Chicago Unbound Chicago Unbound The Roles of Judges in Democracies: A Realistic View The Roles of Judges in Democracies: A Realistic View

What are the “obligations” of judges in democracies? An adequate answer requires us to be realistic both about democracies and about law. Realism about democracy demands that we recognize that electoral outcomes are largely, though not entirely, unrelated to concrete policy choices by elected representatives or to the policy preferences of voters, who typically follow their party based on “tribal” loyalties. The latter fact renders irrelevant the classic counter-majoritarian (or counter-democratic) worries about judicial review. Realism about law requires that we recognize that judges, especially on appellate courts, will inevitably have to render moral and political judgments in order to produce authoritative resolutions of disputes, one of the central functions of a legal system in any society. That means it is impossible to discuss the “obligations” of judges without regard to their actual moral and political views, as well as the moral and political ends we believe ought to be achieved.

is central to civic peace. In any case, let us call making decisions according to pre-existing standards the Minimal Role of a judge.
Green proposes three further "obligations" of judges, following H.L.A. Hart: (1) "Judges have an obligation to apply valid law in making rulings…and make correct findings of fact, know what the law is, keep their knowledge of the law up to date…."; (2) "Judges have an obligation to keep the law 'legally in good shape'" (quoting John Finnis, and meaning judges should resolve unclarity and conflicts); (3) "Judges have obligations to regulate their own processes and the conduct in their courts, and to protect the rule of law and the integrity of their jurisdiction from those who would attack it." 2 I shall call these the Minimal-Plus Roles. Probably judges should do all these things too, though no legal positivist, including Green, thinks that judges have a non-defeasible moral obligation to apply valid law In the realistic spirit suggested by my title, let me be clear at the start that "obligations"understood as moral obligations--do not really exist 3 : when we speak of moral "obligations" (or "rights" or "duties") we are expressing certain attitudes, often very intense and insistent attitudes, about what we feel other people should or should not do. 4 Expressing such attitudes is central to human life: as 2 Green, p. __. 3 Legal obligations, rights and duties do exist: they are certain kinds of social artifacts (like "money" and "stocks"), produced by legal systems, and whether or not they obtain depends on whether they satisfy the existence-conditions for such legal norms in that system. Cf Nietzsche put it, "all of life is a dispute over taste and tasting," 5 and there is value in distinguishing between the strength and character of different tastes, as talk of "obligation" nicely does. Like other people who are products of the post-Enlightenment, post-WWII consensus, my "tastes" run towards individual rights and freedom and the rule of law, in which judges can play a crucial role in constraining behavior by other state actors, since such institutional practices seem to enhance aggregate human well-being. 6 Not having illusions about obligation talk, however, is not simply a matter of philosophical fussiness: imagine if theorists who said judges have "obligations" to do X said, instead, judges are "required by God" to do X? Recognizing that the latter is false-God does not exist, and so, unsurprisingly, has no views about anything, let alone the obligations of judges-means it is possible to express contrary attitudes without violating any moral or rational norm. The conceptual and practical space is open for contrary recommendations. This is a more important point, I think, than is usually recognized in recent academic writing. Assumptions about what is the case exercise a powerful normative force on human behavior and thought-sometimes for very good reason, to be sure. For example, that it is the case that the door is closed and locked gives me a very powerful prudential reason not to try to march straight through the doorway. Let us call this the normativity of fact: that something is taken to be factual or real, to be the case, influences, sometimes quite profoundly, what people think is possible, permissible, and so on. 7 Sometimes, as in the case of locked doors, that's a good thing, but other times, it is not. So when we deny, in a realistic spirit, that moral obligations are metaphysical facts, we are making both a true metaphysical claim (the ontology of the world includes no 5 Thus Spoke Zarathustra, II:13 6 Talk of "taste" like the economists' talk of "preferences" runs the risk of flattening the important differences between our attitudes about what others ought to do. I "prefer" chocolate to vanilla, but I do not simply "prefer" toleration to genocide: those who "prefer" vanilla are not my mortal enemies, those who "prefer" genocide are. 7 [related to what Marxists used to call "reification," but also to a kind of ideological mistake about the natural and the contingent] moral obligations 8 even if it includes some locked doors) and, at the same time, we are defeating the normativity of fact where its normativity has no real justification. 9 When humans speak of moral "obligations" they are speaking of what they really want or expect certain people to do, perhaps no matter what. This is certainly revealing about the feelings of those engaged in such talk, and perhaps also revealing about their social and economic circumstances: Nazis, for example, thought of their actions in terms of obligations, 10 as did Stalinists, 11 and in both cases this included judges-not surprising, since, as Hart argued a half-century ago, where there is a legal system, judges take themselves to have obligations to apply the valid legal norms, even when others might reasonably deem them catastrophically wrong to feel that way. 12 What this should remind us of is that talk of "obligation" is an important feature of human social existence, but it is also compatible with a radical revision of human attitudes and thus a different set of attitudes about what our "obligations" are. Suitably deflating the idea of "obligation" is important if we are to think clearly about the roles of judges: for example, even in cases where judges "promise" through an oath of office to do certain things, it may be that they should sometimes act otherwise. 13 We return to that issue, also, below.
So can we generalize about the "role of judges in democracies"? We can, though subject to some constraints, since judges and democracies seem to differ along various dimensions. For example, there are "civil service" judiciaries in which appointment and promotion depends on evaluation by other judges, as in Britain; there are "political" judiciaries in which appointment depends on political Of course, "justification" itself is normative, but normativity has a "justification" in my deflated sense when it comports with instrumental desires like not walking into closed doors. A promise might be thought to generate a moral obligation to fulfill certain functions in an institutional role. Whether those functions are obligations of the role is separate from the question whether there is a real moral obligation to fulfill those functions simply because of a promise. connections and affiliation, as in the United States; and there are mixed models in which, for example, purportedly "peer" evaluation of judges for appointment and promotion is, in fact, influenced by political partisanship (I am told Italy is such a case). Democracies differ in some respects as well: some are plutocracies masquerading as democracies (for example, the United States); some are incipient authoritarian societies masquerading as democracies (for example, Hungary and Poland); and some are more conventional representative democracies with all their pathologies of class conflicts and political personalities (for example, the current United Kingdom).
A serious discussion about the roles of judges in democracies has to be realistic both about judges and about democracies. It has to acknowledge what judges really do and how democracies actually function, and it has to be clear about what judges can do against those backgrounds. It also has to be explicit about our normative aspirations and ideals (and also whether judges can realize them).
Normative aspirations and ideals tend to be inchoate and not wholly consistent, though in their broad outline they are usually clear: no one needs a moral theory, after all, to know whether they are against racism or in favor of racial equality, against chattel slavery or in favor of human freedom, against cruelty or in favor of treating people in a dignified way, against human misery or in favor of human happiness.
(The problem, after all, with academic friends of Jobbik in Hungary, the National Front in France, or Donald Trump in America is not that they failed to appreciate the correct moral theory! 14 ) I suppose throughout that judges, like other regular post-Enlightenment folk, should try to promote human wellbeing on some reasonable understanding (the theoretical differences between extant accounts rarely matter given the constraints imposed in practice on judicial power)--though the role of judges in helping to promote human well-being will vary quite a bit depending on the nature of the judiciary and the 14 Remember that normative theory delivers no decisive conclusions about what ought to be done, being hostage to "intuitive" starting points that themselves can never be defended. See generally, Brian Leiter, "The Paradoxes of Public Philosophy," Indian Journal of Legal Theory 1 (2016): 51-64. nature of the democracy. My main topic will be the kinds of realistic variations and constraints on exercise of this judicial role. I thus hope this essay will be a contribution to the on-going revival of realist political and legal theory against the moralizing fantasies of so much academic writing about these matters over the last half-century. 15 Theory will be important in what follows-for example, empirical theories about voting behavior in democracies, as well as philosophical claims about the nature of law-but normative theory will not be. 16 I start with a realistic view of democracies, and then turn to judges.

A Realistic View of Democracy
Democracies typically have two features in common: individuals have the right to vote for some number of those who hold political power and they also enjoy rights to certain individual freedoms. So, the one hand, democracies hold elections in which some significant portion of the citizenry can vote for some range of candidates or parties (perhaps two or three or four), the elections are fair and incumbents can actually be voted out of office: this is the crux of the democratic ideal that "the people rule." On the other hand, democracies also typically, in the post-War II era, protect some range of 15 For examples of the realist revival, see Richard Posner, The Problematics of Moral and Legal Theory (Cambridge, Mass.: Harvard University Press, 1999) (2015): 23-50. 16 Realists deny the priority of theorizing about moral truth to theorizing about how political and legal actors should behave, that is, they deny that we should treat political and legal theory as simply "applied ethics," without regard to the strategic questions peculiar to each domain. The central idea is that realists "should theorize about the distinctive forces that shape real politics" and judicial behavior. Enzo Rossi & Matt Sleat, "Realism in Normative Political Theory," Philosophy Compass 9/10 (2014): 689-701, at p.689. Realism, in this sense, is partly about attention to feasibility constraints (id. at 690-691), though not wholly. (An example of realist feasibility constraints in operation is the debate between the moralist Stephen Gardiner and the realist David Weisbach in Debating Climate Ethics [Oxford: Oxford University Press, 2016]) But feasibility does not exhaust the realist approach to questions about what political and legal actors should do. There are normative considerations peculiar to both the political and legal domains: in the former, stability and security; in the latter, authoritative resolution of disputes that arise in any human community. individual rights-rights to expression, privacy, and fair legal process most prominently--though again to varying degrees. 17 Call the combination of electoral rights and rights to individual freedom "Democratic Values." Debates about the role of judicial review of legislation in democracies typically focus on the electoral aspect, and how that is consistent with judicial invalidation of legislation when it violates individual rights: in the United States, this is called the "counter-majoritarian" difficulty, 18 though in parliamentary systems, it could also be called the "counter-democratic" difficulty, since judges, once again, defy the will of a democratic legislature, though less often a majoritarian one. 19 A realistic view of democracy should help us see why this latter concern is without merit, and why the real question for judges who embrace "Democratic Values" should be whether laws really protect the individual rights that are equally central to Democratic Values. 20 The reason we should disregard the supposed counter-majoritarian or counter-democratic difficulty is because there is almost never a plausible sense in which the particular laws passed by a legislature represent the views of an electorate, whose will is then thwarted. That laws reflect the view of the electorate would be true only if two further claims were true: (1) The majority of the legislature intended the legislation to achieve purpose X.
(2) The majority (or plurality) of people who elected the legislature (the "electorate") wanted them to pass legislation which achieved purpose X. 17 The protection of the individual freedoms is typically dependent on allegiance to the family of "rule of law" virtues, such as an independent judiciary, the principle that no one is above the law, the requirement to treat like cases alike, the publicity of legal standards, etc. It is not wholly dependent--contra Lon Fuller and other antipositivists--since the substantive content of the law matters crucially, even if all the other virtues are honored. 18 [cites to Bickel et al] 19 The United States Congress is not a majoritarian institution either as it happens-its Senate is obviously not, and its House of Representatives, due to gerrymandering, is not either. The United Kindgom is a special case, since the courts can not invalidate acts of Parliament on grounds of violating individual rights, but they can declare them incompatible with such rights, and thus prompt a legislative reevaluation. 20 I assume, with Mill among others, that protection of individual rights generally promotes human flourishing and well-being.
The first claim has been subjected to various kinds of critique by economists and public choice theorists, who emphasize that collective intentionality in a legislature is a fiction, and a misleading one. 21 But the more serious problems arise for the second claim, which presupposes something doubtful: that lawmakers are elected because of commitments to pass particular kinds of legislation to achieve particular goals. Even in its most plausible form, known as the "spatial theory of voting" in the empirical political science literature, it supposes only that people vote for the party that is closest (in ideological space, as it were) to their ideological "preferences." But even if voters really did that, this would rarely license the conclusion that any particular piece of legislation that achieved purpose X reflects the will of the voters.
There are, of course, the fundamental theoretical difficulties with this view due to Arrow's Theorem, 22 but even more worrisome for the spatial theory is the overwhelming evidence that voters do not have clear ideologies. For example, in the 1980s-the height of the reactionary Reagan Era in the United States, when the social welfare state was under sustained attack--nearly 65% of Americans thought the government was spending too little on "assistance to the poor," but at the same time only 25% thought too little was spent on "welfare," 23 a term that had been demonized by Reagan, even though it means "assistance to the poor"! 24 What this shows is that expressions of purported "ideology" by the 21 [cite to Easterbrook, but cf. critique in Grinsell, also R. Elkins book; but even the defenses of intention do not really establish the requisite level of specificity of intention that would make the counter-majoritarian difficulty meaningful] 22 (_____, 1960). Spatial voting theorists have produced new evidence in the last decade to try to resist the realist conclusions. An important example is Stephen A. Jessee, "Spatial Voting in the 2004 Presidental Election," American Political Science Review 103 (2009): 59-81. The problem with this literature is that its results are compatible with the realist hypotheses about the role of group identity in driving party identification, and the way in which voters conform their views to the party leaders rather than the other way around. I acknowledge that the argument of the paper is hostage to the soundness of the realist picture, but since realism in every domain I know of-international relations, political theory, legal theory, historical explanation-is always the most explanatorily successful approach, I am willing to rest my argument on the Achen & Bartels version. See my "Classical Realism," Philosophical Issues 23 [Achen & Bartels, 30] 24 We don't know how clarification about the terms would have shifted preferences, of course, but it is easy to see how it could have gone either way: some of the anti-welfare folks would have realized that population are sometimes nothing of the kind, unless by "ideology" one does not mean what voters actually believe about concrete policy questions! This is surprising to university professors, of course, who are disproportionately "ideologues" in the sense defined by the American political scientist Philip Converse in 1964, namely, those who have well-defined views about policy questions (such that, e.g., they realize that "assistance to the poor" is the same as "welfare," and either favor or oppose both). In Converse's study, these "ideologues" were 3% of the population, while 85% of the voting population "seemed to think about parties and candidates in terms of group interests or the 'nature of the times' or in ways that conveyed 'no shred of policy significance whatever.'" 25 Ironically, academic interest in the United States in the "countermajoritarian" difficulty about judicial review of legislation flourished precisely after the period studied by Professor Converse.
If the spatial theory of voting fares poorly in showing why we should worry about judicial invalidation of democratically enacted laws that violate individual rights, the alternatives fare even worse. The "retroactive theory" of democracy suggests that voters express their judgment of the past performance of those they vote for (or against). That sounds somewhat promising, but in fact voters primarily respond to economic events only in the several months preceding an election 26 and, more importantly, they do not discriminate between events for which elected officials bear some plausible causal responsibility and those events which are basically natural disasters: floods, droughts, even shark attacks (on one occasion) have all influenced voters in past elections. 27 Simply put: if voters are "assistance to the poor" was meant to incorporate all the putatively bad aspects of welfare; similarly, some of the pro "assistance to the poor" folks might have been horrified to learn that what they favored really had the putatively bad aspects of welfare. And, of course, we can imagine the converse. Most importantly, settled law creates settled expectations, and unsettling those have costs that judges must consider. In addition, and as we will return to below, invalidating legislation may affect the since it tests an hypothesis that Achen & Baretles do not defend, namely, that shark attacks influence presidential elections. Achen & Bartels focus on a rather singular case-the multiple shark attacks in 1916 along the New Jersey coast, which attracted national and international attention, and which were utterly unprecedented. Their finding was that in the coastal counties closest to the site of the widely covered attacks, Woodrow Wilson's share of the vote declined markedly by comparison to both his 1912 results and his results in New Jersey counties farther removed.
The relevant part of the Fowler & Harris critique centers on disputing choices by Achen & Bartels about the various comparison classes, and most of those critiques simply show that, if you redraw the comparison classes, the statistical effect on election outcomes of the shark attacks is reduced. The notion of "statistical significance" is epistemically vexed, to put it mildly, but what is missing from the Fowler & Harris critique is any serious attention to the question of "mechanism," that is, the causal processes by which a certain event X brought about an outcome Y. Only with a plausible story about mechanism can we really rule out or in "statistically significant" or "statistically insignificant" correlations (since the P value of greater than 95% is both arbitrary and largely meaningless: see David Colquhoun, "An investigation of the false discovery rate and the misinterpretation of p-values," Royal Society Open Science (Nov. 2014): http://rsos.royalsocietypublishing.org]/content/1/3/140216. The basic problem is that 95% confidence that an effect is not random compared to the null hypothesis does not guarantee it is a real effect [a real cause-effect relationship]. What we need to know in addition is how good our test is for detecting real effects and how many real effects there are in the sample studied (the latter, of course, is hard to estimate). But consider: even with the standard p=.05 standard of statistical significance, if our measures only capture 80% of real effects, and real effects are only 10% of our sample, then one-third of our positive results will turn out to be false positives even with the standard P value: cf. id. at 4-5.) Achen & Bartels are aware of the need for a story about causal mechanisms (cf. their discussion of narratives at __-__), and the shark attack case, while amusing and memorable, is only one small data point for their general hypothesis that voters are terrible at assigning causal responsibility for events that affect them adversely. If the underlying causal mechanism in their explanatory story is that ordinary voters have little understanding of causal responsibility, and yet adverse experiences influence their voting behavior, then they have a story about mechanism which Fowler & Harris lack. 28  This is important to emphasize because many of my ultimate recommendations could apply to other official actors, albeit in slightly different terms. But there are aspects of institutional role and public expectations that caution restraint in all cases, including the judicial, but they vary with institutional role. legitimacy of the courts, and it may also provoke backlash from other branches of government. But what is important to notice is that both these reasons for judicial restraint have nothing to do with the moral significance of how the population voted.
An alternative account of voting behavior recently defended most vigorously by self-described "realists" about democracy, Christopher Achen and Larry Bartels, does nothing to restore the moral importance of democratic voting. On their view, it is "group identity" that best explains voting behavior in democracies. 31 By group identity, they mean the religious, racial, ethnic, occupational and cultural identities 32 that are important to people: their "tribes" for short. On their "group theory of politicsthe powerful tendency of people to form groups, the ensuing constructon of 'us' and 'them,' and the powerful role of emotion rather than reason in directing group activity"--"group attachments are easily generated and profoundly felt" (215). 33 What Achen and Bartels argue is that the group identity largely explains the political party with which individuals identify, and we know that, as Professor Converse put it, "expressions of party identification were much more stable" over time than ideological convictions. 34 In other words, for most voters, their tribal loyalties drive party identification, and party identification then drives voting behavior. 35 In some measure, popular commentary on the political process in America has registered this fact: we all know that African-Americans vote Democratic by overwhelming majorities, and that white men do not. But this also explains another important empirical finding, namely, that voters typically 31 Caveat: their evidence comes largely, but not exclusively, from the U.S. 32 The "cultural" identities are often geographically bounded: cf. Achen & Bartels, __-__. 33 Note that there is no reason to voting in accordance with one's group identity means voting in accordance with one's "interests" or well-being. Achen & Bartels do not claim otherwise. On any plausible view of human interests or well-being, voting in the pattern described by Achen & Bartels would stand in only an accidental relationship to the interests or well-being of the voters. 34  adjust their views to fit the ideology of their party (or its most visible representatives), rather than choosing parties based on how well they fit the voter's own views! 36 That fact helps make sense of the peculiarity of the Trump phenomenon in the United States, since he, in fact, rejected many staples of the "Republican brand" since the Reagan Revolution of the 1980s, though he won an overwhelming percentage of Republican voters in the end (I return to Trump below).
Ideology of a party can matter at the margins, of course: at some point a particular "tribe" registers that a particular party is clearly friendly or hostile to the tribe. African-Americans in the United States supported Republicans for a long time (at least when they could vote) since, after all, it was a Republican President who freed them from chattel slavery. (Unsurprisingly, a group freed from slavery noticed that fact.) But then it was a Democratic President, Lyndon Johnson, who ended de jure segregation in America in the 1960s, and since then African-Americans have voted overwhelmingly for Democrats (though, strikingly, as Achen and Bartels document, Southern Whites were moving to the Republicans before the 1960s 37 ). But ideological considerations only go so far. African-Americans continued to support Democrats, including the Clintons (Bill and Hillary), despite the evidence that they were not very good for the African-American community. 38 Of course, it will be fairly noted that the Clintons were always less hostile to interests of the African-American community than the Republicans, but even so, it hardly explains the failure of African-Americans to rally to Senator Bernie Sanders, a Roosevelt New Dealer, in the 2016 Democratic primaries. Reflexive loyalties do seem to explain it.
Ironically, in the end, the absence of a member of their tribe on the ticket also helps explain the fact that African-American turnout was depressed in the 2016 Presidential election compared to elections when Obama was the candidate 39 : African-Americans, highly motivated to turn out to vote when one of their own was a candidate, were not nearly as enthusiastic about voting for a white woman and a white man in 2016. At the same time, some modest number of white working-class voters (not more than 100,000 in the three crucial states of Pennsylvania, Michigan, and Wisconsin), some of whom had supported Obama in 2012, flipped to Trump in 2016. 40 This indicates that at least some highly salient parts of expressed ideology are relevant to some voters (Achens & Bartels do not deny this, but they are, conceptually, sloppy, which is typical of this literature--but they de facto acknowledge the point with their endless hedging about what "mostly" or "often" explains voting behavior). As the rabble-rousing filmmaker Michael Moore observed, 41 when Trump came to Detroit and threatened auto executives that they had better not move their plants outside the U.S. or else, he did something no other candidate of the two ruling class parties in the U.S. had done in living memory. "Free trade" has become a mantra of the ruling class and its intellectual apologists like Paul Krugman in America, but when a Presidential candidate repudiated it in the Mafioso terms in which Trump specializes, some number of voters took note, perhaps contributing to his victory in states with lots of industrial workingclass voters like Pennsylvania, Michigan, and Ohio. 42 And yet this result still confirms the main point of the realist theory of democracy. The working-class voters who flipped to Trump were largely union members, 43 their loyalty to their class as defined by union membership (an increasingly uncommon loyalty in America) having an effect on their votes. If we are realists about democracy, then we need to 39 http://www.nytimes.com/2016/12/23/upshot/how-the-obama-coalition-crumbled-leaving-an-openingfor-trump.html (Dec. 23, 2016). Note that the decline in African-American turnout was consistent across states that had enacted new voter suppression rules and those that had not. 40 Id.

41
[cite] 42 [cite] 43 [cite] acknowledge, as Achen and Bartels document, that tribal loyalties explain a frightening amount about voting behavior, while moral or ideological commitment generally does not, except in the extreme cases or among a small segment of the electorate. 44 It is worth quoting at some length Achen and Bartels' own summary of their findings, since it makes clear the extent to which the popular, and theoretical, conceptions of democracy bear little relationship to any reality. In the summary that follows, they presuppose that party identification drives most voting behavior, and then focus on the case of the "pure" independent voters 45 in the American context, those who do not lean to either the Democrats or Republicans: While they, too, have group ties and social identities, they are often quite unclear about which groups "belong" in which party. Typically less-informed, they may fail to grasp what is at stake in the choice of one party or another, much less where their overall interests lie. Thus, they are often swept along by the familiarity of an incumbent, the charisma of a fresh challenger, or a sense that it is "time for a change," even when the government did not cause the current unsatisfactory situation and cannot greatly alter it. When the party balance is close…election outcomes turn on how these "swing voters" happen to feel when they go to the polls. As Philip Converse [citation omitted] put it, "Not only is the electorate as a whole quite uninformed, but 44 Achen and Bartels are particularly critical of "highly idealistic models of participatory or deliberative democracy" (301). They write: "These models of democracy emphasize-indeed, they often simply assumerationality, mutual consideration, and the patient exchange of publicly justified reasons for supporting specific policies. Thus, they rest on essentially the same unrealistic expectations about human nature embodied in the folk theory [of democracy]" (301). Although I think Achen and Bartels are right to be dismissive of deliberative democracy theories, their reification of "human nature" can not go unchallenged: none of the evidence supports the claim that the unrealistic character of deliberative democracy has anything to do with essential biological and psychological facts about human beings. The evidence they adduce only supports the view that under the conditions in which most people live, they have neither the time nor the interest to engage in deliberative democracy fantasies. 45 They here mean to exclude the "independents" who in fact, on further questioning, lean to one party or the other. They also mean to exclude quite clearly a category they don't mention, namely, "academic independents," who, as ideologues, have such well-defined views on policy questions that no existing party could be adequate to their excruciatingly "nuanced" conception of what should be done. As Nietzsche would say, "The rule is more important than the exception" [cite].
it is the least informed members within the electorate who seem to hold the critical balance of power, in the sense that alternations in governing party depend disproportionately on shifts in their sentiment." The result is that, from the viewpoint of governmental representativeness and accountability, election outcomes are essentially random choices among the available parties-musical chairs.
Elections that "throw the bums out" typically do not produce genuine policy mandates, not even when they are landslides. They simply put a different elite coalition in charge [emphasis added].
This bloodless change of government is a great deal better than bloody revolution, but it is not deliberate policy change. The parties have policy views and they carry them out when in office, but most voters are not listening, or are simply thinking what their party tells them they should be thinking. This is what an honest view of electoral democracy looks like. 46 The topic unaddressed by Achen and Bartels is the mechanism by which a particular "elite coalition" emerges from the "musical chairs" electoral process. Marx had an account of how elite consensus emerges, but an adequate account of the emergence of a particular "elite coalition" would no doubt turn on the mechanisms by which control of extant political parties can be seized. On this Achen and Bartels are silent, but it is the really crucial research question suggested by their realism about democracy.
But back to our main topic: given realism about democracies, what does it mean to say, as I said earlier, that some democracies are merely "masquerading" as such, concealing in reality plutocracies or incipient authoritarian societies? The answer, I believe, is, in one sense, rather simple. It means that the tribal loyalties operate in such a way that actual decisions about policy are taken either by (a) economically dominant groups in accord with their interests, not the interests of the tribal groups that 46 Achen & Bartels, 312. support them; or (b) authoritarian groups in accord with their interests, and not the interest of those tribal groups that support them. 47 The former was true in the Clinton Administration of the 1990s, and the incipient Trump Administration is already beginning to fit both models, with tax cuts for the rich and evisceration of entitlement programs already on the agenda, as well as demonization of minorities in the name of "national security." How that came to be the case in the United States is the hard question, but beyond the scope of this paper. 48 But the descriptive generalization still stands.
So if electoral democracy is, as the realists portray it, a matter of tribal loyalties translating into party identification, and then electing representatives, who may or may not even represent the views of the tribes that elected them, how should that affect our conception of the roles of judges?

A Realistic View of the Judicial Role
Everywhere judges-consistent with what I dubbed earlier their familiar "Minimal Role"--are expected to apply pre-existing standards, meaning usually "the law," though as judicial oaths of office reveal, they are sometimes expected to do more, and often in terms that are not obviously consistent with the Minimal Role. Whether judges should keep those promises is an open question, but before we get to that, let us look at the promises they are actually asked to make when taking office in a variety of modern legal systems. These promises display some striking ambiguities, and also reveal the extent to which even judicial promises leave much of the judge's role open.
In Serbia, for example, judges take the following (rather typical) oath: "I solemnly swear on my honour that I will perform my duties in compliance with the Constitution and the law, according to the 47 The interests of authoritarian elites often track those of economic elites, but not always. 48 A useful documentation of the fact that it is the case is Martin Gilens & Benjamin I. Page, "Testing Theories of American Politics: Elites, interest Groups, and Average Citizens," Perspectives on Politics__ (2014): __-__. The standard Marxist theory of the state under capitalism-as the committee that organizes the affairs of the ruling class-would have predicted as much. Two peculiar factors in the American case are the huge size of the country, which favors those with resources who can access the mass media, and the high level of religiosity, which capitalism usually destroys more effectively than it has in the U.S. The latter complicates the economic elite story. best of my knowledge and ability and in the service of only truth and justice." 49 The promise of Serbian judges embodies the ambiguity in many such promises: namely, between loyalty to the law, and then loyalty to "truth and justice." But how does one serve "truth and justice" if one is supposed to comply with laws that are unjust or contain actual falsehoods? It is unclear from the Serbian oath, though this oath is hardly atypical in this regard.
The German oath, pursuant to the German Judiciary Act, 50 is similar: "I swear to exercise judicial office in conformity with the Basic Law of the Federal Republic of Germany and with the law, to adjudicate to the best of my knowledge and belief, without distinction of person, and to serve the cause of truth and justice alone-so help me God." German jurists, who perhaps know better than most that God does not exist, are allowed to omit the "so help me God" bit at the end. The German oath also adds the interesting demand, one that is otiose on the Basic Law, namely, adjudication "without distinction of person," but that again perhaps reflects the special history of Germany in the last century.
Like the Serbian oath, it poses the puzzle how fealty to false or unjust law is compatible with serving only "the cause of truth and justice." An Egyptian judge must swear that "by God Almighty that I shall adjudicate between people according to justice and respect the laws," 51 which poses the same puzzle about how to resolve conflicts between justice and fealty to the positive law (unless justice is just a different way of saying "fealty to law," in which case it is otiose). Judges in Israel must pledge "allegiance to the State of Israel and its laws" and swear "to dispense true justice, not to pervert the law or to show favor." 52 This promise arguably implies that if "true justice" is at stake, then ignoring positive law would not be a "perversion" of such law, but a requirement. In South Africa, judges of the Constitutional Court and Supreme Court promise to be "faithful to the Republic of South Africa," as well as "uphold and protect the Constitution and the human rights entrenched in it, and will administer justice to all persons alike without fear, favour or prejudice, in accordance with the Constitution and the law." 53 This oath has in it an interesting additional ambiguity: is it the duty of judges to protect actual human rights or only the "human rights entrenched in" the Constitution? The latter is the more natural reading, but at the same time, it would seem an odd oath if it were rewritten to make that explicit ("I promise to uphold and protect the Constitution and only those human rights recognized by it, but other human rights-well, they get no protection"). 54 Perhaps the most interesting oath in the ones I surveyed is the one from Hungary: 53 S. Afr. Const., sched. 2 §6(2) as amended by Sixth Amendment Act of 2001 §18. 54 South African judges frequently rule on whether or not a particular law or practice is consistent with "human rights," appealing to the principles underlying the Constitution. S v Makwanyane and Another (CCT 3/94), for example, struck down the use of the death penalty as inconsistent with the commitment to human rights enshrined in the South African Constitution. (Thanks to Cora True-Frost for calling this case to my attention.) Indeed, the South African Constitution refers to "human rights" without qualification, even though the requirements of human rights under the South African Constitution are not coextensive with those of international human rights law. S v Makwanyane says this directly, as it notes that international human rights law does not ban the death penalty, but holds that the South African Constitution requires otherwise. As the Court stated: "Our Constitution expresses the right to life in an unqualified form, and prescribes the criteria that have to be met for the limitation of entrenched rights, including the prohibition of legislation that negates the essential content of an entrenched right. In dealing with comparative law, we must bear in mind that we are required to construe the South African Constitution, and not an international instrument or the constitution of some foreign country, and that this has to be done with due regard to our legal system, our history and circumstances, and the structure and language of our own Constitution. We can derive assistance from public international law and foreign case law, but we are in no way bound to follow it." I hereby do solemnly swear that I shall assess the cases entrusted to me in fair proceedings, without bias, conscientiously, solely in accordance with the rules of law; in the fulfilment of my duties, I shall be driven by the desire to uphold justice and equity. 55 In Hungary, judicial fealty is not to the country, its constitution or legal system, but to impartiality in adjudication "in accordance with the rules of law" and to "justice and equity." This would appear to give a Hungarian judge ample grounds for going beyond the positive law-assuming they had the requisite desire "to uphold justice and equality"--something such a desire is likely to require under Viktor Orban, let alone Jobbik.
As the preceding examples make plain, judicial oaths leave it uncertain how judges should resolve conflicts between positive law, truth, and justice. But now, with our realist hats on again, we should remember that the idea that promises actually bind has the same status as the idea of "obligation" generally: certainly many people, especially those who have been reading too much Kant, feel very strongly that promises create binding moral obligations. But the promises judges make, as the examples above illustrate, are various and at times vague about the relationship between judicial obligations and obligations to justice and truth. Should one feel that promissory obligations defeat the general "obligations" to do what is just, especially in cases where the promise is ambiguous about how to resolve the conflict between justice and law? Legal philosophers typically note that judges are also human, 56 and insofar as we feel that humans ought to behave morally, then judges should too. From this it follows, as every legal positivist I know believes, that promises to uphold the law notwithstanding, Minimal-Plus Roles of judges. When we remember what democracy is really like, that judges should sometimes do that should seem less worrisome (though it depends, of course, on the judiciary, a point to which we return below).
The central theoretical constraint in thinking about judicial roles has to do with the nature of law itself, since we would not otherwise know how to interpret the command to apply or follow "the law," so central to the various judicial oaths. There is a superficial if popular view according to which law is created by legislators, and judges are the mere mechanical appliers of the law. Whether the society is authoritarian or democratic, this view is a gross simplification, as all lawyers and scholars know. Even if one agrees that norms are legally valid only in virtue of their source (e.g., legislative or judicial enactment), and even if one agrees that where there is a legal system, there exists Hart's naturalized version of a Kelsenian Grundnorm, namely, a practice of officials of treating certain criteria as obligatory criteria of legal validity, that still does not tell us the role of a judge, even on the assumption that judges should "apply the law." Statements of the law, whether by legislatures or courts or any other lawmaker, must be interpreted so that their import with respect to particular facts is clear. Every serious scholar who has written about legal interpretation over the last century--from the American Karl Llewellyn to the Englishman H.L.A. Hart to the Italian Riccardo Guastini, among many others 57 --has recognized that interpretive norms in a given legal system accord judges latitude, perhaps great latitude, in saying what the law is. Even the most fantastical sophist and moralist of the last century's legal philosophy, Ronald Dworkin, admitted that no one may really know what the right interpretation of the law is (though he, of course, believed there was one and he happened to know what it was). Dworkin's main focus was, ironically, the American constitutional order, in which its highest court is rather clearly--to realists--a super-legislature. 58 What that means, of course, is that judicial decisions interpreting the other sources of positive law can be crucial sources of legally valid norms themselves-and not just for the parties before the court, at least in any legal system that treats judicial precedent as a valid source of law.
So what legal constraints are there on interpretation? Here there is little of a general nature we can say, since norms of legal interpretation vary with legal systems just as Hartian rules of recognition do. Since rules of recognition are social rules in Hart's sense--that is, psycho-social artifacts arising from the actual interpretive practices of the judges that they endorse from an internal point of view-it follows that norms of interpretation can evolve and will vary with the nature of the judiciary and its interpretive practices: for example, civil service systems, where judges are promoted by peer evaluation, create more pressure towards adhering to the prevailing interpretive norms, while systems, like the American, which are political, do not. 59 If there are generalizations available about viable norms of interpretation, I suspect they derive more from sociological facts about how law interacts with society and human interests, and the practical constraints that then emerge on interpretive practices, rather than from any true philosophical thesis about the nature of law. I consider only one example.
In the United States, a small number of judges, and many more legal scholars, have argued that the Constitution should be interpreted according to its "original" meaning. Bear in mind that the U.S.
Constitution was written in 1789, though amended in significant part after our Civil War in the 1860s.
Originalism in America has come in two varieties. One version is that constitutional provisions should be provisions should be interpreted in accordance with the original "public meaning" of the provisions, the latter determined not by historical inquiry into the intentions of framers of the provisions, but by old dictionaries and historical evidence about the circumstances and problems to which the provision was responsive. 60 The first version of originalism famously confronted the problem of identifying meaningful intentions of collective bodies, especially about matters they never could have anticipated. 61 The second version was also afflicted by disputes about how one fixes the initial "public" meaning. 62 But there is a deeper practical problem afflicting both versions: the real problem is not that there is never any fact about the original intent or meaning of a constitutional provision (sometimes there is), but that serious commitment to original intentions or meanings will inevitably undermine constitutionalism altogether, that is, the idea that political power should be constrained by a piece of legislation that no one living participated in enacting and that can not be repealed or overridden in the manner of ordinary legislation. "Originalism" makes for good rhetoric, but usually awful politics, something that all jurists in the U.S. recognize, whether explicitly or not. 63 Brest et al.]. In the U.S., a new originalist literature, inspired by the work of Keith Whittington, tries to dodge this problem by distinguishing "interpretation" (determining the ordinary, communicative meaning of the provision) and "construction" (determining its legal effect). The distinction is dubious, since in every legal jurisdiction I am aware of, there are rules of legal interpretation that make ordinary, communicative meaning irrelevant. (Take the U.S.: is it "cruel and unusual" and thus unconstitutional to execute juveniles? As a matter of ordinary meaning, who knows? But it is clearly unconstitutional now given the rules of legal interpretation in the U.S.) Proponents of the distinction, of course, think this is the way interpretation ought to be pursued, but most of them fail to articulate a normative reason for doing so; an exception is Randy Barnett, who thinks the original Constitution is morally superior from a moral and political point of view. See Barnett, _____. If that were remotely plausible, then this would be a good argument. The Real Problem is that constitutional restrictions on what current polities want to do always make current polities deeply unhappy. Now we must remember that, given realism about democracy, this is not the same as saying that constitutional restrictions are restrictions on democratic majorities.
That is false, since elected officials who might find themselves constrained by a constitution do not represent the moral and political views of current democratic majorities (except sometimes by accident), for the reasons already discussed. That being said, there is a limit to how much political elites and perhaps even voters will tolerate frustration of their objectives by an old piece of legislation, that is, the constitution. 64 And constitutions are, as we all know, just old pieces of legislation with special prohibitions on their repeal by later legislators, often a degree of generality about political aspirations, and a certain positive moral valence attached to them. One need not be an historical materialist to recognize that the world changes continuously as its economic and productive power evolves, and that these changes affect the values, aspirations, and goals of human beings and their political associations. 65 (Historical materialists do have the theoretical advantage of explaining this!) What that means, of course, is that serious originalism, of either variety (intentionalist or public meaning), endangers the idea of constitutionalism, of the attractiveness of being bound by old pieces of legislation.
Constitutionalism as an ideal survives only because incremental adjustments to constitutions by courts are possible in every jurisdiction I am aware of, and so the most offensive aspects of being bound by the will of long-dead people-sometimes rather unappealing long-dead people!--are removed or mitigated. 64 In the age of Trump, one might expect that the limit is reached very quickly! 65 Cf. not only Marx's corpus, but more recently Ian Morris, Forages, Farmers, and Fossil Fuels: How Human Values Evolve (Princeton University Press, 2015) and, a bit longer ago, Marvin Harris, Cows, Pigs, Wars and Witches (Vintage, 1968). While Marx treated the level of development of a society's productive powers as a whole as the crucial explanatory variable, Morris (a classical archeaologist) focuses specifically on the degree of energy extraction from nature, while Harris (an anthropologist) looks at short-term material incentives, such as the ability to have enough food to eat in the year ahead. Morris and Harris are, as it were, "micro"-Marxists, accepting the basic Marxist thesis that how people produce the conditions of their material existence determines their beliefs, but filling in the micro-mechanisms that explain the patterns we find. Harris is better than Morris in this regard, since Harris is more alert to what would motivate individuals to act as they do, while Morris operates largely at the level of striking correlations, without sufficient attention to psychological mechanism. Perhaps constitutionalism has no value, though on the limited evidence we have (after all, only the U.S. and Sweden have constitutions that span four centuries), it seems that constitutions may exercise a human welfare-maximizing constraint on current polities.
Canada, the more European-leaning neighbor of my country, is an interesting case study in this regard. The Canadian Charter, or "constitution," was adopted only in 1982, but the courts moved quickly to repudiate the idea that the intentions of its framers controlled its interpretation. In 1985, the Supreme Court of Canada in Re B.C. Motor Vehicle Act 66 made clear that the intention of the framers did not control how the "principles of fundamental justice" should be interpreted. A bit of realistic context is important. The United States Supreme Court claimed final authority for itself to settle the meaning of the U.S. Constitution some two hundred years ago, 67 a point that could not have been lost on the Canadian High Court. Also not lost on them was the fact that the framers of the 1982 Canadian Charter were quite alive, and would be for a long time. In order not to have its authority usurped, it was important to make clear, and make clear quickly, that having written the Charter, the authors were done. What that Charter meant was now in the hands of the courts and the polity. But this was not simply a "power-grab," but a clear-headed recognition of the fundamental fact of "constitutionalism": namely, that a polity (e.g., its political elites) will not tolerate being (at least somewhat) bound by its predecessors unless there is substantial room for adjustment. This is why America's most notorious originalist, the late Justice Antonin Scalia, attracted almost no judicial adherents for his radical originalism over nearly forty years. Most American lawyers and judges are tacit realists, whatever their public ideology. Let me introduce some categories that will help us conceptualize the options. The categories are admittedly crude, but their crudity may highlight what is at stake. I use the terms "progressive," "reactionary" and "status quo" in descriptive senses in what follows, though the terms often have moral valences. "Progressive" may seem the least descriptive of the three, but I use it here in what we can call the descriptive "Hegelian sense." Hegel, a deeply Christian philosopher, 70 thought that the history of humanity was a matter of recognizing (with the help of Hegel's philosophy, of course) God's intentions for His creation. Divine intentions to one side, what was striking, and which Hegel noticed, is that there has been a general tendency in human history of expanding the domain of "freedom." 71 Think only of democratic voting rights in America: first, it was only propertied white men; then men 68 I think the view I defend, below, is not that far away from the view articulated by Judge Posner originally in Richard Posner, The Problematics of Moral and Legal Theory [Harvard University Pres, 1999], though I eschew his terminology about "pragmatism" which conceals more than it explains. without property, then women, then non-whites, and so on. The history of humanity, for Hegel, is the history of human recognition of God's plan for the expansion of freedom. 72 Progressive means--as a shorthand for its clearly Hegelian background 73 --increasing the range of people who enjoy freedom and/or increasing the kind of freedom enjoyed. (Many "progressives," of course, think the range should extend to non-human animals and even non-animal nature. Other "progressives"-I think I am oneare more skeptical.) The boundaries of such freedom are contestable, but the basic pattern is crystal clear: more and more humans (and maybe even non-humans) are entitled to the same moral concern, and the same moral rights, as others. This is the central ideology of post-Enlightenment modernityshared by Kant and Marx, as well as Rawls and Hayek-and one opposed only by Nazis, fascists, reactionaries, and religious traditionalists of all stripes. 74 One might, of course, insist that the real metric of progress is improvement in "well-being," but that is compatible with the point just made: freedom, in the relevant sense here, stands in a constitutive relationship with "well-being"-you can not be well-off unless free, and vice versa--so for ease of discussion I will speak strictly in the former terms.
The other two terms again have descriptive content, though also normative connotations, depending on who is using the terms. Those who favor the "status quo" are those who favor, obviously, the way things are. This can have a pejorative connotation when the status quo isn't very attractive by the lights of the speaker, but it need not have such a connotation, since sometimes the status quo is morally preferable to the available alternatives, even by the speaker's lights. Those who are "reactionary" would like to return society to an earlier state of affairs, especially in terms of the legal 72 Hegel, rather notoriously, seemed to think this expansion reached its perfection with the Prussian state circa 1820, but that tells us more about the parochialism of philosophy than about freedom. 73 I see no way to understand talk of "progress" except against the Hegelian/Christian background and its normative assumptions. 74 [On "traditionalism," see the 20 th -century Italian fascist philosopher Julius Evola, cited favorably by Trump's advisor Stephen Bannon. I do not wish to exaggerate the importance of Evola. Bannon is plainly an idiot, but an idiot who knows something about Evola and has the ear of a powerful ignoramus is not a happy combination] and political status of citizens: a reactionary opposes the expansion of the freedom the progressive favors.
It is important to note that many debates between so-called "conservatives" and "liberals" are not necessarily debates between reactionaries and progressives in my sense, but rather intramural debates between progressives in the Hegelian sense, that is, those who disagree either about the means or about the extension of the relevant concepts of freedom. This is one reason that in American scholarly discourse it seems perplexing to put libertarians on the right ("the conservatives").
Libertarians contest the kind of freedom that modernity should progress towards, and so are often at odds with actual reactionaries. From the left progressive point of view, libertarians are reactionaries because of their naivete about how so-called "economic freedom" operates in practice, but that, of course, is precisely what is disputed. (To be sure, to the extent Hegel is right about the progressive tendency of modernity, we should hardly be surprised to find actual reactionaries trying to obscure their position by framing it, for rhetorical reasons, in "progressive" terms.) In talking about the roles of judges in democracies, I assume-without argument, since there is no useful argument that can be given-that judges should facilitate progress, rather than protect the status quo unreflectively or move us backwards. If the reader grants me that assumption, then the question about the roles of judges in democracies becomes clearer, though it has to be relativized to the background facts about the judiciary and the democracy. Here are, I believe, the relevant parameters.
A Progressive Judiciary is one that wants to expand freedom, and so exercises its interpretive latitude in service of that end. 75 A Status Quo Judiciary takes its promise to uphold the positive law very seriously, and even where there is interpretive latitude, interprets it in favor of the status quo. A 75 Note that I assume that a progressive judiciary can only promote freedom and equality within the constraints of its institutional role. Other progressives will, of course, pursue progressive ends within the parameters of their roles. (Thanks to Omri Ben-Shahar for pressing me on this issue.) Reactionary Judiciary is anti-progressive, looking to restrict freedom to the elite classes or races or sects who had enjoyed its exclusive prerogatives in the past. Given these rough categorizations, here is how I suggest we think of the "obligations" of judges in contemporary democracies. 76 [Poland more complicated, since it has become a more social democratic state] 77 Of course, there are legal systems in which courts officially lack the power to invalidate legislation-Britain, with its doctrine of parliamentary sovereignty may be the most important case, though there is a clear tradition of parliament revisiting legislation when the courts find an "inconsistenty" with fundamental rights.
Progressive Judiciaries in Progressive Democracies should exercise minimal interpretive latitude with respect to positive law, and should honor the Minimal Role of judges, scrupulously in Legislature-Centered Systems, since this will both promote progressive ends and honor public expectations and sustain the legitimacy of judicial review. Progressive Judiciaries in Middling Democracies should exercise their interpretive latitude in order to move the law in a progressive direction-aggressively in a Judge-Centered System, less aggressively in a Legislature-Centered system. Progressive Judiciaries in Reactionary Democracies should exercise their interpretive latitude aggressively, even in a Legislature-Centered system. The loss of legitimacy for the judiciary in aggressive judicial review is worth the risk given the morally defective character of a Reactionary Democracy. Now the preceding admittedly simplifies how a progressive judiciary should think about its obligations, especially in Middling and Reactionary Democracies. A judge who thinks she has an obligation to exercise interpretive latitude in the service of progress (perhaps even to disregard settled law in the service of progress) must also take into account the ramifications of doing so-for example, the likelihood of provoking a reactionary legislature to revisit a question where the judge has ruled in a different direction. The pursuit of progress must of course be strategic, within the parameters in which an agent of progress can operate.. A progressive judge who disregards the "backlash" effect of her decisions is not a reliable friend of progress.
With that important caveat in mind, we can proceed to the other cases. A Status Quo Judiciary should honor the positive law, and exercise minimal interpretive latitude in a Progressive Demoracy, whether its culture is Legislature-Centered or Judge-Centered. In a Middling Democracy, a Status Quo Judiciary in a Legislature-Centered System, should probably act similarly, since it is unlikely to exercise interpretive latitude for progressive ends, and even if it did so, little good would likely result. If the Middling Democracy is Judge-Centered, then a Status Quo Judiciary should exercise its interpretive freedom in progressive directions, to the extent it is able to do so. In a Reactionary Democracy, the obligation of a Status Quo Judiciary is to try to be Progressive! You can infer from the preceding how my account of obligations will go in the other cases. But let me address one case explicitly, namely, that of Reactionary Judiciaries in Reactionary Democracies.
These judges have only one obligation in my view: to resign their position or, equally unlikely, reform their moral views.
The preceding has focused only on the judicial role vis-à-vis legislation, but a similar set of recommendations could be generated for the judicial role in interpreting prior court decisions or constitutional provisions or the edicts of administrative agencies. 78 A Progressive Judiciary will generally produce progressive precedents, though judiciaries are not univocal and there may be outliers of course.
In that case Progressive Judges should utilize their interpretive latitude in much the same was as they do with respect to non-progressive legislation. Here the risks are likely to be different than in the legislation case: that is, the risk is not provoking legislative backlash, but judicial backlash from less progressive colleagues. Once again, the same kinds of considerations apply as in the legislation case: progressive lower courts need to be mindful of their status quo or reactionary higher courts, and so on.
But the basic idea is the same: progressive judges ought to pursue, strategically, progressive outcomes, with due attention to context and ultimate outcome. 79 Now to some all the preceding will sound too facile: surely, the obligations of judges should not depend on the quality of their moral views, or the moral quality of the legislation they must review? It is true that the question is typically approached without regard to these kinds of distinctions, but such an 78 Thanks to Jennifer Nou for pressing me on this point. 79 A nice example of strategic thinking by progressive judges comes from cases in Ohio in the 1840s and 1850s, in which judges were confronted with the question of whether to return escaped slaves to the slave-holding American South. [considerations of comity first dominated, but by 1850s receded] (Thanks to Todd Henderson for the example.) approach is unrealistic both about the judicial role and about democracies in my view. Judges should generally honor their Minimal Role, but not always. And their Minimal Role affords them considerable latitude, and how that latitude should be utilized has to depend on the quality of the judges, i.e., the quality of their moral and political judgment. Insofar as moral and political judgment is essential to exercise of the judicial role, especially at the highest levels of appellate review, and insofar as democracies differ in their moral and political character (as they obviously do), I see no way of discussing what judges should do without regard to their moral and political judgment and the nature of the democracy in which they serve.
To another kind of skeptic this way of thinking about the judicial role will seem to presuppose that we really know what kinds of decisions protect and expand freedom and well-being. Clearly we do not know with certainty what state of affairs would actually maximize real human freedom and wellbeing. But within the horizons of actual judgment, we often do know, even allowing that there will be contested cases. Decisions that protect the freedom of individual choice, within the limits set by some reasonable understanding of the Millian Harm Principle, and decisions which support social-economic policies that protect people from misfortune and encourage productive activity within capitalist relations of production (the current benchmark, of course) are good decisions. All complications to one side, the preceding basically describes the neoliberal consensus of the past half-century. 80 There may be hard cases, of course, where it is unclear, and so the obligation of judges will be less clear. But the only theory of judicial obligation that avoids such problems is one that would be morally ridiculous (e.g., judges should defer to whatever the legislature or executive says, regardless).
In any case, the genuinely hard cases, from a progressive point of view, are few and far between. The Affordable Care Act honored the Hegelian impulse to maximize freedom by extending 80 The neoliberal consensus may be wholly wrong (as I think it is), in which case interpret "progressive" etc. accordingly. access to healthcare-essential to well-being-and the U.S. Supreme Court correctly upheld it, 81 albeit in a way designed to appease the ideological deformities of the current moment in American political life. If there is an argument against the ACA, it has to be that its means were inadequate to its objective-and to the extent they were, it seems to have been because of the need for political compromise in its passage. Legalizing same-sex marriage 82 obviously expanded freedom, with the arguments against it turning on religious and metaphysical fantasies about the alleged "nature" of marriage 83 that bore no relation to the role of marital union in the capitalist West. American free speech doctrine is a notorious moral catastrophe, 84 shielding Nazi sociopaths and pathological liars, in a way no other democracy does. That the U.S. Supreme Court has gone so wrong in the latter cases, as it has on its campaign financing decisions, 85 is easily explicable by the pathologies of American plutocracy (people with money to burn that amplifies their voice over other citizens think such amplification is central to free speech). 86 But the important point is that these kinds of decisions are pretty obviously wrong from the standpoint of progressive judges. The disagreeements on these issues between progressive judges and others who are not explicitly reactionaries are disputes about the empirical facts.
If the facts were otherwise, progressive judges should yield, of course.
In this regard, I should emphasize that my proposed template for thinking about judicial obligations is easily applicable as well by jurists and theorists who dispute the empirical facts. A libertarian--that is, someone who disputes the extension of "freedom" and especially the facts about state regimes that maximize it--can accept the preceding framework, but maintain that, for example, the ACA should have been invalidated, since it did not really maximize human well-being. On this interpretation, invalidation of the ACA is the "progressive" decision, because of a dispute about the nature of freedom and facts about the legislation in application. This view is wrong, but at least we have now located the dispute in the right place, rather than in nonsense about the act/omission distinction as the U.S. Supreme Court did. So, too, an actual reactionary, one who does not think human freedom and well-being without regard to rank or status matters, can also think the ACA should be invalidated, since maximizing the freedom of all human beings has no value-not because such a decision is progressive, but because a reactionary decision is preferable. Obviously, in both cases, it is dubious that the interpretive latitude afforded judges, even in America, would have licensed that outcome on those reasons, but that might not matter if judges at the highest appellate levels choose outcomes that are legally permissible on the basis of their moral and political views, and the boundaries of permissibility are fluid and wide. 87 In short, I think it is a virtue of my account that it locates the dispute at the right place: not about some putatively "generic" judicial obligation, but at the level of substantive disputes about the facts about how particular policies affect human beings. Counting against my approach, however, is the fiction that judges are morally and politically neutral appliers of the law. I do not want to deny that this fiction is taken very seriously in most societies that honor Democratic Values and that it may turn out to be of crucial importance to human freedom and well-being in some circumstances: realism demands as much. But I am not aware of a legal system that has really tested the relevant hypothesis about the fiction. In that system, the polity would actually understand that pre-existing legal standards can never 87 I am skeptical they are wide enough for the reactionary, but if, following Karl Llewellyn, the boundaries of legitimate interpretation are defined by a sociological fact-current opinion among lawyers and jurists-then we can hardly be confident. be adequate to the array of problems that courts are asked to address and would, with H.L.A. Hart, renounce the desire for it to be otherwise: [W]e should not cherish, even as an ideal, the conception of a rule so detailed that the question whether it applied or not to a particular case was always settled in advanced, and never involved, at the point of actual application, a fresh choice between open alternatives….[T]he reason is that necessity for such choice is thrust upon us because we are men, not gods. It is a feature of the human predicament (and so of the legislative one) that we labour under two connected handicaps whenever we seek to regulate, unambiguously and in advance, some sphere of conduct by means of general standards to be used without further official direction on particular occasions. The first handicap is our relative ignorance of fact: the second is our relative indeterminacy of aim….
[H]uman legislators can have no such knowledge of all the possible combinations of circumstances which the future may bring. This inability to anticipate brings with it a relative indeterminacy of aim. When we are bold enough to frame some general rule of conduct (e.g. a rule that no vehicle may be taken into the park), the language used in this context fixes necessary conditions which anything must satisfy if it is to be within its scope, and certaian clear examples of what is certainly within its scope may be present to our minds. They are the paradigm, clear cases (the motor-car, the bus, the motor-cycle)….We have initially settled the question that peace and quiet in the park is to be maintained at the cost, at any rate, of the exclusion of these things….
[B]ut [w]e have not settled, because we have not anticipated, the question which will be raised by the unenvisaged case when it occurs….
[W]hen [it]…does arise, we confront the issues at stake and can then settle the question by choosing between the competing interests in the way which best satisfies us… 88