This file is also available in Adobe Acrobat PDF format
THINKING ABOUT JUDICIAL BEHAVIOR
In 1989, Cincinnati Reds manager Pete Rose faced an investigation of his alleged gambling activities by major league baseball. Rose’s attorneys filed suit to block the investigation, and they steered the case to a Cincinnati judge who faced re-election in 1990. That judge, Norbert Nadel, allowed his announcement of a decision to be televised. When he “started the hearing with a microphone check,” according to one writer, “you knew Pete Rose had the home-court advantage” (Cleveland Plain Dealer 1989). Indeed, the ruling gave Rose what he wanted. (Cincinnati Enquirer 1989)
As George W. Bush ran for president in 2000, commentators speculated about possible candidates for Bush appointments to the Supreme Court.
J. Harvie Wilkinson and J. Michael Luttig, two subjects of the speculation, sat on the federal court of appeals for the Fourth Circuit in Virginia. In two cases decided in June 2000, Luttig wrote opinions attacking relatively liberal positions that Wilkinson was taking. In an environmental case Luttig linked Wilkinson’s position with that of two liberal Supreme Court justices. In a freedom of speech case Luttig used Wilkinson’s name more than fifty times, with four of the mentions coming in a paragraph that described sexually explicit material related to the case. (Urofsky v. Gilmore 2000; Gibbs v. Babbitt 2000)
On the day in 1992 that the Supreme Court announced its decision in Planned Parenthood v. Casey, Justice Anthony Kennedy talked with a legal reporter in his chambers before the announcement. Looking through his window at the crowd of demonstrators on both sides of the abortion issue, Kennedy referred to the impending decision in which he coauthored the decisive opinion. “Sometimes you don’t know if you’re Caesar about to cross the Rubicon or Captain Queeg cutting your own tow line.” Shortly before taking the bench, Kennedy asked to be alone. “I need to brood,” he explained. “I generally brood, as all of us do on the bench, just before we go on. It’s a moment of quiet around here to search your soul and your conscience.” (T. Carter 1992 39–40, 103)
A few weeks earlier, Justice Harry Blackmun spoke for an hour at a luncheon meeting of the Legal Aid Society in San Francisco. During his talk Blackmun read some of the fan mail he received from the public. He also expressed his disappointment about the Supreme Court’s growing conservatism on civil liberties issues. “It seems to get a little lonely. If I had more sense, I suppose I would turn in my suit.” According to a reporter, this statement elicited “a chorus of ‘No!’ ” Blackmun responded that he would “stay there for the moment.” (Holding 1992)
Justice Antonin Scalia frequently speaks in public about his views on legal matters. In these talks he sometimes discusses issues that the Supreme Court addresses, such as capital punishment and the church-state relationship. In a 2003 speech, for example, Scalia criticized a federal court of appeals ruling that inclusion of “under God” in the Pledge of Allegiance violated the First Amendment (Salmon 2003). Nine months later, when the Supreme Court voted to review that ruling in Elk Grove v. Newdow, Scalia acceded to the plaintiff’s request that he recuse himself from the case. (Lane 2003)
THE FIVE EPISODES just described are diverse in form, but they also have something in common: all of these judges were communicating with sets of people and trying to influence the perceptions of those audiences. In quite different ways they all sought to present themselves in a favorable light and, in the case of Judge Luttig, to cast an unfavorable light on a colleague (A. Cooper 2000).
The motivations of the judges in the first two episodes are uncertain, but each judge may have perceived a concrete advantage to his actions. Judge Nadel could anticipate that gaining the maximum publicity for his ruling in favor of a popular figure would enhance his prospects for reelection the next year. Judge Luttig could hope that by calling attention to the differences between his legal positions and those of a possible rival for promotion, he was advancing his candidacy.
Indeed, these judges’ career goals might have influenced the substance of their decisions. Judge Nadel had good reason to think that his chances for reelection would improve if his ruling was consistent with the strong views of many Cincinnati voters. Judge Luttig probably would have taken the same general positions even if he had no interest in promotion, because those positions were consistent with his well-established conservatism. But the value of distinguishing himself from Judge Wilkinson may have led Luttig to accentuate his conservative stances on free speech and environmental protection in these cases.1
Of the scholars who study judicial behavior, most would see these possibilities as credible. These scholars typically give limited attention to judges’ career goals, in part because their field focuses on the Supreme Court. Still, few would reject the proposition that judges with insecure tenure or prospects for promotion sometimes act with those considerations in mind (but see Posner 1995, 111).
The motivations of the Supreme Court justices in the other three episodes do not have as straightforward an explanation. What did these justices expect to gain from their public expressions? With life terms on the highest court, they could not be seeking retention in their positions or promotion to another position—except, perhaps, chief justice.2 Students of the Supreme Court emphasize the justices’ interest in the substance of legal policy, in making good law or good policy, more or less to the exclusion of other goals. Perhaps these three justices thought they were advancing their conceptions of good law or good policy in some fashion. It is difficult, however, to see how Kennedy or Blackmun could do much to achieve those goals through the public expressions I described earlier.
In the case of Justice Scalia, this explanation is more plausible. Scalia may seek to shape attitudes about legal issues within the public and the legal community, ultimately helping to win judicial support for the policies he favors. But this is a strategy with a highly uncertain and distant payoff, and that potential payoff would not seem to merit even the limited efforts required to undertake it. Moreover, as Scalia’s recusal illustrates, the effects can be negative as well as positive.
Yet the behavior of these justices does not defy explanation. Presented with descriptions of the three episodes, most people outside the academic world would have a ready explanation: judges, like other people, get satisfaction from perceiving that other people view them positively. If Justice Kennedy wanted to be seen as a serious and thoughtful jurist, if Justice Blackmun liked to hear that some fellow lawyers appreciated his presence on the Court, if Justice Scalia enjoyed presenting his positions to groups that agreed with him, those motivations accord with what we know about human nature. Nor, I suspect, would most students of judicial behavior dissent from this interpretation. After all, these scholars seek to explain judges’ choices in cases, not the choices they make in other settings.3
But what if a desire for approval actually affected the positions these justices took in cases? Perhaps Kennedy’s unexpected position in Planned Parenthood v. Casey, upholding the bulk of what the Court had decided in Roe v. Wade, reflected an interest in approval from the lawyers to whom he was talking through a reporter. Perhaps Blackmun was reinforced in his shift toward a more liberal record by the favorable reactions of liberal groups such as the one to which he spoke. Perhaps Scalia takes more extreme positions than he otherwise would in order to win accolades from conservative groups with which he interacts.
Students of judicial behavior might agree that this type of influence exists. But their research hardly ever takes it into account, because it involves a motivation that does not fit in the models that dominate scholarship on judicial behavior in political science. In those models judges are impervious to influence from others, or they are susceptible to this influence only for instrumental reasons—in the case of the Supreme Court, as a means to help them achieve good legal policy.
This book departs from those conceptions of the relationship between judges and their audiences. I argue that judges care about the regard of salient audiences because they like that regard in itself, not just as a means to other ends.4 Further, I argue, judges’ interest in what their audiences think of them has fundamental effects on their behavior as decision makers. Through their choices in cases, judges engage in self-presentation to audiences whose esteem is important to them.
Because this argument does not fit within the dominant models, it offers a different perspective. The central purpose of this book is to show how that perspective can enhance our understanding of judges’ choices. This perspective is intended as a means to think in new ways about issues in judicial behavior and, in the process, to strengthen the dominant models.
The book’s inquiry is limited to courts in the United States. I give primary attention to higher courts, especially the Supreme Court.5 That emphasis mirrors the subject matter of scholarship on judicial behavior. Further, it is in higher courts that the perspective of audiences raises the most serious challenges to existing understandings of judicial behavior. However, my interest extends to lower courts. As I discuss in chapter 6, a perspective based on judges’ relationships with their audiences is one means to study lower courts in the same terms as higher courts.6
The place to begin my inquiry is with the dominant models.7 The next three sections of the chapter review those models with the aim of identifying their assumptions, examining the implications of those assumptions, and considering their limitations. The final section sketches out an audience-based perspective as a means to expand our understanding of judicial behavior.
At the outset, I should underline a distinction between the views of scholars and the scope of the models they employ. I emphasize the restrictive assumptions of the dominant models concerning judges’ motivations. Models are intended to simplify reality for analytic purposes, and many scholars who adopt those models would accept broader conceptions of judicial motives. Some explicitly caution that their models do not fully encompass the forces that shape judicial behavior (e.g., Maltzman, Spriggs, and Wahlbeck 2000, 27–28; see Epstein and Knight 1998, 49). Thus I characterize not the views of scholars in the field but the models that dominate research in the field.
MODELS OF JUDICIAL BEHAVIOR
The scholarship on higher courts has depicted three ideal types of judicial behavior, typically labeled legal, attitudinal, and strategic. In a pure legal model, judges want only to interpret the law as well as possible. For this reason they choose between alternative case outcomes and doctrinal positions on the basis of their legal merits. In a pure attitudinal model judges want only to make good public policy, so they choose between alternatives on the basis of their merits as policy. In most pure strategic models judges seek to make good policy, but they define good policy in terms of outcomes in their court and in government as a whole. Thus they may deviate from their most preferred policy position in a case as a way of helping to secure the best outcome.
In practice, the picture is more complicated. Scholars have developed variants of the three ideal types, and only one of those types—the strategic type—currently has many adherents in its pure form. I begin by considering models based on the strategic type and then turn to the mixed models that developed from the other types.8
Like related terms, strategic is used in various ways (see Baum 1997, 90 n. 2; Caminker 1999). My summary of strategic models indicates my own usage of the term. Strategic judges consider the effects of their choices on collective outcomes, both in their own court and in the broader judicial and policy arenas. In other words, they do not simply do the right thing as they see it, such as voting for the most desirable policy on freedom of speech. Rather, they seek to have the right thing triumph in their court’s decision and, more important, in public policy as a whole. For this reason, whenever strategic judges choose among alternative courses of action, they think ahead to the prospective consequences and choose the course that does most to advance their goals in the long term.9
To achieve this result, judges might vote and write opinions that differ from their own conceptions of the right thing. Thus we cannot assume that a judge’s vote in a freedom of speech case fully reflects the judge’s conception of good policy. If the goal of an appellate judge is to advance freedom of speech as much as possible, she might take a more moderate position in a particular case in order to win majority support for a pro– free speech ruling by her court. The judge might also try to avoid a decision that provokes Congress to enact legislation limiting free speech.
Judges could act strategically to advance a variety of goals, not just good policy. State judges who lack life terms might balance policy goals against their interest in remaining judges (e.g., M. Hall 1992; Langer 2002). Judges might seek to advance their conceptions of good law (Ferejohn and Weingast 1992) or strive to achieve both good law and good policy (Spiller and Tiller 1996). But in most strategic models that are applied to federal courts, judges act solely on the goal of achieving good policy.
As suggested already, strategic policy-oriented judges direct their efforts at multiple objects. They consider colleagues on their own court (Maltzman, Spriggs, and Wahlbeck 2000; Hammond, Bonneau, and Sheehan 2005) and judges on other courts (Songer, Segal, and Cameron 1994; Cameron, Segal, and Songer 2000). They also concern themselves with the other branches of government (Spiller and Gely 1992; Schwartz, Spiller, and Urbiztondo 1994; Epstein, Knight, and Martin 2001) and the general public (Epstein and Knight 1998, 157–77). As a result, the strategic judge is subject to influence from a variety of sources. One possible result underlines the impact of strategy: judges who do not care about making good law nonetheless might base their decisions on legal considerations, because they think the public expects them to act on a legal basis. They do so in the belief that if they act in accord with public expectations, the public will be more willing to accept and comply with their decisions (Epstein and Knight 1998, 163–77; see Easterbrook 1992, 287).
Models of strategic judges who are devoted to achieving good policy have some very attractive features. They provide a comprehensive and coherent framework for the analysis of judicial behavior. They also promote rigor in the analysis of that behavior. Primarily because of these virtues, strategic models have become highly influential, and a strategic conception of judicial behavior is now the closest thing to a conventional wisdom about judicial behavior (see Epstein and Knight 2000).
Attitudinal and Quasi-Attitudinal Models
In the terms used in rational choice analysis, judges of pure attitudinal models act sincerely (or naively) rather than strategically. Devoted to good policy as a goal, attitudinal judges act directly on their policy preferences without calculating the consequences of their choices. They cast votes and write opinions that perfectly reflect their own views, regardless of what their court colleagues and other policymakers might do in response. In this model a judge’s vote on an issue involving freedom of speech reflects solely what the judge thinks is good policy on the issue in question.
In the judicial behavior scholarship of the 1960s and 1970s, something like a pure attitudinal model was the leading approach to the study of the Supreme Court (e.g., Schubert 1965; Spaeth 1979). Scholars frequently applied attitudinal models to other courts as well, though often implicitly. At the same time, students of judicial behavior—including adherents to attitudinal models—were analyzing strategy in decision making (Schubert 1963; W. Murphy 1964; Rohde and Spaeth 1976, chs. 8–9). For this reason some accounts of judicial behavior appeared to lack theoretical coherence.
Since the 1990s some scholars have overcome this problem by adopting fully strategic models of judicial behavior. Meanwhile, the leading proponents of the attitudinal approach have made more explicit how they integrate strategy into their approach (Segal and Spaeth 2002). They have set aside the original premise of the attitudinal model that judges act sincerely on their policy preferences: as they now see it, judges think strategically. But they argue that in two respects, strategic considerations have only limited impact on the positions of Supreme Court justices. First, justices seldom have to modify their positions to accommodate the preferences of those outside the Court, such as Congress and the general public (Segal and Spaeth 2002, 326–49, 424–48; Segal 1997). Second, while justices act strategically in regard to their colleagues at preliminary stages of decision (such as case selection), they need not do so when they vote on case outcomes, since final votes on outcomes are the last stage in the decision process (Segal and Spaeth 2002, 96–97).
Segal and Spaeth’s depiction of the Supreme Court, then, is a hybrid of ideal types. Justices think strategically, but for the most part they act attitudinally. While other scholars have not explicitly adopted this formulation, it is implicit in much of the research on judicial behavior.
Models with a Legal Component
The judge who conforms to the legal ideal type seeks only to make good law. In other words, this judge aims to interpret the law accurately, without concern for the desirability of the policies that result. Faced with a free speech case, the law-minded judge simply seeks the best interpretation of the First Amendment.
This explanation of judicial behavior maintains a foothold in law school teaching and exerts some influence on legal scholarship (Cross and Nelson 2001, 1439–43). But at least since the legal realist movement, few scholars have fully accepted this explanation (W. Fisher, Horwitz, and Reed 1993; R. Smith 1994). However, legal realists differed in the role they ascribed to the law. Those who could be called radical legal realists largely rejected the law as a basis for judicial choice (e.g., J. Frank 1930), a position taken more recently by critical legal theorists (Dalton 1985; D’Amato 1989). In contrast, moderate legal realists left room for the law in judges’ decision calculus (e.g., Cardozo 1921). In their conception, judges who decide an issue involving freedom of speech make their choices on the basis of both what they deem to be good policy and how they think the First Amendment is best interpreted.
This moderate version of legal realism is implicit in much of the research by legal scholars. It has also been accepted by some political scientists such as C. Herman Pritchett, a key figure in the movement to analyze judicial behavior systematically (see, e.g., Pritchett 1954). These scholars agree that judges’ policy preferences play a substantial part and perhaps the largest part in judicial decisions. But they argue that legal considerations also play a part, chiefly because judges reach decisions within a legal framework.
To a degree, advocates of the other models have kept adherents to moderate realism on the defensive (Gillman 2001b, 466). But their position has achieved something of a revival in a school that is usually labeled historical institutionalism (R. Smith 1988; Clayton 1999; Gillman 1999; “Courts” 1999; Whittington 2000). Moreover, some scholars from outside this school incorporate legal considerations into their analyses of judicial behavior (H. Perry 1991; Songer and Lindquist 1996; D. Klein 2002; Richards and Kritzer 2002; Kritzer and Richards 2003).
Moderate legal realists who focus on the interplay between legal and policy considerations give limited attention to strategy. However, historical institutionalism emphasizes the links between courts and the broader political regime (Gillman 2004). Gillman (1997) and Graber (2004) made explicit what is implicit in some writings by moderate realists: judges act strategically, but their strategic behavior is in the service of both legal and policy goals. Indeed, it has been suggested, much of the strategic interaction that occurs within appellate courts should be understood as an effort to enhance coherence in the law by achieving an opinion that a majority of judges can support (Kornhauser and Sager 1993, 52–53; Edwards 2003).
SHARED ASSUMPTIONS:THE JUDGE AS MR.SPOCK
The competing models of judicial behavior differ on a number of issues, ranging from the place of law in judging to the public’s influence on the Supreme Court. Much of the scholarship in the field focuses on these differences, as scholars advocate particular positions and conduct research on points of disagreement among models.
Significant as these differences are, the issues on which these models agree are even more fundamental. All the major models share the premise that policy considerations powerfully influence the choices of higher-court judges. Further, each model allows for judicial strategy.
At a deeper level, these models share a basic premise—so basic that it is hardly ever noticed (but see Baum 1997, 27–28; Schauer 2000, 615–17). Each model incorporates the assumption that Supreme Court justices act solely on their interest in the substance of legal policy, whether that interest is centered on policy or on a combination of law and policy. No other goal has any impact on the justices’ behavior. The same assumption is often applied to judges on other federal courts, though not always explicitly.
This is a striking assumption, because it treats judges as people whose choices are based on a very narrow set of goals. If this assumption is accurate, judges’ interest in shaping legal policy must be far stronger than other goals that might affect their decisions. To assess this assumption, then, it is necessary to consider both the strength of good legal policy as a type of goal and the potential impact of alternative goals. I will give primary attention to the Supreme Court, because the argument that judges act only on their interest in legal policy is most widely and most fully accepted for the Court.
The Strength of Legal Policy as a Goal
Central to the dominant models of judicial behavior is the premise that judges on higher courts care a great deal about the content of legal policy. The reasons why judges might care so much are hardly ever spelled out. Rather, having observed patterns of judicial behavior that seem consistent with a strong judicial interest in legal policy, scholars see no need to inquire into the bases for this interest.
This does not mean that such bases cannot be identified. Consider first of all the judge who is devoted to making good policy through decisions. Undoubtedly most lawyers who become judges, like other people in law and politics, have strong preferences about a range of policy issues. As a result, they gain satisfaction from the feeling that they are making or contributing to good policy (W. Landes and Posner 1975, 887; Higgins and Rubin 1980, 130; Posner 1995, 131). If commitment to a vision of desirable public policy can motivate political action in other arenas, such as interest group activity (Moe 1980; Verba, Schlozman, and Brady 1995, ch. 4), surely a similar commitment can influence judges’ choices. Further, judges who act strategically on their policy goals may enjoy the feeling that they are winning victories and exerting influence.
Similar motivations can spur judges’ efforts to make good law as distinct from good policy. Socialized through their legal training and practice, judges gain satisfaction by interpreting the law as well as they can (Stinchcombe 1990; D. Klein 2002, 11–12; Cross 2003, 1473–76). Judge Richard Posner (1995, 131, 133) has suggested that judges enjoy following the rules of the “game” of judging, such as compliance with the strictures of law.
It is very difficult to assess the general strength of these motivations, but it is far from obvious that they are strong enough to dominate judicial decision making. For one thing, a judge does not benefit directly by making good law or good policy. The state of government policy on issues in labor relations or freedom of religion is unlikely to have much effect on a judge’s life. Few judges who vote to limit the use of capital punishment expect to become defendants in murder cases.
Thus, working to achieve legal and policy goals does not serve judges’ self-interest as conventionally defined. “If one looks only at financial incentives,” one legal scholar argues, “it is difficult to explain why most federal judges do not simply decide their cases by flipping a coin, and then take the rest of the day off and go fishing” (Stout 2002, 1606). The same is true of other aspects of self-interest.10
The Impact of Other Goals
It might be, however, that the goal of making good legal policy dominates judicial decision making by default. If other goals are pretty much irrelevant to the task of judging, then judges devote themselves to achieving good law or good policy because they have nothing else to do with their power. This is essentially the position of some of the most influential analyses of Supreme Court decision making (Rohde and Spaeth 1976, 72–74; Epstein and Knight 1998, 36–49; Segal and Spaeth 2002, 92–96).
In considering alternative goals, these analyses give nearly exclusive attention to career maintenance and advancement, an obvious possibility for public officials. Students of judicial behavior accept the relevance of this goal to lower-court judges, especially elected state judges (e.g., M. Hall 1992, 1995). In contrast, Supreme Court scholars argue that it has little relevance to the justices (Rohde and Spaeth 1976, 72–74; Epstein and Knight 1998, 36–39; Segal and Spaeth 2002, 93–96). In their view, justices’ life terms and the attractiveness of their positions rule out self-interest in the form of career considerations. This answer is a bit too sweeping, in that some justices have been interested in nonjudicial positions or—as noted earlier—elevation to chief justice.11 Further, it is possible that this interest affects the positions that ambitious justices take in certain cases. Still, for most justices career goals are indeed irrelevant.
While political scientists say little about other alternatives to the goal of achieving good legal policy, career goals do not exhaust the possibilities. “Indeed, once salary and tenure are guaranteed, ironically the door is open for many other factors to influence judicial behavior” (Burbank and Friedman 2002, 27). Scholars with an economic orientation have posited a multiplicity of goals for judges on various courts (e.g., Anderson, Shughart, and Tollison 1989; Miceli and Cosgel 1994; Cass 1995; Posner 1995, ch. 3; Toma 1996; Drahozal 1998; Gulati and McCauliff 1998; Georgakopoulos 2000; Bainbridge and Gulati 2002). For the Supreme Court some of these goals, like career goals, can be set aside as very weak at best. The same is true of goals whose impact is idiosyncratic. To take an extreme example, one could imagine a situation in which a Supreme Court justice has an incentive to decide a dispute over a presidential election in a way that makes the justice more comfortable about retiring.12 But that incentive, decisive though it might be in a major decision, could not be an important component of any theory of judicial behavior.
In contrast, some other goals might have a more regular impact. One candidate is pleasant working relations with colleagues. On appellate courts, judges must work together. Even in an era with limited face-toface contact among Supreme Court justices, decision making requires considerable interaction among the justices. Frictions are inevitable. The potential for more serious hostility always exists and sometimes comes to fruition (P. Cooper 1995). The justices would be an unusual group indeed if most of them did not prefer to minimize conflict (see Edwards 2003).
In turn, that goal may affect the justices’ behavior. This impact should not be exaggerated: occasional reading of dissenting opinions makes it clear that justices sometimes take actions that could foster conflict. Moreover, it can be very difficult to identify the motivations that underlie efforts to minimize conflict: justices who seek to maintain a pleasant environment and justices who seek to win colleagues’ support for their legal and policy positions might behave in quite similar ways (see W. Murphy 1964, ch. 3). Yet it is quite plausible that concern for interpersonal relations affects behavior such as opinion writing and joining.
Another possibility is usually called leisure (Posner 1995, 124–25), though perhaps a better label is limiting workload. Judges on any court might seek to maximize leisure time or time for other pursuits by working less rather than more at judging. It is noteworthy that a federal court of appeals judge (Posner 1995, ch. 3) has emphasized workload as a goal that can affect judges’ choices (see also Macey 1994; Drahozal 1998, 475– 76; Gulati and McCauliff 1998, 172; Bainbridge and Gulati 2002). Occasionally, a judge is open about holding this goal. “I didn’t get elected to work hard. I worked hard as a trial lawyer” (Flemming, Nardulli, and Eisenstein, 1992, 98). The possibility that Supreme Court justices care about limiting their workloads is supported by the anecdotal evidence that some work less than full days (e.g., Woodward and Armstrong 1979, 270) and by the array of extracurricular activities in which they engage during Court terms, activities that range from law school speeches to duck-hunting expeditions.
Workload considerations affect agenda setting. In courts with discretionary jurisdiction and large numbers of petitions for hearing, there is an unavoidable maximum to the number of cases that judges can accept and handle adequately. The most obvious sign that judges seek to limit their workloads is that they accept substantially fewer cases than that maximum.
This has been the Supreme Court’s practice in the last two decades. There was a precipitous decline in the number of certiorari grants per term during the early years of the Rehnquist Court, and the new lower level has been maintained since that time. Commentators have offered various explanations for this decline (Hellman 1996; O’Brien 1997). Some have suggested (though gently, at least in print) that the Rehnquist Court justices preferred doing less work to more (Lacovara 2003). Evidently aware of this interpretation, Justice Ginsburg protested that “the cutback in opinions doesn’t mean that the court is becoming a lazy lot” (Biskupic 1994c).
In decisions on the merits, judges’ goal of limited workload may result in deference to other people. Judges in some appellate courts strongly defer to the members of central staffs who prepare proposed decisions and opinions (Davies 1981; Stow and Spaeth 1992). Judges may also defer to the colleague who is initially assigned a case, suppressing disagreement in order to avoid writing separate opinions (Wold 1978, 64; Linder 1985, 486; Maltzman, Spriggs, and Wahlbeck 2000, 24). At the extreme, this deference is so great that the outcome of cases largely depends on the identity of the assigned judge (Sickels 1965).
More broadly, workload considerations limit judges’ efforts to get decisions “right” as law or policy. This effect is especially important. The distance between a decision and a judge’s ideal will tend to decline as the judge devotes more time and effort to the decision. Like people in other positions, judges must determine where to draw the line, at what point to stop working on a case because the likely benefits of additional work are outweighed by the value of spending time on other pursuits. If judges give some priority to their workloads, as they surely do, they stop short of achieving what would be the best possible result as they would define it.13
As suggested earlier, still another possible goal for judges is the approval of individuals and groups that are important to them. Under some conditions this goal reinforces judges’ interest in good legal policy. Under others it constitutes an alternative basis for judicial choices. The impact of judges’ interest in the esteem of others will be considered in the last section of this chapter and in the chapters that follow.
The Realism of the Dominant Models
From this vantage point, all the leading models of judicial behavior share a limitation. Each model portrays Supreme Court justices (and, in some formulations, judges on other courts) as single-minded seekers of good legal policy, whether that means good policy or some combination of good law and good policy. If judges’ incentives to pursue this goal are less than overwhelming, and if other motivations are relevant to their work, then the accuracy of this portrayal is questionable.
This limitation is especially relevant to strategic action by judges. As discussed earlier, most students of judicial behavior now believe that judges engage in a good deal of strategic action. Indeed, scholars increasingly accept a model in which judges are fully strategic in acting on their policy goals. For those reasons it is useful to consider the motivational bases for this model in some detail.
One source of the appeal of strategic models is their apparent realism. If judges care about the content of legal policy, then it seems short-sighted or naive for them simply to vote for the alternative that they most prefer in itself. Rather, they should focus on the ultimate consequences of their choices and act to make those consequences as consistent with their preferences as possible. Thus it seems self-evident that judges act strategically to advance their policy goals.
This perception of realism is not necessarily justified. Several scholars have offered formulations of the conditions under which people will adopt the strategies that maximize achievement of their goals (Kreps 1990, 140–45; Kelley 1995, 101; M. Taylor 1995, 225–28; Green and Shapiro 1995, 267; Cameron 2000, 75–78). Most of these conditions fall into two categories, each more or less obvious. The first is the strength of people’s incentives to behave strategically. The second is the extent of their capabilities to identify and follow the best strategy in any situation. Capabilities might seem relevant only to the effectiveness of strategies, not the choice to behave strategically. But capabilities influence incentives: the more difficult it is to play a strategic game well, the less reason people have to play it at all.
Because incentives are so important, we would expect people to be highly strategic in making economic decisions. Entrepreneurs, employees, and consumers all have a strong self-interest spurring their efforts to maximize their benefits and minimize their costs. Because of their limited capabilities, people frequently deviate from the course that would best serve their economic self-interest (Thaler 1991, 1992; Conlisk 1996; Kahneman and Tversky 2000; Barberis and Thaler 2003). Still, their incentives to behave strategically are strong.
In the political arena, some goals are sufficiently important that people have strong incentives to act strategically on their behalf. Perhaps the most obvious example is reelection, which has received considerable emphasis in studies of Congress (Mayhew 1974; Arnold 1990). Retaining office can serve a variety of important purposes. For that reason most members of Congress have good reason to make the considerable efforts that are required to maximize their chances for reelection. Continued tenure in office is not as fundamental as the basic goals of economic actors; among other things, not all members of Congress want to stay there. Even so, a model of concerted strategic action to secure reelection fits the reality fairly well.
As noted earlier, judges and policy goals are different. Ordinarily, judges gain only intangible benefits from advancing their favored policies. These benefits are meaningful, but they do not seem nearly as substantial as those that accrue to economic actors or even to elected officials who want to retain their positions.
There is also a collective action problem. Whatever benefit judges derive from the state of public policy in a particular field, the choices of a single judge ordinarily have only a marginal impact on the totality of policy. Most directly, multiple judges participate in formulating judicial policy on an issue. An appellate judge faces limits imposed by the actions of colleagues and other courts. Justice William Brennan was widely regarded as someone who could exert significant influence over his fellow justices (Eisler 1993), but in the increasingly conservative Court of the 1970s and 1980s he won fewer and fewer victories.
Further, research on the implementation of judicial decisions has underlined the central role of nonjudicial actors in shaping the ultimate impact of decisions (Rosenberg 1991; Canon and Johnson 1999). To take the classic example, the pace of school desegregation in the South after Brown v. Board of Education was determined chiefly by state and local officials and by the other branches of the federal government, not by the Supreme Court. Even with their best efforts, judges can exert only limited influence over the choices of other government officials. In light of this reality, the effort required to act strategically might not be justified by the likely gain for the judge’s policy goals.
In this respect, then, strategic models do not seem uniquely realistic. Judges who act strategically do not secure any tangible benefits that are denied to judges who act sincerely. While they may gain intangible benefits when their favored legal policies come closer to attainment, their limited capacity to affect outcomes reduces the quantity of these benefits. Moreover, judges may gain as much or more satisfaction by simply taking what they identify as the right position, regardless of the ultimate outcome. When there is relatively little to be achieved by acting strategically, judges cannot be assumed to engage in fully strategic behavior.
Indeed, if we take judges’ interest in their workloads into account, strategic models appear less realistic than their competitors. Judges who try to interpret the law as well as possible or to identify the positions that are most consistent with their policy preferences have to work hard, but at least the work of the legal or attitudinal judge ends there. The task of the strategic judge continues, and that judge may need to engage in intensive labor in order to choose something approaching the optimal strategy (see Merrill 2003, 620–21).14
Consider, for instance, the job of Supreme Court justices who try to determine whether they should adjust their interpretations of a statute in order to avoid a congressional override of the Court’s decision. To estimate whether the current Congress would override a prospective decision is an intricate task, and some of the intricacy is reflected in the complex ideological models that scholars have used to make their own estimates (Segal 1997; Bergara, Richman, and Spiller 2003; Sala and Spriggs 2004).
Other complications arise. For example, Congress can reverse a court decision at any time. So the justice who is committed to avoiding overrides must estimate the ideological composition and political environment of future Congresses. That task requires, among other things, a capacity to predict election results that thus far has eluded skilled political scientists. Another complication is obvious but sufficiently daunting that students of judicial behavior generally ignore it (but see Spiller and Spitzer 1995): since individual justices differ in their policy preferences, they need to consider each other in developing strategies aimed at Congress. In practice, Court action to avoid an override may produce a policy no closer to the justices’ preferences than the policy that would result from an override, without the benefit of delay in its adoption (Cross and Nelson 2001, 1454). Finally, because the Court can respond to congressional overrides and shape their effects (Segal 1997, 32), the conscientious justice needs to calculate outcomes in a multistage game.
Of course, Congress is not the only target for strategic behavior. The strategic justice must face the task of predicting public reactions to the Court’s decisions, a task that has its own complexities. One example is the justice who believes that the Court jeopardizes public support for the Court and ultimately its ability to win acceptance for its policies by making decisions that conflict with majority views in the public. If this justice contemplates voting for an outcome that is likely to be unpopular, the justice must estimate how much public support would be lost as a result of that decision and, in turn, how much this loss of support would affect the Court’s effectiveness as a policymaker. That estimated effect would then have to be compared with the estimated benefit to the justice’s policy goals from the decision in question.
The judgments needed to choose effective intra-Court strategies are a good deal easier to make. This is one reason that justices can be expected to behave more strategically toward colleagues than toward targets outside the Court. But even within the Court, the task of predicting the long-term consequences of different actions may be difficult. Would Justice Scalia ultimately win more support for his policy positions if he traded his hard-line advocacy of those positions for William Brennan’s gentler approach? Some very complicated calculations would be required to answer that question with any confidence.15
Most research that employs strategic models assumes that justices are willing to take on the burdens of acting strategically. Justices amass all the relevant information and make the calculations needed to determine whether, and how much, they must modify their preferred positions in light of other participants in the policy process. Justices act in this way, sacrificing a large share of the time they might spend in other pursuits, even though there is a good chance that their predictions will be inaccurate. If they do manage to achieve a high level of predictive accuracy, the gains for their preferred policies are likely to be limited. In any event, they derive no concrete benefits from their success.
In contrast, consider justices who take a different approach. These justices engage in a certain amount of strategic behavior within the Court, aimed at securing majorities for the positions they favor. But for the most part they simply take the positions that accord with their policy preferences. If their positions initially put them on the Court’s minority side, they usually accept their minority status and write or join dissenting opinions to disagree with the majority opinion. Nor do they worry much about responses to the Court’s decisions by the other branches or the public. While these justices are happy when their colleagues and other policymakers agree with them, they derive satisfaction primarily from casting votes and writing opinions that accord with their sense of what is right.
Because of the approach they take, the job of these justices consumes much less time and effort than the job of the fully strategic justice. It also involves much less frustration. As a result, they can work on cases at a deliberate pace and enjoy a reasonably relaxed life. Advocates of strategic models might label these justices as naive, because they largely eschew strategy. But that label does not seem quite right: these justices know what they are doing, and they are happy with the results.
Even these justices or counterparts who care about law rather than policy might be considered idealistic. Such judges make their choices with the goal of adopting good legal policy, a goal whose achievement brings them no tangible benefits. They need to labor to identify the best choices while recognizing that they may fail to do so. Thus there is an element of Mr. Spock16 in the judges of the pure legal and attitudinal models, judges who act without emotion or self-interest in order to advance the general good. As the leading expert on Mr. Spock explained, “In the Vulcan culture, one simply does what is right” (Nimoy 1977, 104).
But it is the judges of strategic models who most resemble Mr. Spock. These judges court exhaustion with their arduous and often futile efforts to advance their conceptions of good policy, efforts they expend only for the personal satisfaction of trying to improve public policy. This is a lot to expect of them. People often behave in ways that are not strictly self-interested (Mansbridge 1990; Monroe 1996). Even so, by standards of ordinary behavior the fully strategic judge seems enormously altruistic. Attractive as Mr. Spock’s behavior may be, most human beings—including most judges—do not act like him.17
LIMITATIONS OF THE DOMINANT MODELS
The most influential models of judicial behavior share not only a basic assumption but also a limitation, the lack of a persuasive theory of judges’ motivations. Each model assumes that judges seek only to achieve good legal policy. Some scholars outside political science have identified motives for pursuing that goal, but these motives seem only moderately strong. Nor does this goal have monopoly status by default; there are other goals that could shape judges’ choices. As a result, the dominant models in their current forms are not entirely realistic. That is especially true of strategic models, which have become the leading models in the field.
For many students of judicial behavior, this critique will seem misguided on either (or both) of two grounds. First, as they see it, empirical evidence demonstrates the dominance of good legal policy as a goal and strategy as a means to that goal for judges on higher courts, especially the Supreme Court. Political scientists have amassed a large body of evidence indicating that Supreme Court justices seek to make good policy (see Segal and Spaeth 2002). Increasingly they are amassing evidence of strategic action by the justices to secure good policy, especially in their interactions with other justices (see Maltzman, Spriggs, and Wahlbeck 2000). Given the volume of this evidence, it could be argued that any concerns about motivation are of no particular importance.
Second, models do not need to accord fully with reality in order to advance our understanding. Indeed, their simplification of reality can be a virtue, because it makes analytic problems more tractable. Even if judges are not fully strategic, for instance, that conception of their behavior produces insights and structures empirical analysis in useful ways.
Both points have considerable validity, and the value of the dominant models for an understanding of judicial behavior is quite clear. Models based on the attitudinal ideal type have told us much about the structure of behavior on appellate courts, and they have established fundamental realities about judicial decision making. Models that posit a mix of legal and policy considerations in decisions illuminate the complexities of judges’ decision processes and integrate competing conceptions of those processes. Strategic models have encouraged scholars to probe judges’ choices in new ways, and they have also brought greater theoretical rigor to the study of judicial behavior. These are hardly trivial contributions, and they continue today with a wave of research that is expanding our understanding of judicial behavior.
Yet these achievements do not dispose of the problem, and both grounds for challenging my critique can be contested. First, the empirical evidence that scholars have amassed proves less than is often thought. This evidence does provide considerable evidence of the kinds of behavior posited by the dominant models. In light of this evidence, to take the Supreme Court as an example, it would be foolish to deny that the justices’ policy preferences have strong effects on their choices or that the justices engage in a good deal of strategic behavior.
But the existing evidence does not establish that justices are motivated solely (or even overwhelmingly) by policy goals. What it shows most clearly is quite important but considerably more limited: differences in the positions that the nine justices take in the same cases are best understood as a product of differences in their policy preferences. The evidence on strategic behavior is even less conclusive. It shows that a good deal of what the justices do is consistent with a strategic interpretation, but it falls well short of demonstrating that all judicial behavior takes the forms we would expect if justices are fully strategic. Moreover, much of the evidence of strategic behavior that scholars have found is consistent with non-strategic (or partly strategic) interpretations as well.18
Second, the significant benefits of simplifying assumptions in models are accompanied by significant costs. A model of strategic policy-oriented Supreme Court justices can encompass any actions that justices take. There is a beauty to simple, comprehensive models of behavior (see Lave and March 1975, 61–64). But by setting complexities aside these models can limit our comprehension of behavior.
Moreover, there is a cost to employing models whose depiction of motivations is more Vulcan than human. As economist Richard Thaler (1996, 227) describes it, individuals in orthodox economic theory are characterized by “rationality, self-interest, and self-control” in that they “can solve problems like an economist, care only about themselves, and never suffer from weakness of will.” As Thaler and others have shown, the departure of this characterization from reality creates difficulties for analysis (Camerer 1997; Thaler 2000; Rabin 2002). Judges in the dominant models of judicial behavior depart further from reality: they share rationality and self-control with orthodox economic actors but act on the basis of complete altruism rather than complete self-interest.
Further, the incorporation of certain judicial goals into the dominant models as basic assumptions limits inquiry into the bases for those goals. This has been a deliberate choice, based in part on scholars’ primary interest in prediction rather than deep explanation (Schubert 1963; Spaeth 1979, 140). This approach is fundamental to most economic analysis (M. Friedman 1953), and it is quite defensible. But it circumscribes explanations of judicial behavior.
Thus the leading models of judicial behavior teach us a great deal about why judges do what they do, but their perspective is narrow. That narrowness limits what these models in their current form can tell us. By expanding the scope of those models, we can gain a fuller comprehension of the bases for judges’ choices. One means to this end is viewing judicial behavior from other perspectives, and that is the approach I take in this book.
AUDIENCE AS A PERSPECTIVE
The perspective that I offer is one based on judges’ relationships with their audiences, people whose esteem they care about. Judges’ audiences usually include colleagues on their own courts, but for the most part these audiences are outside their courts. The impact of external audiences is a point of disagreement among scholars who work within the dominant models. For the leading proponents of a quasi-attitudinal model, at least Supreme Court justices are largely impervious to external influence (Segal and Spaeth 2002). In contrast, some scholars argue that justices are subject to significant influence from the general public (e.g., Mishler and Sheehan 1996; Flemming and Wood 1997) or Congress (e.g., Epstein, Knight, and Martin 2001; Bergara, Richman, and Spiller 2003). Studies of lower appellate courts consider elections as a vehicle of influence for the public (M. Hall 1987, 1992, 1995; M. Hall and Brace 1992; Brace and Hall 1997; Langer 2002) as well as the impact of Supreme Court review on the choices of lower-court judges (Songer, Segal, and Cameron 1994; Cameron, Segal, and Songer 2000).
What scholars who work within the dominant models have in common is the unstated assumption that judges do not care about the regard of their audiences for its own sake. According to this view, judges do not get satisfaction simply from being liked and respected. Rather, judges give attention to audiences as a means to advance their goal of achieving good legal policy. The general public, for instance, functions as a constraint. Judges cannot continue to make policy if they lose elections, and they cannot secure compliance with judicial decisions if they lose legitimacy with the public. But the Spock-like judges of the dominant models have no interest in public approval as an end in itself. If they could best advance what they see as good legal policy by making themselves objects of hatred or contempt, presumably they would do so.
This view of the relationship between judges and their audiences is not universal. One feature of the economic scholarship on judges’ goals is its recurring references to judges’ interest in securing favorable reactions to their work from other people. In part, this interest is described as a means to other ends, advancement of career goals and effectiveness in policymaking. However, several scholars cite an interest in esteem as a goal in itself (e.g., Cass 1995, 971–74; Posner 1995, 119; Drahozal 1998, 475; see Posner 1990). Some treat concern with reputation as a powerful explanation of judicial behavior (Miceli and Cosgel 1994; see Schauer 2000).
The idea that judges care a great deal about what people think of them is not very radical, and it comports with a commonsense notion of human nature (Harsanyi 1969). As I will discuss in chapter 2, it also comports with a large body of relevant research by psychologists. However, this does not establish that judges’ interest in the regard of other people influences their choices on the bench in significant ways. Indeed, because the dominant models of judicial behavior exclude this motivation, arguably the burden of proof lies with anyone who argues for its inclusion.
But perhaps the burden should lie with adherents to these models. After all, the dominant models implicitly treat one of the most fundamental human qualities as irrelevant to judges’ choices. Why would judges’ interest in approval not affect their choices?
Wherever the burden ought to lie, the case needs to be made. In this book I seek to demonstrate the value of taking into account judges’ interest in the esteem of audiences that are important to them. More specifically, I try to show that a perspective based on judges’ interest in the approval of their audiences can enhance our understanding of judicial behavior in several related respects.
First, it supplies some of the motivational basis for patterns of judicial behavior that are incorporated into the dominant models. Most important, concern with audiences helps to explain why judges may commit themselves to making what they see as good law or good policy. In the process, familiar patterns of judicial behavior can be understood in a different way.
Second, this perspective provides ways of thinking about debates between competing models. Students of judicial behavior disagree about the relative weights of legal and policy considerations in judges’ decisions and about the extent to which judges behave strategically. Relationships between judges and their audiences offer a vantage point on those issues.
Third, an audience-based perspective helps to explain patterns of judicial behavior that diverge from all the leading models. The explanations of decision making that flow from those models are incomplete in some respects and simply wrong in others. The relationships between judges and their audiences fill in some of the gaps and correct some inaccuracies.
Fourth, this perspective points to some new lines of inquiry into judicial behavior. These lines of inquiry relate not just to the impact of judges’ audiences but to other issues as well. In this way the scope of empirical research on the bases for judges’ choices can be expanded.
As a consequence of these functions, an audience-based perspective suggests ways to modify and expand the dominant models. The value of these models as bases for inquiry has been amply demonstrated. Moreover, they provide a structure for the accumulation of knowledge about judicial behavior. With adaptations, their contributions can be even greater.19
I should make explicit what is implicit in that list of functions. I believe that judges’ motivation to win the approval of their audiences can explain a good deal about their choices as decision makers, and I seek to demonstrate that explanatory value. But even more important is the value of thinking about judges and their audiences as a means to develop new understandings of judicial behavior.
The book’s exploration of relationships between judges and their audiences begins in chapter 2, which considers audiences in broad terms. The chapter examines and draws from the relevant scholarship in social psychology. The concept of self-presentation is used as a link between people’s interest in approval and their behavior. The chapter then applies the lessons of that scholarship to judges, discussing judicial self-presentation and its part in judicial behavior.
Chapters 3 through 5 probe the relationships between judges and several audiences that might be relevant to them. Chapter 3 examines the audiences that scholars give substantial attention: court colleagues, the general public, and the other branches of government. In the standard images of these audiences, judges respond to them for instrumental reasons, as a means to other ends. I raise questions about the strength of instrumental motives for responding to these audiences. Drawing from the perspective developed in chapter 2, I also consider personal motives for responding to these audiences, motives based on an interest in their esteem for its own sake.
Chapters 4 and 5 discuss other audiences that have received limited attention from scholars but that may exert influence because of judges’ personal motives for seeking their approval. Chapter 4 considers judges’ social groups and the legal community, including judges on other courts. Chapter 5 deals with “policy groups,” people who share views about legal policy, and the news media. The chapter also considers the hypothesis of political conservatives that a “Greenhouse effect” has moved some Supreme Court justices to the left, a hypothesis that pulls together some themes concerning judges’ self-presentation to their audiences.
Chapter 6 examines the ways that an audience-based perspective can inform our understanding of judges’ choices. Drawing on the discussions in earlier chapters, it suggests how to incorporate the book’s inquiries into the study of judicial behavior. As in the book as a whole, I argue that drawing out the implications of relationships between judges and their audiences will lead to richer explanations of that behavior.
Return to Book Description
File created: 8/7/2007
Questions and comments to: firstname.lastname@example.org
Princeton University Press