Introduction

A division of intellectual labor is both understandable and often desirable. The agenda of different scholars—even when they study the same phenomenon, such as law—is sensibly guided by their disciplinary concerns and methodological capabilities, and nothing that follows is intended to quarrel with this truism. Thus, our focus on legal theory does not imply that other theories that inform social science research on law, which come from neighboring disciplines such as economics, sociology, psychology, or anthropology, are not—or not as—important to the understanding of law or to its empirical investigation. We do not deny or minimize the significance of the application of non-legal disciplinary methodology or theory to the data of law. By the same token, we realize that legal empiricists tend to view the law externally and to rely implicitly on background assumptions regarding important features of law, rather than reflect on or be guided by a full-blown theory of law, or even a theoretical perspective on law. And we do not argue that legal empiricists should become legal theoreticians (or vice versa).

Our claim is not one of disciplinary or methodological imperialism. Rather, we hope to point to the distinctive voice of legal scholars within the broader community of interdisciplinary students of law and to highlight the role legal theory can play in usefully informing empirical studies of law. Legal theory, we argue, offers a distinct added value, which might be obscured if either the interdisciplinary turn of legal scholarship or the empirical study of law were to be interpreted as fully captured in non-legal terms. An explicit focus on legal theory, in turn, can be helpful to legal empiricists—it may improve empirical legal scholarship—because it is likely to elucidate their background assumptions and thus integrate existing research agendas and suggest new directions. We pursue this claim of legal theory's added value by articulating a basis for legal theory and by showing how that basis illuminates both the application and design of empirical research on law. We argue that legal theory can be informative for legal scholars who seek to draw on empirical research. We further claim that legal theory can provide helpful guidance both for setting up an empirical research agenda on law and for designing research into specific topics.

The article is our contribution to the New Legal Realism Tenth Anniversary Conference dedicated to “Future Directions for Legal Empiricism.” That group's “manifesto,” published following its inaugural conference, declares that “[l]ike the original realists … new legal realist scholars bring together legal theory and empirical research to build a stronger foundation for understanding law and formulating legal policy” (Erlanger et al. Reference Erlanger, Grant, Larson, Mertz, Nourse and Wilkins2005, 337). However, the relationships between new legal realism (NLR) and legal theory as well as between NLR and the original US legal realism (to which we refer below simply as “legal realism”) have been more complicated. Thus, while we are familiar with (only) one sustained attempt to distill a set of legal-theoretical propositions from the “varieties of new legal realism” (Nourse and Shaffer Reference Nourse and Shaffer2009), we recall the NLR manifesto's “rejection of theory-driven orthodoxies” and its insistence that legal research should “begin[] with practice” (Erlanger et al. Reference Erlanger, Grant, Larson, Mertz, Nourse and Wilkins2005, 345). This proposition has been recently fortified by William Twining's call for propagating a “hedgehog concept” of “one Big Idea,” namely: “the importance of the empirical dimensions of law and justice as part of understanding law” (Twining Reference Twining, Macauley, Mertz and Mitchell2016, 130). Similarly, against NLR's continuity with legal realism (e.g., Erlanger et al. Reference Erlanger, Grant, Larson, Mertz, Nourse and Wilkins2005; Nourse and Shaffer Reference Nourse and Shaffer2009), some of the leaders of legal empiricism and prominent students of realism have insisted on deep differences between empirical and theoretical research (Macaulay Reference Macaulay2005; Leiter Reference Leiter2013; Twining Reference Twining, Macauley, Mertz and Mitchell2016, 130). These statements leave much room for interpretation. We think that to the extent they represent a withdrawal from theory or an attempt to drive a wedge between legal theory and empirical legal study, they are unfortunate and perhaps misguided. To us, it seems clear that the future of legal theory and the future of empirical legal study are intertwined, and we think the connection will be better developed if brought out into the open, articulated, and even fought about.

To vindicate this point we use legal realism as our central example of legal theory and study its relationship to empirical inquiries of the law. As we understand it, legal realism advances a view of law as a going set of institutions distinguished by the difficult accommodation of three constitutive yet irresolvable tensions: between power and reason, science and craft, and tradition and progress (Dagan Reference Dagan2013). We do not contend that legal realism exhausts legal theory. However, we use realism as our central example because the realist view of law helpfully highlights the most defining feature of any legal theory: its interrogation of law as a set of coercive normative institutions. These three components—law's coercion, its normativity, and the institutional settings in which it is manifested—take various forms that generate complex and contested relations. Unlike other types of social norms, law is always instantiated institutionally; unlike morality, it always involves some form of coercion; and unlike brute violence, law always claims authority premised on its compliance with normative standards. Hewing close to these three components gives legal theory enough analytical traction to be useful for both the use and the design of empirical legal research. Indeed, part of what we are pursuing in these pages is an attempt to vindicate the significance of that characterization of legal theory by demonstrating how it can fruitfully inform the empirical study of law.

The structure of this article, which also introduces this issue as a whole, is straightforward. We begin with a brief synopsis of the articles that follow in this issue. We then provide a sketch of our account of legal theory, followed by some reflections on the ways legal theory might inform empirical analysis. The last two sections of this article put these theoretical claims about the practical significance of legal theory to the test by examining two major topics of empirical interest that are addressed by the articles in this symposium: legal compliance, and equality and discrimination.

This Issue

“Elements of the Support Structure for Campaign Finance Litigation in the Roberts Court” by Ann Southworth analyzes the litigation teams, organizational ties, and funding sources for two major campaign finance cases decided by the Supreme Court. Southworth traces organizations' mission statements, the causes they typically advance, and their major funding sources. Regarding the lawyers themselves, she maps campaign contributions, employment history, and participation in events held by advocacy groups. She notes that while there are a few participants who cross lines now and then, for the most part the organizations, the lawyers, and the funders are part of well-defined and highly polarized camps.

“Lawyers and Social Movements: An Integrated Theory,” by Scott Cummings, reframes the discussion over the relationship between lawyers and movements for social change. It argues that the division between two strands of legal research—one dealing with how lawyers use the law to advance causes; the other with the question of what legal reform can actually achieve in advancing those causes—has made it difficult to evaluate lawyers' contributions to social movements. The division feeds into foundational critiques of legal mobilization and unwittingly contributes to reproducing the critiques. Cummings advances an integrated theory that treats how and what together, suggesting that combining the questions is a key to a better understanding of what lawyers can offer movements, and especially a better comparative framework for evaluating the challenges for pursuing movement goals generally. As we detail later, both Cummings's and Southworth's articles showcases our point about the institutional instantiation of law, and the multilevel analysis that institutions call for.Footnote ¹

In “Data Breach, Privacy, and Cyber Liability Insurance: How Insurance Companies Act as Compliance Managers for Businesses,” Shauhin Talesh explores how insurance companies play a critical and, as yet, unrecognized role in assisting organizations with complying with privacy laws and dealing with cyber theft. Talesh draws from and contributes to two literatures that examine organizational compliance with law in different ways: new institutional organizational sociology studies of how organizations respond to legal regulation and socio-legal insurance scholars' research on how institutions govern through risk. Through participant observation at cyber liability insurance conferences, interviews, and content analysis of insurer loss prevention manuals and risk management services, his study bridges these two literatures and highlights how the insurance field acts as a compliance manager for organizations dealing with cyber security threats. Well beyond pooling and transferring risk, insurance companies offer cyber liability insurance and a series of unique risk-management services that influence the form of compliance of organizations with privacy laws. As we argue below, Talesh shows that insurers acting as compliance managers actually shape the meaning of complying with law, and people's understanding of their obligations regarding a range of normative questions.Footnote ²

Finally, in “Law School Climates: Job Satisfaction Among Tenured US Law Professors,” Katherine Barnes and Elizabeth Mertz use a mixed method research design to explore the job satisfaction of US law professors. A survey of a nationally representative sample of tenured American law professors and in-depth interviews of a subsample of them complement each other to better understand their job experiences. Altogether, the tenured professors who participated in the study were satisfied with their jobs. Nonetheless, professors of color and female professors were less satisfied compared to their equally situated white male colleagues. The analyses of both the survey and interviews suggest that both the institutional culture and feelings of lack of voice and respect explain the sex and race differences in job satisfaction among law professors. In our brief account of the interaction between legal theory and equality and discrimination studies, Barnes and Mertz's article highlights the usefulness of combined methodologies: different methodologies are appropriate for different levels of analysis.Footnote ³

Legal Theory

As we noted at the outset, empirical research on the law is surely informed by social science theories, albeit not legal theory, so the first question we must ask is whether legal theory has any distinctive value to offer. Mainstream contemporary understandings of law, which view law as a means to social ends (more on this later), may seem to invite a negative answer. They may suggest that there is nothing distinctive about law, and that therefore the other disciplines that deal with exploring social interactions or achieving social ends would do all the theoretical work in investigating both the background and the effects of law. This is probably what Richard Posner has in mind when he writes that “as the only approaches to a genuinely scientific conception of law are those that come from other disciplines, such as economics, sociology, and psychology, it is appropriate when speaking of ‘legal theory’ at large to confine the term to theories that come from outside law” (Posner Reference Posner2001, 3).

There are two ways to read Posner's claim. The less interesting reading is that law is transparent, that it is simply reducible to social control writ large and therefore merely an accidental locus of investigation that could have been replaced by any other means. This seems like a fairly obvious error because, in fact, the means of achieving social control make a big difference. If armed brigands collect money from me once a year (even if they do so predictably), this is different from living under a statutory regime under which I am required to pay taxes.Footnote ⁴ Likewise, there is a difference between outlawing cigarettes and surreptitiously hypnotizing people when they go to the movies to get them to quit.Footnote ⁵ Similar results do not make these differing approaches to achieving social ends equivalent.Footnote ⁶ Law is distinctive here, and (internal) theories of law contribute to understanding the phenomenon. Missing that entirely would be a significant blindness. Hence, the other, more plausible way of reading the claim that legal theory must be external in which differences in exploring social interactions and modes of meeting social ends indeed exist, but that they can be theoretically analyzed and evaluated only by other disciplines. This version of the claim appears to enjoy some currency among sophisticated participant-observers of legal academia (e.g., Priest Reference Priest1983; Van Zandt Reference Van Zandt2003), but we think it is ultimately unconvincing.

Legal theory, we argue, follows jurisprudence in interrogating the law as a set of coercive normative institutions. The upshot of this approach would be a recognition that an interdisciplinary analysis of law in general, and theory-informed empirical legal research more particularly—as opposed to the application of non-legal disciplinary methodology or theory to the data of law—must rely (explicitly or implicitly) on both a theory of the way law's power and its normativity align, and an account of the way in which this discursive cohabitation manifests itself institutionally. This is admittedly a contestable way to characterize the theoretical core of law as an academic discipline, which two of us have tried to defend elsewhere (Dagan and Kreitner Reference Dagan and Kreitner2011). For our current purpose, we need not rehearse this effort. Rather, it is enough to outline briefly, in this section, the main components of this understanding of legal theory—coerciveness, normativity, institutions—and then, in the next sections of this article, to reflect on and illustrate the implications on legal empiricism. We acknowledge that these three components of legal theory are surely not news to the social science on law. Social scientists offer important insights—some of which we mention below—on all three fronts: the descriptive study of institutions, of power, and of norms occupies significant parts of the social sciences. However, notwithstanding this overlap, legal theory has a distinctive signature, which lies at its simultaneous focus not only on what the law is but also on the standards by which law should be judged (Dickson Reference Dickson, Del Mar and Bankowski2009, 169). For legal theorists, the three features of law—coerciveness, normativity, institutions—and their complicated interactions imply answers to both questions.

Legal theories of the relationship between law's coerciveness and its normativity come in many flavors, with Kantian and natural law theories on one end, Marxist theories on the other, and other schools lodged in between. (Recall, for example, how H. L. A. Hart distinguished himself from John Austin by insisting that Austin overemphasized law's coercive or command dimension at the expense of its normativity [Austin Reference Austin and Hart1832; Hart Reference Hart1961].) While the various theories along this spectrum provide different answers to the question of such a relationship, we think that no legal theory could ignore it. So, for example, Kantian legal theories (or natural law theories) take the possibility of coercion as a mandate for reason to be the sole motivator in law, claiming that for there to be law, all power must be subjugated to reason or, more specifically, the constitutive possibility of the use of force must be subordinated to legitimating reason (Ripstein Reference Ripstein2009). By contrast, some Marxist theories maintain that while law exhibits an internally consistent and reasoned framework, the entire mode of reason flows from and is dependent on (economic) power, namely: that in actuality, law's reason serves power (Pashukanis [Reference Pashukanis1924] 2002). Legal theorists taking their inspiration from legal realism deny the coherence assumed by both Kantians and Marxists, and stake out an alternative position according to which neither element dominates. Reason and power exist as vectors in a field of forces, and the contest among them is never exhausted (Dagan Reference Dagan2013).

The realist view of law offers an articulation of the idea that power and reason are both endemic to law, while at the same time appreciating the difficulties of their cohabitation. Legal realism is preoccupied with law's power not only because judgments prescribed by law's carriers can recruit the state's force to back them, but also because of the institutional and discursive features that disguise or downplay the element of force.Footnote ⁷ Therefore, realism recommends a hermeneutic of suspicion regarding the reasons offered by lawmakers (or those appealing to them); it recognizes these (in part) as instrumental arguments often generated precisely to protect the powerful from claims to substantive justice. But by the same token, legal realism eschews analyzing law only in terms of parochial interests or power politics. It recognizes that modes of legal reasoning—substantive and technical, abstract and contextual—often constrain the sense of choice available to legal decision makers in directions that transcend their self- and group interest. In the best case, legal reasoning must aspire to appeal beyond the parochial, and instances wherein argumentation is exposed as a cover for interest are treated as cases of abuse (Kennedy Reference Kennedy2014).

Legal realism is just one perspective, but our sense is that the realist view of law lays the groundwork for a wide range of familiar current legal theory. We do not claim that the legal scholar's theoretical contribution to the empirical study of law must follow the realist legacy or that all legal empiricists must become legal realists. We have highlighted the realist view here to note an important distinction between, on the one hand, the use of law as raw material for the application of a non-legal disciplinary methodology or theory, and, on the other hand, an interdisciplinary analysis of law: the latter is informed by or at least embedded (explicitly or implicitly) in a theoretical approach to the relationship between law's coerciveness and its normativity. Different positions on these questions will generate different sets of empirical questions, yielding distinct research agendas.

A further contribution of legal theory to the interdisciplinary analysis of law, including legal empiricism, relies on the understanding of law as a set of going institutions; an enterprise that cannot be reduced to its constituent parts. Although many socio-legal empirical scholars approach the law from an institutionalist perspective, they tend to do it descriptively and to highlight the features of law as a social institution (Moore Reference Moore1986; Edelman, Uggen, and Erlanger Reference Edelman, Uggen and Erlanger1999; Edelman et al. Reference Edelman, Krieger, Eliason, Albiston and Mellema2011; Albiston Reference Albiston2010). Legal theorists approach the law internally and tend to highlight the features and expertise that make the law and the legal players unique.

Some legal theorists focus on certain institutional features of law such as objectivity enhancing procedures in common law adjudication and the skill of lawyers in capturing factual subtleties in different types of cases, claiming that these features align with the core of law's legitimacy (Dagan and Kreitner Reference Dagan and Kreitner2011, 677–80). Scholars intent on elucidating this aspect of law would design empirical research to test the possible differences between lawyers and lay participants or observers in understanding fact situations, using experiments or statistical analysis of actual observations (Solan and Darley Reference Solan and Darley2001). Others insist that thinking like a lawyer requires initiation into a particular tradition of thinking, and that the tradition places limits on law's democratic aspirations (Constable Reference Constable2005; Mertz, Reference Mertz2007). A different set of empirical investigations could support this proposition. Our point again is not to adjudicate between these rival views, but to show that attention to the institutional setting in which legal questions are analyzed is the kind of distinctive disciplinary element in legal analysis that can make the interdisciplinary game worth the candle.

This institutional analysis is often unwieldy because of the complexities of every legal institution and their multiplicity. Notwithstanding the traditional jurisprudential focus on adjudication, legal realism opts for a style of jurisprudence that goes beyond adjudication to consider other arenas replete with lawmaking, law applying, law interpreting, and law developing functions (Kreitner Reference Kreitner2010). Some of these institutions actually enforce norms (courts; prosecutors; administrative agencies); some are hierarchical and rule based with readily identifiable agents involved in norm obedience (corporate counsel; tax advisors; etc.); and some are webs of social norms, including the norm of deference to the symbolic power of law. Legal theorists with an institutional focus bring to the interdisciplinary table a non-reductive account of both the discrete parts of this legal universe and of their interactions and possible substitutability. In short, legal theories have differing takes on the importance of institutions, and on the question of how fine grained an analysis of institutions needs to be to generate insights. On the face of things, it seems likely that the more a theory is focused on institutions, the more use it will have for empirical analysis (cf. Pildes Reference Pildes2013). The question of institutional focus, then, is a straightforward illustration that legal theory can drive the design of legal research, empirical legal research included.

Indeed, the core of modern legal theory is the understanding of law as a set of coercive normative institutions, and while various disciplines have something to say about each of these elements, no other discipline specializes in treating the conjuncture. Attention to that combination is the hallmark of legal theory, and characterizes a wide range of theories with different takes on the relationship among the elements. This might, of course, be a semantic argument without much interest: one could understand the common claim as holding that legal theorists gain their insight about the relationship between coercion and normativity from political philosophy, their insights about normativity and institutions from sociology, their insights about institutions from economics, and so on. Although there is some truth to the idea that insights about each element range across disciplines, this says little about the need for and power of internal legal theory. The primary reason is, as mentioned above, that no other discipline thematizes and concentrates on this combination, a jurisprudential combination that is bigger than the sum of its parts. Additionally, the fact that insights travel and that elements of one discipline rely (implicitly or explicitly) on the insights of another is not an argument against the importance or independence or robustness of the discipline or theoretical endeavor. To recall the common example: economics relies on a vision of rationality whose core development lies in the field of psychology; and the psychological inquiry relies, in turn, on a metaphysics of personhood. This chain does not mean there is no internal economic theory, any more than it means that economists should be trained as metaphysicians in order to pursue their scholarly calling (cf. Penner Reference Penner2010, 328).

Legal theory—understanding law as a set of coercive normative institutions—gives scholars a theoretical angle and set of tools to evaluate legal phenomena. It generates perspectives on legal problems that no other discipline specializes in producing. Appreciating the general role of legal theory sets the stage for investigating its particular contribution to both the application of empirical studies of law and their design.

Applying Empirical Legal Studies

Because we think that understanding law through coercive normative institutions is helpful (in other words, regardless of contact with additional disciplines), we believe that this lens may also be helpful for reflective interdisciplinary legal analysis including—our focus here—empirical legal analysis. Much of the discussion that follows is dedicated to flesh out the implications of this perspective on active legal empiricism, studying the input of legal theory on both its agenda and the research design of specific topics in legal scholarship. However, before we hit that road, we offer a few observations about the use of empirical investigations in the study of law in general and the lessons of legal theory on the ways in which such empirical inquiries can be integrated into law more particularly.

As a threshold matter, it seems clear that different types of legal theory will have different needs regarding empirical investigation.Footnote ⁸ Although some jurisprudential scholars have questioned the overall interest of empirical investigation or sociological facts in general, all modern legal theories entail some role for empirical determinations. To begin with an extreme example, consider Hans Kelsen's Pure Theory of Law. The purity of the theory lies in its formality, and thus in its claim that the validity of norms in the system is recognized solely by analyzing the relationship among the norms. On the face of things, this requires only interpretive activity with no basis in empirical facts (i.e., a strict separation of facts and norms), but even Kelsen's theory acknowledges, at its very basis, that the legal system as a whole cannot generate legal norms if it does not enjoy effectivity. Thus, a question of fact about whether the system as a whole is effective—whether there are citizens who consider the system binding, and institutions of enforcement that use its norms in making decisions and carrying them out—must be answered affirmatively in order to engage in the work of legal cognition.Footnote ⁹ And what is true of the system is also true of individual norms: a legal norm, validly enacted, may be “deprived of its validity by ‘desuetudo.’” This deprivation of validity takes time, and only “an enduring lack of efficacy” in which a norm is “permanently inefficacious” will suffice to annul the norm (Kelsen Reference Kelsen1945, 119). These questions of efficacy are empirical questions. They may not be the kinds of empirical questions that require sophisticated statistical study, but they are questions of empirical fact relevant at the very heart of ideal formal theory.Footnote ¹⁰

To be sure, while even extremely formal theories of law have some need for empirical determinations, scholars committed to such theories will typically have only minimal recourse to empirical study of the law. The reason is that wholly formal theories of law eschew purposive analysis of legal arrangements. One way of understanding this is that these theories rebel against what they view as a reduction of law into a mode of social engineering (Weinrib Reference Weinrib1995; Ripstein Reference Ripstein2009).Footnote ¹¹

However, as noted, most current accounts of law follow Rudolph von Jhering's announcement of Law as a Means to an End. This proposition may have been an innovation when he published his book in 1877, but it has become a taken-for-granted commonplace for most theorists today, and once law becomes tightly aligned with social policy, legal theory will almost of necessity be on the lookout for sources of knowledge about the state of the world on which legal norms and institutions impact and about the possible effects of the law on that world. This search for knowledge about the world provides a clear (and longstanding) explanation for interdisciplinary study of the law, and on the face of things, for empirical study in particular. It is thus unsurprising that legal scholars increasingly look to empirical studies to support particular understandings of the law and proposals for reform.Footnote ¹² Legal theory can aid in making difficult choices about what kinds of empirical studies are useful, and about how they should be engaged. An empirically minded legal scholar faces two challenges—selection and integration—and at times also a third one of synthesis (Dagan and Kreitner Reference Dagan2014).

There are many empirical perspectives on any given legal topic, each with its own potential contribution, and the analyst needs to be able to select the disciplinary lens that seems most promising for the task at hand and for integration into legal analysis. Selection is not only about what to include, but also what to exclude. The call to include the relevant is easy: legal scholars often like to be as inclusive as possible regarding interdisciplinarity not simply because other disciplines “contribute” to legal discourse, but because interdisciplinarity has become a defining feature of much (including much of the best) legal scholarship. What is true of interdisciplinarity generally is true of empirical studies as well, but we think there are (contingent, problematic, and arguable) limits to interdisciplinarity. Acknowledging practical limits, such as accessibility, is a simple first step. Slightly more difficult is the recognition that legal analysis should not incorporate pieces of ostensibly relevant materials or conclusions of neighboring disciplines, or that it should modify these inputs to allow for digestion in the metabolism of legal scholarship.

Indeed, integrating empirical insight into legal discourse requires translation. This is often challenging, given the different criteria of validity and weight of empirical findings, a difficulty that increases as the empirical apparatus becomes more technically nuanced, and thus requires more effort for integration or even communication to the uninitiated. To put it bluntly and concretely, counting is different from regression analysis. Just to note a common pitfall: statistical analysis is often concerned with “significance,” which is a way of describing the level of confidence that a particular aspect of the data makes a difference. But statistical significance does not say anything about how big a difference the examined aspect makes. That is obvious to someone versed in statistics, but puzzling for someone relying on the nonprofessional use of the word significance. The NLR Project's website is therefore correct in criticizing the assumption—typifying the “renewed interest” of legal academics in social sciences—that “scholars and policymakers … can simply pick up the findings of social science and use them in legal settings.” Because there are “vast differences in goals and epistemologies that characterize diverse disciplines,” the assumption of transparency is mistaken. Hence, one of the core missions of the NLR: to “pay[] systematic attention to this process of translation and integration” in order to “build a more rigorous and informed framework” that is necessary for bringing “the best of current social science” to bear on “important policy issues of our day” (The New Legal Realism Project 2011).

To select and integrate, academic lawyers need to be able to translate, and thus must be able to read and understand social science, which explains and justifies why legal empiricists seek to incorporate social scientific literacy into legal education (The New Legal Realism Project 2011). But translation is complex. In some cases, legal scholars, for whom law is their professional mother-tongue, can rely on their know-how, namely: the almost intuitive grasp of situations as legal problems, or what some call thinking like a lawyer. But, as is (almost) always the case for academic purposes, such implicit knowledge has a more reflective or theoretical rendition, which refines propositions and elucidates dilemmas. Legal theory elucidates the distinctive theoretical voice legal scholars bring to the interdisciplinary table. The examples below flesh out this claim, which when formulated abstractly, might sound like an empty platitude. The upshot of the examples, however, can be stated up front: the use of empirical studies regarding law almost always passes through a set of assumptions about the core questions of legal theory, including the relationship between coercion and normativity, and the instantiation of that relationship within institutions.

We cannot offer here a full-blown account of the implications of this proposition. But two caveats regarding the application of empirical legal analysis that dovetail with the perspective on law as coercive normative institutions, can illustrate our claim:

1. 1. Appreciating the institutional capacity of pertinent legal actors may suggest that scientifically accurate, but overly complicated, formulae may in fact be counterproductive if translated too readily into the normative realm of prescription. Highly technical analyses may be essential for generating insights unavailable to the naked eye. At the same time, they often require the kinds of idealizations that counsel extreme caution when shifting gears from penetrating insight to broad-based normative conclusions.

2. 2. Respecting law's normativity and recognizing the potential threats of its coerciveness imply that legal prescriptions must meet the demanding justificatory constraints of both the rule of law and public reason. We are not troubled by the use of other disciplines to expand our view of the legal phenomena in question; indeed, there are examples in which interdisciplinary work participates in the evolution of the internal language of law and legal theory. But not all cases take this form; hence the concern with the pitfalls of generating normative conclusions from scholarly methods or forms of analysis too distant from normative inquiry.

We think that the occasional unease of legal scholars regarding importations of interdisciplinary analyses of law—including complex statistical analysis—into legal discourse implicitly reflects these concerns, in a way that sometimes echoes the concerns of legal practitioners in considering expertise. We do not believe there will ever be a technical formula for selection and integration. Deciding on the scope of importation of empirical studies seems to us like an ongoing challenge.

Beyond selection and integration lies the possibility, and at times the imperative, of synthesis. The scope of extralegal importation is a function of the type of task at hand. Many types of legal analyses incorporate a subset of the universe of extralegal, potentially pertinent insights, typically ones that derive from a single neighboring discipline. But as a project shifts toward the evaluative (justificatory or reformist) type, the pressure to incorporate and synthesize the lessons of all the other disciplines that claim to have significant implications strengthens because the responsibility of potentially affecting people's lives forces upon the analyst a duty to doubt as well as a duty to decide, obligations one cannot discharge by endorsing any single perspective on law. Unlike questions regarding coercive normative institutions, synthesis and the methodological challenges it invokes are not part of the core of legal theory. But because a significant subset of legal analyses involve justificatory or reformist inquiries, the effort to engage, incorporate, and synthesize the lessons—including the empirical findings—of these other perspectives is a characteristic, albeit not necessary, feature of legal theory. This feature, which will be exemplified in our section on equality and discrimination, may explain why, typically, good legal theory does not simply accept the surroundings of the other discourses about law in patient resignation; rather, it doggedly seeks out interactions with those surroundings in which law and legal phenomena are investigated using theories and methodologies of the social sciences and the humanities.

Designing Empirical Legal Research

The benefits of legal theory for the application of empirical research on law also imply a similarly important role legal theory can play in designing empirical legal studies. To appreciate this role, it is helpful to recall Thomas Ulen's narrative of the development of empirical work in the study of law (Ulen Reference Ulen2002). Ulen's starting point (which we, of course, contest) is “the puzzling lack of theory in law.” But this, he notes, did not preclude the demand for empirical work thanks to the legal realists' insistence that law's consequences “should form part of the evaluation of [its] worthiness.” And yet, this limited “ground for empirical work has led, until recently, to a particular kind of empirical work,” which has mostly been “descriptive rather than analytical.” This limitation, he claims, is not surprising because “the explanatory work of empirical work,” which is “its highest calling,” requires a theory with some core propositions that then can be—assuming widely accepted standards of evaluation—helpfully confirmed, rejected, or refined by empirical work. We think that Ulen based his understanding of the situation of empirical legal study on a mistaken assumption, according to which legal scholarship proceeds on the basis of a pre-theoretical understanding of law, or what we called above the intuitive grasp of situations as legal problems. While this assumption may characterize some legal scholarship, we believe it is far less dominant than Ulen's account would lead one to believe.

Ulen was optimistic about the future of empirical legal studies thanks to the work of lawyer economists who have “articulated a comprehensive theory—both positive and normative”—and have thus prompted assiduous explorations that examine whether “law is [in fact] efficient.” Ulen refrained from arguing that this is the one and only “method of bringing a comprehensive theory to the study of law,” but he did note the “great attractions and successes” of such “use of microeconomic theory,” adding that competing theories “do not flourish,” and that while law and society “has generated seminal findings,” it still awaits a comprehensive theory of law that can be empirically evaluated. We think that this assessment represents a somewhat insular view from within law and economics, and that it sells legal theory short. The importance of legal theory for empirical study does not require consensus on a comprehensive theory of law. Perhaps the opposite is true. Competing theories, which all focus (as noted) on the “triple” nature of the law as a set of coercive normative institutions, generate different sets of empirical questions, each attempting to test those aspects of the theory that are open to empirical proof or support. Thus, consider the following illustrations: a scholar committed to an expressive theory of law will be interested in empirical examination of how legal messages are conveyed, whether and how addressees of law absorb those messages, reject or adopt them as their own, and reconfigure (or not) their assessments of right behavior. A scholar committed to a feminist theory of law will often be interested in empirical examination of those same issues, but may also be interested in examination of outcomes of behavior and their impact on gender relations, beyond the question of conscious reception of messages. In other words, it is legal theory that will guide scholars in generating empirical questions.

Again, fuller elaboration of this point will have to wait for the development of examples below, but the lesson should be clear even now. Theoretical understandings of law, as opposed to non-reflective assumptions about what the law is or how it might impact on society, can be informative and useful in designing empirical studies. On this level, we might say provocatively that there is no purely descriptive empirical research. Theory, whether well-articulated or mostly implicit, will always drive the kinds of questions we submit to empirical inquiry. Our simple claim here is that explicitly articulated theory will usually do a better job of driving that research than assumed understandings. Ulen, writing over a decade ago, was right in predicting the deepening of economics-inspired empirical work on law. His celebration of empirical research based in efficiency theory has been amplified in the passing years, but that theory is (as he maintained) one of economics, rather than law. That should not, however, lead us to believe that legal theory is not similarly crucial. Ulen noted legal realists' contribution to the growing significance of consequences in legal discourse, but he drew too limited a lesson from the insight. The idea that legal theory was necessary for making facts intelligible and for justifying the study of one set of legal facts rather than another is a venerable realist insight, and still relevant for all those hoping that their empirical work will not yield “a horrid wilderness of useless statistics” (Cohen Reference Cohen1935; Dagan Reference Dagan2013, 47–48).

Ulen was wrong, we think, in one further proposition he made, albeit in passing. He predicted that “the increasing theoretization of law” that can prompt further empirical inquiries “will almost certainly not be jurisprudential” because competing propositions of this sort are “resolvable, if at all, only by hypothetical deductive argumentation” (Ulen Reference Ulen2002, 912). While it is true that empirical inquiry cannot be the sole arbiter in such conflicts, we argue that the distinctive theoretical voice of legal scholars is in fact best captured around the central and persistent questions of jurisprudence, and that empirical studies of law can both contribute to and benefit from scholarly inquiry into these questions. Indeed, various existing studies—of both legal empiricists and non-legal empiricists—implicitly reflect many of the insights that legal theory offers in developing research agendas and designs. Our mission in what follows is to render this theoretical infrastructure explicit in order to organize and clarify the input of legal theory for empirical legal studies. In addition, we hope to highlight theoretical lessons that have been underutilized by empiricists.

We begin with explaining how taking into account the nature of the law as a set of coercive normative institutions would benefit agenda setting and research design and why neglecting any one of these three aspects would result in a failure to account fully for law's manifestations or implications. In the next two sections we turn to illustrate these observations with respect to the extensive empirical research of two topics: the first—legal compliance—is an obvious choice given its immediate connection to the nature of law; our second case study—equality and discrimination—both demonstrates the relevance of legal theory to specific inquiries and provides an example for the significance of synthesis to reform-minded empirical inquiries. Our claim throughout is that the prism of legal theory will often be beneficial by offering important ways of integrating existing empirical research and suggesting further possibilities for useful new directions.

* * *

The understanding of law as a set of coercive normative institutions aligns with law and society scholarship in suggesting that theory-informed empirical studies should address law as a system operating in interaction with other systems.Footnote ¹³ But whereas socio-legal scholars tend to approach the law descriptively and from an external perspective, here we offer to supplement this view with the lessons learned from legal theorists who add normative and internal dimensions to the understanding of the law.

The legal system consists both of the normative schemas and reasoning by which law is enacted and reenacted and of the material distributions of behaviors, resources, and power that result. The interaction of power and reason implies that these distributions will affect the normative schemas and reasoning, which are in turn understood to both legitimize and constrain law's material effects. Singling out coerciveness, normativity, and institutions helps identify the salient features of law's operation, but the whole is greater than the sum of the parts and, ultimately, the parts are not completely distinguishable. Law's coercive action is always conditioned by its normativity and is always instantiated institutionally. The mix of coerciveness, normativity, and institutions in any particular account is in large measure a question of perspective and focus. What we have to say about each of those features separately, then, is always an intermediate step. The hope is to draw out salient points of focus for research, all the while acknowledging that in law, coercive normative institutions form a complex, and none of the components has any pure existence completely distinct from the others.

Understanding law as a set of institutions that operate simultaneously requires attention to three levels of analysis—the micro, the macro, and the meso—as well as to the interactions among them. The micro level consists of individual players like legislators, administrators, regulators, judges, lawyers, jury members, police officers, plaintiffs, defendants, and the accused, as well as their interests, beliefs, moral convictions, professional guidelines, and taken-for-granted ways of doing things. The macro level consists of legal institutions like courts, legislatures, administrative agencies, local governments, or the police. Finally, the meso level comprises the interactions between legal players and thus consists of phenomena like trials, administrative proceedings, regulatory hearings, legislative procedure, or police activities. The three levels of analysis are understood to interact with each other to enact the law, so that the beliefs and behaviors of players at the micro level—like those of judges and lawyers, for example—are both affected by processes at the other levels (like court decisions and legislation, but also the procedures for legislative deliberation) and in return, affect them.

This does not suggest, of course, that every research project should seek to explain and map all processes at all three levels. Because processes at different levels call for different research methodologies, different empirical studies are likely to focus on different levels or aspects of an issue. Thus, exploring processes at the micro level and focusing on individual players would often involve methods designed to assess the beliefs and interests of individuals. Qualitative studies, surveys, and experiments are particularly suited to this set of inquiries. Assessing the macro level, in turn, would tend to involve methods designed to study broader-ranging effects of institutional action, and thus quantitative analysis would often be very useful. Finally, processes at the meso level in which players interact with each other would tend to involve both experiments and observations.Footnote ¹⁴

Southworth's article in this issue, for example, brings together micro- and meso-level analyses. On the one hand, the research digs into the personal details of particular members of litigation teams, from education and employment history all the way through involvement in political causes and personal campaign contributions. And Southworth notes that future research could be directed further on the micro level through interviews of the participating lawyers. On the other hand, the project of mapping the individuals is geared less toward understanding their personal motives, and more toward generating a coherent picture of an organizational field, where types of actors are aligned with types of causes. The amount of data analyzed just for two cases is fairly staggering, so imagining a broader map of the organizational field, scaled up to consider organizational strategies, presents a daunting challenge.

Each such empirical study is valuable in and of itself, but their potential to complement each other may be even more important. This means that a full appreciation of what the law does—in general or in any given specific field (say, marriage law)—requires attention to the cumulative effect of these levels. We could speculate that in a more ideal world, the design of every piece of research would be sensitive to processes operating at each level and across levels of analysis. Finally, on an even broader scale, the institutional lens suggests that these institutions do not operate in a vacuum, but interact with other systems like the market, religion, and the family.Footnote ¹⁵ It follows that interrogation of the relationships between law and these interacting systems is also desirable; at times it may even be indispensable (e.g., Stern Reference Stern2016).

Cummings's article on “Lawyers and Social Movements” takes on the challenge of combining levels of analysis and in so doing generates focused, productive results. Cummings shows that the two strands of literature that have taken lawyer involvement in social movements seriously have been affected by a common malady: that of considering lawyers' intervention in isolation, rather than in comparison to alternatives. One strand of the literature ignores the parallel difficulties in assuring accountable representation between lawyers and movement leaders more generally. The other strand fails to account for the downsides of non-legal movement action (or inaction), making it seem as if the pitfalls of litigation are not balanced by similar problems for other modes of action. By considering how lawyers become involved (comparing that with other movement leaders) at the same time as what legal reform achieves (comparing that and its limitations with non-legal activism), Cummings is able to frame the options for social movements in a much more complete manner. The result is doubly significant: first, it sets out a host of general considerations for social movements that were not quite visible when the focus was too narrowly defined around the pitfalls of lawyerly involvement; second, it suggests that critiques of lawyers' contributions to social movements may be overdrawn. The upshot is that good theory might pave the way for renewed lawyerly energy in social movements.

When focusing on a set of coercive institutions, empirical studies are called on to consider both mechanisms and outcomes. Attention to coercive power as a mechanism invites empiricists to account for naked coercion as well as for its more subtle forms. Research agendas may therefore focus on exploring power and the ways in which it is disguised or obscured in issues as diverse as the various mechanisms and procedures affecting punishment, the incentives of legal players and their effects on outcome, or the hidden implications of gaps between the “law in the books” and the “law in action.”

Attention to power as an outcome, in turn, calls for the investigation of law's effects on people's behavior and beliefs, and of the resulting distribution of power. These kinds of outcomes often (perhaps more often than not) require some account of the interaction between coercion and normativity. Specifically, empiricists are well placed to examine law's blunt incentives as well as the way law affects normative evaluations, social norms, and cognitive scripts. While the law often affects people's decisions by making some actions more costly than others, such blunt incentives by no means exhaust its potential effects. Law may affect people's normative evaluations because, often, people who believe in its legitimacy ascribe normative value to its prescriptions, so that when an action is defined as illegal it also tends to be evaluated as morally wrong. The cumulative consequences of law's blunt incentives and of its effects on what many people evaluate as right and wrong may also come about indirectly: by generating new social norms and changing existing ones. The process is cyclical, so that new social norms may affect the law, which in turn again acts on social norms. Lastly, the law also affects people's behaviors and beliefs by providing cognitive schemas that define the ways in which things are done, so that over time the legal way of doing things becomes “taken for granted” and most people follow it without any questioning (Scott 1995).

All the ways in which the law affects behaviors and beliefs entail significant distributive outcomes (in terms of both material and nonmaterial resources), but not all these distributive implications are equally transparent. Thus, when the law directly affects behavior by changing people's blunt incentives, these consequences are relatively easily observed. By contrast, when the law affects people's normative evaluations and cognitive schemas, distributional outcomes are more hidden. When empirically studying power as a hidden outcome, methodological challenges may arise that require the development of creative ways for assessing people's beliefs and cognitive scripts.

Finally, the focus on law's normativity raises two complementary aspects of the investigation: the constraining effects of legal normativity and its functions as a legitimizing mechanism.

Law's normativity often constrains lawmakers' sense of choice by providing an internally consistent and reasoned schema for legal decision making that in some cases directs lawmakers to transcend their own interests. The schema consists of specific legal doctrines that routinely define the possibilities of lawmakers; it also includes underlying abstract normative principles that come into play in those (relatively infrequent) moments when the inertia of “mechanical jurisprudence” is broken and is replaced with “a jurisprudence of ends” (Pound Reference Pound1908).

It thus follows that in order to understand law—to capture its modes of operation and its typical paths of evolution—empirical studies can utilize qualitative and quantitative research designs that explore legal argumentation and reasoning as data sources. Both specific legal doctrines (or, more precisely, their conventional understanding within a legal community in a given time and place) and abstract normative principles should be explored, and attention should be paid to the different occasions in which they are invoked as well as to their various modi operandi. Thus, for example, empirical studies may utilize legal precedents and textbooks to appreciate the possibilities and constraints that lawmakers face when making their decisions.

Law's normativity constrains lawmakers, but empowers them at the same time. It also tends to grant legitimacy to the distributions of material and nonmaterial resources that law implements. Legitimating action can work in different ways. At one pole of the spectrum, legitimizing comes through reasoned argument, convincing legal subjects of its overall desirability; at the other pole, legal categories become the taken-for-granted backdrop to social arrangements, including social arrangements that might be oppressive. Categories become reified. Here, the emphasis is not on instances where distributions are intentionally disguised or downplayed, but on instances where distribution is explicit, and law's normativity bolsters its perception as reasonable and appropriate, at times even obvious or inevitable. This effect need not necessarily condemn such doctrines; after all, many of the most valuable aspects of our lives are similarly reified. But it does imply that some pieces of the legal landscape become taken for granted and become immune to critique. One task of scholarship is to enliven critique in such situations. Exploring legitimization effects of normativity and the distributions they shield from critical inquiry requires empiricists to design, for example, longitudinal studies that examine changes in perceptions in the wake of legal change, or experiments in controlled settings that can study the effects such changes have on people's views.

* * *

Taking the view of the law as a set of coercive normative institutions seriously, we argued thus far, generates a guiding thread in developing a research agenda and in designing specific research. The directions suggested are surely not new to legal empiricists, but sometimes the most helpful theory of a given activity is one that is already implicit in that practice. Reflection on what we customarily do is useful because articulating our premises can help organize our practice and push it better to deliver on its implicit promise.

But does it work? In other words, does the approach to law as a set of coercive normative institutions help organize existing empirical research in a coherent way? Does it help show how disparate studies complement each other and enhance our understanding of their common subject matter? Would reliance on this theory highlight gaps in our understanding of law by exposing holes in our existing empirical knowledge? Addressing questions like these is the task of the following pages, in which we examine some of the empirical research on legal compliance and on equality and discrimination. Needless to say, we do not purport to provide in these few remaining pages a comprehensive, or even representative, review of the literature on these core topics of legal empiricism, but we hope that the quick sample of existing studies we cite and the directions for extensions we mention give at least a taste of both the implicit alliance of prevailing legal empiricism with legal theory and the potential benefits of rendering this alliance explicit.

Legal Compliance

We now proceed to investigate the input of legal theory to empirical studies by looking into the topic of legal compliance. The issue of compliance raises three distinct yet related questions: What amounts to complying with the law? What are the means by which the law solicits compliance from its subjects? And how does compliance affect the subject who complies? As it turns out, to answer these questions one needs to interrogate the three components legal theory highlights: the multiple institutional settings in which law is manifested, law's coercion, and its normativity.

To see the relevance of legal actors to our first question—of what does it mean to comply with legal prescriptions—we need to recall the most fundamental lessons of legal realism. As realists demonstrated, the availability of multiple potentially applicable doctrinal sources at each juncture in any given case renders pure doctrinalism hopelessly indeterminate. And yet, despite its many caricatures, legal realism does not challenge the felt stability of legal doctrine or its categories at a given time and place. Rather, it suggests that this stability rests on the convergence of lawyers' background understandings (Dagan Reference Dagan2015). In other words, in order to study empirically what amounts to complying with the law, one needs to probe into the processes by which the meaning of compliance with the law evolves, changes, and is diffused, and to the roles that legal institutions (such as courts and regulators) and legal actors (such as lawyers and other professionals) play in its generation. At times, the meaning of compliance is also affected by ideas carried from other sets of institutions (like the market and the family) with which legal institutions interact.

The literature on legal endogeneity provides an illuminating example. This body of research shows how as a response to the ambiguity of employment-related civil rights law, organizations have turned to other organizations around them to understand and develop the meaning of compliance with the law. Professionals (lawyers) played a powerful role in carrying ideas across organizations, leading to a process of mimicking. As a result, many organizations adopted structures such as grievance procedures and anti-harassment policies that over time affected the ways in which courts understood legal compliance. These structures and practices have become symbolic indicia of compliance with the law, regardless of their actual effectiveness (Edelman, Uggen, and Erlanger Reference Edelman, Uggen and Erlanger1999; Edelman et al. Reference Edelman, Krieger, Eliason, Albiston and Mellema2011).

Talesh's article offers an important contribution to this body of literature. His study shows that cyber insurers and the risk management services that accompany cyber liability insurance not only influence the meaning of compliance, but rather actively shape the way organizations deal with cyber threats and comply with privacy laws. As he frames it, cyber insurers—not your obvious dominant legal actor—are acting as external “compliance managers” of organizational behavior in preventing, detecting, and responding to data breaches and complying with various privacy laws. The difference between these two roles—affecting the meaning of compliance and actively managing it—and the passive role insurers play in the context of directors' and officers' professional liability may well be explained by the agency costs that plague the latter (Baker and Griffith Reference Baker and Griffith2010, 125–27). But Talesh's analysis raises an interesting question (maybe a subject for a follow-up inquiry) regarding the possible reasons that can explain why insurers take this managerial role in the cyber liability context as opposed to the more moderate role of shaping the meaning of compliance in others.

Second, dealing with the means by which the law solicits compliance from its subjects requires empiricists to focus on the coercive aspect of the law, namely, the processes by which the law affects people's behaviors and eventually their beliefs. Not surprisingly, there are many studies that belong to this line of investigation.

Some of these empirical inquiries examine the effects of blunt incentives on people's obedience to the law (Chambliss 1969; Salem and Bowers 1972; Gibbs 1975; Zimring and Hawkins 1975; Waldo and Chiricos 1975; Tittle 1980). Specifically, these studies examine whether perceived changes in the certainty and severity of punishment affect people's tendency to comply with the law. Here, the coercive and normative aspects of legal prescriptions are complexly entwined. Scholars interested in the effects of moral evaluations on compliance, for example, focus on instances in which people obey the law because they wish to do the morally right thing. In some of these cases, people believe that a legal prescription reflects a moral value and thus comply with the law because it protects a value they endorse, implying that they would behave similarly even in the absence of that legal prescription. But in other cases, people may obey the law because they think that obeying a legitimate law is the morally right thing to do. In these cases, they may be morally indifferent as to the value that law aims to protect, but comply with it because of moral beliefs regarding the value of compliance with the law (Meier and Johnson 1977; Grasmick and Green 1980; Robinson 1995; Paternoster and Simpson 1996; Robinson and Darley 1997). Finally, an emphasis on the law's cognitive effects on people's behaviors and beliefs has led scholars to study empirically the ways in which the law provides cognitive schemas, so that gradually compliance with the law becomes “taken for granted” and most people comply without questioning the law or consciously taking into account the other possible alternatives (Albiston Reference Albiston2010; Albiston et al. in progress).

These studies shed light on compliance with law and the means by which it is secured. But looking at these four forms—blunt incentives, normative evaluations, social norms, and cognitive scripts—invites empiricists to take a further step and also study their interrelations. Thus, for example, one could imagine empirical investigation of whether the severity and certainty of punishment affect not only blunt incentives but also people's moral evaluations of the law and whether that connection is one of alignment or tension (e.g., Gneezy and Rustichini Reference Gneezy and Rustichini2000). One could explore the ways in which existing cognitive scripts and social norms construct the meaning of law in concrete instances (e.g., Edelman, Erlanger, and Lande Reference Edelman, Erlanger and Lande1993; Nelson and Bridges Reference Nelson and Bridges1999).

Finally, turning to our third question—dealing with the effect of legal compliance on the obeying subject—requires that we focus on the normative aspect of the law, and particularly on the relationship between legitimacy and law's normativity. One body of existing empirical research relevant here is the literature on procedural justice. Works in this vein explore legitimating effects of the perceived fairness (due process) of the legal proceedings. These empirical investigations demonstrate how perceptions of procedural fairness lead people to view the law in general as legitimate and as a result to respect it and obey it (Tyler, Schulhofer, and Huq Reference Tyler, Schulhofer and Huq2010, Reference Tyler and Schulhofer2011; Tyler and Schulhofer Reference Tyler and Schulhofer2011).

These insights are obviously helpful in highlighting how distinct modes of compliance—one that comes after a fair process versus one that follows a process perceived as unfair—differently affect the complying subject. However, they do not provide a full picture of normativity as a legitimizing mechanism. Here, the emphasis is not on instances where people comply with the law because they believe compliance is morally right, but on ones where the legal rules change people's moral views.Footnote ¹⁶ To be sure, people may first comply because of blunt incentives, or because they believe in the general moral value of compliance, or because of social norms that were triggered by these two modes of legal compliance. But the legitimation hypothesis implies that over time they internalize the pertinent prescription to which they were initially indifferent into their own moral system, adopting it as their own, perceiving it as the morally right thing to do.

Empiricists focusing on the legitimizing aspect of normativity and wishing to study compliance could design empirical investigation into the processes leading people to associate “what is legal” with “what is morally right.” Studying how the law changes people's moral beliefs requires longitudinal evidence and methods that enable tracing not only conscious beliefs, but also unconscious ones. One of the most challenging questions that seems ripe for empirical investigation is what role normativity plays in effectiveness. In other words, there is much work to be done in exploring how formal prescriptions, practices, and structures take on the meaning of compliance, how they affect people's perceptions of “what is legal,” “what is morally right,” and how those perceptions contribute to effectiveness over time.

Equality and Discrimination

Looking at the input of legal theory to empirical studies of compliance was an easy task as the phenomenon of legal compliance directly involves the law and its effects on people's behaviors and beliefs. Naturally, investigating compliance can benefit from a theory of what the law is. We now proceed to a less obvious example to show how legal theory informs empirical legal studies, the example of equality and discrimination.

Understanding the law as a set of coercive normative institutions would lead empiricists to focus on the multiple ways in which antidiscrimination law affects people's behavior and beliefs and on whether it manages to reduce inequality. It would lead empiricists to investigate the various processes that operate simultaneously in which individual people, organizations, and social institutions all participate in transforming the meanings of the law and the cultural beliefs associated with it, and the effects such processes have on inequality (Gulati and Nielsen Reference Gulati and Nielsen2006). It would lead empiricists to investigate, for example, how social institutions and cultural beliefs operate to maintain inequality in times of legal change but also how they may sometimes provide opportunities for change (Albiston Reference Albiston2010); to investigate how the various legal players—courts, lawyers, and work organizations—all participate in generating the meaning of compliance with antidiscrimination law (Edelman, Uggen, and Erlanger Reference Edelman, Uggen and Erlanger1999; Edelman et al. Reference Edelman, Krieger, Eliason, Albiston and Mellema2011); and to explore which ways of complying with antidiscrimination law are more effective in reducing workplace inequality than others (Kalev, Dobbin, and Kelly Reference Kalev, Dobbin and Kelly2006).

But more than just illuminating the ways in which legal theory may inform our investigation of antidiscrimination law in action, the example of employment discrimination emphasizes the contribution of incorporating and synthesizing the lessons of other social science disciplines in order to capture, in all its complexity, a phenomenon central to people's core interests. The closer empirical legal studies come to policy analysis (and they often are quite intimate with such analysis), the more important it becomes to synthesize. Indeed, as noted, while the synthesis of extralegal perspective is not, strictly speaking, a defining feature of legal theory, it is a particularly characteristic feature thereof.

Employment discrimination is a complex social phenomenon that involves processes operating simultaneously at multiple levels of analysis. Consider the example of gender. Like other multilevel systems of inequality, the gender system involves cultural beliefs, stereotypes, and distributions of material and nonmaterial resources at the macro level (e.g., by the state), organizational structures, practices, and behaviors at the interactional level (like the hiring and promotion of employees), and selves and identities at the individual level (Ridgeway and Correll Reference Ridgeway and Correll2004). Processes at each level of analysis affect all other levels, so that gender inequality becomes over-determined.

The institutionalized cultural beliefs and stereotypes about gender at the macro level shape the expectations people have of themselves and others' competence and performance. These gendered expectations tend to affect both individuals' actual performance as well as their evaluations of the performance of others, resulting sometimes in gendered distributions of resources and power (like unequal employment outcomes for women and men on the basis of gendered performance and biases in evaluation). As a result, over time, gendered performance and evaluations reinforce the institutionalized cultural beliefs and stereotypes to maintain gender inequality (Ridgeway Reference Ridgeway2011).

Barnes and Mertz's contribution for this issue exemplifies this point. The main focus of their article is at the micro level—at the individual players—of US law professors and at their job satisfaction and work experiences. For this purpose, a survey and in-depth interviews are particularly well suited, especially when they are used to supplement each other. Interestingly, the results of their analysis suggest that processes at the macro and meso levels of analysis might also be important in understanding the job satisfaction of law professors. The authors found that the satisfaction of law professors vary by race and gender, and that the institutional culture (macro level) and feelings of lack of voice and respect (meso and macro levels) may explain why female and minority professors are less satisfied than white male professors. Thus, in a follow-up project that focuses on the meso and macro levels, the authors may further investigate the processes that generate gendered and racialized job experiences for law professors. They may, for example, observe work interactions among law professors, analyze written interactions (e-mails) or protocols of meetings, or use experiments in controlled settings to better understand the differences in the experiences of law professors.

Discrimination and inequality have been subjected to intensive study in psychology, sociology, and economics. Because legal scholars often reach these issues with a reform-oriented perspective, they have a special responsibility to combine the various angles of research and to synthesize them into a coherent framework. Empirical legal analysis of employment discrimination offers an example of an area where the various perspectives have been developed.

Thus, studies geared toward analyzing the way cultural beliefs and stereotypes contribute to decision making have brought a psychological lens to bear on concrete employment issues like hiring, salary recommendations, and promotion, showing the importance of both conscious and unconscious processes (Krieger Reference Krieger1995; Krieger and Fiske Reference Krieger and Fiske2006; Fiske, Cuddy, and Glick Reference Fiske, Cuddy and Glick2007). At the same time, other studies have demonstrated that employment discrimination is not only the result of the concrete decisions made directly by employers, but also of the numerous routine everyday workplace interactions, informal networks, and performance evaluations that are all affected by cultural beliefs and stereotypes people have about themselves and others. These sources of inequality in the employment context are compounded by additional systems of inequality in the market and the family and the effects they have on employment. Thus, legal empiricists have utilized sociological insights to analyze both the institutional components of discrimination and their interaction with inequality outside of employment (Galster and Constantine Reference Galster and Constantine1991; England Reference England1992; Ayres and Siegelman Reference Ayres and Siegelman1995; Pager Reference Pager2007; Albiston Reference Albiston2010; Corell and Ridgeway Reference Ridgeway2014). And because macro-level processes like markets and, especially, the employment market are all crucial to the understanding of employment discrimination, both insights and econometric research methods from the field of economics must find their way into legal empirical studies of discrimination (O'Neill and Polachek Reference O'Neill and Polachek1993; Blau and Kahn Reference Blau and Kahn2000; Goldin Reference Goldin2000; Bertrand Reference Bertrand, Card and Ashenfelter2011).

There is an important difference between the analysis offered by psychologists and sociologists and the one advanced by economists. The former have explained that many unequal employment outcome are generated by (often mistaken) cultural beliefs and stereotypes about the competence and performances of social groups that employers take into account in their interactions with those targeted by these beliefs (Ridgeway and Correll Reference Ridgeway and Correll2004; Correll and Benard Reference Correll and Benard2006; Ridgeway Reference Ridgeway2014). Economists, by contrast, have highlighted that when employers take into account statistically grounded beliefs about social groups in their interactions they engage in statistical discrimination (Aigner and Cain Reference Aigner and Cain1977; Pinkston Reference Pinkston2006). This form of discrimination arises when employers consider in their employment-related decisions the average statistical divergences between different groups of people and do not check whether they are applicable in the case of the individual person. With statistical discrimination, every person who has the particular trait in question is assumed to be the average one and is forced to bear the average costs associated with the trait. From the perspective of employers, statistical discrimination is more efficient than evaluating the competence and performances of each individual employee or potential employee.

Insofar as legal empiricists who investigate these issues are interested in the existing law and in possible reforms (as they typically are), they ought to synthesize insights and methods from all three fields. They should therefore take the distinction between mistaken and accurate cultural beliefs that arises from the synthesis seriously when designing the research, since this distinction may have normative implications for legal reformers who would consider questions such as how to allocate the costs of prohibiting discrimination between individual employers and the public in general.

The context of employment discrimination, then, engenders two lessons that point toward the importance of synthesis: First, on the level of pure knowledge, multifaceted practices require multiple lenses to be understood holistically. No single extralegal perspective encompasses all the relevant insights. Second, and relatedly, the recognition that employment affects people's resources, power, career mobility, and general well-being generates special responsibility for people interested in reforming the law. Like most legal scholars, legal empiricists worry about the law and its effects on people's lives and are thus obligated to investigate the possibilities for explanation and reform from the most inclusive perspective they can muster. They do not enjoy the privilege of merely observing it through the lens of one theoretical perspective or while using one research methodology. The closer they approach reform, the more their mode of investigation must approach an “all things considered” attitude, where “all things” implies multiple methodologies.

Concluding Remarks

Full-blown legal realism of the type we have used as our guide in this article is not a dominant presence on the legal theory stage. Truth be told, it may not be that popular at all. But if there is any truth to the cliché that “we are all legal realists now,” it is that most current theories of law insist on perceiving law as a means to an end. This purposive emphasis explains why many of the original realists attempted to engage in empirical research (Schlegel Reference Schlegel1995), and it certainly justifies the renewed resurgence of the empirical focus in recent years.

Understanding the law as a purposive endeavor should not imply, however, that one look straight past the law to consider only the ends it serves. Neither does it entail that the means can be analyzed only through the methods developed in other disciplines. There is undoubtedly much value in the independent perspectives and toolkits that practitioners from empirically oriented disciplines bring to the study of legal phenomena. But the perspective of legal theory, broadly defined, in which law is interrogated as a set of coercive normative institutions, offers a significant added value, and its contribution is also apparent regarding empirical inquiries of law. Although this article cannot offer a complete catalog of features that empirical legal studies should consider, we think making the place of legal theory in empirical investigation explicit is a valuable first step. Every empirical inquiry makes assumptions about how law works or, in other words, how law is law. Bringing those assumptions to the forefront elucidates the choices involved in the selection and integration of empirical data and facilitates their translation into the grammar of law. By making its legal theory assumptions more explicit, empirical inquiry can clarify its stakes and help its readers gauge its implications. We think such grappling with legal theory will aid in the development of a coherent research agenda on law, or on any particular legal field or phenomenon, as well as in the design of many specific empirical projects. In the best case, empirical inquiry will generate new insights that challenge existing legal theory, opening up the best possibilities of legal realism.