 2137-38 (1990).

(226) See Farber & Frickey, supra note 225, at 47-55 (discussing empirical studies showing a lack of legislative disequilibrium).

(227) See supra text accompanying notes 216-22.

(228) See Farber & Frickey, supra note 225, at 48-49 (discussing single-peaked preferences); Hurley, supra note 34, at 254-56 (same).

(229) See Farber & Frickey, supra note 225, at 49-50 (discussing structure-induced equilibrium); Kenneth A. Shepsle, Institutional Arrangements and Equilibrium in Multidimensional Voting Models, 23 Am. J. Pol. Sci. 27 (1979) (same). For some reservations about this tack, see William H. Riker, Implications from the Disequilibrium of Majority Rule for the Study of Institutions, 74 Am. Pol. Sci. Rev. 432 (1980).

(230) Manning, Second-Generation Textualism, supra note 48, at 1289.

(231) See John David Ohlendorf, Textualism and the Problem of Scrivener's Error, 64 Me. L. Rev. 119, 125 (2011) (noting this danger).

(232) See Barbara Sinclair, Unorthodox Lawmaking 9-56, 218-26 (2d ed. 2000) (describing the complex procedural obstacle course through which modern legislation must pass); McNollgast, Positive Canons: The Role of Legislative Bargains in Statutory Interpretation, 80 Geo. L.J. 705, 711-12 (1992) ("[I]f statutory interpretation is guided by the principle of honoring the spirit of the legislative bargain, it must not focus only on the preferences of the ardent supporters, but also on the accommodations that were necessary to gain the support of the moderates."); Daniel B. Rodriguez & Barry R. Weingast, The Paradox of Expansionist Statutory Interpretations, 101 Nw. U. L. Rev. 1207, 1212-23 (2007) (noting that "passage of legislation usually involves legislative bargaining among ardent supporters and moderates to assure, in the end, sufficient assent"). "McNollgast," a frequent author in the positive political theory literature, is a collective pseudonym for the frequent coauthors Mathew McCubbins, Roger Noll, and Barry Weingast.

(233) See Miriam R. Jorgensen & Kenneth A. Shepsle, A Comment on the Positive Canons Project, Law & Contemp. Probs., Winter/Spring 1994, at 43, 44-46 (describing the conditions under which an enacting coalition might "write [ ] both A and B [two contradictory approaches] into the statute as separate titles, even though the factions fully anticipate that circumstances might arise in which A stands in contradiction to B").

(234) Daniel B. Rodriguez & Barry R. Weingast, The Positive Political Theory of Legislative History: New Perspectives on the 1964 Civil Rights Act and Its Interpretation, 151 U. Pa. L. Rev. 1417, 1487-98 (2003). A briefer discussion by the same authors of the 1969 National Environmental Policy Act reveals a similar process of finely tuned compromise. Rodriguez & Weingast, supra note 232, at 1242-48.

(235) See Eskridge, Jr. et al., supra note 44, at 83-99 (discussing interest group theory); Jonathan R. Macey, Promoting Public-Regarding Legislation Through Statutory Interpretation: An Interest Group Model, 86 Colum. L. Rev. 223, 227-33 (1986) (same).

(236) Chris Frates, Payoffs for States Get Harry Reid to 60 Votes, Politico (Dec. 19, 2009, 7:56 PM), http://www.politico.com/news/stories/1209/30815.html.

(237) See William W. Buzbee, The One-Congress Fiction in Statutory Interpretation, 149 U. Pa. L. Rev. 171, 204-19 (2000) (arguing that positive political theory insights into the process of legislative compromise undermine any treatment of Congress as "a single natural person ... of superhuman omniscience and consistency of style" even assuming "a highly unrealistic static political environment").

(238) Cass R. Sunstein, Incompletely Theorized Agreements, 108 Harv. L. Rev. 1733 (1995).

(239) Id. at 1734.

(240) See supra text accompanying notes 216-20.

(241) Sunstein, supra note 238, at 1739.

(242) Id.

(243) Id. at 1740.

(244) An essentially contested concept is one that embodies at least two separable senses, which flow from differing, overarching theories or "conceptions" of what the best understanding of the concept itself is. See W.B. Gallie, Essentially Contested Concepts, 56 Proc. Aristotelian Soc'y 167, 171-72 (1956).

(245) See Joseph A. Grundfest & A.C. Pritchard, Statutes with Multiple Personality Disorders: The Value of Ambiguity in Statutory Design and Interpretation, 54 Stan. L. Rev. 627, 640-42 (2002) ("Ambiguity can ... be the result of a need to compromise in order to accumulate a majority (or veto-proof supermajority) in support of legislative action."); Nourse, supra note 48, at 1128-34 (discussing "structure-induced ambiguity").

(246) See Sunstein, supra note 238, at 1742 ("Incompletely theorized agreements ... do not try to rationalize the law by showing how an outcome in one case fits coherently with

particular outcomes in the full range of other cases.... [T]hey do not try for global coherence.").

(247) Charles P. Curtis, A Better Theory of Legal Interpretation, 3 Vand. L. Rev. 407, 415 (1950).

(248) Buzbee, supra note 237, at 204 ("[T]he regular congressional and presidential elections and changing committee membership ensure ever-changing political dynamics.").

(249) Joseph Raz, The Morality of Freedom 322 (1986). Some scholars draw a perspicacious distinction between "incommensurability" which

is the inability to rank two values on a cardinal scale, and "in comparability," which is the inability to rank the values on an ordinal scale. See Ruth Chang, Introduction to Incommensurability, Incomparability, and Practical Reason 1, 1-2 (Ruth Chang ed., 1997). While the distinction is helpful, the discussion of these phenomena that has made its way into legal theory has generally used "incommensurability" to do double duty for both failures of ranking. See Larry Alexander, Banishing the Bogey of Incommensurability, 146 U. Pa. L. Rev. 1641, 1642 n.7 (1998). Since I have not insisted that coherence entails cardinal rather than ordinal ranking, see supra note 34, only the stronger claim that values are incomparable in this technical sense poses a challenge to normative coherence as I have defined it. However, because "incommensurability" is the more familiar term in legal theory, I cautiously employ it here, with the stipulation that it serves to identify the failure of both ordinal and cardinal ranking.

(250) John Finnis, Natural Law and Natural Rights 113 (1980).

(251) Bendix Autolite Corp. v. Midwesco Enters., Inc., 486 U.S. 888, 897 (1988) (Scalia, J., concurring).

(252) Finnis, supra note 250, at 111-18; Elizabeth Anderson, Practical Reason and Incommensurable Goods, in Incommensurability, Incomparability, and Practical Reason, supra note 249, at 90, 90-91; Cass R. Sunstein, Incommensurability and Valuation in Law, 92 Mich. L. Rev. 779, 795-812 (1994).

(253) Supra text accompanying notes 28-37.

(254) See Raz, supra note 249, at 325 ("The test of incommensurability is failure of transitivity."); Pildes & Anderson, supra note 225, at 2161 (similar).

(255) The incommensurability thesis is not universally accepted; for some dissents, see Dworkin, supra note 32, at 88-96, 118-20; Hurley, supra note 34, at 254-70 (purporting to accept incommensurability, but Hurley does so by "reining in" the notion of incommensurability beyond recognition); Alexander, supra note 249; Chang, supra note 249.

(256) I am grateful to both Caleb Nelson and Nick Rosenkranz for helping me to feel the force of this argument.

(257) Except for the third, temporal point; this is a conceptual, not empirical, reason for thinking Congress's knowledge is asymmetric in this way.

(258) As it frequently does with respect to state law, for example, by including a "savings clause."

(259) Cf. Adrian Vermeule & Ernest A. Young, Commentary, Hercules, Herbert, and Amor: The Trouble with Intratextualism, 113 Harv. L. Rev. 730, 760 n.160 (2000) (noting that adopting a "coherentist" style of interpreting the Constitution would dramatically increase the costs of amendment by giving each constitutional change unpredictable "ripple effects").

(260) I am indebted to Nick Rosenkranz for pressing me on this point.

(261) Cf. Hart & Sacks, supra note 1, at 97, 496-527, 775-83; Tyler, supra note 55, at 1438-60. I thank Amy Barrett for helping me to see the argument in this way.

(262) The phrase is from Isaiah 28:15 (King James), though it is most famous in our time as Garrison's description of the Constitution. Letter from William Lloyd Garrison to Rev. Samuel J. May (July 17, 1845), in 3 The Letters of William Lloyd Garrison 303 (Walter M. Merrill & Louis Ruchames eds., 1973). For a thoughtful discussion of the appropriate scope and limits of compromise, see Avishai Margalit, On Compromise and Rotten Compromises (2009).

(263) See John F. Manning, Competing Presumptions About Statutory Coherence, 74 Fordham L. Rev. 2009, 2039 (2006) (noting that our constitutional structure "contemplates and, indeed, places a premium on legislative compromise").

(264) Rodriguez & Weingast, supra note 234, at 1454 ("Because the number of northern Democrats was far too small to defeat the filibuster, passing a major new civil rights bill required support of a large Republican bloc, including many quite conservative ones who generally opposed new initiatives proposing government intervention."); see Rodriguez & Weingast, supra note 232, at 1219 ("The wish that Congress had enacted more far-reaching legislation often seriously underestimates the degree to which legislative compromises were necessary to enact any bill.").

(265) Waldron, supra note 217, at 106.

(266) Id. at 101.

(267) Rodriguez and Weingast argue this point persuasively and at great length, concluding that when courts "set aside ... finely crafted legislative compromises in favor of expansionary readings," this "has a feedback effect on the legislature by making new legislation less likely." Rodriguez & Weingast, supra note 232, at 1255.

(268) W. Va. Univ. Hosps., Inc. v. Casey, 499 U.S. 83, 100 (1991).

(269) See 1 Blackstone, supra note 123, at *60 n.12. ("It is an established rule of construction that statutes in pari materia, or upon the same subject, must be construed with reference to each other.").

(270) Roscoe Pound, Common Law and Legislation, 21 Harv. L. Rev. 383, 385 (1908). Pound's suggestion was famously repeated by Landis, see James McCauley Landis, Statutes and the Sources of Law, in Harvard Legal Essays 213, 226-27 (1934), and Stone, see Harlan F. Stone, The Common Law in the United States, 50 Harv. L. Rev. 4, 12-13 (1936). For more modem echoes, see Guido Calabresi, A Common Law for the Age of Statutes 82-83, 85-86 (1982); Hart & Sacks, supra note 1, at 467, 472-73, 758-63, 775, 780, 782-83, 1194, 1210, 1370, 1378; William N. Eskridge, Jr., Public Values in Statutory Interpretation, 137 U. Pa. L. Rev. 1007, 1011-12 (1989).

John David Ohlendorf, Associate, Cooper & Kirk, PLLC, Washington, DC; J.D., Harvard Law School. The views expressed below are of course my own. I would like to thank Randy Barnett, Amy Barrett, Greg Dickinson, Andy Grewal, Erin Hawley, Richard Izquierdo, Josh Kleinfeld, Gabriel Latner, John McGinnis, Caleb Nelson, Nick Nelson, Jim Pfander, Paul Robinson, Nick Rosenkranz, Lochlan Shelfer, Larry Solum, Justin Walker, and Ryan Williams for helpful comments. I am also grateful to the participants in the Georgetown Fellows' Collaborative, particularly Dov Fox, Greg Klass, Miriam Seifter, and Robin West, for thoughtful and probing suggestions.
COPYRIGHT 2014 University of Notre Dame Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Ohlendorf, John David
Publication:Notre Dame Law Review
Date:Dec 1, 2014
Words:25444
Previous Article:Immigration separation of powers and the president's power to preempt.
Next Article:Quasi-inquisitorialism: accounting for deference in pretrial criminal procedure.
Topics:

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters