Suing the Gun Industry: A Battle at the Crossroads of Gun Control and Mass Torts Return to Book Detail Page
Why Regulating Guns through
Litigation Won’t Work
The assault against the gun industry is a peculiarly American form of guerrilla warfare. The insurgents have ‹elded their commandants and foot soldiers, deploying them along diverse fronts. They1 have eagerly and imaginatively recruited spies and turncoats to ferret out the enemy’s documents and strategies. Both they and their industry adversaries have exploited modern tools of communication and persuasion—lobbying, publicity, expert studies, legislative hearings, propaganda, opinion leader vouching,2 and disinformation—to mobilize support among politicians, the media, the uncommitted, and the uninformed. And both sides have wielded a variety of weapons ranging from the traditional to the most technologically advanced.
Legislative tactics having failed either to bring down the gun industry or to produce the more extensive controls that the reformers seek, they have made litigation their new weapon of choice. Although we are by now thoroughly inured to the spectacle of mass tort actions against companies that market controversial products of one kind or another, litigation against an entire industry remains relatively unusual. The main precursor, of course, is the tobacco litigation, which in its current phase seeks to move the focus from the traditional individual claimants to class actions and other forms of mass claims aggregation.3 In view of the remarkable success of the tobacco plaintiffs (and especially of their handsomely compensated lawyers), it is hardly surprising that the gun control advocates regard the tobacco litigation as a template for their own efforts, while also introducing some new wrinkles of their own devising.
(Stephen Sugarman, in chapter 8 of this volume, analyzes the relevance of the tobacco model.)
Litigation, then, is the most remarkable site of gun control efforts today in the United States.4 As of this writing (November 2004), the plaintiffs’ lawyers in the gun litigation have invested a great deal of time and money on this campaign5 with very little to show for it in terms of either new liability-promoting legal doctrine or jury awards sustained on appeal.6 Indeed, this litigation may be producing the worst possible outcome from the reformers’ perspective. Their lawyers, after all, are drilling dry holes at great cost, creating adverse legal precedents, and further energizing already militant pro-gun groups. State legislatures are adopting statutes that bar or limit such litigation, and Congress is actively considering legislation that would grant the industry nationwide immunity from liability. Such controls are a direct, categorical response to the plaintiffs’ litigation campaign.7 From a purely consequentialist point of view, this campaign must be considered an almost total failure.
I say almost because the gun litigation has produced certain outcomes cheered by the plaintiffs. It has forced at least one small gun manufacturer (Davis Industries) into bankruptcy;8 induced another (Colt) to discontinue production of seven handgun models;9 convinced a third (Smith & Wesson) to agree to alter its design, marketing, research, and distribution practices to reduce gun-related risks;10 and obtained some concessions from some small gun dealers.11 Whether these outcomes will in fact reduce gun-related risks or merely shift the demand to other brands and dealers12 remains to be seen, but gun control advocates regard them as victories. In addition, the litigation has unearthed information from industry ‹les that helps to discredit its public position,13 and Judge Weinstein’s decision in July 2003 dismissing the most prominent private tort case against the industry was written in a way that creates a roadmap for plaintiffs’ lawyers in other cases.14 Finally, as Philip Cook and Jens Ludwig argue, litigation may succeed in altering the political starting points and encouraging compromise solutions that would otherwise be more elusive.15
The analysis that follows constitutes an exercise in comparative institutional analysis. My purpose is to illuminate the relative advantages and disadvantages of undertaking gun control through common law litigation versus other forms of social policy-making and implementation. In doing so, I shall take as a premise—although this premise is in fact highly contestable16—that gun control (in some form) is a desirable public policy and that the relevant question is not whether to undertake such control (in some form) but whether to do so through litigation or instead through some other public law technique that I shall call “risk regulation.”17 When my answer to this question depends either on which particular form of gun control the decision maker chooses or on how its choice would be implemented, I shall note this conditionality. In short, I mean to focus this analysis as much as possible on the institutional differences between judicial and nonjudicial regulation. To this end, I ‹nesse the substantive issues of gun control policy as much as my focus on comparative institutional analysis—which prefers the management of gun risks through legislative or administrative regulation rather than adjudication—will permit. Being only human, however, I cannot resist using the conclusion to state very brie›y my own policy views, which would place a priority on developing improved technologies to detect illegal possessors and to identify illegal users rather than extending the kinds of existing controls that have been demonstrably ineffective.
Needless to say, the gun control movement is reluctant to forswear litigation about gun risks in favor of legislative and administrative regulation. After all, they have already tried and failed in their political pursuit of the latter. Nor is it necessarily true that regulation and litigation are mutually exclusive. Indeed, as Timothy Lytton argues in chapter 10, these two forms of social control may in some circumstances actually be complementary. My analytical project, however, will be that of an Olympian designer of an optimal regime for controlling gun-related risks, one who compares the litigation option as it now exists to the regulatory option that would become viable were the gun control movement to achieve greater political success in the future. Finally, my discussion will refer to legislators and administrative of‹cials collectively as risk regulators unless I mean to distinguish between the two categories for analytical purposes.
The Logic of Institutional Comparison
Institutional comparison is rooted historically in the legal process school of jurisprudence strongly identi‹ed with the work of Henry Hart and Albert Sacks at Harvard Law School in the 1950s.18 The methodological premises of the legal process approach are well established in a rich literature with linkages to work in political science, public administration, and economics. At the most foundational level, this approach is concerned with the structure and behavior of institutions. Political theorists have recently elaborated a deep conceptual understanding of the nature of institutions,19 but such abstraction is unnecessary for my purposes. The institutions of chief interest here are the three standard structures of political governance and policy development—legislatures, courts, and administrative agencies—as well as the less formal institutions that surround them: the plaintiffs’ and defense bars, the media, the gun industry, gun control groups, and the market.
The legal process approach holds that different institutions exhibit distinctive structural elements that shape the behavior of those individuals and groups that constitute them in patterned, recurrent, recognizable ways. Some of these elements—such as how the institution’s leaders are appointed, how they must exercise their authority, and how the institution relates to other institutions—are constitutional in nature. Others are prescribed by external statute or internal regulation. Still others are informal, generated by the goals, incentives, and interactions of the of‹cials and of outsiders with interests in the institution’s decisions. These structural elements, taken together, fashion an institution’s routines and tendencies—and, over time, its traditions, organizational values, self-conception, and sense of mission.
All of this institutional development is designed to legitimate its decisions and authority as it competes against other institutions for resources and in›uence in a political world where they are always being hotly contested. In this competition, each institution exhibits characteristic (because largely structural) strengths, weaknesses, and trade-offs. Simply stated, each institution is good at some things and poor at others—compared to other institutions. Such judgments about comparative institutional performance or competence, of course, are largely in the eye of the beholder; they depend on her particular values and empirical assessments. There simply is no Archimedean point outside the system of evaluation from which an observer can derive and render authoritative, uncontroversial judgments about how institutions perform.
These judgments about comparative institutional competence are inherently dif‹cult. First, the notion of competence, properly understood, extends far beyond mere technical expertise. It also requires an assessment of an institution’s capacities across a number of criteria: normative (its ability to instantiate the distinctive values for which it stands), political (its ability to generate and sustain the necessary public support), and instrumental (its ability to effectively and ef‹ciently implement the decisions it makes). Second, such assessments are necessarily contingent; they vary according to the nature and history of the speci‹c issue being addressed by each institution, the resources available to it, political factors outside its control, and many other variables. Accordingly, an assessment must be quali‹ed and context sensitive, not categorical and timeless. Finally, institutional performance is inevitably suboptimal, which means that an institutional comparison is inevitably a comparison among what Neil Komesar calls “imperfect alternatives” (which is the title of his important analysis of how society should choose among different institutions to achieve its goals).20 Thus, the fact that an institution is less than ideal for a particular policy task is not necessarily a reason to reject it and choose another; for all its limitations, it may still be better than the alternatives.
Although the particular assessments required by the institutional comparison approach are likely for these reasons to be controversial, the approach itself might seem self-evidently correct. After all, who can object to judging and comparing institutions according to their capacities to handle the kinds of questions that come before them? The answer, however, is somewhat more complicated—at least where important legal rights are concerned. Some commentators indeed object to basing the assignment of responsibility for determining and protecting legal rights on an institutional assessment that takes this rationalistic, instrumentalist, cost-bene‹t balancing form. In particular, those legal scholars and others who tend to prefer judicial fora to legislative and administrative ones resist this approach, celebrating the courts’ traditional role in recognizing and enforcing individual rights, whether those rights be constitutional, statutory, or judicial in origin. They are often beguiled by the common law process, which purports to elaborate law from previously adopted principles rather than making a synoptic judgment at the threshold of decision about which of the various institutional possibilities for lawmaking is best suited to the task. Courts engaged in common law adjudication seldom ask institutional competence questions unless, as in many gun liability cases, the novelty of the plaintiffs’ claims at common law is clear and the claims’ broad policy implications make such questions impossible to ‹nesse.21 And even when the courts do ask institutional competence questions in gun cases, they seldom analyze them in a systematic or thorough fashion. In some cases, this juridico-centrism (as I have called it)22 may be the outcome of a self-conscious institutional comparison by the court, but more often it represents either a kind of re›exive or ideological preference for judicial resolution or a prudential judgment that the particular right in question will fare better (or worse, as the case may be) in adjudication than in a political or bureaucratic forum.
The institutional competence literature is now large. Much of it has been a response to earlier waves of class action litigation designed to reform large public bureaucracies like school, prison, police, and mental health systems where the legal claims and especially the legal remedies being urged upon courts strain, and perhaps exceed, their institutional capacities.23 Less of this literature has been concerned with the challenge of designing a process to manage social problem solving more generally.24 The analysis that follows is written in this broader spirit.
The Criteria for Institutional Comparison
In order to conduct an institutional comparison, of course, one ‹rst needs to specify the criteria according to which the competing institutions are to be appraised. Although the relative weights assigned to each criterion in the overall judgment are bound to be controversial, the criteria themselves should elicit general agreement. I propose six of these criteria; obviously I could identify many more. Each of these six criteria pertains to the institutions’ propensity to develop and implement sound gun control policies. Again, this is true almost without regard to the substance of those particular policies. In order for an institution to make and implement policy effectively, it must (1) generate the technocratic information needed for gun-related policy-making; (2) generate the political information needed to frame an acceptable policy; (3) mobilize the array of different policy instruments necessary to establish and implement the policy; (4) promote social learning (short feedback loops) and ›exible adaptation to new conditions; (5) generate predictable rules; and (6) secure and sustain the policy’s legitimacy.25
In the discussion that follows, I shall often speak of regulation in general terms as a kind of useful shorthand. Even at this abstract level, however, it is important to emphasize that the properties and propensities of various regulatory systems may differ signi‹cantly from one another. We might broadly distinguish between two types of such systems. One is a regime in which an agency supervises and speci‹es a multitude of design, production, transportation, and other decisions by the industry in minute, technical detail and on a continual, dynamic basis. Air pollution control is a familiar example. A quite different regime is one in which the regulator imposes relatively few limitations on what the industry may or may not do. These standards tend to be fewer and more static, and their enforcement does not require the elaborate regulatory machinery of an Environmental Protection Agency but relies on more conventional agencies such as police departments and the Bureau of Alcohol, Tobacco, Firearms and Explosives (BATFE). I imagine that the regulation of gun-related risks will likely take this simpler form. It already does so with regard to registration requirements, background checks, bans on the sale and transport of speci‹c weapons, and so forth, and may do so in the future if regulation requires, say, ballistic markings or certain new safety features. I do not mean to prejudge the speci‹c forms that gun regulation might take. My point, rather, is to re‹ne the discussion of gun regulation by contrasting it, even at this abstract level, to the far more complex regimes found in other areas of risk regulation policy.
Information is the lifeblood of intelligent lawmaking. A legal system’s ability to mobilize high-quality policy-relevant facts for the lawmakers at a relatively low cost is perhaps the most important precondition for the effectiveness of its policies—and ultimately for their legitimacy, as I discuss later. The term facts here encompasses empirical evidence concerning, among other things, the nature and magnitude of gun-related risks; how well existing policies work; which public and private resources can be brought to bear on which aspects of the problem; how alternative policies might affect the world and through which causal pathways and with what substitution effects;26 the magnitude of the costs and bene‹ts created by these alternatives and the intergroup distribution of these costs and bene‹ts; and how markets will respond to particular policy changes. I call the information about such facts “technocratic,” distinguishing it from the “political” information that is the subject of the next section.
For any complex policy problem, of course, the available information, technocratic or otherwise, is always incomplete, imperfect, subject to competing interpretations, and costly to obtain or improve. This is simply a fact of life for policymakers. The relevant question is this: Which institution or combination of institutions27 is best equipped to minimize this informational de‹cit with respect to this particular policy problem?
We are in a better position to answer this question today after experiencing more than three decades of legislation, administrative regulation, and mass tort litigation directed at major public health risks—especially asbestos, tobacco, and gun-related violence.28 During the 1960s and 1970s, federal and state legislation established regulatory agencies designed to mobilize technical expertise, to identify signi‹cant public health risks, and then to act to reduce them. These regulatory schemes, it was thought, could render private tort lawyers ancillary, if not super›uous, to the regulators engaged in risk management.
Alas, as has frequently been true in the history of the administrative state, this promise proved to be false, or at least vastly exaggerated. Environmental, occupational, and product risks are actually identi‹ed through complex processes in which many actors—endowed with different resources, responding to different incentives, and employing different methodologies—play a role. Often, these risks are ‹rst revealed by research scientists, public health and environmental organizations, specialized science and health journals, and other private groups. These risk monitors then communicate their research ‹ndings to those who are in a position to act upon this information—mass media, industry-speci‹c newsletters, labor unions, policymakers, and services that publish and distribute this information to lawyers.
Sometimes, plaintiffs’ lawyers are alerted to litigation-worthy health risks by an accumulation of workers’ compensation awards, consumer complaints, public health agency reports, or media coverage. This information ›ow triggers what I have called a “lawyerization” of risk, in which plaintiffs’ lawyers drive the policy-through-litigation process forward. Only seldom does formal regulatory action spawn mass tort litigation; in cases as diverse as asbestos, Dalkon Shield, DES, Agent Orange, auto safety, and tobacco, the regulatory agency took action, if at all, only years after the risk had been at least partially lawyerized.
Gun litigation is quite different. Whether and how to control gun violence have been prominent public policy issues for a very long time—since 1927 at the federal level and at various times at the state and local levels (1911 in New York)29—yet the tort lawyers did not get into the act in a serious way until the late 1990s.30 They cannot claim responsibility for developing the policy-relevant information about the nature or magnitude of gun-related risks, which have long been well known.
It is true that the litigation process tends to produce a good deal of information from and for the parties. After all, discovery can be notoriously aggressive and intrusive, and the plaintiffs’ lawyers have every incentive to demand as much existing information from the defendants as the discovery rules, the judge or discovery master, and their photocopying budgets will allow. But this access to existing ‹les hardly translates into providing the decision maker, be she a judge or juror, with the kind of information that she would want if she were a rational risk regulator, and, in any event, few litigants possess the resources needed to adduce it. The adage “there is many a slip between cup and lip” has its tort litigation counterpart in the many screens that lawyers impose between discovery documents and presentation of evidence.
It is also true that the plaintiffs’ lawyers have discovered documents bearing on the industry’s allegedly negligent distribution and marketing practices.31 Although these documents have not yet succeeded in establishing tort liability, they are surely useful to risk regulators seeking to identify available pressure points for reducing gun-related harms. Nevertheless, it is hard to believe that risk regulators engaged in this task would not have developed this kind of information through the conventional information-gathering techniques available to them. The publicity surrounding the litigation certainly raised the issue of gun-related harm to greater public prominence, but this is not the same as producing policy-relevant information that regulators would not otherwise have adduced. Then again, my belief on this point may only re›ect the advantages of hindsight.
Indeed, it is quite plausible to suppose that litigation, or the prospect of it, may actually discourage the gun industry from undertaking new risk-related research. Manufacturers are the only group with a possible economic incentive to invest in such research, but at least two obstacles exist. The new risk information (unless protected as intellectual property) would be a public good, so no individual manufacturer would be able to recover the costs of producing it, while any research cost-sharing agreement among industry competitors would court potential antitrust liability. More to the present point, any new information that the research might yield—for example, ways to manufacture or market in risk-reducing ways—would likely be used by gun control groups to press for expanded civil liability or government regulation, at least insofar as the rules of evidence permit them to use this new information against the manufacturers.32
In contrast, risk regulators can encourage or even subsidize research on gun-related risks, as they have for many other public health problems like lead paint poisoning, pollution, teenage pregnancy, and a host of diseases. They can also direct which particular research issues should be investigated rather than rely on the probably vain hope that the tort system’s liability signals will indirectly induce manufacturers to research the speci‹c issues that, from regulators’ more disinterested point of view, are likely to have the greatest public health payoff. In particular contexts, of course, risk regulators may fail to undertake the needed research for political, technical, or other reasons—and some would say that this has been the case with respect to the gun industry’s marketing and distribution practices. To the extent that this is so, litigation-induced research may be all the more valuable—although the fact that it is driven by litigation or funded by the parties may adversely affect its quality and reliability as an instrument of policy.
Judges as policymakers are technocratically disadvantaged in several other ways relative to their risk regulator counterparts. They are trained as generalist lawyers (usually as litigators), not as policy specialists, and thus are unlikely to acquire, or to know how to exploit, the kinds of information that a competent policy analysis requires. As already noted, judges only receive the information that the litigants choose, for their own self-interested reasons, to provide. Moreover, these litigants represent only a fraction of the many social interests probably affected by the legal rule, and the selection biases that drive the decision to litigate mean that the relatively few tort cases that reach trial are likely to be unrepresentative of the set of all tort disputes, much less the social reality that a policy must address.33 The tort case, then, is a most unpromising vehicle for generating public policy that will govern a relatively large number of often disparate situations.
But the system contrives to keep judges even more ignorant than this. The rules of evidence further limit—and, in the name of protecting other legal values, distort—the kinds of information that judges can receive or that they may take into account if proffered. Other legal rules render certain factual issues legally irrelevant to a tort claim and thus unlikely to be developed by the parties and introduced into evidence. A rational regulator of gun risks, for example, might want to know about the industry’s revenues, costs, and pro‹ts in order to assess the feasibility of alternative policy interventions, yet this kind of economic information might be inadmissible in a tort case where the issue is whether and how the defendant harmed the plaintiff.34 Similarly, the same risk regulator would regard information about liability insurance practices in the industry as highly relevant. Yet this information, which judges might or might not take into account in deciding whether a manufacturer owes a legal duty to a remote victim, would be inadmissible before a jury and, if hinted at by a lawyer, might even prompt a mistrial.
More generally, risk regulators are less interested in the adjudicative facts concerning a particular gun manufacturer and the people its products injured than about the legislative facts bearing on the much larger risk environment. Indeed, unlike a judge who is con‹ned to assessing the risk presented by a particular case that happens to come before her and cannot easily choose not to decide, a rational risk regulator considers not only gun risks but all other risks that compete with it for regulatory resources. After reviewing the larger risk environment, the regulator may decide that it could better use those resources to regulate, say, smoking rather than guns or to regulate guns of one type rather than the type used in the case before the court. Although in reality regulators seldom use risk information as synoptically as this, in principle they should do so and in practice they sometimes do. Common law judges almost never do, and in principle they should not.
Again, the distinction between the production of adjudicative and legislative facts in different fora should not be overdrawn. The gun litigation, for example, has generated useful legislative-type facts about the feasibility of alternative designs and marketing practices, while risk regulators are often in›uenced by individual narratives that they believe, rightly or wrongly, exemplify more general problems. The important point here is about the general tendencies and distinct properties of different institutions.
Policymakers need good political information at least as much as the technocratic information. This category encompasses information about, among other things, the various outcomes different groups of voters want; the intensity of those preferences within those groups; the stakes that different policy proposals would create for different groups and the political resources that they will invest in the policy process; the levels and modes of their participation; the support enjoyed by the groups’ leaders; the media’s likely take on the issue; the legislative, bureaucratic, and organizational alliances that policymakers can form or mobilize; the likely sources of opposition; the carrots and sticks they can deploy to elicit the necessary support; and a variety of local, often arcane knowledge about what motivates the key political actors, which concessions they would accept and on what terms, and how they will behave in particular situations.35
Merely to recite this (incomplete) list of political information, of course, is to suggest how elusive and intricate it will be for any policy proposal, and all the more so as its importance and complexity increase. Consider just one of these factors—the intensity of voters’ preferences.36 Policymakers need to know not what voters want in the abstract but how strongly they want it and what they are willing to sacri‹ce in order to obtain it. This vital political intelligence is hard to obtain. While a competitive marketplace can readily register the intensity of consumers’ preferences, only a political bargaining process in which alternative outcomes are “priced” through exchanges among the participants can measure how intensely voters feel about policy outcomes. Interest groups provide an important index of voters’ willingness to “pay” for their policy goals. Being a member of a group usually entails some cost; members’ actions on behalf of the group such as letter writing, grass roots work, reading-group materials, and lobbying are even more costly to them. For this reason, voters’ group memberships are a crude proxy for the intensity of their policy preferences, one whose accuracy will depend on the cost of membership, the ease of exit, and so forth. Skillful policy-makers learn to gauge these factors as they gain knowledge about the various groups that are actively concerned with particular policy issues, but it is harder for them to do so with respect to preferences around which groups cannot easily organize, either because the preferences are diffusely or weakly held or because the individuals who share them lack political resources.
Common law courts are walled off from this vital political information by institutional design, lack any expertise in interpreting it, and are precluded by both tradition and doctrine from integrating it into their decisions. To observe this is emphatically not to say that judges are political innocents. In fact, they often have political experience and by profession and inclination are interested in and aware of political developments. Nor is it to suggest that judges do not advert to what they imagine are the political consequences of their decisions or take those consequences into some account. They do both. The important point for my purposes is that they are singularly disabled from gathering and using this kind of information systematically or effectively—and anything less than this information is likely to lead them astray, from a political point of view.
The gun litigation may be an example of this political ineptitude, although for reasons about to be explained one cannot be certain. The ‹rst decision to hold a gun manufacturer strictly liable in tort for injuries in›icted by a third party, Kelley v. R.G. Industries, Inc.,37 was quickly rejected by the Maryland legislature, which overturned the decision and adopted a statutory nonstrict liability alternative. Subsequent gun litigation has generated a political backlash against these claims in many other states and, most notably, in Congress. These political actions do not necessarily mean that courts upholding such claims were wrong as a matter of tort law. After all, courts are free to interpret tort law as they see ‹t and to let the political chips fall where they may, while legislatures are entitled to revise the courts’ understandings of tort law through statute. This kind of pas de deux is by no means uncommon or necessarily unwelcome; the California legislature, for example, has rejected its courts’ imposition of social host liability to victims of alcohol-related accidents.38
Nor do such political responses preclude the possibility that the judicial decision helps to produce some important policy changes. Indeed, Timothy Lytton demonstrates in chapters 6 and 10 that this has occurred in the wake of some unsuccessful gun litigation. One suspects, however, that some courts have not fully grasped the extent of political opposition to the new doctrine. Institutionally speaking, political intelligence gathering is not part of their job. This limitation, far from being a cause for regret, is generally salutary for the legal system.
Effective policymaking is impossible unless the implementing of‹cials can deploy the variety of policy instruments to create and shape the incentives necessary to secure compliance by the members of the public and other of‹cials who are the policy’s targets. A risk regulator’s kit bag contains a large number of these instruments, enabling her to impose ‹nes or taxes, subsidize, educate, reorganize, inform, hire, ‹re, insure, establish bureaucracies, build political coalitions, or coerce third parties—to name just a few of the tools available to her. In contrast, judges in tort cases possess few instruments for securing compliance, and they tend to be weak, in›exible, or both. With the exception of nuisance cases, where injunctive relief may be possible, all that judges can do is order A to pay B a prescribed sum of money for breaking a rule. The damage remedy that they can deploy is undeniably important in shaping some kinds of behavior—including the pricing decisions of insurers, as Tom Baker and Thomas Farrish explain in chapter 12—but it affects quite a narrow band on the broad spectrum of human motivation.
Another instrumental factor that severely limits courts’ ability to implement their policy views is the institutional imperative to decide cases on ostensibly principled grounds rather than on the basis of legislative-type choices. For example, modern courts that decide to abandon the traditional rule of contributory negligence nearly always feel constrained to adopt the alternative approach of “pure” comparative fault. One reason may be that courts think they can derive a “pure” regime from general tort principles, whereas any of the “modi‹ed” approaches—which might make better policy and which legislatures usually choose—would require judges to select arbitrary percentage cutoffs as legislators routinely do but traditional, legitimacy-conscious judges shrink from doing. This judicial bias in favor of principled reasoning, of course, does not prevent courts from importing policy considerations into their decisions, as a vast number of contemporary tort law attests, and some of them produce rules that are necessarily arbitrary even when policy justi‹ed—for example, the multipart tests for bystander and economic loss recoveries. My point is simply that this concern to decide according to general principles of justice constitutes an important institutional constraint, largely actuated by legitimacy concerns, on their practical ability to engage in policy-making in an open, unapologetic, and systematic or thoroughgoing fashion.
Nor are general tort standards likely to be as effective as speci‹c regulations in targeting the particular behaviors that are most pertinent to sound gun control policy. In part, the relative ineffectiveness of these standards is because it is juries who apply them. Of course, juries apply general standards such as reasonableness, proximate cause, contributory fault, and knowledge of risk to conduct in the normal course, but as an institutional matter they do so in an opaque and conclusory fashion— through general verdicts and without explanation. Especially when applied in this delphic fashion, such standards lack the capacity of highly speci‹c regulations to focus the attention of ‹rms and of‹cials on the behavioral variables that actually shape gun-related risks—variables such as record keeping, advertising, selection and monitoring of dealers, ‹rearms education, weapons storage, threatened criminal violence, compliance with licensing, and technological innovation. I do not mean either to deny the superiority of standards in many contexts or to suggest that rules do not also entail endemic disadvantages. My point, rather, is that courts do not have the same range of choice between rules and standards that other risk regulators possess.
The common law damage remedy, of course, is often perfectly adequate for the purpose of inducing defendants’ straightforward compliance. Where gun litigation is concerned, however, recalcitrance is more predictable. First, many gun manufacturers—in sharp contrast to the cigarette producers—are relatively small, thinly capitalized businesses that may be rendered insolvent by a large tort judgment.39 Second, the manufacturers suspect that the real motivation behind the gun litigation, as it was in the cigarette litigation, is less to reform their production or marketing practices than to end their nongovernmental business.40 The plaintiffs’ assumption, on this account, is that there is no such thing as a safe gun in a world where guns are routinely stolen and ‹nd their way, despite the risk of tort liability and criminal law enforcement, into the hands of violent criminals, vulnerable children, sociopaths, and civilians who misuse them. In such an environment, the gun industry concludes, it is literally ‹ghting for its economic life and must resist compliance as strenuously and resourcefully as it can. This may be one reason why Smith & Wesson, after agreeing to a widely publicized settlement praised by the president of the United States, backpedaled and ended up altering its marketing practices only marginally.41
Third, the behavioral response of a gun defendant, even one that pays the tort judgment or alters its practices in order to minimize its liability, can take a number of different forms. For my purposes, the crucial point is that a court cannot control which of these choices the manufacturer makes, and some of its choices may simply shift the risk of gun-related harm around or even increase it.42 Gun control scholar James Jacobs speculates, for example, that unscrupulous manufacturers might respond by selling only to wholesalers or only to wholesalers who promise to distribute guns responsibly, which might encourage shady dealers to cash in by becoming wholesalers for these manufacturers. Alternatively, they could divert their guns into the secondary market or simply go out of business, thereby leaving the manufacture of guns to the black market, as has occurred with certain illicit drugs.43
In contrast, a risk regulator can decide to regulate particular behaviors of gun ‹rms with some con‹dence that the ‹rms will either comply with the regulations or violate them in ways that the regulator can then identify, address, and perhaps punish. The regulator can encourage or mandate new behaviors that it thinks will best address the gun-related risk it wants to alter, or it can discourage or prohibit existing ones. Again, I am not suggesting that regulators always secure compliance while judges cannot—far from it—but rather that the former are institutionally more capable of getting their way. The list of regulatory failures is long indeed—and, as Timothy Lytton argues, BATFE’s failures may be among them—but those failures are more often instances of regulators choosing the wrong policies in the ‹rst place than they are instances of choosing the right ones but being unable or unwilling (given their resources) to implement them.44
Finally, any institutional comparison of policy instruments must consider the crucial role of markets. The effectiveness of every public policy is signi‹cantly affected, if not determined, by how that policy interacts with the market forces that in›uence those on whose behavior the policy’s success depends. Strong market forces, particularly when reinforced by powerful social norms, may simply overwhelm law, distorting or disabling it. If those forces are strong enough, they will spawn corruption, smuggling, or other illegal markets. On the other hand, well-designed policies can harness market dynamics to the law, thereby strongly advancing the law’s goals.
These considerations are particularly relevant to gun control policy. There are now about 275 million ‹rearms in the United States, distributed among 35–45 percent of all American households; those households possess between 2.5 and 4.5 weapons on average.45 This bespeaks immense and rapidly growing ‹rearms markets, legal and illegal—markets that are notoriously dif‹cult for the government to limit. Indeed, Jacobs shows that regulatory control efforts have largely served to expand gun markets by driving them underground, where they defy even conventional law enforcement. My point here, then, is not that risk regulators possess policy instruments for effectively controlling gun markets; rather, it is that courts do not. As Jacobs’s analysis suggests, their only tool, the damage remedy, could actually strengthen illicit markets.
Perhaps no resource is more essential to a society’s policy wisdom than its capacity to learn and to adapt swiftly and creatively to changing conditions. This learning capacity in turn depends on the society’s ability to generate, aggregate, process, disseminate, deploy, and (as necessary) correct the information it needs in order to discover what its collective purposes are and might be and then to pursue them effectively.
Courts lack any reliable way to obtain feedback on their policies’ real-world effects, and, as noted earlier, they are poorly equipped institutionally to assess the policy implications of the new evidence. Judges depend for their information largely on the particular litigants and disputes that happen to come to their courtroom, and the later case is an appropriate vehicle for policy change only if it raises the same policy issues that the earlier case raised but manages to adduce better information for decision. Even if the precedent can simply be distinguished, disregarded, or overruled, a court that realizes belatedly that it erred earlier must ordinarily wait for a properly framed new case to come along that provides it with a vehicle for changing doctrinal course. In its recent Lawrence v. Texas decision,46 for example, the U.S. Supreme Court announced that the rule it had adopted in Bowers v. Hardwick was wrong not only in 2003 but from Bowers’s inception. Nevertheless, it took the Court twenty years to ‹nd a case (and the necessary votes) that enabled it to correct its error. This situation, moreover, is not at all typical; as the Court emphasized in overruling Bowers, that decision had been relentlessly and severely criticized from its ‹rst day. In contrast, the court in the usual tort case is far less likely to learn about the nature and extent of its error.
A risk regulator, in contrast, has numerous opportunities to revisit its earlier decisions and revise them whenever it seems appropriate. Indeed, regulators have institutionalized formal techniques for doing so—oversight hearings, notice-and-comment rule making, reauthorizations, and the like—and access to the information and expertise needed to assess the case for change and the various reform options that are available.47 And interest groups such as the National Ri›e Association (NRA) and gun control advocates engender rapid social learning by giving regulators, the media, and the general public almost instant feedback about the costs of public policies to the groups that claim to bear them. Whether one regards their efforts as examples of valuable public education or of irresponsible distortion—the gun control debate has many examples of both—may depend largely on one’s views about the merits of these controversies. It is also a question in any particular situation of whether the risk regulators were actually informed by this feedback, ignored it, or merely used it as a pretext for decisions reached on other grounds. What is certain is that such regulators are more likely than courts to receive such feedback from a wide variety of sources and in a forceful and timely fashion.
An effective public policy must be predictable. That is, those who will interpret, enforce, or comply with the policy must be able to determine—in advance of its application, with a high degree of accuracy, and at low cost—what it means and (which is much the same thing) how it applies in a variety of factual circumstances. This in turn may require that its purposes be fairly transparent. Predictability is vital in any legal system, but particularly in a liberal one where citizens are expected to govern themselves with maximum freedom and a minimum of of‹cial intrusion. Predictability also advances certain other values such as the rule of law (by facilitating the treatment of like cases alike) and ef‹ciency (by minimizing transaction and enforcement costs). The need for predictable rules on guns is particularly great because of their ubiquity in the United States. These rules, then, must be determinate and transparent enough to enable a vast number of ordinary citizens to know their gun-related legal rights and obligations.
Tort rules are much less determinate and transparent than regulations, other things being equal. In part, this re›ects tort law characteristics that have already been discussed: the generality of its rules; their “black box” application by juries; the state-by-state variations; and the glacial and somewhat opaque process by which tort decisions accumulate, congeal into “rules,” and become recognized as authoritative precedents. In addition, appellate decisions are contingent on the speci‹c facts of those cases—or so the courts, eager to maintain their decisional leeway for the future, insist—yet common law legal reasoning means that we can never know which of the many facts were pivotal unless and until some court writes an opinion in some future case that tells us. A related feature of tort law, lack of codi‹cation, makes it harder for even lawyers to access and interpret its doctrines.
In short, a manufacturer, dealer, user, or victim of guns would ‹nd it much easier to ‹nd and comprehend what the regulatory law of guns prescribes—with regard, say, to licensing or background checks or concealment rules—than what tort law rules prescribe with respect, say, to distribution patterns, intervening criminal use, or gun technology. In such complex contexts, the standard of reasonableness is not a very useful guide to manufacturers that must predict what the courts (or, more precisely, juries) will require and make their long-term investments in technology and distribution networks accordingly. Tort law’s uncertainties create costs for the industry and legitimate users—and, as Jacobs points out, opportunities for illicit gun market participants. To be sure, regulations may immunize practices that comply with the letter of the law but perhaps not with its spirit, while the indeterminacy of tort standards may sometimes induce compliance beyond what the law would have required. These problems, however, can be solved better through regulatory drafting that is more precisely tailored to the desired outcomes than through the uncertain compliance (too much or too little) that vague tort standards may induce.
Legitimacy is a much-discussed but ill-de‹ned criterion. It is easier to describe why it is necessary in a democratic polity than to explain where it comes from, how it is sustained or lost, and why some institutions enjoy more of it than others do in some situations while the reverse is true in other situations. What seems clear, however, is that legitimacy is not simply the sum of the other criteria of competence that I have discussed here. Criteria-based assessments certainly affect an institution’s legitimacy, but it does not exhaust its meaning.
The public perception that law is legitimate constitutes a social resource of incalculable value. It nourishes respect for both polity and principle and hence for the claims of one’s fellow citizens. It induces people to comply with law even when, as often occurs, narrow self-interest counsels disobedience. Indeed, it invites people to broaden their conception of what their interests are and perhaps to identify them with the interests of others. It re›ects an often compelling vision of social justice. Law’s legitimacy, however, is always fragile in a vibrant, liberal society. Political and economic actors attack laws that constrain them while supporting those that constrain their competitors. Law’s moral claim to obedience rests ultimately on the public’s perceptions of its procedural fairness, its substantive justice, and its practical ef‹cacy. Because the lawmaking process invariably compromises these values, its legitimacy is always contestable. Law’s legitimacy is most vulnerable when it seeks to transform a deeply embedded social practice like gun ownership that enjoys its own public legitimacy, for here the gap between the law’s promise and its performance is likely to be widest. Indeed, this gap can become a chasm whose breadth contributes to the decline in public respect for law and government.
These general re›ections on legitimacy, and particularly on the relation between the law’s practical ef‹cacy and its legitimacy, are highly pertinent to an institutional comparison of risk regulators and courts in addressing the problem of gun-related violence. There is something of a paradox here. In general, judges are more highly esteemed in the public mind than are politicians and bureaucrats, yet when it comes to solving this particular problem through tort law, judges are (so I argue) comparatively less competent and less legitimate as an institutional matter.
The paradox is resolved, I think, by the special policy-making demands that contemporary gun litigation places on the courts, demands that I have analyzed in this section.48 The relatively high repute of courts is probably more a vestige of what courts traditionally did than a tribute to what some courts are now attempting to do. Even today, the usual task of a court (and jury) in a tort case is the relatively straightforward one of applying given standards, vague though they may be, to a given set of facts. In contrast, the tasks of risk regulators tend to be more intellectually challenging, involve more complex and ›uid problems, demand more information, arouse more political controversy, require more skill at persuading and mobilizing constituencies, and are harder to implement. Unlike federal and many state judges, legislators must also face at the polls those who bear the costs of their policies as well as those who bene‹t from them. It should come as no surprise, then, that risk regulators have seemed to fail while courts remain popular and successful, for they are playing very different games by altogether different rules. When courts take on the kinds of comprehensive problem-solving, policy-making functions that risk regulators routinely exercise, their failure rate—and the level of public disappointment in them—is likely to rise as well. Courts may believe that inaction by the risk regulators has forced their hand, but this belief begs the essential question as to the reasons and justi‹cations for inaction in the face of what are often daunting policy dilemmas, and it elides the question of the institutional role.
The problem of gun-related violence implicates both of these questions. The risk regulators’ failure to solve the problem during the last four decades speaks volumes about its political and practical intractability. But if this failure speaks, what does it say? Gun litigation advocates offer some answers. This failure, they argue, says that risk regulators, particularly legislators, lack the imagination and, more important, the political will necessary to solve it. Exhibit A in their argument is the fact that the NRA is by many accounts the single most powerful lobbying group in the nation.49 This argument, however, simply pushes the question back a step: why is the NRA so politically powerful? The most straightforward answer is that the NRA is powerful because its more than three million members and other gun owners constitute a bloc, according to polling data, of from twenty-three to twenty-seven million voters in all parts of the country, and these voters intensely oppose most forms of gun control.50 In a democracy, politicians are supposed to respect these voters’ preferences on gun issues and the legislative procedures that often have the effect, if not the purpose, of protecting the status quo from impulsive reforms—and they do. Needless to say, this does not mean that the NRA’s positions are necessarily correct or that politicians who accept those positions do so for public-spirited reasons. It means, rather, that one cannot infer a failure of democratic process from the NRA’s frequent political ef‹cacy.
A second inference that litigation proponents draw from the politicians’ failure to enact more stringent gun control is this: even if the politicians’ opposition to gun control is legitimate, the courts are legitimate legal institutions too, and their institutional duty is to adjudicate the cases that come before them on the basis of settled legal principles (here, tort law) as applied to new, or at least newly compelling, social problems (here, gun-related violence). But this position also begs the more fundamental questions of whether it would be just for courts to stretch those principles to impose liability in these cases and whether doing so would squander the courts’ legitimacy.
On the question of justice, almost every appellate court that has addressed the issue has concluded that existing legal principles cannot properly be stretched so far, and I am inclined to agree with them—even assuming that manufacturers have not controlled their distribution networks as well as they might. I would expect that the legislative reversals of such a stretch would be swift and sure,51 and the courts’ public reputation would suffer accordingly—and, in terms of practical results, for naught. Indeed, by circumventing the essential process of cultural accommodation, such judicial decisions could exacerbate social con›icts over guns rather than ease them.52 Having said this, I hasten to add a caution that I offered some years ago about how the Supreme Court’s legitimacy has fared when it entered political thickets.
In a long line of cases, the Court has turned aside warnings of just this kind, and has instead proceeded to take on project after daunting project of broad political and social reform. At the time, even sympathetic observers often believed that these projects [long list omitted] would strain judicial capacities to the breaking point. Yet more often than not, the Court seemed to sustain its projects without losing either perceived legitimacy or prestige. Indeed, the Court has returned from these engagements bearing the battle ribbons and the public esteem of a military hero. Its struggles with the politicians and the bureaucrats have only served to enlarge its reputation for moral purity, constitutional wisdom, and real-world effectiveness.53
It is hard to say how applicable this cautionary tale is to the courts in gun litigation. State courts do not enjoy the same iconic reverence as the
U.S. Supreme Court, and they presumably have less accumulated prestige to draw down. Because most gun cases involve only common law principles, not constitutional ones, the courts’ pronouncements lack the oracular, almost hieratic quality of constitutional rulings. Because their rulings in gun cases can be, and often are, overridden by the legislature, they may seem less authoritative and worthy of respect. My sense—and it is no more than that—is that members of the public who favor gun control will praise the courts for exhibiting courage and independence in contrast to what they see as the politicians’ abject obeisance to the NRA, while pro-gun forces will denounce the courts for eroding what they view as a precious constitutional right. Beside professors and other (high-minded) pundits, few will devote much attention to issues of institutional legitimacy, competence, or separation of powers, issues that will strike them as academic.
Since my comparative institutional analysis leads to the conclusion that risk regulators are far better equipped than judges to address and manage the problem of gun-related violence, the obvious question remains: what should the regulators do about the problem? Although my writ in this chapter does not run to that question, I do have some views on it, with which I shall conclude.
First, I am convinced by Jacobs’s recent and comprehensive review of existing and proposed laws to keep guns out of the hands of those who should not have them that such laws—call it a “Brady-plus” approach—are unlikely to succeed. There are many reasons for this pessimism: the existence of a vast secondary market that is probably impossible to regulate as a practical matter; a large and elusive black market; the powerful incentives and ample opportunities for criminals to obtain guns in the secondary and black markets; the approximately 275 million guns that are already in people’s hands; the spur to new gun purchases that any reform proposal would create during the periods before enactment and before it becomes effective; the marginal (at best) effectiveness of the Brady Act itself and of other laws along the same lines; the massive noncompliance with existing controls on felons’ access to guns; the large-scale exemptions that any control regime must contain; and so on.54
This pessimism does not mean throwing up one’s hands in futility, but it does suggest a need for more modest expectations and greater realism about Brady-plus reforms. A more promising approach is technological in nature—making guns safer (although relatively few gun-related injuries are in›icted on or by careless, innocent users)55 and more easily traceable to those who ‹re them. Further research on “smart” guns, ballistic ‹ngerprinting, trigger locks, and other such ‹xes is clearly warranted. Whether and to what extent this research should be publicly subsidized is a separate question, but the possibility that the large and growing costs of gun litigation may on balance be diverting manufacturers’ resources away from such research is a disturbing one.
An even more promising approach is the use of surveillance technology to enable the police to detect guns in public spaces and at safe distances while obviating the need for making dangerous, discriminatory, and intrusive “stop and frisks” of individuals without probable cause for suspicion. This would vastly increase the ability of police of‹cers safely and effectively to disarm people who have no legal right to carry weapons. It is only a matter of time before such technology is available. If it is to be deployed, however, it must safeguard individual privacy rights under the Fourth Amendment. Whether and how this can be done is already the subject of some scholarly debate.56 My guess (and it is only a guess) is that the vast majority of Americans will view the use of a reliable surveillance technology limited to concealed guns in public spaces and subject to strict legal controls against unwarranted pro‹ling and intrusion as constituting a reasonable search, one whose crime-and violence-reduction bene‹ts would more than justify the small invasion of privacy entailed. If so, and if the politicians adopt such a measure, the courts will probably uphold it. As an institutional matter, however, the courts are ill-equipped to promote or appraise such technological developments.
Compared with these kinds of direct, narrowly targeted, legally manageable, technology-based forms of risk regulation, tort litigation is a remarkably indirect, indiscriminate, crude, and unpromising remedy for the plague of gun-related violence. It has advanced valuable social ends by generating new policy-relevant information about gun-related violence and by capturing the industry’s undivided attention. These are signi‹cant achievements, and perhaps—the question is an important one—we would not have gained them had the litigation not been prosecuted. At this point, however, it amounts to a costly and institutionally inappropriate distraction from the real political and policy tasks before us.