Law vs. democracy?
Law vs. democracy?
Abstract and Keywords
Together with Chapter 8, this chapter applies the dispute-resolution account of law to the countermajoritarian difficulty—the apparent tension between democratic government and judicially interpreted constitutional constraints on democracy. It first describes the difficulty and canvasses four popular types of approach to resolving it, including those endorsed by theorists such as Ronald Dworkin, John Hart Ely, Alexander Bickel, and Bruce Ackerman. It then outlines the influential version of the countermajoritarian critique articulated by Jeremy Waldron and presents a dispute-resolution defense of constitutional law. Constitutional law is necessary not only to literally constitute democratic government, but also to prevent predictable types of breakdowns in democratic authority—including agency failure and factions—and to establish extrademocratic means to resolve disputes about whether a breakdown has occurred. These tasks must be performed by constitutional law, the chapter argues, because disputes about democratic malfunctions often cannot be resolved impartially by ordinary democracy itself.
1. The countermajoritarian difficulty
I’ve argued that the Dispute-Resolution (DR) account endorses majoritarian democracy as a way to generate legal rules, and participatory, common-law adjudication as a way to apply those rules. But in the United States, and increasingly in other parts of the world, courts sometimes apply legal rules against majoritarian democracy. Americans expect their government to adhere to a written Constitution—a charter, originally adopted more than 200 years ago and amended periodically since, that establishes the structure and powers of the national government, regulates the federal relationship between the national and various state governments, and lists certain individual “rights” that government must respect. And Americans recognize the authority of the Supreme Court—a judicial body consisting of nonelective judges who can serve for life—to render binding interpretations of the Constitution’s (often highly indeterminate) general rules.
Constitutional law and judicial review are often said to create a “countermajoritarian difficulty”—a clash between the broadly participatory, presumptively majoritarian mechanisms of democracy and the entrenched, often archaic democracy-constraining rules of constitutional law as enforced by an elite cadre of politically unaccountable judges.1 And the DR account, at first blush, seems only to make the countermajoritarian difficulty more difficult. The DR account holds out legal rules as the authoritative product of democratic processes, and adjudication as a sort of second-best means to authoritatively apply legal rules (given the impossibility of using the full-dress democratic process to do so). But constitutional law seems to invert this hierarchy; it seems to give (p.216) extrademocratic legal rules priority over democracy, and semidemocratic adjudication a trump card against democratic politics. Constitutional law claims to resolve very important disputes by apparently nondemocratic means, and thus its authority, on the DR account, is rather saliently open to question.
I will argue in this chapter and the next that constitutional law, including judicial review, can in fact be authoritative on the DR account, and that the American practice of constitutional law in particular (though not exclusively) possesses general authority. The special authority of constitutional law lies in its capacity to mitigate partiality in the resolution of disputes about democracy itself. But that claim to authority would risk being swamped by a tide of antidemocratic entrenchment and elitism, were it not for the ability of constitutional law to respond in various ways to political and popular inputs. Authoritative constitutional law, I’ll argue, ultimately is a balancing act between impartiality and competence, between fairness and participation, and (most fundamentally) between the authority of law and the moral imperative of the moment.
I begin by canvassing the most influential kinds of response that have been offered to the countermajoritarian difficulty. All of them have weaknesses; most of them have significant strengths that will inform the DR response I will offer. The chief advantage of the DR response over the others will be its capacity to unite the strengths of these rivals into a single compelling defense of constitutional law.
2. Strategies for resolving the difficulty
The countermajoritarian difficulty has prompted four basic types of response from defenders of constitutional law and judicial review. There is what I will call a substantive response, which holds that certain fundamental values trump democratic processes and thus require extrademocratic protection from democracy. There is a proceduralist response, which denies that substantive values trump democracy but asserts that extrademocratic procedures are necessary to keep democracy itself in good working order. There is a dualist response, which holds that constitutional law is just another form of democracy, one with special authority as compared to its “ordinary” form. And there is a reductionist response, which asserts that constitutional law is simply part of ordinary democratic politics.
A. The Substantive Strategy
A long tradition in American constitutional theory holds that democracy, when push comes to shove, must yield to certain substantive values, such as equality or justice. On this view, it is all well and good to subject garden-variety disputes to democratic politics; but on fundamental matters of equality or justice, what is democratic must give way to what is right. Constitutional law, this approach (p.217) holds, trumps democracy because (and to the extent that) it protects fundamental substantive values from impairment by the democratic process.
This view, which has affinities with John Locke’s notion of natural rights,2 fairly can be attributed to many if not all of the American Founders3—and in particular to James Madison and Alexander Hamilton, the principal authors of the Federalist Papers, which proved to be the most influential and enduring defense of the Constitution of 1789. The Declaration of Independence, of course, opened its indictment of the British Crown by invoking the “unalienable rights” of men, and the Federalists worried that unfettered democratic government would threaten those rights just as an abusive monarchy had. They therefore (p.218) extolled the proposed Constitution, with its multiple layers and divisions of government (including its independent judiciary), in part for its capacity to protect individual and minority rights from the “passions” and “interests” of the majority.4 The addition of the Bill of Rights in 1791 was a concession to those who believed (among other things) that individual and minority rights would not sufficiently be guarded by the structural features of the original Constitution.5
We can take the American Founders’ understanding of natural rights, as exemplified in the Declaration and in the Federalist Papers, as a paradigmatic example of a substantive approach to the countermajoritarian difficulty. The Declaration, paraphrasing Locke, understood the purpose of government itself as the preservation of certain “unalienable” natural rights (including “Life, Liberty, (p.219) and the pursuit of Happiness”). On this understanding, government vitiates its raison d’être if it violates these natural rights, and thus mechanisms to prevent government from violating rights are justified, even necessary. Democracy itself might be, or consist of, precisely such mechanisms; it might be better than other forms of government (such as Hobbesian absolutism) at the task of protecting natural rights. But to the extent democracy is imperfect at this task, it must be constrained by nondemocratic means: structural devices like those defended in the Federalist Papers (federalism, the separation of powers, an independent judiciary), and perhaps a written Bill of Rights. The constraint of democracy by constitutional law (on this view) is justifiable because constitutional law holds democracy to its purpose, namely the protection of natural rights. The countermajoritarian difficulty is no difficulty at all: The very purpose of democracy requires that the majority be countered where fundamental rights are at stake.
This substantive justification of constitutional law finds contemporary expression, among many other places, in Ronald Dworkin’s account of constitutionalism and judicial review. Dworkin’s “constitutional conception of democracy” holds that “government [is] subject to conditions… of equal status for all citizens.”6 Some of those conditions are essentially procedural: They relate to the terms on which citizens can participate in government. But many of them are substantive in the same sense as the Declaration’s “unalienable rights”: They specify what individual citizens may demand of the community or what the community may not demand of its citizens. According to Dworkin, one “democratic condition” is a condition of equality: “[T]he political process… must express… equal concern for the interests of all its members.”7 Another is a condition of liberty or autonomy, or what Dworkin calls “moral independence”: The community must respect “an individual’s own responsibility to decide for himself what life to live given the resources and opportunities that… collective decisions leave to him.”8 These values are, perhaps not accidentally, redolent of the Declaration’s avowal that “all men are created equal,” endowed with the rights to “Life, Liberty, and the pursuit of Happiness”; they are prepolitical values, values whose advancement is, on Dworkin’s view as in the Declaration’s, the very point of democratic government. Constitutional law, for Dworkin as for many of the Founders, is a codification of these values against encroachment by the form of government that is supposed to protect them.9
(p.220) These substantive approaches attempt to resolve the countermajoritarian difficulty by prioritizing constitutional law—as the entrenched protection of prepolitical rights or other values—over the procedures of majoritarian democracy. And it should be readily apparent that such approaches are anathema on the DR account, for the same reason that outcome-based accounts of authority are problematic more generally: They rely on content-dependent reasons for obedience. People disagree about the existence and content of prepolitical rights, and it is precisely those disagreements that legal authority, including the authority of both democratic legislation and constitutional law, is (on the DR account) supposed to avoid or resolve. The countermajoritarian difficulty is a tension between two alternative methods of resolving disputes about rights; that choice cannot turn on the question of which side of a dispute is correct.
Imagine, for example, a public disagreement over whether people have a right of what Dworkin might call “moral independence”—or what the Founders would call “liberty and the pursuit of happiness”—to engage in consensual sexual relations with others of the same sex. The countermajoritarian difficulty poses the question of which set of procedures—the majoritarian democratic process or judicial interpretation of a written Constitution—is capable of resolving this dispute most acceptably. The answer cannot be that whichever procedures resolve this particular dispute correctly—by getting the normatively right answer—are the procedures that should be used. “[T]he design of a decision-procedure must be independent of the particular disagreement it is supposed to settle,” Jeremy Waldron reminds us; “it is no good if it simply reignites it.”10
The DR account, of course, is built on the plausibility of endorsing procedures on the ground that their results will be generally perceived as reasonably correct over the long run. But that is not what substantive approaches like Dworkin’s assert with respect to constitutional law. Dworkin and other substantive theorists hold that the Constitution is authoritative because (and to the extent that) it protects certain values from the democratic process, that is, because it picks the correct side in particular disputes over values. Substantive approaches (p.221) to the countermajoritarian difficulty thus beg the very questions that create the need for democracy and, arguably, constitutional law in the first place.
B. The Proceduralist Strategy
A more recent tradition, and ultimately a far more promising one, attempts to dissolve the countermajoritarian difficulty by envisioning constitutional law not as a constraint on democracy, but as simply the rules of the democratic game, designed to make democracy more fair and participatory. This procedural approach has roots as far back as McCulloch v. Maryland, the 1819 decision in which the Supreme Court held, among other things, that the state of Maryland could not constitutionally impose a tax on the federally chartered Bank of the United States.11 Writing for the Court, Chief Justice John Marshall noted that a state tax on the federal Bank would amount to a tax on citizens of other states who could not vote in Maryland—a literal example of “taxation without representation.” Marshall thus used constitutional law to enforce the democratic principle of fair participation in binding decisions. So understood, the Constitution was not so much a constraint on the democratic process in the state of Maryland as a guarantor of democratic safeguards in the politics of the nation as a whole.
The watershed moment for procedural approaches to constitutional law, however, did not arrive until 120 years after McCulloch, with the famous “Footnote Four” of Justice Harlan Fiske Stone’s 1938 opinion for the Court in United States v. Carolene Products Co.12 The Carolene Products decision continued the New Deal Court’s recently discovered policy of deference to the political branches on matters of economic regulation, a policy that reversed roughly three decades of occasionally aggressive judicial use of the Constitution’s Commerce and Due Process Clauses to invalidate Progressive-era legislation. As the Court committed itself to staying out of economic affairs, however, it simultaneously began to sense the need to intervene more forcefully in other controversial areas, particularly on matters involving political participation and social equality.13
(p.222) In Carolene, the Court attempted to accommodate these competing impulses: judicial deference to politics in economic matters on the one hand, aggressive judicial enforcement of political and egalitarian rights on the other. The Carolene Court, upholding a federal prohibition on so-called “filled milk,” declared that it would not invalidate “regulatory legislation affecting ordinary commercial transactions… unless… it is of such character as to preclude the assumption that it rests upon some rational basis.”14 But it qualified this “rational basis” test in a lengthy footnote—Footnote Four—suggesting that “more exacting judicial scrutiny” might be appropriate in two types of circumstance: where “legislation… restricts those political processes which can ordinarily be expected to bring about repeal of undesirable legislation,” and where “prejudice against discrete and insular minorities… tends seriously to curtail the operation of those political processes ordinarily to be relied upon to protect minorities.”15
Footnote Four implies a procedural account of constitutional rights, by which judicial enforcement of rights prevents or cures defects in the democratic process itself rather than simply trumping democratic outcomes with results the Court prefers. One way the democratic process might go wrong is if those (p.223) holding temporary power—a momentary political majority, or elected officials currently in office—rig the system to entrench that power and prevent others from legitimately taking or sharing it. A democratically elected government, for example, might decide to prosecute dissenters for speaking against government policies, thus depriving voters of information and arguments critical of those policies and increasing the chance that the government will retain power at the next election. Constitutional law targeted at preventing these abuses—a judicially enforced guarantee of freedom of political speech, for instance—would ensure that democratic mechanisms work properly, allowing meaningful citizen participation by maintaining the flow of information and opinion upon which participation can be based. It would be strange to think of constitutional law, used this way, as contradicting or overriding democratic decisions; constitutional law, rather, would be a means of making sure that political decisions truly are democratic.
Another way the democratic process might go wrong is if irrational prejudice or bias leads to the formation of artificial political minority groups. As we saw in Chapter 4, democratic majorities and minorities typically are in flux, in two related senses. First, political majorities and minorities coalesce around particular issues or sets of issues, and thus a minority with respect to one issue may be a majority with respect to other issues. Second, even with respect to particular issues, political majorities and minorities often change over time, as dominant social attitudes about those issues evolve.
Suppose, however, that a certain subset of the population shares some nonpolitical characteristic—race, say, or religion or gender—that triggers irrational negative bias on the part of others; and suppose members of that group also (thanks perhaps to the long-term effects of that bias) tend to share certain political interests and viewpoints. The existence of the irrational bias might prevent those political interests and viewpoints from having a fair hearing in the political process. The interest of African Americans in more funding for inner-city schools, for example, might be discounted by others precisely because of its association with African Americans, a group subject to pervasive racial prejudice. A minority religious group’s argument for an exemption for its practices from generally applicable laws—say, an exception to the drug laws for the sacramental use of peyote by Native Americans—might fall on deaf ears thanks to the majority’s religious bias against its proponents.
But constitutional provisions aimed at protecting “discrete and insular minorities” from the effects of majority animosity—a guarantee that the laws will provide “equal protection,” for example, or a prohibition against laws targeting particular religions or religious practices—might force democracy back on its track, preventing the formation of discrete and insular minorities and limiting their significance if they form. Again, constitutional law viewed this way seems less like a trump on majoritarian democracy and more like a corrective against democratic pathologies.
(p.224) This Footnote Four approach—later expanded and popularized by constitutional theorist John Hart Ely16—seeks to evade the countermajoritarian difficulty, not by prioritizing rights over democracy as the substantive approach does, but by conceiving of constitutional rights as elements of a properly functioning democratic system. The approach sees judges as mechanics, armed with the tools of constitutional doctrine and charged with keeping the machinery of democracy humming. An invalidation of a statute on constitutional grounds, this approach holds, is not so much a rebuke of the machine itself as a jolt designed to nudge the machine back into working order, a good whack with the judicial wrench.
The Footnote Four proceduralist strategy has much in common with the DR approach I will outline below. But for our purposes it has two important limitations, both of which flow from a failure to fully appreciate the connection between legal authority and the resolution of disputes. First, the proceduralist account of democratic malfunctions envisions a rather narrow scope for constitutional law. Footnote Four, and Ely’s subsequent elucidation of it, focus somewhat myopically on the danger that irrational bias against “discrete and insular minorities” will taint the democratic process. As I will explain, however, democracy can malfunction whenever a majority (or even a well-organized minority) becomes partial with respect to a particular issue—whenever the majority is “united and actuated by some common impulse of passion, or of interest,” in Madison’s famous phrase.17 As Madison’s description suggests (and as a reading of Federalist No. 10 makes quite clear), a majority united by self-interest can be every bit as partial as a majority united by irrational bias. By spotlighting the role of bias in politics, however, Footnote Four and Ely emphasize the risk of majority “passion” and ignore the danger of majority “interest.” Thus they underestimate the value of constitutional law in preventing and remediating self-interested abuses of power by the majority—a function, not coincidentally, that has often been performed in the American system by the Due Process Clause, whose scope the Carolene Products Court was attempting to curtail.
Second, and more fundamentally, Footnote Four proceduralism is insufficiently attentive to questions of process. As espoused by Ely and the Carolene Products Court, proceduralism is, somewhat counterintuitively, concerned primarily with the substance, or content, of constitutional law—with the kinds of disputes that should be removed from ordinary democratic politics. But Footnote Four proceduralism pays precious little attention to the questions of how those disputes, once removed, should then be resolved, and of why constitutional means of resolving them are superior to ordinary democratic ones.
(p.225) Consider two manifestations of this inattentiveness. The first and most glaring is proceduralism’s lack of an explanation for the constitutional text itself, and for the process that generated the text. Proceduralism holds that courts should closely scrutinize political decisions that tend to entrench those in power or that reflect, or promote, bias against discrete and insular minorities. But why should courts be bound by textual provisions, particularly centuries-old textual provisions, in doing so? The project of tethering democracy-reinforcing judicial decisions to the language of an eighteenth- or nineteenth-century clause seems at best gratuitous on a proceduralist theory, and at worst affirmatively obstructionist. The fact that a collection of eighteenth-century politicians chose a particular verbal formula—“freedom of speech, or of the press”—hardly seems relevant to the question of how best to protect political participation and fairness under contemporary conditions. The constitutional text, and indeed the efforts of the constitutional Framers generally, appear to be little more than annoying distractions to a Footnote Four proceduralist (which is why Ely dismisses the significance of the text with the epithet “clause-bound interpretivism”18).
A second aspect of the paradoxical proceduralist inattention to process is a decidedly shallow appreciation of courts themselves and of the constitutional tasks they are asked to perform. In defending constitutional courts as referees of democracy, Ely contrasts the relative political independence of judges with the interest-driven processes of ordinary politics;19 but he pays almost no mind to the procedures of constitutional adjudication themselves. As a result, he tends to lose sight of the fact that constitutional courts are not simply “calling balls and strikes,” to use a recent and inapt metaphor,20 but are in fact attempting to resolve difficult and contentious disputes. For example, since the mid-1970s the Supreme Court has struggled with the First Amendment validity of laws regulating spending in connection with political campaigns.21 Opponents of regulation generally argue that it stifles the communication of political information and opinion, thus frustrating a core purpose of the Free Speech Clause; proponents (p.226) generally argue that it mitigates distortions in the marketplace of ideas by which wealthier speakers are disproportionately able to communicate their views. This is a genuine and important disagreement, and its difficulty is not diminished by the fact that it involves the procedures of democracy rather than some purely “substantive” concern. But Ely does not explain why the process of adjudication is a better way to resolve these disagreements than the process of politics; instead he naively offers proceduralism as a way to prevent judges from imposing their own values on the rest of us.22
The DR approach I will offer here draws heavily on the Footnote Four/Ely strategy, in particular by justifying constitutional law as a response to predictable kinds of defects in the democratic process. But the DR approach pays closer attention than the proceduralist strategy to the actual procedures of both constitutional law and ordinary democracy. It explains typical democratic defects as failures of impartiality, and it describes in some detail how a written constitution and judicial review can avoid or remediate those failures. It thus offers an account of constitutional process as well as constitutional doctrine.
C. Dualist Strategies
A third influential family of approaches to the countermajoritarian difficulty holds that constitutional law is, in essence, simply another form of democracy. According to these dualist strategies, constitutional law, while not undemocratic, trumps ordinary democratic politics—either because constitutional law is better than ordinary democracy at deciding issues of “principle” or “values,” or simply because its procedures are more democratically legitimate than everyday democracy.
(i) Bickelian Dualism
Alexander Bickel’s defense of judicial review was a progenitor of one type of dualist approach. Bickel distinguished between “expediency” on the one hand and “principle” on the other—between “immediate results” and “enduring general values.”23 Citizens of a democracy, Bickel suggested, are likely to value both halves of this dichotomy; they are likely to want favorable immediate results but also to care about enduring values, and thus they are likely to want to act, as a collective governing body, in ways that properly balance the former with the latter. But the procedures of the ordinary “political marketplace”—elections, lawmaking by representative legislatures, and enforcement by an elected executive, where “the pressure for immediate results is strong enough and emotions ride high”—tend to ignore enduring values in the pursuit of expediency.24 A different set of procedures, therefore, might be necessary in (p.227) order to supply the “principle” side of the equation. And (Bickel suggested), constitutional review by judges, who “have… the leisure, the training, and the insulation to follow the ways of the scholar in pursuing the ends of government,” and who “are concerned with the flesh and blood of an actual case,” can be defended as just such a necessary set of procedures for articulating enduring values.25
The recent work of Christopher Eisgruber offers a more fully developed Bickelian account of constitutionalism and judicial review.26 Like Bickel, Eisgruber distinguishes between two kinds of substantive democratic goals, which Eisgruber calls “values” (“views about how [people] ought to behave,” about matters of “moral duty”) and “interests” (“views about what [people] want or desire”).27 Values and interests sometimes conflict with each other, and “[w]hen such conflicts occur, people commonly believe they ought to subordinate their interests to their values.”28
Constitutionalism and judicial review, Eisgruber argues, are institutional strategies for subordinating interests to values in the realm of democratic politics. Ordinary majoritarian politics is likely to do a poor job of this, for a number of related reasons. It might not distinguish issues of values or principle from issues of mere self-interest, and when it does it might tend to prioritize the latter over the former. And in addressing matters of value, ordinary majoritarian politics is likely to “merely serve the majority” rather than “respond to the… opinions of all the people.”29
By contrast (Eisgruber contends), a relatively obdurate constitution identifies certain topics of dispute that are not subject to ordinary politics. In doing so, it tags those topics as matters of value, not merely of interest, thus increasing the likelihood that they will be resolved through serious public deliberation; and it makes it more difficult for temporary political majorities to consolidate their power (and entrench their own views on matters of value) against the minority. Trusting the development of constitutional law to the process of judicial review, moreover, enhances the probability that issues of value will be treated as such, resolved according to “the right kind of reasons—reasons of moral principle rather than self-interest”—but also in a way that credibly can be claimed to “speak on behalf of the people.”30 Judges (in the American federal system) are, thanks to their lifetime tenure, relatively “disinterested” with respect (p.228) to politics and thus free to decide based on genuine moral convictions rather than estimations of short-term political gain. Their authority to resolve issues alone or in small groups, and their obligation to defend those resolutions publicly, encourages judges to take moral responsibility for their decisions. And yet the political process by which they are appointed ensures that judicial decisions, over time, tend to hew closely enough to mainstream public opinion that the judges can claim to represent the people in their decisionmaking.31
The Bickel/Eisgruber defense of constitutional law shares with substantive accounts (like Dworkin’s) a belief in the priority of “principle” or “values” over mere “expediency” or “self-interest” and a distrust of the capacity of regular majoritarian politics to enforce that priority. Where Bickelian dualism differs from substantive approaches, however, is in its refusal to endorse any particular set of sanctified, prepolitical substantive values. The point of constitutional law on the Bickelian account is not to entrench any particular collection of values against democracy, but rather to identify, prioritize, and enforce whatever set of values a democratic community chooses to recognize. Constitutionalism and judicial review are, for Bickel and Eisgruber, simply means of holding the democratic “people” to its (democratic) commitment to values, even (indeed especially) where momentary exigencies make it difficult to fulfill that commitment. The democratic “people” are Ulysses; constitutional law is Ulysses tied to the mast.
(ii) Problems with Bickelian Dualism
Like Footnote Four proceduralism, Bickelian dualism sees constitutional law as a response to certain shortcomings of ordinary democracy. In fact Bickel and Eisgruber improve upon proceduralism in one key respect: They offer a procedural account of constitutional law itself, one that attempts to explain how constitutional courts can decide some issues better than ordinary democracy can. The device Bickelian dualism employs to that end, however—the supposed distinction between matters of “value” or “principle” on the one hand and matters of “interest” or “expediency” on the other—turns out to be inadequate to the task.
The first difficulty with the distinction is that it seems likely to be substantively controversial. As Eisgruber acknowledges, a distinction in kind between “values” and “interests” does not exactly operate at the level of undeniable moral truism—a reasonable person might well reject its validity.32 And someone who rejects the distinction will, by virtue of her skepticism, lack any good reason to accept the authority of constitutional law to resolve (what she believes to be) nonexistent issues of “principle.” The Bickelian account fails to justify the exercise of constitutional authority over such a person, just as substantive accounts (p.229) fail to justify constitutional authority over someone who disagrees with the particular constitutional values being enforced. Bickelian dualism is an improvement over substantive strategies in that it abstracts away from disputes about particular results; but its “values/interests” dichotomy might not be abstract enough to avoid collapsing into content-dependence and thus failing as an account of constitutional authority.
A second, related problem is that the “values/interests” distinction, even if it is widely accepted in theory, is so vague as to invite frequent controversy in practice. Consider, for example, the problem of how to balance civil liberties with national security during a struggle against international terrorism. Which side of this balance represents “values” or “principle” and which side represents “interests” or “expediency”? The “expedient” imperative to keep people safe might also be understood as an enduring command of “principle”; the “value” of protecting civil liberties might be dismissed as simply a matter of convenience or of narrow “interests.” One person’s “principle” is another’s “expediency,” and there is no readily apparent algorithm for distinguishing the former from the latter. Of course, such boundary-line issues probably will haunt any strategy that attempts to demarcate different spheres of authority—the DR account will be no exception; but the “values/interests” or “principle/expediency” distinction seems indeterminate to a striking degree.
A third problem, one that flows to some extent from the first two, is the shallowness of the narrative of constitutional process offered by Bickelian dualism. True, that narrative is an improvement on Footnote Four proceduralism, which justifies constitutional judicial review principally on the ground that it is not ordinary democratic politics. Bickelian dualism at least describes certain features of constitutional courts—the relative political independence of judges, the professional norms associated with legal argument—that might generate more accuracy than politics can with respect to some issues. But the Bickel/Eisgruber model of constitutional procedure, like the values/interests distinction to which it responds, remains notably undertheorized. The question for constitutional procedure is not simply whether it performs better than ordinary politics according to certain metrics, but whether its existence improves overall accuracy as compared to a purely democratic system. And Bickelian dualism gives us precious little sense of what the answer to this query might be, or even of how to go about answering it. Are the gains in “principle” from constitutional judicial review likely to outweigh the costs in terms of “expediency”? Bickel and Eisgruber cannot say—in part because their treatment of constitutional procedure itself is relatively superficial, and in part because (again) the standards of “principle” and “expediency” are so vague as to defy measurement.
Finally, there is the difficulty that Bickelian dualism, like Footnote Four proceduralism, seems to give short shrift to the constitutional text and the process of its creation. Bickel and Eisgruber focus on the supposed procedural advantages of courts in identifying and enforcing values or principles, but they fail to (p.230) explain why courts should feel themselves bound by the constitutional text in the process of doing so. Why should the centuries-old text be a better source of enduring values than the judges themselves, or for that matter the political process upon which the judges are intruding? As with Ely, the text for Bickel and Eisgruber seems little more than an annoyance.
The DR account I will offer here accepts the Bickelian notion that there is something procedurally special about constitutional judicial review that justifies its existence. It improves on the Bickelian strategy in two ways, however. First, it rejects the problematic “values/interests” or “principle/expediency” distinction as the basis for constitutional authority. Constitutional authority, I’ll argue, is not issue-dependent in this sense; it does not depend on the normative status or weight of the matter being decided. It is, rather, dispute-dependent; it exists as a relatively impartial way to resolve disputes that otherwise would be tainted by democratic self-judging. Second, the DR account explains in sufficient detail what exactly is advantageous about constitutional process (including both judicial review and the enactment of an original text)—how it is that these procedures can avoid or remediate democracy’s self-judging problems.
(iii) Ackerman’s “Dualist Democracy”
Bickelian dualism, however, is not the only type of dualist strategy. There is another brand of dualism that avoids some of the Bickelian shortcomings: the “dualist democracy” offered by Bruce Ackerman.33 Like Bickel and Eisgruber, Ackerman distinguishes between different institutional expressions of democracy: an everyday “normal politics” and a separate track, which Ackerman calls “higher lawmaking.” Higher lawmaking occurs during periodic “constitutional moments”—really constitutional phases—in which proponents “convince an extraordinary number of their fellow citizens” to support an initiative, opponents have “a fair opportunity to organize” against the initiative, and “a majority of… Americans [become convinced] to support [the] initiative as its merits are discussed, time and again, in… deliberative fora.”34 Often—but not always—these episodes of higher lawmaking result in one or more foundational constitutional amendments, as during the Framing itself and in the immediate aftermath of the Civil War; sometimes they amend the Constitution functionally (by means of “transformational judicial opinions”) if not formally, as during the New Deal.
Ackerman asserts that the special characteristics of higher lawmaking give it priority over the results of “normal politics,” which is simply the run-of-the-mill process of representative government in which elected officials make policy and (p.231) periodically stand to account by means of regular elections. And Ackerman defends judicial review as necessary to preserve the superauthoritative results of higher lawmaking against backsliding during periods of normal politics. Ackerman’s dualism differs from the Bickelian version, however, in that Ackerman’s hierarchy of higher lawmaking over normal politics is not (at least not openly) driven by the premise that there are some issues (matters of “values” or “principle”) that the former can decide better than the latter can. Ackerman prioritizes the “rare occasions” of higher lawmaking, not because they are likely to generate more-accurate resolutions of certain types of questions, but because they (and they alone) “earn the special recognition accorded the outcomes of mobilized deliberation made in the name of We the People.”35 It is the procedural pedigree of the results of higher lawmaking that confers its special status, not the subject matter to which it applies. If this understanding is right, then Ackerman’s brand of dualism can avoid the difficulties of controversy and indeterminacy that plague the “values/interests” or “principle/expediency” distinction. It also, not incidentally, can explain the role of text in constitutional decisionmaking in a way that neither Bickelian dualism nor Footnote Four proceduralism can.
(iv) Problems with Ackerman’s Dualism
In fact, however, Ackerman’s dualism is ambiguous in a number of respects that are crucial to its consistency with a DR account of legal authority. At the core of the difficulty is the question whether Ackerman’s account is entirely process-based and noninstrumental or rather outcome-based and instrumental, at least in part.
Dualist democracy as an instrumental theory. We can get at this difficulty by probing more deeply into Ackerman’s reasons for prioritizing higher lawmaking over normal politics. One plausible type of reason (the most plausible, in fact) is instrumental: Ackerman might believe that higher lawmaking produces (or tends to produce) better decisions than normal politics can. If this premise applies only to certain kinds of decisions—if, for example, Ackerman believes that higher lawmaking is better than normal politics at deciding issues of principle—then Ackerman’s account is really a version of Bickelian dualism, and it is subject to most of the shortcomings of Bickel’s approach canvassed above. But let us suppose that Ackerman’s account, unlike Bickel’s and Eisgruber’s, does not in fact rely on a distinction between principle and expediency or some similar dichotomy between types of normative issues. (Nothing in Ackerman’s articulation of his account suggests reliance on such a distinction.)
If Ackerman’s dualism is not based on a dichotomy between different types of normative decisions, it might nonetheless be a more general kind of instrumental account: Ackerman might believe that higher lawmaking produces (or tends to produce) better decisions across the board, on most or all types of (p.232) questions, than normal politics does. We should recognize by now that a case-specific outcome-based account along these lines will run smack into the Encroachment Flaw: If Ackerman’s assertion is that normal politics should defer to higher lawmaking in a given case because, in that case, higher lawmaking is more likely to have generated the correct result, it is a variety of Epistemic-Guidance (EG) account of authority and, as such, fails to provide a strong reason for normal politics to defer in cases of disagreement. But of course Ackerman’s theory might instead be systemically instrumental in the same sense as the DR account of authority—it might assert, not that higher lawmaking is more likely to get it right in every case, but that it is entitled to deference on the ground that it is reasonably likely to get it right over the run of cases.
Understood this way, however—as a claim about the general systemic accuracy of higher lawmaking versus normal politics—Ackermanian dualism becomes deeply unpersuasive. It is true that one tempting reason to be skeptical of such a claim turns out to be unfounded. We might initially think that Ackerman’s broad instrumental claim, if true, proves too much: If higher lawmaking (i.e., constitutional law) is a better way to make most or all decisions than ordinary democratic politics, why should we bother with ordinary politics at all? Why not simply subject every decision, or at least every reasonably important one, to the constitutional process?
There is a two-part response to this objection (although the first part suggests a more trenchant objection, which I explore below). First, deciding every important question by means of higher lawmaking—making every legislative moment a “constitutional moment”—would be prohibitively cumbersome and time-consuming. An Ackermanian “constitutional moment” features a complex series of proposals, mobilizations, deliberations, reactions, and (ultimately) canonical legal codifications, the whole of which takes considerable time, effort, and collective expense to complete.36 Relying on that process for the bulk of a society’s lawmaking would be impractical in the extreme; efficiency imperatives therefore impose natural limits on the scope of constitutional law.
As I discuss below, this efficiency rejoinder opens the door to another worry about Ackermanian dualism, namely that its potential instrumental advantages are confined to the work product of constitutional moments themselves—a potentially crippling limitation, given the inevitably substantial interpretive role of subsequent courts. But there is a related rejoinder as well: Higher lawmaking as Ackerman envisions it would, if used as the primary or only source of legislation within a system, give disproportionate power to political minorities. It is one thing to claim, as Ackerman might be understood as claiming, that the results of supermajoritarian consensus, where they exist, deserve priority over the results of simple majoritarian politics. It is something else entirely to claim (p.233) that every political decision, or at least every important one, should be subjected to the kind of supermajoritarian process Ackerman attributes to constitutional moments. A supermajority requirement gives a minority, perhaps a relatively small minority, the power to block legislation; the exercise of this power (or the threat of its exercise, for the purpose of bending the legislation in the minority’s favor) operates not as democracy, but as a sort of oligarchy. Systematic supermajoritarianism thus undercuts the evenhanded participatoriness that gives democracy its claim to reasonable accuracy.
The Ackermanian response to the “baby with the bathwater” objection—that is, to the worry that the priority of higher lawmaking implies the worthlessness of ordinary democratic politics—therefore is, in essence, that the perfect is the enemy of the good. Where higher lawmaking actually succeeds—where its supermajority requirement and other hurdles are overcome—then its progeny has a strong claim to accuracy and thus to priority over the results of normal politics. But higher lawmaking is entitled to no such priority where it fails; its failure does not preclude ordinary politics from filling the gap, any more than the absence of ordinary legislation precludes unregulated private activity. To rely on higher lawmaking—that is, on constitutional law—for the bulk of a society’s legislative output would be, in essence, to preclude legislation through ordinary politics in the absence of constitutional law on an issue, and that, in addition to being prohibitively inefficient, would give way too much power to political minorities.
If we understand Ackerman’s dualism in broad instrumental terms, then, we cannot reject it on the ground that it proves too much. But we may be able to reject it on the ground that it proves too little. There are two closely related arguments here, the first of which hearkens back to the discussion of indeterminacy in Chapter 5.
Recall from that discussion that legal rules inevitably are indeterminate to some extent. The frequency and degree of indeterminacy in the case of constitutional rules are likely to be especially great; constitutions typically are intended to last for a long time, and so their Framers tend to draft their provisions in relatively broad, open-ended language that can be applied in a myriad of unforeseen circumstances. This certainly is true of many provisions in the U.S. Constitution; consider the vague guarantees of “freedom of speech” in the First Amendment, of “due process of law” in the Fifth and Fourteenth Amendments, and of “equal protection of the laws” in the Fourteenth Amendment. And the relative textual indeterminacy of constitutional provisions makes more work for courts and other ex post interpreters, who must specify a provision’s meaning as applied to the facts of a great many cases unforeseen by the constitutional Framers.
If Ackerman’s dualist theory is in fact an instrumental one, however, then his case for the greater accuracy of higher lawmaking rests on the character, not of the process of judicial interpretation of constitutional provisions, but of the framing of those provisions themselves—or, in cases where a “constitutional moment” generates de facto but not formal constitutional amendment (such as (p.234) the New Deal), of the struggles leading to “transformative judicial opinions” that innovatively interpret existing provisions.37 Ackerman’s is a theory about the special authority of those “rare occasions” involving “mobilized deliberation made in the name of We the People,”38 occasions (again, really phases or intervals) featuring a complex series of procedural steps. It is plausible to think that the extraordinarily deliberative, dynamic, supermajoritarian character of these “constitutional moments” gives them special accuracy in deciding important political questions. But it is much less plausible to imagine that garden-variety judicial review—decision by a small cadre of unelected judges (albeit following a process that is, as I’ll explain in Chapter 8, meaningfully and broadly participatory in a number of respects)—possesses the same kind or degree of special accuracy.
And yet the endemic indeterminacy of written constitutional provisions means that a large percentage of actual on-the-ground constitutional decisionmaking—including virtually the entire business of judging the constitutional validity of actual statutes or other government actions—will be performed, not by the full-dress process of Ackerman’s higher lawmaking, but by the considerably less-imposing means of everyday judicial review. The premise that normal politics should yield to higher lawmaking thus does not necessarily, or even very plausibly, lead to the conclusion that normal politics should give way to the specific, highly contingent interpretations of extremely indeterminate higher law that are rendered by unelected judges.
The first reason to be skeptical of Ackermanian dualism as a claim about relative systemic accuracy, then, is that the procedural conditions that would make such a claim persuasive will not actually exist, or will exist only in traces, in most instances of constitutional decisionmaking. The second reason is closely related to the first: It is simply that higher lawmaking’s claim to special accuracy seems likely to fade with the passage of time.
Let us assume away the first reason as best we can by considering a relatively determinate constitutional provision—say, the requirement in Article II of the U.S. Constitution that a person “have attained to the Age of thirty five Years” in order to be eligible for the office of President. Suppose a majority of Americans, in the 2012 election cycle, favors electing a thirty-three-year-old as President. Does the relative accuracy of the process of framing and ratifying Article II in the late 1780s give the minimum-age requirement a plausible claim of authority over the popular majority in 2012?
The argument seems problematic at best. Even if we assume that the obvious defects in the 1780s ratification process—its total exclusion of women and (p.235) near-total exclusion of nonwhites and nonproperty owners—were not enough to overcome the benefits of the dynamic, deliberative, supermajoritarian character of that process, the fact remains that ratification occurred more than 220 years ago. Life expectancies are much longer now; the regular duties of the President have been refined to a much greater degree of detail; people’s expectations of leadership may well have altered a good deal over the past two centuries. In short, a simple majority in 2012 might seem (to that majority) every bit as likely as, perhaps more likely than, a supermajority in 1789 to render a good decision regarding the minimum age for a President. The accuracy-generating advantages of the higher lawmaking process, plausible as they may be if we ignore the factor of time, become less and less plausible as time marches on.
So: If we give Ackermanian dualism its most plausible understanding—as an instrumental theory holding that constitutional law (“higher lawmaking”) is generally more accurate than normal politics—it loses traction where the rubber of high theory meets the road of practical necessity. No society could (or would want to) rely on constitutional moments for most or all of its lawmaking; that would be impossibly cumbersome and would in any event give minorities too much power. But if constitutional moments are relatively rare, the legal rules they generate—in the form of constitutions, constitutional amendments, or “transformative” judicial decisions—will of necessity be quite indeterminate. High indeterminacy in constitutional rules will require that courts be given substantial authority to interpret and apply those rules in particular cases; and this will mean that the bulk of constitutional law will in fact be attributable to the everyday work of elite, relatively unaccountable judges rather than to supermajoritarian constitutional moments. As a constitutional moment fades more distantly into the past, moreover, its own authority as a specially accurate procedure is likely to fade along with it. The upshot is that the instrumental reading of dualist democracy becomes, in any real democratic system, considerably less than compelling.
Dualist democracy as a noninstrumental theory. But what if we conceive of Ackerman’s brand of dualism in noninstrumental rather than instrumental terms? What if we understand Ackerman as claiming, not that higher lawmaking is a better (more accurate) way to generate important political decisions, but simply that it is more legitimate than normal politics without regard to the quality of the decisions it generates? This would be a purely process-based theory of constitutional authority.
We got a sense in Chapters 2 and 3 of the nature of the core challenge facing any purely process-based account of legal authority. Such an account must plausibly explain what it is, aside from the quality of the results a process produces or tends to produce, that creates a strong moral reason to defer to that process. As we saw in those chapters, the most plausible basis of such an account appears to be consent; but Ackerman’s dualist theory of constitutional democracy cannot reasonably be interpreted as turning on consent in any of its forms. (p.236) A noninstrumental reading of Ackerman thus faces a substantial handicap from the very beginning.
One implication of that handicap is the difficulty in imagining what in particular it might be about the processes of higher lawmaking (aside from their results) that would confer authority upon them. But we need not resolve this puzzle in order to expose the shortcomings of dualist democracy as a noninstrumental theory. Those shortcomings mirror the weaknesses of the instrumental version of the theory. First, whatever are those features of higher lawmaking that give it a special claim to procedural legitimacy, surely they are diluted when higher law is strained through the medium of judicial review. Ackerman describes higher lawmaking as lawmaking “by the American people,” in contrast to normal politics, which is lawmaking “by their government.”39 While the processes of a constitutional moment might plausibly be understood as an action of “the People” themselves, the same can hardly be said for subsequent judicial interpretations of the products of such a moment. And as we have seen, there is no way around giving substantial authority to courts or some other institution to interpret inevitably vague constitutional provisions.
Second, the legitimating features of the higher lawmaking process (whatever they are) are themselves likely to lose their potency—relative to the features of normal politics—over time. The People who act more-or-less directly through higher lawmaking (however they may do that) are the People of a particular time and place; as that time fades further into the past, those People are likely to have less and less in common with Bickel’s “people of the here and now.” There are theories by which we might envision the corporate People of the framing and the corporate People of the here and now as the same People,40 but it is difficult to articulate a convincing process-based account of why a supermajoritarian consensus that occurred 150 or 200 years ago is more legitimately binding than a majoritarian consensus that exists today. None of today’s People participated in the framing or ratification of centuries-old constitutional provisions; and as the cultural gap between today’s People and those who did participate grows wider with time, it becomes more and more difficult to imagine that the interests of many of today’s People were well represented in the framing and ratification process.
It is possible, though it seems unlikely, that there is a persuasive process-based explanation of why a current democratic majority should consider itself bound by unelected judges’ interpretations of constitutional provisions laid (p.237) down by long-ago supermajorities. But such an explanation, if it exists, is far from obvious. As things stand, a noninstrumental understanding of Ackerman’s dualist democracy appears subject to essentially the same shortcomings as an instrumental understanding.
Rescuing Ackerman’s dualism. The difficulties with Ackerman’s version of dualism flow from his premise, largely implicit, that higher lawmaking is procedurally superior to ordinary democracy (albeit practicable only at infrequent “constitutional moments”). Even if that is so, it fails to explain why the authority of constitutional law does not diminish with time, or why the decisions of constitutional courts should carry the same weight as the efforts of constitutional Framers.
The DR account of constitutional law will not share that premise. The DR account begins with the assumption that ordinary democracy is, generally speaking, the most accurate (and thus the most authoritative) way to resolve disputes about the prospective content of the law. The role played by constitutional law on the DR account is not that of an extraordinary, idealized version of democracy that occasionally appears to set regular democracy straight, like an Olympian god descending to walk for a while among mortals. Its role, rather, is simply that of a backstop: Constitutional law steps in to resolve disputes where some salient flaw renders ordinary democracy incapable of doing so. On the DR account, therefore, what matters is not whether constitutional procedures can always live up to a “higher lawmaking” ideal, but whether constitutional procedures typically can resolve those disputes more accurately than ordinary democracy can. I will suggest in this chapter and the next that constitutional process, considered as a whole, generally is capable of meeting that challenge.
D. The Reductionist Strategy
A fourth prominent response to the countermajoritarian difficulty attempts to deny or downplay the existence of a meaningful difference between constitutional law and ordinary democratic politics. There is a critical version of this strategy, which holds in essence that constitutional adjudication is simply politics in disguise and thus lacks the legitimacy of democratic politics in its more overt forms (which legitimacy some of these same critics, incidentally, also question).41 My interest here is in the use of this approach, which we might label reductionism, to defend constitutional law and judicial review. Reductionists argue that the procedures of constitutional law are infused with political elements, and subject to political constraints, to such an extent that they are in effect part of the system of ordinary democratic politics and thus are not substantially countermajoritarian at all. Reductionists emphasize such factors as the (p.238) political appointments process for federal judges; Congress’s power to control the budget, size, jurisdiction, and procedure of the federal courts; and the need for political officials and lower-court or state-court judges to implement judicial decisions. These features of constitutional decisionmaking, they contend, strongly limit judges’ typical willingness or ability to reach constitutional results that stray far from the American political mainstream. Reductionists back up this claim by pointing to the historical record, which suggests that constitutional doctrine eventually, if not immediately, tends to “follow the election returns,” to paraphrase Mr. Dooley.42
The purest examples of reductionist theory come from positive political science, many practitioners of which tend to view the Supreme Court as just another political institution, albeit an idiosyncratic one.43 In the legal academy, semireductionist theorists like Neal Devins, Louis Fisher, and Barry Friedman take a more nuanced view, emphasizing the presence of “dialogue” between the judiciary and the political branches and asserting that constitutional doctrine tends, in the long run, to reflect input from both directions.44 These semireductionists agree with the political scientists, however, that the Supreme Court’s constitutional decisionmaking ultimately is constrained by popular and political forces, not the other way around.
There is, as I’ll argue in the next chapter, considerable and important truth in the reductionist insight. As a defense of constitutional law and judicial review, however, it suffers from two complementary shortcomings. First, as a normative matter it treads perilously close to critique; for it fails to stake out a strong distinctive role for constitutional law and thus a persuasive justification for it. However closely the Court follows the election returns, surely it almost always lags behind the elected branches themselves in this regard—begging the question why we need the Court or the Constitution at all, if what they provide is just a less-efficient form of ordinary majoritarian democracy. One might venture an explanation that turns on the advantages of divided power or of a relatively nonpopulist brake to promote deliberation and prevent sudden fluctuations in (p.239) the system, but the idea of the Court as a sort of backup Senate hardly seems sufficient to justify the enterprise. In any event, it utterly fails to make sense of the Constitution itself, which can only be interpreted as an attempt, not simply to slow the pace with which the election returns are implemented, but to impose real limits on what the election returns can accomplish.
This last point suggests the second major shortcoming of the reductionist defense, namely that as an interpretation of American constitutional practice it is noticeably thin. As I’ll suggest in the next chapter, the relationship between the election returns and the constitutional doctrine generated by the Court is considerably more complicated than one of cause and effect. To be fair, theorists I’ve labeled as “reductionists” almost never claim a simple cause-and-effect connection. But for reductionism to work as a normative defense of constitutional law, it must give short shrift to two central features of American constitutionalism. The first, just mentioned, is the existence of the Constitution itself, which—regardless of how it is implemented—clearly represents an effort to constrain ordinary democracy. To envision constitutional law as simply a subset of democratic politics thus is to deny any significance to the fact that the Constitution exists.
The second feature, following closely on the first, is the fact that the Court frequently interprets the Constitution in ways that are widely unpopular, obstructive of the immediate goals of the political branches, or both. There are many well-known examples here, from the Court’s holding in 1793 that the states are not immune from damages suits in federal court,45 to its Dred Scott decision essentially forbidding Congress to compromise on the issue of slavery in the territories,46 to its long string of decisions invalidating popular Progressive-era and New Deal economic regulations,47 to its 1967 ruling striking down (p.240) antimiscegenation laws,48 to its 2010 holding that corporations can spend freely in election campaigns.49 In each of these areas (with the exception of campaign finance, which is still very much in flux), the Court’s current doctrine coincides with broad popular opinion. On some issues, however (e.g., state sovereign immunity, slavery), it took one or more constitutional amendments to accomplish this;50 in other cases (e.g., economic regulation) the Court came around only after a long period of sustained political movement in a particular direction.51 And in some instances (e.g., antimiscegenation laws) it appears to be the public’s opinion that has evolved to match the Court’s, not the other way around.
As these examples show, there is some significant functional substance to American constitutional law that transcends, and sometimes confounds, the workings of normal democratic politics. At the very least, constitutional law and judicial review “may clog the administration” and “convulse the society,” as Madison said of minority factions.52 And sometimes they might affirmatively change the course of societal values. These effects cannot be explained away by classifying constitutional law as simply a subset of politics; if they are to be justified, they will require something substantially more robust.
3. Waldron’s challenge
The DR account rejects outright the substantive response to the countermajoritarian difficulty, and it can patch the significant holes in the proceduralist, (p.241) dualist, and reductionist responses. I want to set up the DR approach to constitutional law by first describing the most articulate contemporary statement of the countermajoritarian difficulty, which belongs to Jeremy Waldron.
Waldron’s critique of constitutional law follows two tracks, only one of which will be my central focus here. Waldron’s first line of attack (and, one senses, the approach he thinks most important) is, in his phrase, “process-based”: Waldron argues that noninstrumental considerations—reasons of what John Rawls called “pure procedural justice”53—favor decisionmaking by majoritarian democracy over decisionmaking by supermajoritarian constitutional procedures supplemented by elitist judicial review. Waldron’s theory of legal authority (including constitutional authority) is a type of DR account: He believes law must be justified, if at all, as a response to the problem of disagreement.54 Waldron’s account, however, turns primarily on what he believes to be the inherent value of procedure—on the supposed capacity of a decision-procedure to generate strong reasons for obedience to its results, quite apart from the quality or nature of those results themselves. Such “process-based” reasons, Waldron notes, would (assuming they exist) fulfill the requirement that legal authority be justified only by reference to content-independent considerations; process-based reasons are types of content-independent reasons.55 And Waldron asserts that fair participation among equals is a procedural feature capable of generating process-based reasons to obey the procedure’s results—that is, of clothing those results with legal authority.56 But, Waldron argues, majoritarian democracy is far more fairly and equally participatory than countermajoritarian constitutional law and judicial review. Process-based considerations therefore favor ordinary democracy over constitutional law.57
There is cause to be skeptical of this process-based attack. As I suggested in Chapter 3, it is difficult to discern just what inherent value—value without any regard to outcomes—that is capable of supporting legal authority might lurk within features of a procedure. The most plausible candidate as a procedural feature that can generate process-based reasons for obedience appears to be the (p.242) kind of fair and equal participation that Waldron cites, but it is much easier to understand the value of participation in the instrumental, outcome-based terms I outlined in Chapter 3. Without some reference to outcomes, it’s hard to see why people ought to care about procedure at all, at least in the ways necessary to generate legal authority. And even if we take participation to be a purely process-based value, majoritarian democracy does not have a monopoly on fair participation, as the discussion of adjudication in Chapter 5 suggests.
I will focus here, however, on Waldron’s second line of attack, which seeks to undermine the “outcome-based” or instrumental defense of constitutional law and judicial review. As I noted in Chapter 3, Waldron accepts the possibility that legal authority, on a DR account, might derive from a certain kind of “outcome-based” reason, one that turns not on the substance of any particular result generated by a procedure, but rather on the procedure’s tendency, over the run of cases, to generate good results.58 This is the type of reason to which I’ve been referring as one of procedural accuracy (what Rawls calls a matter of “imperfect procedural justice”59). Waldron considers some prominent arguments that constitutional judicial review is more accurate, on questions of individual rights, than ordinary democracy and finds them all wanting. In response to points offered by Bickel and others, he asserts that by the time constitutional cases reach the Supreme Court or other high-level constitutional tribunals, their original basis in concrete facts has become obscure and their issues tend to be presented in a form that is at least as abstract as those the legislature decides.60 Waldron also contends that the presence of a canonical constitutional text tends to “distract” courts, and others, from substantive arguments about rights, and that judicial attentiveness to matters of constitutional “principle” thus often misses the point, focusing on interpreting archaic language or divining the “original intent” of its authors rather than the underlying normative issues themselves.61 His conclusion is that there is little or no reason to believe that (p.243) constitutional law and judicial review are more-accurate ways to identify and enforce rights than the procedures of ordinary democracy.
In what follows, I will have relatively little to say about Waldron’s specific points of comparison between judicial review and democratic legislation.62 Waldron’s outcome-based arguments turn on considerations of what I’ve called competence: Waldron offers reasons to think judicial review is no more competent (and perhaps less so) than democratic legislation in resolving disputes about rights. My own outcome-based case for constitutional law and judicial review, by contrast, will be based mostly on considerations of impartiality, an ingredient of procedural accuracy that Waldron largely ignores.63 I will argue that constitutional law and judicial review, despite their flaws, are necessary responses to the problem of partiality that inevitably crops up when democracy is asked to resolve fundamental disagreements about itself. Some of the features of which Waldron complains—particularly the focus on divining principles from archaic texts—turn out, I’ll contend, to play a central role in the relative impartiality of constitutional law. And I will suggest that American-style constitutional adjudication can be acceptably competent, too—in ways that, not incidentally, bring it into closer contact than Waldron acknowledges with the workings of ordinary democracy that he rightly admires.
4. Constituting democracy
To begin a response to Waldron’s statement of the countermajoritarian difficulty, let us assume that the alternative to constitutional law and judicial review is to rely solely or primarily on democracy in the general form outlined in Chapter 4. As I argued in that chapter, democratic government derives its authority from the three central features of broad participation, majority rule, and regular elections; so let’s assume that the alternative to submitting some issues for constitutional resolution—the “majoritarian” end of the countermajoritarian difficulty—is to resolve them by a democratic process that includes these features.
Note, however, that rather open-ended, even aspirational concepts like “broad participation,” “majority rule,” and “regular elections” are not by themselves sufficient actually to operationalize democratic government. Before we can have (p.244) a working democracy, we will need to mold these general ideas into concrete forms—legislatures, electoral systems, methods of enacting, enforcing, and interpreting laws, and so on. Democracy (however defined) presupposes certain procedures and institutions—“the people [cannot] magically speak for themselves unfettered by institutions or procedures,” as Christopher Eisgruber (citing Stephen Holmes) puts it64—and those procedures and institutions cannot operate or even exist, at least not for long, without relatively stable rules to govern them. Constitutional law in some form thus seems necessary to democracy in the same way that rules of grammar are necessary for speech (or, to revisit an example from Chapter 2, that rules of the road are necessary for driving): It “set[s] out the rules by which political discussion will occur,” as Cass Sunstein notes.65
At a basic, literally constitutive level, then, constitutional law is not only consistent with democracy but essential to it. The existence and functioning of anything we can call “democracy” depend on the existence of constitutive rules. They also depend on a certain degree of obduracy in those rules. Like the rules of driving or of grammar, constitutive rules would fail to perform their function—namely to establish and facilitate some activity, such as democracy—if participants in that activity spent most of their time debating the rules themselves. If the activity is to succeed, the understanding among most of the participants must be that its constitutive rules should be presumptively resistant to sudden change. Waldron’s democratic premises thus depend to some extent on the presence of what can fairly be called constitutional law—a set of rules, or norms, that establish the democratic process, are relatively fixed and stable, and thus constrain the process they have established.
Waldron, for one, appears to accept roughly this degree of constitutional law. His critique assumes “a broadly democratic political system with universal adult suffrage” that includes “a representative legislature, to which elections are held on a fair and regular basis.”66 (In fact Waldron makes a few more-robust assumptions as well, including bicameralism and a committee structure in the legislature and the existence of political parties.67) “None of this is meant to be (p.245) controversial,” Waldron asserts; these assumptions merely “pick out the way in which democratic legislatures usually operate.”68
Note, too, that the need for constitutive democratic rules presupposes the need for some extrademocratic institutions and procedures that will create those rules. Constitutive rules bring democracy into existence; those rules cannot themselves be fashioned using (as yet nonexistent) democratic procedures. This is not to say that the mechanisms for generating constitutive rules cannot be democratic, in the sense of being broadly participatory, utilizing majority rule, and so on; the point is simply that the democratic system that will be constituted by the rules cannot, by definition, be the system that creates the rules constituting it. Constitutive rules must be generated by some mechanism that preexists, and thus is external to, the institutions and procedures of democratic government they make possible.
And note, finally, that the function of rules, including constitutive rules, requires mechanisms to interpret and apply those rules to particular cases. (This was of course a central theme of Chapter 5.) I will argue below that those mechanisms too must be in some meaningful way external to the democratic system whose constitution they are interpreting. For now the point is only that some standing apparatus for interpreting and applying constitutive rules must exist in a functioning democracy.
We have, then, a sort of bare-bones case for constitutional law, one that is implicit in the very idea of democratic government. Constitutional law at some level is necessary simply to constitute democracy, to bring it into existence and give it form. Constitutional lawmaking procedures—procedures separate from the democracy being constituted—are required to carry out this genesis, and standing procedures are required to interpret and apply the constitutive rules of democracy going forward.
(p.246) All of this, I take it, is unobjectionable to Waldron and most other critics of constitutional law. The problem for Waldron and others arises once we move beyond basic constitutive rules and begin using constitutional law as a way to constrain rather than simply constitute democracy. Of course democratic institutions and procedures require rules to bring them into existence, and of course those rules will constrain in the sense of defining those institutions and procedures—in the same way the rules of grammar constrain the activity of speech. The crisis, for Waldron and other critics, appears when constitutional law is used not to constrain how democracy works, but rather to constrain what democracy does. The fact that rules of grammar necessarily constrain the process of speech does not justify separate rules that constrain what can be said—what topics are off limits, what topics must be discussed, and so on. It is these substance-constraining rules—which Waldron describes, in shorthand, as constitutional rights 69—that create the countermajoritarian difficulty.
It turns out, however, that the need to define (and thus to constrain) how democracy works is a rather broad imperative indeed—so broad, in fact, that often the line between how democracy works and what democracy does becomes blurred beyond recognition.
5. Resolving disputes about democracy
The problem, in a nutshell, is that there is plenty of room for disagreement about what exactly the institutions and procedures of democracy should look like and whether, in any given instance, they are functioning correctly. Those disagreements will have to be avoided, mitigated, or resolved in most cases in order for democracy to function, that is, to generate laws and other decisions that are generally recognized as authoritative. But the institutions and procedures of democracy themselves—the things people may disagree about—are unlikely to be capable of resolving those disagreements authoritatively. A more expansive role thus begins to emerge for constitutional law—a role including not just the establishment of basic constitutive rules, but also the expansion of those rules to avoid disagreements about democratic authority, the interpretation and enforcement of those rules in a way that constrains the institutions they define, and the resolution of disputes about whether the rules are being followed in any given instance.
We can begin to grasp this broader dispute-resolving function of constitutional law by recognizing the possibility that one or more of the features that makes democracy relatively accurate will be missing, or malfunctioning, in a particular instance.
Consider the two basic types of democratic dysfunction identified in Footnote Four of the Carolene Products decision and elucidated by John Hart Ely. The first is legislation that “restricts… political processes,” thus artificially entrenching those currently holding power (the “ins,” as Ely calls them).70 A paradigmatic example is the Sedition Act of 1798, which outlawed, among other things, conspiracies “to oppose any measure or measures of the government of the United States” and the knowing publication of “any false, scandalous and malicious writing… against the government of the United States… with intent to defame the said government.”71 The Sedition Act made much antigovernment speech and opposition political activity punishable as a crime, thus protecting incumbent government officials from criticism and insulating them artificially against the possibility of removal at the next election. Had the Act remained in force—thankfully it was allowed by the incoming Republican Congress to expire in 1801—it might well have disrupted the normal cycling function of democracy, the process (discussed in Chapter 4) by which regular elections can launder the effects of occasional partiality. It also, and as a result, might have concentrated political power in a group of like-minded officials, frustrating the partiality-atomizing function of broad participation and majority rule and compromising the competence that normally flows from diversity. The Act, in short, threatened to undermine many of the qualities that give democracy its authority.
The other type of dysfunction identified by Footnote Four and Ely is legislation that reflects, or perpetrates, “prejudice against discrete and insular minorities” who can’t get a fair shake in the democratic process.72 Here the classic (tragic) examples are the mandatory school segregation and other Jim Crow laws common in the American South during the first half of the twentieth century. Those laws, born of a shared conviction on the part of most southern whites that blacks were inferior, had the effect of isolating blacks, ensuring that their inability to participate fully in democratic politics would continue indefinitely. Again both the impartiality and the competence of democracy were compromised: impartiality, because the white majority acted (and entrenched its power to act) as a self-interested faction on issues relating to race; and competence, because the benefits of broad-based and diverse decisionmaking were lost as a result. For all intents and purposes, southern blacks during Jim Crow (p.248) (like South African blacks during apartheid) were not citizens of a democracy; they were subjects of a white oligarchy.
These examples are relatively extreme and, in a reasonably well-functioning modern democracy, relatively uncommon. But they demonstrate that dysfunction can occur even in democratic systems. As Waldron notes, “[n]o decision-procedure”—democracy included—“will be perfect.”73 And the mere possibility of democratic dysfunction also raises the possibility—indeed the inevitability—of reasonable disputes about whether, in any given case, such a breakdown has occurred.
B. The Inevitability of Democratic Disputes
Imagine that a small postcolonial country—Carolenia—establishes basic democratic procedures and institutions along the lines sketched in Chapter 4, featuring regular elections, broad participation, and majority rule. And suppose the newly formed legislature of Carolenia enacts a statute substantially similar to the Sedition Act. A citizen of Carolenia (we can call him Quentin) believes the legislation represents a democratic dysfunction, a failure of the reasonable competence and impartiality that give democracy its authority. But suppose there are reasonable arguments on the other side of the question: At the time the Carolenian Sedition Act is adopted, Carolenia is under threat of invasion by a hostile foreign power with many agents and sympathizers within Carolenia itself, and many Carolenian citizens believe the law is justifiable as a temporary response to this serious danger.
Notice that what we now have is a dispute among the citizens of Carolenia about the existence of democratic authority. Quentin (and others of like mind) believe the Carolenian version of the Sedition Act is an abomination, an utter failure of democratic process that, as a result, lacks democratic authority; they believe, moreover, that the democratic authority of the Carolenian government as a whole is put at risk by the artificial entrenchment the statute threatens to create. But other Carolenians disagree; some think Quentin’s complaints about the statute’s supposed threat to democratic authority are overblown, while others recognize the threat but think it is outweighed by pressing requirements of national security. There is, then, a reasonable, good-faith dispute over whether the sedition statute in fact embodies a form of democratic dysfunction.
So long as democratic dysfunction is possible, disagreements like this one inevitably will arise in a democratic society. People will disagree, reasonably and in good faith, about what constitutes a malfunction or breakdown in democratic authority; and they will disagree about whether, and when, malfunctions can be tolerated in the name of other important goals. Indeed—and this is an important point—this phenomenon of prevalent disagreement inevitably will extend (p.249) beyond the question of democratic breakdowns to the constitutive features of democracy itself. People will disagree not just about whether democracy has failed in some particular instance, but more generally, about what democracy even is and what it entails. They will disagree about what constitutes reasonable democratic impartiality and competence, and thus reasonable democratic accuracy. They will disagree about how “broad” democratic participation must be, about the applications of and exceptions to majority rule, and about whether elections are sufficiently “regular.” And they will disagree about the likely factual consequences of particular legislation or other government actions for democratic authority. These many manifestations of democratic disagreement will be unavoidable products of the Rawlsian “burdens of judgment,” discussed in Chapter 2, that the members of any liberal society will face.74
As Waldron notes with respect to disputes about rights,75 the lion’s share of these disagreements must be resolved peacefully if a democratic society is to function. Legislatures and other institutions of democracy cannot operate for long without knowing how they are to operate and what kinds of laws they may enact. A society cannot remain a society amidst pervasive uncertainty about the legitimacy of its laws and the duty of its members to obey them.
The question then becomes what sort of institutions and procedures are best able to resolve these (inevitable) democratic disagreements in a way that is acceptable to the disputants. And there is good reason to think the answer is not the very institutions and procedures that are themselves the subjects of dispute.
C. The Partiality of Democratic Self-Judging
To see why, suppose someone suggests to Quentin, our dissenting Carolenian citizen, that he take his concerns about the sedition legislation to the Carolenian legislature itself, petitioning the body to repeal the statute on the ground that it entrenches incumbents and thus undermines democratic authority. Quentin might quite reasonably object that the very institution that enacted the questionable statute is unlikely to be an impartial arbiter of its democratic propriety. He might worry that the members of the legislature—who, after all, stand to benefit personally from continued enforcement of the statute—would not provide a truly impartial hearing for his complaint.
The problem Quentin faces is akin to one of the chief reasons a community like Carolenia might choose democracy over autocracy or oligarchy in the first place. People with their own strong interests in an issue are unsuitable to impartially resolve disputes about that issue; judges should not decide cases that affect their pocketbooks, jurors should not vote on the fate of friends and (p.250) relatives, and so on. As we’ve seen, the principle nemo iudex in sua causa—no one should serve as judge of his own case—is a central tenet of acceptable dispute resolution. An autocrat or small collection of oligarchs often will be self-interested in the issues they must decide, thus posing an obvious threat to the nemo iudex principle. Of course, democratic citizens often are self-interested as well; but, as we saw in Chapter 4, democracy’s technique of blending each citizen’s (potentially self-interested) vote on an equal basis with every other citizen’s (potentially self-interested) vote tends to launder that partiality, to mitigate the force of each individual’s self-interest by atomizing it among that of many others. This stands in contrast to autocracy or oligarchy, in which the likelihood of some shared self-interest on the part of those in power—a single individual or small group of them—becomes proportionately higher.
With respect to challenges to its own power, a democratic legislature is, or at least is quite likely to be, a self-interested oligarchy. Legislators, like other human beings, are inclined to look sympathetically on measures that would extend or preserve their own authority; perhaps more to the point, the rest of us are unlikely to trust legislators (as human beings) to assess those measures impartially. “If men were angels,” Madison reminded us, “no government would be necessary”;76 nor must we assume that legislators are devils in order to distrust their judgment on questions of their own power. As Ely puts it, “[p]erspective is critical, and one whose continued authority depends on the silencing of other voices may well in all good faith be able to convince himself” that doing so is justified on appropriate grounds.77
Notice that the threat to legitimate authority here is not that the legislature is an oligarchy in the sense of being an elite subset of the citizenry as a whole. The problem, rather, is that the legislature is an oligarchy in the sense that its members share the same interest in an issue they have the power to decide. We trust democratic legislatures (elitist though they may be) to resolve most policy issues because we know there is likely to be a cacophony of interests, and thus of voices, regarding such issues in the body—and in the larger community the body represents. Self-interest on one side is balanced by self-interest on the other; the result, if things work well, is a lively debate that generates an outcome better than any side on its own could have produced. But the legislature’s (p.251) interest in retaining power is likely to be monolithic, or nearly so. Some members may pursue that interest purposefully, others may allow it to sway them subconsciously, but most are likely to be moved by it somehow.
Note, too, that the crucial question is at least as much one of perception as of reality. The appeal of democracy (on the DR account) is not merely that is likely to be more competent and impartial than the alternatives, but that it is—mostly as a function of these real advantages—more likely to be widely perceived as competent and impartial. Acceptable dispute resolution depends ultimately upon people’s beliefs about the procedure to which they are submitting. So the issue, with respect to democratic self-judging, is not whether any given set of legislators at any given time is capable of remaining objective and neutral on questions respecting their own power. The issue is whether a democratic citizenry is likely to accept, as reasonably impartial, a system in which democratic legislatures routinely have final authority to decide those questions. And the principle nemo iudex in sua causa strongly suggests that the answer might well be no.
D. The Promise of Constitutional Dispute Resolution
If disputes about democratic authority inevitably will arise (as both the possibility of democratic dysfunction and the indeterminacy of concepts like “democracy” ensure), a democratic community will have to develop means for resolving them acceptably. But there is good reason to think the institutions that are the subjects of these disputes—legislatures and other units of ordinary democratic government—will be saliently partial, or will be seen to be so, and thus will not be suited to the task. A democratic community therefore will have cause to submit these democratic disputes to institutions and procedures that are meaningfully external to democracy—separate from and independent of the institutions and procedures whose authority is at stake—and thus relatively impartial with respect to the issues being resolved.
The idea here is familiar from Chapters 3, 5, and 6; it is the notion of a neutral third party to serve as a reasonably impartial arbiter of disagreements. We can now begin to grasp a central justification of the “countermajoritarian” nature of constitutional law, and in particular of a constitutional process that includes judicial review. Constitutionalism in the American mode assigns important decisions about democratic authority to relatively politically unaccountable judges. This comparative unaccountability is precisely the thing that Waldron and many other critics of judicial review find most offensive. On the emerging DR account of constitutional law and process, however, the relative political independence of judges—their externality with respect to democratic politics—is at the core of their justification, of their raison d’être. The political insularity of constitutional judges gives them an impartiality with respect to democratic disputes that more-accountable democratic decisionmakers—those whose authority is in dispute—saliently lack.
(p.252) In Chapter 8, I will back up this admittedly rather cursory assertion with a more extensive narrative of judicial impartiality. And I will combine that narrative with another one that is equally important: an account of how constitutional adjudication, in addition to being relatively (and crucially) impartial, also can be reasonably competent as a way to resolve disputes about democratic authority. (Remember that impartiality can’t stand alone as a marker of systemic accuracy; competence is required too.) The point for now is a rather general one, namely that the inevitability of democratic disputes, and the obvious partiality of ordinary democratic institutions as a way to resolve those disputes, sets the stage for some relatively apolitical, external mechanism for settling them.
E. The Advantages of Constitutional Rules
We have now the bare-bones outline of a case for extrademocratic dispute resolution in a democratic community. But constitutional law typically consists of more than simply ad hoc acts of ex post dispute resolution by courts; it consists also of constitutional rules, of a body of legal doctrine to which the community adheres (or tries to) even absent an active dispute, and which constitutional courts apply (or try to) to disputes when they arise. By now the existence of constitutional rules should be no mystery, for their advantages in a number of respects have already been made clear.
Recall, first of all, the need for some relatively obdurate constitutive rules establishing the basic institutions and procedures of a democracy. Democracy cannot get up and running, and keep itself running, without reasonably stable understandings of the fundamental institutions and procedures of which it consists. Even critics of constitutionalism like Waldron rarely object to the existence of a set of baseline constitutive rules for democratic government to follow (and thus for courts, as relatively impartial arbiters of democratic authority, to enforce).
Note, however, that the need for constitutive democratic rules is simply a special case of the more general argument for legal rules advanced in Chapter 4. Legal rules, generally speaking, are advantageous—and, in a complex modern society, probably necessary—because they reduce the incidence and costs of ex post disputes about what should be done. They accomplish this by stating ahead of time, with some measure of determinacy, the consequences of particular actions. In the constitutional context, constitutive democratic rules bring relative determinacy to the structure of democratic government, thus reducing the incidence and costs of disputes about that structure, disputes that would prevent democratic government from functioning if they were too frequent or costly. Of course, as we saw in Chapter 5 and earlier in the present chapter, legal rules can never be perfectly determinate, and thus disputes about their application inevitably will arise. This seems even more likely to be true of constitutive democratic rules, which, if they truly are to be confined to a relatively (p.253) foundational level, will tend to be exceptionally open-ended and vague. I will have more to say on this point in Chapter 8. But even relatively open-ended rules can provide some determinacy and thus serve a valuable dispute-mitigating function.
Notice, too, that (as we saw in Chapter 5) legal rules, including constitutive democratic rules, can serve another important purpose as well: They can constrain the discretion of ex post dispute-resolvers such as courts. A judge who attempts, in good faith, actually to apply a constitutive rule to a dispute about democratic authority is not, to that extent, simply imposing her own views or preferences on the matter. If we are tempted to worry about the unintended consequences of ceding important decisional authority to politically unaccountable judges—and how can we not have some concern on this front?—the presence of constitutional rules, and of a strong judicial ethic of interpreting them faithfully, may ease our fears somewhat.
Given all that rules can do, it is important, as Chapter 5 explained, that they be generated by a process that is itself reasonably accurate, that is, reasonably competent and impartial. I will have much to say in the next chapter about the American-style process of generating constitutional rules. For present purposes, the crucial point in this regard is as follows: Just as the institutions of ordinary democracy are, thanks to their salient partiality, inapt at resolving ad hoc disputes about their own authority, so they are likely to be ill-suited to establishing general legal rules defining and governing that authority. The same element of self-judging rears its ugly head in the latter context as in the former. There will, then, be good reason for a democratic society to establish some process for creating constitutive and other constitutional rules that is meaningfully external to the processes of ordinary democracy those rules will govern.
There is an important caveat here, though. The perspective of someone—a person, a government official, a legislature or other decisionmaking body—seeking to define what its own authority will be in future cases may well be crucially distinct from the perspective taken by that same actor trying to assess its own authority in an immediately pressing case. It may be easier for Ulysses, anticipating his encounter with the Sirens, to lay down ahead of time a rule requiring that he be tied to the mast than for Ulysses to resist the Sirens’ song while it dances around in his ears. (This is precisely the point of the rule.) To borrow a slightly different trope, a group of people deciding on rules from behind what John Rawls called a “veil of ignorance”—that is, without knowing precisely how they will be affected by the rules they are adopting—may be a more reliable decisionmaker than the same group deciding how to act ex post, with knowledge of precisely how the decision will affect them.78 Distance can bring ignorance (p.254) and thus relative partiality. And so it might be the case that nemo iudex in sua causa will be of less concern when legislatures or other democratic institutions lay down rules to define or constrain their authority in the future, than when they attempt to judge the existence or scope of their authority on some particular, immediate occasion. In fact, as I’ll explain in Chapter 8, the constitutional rulemaking process on the American model incorporates the input of the democratic institutions whose authority is in question; but it tends to do so by using the passage of time to simulate Rawls’s veil of ignorance and thus to promote relative impartiality.
There is, then, a plausible case for the establishment, not just of an external means to adjudicate disputes about democratic authority, but also of constitutional rules that can be applied in those disputes. There is a case as well for the creation of those rules by a process that is meaningfully external to the democratic institutions and procedures they will govern.
And, finally, there is a case for expanding the scope of constitutional rules beyond the merely constitutive. Disputes about democratic authority, remember, will not be limited to clearly constitutive matters like the structure and basic responsibilities of government institutions. Disputes also will go to the kinds of issues identified in Footnote Four—issues like whether those in power are attempting to entrench that power in a way that frustrates democratic authority, or whether a political majority is behaving as a faction with respect to a minority. There will be good reason to liquidate ahead of time as many of these disputes as possible, too, by means of legal rules.
Suppose, for example, that the citizens of our imaginary nascent democracy, Carolenia, want to reduce the threat of the kind of factional entrenchment embodied in the Sedition Act. Now they might hope that the constitutive rules they’ve put into place—rules providing for things like separation of powers, perhaps, and specifying regular intervals for elections—will be enough to do the trick; but the example of the Sedition Act (which occurred, after all, against a democratic backdrop featuring these and other structural safeguards against entrenchment) will suggest otherwise. So the citizens of Carolenia might take the additional step of adopting a rule that forbids the government to punish citizens for engaging in political speech (a rule that might even be phrased as a guaranteed “right” of free speech). Such a rule would not be strictly constitutive in the sense of creating basic democratic institutions and procedures. It would, rather, take the form of an external constraint on the democratic institutions and procedures that have been constituted—a limit on what democracy may do, not just one of the rules allowing democracy to exist in the first place. The purpose of such a constraining rule would be prophylactic: It would be to prevent the occurrence of a certain identifiable form of democratic dysfunction.
Suppose the citizens of Carolenia also want to avoid the emergence of laws motivated by (and tending to perpetuate) irrational majoritarian prejudice, like (p.255) the Jim Crow laws in the American South. Here again, they might be tempted to hope that the structural features of the government they are constituting (separation of powers, federalism, and so on) will alone prevent such a malfunction from occurring; but history will suggest otherwise. So they might, again, adopt prophylactic rules imposing external constraints on what the democratic process can do: a rule, for example, that requires the government to afford “the equal protection of the laws” to every one of its citizens.
Constitutional law, in other words, might well consist of both constitutive and prophylactic rules. The constitutive rules would define the basic institutions and procedures of democratic government (a bicameral legislature, for example, or a system of separated powers or of federalism); the prophylactic rules would constrain those institutions and procedures in ways designed to prevent certain predictable types of democratic malfunctions (perhaps by guaranteeing “rights” of free speech or equal protection). Of course, the constitutive rules themselves also would serve a prophylactic function: By defining democracy in a certain way, they would represent attempts, not just to get democratic institutions up and running, but to design those institutions so as to maximize their authority-generating features (their impartiality, their competence) and minimize the risk that those features will break down.
These constitutive and prophylactic rules—constitutional law—and the extrademocratic means for creating and interpreting them—constitutional process—will of course constrain the ordinary mechanisms of democratic government. They could hardly constitute democracy, or prevent democratic malfunctions, or impartially resolve disputes about democratic authority if they were not constraining in this sense. At the same time, though, to describe them as “countermajoritarian”—or at least as counterdemocratic—would be to shortchange the extent to which constitutional law and process are necessary in order for democracy to serve its assigned function, namely to produce authoritative general laws.
In the next section of this chapter, I will explore in a bit more detail the types of democratic dysfunction that constitutive and prophylactic constitutional rules might be designed to avoid. (I will suggest that the list proffered by Footnote Four and Ely, while immensely insightful, is somewhat incomplete.) Then, in Chapter 8, I will describe how constitutional process—the means of creating constitutional rules and applying them to disputes—can avoid the partiality problem that afflicts ordinary democratic institutions while still retaining much that is meaningfully democratic.
6. Democratic malfunctions (and constitutional responses)
Recall from Chapter 4 that democratic authority rests, according to the DR account, on three basic properties of democratic government. Broad participation (p.256) brings competence in the form of diversity, deliberation, and aggregation of viewpoints and interests. Majority rule minimizes partiality by balancing each potentially partial vote against many other votes. And regular elections, by allowing democracy periodically to revisit its decisions, reduce the impact of partial decisionmaking over time and promote competence through trial and error.
Malfunctions in democratic authority occur when one or more of these properties is substantially compromised. I will suggest here that this is likely to happen in two basic categories of circumstance. The first involves the potential for divergence between the interests of the democratic citizens themselves and those of their elected or appointed representatives in government: Government officials may seek to use their delegated power to enlarge or entrench that power. The second category involves the danger of majority (or, in some cases, well-organized minority) factions: Groups of citizens with the capacity to control public policy might find themselves united by a common bias or self-interest. Either of these circumstances can vitiate democratic authority by frustrating one or more of the three crucial components of democratic accuracy, and constitutional law and process are means of preventing this from happening.
A. Agency Failure
One type of democratic malfunction is agency failure: Government officials might use their positions to augment or entrench their own power without regard to (and possibly in opposition to) the interests of the public.
We have already seen an example (at least an arguable one) of agency failure: the Sedition Act of 1798, discussed above, which made it a federal crime “to oppose any measure or measures of the government of the United States” or to knowingly publish “any false, scandalous and malicious writing… against the government of the United States… with intent to defame the said government.” By restricting speech that opposed them or their policies, Congress and the President (who was, ironically, our old acquaintance John Adams, advocate of “a government of laws and not of men”) made it more difficult to unseat them at the next election, thus entrenching their own power and frustrating the normal cyclical pattern of democratic politics. The Sedition Act therefore disabled the typical capacity of democracy to launder partiality and promote trial-and-error competence through repeated election cycles. This in turn threatened to shift political power from the democratic majority to an elite cadre of incumbent public officials. The resulting quasi oligarchy, by concentrating power in a relative few, would have amplified the potential partiality in each individual’s decisionmaking and diluted the competence advantages of broad participation.
Fortunately this parade of horribles did not come to pass as a result of the actual Sedition Act; the Federalists nonetheless lost both Congress and the Presidency in the 1800 elections, and the new Republican-controlled Congress allowed the Act to lapse in 1801. And in fact there is reason to suppose that Adams and the Federalist Congress adopted the Act, not as a way to punish (p.257) political opponents and entrench their own power, but out of real (if overblown) fear that French agents provocateurs would foment domestic unrest. But the democracy-disrupting danger of legislation like the Sedition Act is readily apparent. So too, as we’ve seen, is the unsuitability of the democratic branches themselves to assess this danger. Reasonable minds can differ (and did) on the propriety of the Sedition Act, but the President and Congress themselves, as saliently self-interested parties, could hardly be trusted to resolve the dispute impartially. (The fact that the Federalists, whatever their motives for adopting the Act, used it primarily to prosecute their political enemies only underscores this point.79)
Constitutional law can manage the risk of agency failure through constitutive and prophylactic rules that regulate the agent-principal relationship between democratic officials and the citizenry they are supposed to serve. The regulation might be direct—for example, a provision for regular elections or a prohibition on laws restricting political speech. It might also be indirect: division of power among different groups of officials who are selected, evaluated, and controlled by the citizenry in different ways, for example, thus allowing “[a]mbition… to counteract ambition.”80 Such means of “oblig[ing]” the government “to control itself”81 are familiar in other agency contexts within modern law—in the law of corporations, for example, where officers and directors owe legally regulated fiduciary duties to shareholders, and in the law of professional responsibility, where experts like lawyers and doctors owe duties to clients and patients.
As Ely documented,82 a large portion of American constitutional doctrine in fact can be understood as a series of hedges against the risk of agency failure. Articles I and II impose specific limits on the terms of office of elected representatives, thus requiring regular elections. The rights of speech, petition, and assembly guaranteed by the First Amendment limit incumbents’ ability to insulate themselves from political challenges. Article I, § 5 requires each House of Congress to “keep a Journal of its Proceedings,” thus enhancing the public’s capacity to monitor what its representatives are doing. The system of separation of powers, combined with a bicameral legislature and including staggered terms of office and different means of selection for members of the various branches, (p.258) greatly increases the transaction costs involved in the entrenchment of official power. The federal structure of the government has a similar effect.83
B. Majority (and Well-Organized Minority) Factions
A second kind of democratic malfunction is the development of majority factions: political majorities “united and actuated by some common impulse of passion, or of interest,” in Madison’s famous description.84 Like official partiality, the development of a majority faction frustrates the atomizing function of democracy, concentrating a single type of partiality to the point where it dominates political decisionmaking on certain issues. The dominance of partiality in a majority faction also is likely to undercut the “many minds” benefits of democracy for decisionmaking competence. And sometimes majority factions (like factions of self-interested officials) can artificially expand or entrench their own power, disabling the usual cycling property of democracy.
We have already discussed an example of majority factionalism as well: the school-segregation and other Jim Crow laws endemic in the American South for nearly a century following the Civil War. The white majority population in each of the southern states, united by a shared racism and distrust toward the black minority, elected like-minded representatives who enacted these segregationist measures in an attempt to isolate and subjugate the black minority.
Note that the chief democratic malfunction embodied in Jim Crow, from the perspective of the DR account, was not the racist substance of the segregationist laws (although to the extent those laws had the effect of entrenching the white majority’s power, that constituted a democratic failure too; more on this point below). The central problem was again one of partiality. Like the self-interested Congress and President who adopted the Sedition Act, the white southern majority was not a reasonably impartial judge on issues involving relations or allocation of resources between the two racial groups. This partiality problem risks being obscured by what are, to our modern sensibilities, obviously racist and inappropriate laws. But imagine a proposed law implicating race that is not so saliently venal: perhaps a bill requiring that affirmative action be used to remedy the inadequate representation of members of the black minority in public universities and employment. The partiality problem would remain with respect to such a proposal. If the members of the white majority share racist views, that majority hardly could be a reasonably impartial judge on the wisdom of the proposed law, any more than the members of Congress could impartially judge the wisdom of the Sedition Act.
(p.259) Note, too, that the partiality problem with Jim Crow laws was not simply a function of the fact that some individual members of the political majority voted based on racist views. As Madison recognized, and as I discussed in Chapter 4, individual or majority partiality can, in a democratic polity of reasonably large size, be cleansed by the votes of others in any given election and by a succession of elections over time; “relief” for such minority factions “is supplied by the republican principle, which enables the majority to defeat its sinister views by regular vote.”85 The democratic dysfunction of Jim Crow was the fact that the same source of salient partiality—a racist disdain for or distrust of members of the black minority—was shared by a majority of the citizens in the South. On matters relating to race, the southern white majority acted in essence like a Hobbesian autocrat, motivated by a single dominant set of biases and interests. It was this element of a majority acting as a single entity with respect to the minority that created the partiality problem by frustrating the usual atomizing properties of democracy.
(i) The Nature(s) of Majority Partiality
Of course, majorities always in some sense “act as a single entity” in voting by majority rule: Their members come together and speak with a single voice on whatever is the issue at hand. What makes examples like Jim Crow problematic from the perspective of democratic authority—examples of majority factions in Madison’s sense—is that the white majority shares a common source of partiality, a common “passion” or “interest” that deforms its judgment on certain issues. In fact this partiality might assume one or both of two distinct but related forms.86
Partiality as substantive prejudice. Racist partiality might take the form of prejudgment (literally “prejudice”) regarding issues relating to race—for example, a firm conviction that racial segregation is justified or necessary. The essence of this type of partiality is a resistance, a relative imperviousness, to the kinds of evidence and arguments that normally would be offered and considered in deciding such issues. A majority whose members share this kind of resistance is distinct from an ordinary majority that happens, after attending to the facts and arguments on either side, to reach the same conclusion on an issue involving race. Its substantive prejudice serves as the “common impulse of passion, or of interest” that makes the majority a faction.
Note, by the way, that we can recognize this prejudice as a type of partiality without taking a position in legitimate debates about issues involving race.
(p.260) We might reasonably disagree on the propriety of affirmative action—even on the propriety of racial segregation—while still acknowledging that a person or group that is impervious to facts and arguments on those issues is insufficiently impartial to authoritatively resolve our disagreement.
And note, relatedly and importantly, that this type of prejudice need not be the product of a belief, like that in racial superiority, that most of us are likely to consider irrational. Prejudice might be motivated by “interest” as well as “passion.” Consider, for example, the prodairy legislation enacted in many states during the first half of the twentieth century, such as laws prohibiting the sale of margarine.87 In passing these laws, political majorities in the dairy-heavy states might well have prejudged the issue of whether the sale of margarine should be permitted—not out of some irrational bias against margarine producers, but out of simple pecuniary self-interest. As Madison and his fellows understood, the ego as well as the id might render a person, or a group of people with similar interests, impervious to evidence and reasoned arguments.
Partiality as personal bias. The majority partiality behind Jim Crow–style laws might instead—or in addition—assume a related but subtly different form: It might amount to an animosity against a particular person or group of people, one that prevents the decisionmaker from treating that person or group fairly in the decision process. Members of the southern white majority might have supported racial segregation or opposed affirmative action, not because (at least not only because) they had formed relatively immutable views on the substance of those issues, but because they had a desire to harm members of the black minority or an ingrained resistance to taking their views and interests seriously. The adjudicative analogue here would be a judge or juror who has a strong preexisting dislike for one of the litigants.
This sort of interpersonal animosity seems to be what the Footnote Four Court had primarily in mind in worrying about “prejudice against discrete and insular minorities,” and it (rather than the kind of substantive prejudgment discussed above) is the focus of Ely’s elaboration of this Footnote Four concern.88 As with substantive prejudgment, though, the concern with racial or other types of interpersonal animosity is, from the DR perspective, chiefly a concern about the sanctity of the procedure, not about controversial substantive outcomes. We need not take a position on substantive questions of racial merit—or, to invoke a more current and thus still-controversial topic, on substantive questions regarding the morality of sexual orientation—to agree that a decisionmaker who (p.261) harbors an active animosity toward members of the relevant minority group cannot impartially resolve those issues.89
The malfunction of majority factions. More generally, a focus on “discrete and insular minorities” tends to obscure the possibility of majority factions united by self-interest instead of, or in addition to, irrational bias based upon race or some other arguably irrelevant characteristic. A majority faction is problematic even if its animating impulse is indisputably logical—a desire to protect its members from economic competition, for example. Where the majority (or, as we’ll see below, the minority) that holds political power shares the same self-interest, however rational, it lacks the capacity to impartially judge policy questions implicating that interest. The resulting democratic breakdown goes deeper than the (often arguable) presence of some unfounded psychological animosity—a fact neglected in Ely’s elucidation of Footnote Four proceduralism.
If a majority shares one or both of these types of partiality—substantive prejudgment of an issue or psychological animosity toward members of the minority—it thus lacks the typical authority of democratic government to resolve relevant disputes. The fact that a majority shares the same impulse of partiality blocks the usual capacity of democracy to atomize the partiality of individual voters. And the fact that the impulse is one of partiality—a prejudgment or bias rather than a joint conclusion based on evidence and arguments—separates a majority faction, and thus a democratic malfunction, from the ordinary phenomena of majority rule.
(ii) Minority Factions
While Madison and his colleagues focused many of their considerable faculties on the problem of majority factions, they discounted the danger of minority factions, which may “clog the administration” and “convulse the society” but, Madison confidently asserted, would “be unable to (p.262) execute and mask [their] violence under the forms of the Constitution.”90 This confidence has turned out to be misplaced. As the modern field of public-choice theory—an intersection of political science and economics—has demonstrated, well-motivated, well-organized minority interest groups often succeed in extracting “rents” from the democratic political process: legislation that benefits their interests without a corresponding benefit to (and often at the expense of) the interests of the broader public.91
Consider a refinement of the prodairy legislation example. Suppose a majority of the voters in a state would prefer to have a choice between butter and margarine, a less-expensive butter alternative, while voters affiliated with the dairy industry—a minority—want, out of self-interest, to exclude margarine from the market. Each dairy-industry voter—a dairy worker, a dairy owner or major investor, a supplier of goods or services to the dairy industry—is likely to have a relatively strong interest in excluding margarine, namely that the possible alternative—a market shift toward margarine—will greatly impact each dairy voter’s bottom line. This creates a powerful incentive for dairy voters to devote resources to the cause of excluding margarine and to other prodairy efforts. The most efficient way to do that might involve forming an industry organization that can lobby the state legislature, make targeted campaign contributions, and the like. And if the dairy industry is relatively small, the costs of organizing such a group will be relatively minimal. In contrast, any given member of the dairy-consuming public is unlikely to have a particularly strong interest in preserving the choice between butter and margarine; the marginal benefit of such a choice (and the marginal detriment of its loss) in the life of each consumer will not be worth the substantial investment of resources necessary to preserve (or avoid) it. The costs of organizing the huge, diffuse mass of the dairy-consuming public, moreover, into an effective political action organization will be quite high.
As these dynamics interact, the likely result will be the rise of the dairy industry as a politically powerful minority interest group with the capacity to extract “rents”—favorable legislation—from the political process in the state, including (p.263) a law taxing or prohibiting the sale of margarine. Legislators and other officials, for their part, will have strong incentives to do the dairy industry’s bidding, thanks to that industry’s well-organized lobbying and campaign-contribution efforts. And while legislators will not want to alienate the majority of voters who control their reelection, they will understand that the margarine issue matters relatively little to any individual (nondairy) voter and thus is relatively unlikely to influence many (nondairy) votes. The legislature, then, may well agree to enact prodairy, anticonsumer laws, including a tax or a ban on margarine, thanks to a rational calculation that the costs of displeasing the well-organized dairy industry will exceed the costs of displeasing the diffuse, disorganized majority of consumer-voters.92
Public-choice theory thus demonstrates that the problem of factions can combine with a sort of agency failure to allow minority interest groups to dictate public policy under certain conditions. This phenomenon is even more troubling on a DR account of democracy than the problem of majority factions. Empirically, the problem of rent-seeking minority factions appears substantially more likely to occur than the problem of majority factions: While the combination of a majority into a singly-motivated group probably will be infrequent in a large, diverse polity, the size and diversity of a polity actually is likely to enhance the effectiveness of minority interest groups by making it less likely that a diffuse majority will combine to offer resistance. And normatively, the possibility that policy issues will be determined by a self-interested or otherwise partial minority undercuts the impartiality and competence advantages of democracy even more than the danger of majority factions does.
(iii) Factional Entrenchment
As Footnote Four proceduralism recognizes, the possibility that a faction, majority or minority, will assume legislative power in a democracy raises the further danger of factional entrenchment: The faction may attempt to enact legislation that would have the effect of entrenching its power. Consider the following examples, loosely based on real cases:
• A state’s rural majority, predicting that the population will continue to shift toward urban areas, amends the state’s constitution to establish legislative districts of roughly equal geographic size and provide for the election of one representative from each district, resulting in increasing proportional overrepresentation of rural voters.93
• A state legislature controlled by one political party redraws, or “gerrymanders,” its electoral districts along demographic lines to disadvantage the opposing party.94
• A state’s white majority institutes a “poll tax” and a literacy test as requirements for voting in state elections, both of which disproportionately prevent voting by blacks and other racial and ethnic minorities.95
In these cases, as with cases of agency failure, the democratic malfunction consists in the artificial entrenchment of current majorities by impeding the ability of minority groups to gain influence through democratic elections. And, as in the case of the Sedition Act, the majority in power, because of its self-interest or bias—that is, because of its status as a faction—is unqualified to (p.265) impartially balance the competing considerations and assess the wisdom of the policies.
(iv) Constitutional Inoculation Against Factionalism
A democratic community might develop constitutional legal rules designed to frustrate the formation of majority or powerful minority factions, by dividing the instrumentalities of power among different institutions, for example. (The American Constitution accomplishes this through separation of national powers, bicameralism, and federalism.) It might also seek to block or limit the power of factions when they do form, by requiring that laws be justifiable by reference to broad public interest and forbidding or strongly discouraging legislation that turns on race, religion, or other common markers of bias. (American due-process and equal-protection doctrine serve these functions by requiring that virtually every law meet a rationality requirement96 and that laws employing “suspect” or “quasi-suspect” classifications—race, ethnicity, national origin, or gender—be subjected to heightened judicial scrutiny.97 First Amendment doctrine similarly (p.266) requires special justification for laws that hinge on religion.98) And it might regulate the terms of political participation in a way that discourages majority entrenchment (through, for example, provisions for regular elections, guarantees that members of minority groups can vote and hold office,99 and strong protections for political speech100).
The defense of constitutional law and process that I’ve outlined here, and will continue in the next chapter, resembles the Footnote Four/John Hart Ely approach in a crucial respect: Both approaches turn on the notions that ordinary democracy often suffers from certain kinds of flaws, and that constitutional law and process are reasonable responses to those problems. I will have a bit more to say in Chapter 8 about the relationship between my DR approach and Footnote Four proceduralism. For the moment, however, I want to address a worry that might seem to undermine any justification of constitutional law as a response to democratic dysfunction. The worry is that some elements that seem important to the actual practice of American constitutional law cannot be explained by such an account.
The paradigmatic object of this worry is the body of doctrine known as “substantive due process.” The Due Process Clause—really two clauses, one in the Fifth Amendment that applies to the federal government, the other in the Fourteenth Amendment that applies to the states—prohibits deprivation of “life, liberty, or property, without due process of law.” For more than a century the Supreme Court has interpreted the Clause to impose substantive limits on government’s authority to deprive someone of her life, liberty, or property, limits that apply without regard to the process by which government effectuates the deprivation.101 Prior to the New Deal, the Court used this “substantive due process” doctrine primarily to constrain the government’s ability to regulate (p.268) economic interactions.102 The Court abandoned that approach in the 1930s,103 but it now protects certain “fundamental” noneconomic rights—“privacy,” reproductive choice, sexual “autonomy”—under the banner of substantive due process.104
The account of constitutional law I’m offering here is intended to be interpretive—to fit reasonably, if not always perfectly, our actual practices of constitutional law and procedure. If the account fails to justify substantive due process, its claim to a reasonably good fit might thereby be diminished. And there is indeed reason to think that the DR account cannot fully explain substantive due process.
Consider Roe v. Wade, which is probably the best known (and certainly the most controversial) substantive due process decision. In Roe, the Court held that a “right of privacy” is “founded in the… concept of personal liberty” protected by the Due Process Clause and “encompass[es] a woman’s decision whether or not to terminate her pregnancy.”105 The Court seemed to treat this privacy right, and its reproductive-choice component, as staking out an area of activity that is immune from government regulation, a sphere into which even democratic procedures cannot intrude. (This view parallels those the Court has taken in other substantive due process cases involving activities like making labor contracts106 and engaging in private, consensual sex.107)
But it is far from obvious how a constitutional rule protecting individual privacy or choice from government intrusion serves the sort of constitutive or (p.269) prophylactic function that drives constitutional law on the DR account. To prohibit government from regulating abortion is not to constitute the basic institutions and procedures of democracy. Nor does such a prohibition seem directed at preventing agency failure, factionalism, or some other type of democratic breakdown (in contrast to, say, a prohibition on government regulation of political speech). More broadly, it is hard to see how a debate about whether the government should regulate abortion qualifies as a dispute about democratic authority in the appropriate sense. The prochoice claim, at least as the Roe Court seemed to understand it, is not a claim that one of the key ingredients of democratic authority—broad participation, regular elections, majority rule—is missing or malfunctioning. Instead it is a claim that those ingredients, functioning properly, have generated a morally wrong or harmful law. This is a claim about the substance of the legislation, not the authority of the process that enacted it; and it is precisely these kinds of substantive disagreements that ordinary democracy is designed to resolve acceptably.
If we understand substantive due process as the Roe Court did, then—as a position in a debate about substantive political morality, not about legal authority—then the doctrine has no apparent place in a DR account of constitutional law. But in fact it might be possible to understand much of substantive due process in other than purely substantive terms. Consider the doctrine’s first significant incarnation, as a means to protect laissez-faire economic values during the first third of the twentieth century.108 This “economic” due process was unjustifiable, and ultimately a failure, to the extent it attempted to use the instrumentalities of constitutional law simply to impose one side of a substantive dispute. But economic due process had one redeeming feature: It made it more difficult for minority factions to extract rents from the political process.109 In order to preserve an economic regulation from a due process challenge, the government had to convince the court that the legislation clearly benefited a broad spectrum of the public or, if it seemed to benefit only a subset of the public, that the benefited group had some special need for protection. Purely rent-seeking legislation typically had trouble clearing these hurdles. Supposedly “substantive” due (p.270) process, in other words, served a sanguinary procedural function, one that fits into the DR account offered here: It limited the influence of minority factions. (The Court’s ultimate rejection of economic due process—really its dilution in the form of a “rational basis” test for economic regulation—might be understood as a strategic calculation that the doctrine’s substantive costs (in the form of judicial value-imposition) outweighed its procedural benefits (in the form of mitigation of special-interest influence).)
We might similarly understand much of modern “privacy”-based substantive due process as in fact serving DR-related procedural functions. Constitutional protection of an abortion right, for example, has the effect of eliminating or reducing a significant barrier to women’s full and fair participation in democratic politics, namely the responsibilities of bearing and raising children, which, for a variety of reasons (some biological, some sociological), tend to devolve solely or predominantly on women.110 The right to use contraceptives might be justified on similar grounds.111 Constitutional protection of a right to same-sex intimacy might be necessary to counteract the work of factions, majority or minority, animated by personal bias against homosexuals or substantive prejudice (perhaps religiously inspired) with respect to the relevant issues. In this regard, the decision in Lawrence v. Texas, in which the Court used due process to invalidate a Texas law criminalizing same-sex sodomy, would be analogous to the requirement that economic regulations be rationally related to a legitimate governmental interest: In both contexts, the worry is that factions will legislate based on their own parochial interests or values rather than considerations of the broader public good.112
(p.271) On the other hand, not every aspect of contemporary substantive due process doctrine is comfortably justifiable on procedural grounds. In its 2000 decision in Troxel v. Granville, the Court invalidated a state law allowing “[a]ny person” to petition a court for the right of visitation of a minor child and requiring the court to order visitation in “the best interest of the child”; the Court invoked the “fundamental liberty interest” of parents “in the care, custody, and control of their children.”113 There certainly is a process-related danger in allowing the state to dictate a uniform educational program for every child; homogeny in education threatens subsequent homogeny in political views, thus compromising the central democratic advantage of broad political participation. And in fact the Troxel Court relied heavily on earlier decisions invalidating state bans on private-school education and the teaching of foreign languages.114 But the decision in Troxel itself seems hard to interpret as an expression of this procedural concern. Allowing a court to grant visitation rights to nonparents without special deference to the parents’ wishes poses no evident threat of rigid official indoctrination of uniform beliefs; nor is there reason to think that the law in question reflected some systemic bias against particular groups or prejudice with respect to the relevant issues.
(p.272) At bottom, though, it seems unlikely that the prominence of substantive due process doctrine in contemporary American constitutional law poses a dire threat to the DR account as an interpretation of our actual practice. If we fetishize the “substance” in substantive due process, viewing the doctrine as involving nothing more than judicial choices among controversial substantive values, then the doctrine will seem at odds with the DR approach. Typically, however, the doctrine need not be understood this way; much or most of it is consistent with the notion of constitutional law as a hedge against breakdowns in democratic authority. Of course, this is not to say that the content of substantive due process is somehow immune to reasonable disagreement. Obviously people disagree strongly about matters like abortion and gay rights; the point is that there is reason to think these disagreements cannot always be resolved acceptably using ordinary democracy alone.
8. From law to process
I’ve argued here that constitutional authority can be justified as a way of preventing democratic malfunctions and of resolving disputes about democratic authority. A huge part of this account, however, has to do with process; at bottom, constitutional authority is justified, if at all, as the result of a procedure, or a series of procedures, that avoids the salient partiality problems that sometimes afflict ordinary democratic politics. But those procedures are part of an overall system that includes ordinary politics; and their justification depends on their capacity, not just to avoid occasional democratic partiality, but to do so in a way that does as little damage as possible to the other accuracy-promoting virtues that democracy embodies.
In the next (and final) chapter, I argue that the processes of American constitutional law—of framing, interpreting, and implementing constitutional rules—generally do an acceptable job of balancing externality from politics with responsiveness to it, and thus of accommodating reasonable impartiality with reasonable competence. American constitutional process, I’ll contend, avoids the self-judging pathologies of democratic politics while ultimately honoring the primacy of democracy as a source of authoritative legal rules.
(1.) The term “counter-majoritarian difficulty” was coined in mid-century by the highly influential constitutional theorist Alexander Bickel. See ALEXANDER M. BICKEL, THE LEAST DANGEROUS BRANCH: THE SUPREME COURT AT THE BAR OF POLITICS (2d ed. Yale Univ. Press 1986) (1962). The term has since become synonymous with worries about the democratic legitimacy of judicial review. See generally Barry Friedman, The Birth of an Academic Obsession: The History of the Countermajoritarian Difficulty, Part Five, 112 YALE L.J. 153 (2002). The “difficulty” applies not just to judicial review, however, but to constitutional law more generally. Even if binding interpretations of the Constitution were rendered by the political branches of government rather than the Court, those interpretations still would be interpretations of—attempts to understand and apply—legal norms laid down by previous generations, norms that cannot be changed by “the actual people of the here and now,” in Bickel’s phrase, through ordinary democratic processes. See BICKEL, supra, at 17.
(2.) The straightforward view that government exists to protect natural rights, and that a government’s failure to do so justifies replacing that government, often is attributed to Locke. See, e.g., RANDY E. BARNETT, RESTORING THE LOST CONSTITUTION: THE PRESUMPTION OF LIBERTY 70–71, 75, 325–27 (2004); GORDON S. WOOD, THE CREATION OF THE AMERICAN REPUBLIC, 1776–1787, at 282–304 (1969). There is, however, a somewhat more nuanced plausible reading of Locke’s understanding about the relationship between natural rights and government, a reading that centers on the procedures of dispute resolution rather than the substance of natural rights. Locke emphasized that a pregovernmental state of nature was insufficient, not just to protect natural rights, but to resolve disputes about the content and application of natural rights. In Chapter 9 of the Second Treatise, Locke listed, among the “many things wanting” in the state of nature
First,… an establish’d, settled, known Law, received and allowed by common consent to be the Standard of Right and Wrong, and the common measure to decide all Controversies between them.
Secondly,… a known and indifferent Judge, with Authority to determine all differences according to established Law.
Thirdly,… Power to back and support the Sentence when right, and to give it due Execution.
JOHN LOCKE, Second Treatise of Government, in TWO TREATISES OF GOVERNMENT 395, §§ 124–26, at 396–97 (Peter Laslett ed., 1960) (emphasis in original). The problem with the state of nature, for Locke, was not simply (or even primarily) that people would viciously violate the rights of others, but rather that people, “being biased by their Interest, as well as ignorant for want of study,” will do so innocently. See id. § 124, at 396. Recall from the discussion in Chapter 4 that Locke’s rejection of Hobbesian absolutism turned on his recognition that a Hobbesian monarch would himself be biased by self-interest and thus ultimately would fail as an “indifferent judge” of society’s disputes. The “long train of abuses” that would, for Locke, spark a justified revolution can be understood, not as a series of clear monarchical violations of the dictates of natural law, but as a collection of instances in which the monarch displays his partiality by taking his own side in disputes about what natural law requires. See id. § 225, at 463. Locke, in other words, might be understood as advocating government (and critiquing absolutist government) not on the ground that some agreed set of natural rights must be protected, but for the reason that disputes about natural rights cannot acceptably be resolved without (reasonably competent, reasonably impartial) government.
(3.) See BARNETT, supra note 2, at 70–71, 75; WOOD, supra note 2, at 282–304.
(4.) For example, in Federalist No. 10, Madison rebutted objections to the relatively powerful central government the Constitution would create by arguing that a large republic can “control [the] effects” of “factions”—groups “of citizens, whether amounting to a majority or a minority of the whole, who are united and actuated by some common impulse of passion, or of interest, adverse to the rights of other citizens, or to the permanent and aggregate interests of the community.” THE FEDERALIST NO. 10, at 122, 125, 128 (James Madison) (Isaac Kramnick ed., 1987) [hereinafter Madison, FEDERALIST NO. 10].
Madison reemphasized the point in Federalist No. 51, arguing that the Constitution’s combination of vertical division of powers (between the states and the national government) and horizontal separation of powers (at the national level) would create “a double security… to the rights of the people…. [T]he society itself will be broken into so many parts, interests, and classes of citizens, that the rights of individuals, or of the minority, will be in little danger from interested combinations of the majority.” THE FEDERALIST NO. 51, at 318, 321 (James Madison) (Isaac Kramnick ed., 1987) [hereinafter Madison, FEDERALIST NO. 51].
In Federalist No. 78, Hamilton explained the (relatively) politically independent judiciary created by Article III in part as a way “to guard the Constitution and the rights of individuals from the effects of those ill humors which the arts of designing men, or the influence of particular conjunctures, sometimes disseminate among the people themselves, and which… have a tendency… to occasion dangerous innovations in the government, and serious oppressions of the minor party in the community.” THE FEDERALIST NO. 78, at 436, 440 (Alexander Hamilton) (Isaac Kramnick ed., 1987).
(5.) See Centinel, Letter to the Editor, Centinel I, PHILADELPHIA INDEPENDENT GAZETTEER, Oct. 5, 1787, reprinted in 2 THE COMPLETE ANTI-FEDERALIST, 136, 142–43 (Herbert J. Storing ed., 1981) (“The framers of [the proposed Constitution] have made no provision for the liberty of the press, that grand palladium of freedom, and scourge of tyrants; but observed a total silence on that head.”) (emphasis in original) (authorship attributed to Samuel Bryan); Brutus, Letter to the Editor, Brutus II, NEW YORK JOURNAL, Nov. 1, 1787, reprinted in 2 THE COMPLETE ANTI-FEDERALIST, supra, at 372, 373 (“[I]n forming a government on its true principles, the foundation should be laid… by expressly reserving to the people such of their essential natural rights, as are not necessary to be parted with.”) (authorship attributed to Robert Yates). See also James Madison, Speech in Congress (June 8, 1789), available at http://www.constitution.org/jm/17890608_removal.htm (introducing the Bill of Rights and acknowledging Anti-Federalist objections).
(6.) RONALD DWORKIN, FREEDOM’S LAW: THE MORAL READING OF THE AMERICAN CONSTITUTION 17 (1996).
(7.) Id. at 25.
(8.) Id. at 26.
(9.) Dworkin glosses his substantive understanding of constitutionalism with proceduralist rhetoric, describing his “constitutional conception” as a superior version of democracy itself. See id. at 15–19. Dworkin is of course entitled to apply the label “democracy” to whatever set of arrangements or values he likes; but for our purposes Dworkin’s maneuver is merely semantic. Recall that the understanding of “democracy” I outlined in Chapter 4 is a procedural one, in the sense of John Rawls’s “imperfect procedural justice”: It values democratic procedures as a way to generate, over time, results that are generally perceived as reasonably accurate. But it does not value democracy as a way to generate particular results. (On the distinction between purely substantive, or outcome-based, accounts of procedure and accounts of “imperfect procedural justice,” see the discussion in Chapter 3, section 2.) Dworkin’s “constitutional conception of democracy” is substantive, not procedural, in this latter sense: Dworkin values democracy as a way to generate particular results, namely the advancement of his “democratic conditions” of equal concern and moral independence.
(10.) Jeremy Waldron, The Core of the Case Against Judicial Review, 115 YALE L.J. 1346, 1373 (2006) [hereinafter Waldron, Core Case].
(11.) McCulloch v. Maryland, 17 U.S. 316 (1819).
(12.) United States v. Carolene Products Co., 304 U.S. 144, 152 n.4 (1938).
(13.) In a series of cases beginning during the First World War, the Court had addressed First Amendment challenges to laws criminalizing speech against the war or the draft and, later, speech espousing socialist or communist ideology. Most of these decisions upheld convictions, but from them emerged a rhetorically powerful series of opinions (mostly dissents) in which Justices Oliver Wendell Holmes and Louis Brandeis articulated a free-market theory of the Free Speech Clause, one advocating tolerance of dissenting viewpoints (and owing much to the ideas of John Stuart Mill). See Whitney v. California, 274 U.S. 357, 372, 377 (1927) (Brandeis, J., concurring); Gitlow v. New York, 268 U.S. 652, 672 (1925) (Holmes, J., dissenting); Schenck v. United States, 249 U.S. 47 (1919); Abrams v. United States, 250 U.S. 616, 624, 630 (1919) (Holmes, J., dissenting). Cf. JOHN STUART MILL, On Liberty, in UTILITARIANISM, ON LIBERTY, AND CONSIDERATIONS ON REPRESENTATIVE GOVERNMENT 69, 83–123 (H. B. Acton ed., 1972) (1859).
While the Holmes/Brandeis approach would not be accepted, more or less fully, by the Court for another thirty years (see Brandenburg v. Ohio, 395 U.S. 444 (1969)), the Court in 1938, when Carolene Products was decided, must have grasped the possibility that the Free Speech Clause might be applied relatively aggressively to protect dissident speech. Indeed, during the succeeding two terms the Court used the Clause to strike down a city ordinance prohibiting the distribution of leaflets on the public streets and sidewalks, see Schneider v. New Jersey, 308 U.S. 147 (1939), and to reverse the disturbing-the-peace conviction of a Jehovah’s Witness who played an anti-Catholic phonograph record on a street corner, see Cantwell v. Connecticut, 310 U.S. 296, 308 (1940).
The Carolene Court also stood on the cusp of the revolution in Equal Protection jurisprudence that eventually produced Brown v. Board of Education and outlawed Jim Crow. Later in the same term the Court would hold, in Missouri ex rel. Gaines v. Canada, 305 U.S. 337 (1938), that a state could not maintain its all-white public law school by paying for black citizens to attend school in neighboring states. This chink in the armor of the “separate but equal” doctrine—the principle, embodied in Plessy v. Ferguson, 163 U.S. 537 (1896), that equal protection was satisfied so long as racially segregated public facilities are materially equivalent—would be widened in subsequent decisions before the armor eventually fell away entirely in Brown. See Sipuel v. Bd. of Regents, 332 U.S. 631 (1948) (reaffirming Gaines); Sweatt v. Painter, 339 U.S. 629 (1950) (holding that a new blacks-only public law school was not materially equivalent to the established whites-only school); McLaurin v. Oklahoma State Regents, 339 U.S. 637 (1950) (holding that the isolation of black students within a formally “integrated” graduate school violated equal protection).
(14.) Carolene Products, 304 U.S. at 152.
(15.) Id. at 152 n.4. Footnote Four also suggested a third category of case, irrelevant for our purposes here, in which “[t]here may be a narrower scope for operation of the presumption of constitutionality”: cases in which “legislation appears on its face to be within a specific prohibition of the Constitution, such as those of the first ten Amendments.” Id.
(16.) See JOHN HART ELY, DEMOCRACY AND DISTRUST: A THEORY OF JUDICIAL REVIEW (1980).
(17.) Madison, FEDERALIST NO. 10, supra note 4, at 123.
(19.) See id. at 101–04,106–07, 151–52.
(20.) Chief Justice John Roberts opined at his confirmation hearing that “[j]udges are like umpires. Umpires don’t make the rules, they apply them. They make sure everybody plays by the rules, but it is a limited role. Nobody ever went to a ball game to see the umpire.” Confirmation Hearing on the Nomination of John G. Roberts, Jr. to Be Chief Justice of the United States: Hearing Before the S. Comm. on the Judiciary, 109th Cong. 55 (2005) (statement of John G. Roberts, Jr., Nominee to be Chief Justice of the United States).
(21.) See, e.g., Buckley v. Valeo, 424 U.S. 1 (1976); First Nat’l Bank of Boston v. Bellotti, 435 U.S. 765 (1978); Austin v. Michigan Chamber of Commerce, 494 U.S. 652 (1990); McConnell v. Fed. Election Comm’n, 540 U.S. 93 (2003); Fed. Election Comm’n v. Wisconsin Right to Life, 551 U.S. 449 (2007); Citizens United v. Fed. Election Comm’n, 130 S.Ct. 876 (2010).
(23.) BICKEL, supra note 1, at 24–27.
(24.) Id. at 25.
(25.) Id. at 25–26.
(26.) See CHRISTOPHER L. EISGRUBER, CONSTITUTIONAL SELF-GOVERNMENT (2001). The description of Eisgruber’s account as “Bickelian” is mine, not Eisgruber’s; in fact Eisgruber cites Bickel only once in his book. See id. at 225 n.50.
(27.) Id. at 5, 53.
(28.) Id. at 53.
(29.) Id. at 19.
(30.) Id. at 55, 3.
(32.) See id. at 55 (“Of course, people might believe none of this. They might believe that moral positions are nothing more than tastes which people happen to have…. But I think that most Americans do, in fact, believe that one should have moral reasons for taking moral positions, and that it is productive to discuss these reasons.”).
(33.) The canonical statement of Ackerman’s theory appears in 1 BRUCE ACKERMAN, WE THE PEOPLE: FOUNDATIONS (1991) [hereinafter ACKERMAN, FOUNDATIONS]. See also 2 BRUCE ACKERMAN, WE THE PEOPLE: TRANSFORMATIONS (1998).
(34.) ACKERMAN, FOUNDATIONS, supra note 33, at 6.
(35.) Id. at 7.
(36.) Ackerman describes the process in detail in Chapter 10. Id. at 266–94 (Chapter 10).
(38.) Id. at 7.
(39.) Id. at 6.
(40.) See, for example, the creative theory offered in JED RUBENFELD, FREEDOM AND TIME: A THEORY OF CONSTITUTIONAL SELF-GOVERNMENT (2001). For a powerful argument against such an approach (which does not, however, mention Rubenfeld’s theory in particular), see JEREMY WALDRON, LAW AND DISAGREEMENT 255–81 (1999) [hereinafter WALDRON, LAW AND DISAGREEMENT].
(41.) See, e.g., DUNCAN KENNEDY, A CRITIQUE OF ADJUDICATION (Fin de siècle) (1997); Girardeau A. Spann, Pure Politics, 88 MICH. L. REV. 1971 (1990).
(42.) See PETER FINLEY DUNNE, MR. DOOLEY’S OPINIONS 26 (1901) (“[N]o matter whether th’ constitution follows th’ flag or not, th’ supreme coort follows th’ iliction returns.”).
(43.) The patron saint of reductionism is the political scientist Robert Dahl. See Robert A. Dahl, Decisionmaking in a Democracy: The Supreme Court as a National Policy Maker, 6 J. PUB. L. 279 (1957). Other key works in this vein are David Adamany, Legitimacy, Realigning Elections, and the Supreme Court, 1973 WISC. L. REV. 790 (1973), and Richard Funston, The Supreme Court and Critical Elections, 69 AM. POL. SCI. REV. 795 (1975).
(44.) See, e.g., NEAL DEVINS & LOUIS FISHER, THE DEMOCRATIC CONSTITUTION (2004); NEAL DEVINS, SHAPING CONSTITUTIONAL VALUES: ELECTED GOVERNMENT, THE SUPREME COURT, AND THE ABORTION DEBATE (1996); LOUIS FISHER, CONSTITUTIONAL DIALOGUES: INTERPRETATION AS A POLITICAL PROCESS (1988); BARRY FRIEDMAN, THE WILL OF THE PEOPLE: HOW PUBLIC OPINION HAS INFLUENCED THE SUPREME COURT AND SHAPED THE MEANING OF THE CONSTITUTION (2009).
(45.) See Chisholm v. Georgia, 2 U.S. (2 Dall.) 419 (1793). The decision in Chisholm was overturned by the ratification of the Eleventh Amendment in 1795.
(46.) See Scott v. Sandford, 60 U.S. (19 How.) 393 (1857).
(47.) See, e.g., United States v. E.C. Knight Co., 156 U.S. 1 (1895) (applying the Commerce Clause to prevent the application of the Sherman Antitrust Act to a combination of sugar refineries); Lochner v. New York, 198 U.S. 45 (1905) (applying the Due Process Clause to invalidate a state regulation of the working hours of bakery employees); Hammer v. Dagenhart, 247 U.S. 251 (1918) (applying the Commerce Clause to invalidate a federal prohibition on the interstate sale of goods produced by child labor); Adkins v. Children’s Hosp., 261 U.S. 525 (1923) (applying the Due Process Clause to invalidate minimum-wage regulations in the District of Columbia); A.L.A. Schechter Poultry Corp. v. United States, 295 U.S. 495 (1935) (applying the Commerce Clause and general separation of powers principles to prevent the president from issuing “codes of fair competition” to regulate prices and trade practices in the poultry industry); Carter v. Carter Coal Co., 298 U.S. 238 (1936) (applying the Commerce Clause to prevent Congress from regulating labor standards and wages in the coal industry).
(48.) See Loving v. Virginia, 388 U.S. 1 (1967). Some commentators have noted the unpopularity of the Court’s decision in Loving. See David R. Stras & Ryan W. Scott, Retaining Life Tenure: The Case for a “Golden Parachute”, 83 WASH. U. L.Q. 1397, 1430–31 (2005) (noting that Loving v. Virginia “encountered fierce resistance when decided”); Kevin Noble Maillard, The Color of Testamentary Freedom, 62 SMU L. REV. 1783, 1793–94 (2009) (pointing out that even though Loving v. Virginia declared antimiscegenation laws unconstitutional in 1967, South Carolina’s prohibition on interracial marriage was “in its books until 1999”).
(49.) See Citizens United v. Fed. Election Comm’n, 130 S. Ct. 876 (2010). An ABC News/Washington Post poll conducted shortly after the decision suggested that roughly 80 percent of Americans opposed it, with broad agreement across the political spectrum. See Dan Eggen, Poll: Large Majority Opposes Supreme Court’s Decision on Campaign Financing, WASHINGTON POST, Feb. 17, 2010, http://www.washingtonpost.com/wp-dyn/content/article/2010/02/17/AR2010021701151.html?sid=ST2010021702073.
(50.) See U.S. CONST. amend. XI (denying federal jurisdiction over suits against states by citizens of other states or foreign nations); amend. XIII (outlawing slavery); amend. XIV, §1 (giving citizenship to former slaves and guaranteeing them due process, equal protection, and the “privileges” and “immunities” of citizenship).
(51.) For an excellent narrative of the Court’s longstanding resistance, and ultimate capitulation, to the Progressive Era and New Deal tide of economic regulation, see FRIEDMAN, supra note 44, at 167–236 (Chapters 6–7).
(52.) Madison, FEDERALIST NO. 10, supra note 4, at 125.
(53.) JOHN RAWLS, A THEORY OF JUSTICE 86 (1971) (“By contrast, pure procedural justice obtains when there is no independent criterion for the right result: instead there is a correct or fair procedure such that the outcome is likewise correct or fair, whatever it is, provided that the procedure has been properly followed.”).
(54.) This is the central theme of Waldron’s book Law and Disagreement. See generally WALDRON, LAW AND DISAGREEMENT, supra note 40; see also Waldron, Core Case, supra note 10, at 1371–76.
(55.) See WALDRON, LAW AND DISAGREEMENT, supra note 40, at 232–54 (Chapter 11); Waldron, Core Case, supra note 10, at 1371–73, 1375.
(56.) See WALDRON, LAW AND DISAGREEMENT, supra note 40, at 232–54 (Chapter 11); Waldron, Core Case, supra note 10, at 1386–89.
(57.) See WALDRON, LAW AND DISAGREEMENT, supra note 40, at 232–54 (Chapter 11); Waldron, Core Case, supra note 10, at 1386–95.
(58.) See Waldron, Core Case, supra note 10, at 1373–74.
(59.) RAWLS, supra note 53, at 85–86.
(60.) See Waldron, Core Case, supra note 10, at 1379–80. Although Waldron does not expressly cite Bickel, he can be understood as responding to an argument in favor of judicial review presented in BICKEL, supra note 1, at 26:
Another advantage that courts have is that questions of principle never carry the same aspect for them as they did for the legislature or the executive. Statutes, after all, deal typically with abstract or dimly foreseen problems. The courts are concerned with the flesh and blood of an actual case. This tends to modify, perhaps to lengthen, everyone’s view. It also provides an extremely salutary proving ground for abstractions; it is conducive, in a phrase of Holmes, to thinking things, not words, and thus to the evolution of principle by a process that tests as it creates.
(61.) See Waldron, Core Case, supra note 10, at 1380–86. These points too can be understood as responses to the defenses of constitutional law and judicial review offered by the likes of Bickel and Eisgruber.
(62.) Some arguments I made in Christopher J. Peters, Assessing the New Judicial Minimalism, 100 COLUM. L. REV. 1454, 1492–1513 (2000), suggest there is more to be said for judicial review in these respects than Waldron allows.
(63.) I suspect Waldron largely ignores the question of impartiality because his own allegiance is to a purely process-based or noninstrumental account of legal authority. See WALDRON, LAW AND DISAGREEMENT, supra note 40, at 249–54 (arguing against an instrumental account of the right of participation).
(64.) EISGRUBER, supra note 26, at 12 (citing STEPHEN HOLMES, PASSIONS AND CONSTRAINTS 167–69 (1995)).
(65.) CASS R. SUNSTEIN, DESIGNING DEMOCRACY: WHAT CONSTITUTIONS DO 98 (2001).
(66.) Waldron, Core Case, supra note 10, at 1361.
(67.) Here is Waldron’s full statement of his assumptions about the democratic process in question:
Id. (citations omitted).
I assume that the society we are considering is a democratic society and that, like most in the modern Western world, it has struggled through various forms of monarchy, tyranny, dictatorship, or colonial domination to a situation in which its laws are made and its public policies are set by the people and their representatives working through elective institutions. This society has a broadly democratic political system with universal adult suffrage, and it has a representative legislature, to which elections are held on a fair and regular basis. I assume that this legislature is a large deliberative body, accustomed to dealing with difficult issues, including important issues of justice and social policy. The legislators deliberate and vote on public issues, and the procedures for lawmaking are elaborate and responsible, and incorporate various safeguards, such as bicameralism, robust committee scrutiny, and multiple levels of consideration, debate, and voting. I assume that these processes connect both formally (through public hearings and consultation procedures) and informally with wider debates in the society. Members of the legislature think of themselves as representatives, in a variety of ways, sometimes making the interests and opinions of their constituents key to their participation, sometimes thinking more in terms of virtual representation of interests and opinions throughout the society as a whole. I assume too that there are political parties, and that legislators’ party affiliations are key to their taking a view that ranges more broadly than the interests and opinions of their immediate constituents.
(69.) See generally id.; WALDRON, LAW AND DISAGREEMENT, supra note 40, at 209–312 seriatim.
(70.) See Carolene Products, 304 U.S. at 152 n.4; ELY, supra note 16, at 103, 105–34.
(71.) Sedition Act of 1787, ch. 74, 1 Stat. 596 (1798) (expired 1801).
(72.) See Carolene Products, 304 U.S. at 152 n.4; ELY, supra note 16, at 135–79.
(73.) Waldron, Core Case, supra note 10, at 1372.
(74.) RAWLS, supra note 53, at 54–58.
(75.) See Waldron, Core Case, supra note 10, at 1368–69.
(76.) Madison, FEDERALIST NO. 10, supra note 4, at 319.
(77.) ELY, supra note 16, at 107 (emphasis in original). And remember that worries about the good faith of legislators, or of members of a voting majority of citizens, do not contradict our standing assumption that legal subjects will act as good-faith moral reasoners in deciding whether to obey the law. Legislators and voting citizens are not (in those capacities) legal subjects; they are rather legal officials, deciding how to exercise (or to claim to exercise) legal authority, not whether to obey it. (But legislators and voting citizens are legal subjects with respect to the commands of laws that purport to govern them in their capacities as legal officials—that is, with respect to constitutional law.)
(78.) See RAWLS, supra note 53, at 11–22.
(79.) For an account of the events surrounding the adoption, enforcement, and eventual abandonment of the Sedition Act, with particular attention to the free-speech issues involved, see GEOFFREY R. STONE, PERILOUS TIMES: FREE SPEECH IN WARTIME: FROM THE SEDITION ACT OF 1798 TO THE WAR ON TERRORISM 15–78 (2004).
(80.) Madison, FEDERALIST NO. 51, supra note 4, at 319.
(82.) See ELY, supra note 16, at 105–34.
(83.) The Framers were quite explicit about their belief that dividing power would reduce the risk of agency failure. See, e.g., Madison, FEDERALIST NO. 51, supra note 4, at 321 (describing the combination of separated national powers and federalism as “a double security… to the rights of the people”).
(84.) Madison, FEDERALIST NO. 10, supra note 4, at 123.
(85.) Madison, FEDERALIST NO. 10, supra note 4, at 125.
(86.) Justice Antonin Scalia, in an interesting discussion in his opinion for the Court in Republican Party of Minnesota v. White, 536 U.S. 765 (2002), drew a similar distinction between different possible meanings of the concept of judicial “impartiality.” Impartiality, Justice Scalia speculated, might mean “the lack of bias for or against either party to the proceeding.” Or it might mean a “lack of preconception in favor of or against a particular legal view.” Id. at 775–78.
(87.) For a narrative of dairy-industry influence in legislation during the early twentieth century, see Geoffrey Miller, Public Choice at the Dawn of the Special Interest State: The Story of Butter and Margarine, 77 CAL. L. REV. 83 (1989).
(89.) This realization—that animosity toward a group with an interest in an issue destroys impartiality, regardless of which resolution of the issue is proper—probably motivated the Supreme Court’s reliance on the concept of animosity in deciding at least one recent case involving the rights of homosexuals. In Romer v. Evans, 517 U.S. 620 (1996), the Court held that a Colorado constitutional amendment excluding homosexuals from antidiscrimination laws could not survive “rational basis” scrutiny under the Equal Protection Clause because it was “born of animosity toward” homosexuals. Id. at 634–35. In rejecting “a bare… desire to harm a politically unpopular group” as a legitimate governmental interest, the Romer Court drew on earlier equal protection decisions involving discrimination against “hippies” (U.S. Dept. of Agriculture v. Moreno, 413 U.S. 528 (1973)) and people with mental disabilities (City of Cleburne v. Cleburne Living Center, Inc., 473 U.S. 432 (1985)). Id. Shades of this reasoning also appeared in the Court’s decision invalidating a Texas prohibition of same-sex sodomy under the Due Process Clause. See Lawrence v. Texas, 539 U.S. 558, 574–75 (2003) (citing Romer’s anti-animosity principle). Justice O’Connor relied explicitly on that rationale in concurring in Lawrence on equal protection grounds. See id. at 579–82 (O’Connor, J., concurring). See also infra note 112 and accompanying text.
(90.) Madison, FEDERALIST NO. 10, supra note 4, at 125.
(91.) Classic studies of minority interest-group influence in democratic politics include ARTHUR F. BENTLEY, THE PROCESS OF GOVERNMENT: A STUDY OF SOCIAL PRESSURES (1908); E. E. SCHATTSCHNEIDER, POLITICS, PRESSURES, AND THE TARIFF (1935); DAVID B. TRUMAN, THE GOVERNMENTAL PROCESS: POLITICAL INTERESTS AND PUBLIC OPINION (1951); EARL LATHAM, THE GROUP BASIS OF POLITICS: NOTES FOR A THEORY (1952); ROBERT A. DAHL, WHO GOVERNS? DEMOCRACY AND POWER IN AN AMERICAN CITY (1961); and KAY LEHMAN SCHLOZMAN & JOHN T. TIERNEY, ORGANIZED INTERESTS AND AMERICAN DEMOCRACY (1986). The conventional economic theory explaining minority political influence was first articulated in MANCUR OLSON, THE LOGIC OF COLLECTIVE ACTION: PUBLIC GOODS AND THE THEORY OF GROUPS (1965). A good critical overview of interest-group theory can be found in DANIEL A. FARBER & PHILIP P. FRICKEY, LAW AND PUBLIC CHOICE: A CRITICAL INTRODUCTION 12–37 (1991).
(92.) This hypothetical is based on the actual rise of the dairy industry as a powerful force in state and, eventually, national politics in the late nineteenth and early twentieth centuries. The rise is documented in Miller, supra note 87.
(93.) This is a simplified version of the state apportionment schemes struck down by the Supreme Court in Gray v. Sanders, 372 U.S. 368 (1963), and Reynolds v. Sims, 377 U.S. 533 (1964).
(94.) There is a long and sordid history of gerrymandering in the United States. The term “gerrymander” itself derives from a districting bill signed by Massachusetts governor Elbridge Gerry in 1812, which heavily favored Gerry’s own Democratic-Republican party and included a district whose shape resembled that of a salamander. See Davis v. Bandemer, 478 U.S. 109, 164 & n.3 (1986) (Powell, J., concurring in part and dissenting in part). In Gomillion v. Lightfoot, 364 U.S. 339 (1960), the Supreme Court held that Alabama’s redrawing of the Tuskegee city boundaries to form a 28-sided figure that excluded almost all black voters violated the Equal Protection Clause. In League of United Latin Am. Citizens v. Perry, 548 U.S. 399 (2006), the Court dismissed on procedural grounds most of an equal-protection challenge to Texas’s redrawing of federal congressional districts, the result of which strongly favored the Republican party, which controlled the state legislature and held the governorship during the redistricting process.
(95.) After ratification in 1870 of the Fifteenth Amendment, which protected the right to vote of newly freed slaves, many southern states adopted poll taxes with “grandfather clauses” exempting citizens eligible to vote before slavery was abolished (i.e., white citizens). See 2 A. LEON HIGGINBOTHAM, JR., SHADES OF FREEDOM: RACIAL POLITICS AND PRESUMPTIONS OF THE AMERICAN LEGAL PROCESS 172–74 (1996); see also South Carolina v. Katzenbach, 383 U.S. 301, 310–14 (1966). Many of the same states later added literacy requirements, as did some northern states with large immigrant populations. See HIGGINBOTHAM, supra, at 178–79; Katzenbach, 383 U.S. at 312. The combined effect of poll taxes and literacy tests disenfranchised large portions of the southern black population until well into the mid-twentieth century. See HIGGINBOTHAM, supra, at 178–79. In 1964 the Twenty-Fourth Amendment, which banned poll taxes in federal elections, was ratified, and in 1966 the Supreme Court invalidated poll taxes in state elections as violations of equal protection. See Harper v. Virginia Bd. of Elections, 383 U.S. 663 (1966). The Voting Rights Act of 1965 effectively eliminated the use of literacy tests; its validity was upheld by the Supreme Court in South Carolina v. Katzenbach, supra, and Katzenbach v. Morgan, 384 U.S. 641 (1966).Recently some voting-rights advocates have argued that state voter-identification requirements amount to a functional poll tax by effectively limiting access to the polls by many impoverished citizens. In 2008, the Supreme Court rejected a facial equal-protection challenge to Indiana’s voter-identification requirement, although the Court left open the possibility that such requirements might be challenged by individual voters who could establish harm. See Crawford v. Marion County Election Bd., 128 S.Ct. 1610, 1621–23 (2008).
(96.) “Rational basis” scrutiny under both the Due Process and Equal Protection Clauses requires that the government demonstrate a “rational relationship” between the law at issue and some “legitimate” objective. Most laws survive rational-basis scrutiny. See, e.g., United States v. Carolene Products Co., 304 U.S. 144 (1938) (upholding a federal prohibition on the shipment of “filled milk” against a due process challenge despite obvious special-interest influence); Williamson v. Lee Optical, 348 U.S. 483 (1955) (upholding a state law requiring prescriptions for duplicating lenses against both due process and equal protection challenges); see also Geoffrey Miller, The True Story of Carolene Products, 1987 SUP. CT. REV. 397 (portraying the Filled Milk Act as the product of special-interest rent-seeking) [hereinafter Miller, Carolene Products]. Some do not survive it, however. See sources cited supra note 89.
(97.) “Strict scrutiny” under due process and equal protection requires the government to demonstrate that a law is “narrowly tailored” to serve a “compelling” objective. Equal protection strict scrutiny is triggered by laws that discriminate based on race, ethnicity, or national origin; due process strict scrutiny is triggered by laws that infringe some “fundamental interest.” Most laws fail strict scrutiny. See, e.g., Gratz v. Bollinger, 539 U.S. 244 (2003) (holding that a public university’s program of affirmative action in admissions was insufficiently narrowly tailored to survive strict scrutiny under the Equal Protection Clause); Troxel v. Granville, 530 U.S. 57 (2000) (holding that a state law allowing visitation by nonparents in “the best interest of the child” lacked a compelling interest and thus failed strict scrutiny under the Due Process Clause). Some survive it, however. See Grutter v. Bollinger, 539 U.S. 306 (2003) (upholding a public law school’s program of affirmative action in admissions as sufficiently narrowly tailored to survive strict scrutiny under the Equal Protection Clause).
The Court also has identified another form of heightened scrutiny under the Equal Protection Clause, typically referred to as “intermediate scrutiny,” which requires the government to demonstrate that a law is “substantially related” to an “important” government interest. Intermediate scrutiny so far has been limited to laws that classify based on gender, and the results have been mixed, although most gender classifications have been struck down. See, e.g., Craig v. Boren, 429 U.S. 190 (1976) (applying intermediate scrutiny to invalidate an Oklahoma law prohibiting males but not females to purchase 3.2 percent beer at age eighteen); United States v. Virginia, 518 U.S. 515 (1996) (applying intermediate scrutiny to invalidate a Virginia policy of excluding women from its state military college); but see Nguyen v. INS, 533 U.S. 53 (2001) (applying intermediate scrutiny to uphold immigration rules favoring mothers over fathers).
(98.) See Church of Lukumi Babalu Aye v. City of Hialeah, 508 U.S. 520 (1993) (invalidating a city’s ban on the slaughtering of animals on the ground that it was intended to prohibit the practices of the Santeria religion).
(99.) While the original Constitution was notably silent on the question of who could vote, six subsequent amendments had the direct or indirect effect of expanding the franchise. The Fifteenth Amendment (ratified in 1870) prohibited impairment of the right to vote “on account of race, color, or previous condition of servitude.” U.S. CONST. amend. XV, § 1. The Seventeenth Amendment (ratified in 1913) replaced the original scheme of Senate appointments by the state legislatures with a system of direct popular election. Id. amend. XVII. The Nineteenth Amendment (1920) gave women the vote. Id. amend. XIX. The Twenty-Third Amendment (1961) assigned presidential Electors to the District of Columbia, thus effectively allowing D.C. residents to vote in presidential elections (though they still have no voting representatives in Congress). Id. amend. XXIII. The Twenty-Fourth Amendment (1964) eliminated use of the poll tax in federal elections, removing a powerful tool for disenfranchising the poor and especially African Americans; the Supreme Court quickly extended its principle to state and local elections. Id. amend. XXIV; see also Harper v. Virginia State Bd. of Elections, 383 U.S. 663 (1966) (invalidating a state poll tax on equal protection grounds). And the Twenty-Sixth Amendment (1971) effectively lowered the minimum voting age in both federal and state elections to a uniform eighteen years. U.S. CONST. amend. XXVI.
Article VI expressly bars “religious Test[s]” as requirements for federal office. U.S. CONST. art. VI. There is little doubt that state and local religious tests are banned by virtue of the Free Exercise and Establishment Clauses of the First Amendment, or that the Equal Protection Clause subjects discrimination among citizens for purposes of voting or holding office to strict scrutiny. See McDaniel v. Paty, 435 U.S. 618 (1978) (holding a state constitutional provision which prohibited ministers or priests from serving in the legislature violated the Free Exercise Clause of the First Amendment as applied to the states by the Fourteenth Amendment); Hill v. Stone, 421 U.S. 289 (1975) (striking down a state provision limiting the right to vote in city bond issue elections to persons who had “rendered” property for taxation as a violation of the Equal Protection Clause); Kramer v. Union Free Sch. Dist., 395 U.S. 621 (1969) (invalidating under strict scrutiny a state requirement that voters in school district elections either own or lease taxable real property within the relevant district or have children attending the local public schools); Harper, 383 U.S. 663 (1966) (invalidating under strict scrutiny a state poll tax); cf. Sugarman v. Dougall, 413 U.S. 634 (1973) (invalidating under strict scrutiny a state law prohibiting noncitizens from holding low-level civil-service positions).
(100.) Since at least the 1960s, the Court fairly consistently has used the Free Speech Clause of the First Amendment to protect expression that communicates a political message. See, e.g., New York Times v. Sullivan, 376 U.S. 254 (1964) (requiring that public officials demonstrate “actual malice” to recover for defamation); Brandenburg v. Ohio, 395 U.S. 444 (1969) (requiring government to meet a stringent test before regulating speech on the ground that it might incite violence); Cohen v. California, 403 U.S. 15 (1971) (invalidating a conviction for disturbing the peace of a defendant who wore a jacket reading “Fuck the Draft” in a courthouse); Buckley v. Valeo, 424 U.S. 1 (1976) (applying strict scrutiny to laws regulating political campaign expenditures); Texas v. Johnson, 491 U.S. 397 (1989) (applying strict scrutiny to invalidate a state law prohibiting burning the U.S. flag); R.A.V. v. City of St. Paul, 505 U.S. 377 (1992) (applying strict scrutiny to content-based bans on “fighting words” that carry certain messages); but see United States v. O’Brien, 391 U.S. 367 (1968) (applying intermediate scrutiny to uphold a statute criminalizing the destruction of draft-registration cards as applied to a defendant who publicly burned his card in protest).
(101.) The Court’s first “substantive” use of the Due Process Clause occurred in Dred Scott v. Sandford, 60 U.S. (19 How.) 393 (1857), where it suggested that Congress’s attempt to ban slavery in parts of the territories deprived slaveowners of their property without due process of law.
(102.) See, e.g., Adkins v. Children’s Hosp., 261 U.S. 525 (1923) (using due process to invalidate a state law requiring minimum wages for women); Lochner v. New York, 198 U.S. 45 (1905) (using due process to invalidate a state limitation on the working hours of bakers).
(103.) See, e.g., United States v. Carolene Products Co., 304 U.S. 144 (1938) (upholding against a due process challenge a federal prohibition on the shipment of “filled milk” in interstate commerce); West Coast Hotel Co. v. Parrish, 300 U.S. 379 (1937) (overruling Adkins to uphold a minimum wage for women); Nebbia v. New York, 291 U.S. 502 (1934) (upholding price controls for milk against a due process challenge).
(104.) See, e.g., Griswold v. Connecticut, 381 U.S. 479 (1965) (invalidating a law criminalizing the use of contraception by married couples as violating a right of “privacy” gleaned from “penumbras” emanating from several provisions of the Bill of Rights); id. at 499–501 (Harlan, J., concurring) (concurring in the judgment on substantive due process grounds); Roe v. Wade, 410 U.S. 113 (1973) (invalidating state laws prohibiting abortion as violating the “liberty” component of substantive due process); Lawrence v. Texas, 539 U.S. 558 (2003) (invalidating a state law criminalizing same-sex sodomy as violating a right to “autonomy” protected by substantive due process).
(105.) Roe, 401 U.S. at 153.
(106.) See, e.g., Lochner, 198 U.S. at 45.
(107.) See, e.g., Griswold, 381 U.S. at 479; Lawrence, 539 U.S. at 558. Although the majority opinion in Griswold did not expressly rely on the Due Process Clause, I think it is fair to treat the case as involving substantive due process, for reasons I explain in note 112, infra.
(108.) See sources cited supra note 102.
(109.) As David Bernstein points out, the statutory provisions at issue in Lochner—limitations on the working hours of bakery employees—can be understood as the products of rent-seeking by politically powerful large bakeries and bakery unions, at the expense of less-powerful smaller bakeries and nonunionized (and largely immigrant) bakery workers and of disorganized consumers. See David E. Bernstein, The Story of Lochner v. New York: Impediment to the Growth of the Regulatory State, in CONSTITUTIONAL LAW STORIES 326, 328–35 (Michael C. Dorf ed., 2004). If this view is accurate, then the Court’s close scrutiny of that and similar laws often had the effect of frustrating such rent-seeking behavior—and the Court’s abandonment of close scrutiny during the New Deal had the opposite effect. For a rent-seeking account of the Filled Milk Act upheld in Carolene Products, a central New Deal decision, see Miller, Carolene Products, supra note 96.
(110.) This is the central insight of a number of recent scholarly attempts to “rewrite” Roe as an equal-protection decision. See, e.g., WHAT ROE V. WADE SHOULD HAVE SAID 31, 42–45 (Jack M. Balkin ed., 2005) (“opinion” of Jack M. Balkin); id. at 63, 63–82 (“opinion” of Reva B. Siegel). Hints of it also appear in the Court’s Casey decision upholding the “central holding” of Roe. See Planned Parenthood v. Casey, 505 U.S. 833, 835 (1992) (“The ability of women to participate equally in the economic and social life of the Nation has been facilitated by their ability to control their reproductive lives.”).
(111.) See generally Griswold, 381 U.S. at 485–86. In fact Justice William O. Douglas’s opinion for the Court in Griswold expressly eschewed reliance on substantive due process in favor of a theory involving “penumbras” of privacy emanating from a number of provisions of the Bill of Rights. See id. at 481–82. Only the concurrences of Justices John Marshall Harlan and Byron White relied explicitly on substantive due process. See id. at 500, 503–04 (Harlan, J., concurring) (White, J., concurring). By virtue of its precedential influence in subsequent substantive due process cases, however, Griswold is now almost universally considered a de facto substantive due process decision. See, e.g., Roe v. Wade, 410 U.S. 113, 129,155 (1973); Bowers v. Hardwick, 478 U.S. 186, 190–91 (1986); Washington v. Glucksberg, 521 U.S. 702, 720 (1997); Casey, 505 U.S. at 848–49; Lawrence, 539 U.S. at 564–66.
(112.) This way of understanding Lawrence illuminates the majority’s caginess in that decision regarding the level of scrutiny it was applying. Much of Justice Anthony Kennedy’s opinion for the Court was devoted to undermining the Court’s earlier holding in Bowers v. Hardwick, 478 U.S. 186 (1986), that same-sex intimacy was not a “fundamental right” and thus was undeserving of strict scrutiny pursuant to the Due Process Clause. See Lawrence, 539 U.S. at 567–68. But the Lawrence Court never actually declared that a “fundamental right” was involved in that case and thus never overtly applied strict scrutiny; instead it held that the Texas statute involved “furthers no legitimate state interest,” thus employing the language of “rational basis” scrutiny. See id. at 578.
Justice Kennedy’s opinion also made a point of citing Romer v. Evans, 517 U.S. 620 (1996), in which the Court had held that a Colorado ban on laws protecting homosexuals against discrimination was motivated by “animosity” toward homosexuals and thus could not survive rational-basis scrutiny under the Equal Protection Clause. See id. at 574–76. If we interpret Lawrence (and perhaps Romer) as responses to the problem of faction, the Court’s use of rational-basis scrutiny to invalidate the laws in question—sometimes called “rational basis review with teeth”—makes some sense. On this view, the problem in Lawrence and Romer was that the political process had been compromised by bias against a particular minority group or prejudice with respect to the issues involved, in a way analogous to rent-seeking by economically motivated special interests.
(113.) Troxel v. Granville, 530 U.S. 57, 60, 65 (2000).
(114.) See Pierce v. Soc’y of Sisters, 268 U.S. 510 (1925) (invalidating a state law requiring that children be educated in public schools); Meyer v. Nebraska, 262 U.S. 390 (1923) (invalidating a state law prohibiting the teaching of foreign languages to children). We might classify Brown v. Board of Education as a response to similar concerns, except that enforced racial segregation in public schools fosters not homogeneity, but rather a sense of inferiority in students of the minority race and of superiority in students of the majority race (and thus a preordained partiality in the political process in which those students eventually will take part).