Printer Friendly

Deliberative autonomy and legitimate state purpose under the First Amendment.

The right to free speech is usually thought of as rooted in the autonomy of the speaker: An autonomous individual may speak freely and the state may not stop her. Some commentators have found such an autonomy-based view of our free speech rights abstract and divisive. Turning instead to the interests of the audience, they start with our commitment to democracy and regard the right to free speech as instrumental to our democratic interest in open public dialogue. (1) I argue here that free speech law does--and should--revolve around an alternative organizing principle: the autonomy of the listener.

The particular conception of autonomy I advance here is deliberative autonomy. Deliberative autonomy looks to the purposes of state regulation and asks whether they are consistent with the liberal conception of the person that underlies the constitution. In the context of free speech, deliberative autonomy precludes the government from denying citizens access to others' speech based on presumptions about how listeners will react. This point is especially important today, when it is tempting to shut down the speech of fundamentalists who do not believe in the constitutional rights that would protect them. (2) The reason that the United States government may not prevent such individuals from proclaiming the evils of a religion, a society, or a nation lies neither in their right to express hatred nor in the public interest in hearing or 'airing' such hatred. Rather, the government must allow speakers to express disdainful and even indirectly dangerous views because to disallow speech out of fear that the speech will be persuasive violates the deliberative autonomy of those whom censorship aims to protect.

Deliberative autonomy is deliberative not in the general sense that government should not interfere with deliberation, but in the more specific sense that the government may not act in a manner that denies our capacity to reason and deliberate. (3) As a consequence of its nature as a demand for recognition, in the first instance, deliberative autonomy constrains state purpose. State purpose is not identical with the reasons that a government gives a court; it must be inferred from the social meaning of the state's justifications.

I begin in Part I by developing the concept of deliberative autonomy and contrasting it with alternative conceptions of autonomy that have been proposed under the First Amendment. In Part II, I show how deliberative autonomy already manifests itself in free speech doctrine. Part III explores the relationship between deliberative autonomy and democracy, and argues that, contrary to theories that treat autonomy as merely instrumental for democracy, autonomy motivates democracy and is in that sense prior to it. In Part IV, I defend the focus on state purpose and advocate the interpretation of state purpose by way of social meaning. State purpose is essential to a number of constitutional doctrines, including free speech, but the interests of autonomy are best served where, for historical reasons, the Court is most sensitive to the social meaning of state action. Part V responds to two frequent challenges to autonomy-based theories of the First Amendment, by showing how, because deliberative autonomy is only a regulative principle, (4) it depends neither on empirical conditions nor on a holistic moral philosophy. Finally, in Part VI, I discuss how the application of the principle of deliberative autonomy can place negative and positive burdens on the state.


Joel Feinberg identified four types of autonomy: first, "the capacity to govern oneself"; second, "actual condition[s] of self-government"; third, "an ideal of character"; and fourth, "sovereign authority." (5) An authority conception of autonomy may derive from a view about our capacities and virtues, but as a principle, it asserts a kind of jurisdictional right of persons over certain aspects of themselves. Deliberative autonomy is such an authority conception of autonomy.

The concept of deliberative autonomy is familiar: Individuals should be treated in a manner consistent with their capacity to think for themselves. As a general moral principle, deliberative autonomy generates a number of interpersonal duties--between friends, between parents and children, as well as between strangers--that demarcate the boundaries of acceptable paternalism. Although the conception of deliberative autonomy employed here refers not to the positive capacity of individuals to think for themselves, but instead to the normative duty to respect their right to do so, the plausibility (though not the validity) of deliberative autonomy as a moral principle probably hinges on our every-day recognition that people exercise reason and that they expect--and are expected--to do so. Just as we justify the imposition of a number of legal and social responsibilities on persons as a result of the presumption that they are capable of deliberation, the presumption generates a number of rights. The right to free speech under the First Amendment is one such right. The likely origins of the principle of deliberative autonomy in each of our insistence on our own capacity for reason may lead to the question of why we are obligated, in our capacity as citizens, to presume the deliberative capacity in others. Indeed, those most confident about their own deliberative capacities sometimes appear most skeptical about the capacities of others.

The obligation to recognize deliberative capacity in others results from the fact that in our capacity to deliberate lies the ability to confer value, and our democratic form of government presupposes citizens' capacity to confer value. The capacity to deliberate is not constitutionally important because it is the means by which we access objective truth; without a common notion of objective truth we have little public reason to believe that we or any agreed group of individuals is able to access objective truth through public or private deliberation. Instead of objective truth, it is disagreement and separate commitment to those values over which we disagree which endure as "permanent condition[s] of democratic politics." (6) What we know is that we each do value ideas--and from the point of view of those who do not value the same ideas, we confer value on them.

From the public perspective, the common capacity to assign value is both the source of disagreement and the basis for mutual recognition. The value we attach to certain ideas, from the point of view of other citizens, is not an objective, independent fact. The presumption of deliberative autonomy involves a certain concept of specifically public or political value. From the public perspective, it does not matter whether ideas are valued by citizens because they are already valuable in some other sense. Ideas are valuable because they are valued by individuals with the authority to confer political value, i.e. citizens. Recognition of this authority to confer value through deliberation is the basis for one's status as an equal member of the political community. It is also the basis for democratic government, which presumes that citizens can confer authority on the state. (7) The political aspect of our moral capacity for deliberation is not the most important aspect of the self for many people, but it is something that we must expect to secure though the basic structure of society if we are at once to recognize political authority, and to recognize ourselves as moral agents. (8) The presumptions imposed on the government by free speech law help constitute the individual qua citizen. The First Amendment prevents the state from exercising its regulatory authority in a manner inconsistent with the conception of the person underlying its democratic form.

A. Ambiguity in the Principle of Autonomy

In Stanley v. Georgia, the Supreme Court observed that "[o]ur whole constitutional heritage rebels at the thought of giving government the power to control men's minds." (9) Most Americans probably can agree that there are limits to government power over individuals; what lies within the boundaries that government cannot cross constitutes our autonomy. Despite abiding popular support for some notion of autonomy, and likely popular agreement on the centrality of autonomy to constitutional constraints on government, the precise content of this aspect of our constitutional heritage--or the best contours of autonomy as a constitutional principle--are deeply contested.

Free speech doctrine could interpret the principle of autonomy in two importantly distinct ways. The First Amendment may be read to prohibit regulation that has the effect of controlling people's minds. Alternatively, it could reject only (or even) those regulations that have the purpose of controlling people's minds. For those who see free speech and autonomy as mere instruments in the service of democracy, the choice is clear: "what matters is not what the agency is trying to do but what it has in fact done." (10) The Millian version of the autonomy principle also would focus free speech law on the effects of speech regulation. Mill celebrates our deliberative capacities and encourages people to cultivate the "moral courage of the human mind." (11) The doctrine of free speech makes possible "an intellectually active people." (12) Mill's emphasis on the negative effects of speech regulation on intellectual life is consistent with his utilitarian philosophy. (13)

But even avowed Kantians, though they will add that our capacity to think freely is basic to moral agency, do not consistently hold that the harm in speech regulation results not from the fact of restraint but from the presumption from which the state acts. For example, Thomas Scanlon has forcefully defended deliberative autonomy as a jurisdictional limit on state authority. (14) He further held that "the authority of governments to restrict the liberty of citizens in order to prevent certain harms does not include authority to prevent these harms by controlling people's sources of information to insure that they will maintain certain beliefs." (15) But Scanlon did not make clear whether the government is prohibited from using certain methods or whether the government cannot try to do certain things. That choice depends on whether the goal of free speech norms should be to protect opportunity for deliberation for the sake of its private moral significance, or whether the mere attempt to intervene in deliberation--even where it does not actually deprive anyone of moral agency--violates an independent (if related) political-moral principle. (16) Even under his Kantian interpretation of the Millian principle, it is not clear whether Scanlon would limit de facto cognitive interference or only (or even) limit the attempt to extend the scope of government authority to our minds.

Another sophisticated theorist of autonomy, Joseph Raz, also looks to the effects of speech regulation on individual realization of autonomy. Raz locates the harm of government regulation of speech in the attitude it expresses toward individuals, but is concerned only with the government's attitude toward individual conceptions of the good--not the government's attitude toward individuals' underlying capacity to endorse one conception of the good over others (17) Because a ban on speech condemns a way of life associated with that speech, and because we have a "fundamental need for public validation of [our] way of life," Raz identifies the harm from regulation as the actual effect that denial of recognition has on individuals. (18) Individuals whose lives are condemned through speech regulation will feel alienated from society and "[i]t is normally vital for personal prosperity that one will be able to identify with one's society." (19) According to Raz, the government should respect ways of life so as to maximize the possibilities available to all persons. (20) Such an operative principle would enhance the achievement of autonomy as an ideal; but it too has little to do with respecting deliberative autonomy as sovereignty. (21)

Freedom of speech doctrine can be unduly effects-oriented not only when it concerns itself with promoting autonomy as a way of life, a la Raz, but also when it protects speech categorically without regard for state purpose. The Court sometimes talks as though it is concerned with preserving expression per se, as in Texas v. Johnson, where it states that the first question to be answered must be whether the defendant was engaged in "expressive conduct." (22) Brennan, in his dissent in Columbia Broadcasting v. DNC, admonishes the Court for insufficient attention to "the independent First Amendment interest of groups and individuals in effective self-expression," (23)

Edwin Baker defends the notion of autonomy as uninhibited self-expression, maintaining that speech should be protected when it is an expression of the speakers' individual autonomy. (24) He describes even the listener's constitutional interest as one of self-expression:
 By outlawing the speech, the law does not treat the listener
 as an agent presumptively capable of responsible choice. But
 to decide how to act is self-defining. Restricting access to the
 speech paternalistically denies her the opportunity to obtain
 perspectives relating to how she will define herself, even in
 dissent from the law. (25)

Although Baker articulates the wrongful presumption in speech regulation, it is not the state's presumption in such cases that leads him to condemn the regulation; he is concerned instead with the effect of regulation on opportunities for expression.

Baker's view cannot explain the boundaries of free speech. The right to exposure to different views is imperfect. There are any number of things the government does and fails to do that reduce the views to which we are exposed. Baker's principle does not explain how we can label some action or inaction unconstitutional while accepting others. Interpreting the principle of autonomy as the right not to have the government act in denial of our presumptive capacity for reason, because it is more specific, is better able to distinguish between acceptable and unacceptable regulation of speech.

Notably, each of the last three theories of speech considered here is committed to a principle of autonomy. None is ultimately consequentialist. Scanlon, Raz and Baker are all concerned with the effects of public regulation on private moral agency and are motivated more broadly by classically nonconsequentialist moral theory. Moreover, each would probably agree to the moral significance of something like deliberative autonomy. But while each would expect the state to promote or secure that autonomy, the principle of individual autonomy operates only as a reason for action or inaction for the state, not a constraint on the totality of reasons. Their theories of free speech are effects-oriented at the state-level because each looks to the effects of, rather than the purpose behind, state action. Under those views, deliberative autonomy does not limit state purpose per se.

Nevertheless, the idea that we object to the state's reasons for certain kinds of regulation is intuitive and widespread. (26) Fiss, who explicitly opposes autonomy-based free speech norms and purpose-based free speech norms in particular, observes that the silencing that results from allocation of a subsidy is different from the silencing that results from a ban. He observes that "silencing is a necessary concomitant of every allocative decision." (27) But this fact is relevant precisely because an individual denied the subsidy allocated through a 'neutral' process normally has no grounds to think that the government acted from an illicit motive. For example, in the case of subsidies for artists, if an individual believes that the government acted in order to favor some cultural values and artistic views over others, then the individual would feel wronged. (28) The wrong would lie not in the actual deprivation of government funds, which would be no different whatever the government's reasons for allocating funds elsewhere, but would lie in the government's usurpation of citizens' right to value various artistic views for themselves. (29) Whether individual lawmakers intend such a meaning or not, systematically favoring certain aesthetic theories in the allocation of artistic subsidies assumes that the government has authority to assess the merits of those theories before they reach us. It is the assertion of such an authority over the deliberative process that wrongs us. (30)

Deliberative autonomy is a moral principle. Although it rests on certain background views about the capacity of individuals to reason and deliberate, the thrust of the principle is not to promote the cultivation of these capacities as a political or ethical ideal, but to present them as a moral fact about ourselves with consequences for how we must treat each other, and therefore, how the government must treat us. This is not to conflate individual moral principles, which in the Kantian tradition at least, emphasize motive, with the moral-political principles that should regulate state behavior. Even a thin liberal conception of the person is concerned with mutual recognition, and in the context of evaluating state action, with state purpose.


Deliberative autonomy illuminates certain aspects of current First Amendment doctrine. It helps to explain why the Court distinguishes between incitement and mere advocacy; (31) "fighting words" and offensive speech; (32) content and noncontent regulation; (33) primary and secondary effects; (34) picketing and handbilling; (35) and regulation of commercial advertising designed to restrict deceptive marketing versus a total ban on truthful advertising. (36) There are other explanations for each of these parts of the law, but deliberative autonomy provides a consistent and compelling justification for each.

The "clear and present danger" test was an attempt to narrow the conditions under which the government can assume that people are predictable: only when no countervailing discussion can prevent other ills that the government is responsible to prevent. (37) The government may not ban speech because it expresses a worthless idea, or even an idea antithetical to democratic government. Mere advocacy is harmless unless an audience is persuaded, and the government may not judge an idea valueless that at least some citizen believes to be of value. Similarly, words cannot be banned because they are insulting or offensive, but they can be banned because the insult or offense cannot but result in violence, i.e. they are fighting words. (38) The court is reluctant to license the government to preempt the speaker and listener's own valuations of speech. Only where the presumption of deliberative autonomy is incompatible with other fundamental principles can it be given up. (39) The Court will sanction the denial of the listener and speaker's right to ascertain for themselves the value of speech through their own deliberative capacities only when outweighed by some other immediate harm.

While the essence of the clear and present danger and fighting words doctrines reflects an effort to respect deliberative autonomy at all but the highest costs, application of the doctrines often results in something less than the highest degree of tolerance. What is labeled incitement or fighting words ultimately depends on the government's assessment of the risk of violence. Justice Holmes recognized the difficulty of drawing a line in his dissent in Gitlow v. New York when he wrote that "[t]he only difference between the expression of an opinion and an incitement in the narrower sense is the speaker's enthusiasm for the result." (40) Even though the very concept of a mob denies individual reflective capacities, the government regularly restricts speech made to groups of emotional people, characterized as mobs, in part for fear it would be blamed for failure to prevent avoidable violence if it did not. (41)

A similar combination of a noble theoretical commitment to deliberative autonomy and practical erosion of that commitment exists in the doctrine of fighting words. Chaplinsky v. New Hampshire held that certain harsh but clearly political insults to a city marshal "are no essential part of any exposition of ideas, and are of such slight social value as a step to truth that any benefit that may be derived from them is clearly outweighed by the social interest in order and morality." (42) Although deliberative autonomy allows for an analytic distinction between mere insults and fighting words, the court's reasoning seems to draw on its own ideas about what speech is valuable and what mode of communication is preferable. As stated in American Booksellers Ass'n v. Hudnut, "[a] power to limit speech on the ground that truth has not yet prevailed and is not likely to prevail implies the power to declare truth"; (43) the Court's characterization of fighting words sometimes risks the exercise of such power. The particular insults delivered in Chaplinsky, for example, were ones we could expect a mature government official to ignore. (44) Nevertheless, the rule given by the Court ostensibly covers only those words so viscerally hateful that they are more like a physical assault than a content-based insult. To draw a line makes sense even if the line the Court actually draws in particular cases does not.

Deliberative autonomy also makes an appearance in the central distinction between content and noncontent regulation. Under United States v. O'Brien, noncontent regulation is subject only to the substantial interest test. (45) Regulation qualifies for this test if, inter alia, the government has a substantial interest in regulation that is unrelated to the content of speech. (46) The effect on speech may be no less than it would be if the sole purpose were to ban the speech. The standard of scrutiny turns on whether the regulation is directed at the communicative impact of speech. This captures the principle that the government does not have the right to intervene in an individual's valuation of an idea.

Again, the application of the concept leaves something to be desired. In the principal case, O'Brien, the substantial interest given does not seem substantial at all. While the threat of widespread disobedience may be a greater threat than missing draft cards, the former only results if others decide that they too will refuse to cooperate. The government cannot preempt their deliberation. Moreover, it is not clear why recognizing that there is a deeper problem with content regulation than with noncontent regulation entails such little protection for the latter--especially in light of the difficulty in ascertaining the substantiality of government purpose. Nevertheless, at the doctrinal level, deliberative autonomy explains the court's special wariness about content regulation.

Similarly, deliberative autonomy explains the Court's distinction between primary and secondary effects, (47) which saved the regulation of adult films at issue in City of Renton v. Playtime Theatres, Inc. (48) The Court allowed the government to act on the grounds that adult film theaters produced a range of negative secondary effects in their neighborhoods. (49) The Court seemed satisfied that it was not the city council who fled the area and drove down property values; the government did not substitute its own judgment about the films for those of individuals. (50) However, this is an especially problematic application of deliberative autonomy since the secondary effects that the government acts to prevent are the consequence of aggregate private response to the content of those films. (51) There is no fundamental difference between acting directly as an agent of the majority's opinion and registering its opinion in law in anticipation of its collective reaction; either way, the government helps the dominant point of view to overpower the minority view. Nonetheless, it is important that the government at least does not claim to endorse the dominant view and that, while some individuals become less likely to view the films at the theater, others may be more likely to do so as a result of the relocation of the theater.

The Court has evoked deliberative autonomy more explicitly in the economic context, perhaps because concerns about public discourse and political truths are less eminent. In the labor union context, the Court has privileged handbilling over picketing because the former relies more on the persuasive force of its message, even where the harm to the neutral party is the same. (52) Although the Court's empirical claim that consumers and neutral workers respond unreflectively to picketers is questionable, because the Court does not find handbills similarly intimidating, it requires, in at least that context, that the government protect the right of people to decide for themselves how to respond to union allegations. (53)

Finally, the Court justifies its protection of commercial speech in large part on the concerns of deliberative autonomy. While most economic regulation is constitutionally fair game, regulation that presumes that people cannot think for themselves is not. While in general the government is free to treat citizens as objects of economic regulation, in their capacity as speakers and listeners, citizens must be recognized as agents capable of deliberation. Thus, the Court overturned advertising restrictions in Virginia State Board of Pharmacy v. Virginia Citizens Consumer Council because "the State's protectiveness of its citizens rests in large measure on the advantages of their being kept in ignorance.... [The advertising ban] affects them only through the reactions it is assumed people will have to the free flow of drug price information." (54) Likewise, it struck down advertising restrictions for alcohol in 44 Liquormart, Inc. v. Rhode Island because the regulation was based on the "offensive assumption that the public will respond 'irrationally' to the truth." (55) The court found fault not with the effect of the regulation but with the intent behind it.

Deliberative autonomy has never been the sole principle motivating the Court. Nor is the claim here that deliberative autonomy should be the sole principle of free speech law. Not only are other values obviously important, but the application of deliberative autonomy would benefit from guidance derived from other principles; related principles of free speech usually point in the same direction, and when they point more clearly, they may clarify the import of deliberative autonomy itself. But autonomy is nevertheless fundamental in the sense that it cannot easily be sacrificed. It provides a floor. Where it is not clear whether deliberative autonomy forbids some regulation, we might ask whether the regulation will advance the search for truth or enhance the quality of public discourse. But where it is clear that deliberative autonomy is inconsistent with regulation of some speech, doctrine has tried and should try to protect that speech.


The status of deliberative autonomy as a constitutional value hinges in large part on its relationship to the other major constitutional value celebrated in free speech doctrine: democracy. Autonomy might be protected not "as an end in itself, nor as a means of individual self-actualization" but only "as a way of furthering the larger political purposes attributed to the first amendment [sic]. It is assumed that the protection of autonomy will produce a debate on issues of public importance ... that is 'uninhibited, robust, and wide open'" (56) While the fact that democracy will generally suffer from violations of autonomy is a good reason to protect autonomy, deliberative autonomy can also carry its own weight. In fact, as a central reason for democratic government, autonomy is a first premise of our constitutional system.

The Court's own inconsistency about the relative priority of the principles makes autonomy and democracy sound like the chicken and the egg. In Cohen v. California, the Court ambiguously spoke of the "hope that use of [the freedom of speech] will ultimately produce a more capable citizenry and more perfect polity" but also "the belief that no other approach would comport with the premise of individual dignity and choice upon which our political system rests." (57) It is not clear whether autonomy acts as a functional or a normative premise, i.e. whether it must be protected because the political system depends on it or because our system of government assumes its value.

Elsewhere the Court seems to celebrate freedom of speech mainly for its contribution to the search for truth and the quality of public discourse. The classic source of this view is Mill's claim that "the peculiar evil of silencing the expression of an opinion is, that it is robbing the human race." (58) In his dissent in Abrams v. United States, Holmes speaks of a "free trade in ideas," arguing "that the best test of truth is the power of the thought to get itself accepted in the competition of the market.... That at any rate is the theory of our Constitution." (59) It seems implausible that the Constitution embodies a deep epistemological theory of truth in general, but it seems likely that democratic government assumes that, insofar as political ideas ever rise to the level of truths, they are best identified through public discourse.

But the importance of public discourse does not settle its relation to deliberative autonomy. We must ask "what basic value or values the democratic process was designed to serve." (60) As several commentators have noted, democracy itself derives its normative force from premises about the moral capacities of citizens and their right of self-determination. (61) We respect the decisions that arise out of the democratic process because we respect the deliberative autonomy of the individuals who participate in that process. "Public discourse could not serve the project of self-determination if the opinions and attitudes of speakers were deemed to be merely the effects of external causes." (62) Democracy as a normative principle is analytically derivative from autonomy.

As a state of affairs, democracy advances respect for autonomy within a given political community: individuals cannot respect one another's autonomy without subjecting themselves to a democratic system of governance. Democratic decision-making, or the attempt to fairly reconcile opposing views, presumes that individuals can act from moral and other reasoned motivation and not just from fear of coercion. If individuals lacked deliberative autonomy, there would be no point to elaborate democratic procedures. It would not be wrong to govern through power and coercion alone. But because of deliberative autonomy, "a legitimate government is one whose authority citizens can recognize while still regarding themselves as equal, autonomous, rational agents." (63) This means that citizens can recognize a democratic government as a legitimate authority over themselves, but it also means that by recognizing and participating in government together, citizens recognize one another's autonomy.

In a well-structured democracy, public discourse is of consequence. Citizens will therefore invoke public reasons for their agenda. Often, this is merely good strategy because policies are more likely to succeed if they can win the support of those who will not directly benefit from them. But this is not always true, as where the self-interested group is a powerful majority. In a well-socialized democracy, the majority will nevertheless be reluctant to claim benefits openly on the sole grounds that it has the power to do so. The presumption is that the politically weak are also entitled to be addressed. This deep presumption of democratic decision-making stems from an idea about how one can exercise legitimate authority over agents capable of acting on reasons, not just interests.

Despite their basic relationship, not all democracies serve autonomy equally well. When an ostensibly democratic government fails to respect the autonomy of some individuals, it undermines the premise of its own authority. That is because "respect for individual autonomy is a presupposition of the legal order's assertion that people have an obligation to abide by its demands." (64) Insofar as its policies fail to recognize the deliberative autonomy of individuals, the state fails to address them as beings capable of moral obligations at all. The state treats them only as beings capable of experiencing sensations and causing harm, not as moral agents capable of thinking. Democracy without autonomy is not an attractive constitutional principle. If the Constitution sets up democratic government, it does so in order to recognize the autonomy of citizens. This recognition, manifest in macro structures, must pervade the practice of government.

This is not to deny that the political process is sacred in a system of democratic government, or that liberties that serve that process are not especially prized. As a practical matter, without them the autonomy of many minorities (and, perhaps in the end, the majority) is unlikely to be recognized. The principle of deliberative autonomy generates competing demands, and autonomy interests relating to public discourse and the political process may be privileged due to their background role. But the latter are nevertheless derivative from the more fundamental principle of deliberative autonomy, prior even to democratic self-government.


In the widely influential Kantian tradition, the morality of an act hinges on the motive with which an individual performs it. Similarly, the constitutionality of state action in the United States frequently depends on the state's purpose. Many commentators are wary of so unwieldy a constitutional standard. Reference to state purpose allegedly anthropomorphizes the state. (65) Moreover, because we cannot reduce the state to a singular moral agent, some conclude that it is conceptually incoherent to assign normative significance to any constructed purpose we may attribute to the state. (66)

There are clearly limits to the parallel between state and individual purpose. Kant argued that individuals must conform their wills to the categorical imperative in order to be autonomous, because only by acting in accordance with our faculty of reason, at the expense of our worldly interests, can we assert our free will. (67) The state is a fundamentally different creature: it has no faculties, only institutional practices; and it has no stake in the metaphysics of free will. But the relationship between subject and action can produce a regulative principle (68) in more ways than one. State purpose is important to constitutionality for related, though importantly different, reasons than those which make motive essential to individual morality.

The individual abides by the categorical imperative, or moral law derived from practical reason, because doing so is consistent with the presumption of free will on which she must operate. To advance her own heteronomous interests at the expense of others would allow external facts in the world to determine the course of her conduct; yet to act at all she must presume herself capable of acting on her own will, on grounds supplied not by contingent fact but by her own faculty of reason. (69) Similarly, the democratic state must act on reasons consistent with the premises of its own being. The democratic state derives its legitimacy from the combined voice of individual wills. Without individual wills separate from its own, the state lacks a source of democratic authority. In the context of free speech and public discourse, Richard Post observed that:
 some form of public/private distinction is necessarily implied
 by democracy understood as a project of self-determination.
 This is because the state undermines the raison d'etre of its
 own enterprise to the extent that it itself coercively forms the
 'autonomous wills' that democracy seeks to reconcile into
 public opinion. (70)

Elsewhere, he argued that "[c]itizenship ... presupposes the attribution of freedom. The ascription of autonomy is ... the transcendental precondition for the possibility of democratic self-determination." (71) To sustain the premise of its own legitimacy, the state must act in a manner consistent with the presumption that individuals are capable of self-construction.

Like the "categorical imperative" in the context of individual morality, regulative presumptions in the context of democratic legitimacy allow for consistency in our self-understanding. Consistency between the premises and practices of a democratic state is important because it allows us, as individual moral agents, to make the best sense of our relationships with the state, and thereby, our political relations with others in our community; it best reconciles our respect for state authority with respect for ourselves. Because we are inter-subjectively constituted, the conditions for individual reconciliation with authority are of a social (i.e. collective) character. The state's presumption regarding our capacity for self-construction can only do cognitive work for us as individuals if the presumption is public. The question of whether the state acts consistently with the presumption of self-construction is therefore inseparable from the question of whether it is understood to do so.

The social meaning attached to state purpose is important because the state acts for the public, especially in a democracy. Its actions are attributable to the public. What the state condones, society appears to condone. What the state condemns, the entire community appears to condemn. The government is an agent for the public. It does not embody the public or even perfectly represent its views (after all, in a pluralistic society there is no singular viewpoint the government could represent); but the government's actions carry the weight of more people than the few who actually act on its behalf. Its acts help define the community's relationship with the individual, and help constitute the inter-subjective identity of citizens. Its presumptions about a person make that person in her own and others' eyes. Violations of autonomy undo the person who, as citizen, is in part constituted by the public presumption of deliberative autonomy. Thomas Nagel made the point when he said that "[t]o be tortured would be terrible; but to be tortured and also to be someone it was not wrong to torture would be even worse." (72) Likewise, to be silenced by anyone is a harm, but to be deemed incapable of communication and deliberation is even worse.

Government acts that fail to respect the autonomy of individuals are thus more weighty wrongs than isolated snubs by private individuals. A denial of autonomy by the government suggests a refusal of recognition by the political community in which one lives. This is the community through which one acquires a sense of self. Further, unlike other individuals, the government claims authority over citizens. The presumption of deliberative autonomy is a basis for this claim of authority; without it, citizens could not be obligated to respect authority, only controlled. The government therefore always implicitly claims that what it does it does for right reasons; or, that it can justify itself to those it governs. Because authorities claim to bind us with their judgments, they bear a greater burden of justification than do individuals without the mark of authority. The very legitimacy of the state's authority on a matter may be judged on the legitimacy of the public justification for the state's action (i.e., the legitimacy of the apparent purpose behind the action). Because a legitimate state, in justifying its judgments and actions, must appeal to allegedly objective truths (as opposed to subjective or purely internal motives), a normative construct of state purpose must look to the social meaning of the state's actions. (73)

I have argued for a particular constraint on the purposes that a democratic government may espouse: democratic states must presume that individuals are capable of self-construction. That presumption manifests itself not only in free speech law but also in other bodies of constitutional law. But as a regulative principle, (74) it constrains always the purpose with which the state acts. The purposes prohibited by the First and Fourteenth Amendments are always ones that impose or rely on value judgments that deny the capacity of individuals to deliberate and make choices for themselves. In order to illuminate the centrality--and coherence--of state purpose in constitutional doctrine, I explore how state purpose is constructed with respect to not only free speech but also religion and equal protection.

A. Purpose in Constitutional Doctrine

Despite the skepticism of many commentators, existing doctrine regularly incorporates state purpose into tests of constitutionality. (75) Several possibilities are available to the Court in its construction of state purpose. It might conceive of state purpose as the reasons the government gives to justify a law; these might be any reasons given in court, or else the reasons publicly proffered by the various advocates of a law, whether legislators or key lobbyists. The Court could require sincerity in the government's account of its own motivation, or it could deem good faith unnecessary. On the other hand, it might deem subjective good faith insufficient. Finally, as advocated here, the Court could look to the social meaning of a law, which may or may not be consistent with either the stated or actual motivations for it.

Any social meaning test must look to the popular understanding of a law's purpose, not just its effects. (76) Otherwise the test will be at once over- and under-inclusive. It may be that the social meaning of a law is implicit but not self-conscious on the part of either legislators or the public. An invidious social norm is nevertheless part of the state's purpose if it conditions the more overt purpose of the law. (77) On the other hand, it may be that some part of the public assigns a purpose to a law that is not the motivating purpose, and that compelling alternative justifications are available. Those laws cannot be invalidated under a social meaning test without conflating the foreseeability element of legislative intent with the motivation element of legislative purpose. The state need not act in order to pursue an end inconsistent with the presumption of autonomy to violate constitutional principles; but ideas inconsistent with the regulative presumption must be 'but for' conditions of state action. They may be unspoken elements of the state's motive, which can (or must) remain unsaid because they are so deeply embedded in the background against which the state acts. But they must be 'real' rather than merely alleged.

Constructing state purpose may be most difficult where the 'but for' premise of government action is not supplied by political process. The government does not just react to majority values that control it via electoral politics; the government also imposes majority norms by anticipating market behavior. In these cases, value judgments are preconditions for state action, but the operative value judgments are not those of government actors; they are not a product of the state's decision-making apparatus. In these cases, the Court is forced to decide most clearly whether it will accommodate state purposes that are conditioned by, though not the direct product of, majority ideology.

Those who view the Constitution entirely as a check on the powers of government, with no further or underlying purposes, are unlikely to recognize the work of ideology as a background condition for state action. Charles Fried maintains:
 The Constitution is concerned only with limits on
 government, even though a person's autonomy may be
 assaulted as much if an employer, a neighbor or a family
 member silences him or stops his access to speech. Other
 legal norms take care of non-governmental offenses.... Free
 speech values are preserved in [private law] because of the
 neutrality of [the] ordinary background systems of tort,
 property and criminal law. (78)

Others are less confident about the neutrality of those background systems.

Richard Fallon argues that beneath differences regarding the scope of constitutional protections are varying assessments of how coerced our daily existence really is. Although he distinguishes between descriptive autonomy (causal control) and ascriptive autonomy (moral right), he suggests that negative libertarians assume "that normal adults have the requisite capacities to lead autonomous lives and that, in the absence of extraordinary interferences, they in fact do so." (79) It may be a part of their optimism to locate majoritarian pressures entirely in the state.

The Court is not always so optimistic. It recognizes that apparently neutral laws may be indirectly motivated by impulses inconsistent with the presumption of autonomy. Whether state purpose will be burdened by its ideological origins in the private sphere depends on whether there is a perceptible, meaningful difference between the ideological force of the state and the ideological force of the market. The Court is more willing to tie state purpose together with private purpose where history suggests the illiberal norm at issue is pervasive and therefore inescapable for the state.

In doctrinal terms, the level of suspicion with which a Court approaches the purported purposes of a regulation determines the level of scrutiny applied. Some constitutional "doctrines employ tests that can be viewed as surrogates for purpose tests." (80) Illicit motives are hard to prove from scratch and a determination of unconstitutional purpose on a case-by-case basis may be insulting to the government officials involved. "[H]istory and familiar psychology" sometimes warrant a rebuttable presumption that the state acts on forbidden reasons when it makes certain kinds of distinctions. (81)

B. Ascertaining Purpose

The theoretical merits and demerits of conditioning constitutionality on state purpose turn, to a significant extent, on the manner in which the law ascertains state purpose. I will first explore certain difficulties in defining the purpose that is the object of free speech and other constitutional norms. I will then discuss how, though the presumption of self-construction is consistent across a number of constitutional norms, those doctrines which, for historical reasons, more effectively incorporate social meaning into state purpose also better embody our political commitment to the principle of autonomy.

Perhaps the most controversial question has been the extent to which discriminatory effects should be treated as evidence of discriminatory purpose. The Court has refused to infer discriminatory purpose directly from discriminatory effect; (82) even so, it has treated discriminatory impact as compelling evidence, that combined with other facts, might establish what is usually impossible to establish directly, i.e., subjective discriminatory purpose on the part of lawmakers. (83)

The Court has rejected two extreme methods of constructing state purpose. It will not accept any government interest presented in a government brief; (84) nor will it look only to the objective consequences of state action. (85) On the one hand, it is problematic for the judiciary to imply bad faith on the part of the legislature; judgments about the boundaries of reasonableness are inevitable. "A peremptory rejection of proffered state purposes strongly suggests a value-laden appraisal of the legitimacy of ends." (86) On the other hand, disallowing all state action with disparate impact would preclude whole categories of desirable state action. A tempting compromise is to look to the legislative record to assess whether the prospect of disparate impact motivated the legislature; but "the very concept of 'the legislative record,' as employed by the Court, is a fiction. The nature of the legislative process belies the existence of comprehensive explanatory materials." (87) The Court has held further that "because we never require a legislature to articulate its reasons for enacting a statute, it is entirely irrelevant for constitutional purposes whether the conceived reason for the challenged distinction actually motivated the legislature." (88) Indeed, any attempt to solve the "epistemological problem" of ascertaining legislative motive by requiring express statements of motive only exacerbates the "futility problem," whereby legislators can pass identical legislation with apparently clean motives. (89)

In light of these obstacles, the Court deems evidence of subjective discriminatory purpose neither necessary nor sufficient. "[D]iscriminatory intent need not be proved by direct evidence.... [D]etermining the existence of a discriminatory purpose 'demands a sensitive inquiry into such circumstantial and direct evidence of intent as may be available.'" (90) At the same time:
 Th[e] Court will not strike down an otherwise constitutional
 statute on the basis of an alleged illicit legislative
 motive.... What motivates one legislator to make a speech
 about a statute is not necessarily what motivates scores of
 others to enact it, and the stakes are sufficiently high for us
 to eschew guesswork. We decline to void essentially on the
 ground that it is unwise legislation which Congress had the
 undoubted power to enact and which could be reenacted in
 its exact form if the same or another legislator made a 'wiser'
 speech about it. (91)

The recognition in the O'Brien case of mixed legislative motive alludes to a second issue. Besides determining whether discriminatory purpose is present, the Court must decide what weight to assign discriminatory purpose as compared to other legitimate purposes that may have motivated a particular piece of legislation. (92) The question is whether an illicit purpose must be the dominant purpose to render state action unconstitutional, or whether it is enough that an illicit purpose has played any role at all. In other words, "[a] court must construe the centrality of the impermissible interests in light of the harms attending such purposes." (93) In the case of interests that are clearly "constitutionally unacceptable," such as "denaturalizing racial minorities," even a weak motivational presence may be enough to undo a law. (94) But in the case of "weak forbidden interest rule[s], [such as those] against gender-based stereotypes ... policies can survive ... if they promote some other government interest that is not based on gendered stereotypes." (95) Thus, evidentiary and centrality requirements of legitimate state purpose vary with the social background against which the state acts, and in particular, the nature of the "forbidden interest" served by the government action.

I now explore similarities and differences in the construction of state purpose across three bodies of constitutional doctrine: free speech law, the Establishment Clause, and the Equal Protection clause. The debate about the constitutional significance of market-based policymaking and disparate impact to state purpose can be best understood as ultimately about the role of social meaning in defining state purpose.

C. Speech

The purpose of a restriction on speech is essential to its constitutionality. "Th[e] Court has long held that regulations enacted for the purpose of restraining speech on the basis of its content presumptively violate the First Amendment." (96) Non-content regulations are also invalid unless the "governmental interest is unrelated to the suppression of free expression." (97) Thus, the purpose behind government regulation of speech is an important factor in determining its constitutionality.

I have argued more generally that a purpose for state restriction of expression is illegitimate when it interferes with individual deliberation. By preempting the deliberative process, the state threatens to narrow the range of ideas and values viable under its authority to those ideas which the state itself endorses. But as the legitimacy of authority in a democratic state rests on the endorsement of its agenda by the citizenry, rather than the other way around, "[n]o particular objective can justify the coercive censorship of public discourse without simultaneously contradicting the enterprise of self-determination." (98) "[T]he government may not suppress speech on the ground that the speech is likely to persuade people to do something that the government considers harmful." (99)

The Court has adopted a narrower limitation on purpose and prohibits only those restrictions motivated by a desire to suppress expression. That standard has not served adequately the fundamental principle that the state ought not to interfere with the autonomous development of ideas and values. In Barnes v. Glen Theatre, Inc., the Court concluded that a statute prohibiting nude dancing served the "purpose of protecting societal order and morality." (100) Because it did not target the "expressive activity" of erotic dancing, it was upheld as constitutional. (101) Yet as Justice White observed in his dissent:
 The purpose of the proscription in these contexts is to protect
 the viewers from what the State believes is the harmful
 message that nude dancing communicates ... apparently on
 the assumption that creating or emphasizing such thoughts
 and ideas in the minds of the spectators may lead to
 increased prostitution and the degradation of women. But
 generating thoughts, ideas, and emotions is the essence of
 communication. (102)

The desire to promote morality through the suppression of nudity has the same social meaning as a desire to inhibit erotic expression. Not only did the state effectively restrict access to certain views about the significance of nudity and the value of eroticism, but also, in communicating condemnation, it violated the presumption that individuals can decide for themselves whether to reject the erotic message of nude dancing on the grounds that nudity is immoral. The Court's construction of state purpose was not informed by the set of social statements available to the state. The Court attributed to the state legislature too nuanced a purpose to be feasible where the social meaning of suppressing nudity and eroticism are indistinguishable. State legislators surely knew that, however constitutional their subjective legislative desires, the purposes embodied in their law were bounded by the social world on which they acted. The Court should not have ascribed to them a purpose different than what everyone would understand the state to be doing.

The social meaning of state purpose becomes especially tricky to decipher where majorities register the force of their own values at least in part outside the realm of politics, and the Court has been no more nuanced in its receptivity to social meaning in that context. In Renton, the District Court found that the local city council was concerned primarily "with the secondary effects of adult theaters, and not with the content of adult films themselves." (103) Among those anticipated secondary effects was a decline in the city's retail trade and local property values. (104) Those secondary effects saved the statute as they were deemed legitimate government interests unrelated to the content of the films. (105) But as Justice Brennan argued in his dissent, "[t]hat some residents may be offended by the content of the films shown at adult movie theaters cannot form the basis for state regulation of speech." (106) Strictly speaking, it seems plausible (though unlikely) that the city council was motivated by the economic consequences of an adult theater industry. But those economic effects hinged on the force of majority values in the marketplace, and the marketplace itself is a political construct with sometimes foreseeable consequences. The constitutional problem arises because the state has empowered private actors to impose their will on minorities, to require that minorities conform their lifestyles to majority tastes and even to foreclose the development of certain ideas in others. If the purpose behind state action is constituted by its social meaning, then it seems clearly unconstitutional. The purpose of the regulation assumed domination by one set of values.

A similar, but harder question arises in the context of media corporations and access to public discourse en masse. If the public traces the dominance of one ideology in the media to the state, then the state would be acting--by maintaining a univocal public discourse--to promote one set of views. Fiss has argued that the "the state's obligation of neutrality requires that it make certain that the public debate is as rich and varied as possible." (107) But even if it agreed that debate is neither rich nor varied, the Court would also have to find that the public associated that homogeneity with the state's own designs. If the Court orders a remedy, it may require the state to construct public discourse even more overtly. Even if the media resulting from state reforms were more objectively balanced, it would be problematic if the public perceived the newfound "balance" to be a product of majoritarian politics. The social meaning that is of constitutional concern (from the standpoint of the judiciary, though not necessarily legislators or regulators) is not whether discourse is ultimately rich and varied, but whether the state either has set up a thin, homogenous discourse, or has attempted to impose its own preferred conception of richness and diversity on its citizens.

D. Religion

Just as the state's purpose to interfere with private deliberation captures unconstitutionality under the First Amendment, a state purpose to advance or suppress religion is unconstitutional under the Establishment and Free Exercise Clauses of the First Amendment. (108) The first prong of Lemon v. Kurtzman holds that a law must have a secular purpose under the Establishment Clause. (109) In Church of Lukumi Babalu Aye, Inc. v. City of Hialeah, a city ordinance was struck down under the Free Exercise Clause because it had the "impermissible object" of preventing the Santeria from practicing their religion in the area. (110) The Court rejected the city's claims of alternative purposes, maintaining that the ordinance was "gerrymandered" to oppress the Santeria church and observing that "[t]he Free Exercise Clause protects against governmental hostility which is masked as well as overt." (111)

Just as content regulation flagged the possibility of illicit government purpose in the context of free speech, laws that impact primarily one religious group are suspect for illicit purpose under the Free Exercise Clause. (112) By contrast, "a law that is neutral and of general applicability need not be justified by a compelling governmental interest even if the law has the incidental effect of burdening a particular religious practice." (113) As in the free speech and equal protection contexts, levels of scrutiny are really proxies for levels of suspicion about legislative purpose. The Court in Lukumi found it "not unreasonable to infer, at least when there are no persuasive indications to the contrary, that a law which visits 'gratuitous restrictions' on religious conduct seeks not to effectuate the stated governmental interests, but to suppress the conduct because of its religious motivation." (114) Applying the highest scrutiny, the Court set out to:
 determine the city council's object from both direct and
 circumstantial evidence. Relevant evidence includes, among
 other things, the historical background of the decision under
 challenge, the specific series of events leading to the
 enactment or official policy in question, and the legislative or
 administrative history, including contemporaneous
 statements made by members of the decisionmaking body. (115)

In the Lukumi case the Court found "[t]hat the ordinances were enacted 'because of,' not merely 'in spite of,"' the religious burdens it imposed. (116)

Under strict scrutiny, illicit purposes are forbidden regardless of whether legitimate purposes would have otherwise sufficed. State officials "must ensure that the sole reasons for imposing the burdens of law and regulation are secular. Legislators may not devise mechanisms, overt or disguised, designed to persecute or oppress a religion or its practices." (117) But in fact, the form that the law takes is important; the Court infers from targeted restrictions that the law would have looked differently had secular interests motivated lawmakers. (118)

Texas Monthly v. Bullock further refined the construction of state purpose under the Establishment Clause. It explained that:
 [t]he core notion animating the [Establishment Clause's]
 requirement that a statute possess 'a secular legislative
 purpose' and that 'its principal or primary effect ... be one
 that neither advances nor inhibits religion,' is not only that
 government may not be overtly hostile to religion but also
 that it may not place its prestige, coercive authority, or
 resources behind a single religious faith or behind religious
 belief in general.... (119)

Whether the state communicates such a message hinges in part on the objective 'statement' made by the state's action, but like all communicative acts, the meaning of the statement hinges on the audience that receives it. Whether the government endorses any kind of religious truth through state action depends crucially on whether it appears to do so.

In her concurring opinion in Amos, Justice O'Connor argued that:
 [T]he Lemon test should be 'whether government's purpose is
 to endorse religion and whether the statute actually conveys
 a message of endorsement.' To ascertain whether the statute
 conveys a message of endorsement, the relevant issue is how
 it would be perceived by an objective observer, acquainted
 with the text, legislative history, and implementation of the
 statute.... The determination whether the objective observer
 will perceive an endorsement of religion 'is not a question of
 simple historical fact. Although evidentiary submissions
 may help answer it, the question is, like the question
 whether racial or sex-based classifications communicate an
 invidious message, in large part a legal question to be
 answered on the basis of judicial interpretation of social
 facts. (120)

Her analysis seems to recommend a case-by-case, fact-based inquiry. Similarly, in Lynch, Justice O'Connor argued that "[e]very government practice must be judged in its unique circumstances to determine whether it constitutes an endorsement or disapproval of religion." (121) The outcome in Lynch depended on "the notion that Christmas was a cultural as well as a religious holiday." (122) Once the focus turns on the communicative impact of government action, purpose becomes a necessary component of the analysis because language cannot be understood without reference to the apparent intention of the speaker. But the state does not speak a language unto itself; the purpose of the state is also bound up with the meaning its actions take on in the social field in which it acts. The requirement that the "secular purpose" of the state be successfully communicated is consistent with the requirement that its purpose be "sincere." (123) Legislators can be presumed fluent in the social language in which they speak. If the purpose the state retroactively ascribes to its choice of language is entirely at odds with the social meaning of its action, then the government defense will be deemed in bad faith.

The assumption of fluency only applies, however, where the communication is clear to those whose understandings constitute the language. Where state action has no definite social meaning, because there is no common grammar from which its actions can be interpreted, then governments are substantially relieved of the high burden of establishing implausible secular purposes as their own. This suggests that laws may be religiously motivated in the sense that some lawmakers and some members of the public may support laws on religious grounds. But, so long as the meaning of the state's action is not religious endorsement, it cannot be the state's purpose. While intent does not imply purpose, in this context, purpose does require intent. If the failure of a message of endorsement is foreseeable, we can surmise that endorsement was not the purpose of the law. At other times, however, "[t]he social meaning of a law may ... be so clear that it swamps the diversity of perceptions, so that nearly any member of society will agree about what the law signifies." (124) In these cases, the government has effectively endorsed one view point, and the Court will apply the highest level of scrutiny to the question of whether it was motivated to do so.

But then the same dilemma regarding the state's relationship to the market arises in this context as in the free speech context. Just as state action may restrict speech on the basis of the secondary effects of private choices, state action may advance religion through private choice. As in the free speech context, the extent to which the state may successfully disassociate its reasons from those motivating the private choices the state anticipates depends on whether private choices "swamp" the social meaning of what the state does. The Court has increasingly leaned toward allowing private choice to wash the hands of the state under the Establishment Clause despite the foreseeability of disparate impact and the resulting 'grammatical' constraints on the state's message. For example, under Agostini v. Felton, state financing of religious indoctrination is acceptable if "aid is provided to students at whatever school [they] choose to attend." (125) In Mitchell v. Helms, the Court observed that "[i]f the government is offering assistance to recipients who provide, so to speak, a broad range of indoctrination, the government itself is not thought responsible for any particular indoctrination." (126) The Court found it important that the presence of choice allegedly prevented the appearance of endorsement. In Amos, where religious organizations were permitted to discriminate on the basis of religion, the Court likewise took comfort in the fact that it could "not see how any advancement of religion achieved by the Gymnasium can be fairly attributed to the Government, as opposed to the Church." (127) The relevant doctrinal question in these cases was not whether the resulting social system would be pro-religion, but rather, whether the purpose of advancing religion would be assigned to the state. (128) Whether this purpose would be assigned to the state depended in part on whether the public would attach political significance to the background markets that conditioned both of these cases. Although the labor market is largely secular, the market for private education is predominantly sectarian. Nevertheless, the Court essentially found that the message of government endorsement was sufficiently ambiguous as to pass constitutional muster. (129)

E. Equal Protection

The constitutional doctrines considered in this Part are constructed differently, but they can be understood to do similar work. For example, the First Amendment prohibition against viewpoint discrimination can also be understood in equal protection terms, with the First Amendment itself constraining only the scope of legitimate government interests. (130) "Similarly, the attention paid in [the context of the Establishment Clause] to 'insiders' and 'outsiders' rings with equal protection considerations." (131)

Post suggests that "[t]he ideal of autonomy essentially distinguishes First Amendment jurisprudence from other areas of constitutional law, which are most often associated with specific visions of collective identity" like, "national values of equality." (132) But, just as the First Amendment can be understood in terms of equal protection, equal protection can be understood to promote the free development of individual personality that is central to an autonomy-driven interpretation of the First Amendment. (133) The presumption of deliberative autonomy and the possibility of self-construction animate the law of not only free speech and religion, but also of equal protection.

Andrew Koppelman attempts to differentiate the religion clauses from the Equal Protection Clause by suggesting that subjective purpose is more important in the equal protection context: "The basic premises of democracy condemn a political process in which the decisionmakers are racist, but not a political process in which some of the decisionmakers have religious views and allow those views to influence their political positions." (134) But though racism on the part of legislators is surely to be condemned, it is not the fundamental object of constitutional scrutiny.

Koppelman is right to suggest that equal protection is different. It may be "the most purpose-conscious area of the Court's jurisprudence." (135) The legitimacy of state purpose is central to equal protection analysis even for groups that are not entitled to any heightened scrutiny. Where a law serves no apparent legitimate purpose but singles out one group for disadvantage, the Court will infer the purpose of signaling the greater worthiness of some people, and some ways of life, over others. (136)

It is the exceptional case where there is no "conceivable state of facts that could provide a rational basis for the classification." (137) Since most laws can be rationalized ex post, the Court's inquiry into state purpose usually turns on the level of scrutiny applied to a classification. (138) The level of scrutiny applied depends on the level of suspicion with which the Court views classifications of a given type. (139) Where it believes lawmakers are generally to be trusted in their treatment of a group, either because there is positive history, a positive trend, or a perceived 'good reason' for classifications, the Court will not provide heightened scrutiny. (140) In the absence of "some reason to infer antipathy," the Court will assume that "even improvident decisions will eventually be rectified by the democratic process and that judicial intervention is generally unwarranted no matter how unwisely we may think a political branch has acted." (141)

Where the Court has taken notice of a troubled history, it attempts to cipher out the hidden presumptions behind discriminatory practices. In United States v. Virginia, for example, Justice Ginsburg extensively discusses the nation's and Virginia's long history of sex discrimination. (142) In this context, "precedent instructs that 'benign' justifications proffered in defense of categorical exclusions will not be accepted automatically; a tenable justification must describe actual state purposes, not rationalizations for actions in fact differently grounded." (143) In that case, she "found no close resemblance between 'the alleged objective' and 'the actual purpose underlying the discriminatory classification." (144) The Court's inquiry was to ensure that sex classifications not be used "as they once were, to create or perpetuate the legal, social, and economic inferiority of women." (145) The same constitutional interest in preventing social stereotypes to inform state purpose and thereby violate the presumption of self-definition led the Court to strike down discriminatory social security provisions based on sex-based generalizations. (146)

The Court has applied intermediate scrutiny even where the disadvantage at issue applied to a traditionally advantaged group. Due to doubts about the true motives of the legislature in enacting sex-based drinking ages, (147) the Court in Craig v. Boren looked into the statistical basis for the policy and found them unsatisfactory. (148) Though Justice Rehnquist objected to the application of strict scrutiny given the lack of a history of discrimination against men, stereotypes about the personality and behavior of individuals of certain classes are inconsistent with the presumption of self-construction regardless of whether they are perceived by some or by all as positive or generous. (149)

While the Court may disagree about whether the degree of attention afforded the social meaning of state action should sometimes turn on whether the group hindered by a classification has a history of disadvantage, it seems largely to agree that "[c]ertain classifications ... in themselves supply a reason to infer antipathy. Race is the paradigm." (150) Some classifications have so much cultural baggage that it is difficult to believe that a law incorporating those classifications was not motivated in part by the animus and stereotypes that have traditionally prevailed. (151)

But even where there is historical reason to suspect that the state is motivated by assumptions about persons inconsistent with the presumption of self-construction, evidence that a law interferes with autonomy as a positive ideal does not establish illicit state purpose. The Court's construction of state purpose allows for the private sphere to undermine the positive realization of autonomy, even while the presumption of autonomy is binding on the state. Where evidence of disparate impact makes a prima facie case for invidious state purpose, the burden of proof does shift to the State to "rebut the presumption of unconstitutional action by showing that permissible racially neutral selection criteria and procedures have produced the monochromatic result." (152) As the Court acknowledged in Washington v. Davis, "in various circumstances the discrimination is very difficult to explain on nonracial grounds." (153) Nevertheless, though "[d]isproportionate impact is not irrelevant ... it is not the sole touchstone of an invidious racial discrimination forbidden by the Constitution. Standing alone, it does not trigger the rule that racial classifications are to be subjected to the strictest scrutiny and are justifiable only by the weightiest of considerations." (154)

In other words, unconstitutional purpose must motivate state action in order to render the action unconstitutional, even if it is but one motivating factor among others. Under Feeney, it is not enough that discriminatory impact was foreseeable. '"Discriminatory purpose' ... implies more than intent as volition or intent as awareness of consequences. It implies that the decisionmaker ... selected or reaffirmed a particular course of action at least in part 'because of,' not merely 'in spite of,' its adverse effects upon an identifiable group." (155) This standard does not imply that the state only acts unconstitutionally when its object is to disadvantage one group. State action may also violate Equal Protection when the state would not have so acted but for background social norms that disadvantage one group.

As in the free speech and religion contexts, unconstitutionality lies in state purpose itself, not the social facts produced by state practices. (156) In the equal protection context, the antisubordination principle may be understood to target the illicit purpose most common in racial classifications: not only a blunt desire to oppress certain groups, but also views about persons inconsistent with their status as equal citizens, or authorizing agents in a democratic state. It is easy to carry this prohibition on state purpose into a direct prohibition on types of state action. For example, Meyers moves from the claim "that a governmental interest in harming a 'politically unpopular' group or accommodating the fears and negative biases held toward a group constitutes an unconstitutional purpose," to the quite distinct claim that "[l]egislation creating virtually insurmountable obstacles to the social advancement of other groups similarly reflects impermissibility." (157) But the latter only follows from the former if all such legislatively-created obstacles are purposeful.

The Court looks to social facts to give meaning to what the state does, but its commitment to a boundary between the state and civil society prevents it from assigning the entire social impact of a law to the state. It cannot presume that individuals will respond to state action in a particular way; it can only interpret the state's actions in order to police the state's motivations. There are different ways of understanding the Court's commitment to antisubordination. All will involve contestable "interpretive judgments about social meaning, status, and the like" that are contingent on "factual and historical contexts, and, in particular, on the laws and social mores that prevail in a given society at a given moment in history." (158) In a given case, this record must speak with great clarity to establish that the state did not merely causally bring about subordination but acted in a manner that presumed the desirability of subordination, or made other assumptions about group members that are anathema in a liberal state. Social meanings and practices are a grammar with which to interpret what the state is trying to do, rather than instruments by which the Court. should assess the desirability of what the state has done against substantive egalitarian goals. While there are no doubt substantive goals suggested by the Equal Protection clause, as well as by each of the other clauses discussed here, the Court is not in a position to favor some means to their attainment over others. It can only regulate purposes inconsistent with the authority of a democratic state in the first place.

The unique social background for application of the Equal Protection clause appears to have led the Court to a more nuanced conception of state purpose in this context than under the First Amendment. Since the purpose constitutionally proscribed is the imposition of values and ideas on a group in denial of their capacity to assign value to ideas for themselves, the Court again faces a dilemma in the Equal Protection context about whether to assign responsibility to the state for heeding market pressures that themselves reflect the ideology of the majority. But unlike in the First Amendment context, the Court seems unwilling to presume the autonomy of private decision-making at the risk of allowing the state to indirectly bless the subordination of one class within it.

In Palmer v. Thompson, the Court rejected the city's explanation for desegregating everything except swimming pools, which were closed on the alleged belief that closure was necessary for peace and order and that integrated pools could not be operated economically. (159) The Court's decision turned on the fact that it would not have been possible as an evidentiary matter to impeach the motives avowed by the city. It would be too easy for the city to obscure its own purposes by pointing to the market consequences of private prejudice.

The Court later held more explicitly that the state cannot accommodate, even in 'good faith,' the prejudice of private actors. In Cleburne, the Court was unsympathetic to the Council's concerns about the "negative attitude[s]" of neighboring property owners: "mere negative attitudes, or fear, unsubstantiated by factors which are properly cognizable in a zoning proceeding, are not permissible bases for treating a home for the mentally retarded differently from apartment houses, multiple dwellings, and the like." (160) Similarly, in Palmore v. Sidoti, the Court observed that "[c]lassifying persons according to their race is more likely to reflect racial prejudice than legitimate public concerns" and chose not to distinguish between the racial prejudice of private and public actors. (161) Instead it stated:
 The question ... is whether the reality of private biases and
 the possible injury they might inflict are permissible
 considerations for removal of an infant child from the
 custody of its natural mother. We have little difficulty
 concluding that they are not. The Constitution cannot
 control such prejudices but neither can it tolerate them.
 Private biases may be outside the reach of the law, but the
 law cannot, directly or indirectly, give them effect. (162)

In Nipper v. Smith, the Eleventh Circuit held that districts cannot be drawn to reflect voter prejudice. (163)

The special history of the Equal Protection clause, as well as Congressional pressure to break down the immunity of state accommodation of private prejudice, has led the Court to recognize social meaning with a breadth lacking in the First Amendment contexts. Moreover, the Court is clearly more confident that it is 'right' about those values backed by its Equal Protection interventions than those that might be protected by its intervention in the First Amendment contexts. Despite the basic consistency in ambition across these amendments, the rigor with which the Court enforces the presumption of individual self-construction varies depending on how clear and unambiguous the Court perceives the social meaning of the State's purpose to be.

Variation in application does not detract from the legitimacy or feasibility of a constitutional doctrine centered around state purpose. Rather, it shows how, like any workable doctrine, free speech doctrine aimed at protecting even so abstract a concept as deliberative autonomy--precisely because it centers around state purpose--can be responsive to social and historical context. (164) It also shows that understanding free speech law by way of the principle of deliberative autonomy may offer a coherent, broad picture of how free speech rights relate to other constitutional norms.


Autonomy-based theories of the First Amendment regularly face two challenges. First, how can a constitutional doctrine be premised on an assumption about persons that is empirically false? Individual decision-making is often significantly constrained by external factors or simply impaired in itself. Second, how can constitutional doctrine be premised on a thick moral principle which many Americans do not share? Not all Americans subscribe to Kantian moral philosophy; many do not constrain their understanding of their interpersonal duties with reference to the moral autonomy of others. These two challenges to deliberative autonomy as a constitutional doctrine would rest on a misunderstanding of the conceptual structure of deliberative autonomy, which can be clarified with reference to the Kantian concept of a regulative principle.

A. Relationship of Moral Principle to Empirical Fact

Deliberative capacity, in a theory of deliberative autonomy, operates as a source of moral claims by the citizen, not as a talent to be cultivated by the individual or the State--though no doubt the worldly ideal of autonomy is important, it is not a first constraint on free speech regulation. But if the moral presumption has its roots in an actual capacity, the question arises: What happens when the presumption is not descriptively true?

The question is important because of two facts about the world: first, many people do not have the control over their lives that one associates with autonomy as a positive state; and second, they and others want to increase their autonomy, which does not make sense when one treats it as a binary concept (you either have it or you do not). Joseph Raz has accordingly declared that "[a]utonomy is a matter of degree" and can be "construed as a kind of achievement." (165) Cass Sunstein has been more skeptical, arguing that it is not apparent that we should respect decisions that may themselves "be a product of social conditions that fail to allow for autonomy, rightly understood." (166) More specifically, "[t]he connection between any particular decision to speak and individual autonomy, rightly understood, will not always be clear. Some speech ... may not reflect speaker autonomy at all. Moreover, there are many real-world constraints on the autonomy of listeners, stemming from lack of sufficient education, information, and opportunities." (167) We should not equate choice with autonomy because "some choices reflect an absence of autonomy, as in cases of insufficient education, manipulation, or lack of options." (168) Even Robert Post, who defends an ascriptive notion of autonomy, allows for the possibility that its presumptions "be negated in discrete and local ways where First Amendment presumptions of autonomy have come to seem merely 'fictions' masking particularly intolerable conditions of private power and domination." (169)

Richard Fallon explains the persistence of autonomy as a principle even in the face of such empirical contradiction by observing that "negative libertarians" (with whom he associates ascriptive notions of autonomy) "typically assume descriptive autonomy to be the natural and predominant condition of normal adults in a free society." (170) But deliberative autonomy does not hinge on such false idealism. We clearly are limited externally in a number of ways that limit our ability to advance our life projects and advance our own values. Deliberative autonomy is a way of regarding one another that allows us to distinguish those constraints, within which we can pursue our conception of the good with integrity, from socio-political conditions which deny our very capacity to regard others or ourselves as conferring value on ideas. We can make sense of the presumption of deliberative autonomy as something more than false hope if we distinguish between a principle that aspires to capture some actual fact about the sensible world, on the one hand, and a regulative principle, on the other. Deliberative autonomy is a regulative principle.

A regulative principle is an assumption about the world that we cannot know to correspond to empirical reality--which, indeed, we may already know not to correspond to empirical reality--but which we justifiably regard as true for other reasons. Regulative principles do not derive strictly from the nature of their apparent object. For example, in the case of deliberative autonomy, the content of the principle does not derive from the quality of our cognitive capacities. Because regulative principles are adopted instrumentally--either necessarily because they are the only means by which to attain necessary (i.e. moral) ends, or because they are the best means of advancing cognitive inquiry (171)--their value does not hinge on their manifestation in the world of mechanical causality.

Understood this way, the apparent conflict between the scope of positive and normative autonomy "is not a true conflict." It is "merely a different interest of reason that causes a divorce between ways of thinking." (172) The nature of regulative principles, in contrast to constitutive principles, (173) explains why the validity of deliberative autonomy as a constitutional (political-moral) principle does not hinge on the satisfaction of any empirical hypotheses.

Deliberative autonomy is not unique as a regulative constitutional principle. The principle of equality is similarly situated in our discourse. Although the idea of equality may have empirical origins, (174) the principle of equality does not apply only to those whose actual circumstances demonstrate their equality. (175) Equality in constitutional discourse, like autonomy, is a moral principle underlying the political order. The principle of equality is true in an altogether different--also necessary and objective, but different--sense than statistical facts about our actual condition. (176) Prevailing 'actual' conditions are but one aspect of our collective existence and fail to explain all the constitutional principles by which its governance is constrained.

B. Objective and Subjective Variations

The second challenge to a free speech doctrine based on the principle of deliberative autonomy is that it would rest on too thick a conception of the person. We should not "turn freedom of expression into a sectarian political position." (177) "[A] strong commitment to expressive liberties" should not be available "only to those who endorse the idea that autonomy is the fundamental human good." (178) If the concept of autonomy requires too many metaphysical assumptions, democratic self-government--as an alternative organizing principle for free speech doctrine--may be, if not a first premise, at least a safe common denominator.

But the defense of deliberative autonomy as an ascriptive, moral concept did not appeal to a particular view of what good we should all pursue. In fact, on its own terms, the argument for deliberative autonomy denies the possibility of such a universal good for political purposes. The roots of the political principle of autonomy in Kantian moral theory are important to understanding its historic evolution, but the origins of the idea are not inseparable from its particular content. The Millian claim that "it is only the cultivation of individuality which produces, or can produce, well-developed human beings" need not be endorsed by all those who, precisely because they are unsure what a well-developed human being is, claim (but do not necessarily celebrate) a right to decide for themselves. (179) Nor must the thin liberal conception of the person underlying deliberative autonomy exhaust each individual's conception of the person. "Within different contexts we can assume diverse points of view toward our person without contradiction so long as these points of view cohere together when circumstances require." (180)

The regulative principle of deliberative autonomy is a thin moral principle because it can be regarded as either objective or subjective. Those of us who see the constitutional principle of deliberative autonomy as the political aspect of a deeper moral principle can regard it as objective. Those who perceive it as contingent and specific to our constitutional order can regard it as subjective.

Regulative principles that do not manifest themselves in objective (empirical) reality can nevertheless be objective in the sense that they are practically (morally) necessary. Moral laws are regulative principles because, as imperatives, they refer not to what 'is' but what 'ought to be'. (181) They can be objective within their own moral sphere, in the limited sense that they are not contingent and apply universally. (182)

Deliberative autonomy is an objective regulative principle only for those for whom the principle of deliberative autonomy is a true and universal moral principle. Given the centrality of the principle of autonomy to modern liberalism, and the centrality of modern liberalism to American culture, many Americans probably do espouse some variant of the thick principle of deliberative autonomy. But certainly not all.

Those for whom deliberative autonomy fails as a general moral principle may nevertheless embrace it as a regulative principle for constitutional purposes. While Kant employed the concept of a regulative principle to explain the status of various individually-held beliefs, and therefore labeled regulative beliefs that lacked universal validity "subjective," (183) the counterpart to subjective regulative principles in the context of a political system are principles that are internal to the constitutional order. Individuals can accept that deliberative autonomy animates democracy without conceding that it governs interpersonal relations in other spheres. Kant explains that we:
 can have a satisfactory reason for assuming something
 relatively (suppositio relativa) without being warranted in
 assuming it absolutely (suppositio absoluta). This
 distinction is pertinent when we have to do merely with a
 regulative principle, which we recognize as necessary, but
 whose source we do not know, and for which we assume a
 supreme ground merely with the intention of thinking the
 universality of the principle all the more
 determinately.... (184)

Kant emphasizes the special character of regulative principles here in the course of defending belief in a higher intelligence because the belief unifies otherwise segmented ideas and structures scientific inquiry--even while conceding that we have no reason to believe that such a being actually exists, and that the objective validity of the concept of a higher being is "excluded by the idea itself" (185) While it is "reason's speculative interest" (186) that justifies for Kant the regulative principle of a God, the principle of autonomy can bring theoretical coherence to democratic governance in much the same way. The unity is less comprehensive to be sure, but so too is the sphere in which the principle would apply. Those who reject autonomy as a true and general moral principle may recognize it as foundational for our form of government. (187) For them, the notion of autonomy can serve, in effect, as one tenet of America's public religion.


Thus far, I have defended deliberative autonomy primarily at a theoretical level. In discussing how the principle is implemented, I now consider whether it can only be applied negatively, i.e., to bar government action.

To begin with, the principle of deliberative autonomy has only a contingent relationship with libertarianism. Theories of autonomy are especially sensitive to government wrong, but not without reason. David Strauss deems the bias partially justified because manipulation by the government is more dangerous than manipulation by private parties. (188) But as Strauss also observes, sometimes "private action ... present[s] comparable dangers" or "the government might help overcome the dangers created by private action." (189) Sometimes "contemporary social structure is as much an enemy of free speech as is the policeman." (190)

The Court's special concern for state threats to deliberative autonomy derives from the government's unique position as the official interpreter of social meaning. (191) Although unique, the state is clearly not alone in this role. Accordingly, we should not blindly emphasize the dangers posed by state action at the expense of those posed by certain types of private action. Whether deliberative autonomy as a political principle governs an action depends on whether the public or private actor's presumptions about a listener (or, more likely, category of listeners) threatens to undermine public recognition of the listeners as individuals capable of reason and deliberation, no more or less capable than anyone else of conferring value (for political purposes) on ideas. Random insults by individual private actors are not likely to affect the political identity of those insulted. But restrictive choices by mass media that influence large numbers of people and claim to respond to the views of the public do pose a substantial threat to those excluded from their forums. This is not to say that these media must affirm all viewpoints, but the rules governing access--rules affirmatively enacted by the government--should ensure that each citizen can consider herself a participant in public discussion.

Deliberative autonomy thus places both negative and positive burdens on state action. When private actors wield disproportionate power over public discourse, the state should ensure that all citizens retain the access necessary for their voices or views to count. But while a court is well-positioned to declare government action unconstitutional because it fails to conform to the principle of deliberative autonomy, it is not well-situated to declare a state of affairs unconstitutional. The principle of autonomy often raises constitutional questions about social structures without demanding any particular response. Although a court could prescribe the remedy it thinks best, there is little gained by having a court do what legislatures are supposed to do. (192) Where the problem is narrow enough, a court could issue a structural injunction. But when the options are wide and the solution is necessarily an extensive and complex reform program, a court remedy would trivialize the relevance of public debate in the name of free speech.

This asymmetry in the Court's jurisdiction over negative and positive duties arising under the principle of deliberative autonomy does not leave us without remedy when powerful private actors endanger autonomy or other important public values. Deliberative autonomy is not as blunt a principle of free speech as libertarianism in general. It does not foreclose government regulation of speech per se. Good intentions go a long way. Regulations that are designed to provide access to public forums, whether airports or television, do not rest on insidious presumptions about our deliberative capacities.

Campaign finance reform is one context in which autonomy theories of free speech have gained notoriety. (193) Strauss concludes that (what amounts to) deliberative autonomy forbids campaign expenditure restrictions, because "when the government restricts expenditures on campaigns ... its concern is with the persuasiveness of the speech." (194) However, the government's purpose in regulating campaign finance clearly is not to preempt deliberation but to promote it. Campaign finance reform is therefore perfectly consistent with deliberative autonomy. In fact, the principle explains in part why such reform is imperative. To say that the Court should not mandate any particular regulatory regime in the area of campaign finance is not to say that it should interfere in whatever regime the legislature enacts, unless the enacted reforms interfere with deliberative autonomy in some particular way.

I have argued here that the direct object of free speech law should be state purpose rather than the effects of speech regulation. But nothing about the principle of deliberative autonomy as a constraint on state purpose suggests that the Court should refuse categorically to concern itself with the practical effects of state regulation. As is most evident in the context of equal protection (though not fully recognized there either), courts can often discern intent through patterns, and should in any case construct purpose with reference to social meaning. Looking to the working reality of speech regulation does not entail rejection of the underlying notion of purpose-based harm. Nothing in the moral principle of autonomy requires that we ignore the distribution of power in which people operate. It is the concentration of power in the government that makes us wary of its potential to override deliberative autonomy, and the same concern applies to private actors who are more influential than accountable. The principle of deliberative autonomy gives legislatures a mandate to curb the abuse of private influence and gives courts no reason to thwart those legislative efforts.

(1) See, e.g., Owen M. Fiss, Why the State?, 100 HARV. L. REV. 781, 785 (1987) (arguing that the First Amendment protects autonomy as a means of encouraging public debate, rather than "as an end in itself"). See also infra Part III (discussing the relationship between autonomy, democracy and free speech).

(2) Even the Supreme Court has given in to this temptation. See, e.g., Abrams v. United States, 250 U.S. 616, 617 (1919) (affirming the convictions of anarchists and socialists who had distributed leaflets attacking the government and calling for a general strike during World War I); Gitlow v. New York, 268 U.S. 652, 668 (1925) (upholding defendant's conviction for publishing documents advocating anarchy on the premise that free speech law does not deprive the State of its "primary and essential right of self-preservation").

(3) Cf. Charles Fried, The New First Amendment Jurisprudence: A Threat to Liberty, 59 U. CHI. L. REV. 225, 233 (1992):
 Our ability to deliberate, to reach conclusions about our good, and
 to act on those conclusions is the foundation of our status as free
 and rational persons.... IS]elf-respect forbids that I cede to the
 state the authority to limit my use of my rational powers ...
 [That is] why the state has no claim to dominion over our minds: what
 we believe, what we are persuaded to believe, and (derivatively) what
 others may try to persuade us to believe.


(4) A regulative principle is a presumption about the world adopted on non-empirical grounds. See it infra Part V.

(5) Joel Feinberg, Autonomy, in THE INNER CITADEL: ESSAYS ON INDIVIDUAL AUTONOMY 27, 28 (John Christman ed., 1989).


(7) See, e.g., Thomas Jefferson, Government Founded on the Will of the People, in THE POLITICAL WRITINGS OF THOMAS JEFFERSON 79, 80-81 (Edward Dumbauld ed., 1955).

(8) Some moral aspects of the person do not require public recognition. The deep moral principle of deliberative autonomy which we might use to guide our daily conduct does not depend on it. But I defend here only the political counterpart to that moral principle, or the political-moral principle of deliberative autonomy, and this principle constrains not individual but rather state action.

(9) 394 U.S. 557, 565 (1969).

(10) Owen M. Fiss, Free Speech and Social Structure, 71 IOWA L. REV. 1405, 1421 (1986). But see CABS R. SUNSTEIN, DEMOCRACY AND THE PROBLEM OF FREE SPEECH 155 (1993) (advancing a First Amendment framework which would prohibit regulations of speech that have "impermissible governmental justifications," or, in other words, illegitimate purposes).


(12) Id. at 34.

(13) Others concerned with the effects of speech regulation on public discourse are less clearly effects-oriented. For example, Christina Wells writes that "[t]he State can and should regulate speech that, by attempting to override the thought processes of other individuals, disrespects their rational capacities. Such speech does not facilitate, but rather detracts from, the public exercise of reason and is therefore the proper subject of the State's coercive powers." Christina E. Wells, Reinvigorating Autonomy." Freedom and Responsibility in the Supreme Court's First Amendment Jurisprudence, 32 HARV. C.R.-C.L.L. REV. 159, 170 (1997). Wells is ambiguous as to whether, in her view, unconstitutionality follows from any attempt by the state to override individual reason, or whether the state must have some measure of success, i.e., whether only interventions that actually skew public deliberation are unconstitutional.

(14) Thomas Scanlon, A Theory of Freedom of Expression. 1 PHIL. & PUB. AFF. 204, 217 (1972):
 [T]here are limits on the kind of obligation which autonomous
 citizens could recognize. In particular, they could not regard
 themselves as being under an 'obligation' to believe the decrees
 of the state to be correct, nor could they concede to the state the
 right to have its decrees obeyed without deliberation.


(15) Id. at 222. Scanlon later revises this position and treats autonomy "as a good to be promoted" rather than a mere "constraint on justifications of authority." T.M. Scanlon, Jr., Freedom of Expression and Categories of Expression, 40 U. PITT. L. REV. 519, 533-34 (1979).

(16) Of course, the "private" good that results from free speech, i.e., the opportunity to exercise moral agency, is closely related to--and supports--the political-moral principle that requires governments to presuppose the capacities that underlie moral agency. While application of the regulative principle of deliberative autonomy does not turn on the consequences of speech regulation for individual moral agency, the latter can be a positive reason for state action in its own right.

(17) See Joseph Raz, Free Expression and Personal Identification, 11 OXFORD J. LEGAL STUD. 303, 323 (1991).

(18) Id. at 324.

(19) Id. at 313.

(20) Id. at 313-14.

(21) See Feinberg, supra note 5, at 46-49 (discussing autonomy as a matter of personal sovereignty).

(22) 491 U.S. 397, 403 (1989) (holding that defendant's "burning of the [American] flag constituted expressive conduct" protected by the First Amendment).

(23) 412 U.S. 94, 192 (1973) (Brennan, J., dissenting) (criticizing the majority for holding that the First Amendment does not require broadcasters to accept paid advertisements).

(24) C. Edwin Baker, Harm, Liberty and Free Speech, 70 S. CAL. L. REV. 979, 990-91 (1997).

(25) Id. at 991.

(26) See, e.g., Thomas Nagel, Personal Rights and Public Space, 24 PHIL. & PUB. AFF. 83, 97 (1995) ("[T]he suppression of dissenting opinion because of the danger that it may persuade people ... is the ultimate insult"); GERALD DWORKIN, THE THEORY AND PRACTICE OF AUTONOMY 31 (1988) ("Underlying the various ideas of moral justification is a prohibition against treating people in such a way that they cannot share the purposes of those who are so treating them.").

(27) Owen M. Fiss, State Activism and State Censorship, 100 YALE L.J. 2087, 2097 (1991).

(28) Id.

(29) A violation of autonomy is a kind of wrong, or harm. As Richard Fallon suggested, harm can be an ascriptive concept understood in reference to the nature of autonomy. Richard H. Fallon, Jr., Two Senses of Autonomy, 46 STAN. L. REV. 875, 897 (1994).

(30) Cf. Fiss, supra note 27, at 2100 (noting that "[k]eeping ideas and information from the public ... is the gist of the constitutional wrong").

(31) See, e.g., Brandenburg v. Ohio, 395 U.S. 444. 448-49 (1969) (striking down Ohio's Criminal Syndicalism Act for failure to distinguish "mere advocacy ... from incitement to imminent lawless action").

(32) See, e.g., Street v. New York, 394 U.S. 576, 592 (1969) ("Though it is conceivable that some listeners might have been moved to retaliate upon hearing appellants disrespectful words, we cannot say that appellants remarks were so inherently inflammatory as to come within the small class of 'fighting words' which are likely to provoke the average person to retaliation, and thereby cause a breach of the peace.").

(33) See, e.g., United States v. O'Brien, 391 U.S. 367, 376-77 (1968) (holding that the government may regulate conduct with both 'speech" and 'nonspeech' elements if. among other requirements, "the government interest is unrelated to the suppression of free expression").

(34) See, e.g., City of Renton v. Playtime Theatres, Inc., 475 U.S. 41, 49 (1986) (treating zoning laws "designed to combat the undesirable secondary effects" of businesses in the sex industry as being subject to a special "content neutral" analysis under the First Amendment).

(35) See, e.g., Edward J. DeBartolo Corp. v. Fla. Gulf Coast Bldg. & Constr. Trades Council, 485 U.S. 568, 580 (1988) (holding that peaceful handbilling did not constitute coercive secondary boycott under the National Labor Relations Act because handbilling, unlike picketing, "depend[s] entirely on the persuasive force of the idea").

(36) See, e.g., 44 Liquormart, Inc. v. Rhode Island, 517 U.S. 484, 501 (1996) ("When a State entirely prohibits the dissemination of truthful, nonmisleading commercial messages for reasons unrelated to the preservation of a fair bargaining process, there is far less reason to depart from the rigorous review that the First Amendment demands.").

(37) See Schenck v. United States., 249 U.S. 47, 52 (1919) (declaring that speech can only be proscribed when the words will create a clear and present danger of bringing about "substantive evils that Congress has a right to prevent").

(38) See Chaplinsky v. New Hampshire, 315 U.S. 568, 572-73 (1942); see also Texas v. Johnson, 491 U.S. 397, 409 (1989) ("asking whether the expression is directed to producing imminent lawless action and is likely to make or produce such action"); Terminiello v. Chicago, 337 U.S. 1, 4 (1949) ("Speech is often provocative and challenging ... [But it] is nevertheless protected ... unless shown likely to produce a clear and present danger of a serious substantive evil that rises far above public inconvenience, annoyance, or unrest.").

(39) In particular, the Supreme Court has rejected the contrary presumption that:
 an audience that takes serious offense at particular expression is
 necessarily likely to disturb the peace and that the expression may
 be prohibited on this basis ... On the contrary ... a principal
 foundation of free speech under our system of government is to
 invite dispute. It may indeed best serve its high purpose when it
 induces a condition of unrest, creates dissatisfaction with
 conditions as they are, or even stirs people to anger.

Texas v. Johnson, 491 U.S. at 408-09 (internal quotation omitted). The Court reiterated the point in Cohen v. California, 403 U.S. 15 (1971):
 We have been shown no evidence that substantial numbers of citizens
 are standing ready to strike out physically at whoever may assault
 their sensibilities with execrations like that uttered by Cohen.
 There may be some persons about with such lawless and violent
 proclivities, but that is an insufficient base upon which to erect,
 consistently with constitutional values, a governmental power to
 force persons who wish to ventilate their dissident views into
 avoiding particular forms of expression. The argument amounts to
 little more than the self-defeating proposition that to avoid
 physical censorship of one who has not sought to provoke such a
 response by a hypothetical coterie of the violence and lawless,
 the States may more appropriately effectuate that censorship

Id. at 23.

(40) 268 U.S. 652, 673 (1925) (Holmes, J., dissenting) (opining that "[e]very idea is an incitement" if acted upon).

(41) See, e.g., Milk Wagon Drivers Union of Chicago, Local 753 v. Meadowmoor Dairies, Inc., 312 U.S. 287, 293-95 (1941) (observing that an "utterance in a context of violence can lose its significance as an appeal to reason and become part of an instrument of force," and deferring to the state court's assessment of the risk of violence in the context of industrial conflict).

(42) 315 U.S. 568, 572-73 (1942).

(43) 771 F.2d 323, 330-31 (7th Cir. 1985) (striking down a state ordinance which sought to prohibit only certain types of pornography and stating that "[u]nder the First Amendment... there is no such thing as a false idea").

(44) In Chaplinsky, a Jehovah's witness called his arresting officer, together with the rest of the municipal government, a "damned fascist" and a "damned racketeer." The incident took place at the height of World War II, a period in which no doubt a broader ranger of speech was deemed injurious to personal honor and in which violence was still widely considered an acceptable means by which a man could avenge his honor in the face of certain insults. And of course, because the country was at war with fascists and fascism, Chaplinsky's epithets of choice had special significance. Hence, while the Court in that case concluded that "[a]rgument [was] unnecessary to demonstrate that [Chaplinsky's appellations were] epithets likely to provoke the average person to retaliation, and thereby cause a breach of the peace," Chaplinsky has been interpreted narrowly. Chaplinsky, 315 U.S. at 574. Only a very limited range of statements are reasonably construed as 'fighting words' today.

(45) 391 U.S. 367, 376-77 (1968).

(46) See Clark v. Community for Creative Non-Violence, 468 U.S. 288, 293 (1984) (applying intermediate scrutiny to content-neutral regulation).

(47) See Seana Valentine Shiffrin, Speech, Death, and Double Effect, 78 N.Y.U.L. REV. 1135, 1161 (2003):
 [S]peaker liability [for secondary effects] is rejected primarily
 because the point of the activity and our protection of it depend
 on a separation of responsibility between speakers and audiences.
 It depends on a view of the role of audience members that conflicts
 with assigning a large amount of responsibility to the speaker,
 namely a view that audience members are, or are able to act as,
 independent minds who react to the content of [the] speech for
 themselves and perform their own evaluations of these ideas
 and their relevance to action.

Id. However, Shiffrin insists that "[s]ome part of the trouble has to do with the consequences & suppression," Id. at 1170.
 Speech that produces intentional harm by stimulating others into
 endorsing and acting upon the speech's content is operating as
 it should--as it is supposed to.... A central reason for protecting
 speech is to permit a variety of ideas to be promulgated to, and
 evaluated and tested by, independent agents.... If we value speech
 as a communicative enterprise and we value the voicing of a range
 of ideas ... we have strong reason not to use the effectiveness of
 speech qua speech, in the context of its normal mode of operation,
 as grounds for its suppression.

Id. at 1163. In this respect, her argument may be Millian. See supra Part I. Such an instrumental account of free speech may locate the value of free discourse either in the benefits it confers on individuals or in the common good, i.e., the public state of affairs brought about by the right to free speech. See, e.g., Richard H, Pildes, Why Rights Are Not Trumps. Social Meaniugs, Expressive Harms, and Constitutionalism, 27 J. LEGAL STUD. 725, 744 (1998) (noting that First Amendment analysis is guided by conceptualizations of the common good rather than by "analysis of rights").

(48) 475 U.S. 41, 50-51 (1986) (upholding a statute limiting adult movie theaters to certain areas because the aim of the statute was to protect the community from the "secondary effects" of the theaters, not from the movies themselves).

(49) Id. at 50.

(50) Id. at 51.

(51) See infra Part IV (discussing the concept of state purpose).

(52) Edward J. Debartolo Corp. v. Fla. Gulf Coast Bldg. & Constr. Trades Council, 485 U.S. 568, 580-81 (1988).

(53) Id. at 584.

(54) 425 U.S. 748, 769 (1976).

(55) 517 U.S. 484, 503 (1996) (citing Linmark Associates, Inc. v. Township of Willingboro, 431 U.S. 85, 96 (1977)).

(56) Fiss, supra note 1, at 785. Fiss borrows the phrase "uninhibited, robust, and wide open" from Justice Brennan's opinion in New York Times Co. v. Sullivan, 376 U.S. 254, 270 (1964).

(57) 403 U.S. 15, 24 (1971).

(58) MILL, supra note 11, at 19.

(59) 250 U.S. 616, 630 (1919) (Holmes, J., dissenting).

(60) Martin H. Redish, The Value of Free Speech, 130 U. PA. L. REV. 591, 601 (1982).

(61) See, e.g., Robert Post, Managing Deliberation: The Quandary of Democratic Dialogue, 103 ETHICS 654, 671-72 (1993) (posing that public discourse cannot serve a self-determinative function if speakers are deemed to be merely reacting to external stimuli).

(62) See id. at 671.

(63) Scanlon, supra note 14, at 214.

(64) Baker, supra note 24, at 1015.

(65) See William N, Eskridge, Jr., Dynamic Statutory Interpretation, 135 U. PA. L. REV. 1479, 1507 (1987) (describing realists' rejection of the "anthropomorphism of legislative 'intent"').

(66) Id.

(67) See generally IMMANUEL KANT, GROUNDWORK OF THE METAPHYSICS OF MORALS (Mary Gregor trans. & ed., 1998).

(68) See infra Part V for discussion of the Kantian concept of a regulative principle.

(69) See KANT, supra note 67. at 49-62.

(70) Robert C. Post, Racist Speech, Democracy, and the First Amendment, 32 WM. & MARY L. REV. 267, 284 (1991).

(71) Post, supra note 61, at 672.

(72) Nagel, supra note 26, at 93.

(73) Thus, the state not only makes social meaning because it is a public agent, it also reinforces existing social meanings when it relies on them as objectively true and therefore the legitimate basis for exercise of state authority. However, what is relevant here is not the effect of state action on social meanings but the social meaning of state action. Cf. Cass R. Sunstein, On the Expressive Function of Law, 144 U. PA. L. REV. 2021, 2045 (1996) (locating the significance of the "statements" law makes in their effect on social norms); Lawrence Lessig, The Regulation of Social Meaning, 62 U. CHI. L. REV. 943, 1039-42 (1995) (describing how current First Amendment law constrains some but not other ways in which the state shapes social meanings).

(74) See infra Part V (addressing the common criticism of autonomy-based First Amendment theories as falsely assumptive and over-extensive).

(75) Although here I only discuss purpose in the contexts of free speech, religion, and equal protection, purpose is also important to the commerce clause and substantive due process. See Richard H. Fallon, Jr., Foreword: Implementing the Constitution, 111 HARV. L. REV. 56, 93-94 (1997). It is also important in the context of criminal punishment under the Eight Amendment. See generally Michele Cotton, Back with a Vengeance: The Resilience of Retribution as an Articulated Purpose of Criminal Punishment, 37 AM. CRIM. L. REV. 1313 (2000). In the context of substantive due process, purpose is important for reasons very similar to those discussed here. See, e.g., Washington v. Glucksberg, 521 U.S. 702, 782 (1997) (Souter, J., concurring) (distinguishing third justification for anti-euthanasia law because it deals with recognizable third-party interests).

(76) See Elizabeth S. Anderson & Richard H. Pildes, Expressive Theories of Law: A General Restatement, 148 U. PA. L. REV. 1503, 1523-24 (2000).

(77) The scope of state purpose therefore encompasses not just the end but also the premises on which the state adopts an end or the means to that end. Although this may seem counterintuitive, it is increasingly well-accepted in the context of race and gender. For example, an unconstitutional purpose encompasses not only the end of subjugating women, but also underlying gender stereotypes. See Miss. Univ. for Women v. Hogan, 458 U.S. 718, 728-30 (1982) (stating that "MUW's policy of excluding males from admission to the School of Nursing tends to perpetuate the stereotyped view of nursing as an exclusively woman's job"); United States v. Virginia, 518 U.S. 515, 535-36 (1996) (likening the Virginia Military Institute's exclusion of women to MUW's similar policy in Miss. Univ. for Women). See also Kenneth L. Karst, Constitutional Equality as a Cultural Form: The Courts and the Meanings of Sex and Gender, 38 WAKE FOREST L. REV. 513, 521-22 (2003) (discussing Miss. Univ. for Women and United States v. Virginia). Just as there are assumptions the state cannot make about particular groups, there are assumptions the state cannot make about persons (or, there are mandatory assumptions about persons). These more general impermissible assumptions about persons are, of course, substantively linked to the more specific assumptions we regularly ask courts to reject in the context of race or gender. For example, it is because every person can think for themselves that we reject gerrymandering that presumes all black people think alike in the context of an election.

This is not to say that all unconstitutional presumptions can be reduced to the principle of deliberative autonomy. Some assumptions relate to the basic constitutional assumption of equality. But the First Amendment revolves around the autonomy principle, and so I focus on it here.

(78) Fried, supra note 3, at 234-35.

(79) Fallon, supra note 29, at 880 (citations omitted).

(80) Fallon, supra note 75, at 90.

(81) Id. at 95. See also Elena Kagan, Private Speech, Public Purpose: The Role of Governmental Motive in First Amendment Doctrine, 63 U. CHI. L. REV. 413, 414 (1996).

(82) See Washington v. Davis, 426 U.S. 229, 242 (1976) (stating that "[d]isproportionate impact is not irrelevant, but it is not the sole touchstone of an invidious racial discrimination").

(83) See Vill. of Arlington Heights v. Metropolitan Hous. Dev. Corp., 429 U.S. 252, 266 (1977) (stating disproportionate effects "may provide an important starting point" for proving discriminatory purpose).

(84) See, e.g., Rubin v. Coors Brewing Co., 514 U.S. 476, 485-86 (1995) (refusing to recognize the federal government's asserted interest in "faciltat[ing] state efforts to regulate alcohol" as sufficient to justify a labeling ban on alcoholic beverages).

(85) See Arlington Heights, 429 U.S. at 266 (stating that "impact alone is not determinative").

(86) Gerald Gunther, Foreword: In Search of Evolving Doctrine on a Changing Court: A Model for a Newer Equal Protection, 86 HARV. L. REV. 1, 35 (1972).

(87) William W. Buzbee & Robert A. Sehapiro, Legislative Record Review, 54 STAN. L. REV. 87, 87 (2001).

(88) FCC v. Beach Communications, Inc. 508 U.S. 307, 315 (1993). See also U.S. R.R. Ret. Bd. v. Fritz, 449 U.S. 166, 179 (1980).

(89) Alan K. Chert, Statutory Speech Bubbles, First Amendment Overbreadth, and Improper Legislative Purpose, 38 HARV. C.R.-C.L.L. REV. 31, 78 (2003).

(90) Rogers v. Lodge, 458 U.S. 613, 618 (1982) (quoting Vill. of Arlington Heights v. Metropolitan Hous. Dev. Corp., 429 U.S. 252, 266 (1977)). The Court affirmed a lower court holding which had used evidence of past racial discrimination and "unresponsive[ness] and insensitiv[ity] to the needs of the black community" on the part of white elected officials, among other things, to conclude that an at-large voting system was maintained in order to dilute the black vote. Id. at 625-26. 628.

(91) United States v. O'Brien, 391 U.S. 367, 383-84 (1968) (finding the legislation in question--the Universal Military Training and Service Act of 1948--to have mixed motives).

(92) Melanie E. Meyers, Impermissible Purposes and the Equal Protection Clause, 86 COLUM. L. REV. 1184, 1207-08 (1986) (discussing the need for the Court to consider the "centrality of the impermissible interest[s]" in government decisions that create classificatory schemes).

(93) Id. at 1208.

(94) Roger Craig Green, Note, Interest Definition in Equal Protection: A Study of Judicial Technique, 108 YALE L.J. 439, 443 (1998).

(95) Id.

(96) City of Renton v. Playtime Theatres, Inc., 475 U.S. 41, 46-47 (1986).

(97) O'Brien, 391 U.S. at 377.

(98) Post, supra note 61, at 662.

(99) David A. Strauss, Persuasion, Autonomy, and Freedom of Expression, 91 COLUM. L. REV. 334, 335 (1991).

(100) Barnes v. Glen Theatre, Inc., 501 U.S. 560, 568 (1991).

(101) Id. at 570-71.

(102) Id. at 591-92 (White, J., dissenting).

(103) City of Renton, 475 U.S. at 47.

(104) Id. at 48.

(105) Id. at 54-55

(106) Id. at 59 (Brennan, J., dissenting).

(107) Fiss, supra note 10, at 2100.

(108) Anderson & Pildes, supra note 76, at 1545-46 (explaining that "laws that express hostility to religion ... violat[e] ... the Free Exercise Clause, and laws that endorse religion ... violat[e] ... the Establishment Clause") (citations omitted).

(109) 403 U.S. 602, 612 (1971).

(110) 508 U.S. 520, 524 (1993).

(111) Id. at 534.

(112) Id.

(113) Id. at 531.

(114) Id. at 538 (citation omitted).

(115) Id. at 540 (citation omitted).

(116) Id. (quoting Pers. Adm'r of Mass. v. Feeney, 442 U.S. 256, 279 (1979)).

(117) Church of Lukumi Babalu Aye, Inc., 508 U.S. at 547.

(118) See Texas Monthly, Inc. v. Bullock, 489 U.S. 1, 8-9 (1989).

(119) Id. at 9 (citation omitted).

(120) Corp. of Presiding Bishop v. Amos, 483 U.S. 327, 348 (1987) (O'Connor, J., concurring) (citations omitted).

(121) Lynch v. Donnelly, 465 U.S. 668, 694 (1984) (O'Connor, J., concurring); see also Wallace v. Jaffree, 472 U.S. 38, 73-74 (1985) (O'Connor, J. concurring) (arguing that "[t]he crucial question is whether the State has conveyed or attempted to convey the message that children should use the moment of silence for prayer. This question cannot be answered in the abstract, but instead requires courts to examine the history, language, and administration of a particular statue to determine whether it operates as an endorsement of religion.").

(122) Ira C. Lupu, The Lingering Death of Separationalism, 62 GEO. WASH. L. REV. 230, 239 (1994).

(123) Wallace, 472 U.S. at 64 (Powell, J., concurring) ("this secular purpose must be 'sincere': a law will not pass constitutional muster if the secular purpose articulated by the legislature is merely a 'sham'").

(124) Andrew Koppelman, Secular Purpose, 88 VA. L. REV. 87, 114 (2002).

(125) 521 U.S. 203, 228 (1997).

(126) 530 U.S. 793, 809-10 (2000).

(127) Amos, 483 U.S. at 337.

(128) See, e.g., Mitchell, 530 U.S. at 809.

(129) In fact, the Court found the message "neutral." Amos, 483 U.S. at 339. But neutrality in such cases is not passive, or suggestive of 'silence' on an issue. Rather, the state's actions are understood to have multiple meanings, i.e., they are known to all to be of varied, even conflicting, significance to different groups. The resulting ambiguity in the social meaning of a legislative act renders its communicative import inert, or "neutral."

(130) R. A. V. v. City of St. Paul, 505 U.S. 377, 384 n.4 (1992).

(131) Lupu, supra note 122, at. 241.

(132) Post, supra note 61, at 666.

(133) By contrast, expressive theories of law emphasize the equality presumption that constrains state purpose primarily under the Establishment and Equal Protection Clauses. See Anderson & Pildes, supra note 76, at 1521, 1570.

(134) Koppelman, supra note 124, at 118.

(135) Ashutosh Bhagwat, Purpose Scrutiny in Constitutional Analysis, 85 CAL. L. REV. 297, 357 (1997).

(136) See Romer v. Evans, 517 U.S, 620. 634-35 (1996) (finding "laws of the kind now before us raise the inevitable inference that the disadvantage imposed is born of animosity toward the class of persons affected").

(137) FCC v. Beach Communications, Inc. 508 U.S. 307, 313 (1993).

(138) See Jed Rubenfeld, Affirmative Action, 107 YALE L.J. 427, 432 (1997).

(139) See id. at 428 (noting that "[o]ne powerful function of strict scrutiny has always been that of 'smoking out' invidious purposes masquerading behind putatively legitimate public policy").

(140) See City of Cleburne v. Cleburne Living Ctr., Inc., 473 U.S. 432, 443 (1985) ("lawmakers have been addressing their difficulties in a manner that belies a continuing antipathy or prejudice and a corresponding need for more intrusive oversight by the judiciary"). The court explained that:
 Because mental retardation is a characteristic that the government
 may legitimately take into account in a wide range of decisions,
 and because both State and Federal Governments have recently
 committed themselves to assisting the retarded, we will not
 presume that any given legislative action, even one that
 disadvantages retarded individuals, is rooted in considerations
 that the Constitution will not tolerate.

See id. at 446. But see Justice Marshall's opinion: "Courts ... do not sit or act in a social vacuum ... what once was a 'natural' and 'self evident' ordering later comes to be seen as an artificial and invidious constraint on human potential and freedom." Id. at 466 (Marshall, J., concurring in part and dissenting in part). The discrimination was ultimately struck down on the grounds that it served no rational purpose. Id. at 450.

(141) Vance v. Bradley, 440 U.S. 93, 97 (1979) (citation omitted),

(142) 518 U.S. 515, 531, 536-37 (1996).

(143) Id. at 535-36.

(144) Id. at 536 (quoting Miss. Univ. for Women v. Hogan, 458 U.S. 718, 730 (1982)).

(145) Virginia. 518 U.S. at 534 (citation omitted).

(146) Weinberger v. Wiesenfeld, 420 U.S. 636, 637-39 (1975).

(147) Craig v. Boren, 429 U.S. 190, 199-200 n.7 (1976) (noting that the district court found the purpose of the statute to be promoting the safety of young persons while the attorney for Oklahoma argued that the purpose was traffic safety).

(148) Id. at 204 ("proving broad sociological propositions by statistics is a dubious business. and one that inevitably is in tension with the normative philosophy that underlies the Equal Protection Clause"). Justice Stevens concurred, finding it "difficult to believe that the statute was actually intended to cope with the problem of traffic safety." Id. at 213 (Stevens, J., concurring). "[E]ven assuming some such slight benefit, it does not seem to me that an insult to all of the young men of the State can be justified by visiting the sins of the 2% on the 98%." Id. at 214 (Stevens. J., concurring). "There is, of course, no way of knowing what actually motivated this discrimination, but I would not be surprised if it represented nothing more than the perpetuation of a stereotyped attitude...." Id. at 213 n.5 (Stevens. J., concurring).

(149) Id. at 219 (Rehnquist. J., dissenting). Alternatively, one could argue that classifications between groups to the apparent advantage of the disadvantaged group only reinforce more general stereotypes, thereby subjecting them to social stigmatization. See, e.g., Melissa L. Saunders, Reconsidering Shaw: The Miranda of Race-Conscious Districting. 109 YALE L.J. 1603. 1613-15 (2000) (observing that in the context of race, "minorities do not need to show special disadvantage to strike down racial classifications, the presumption is that they subject them to social stigmatization"). But if the constitutional standard points to state purpose, then stereotypes that are reinforced by, but do not underlie that purpose are only of policy, not constitutional, interest.

(150) Feeney. 442 U.S. at 272.

(151) Brenda Swierenga, Still Newer Equal Protection: Impermissible Purpose Review in the 1984 Term, 53 U. CHI. L. REV. 1454, 1475 (1986) ("Since the passage of the fourteenth amendment, the Court's opinions have reflected the notion that classifications based on race almost always reflect unfounded stereotypes.").

(152) Washington v. Davis, 426 U.S. 229, 241 (1976) (quoting Alexander v. Louisiana, 405 U.S. 625, 632 (1972)).

(153) 426 U.S. at 242.

(154) Id. (citation omitted).

(155) Feeney, 442 U.S. at 279 (citation omitted). But see id. at 284 (Marshall, J., dissenting) ("Where the foreseeable impact of a facially neutral policy is so disproportionate, the burden should rest on the State to establish that sex-based considerations played no part in the choice of the particular legislative scheme.").

(156) Id.

(157) Meyers, supra note 92, at 1193 (citations omitted).

(158) Jack M. Balkin & Reva B. Siegel, The American Civil Rights Tradition: Anticlassification or Antisubordination?, 58 U. MIAMI L. REV. 9, 14-15 (2003).

(159) 403 U.S. 217 (1971).

(160) Cleburne, 473 U.S. at 448.

(161) 466 U.S. 429, 432 (1984).

(162) Id. at 433.

(163) 139 F.3d 1494, 1546-47 (11th Cir. 1994).

(164) The current court appears willing to give cultural context its due. See Robert C. Post, Foreword: Fashioning the Legal Constitution: Culture, Courts, and Law, 117 HARV. L. REV. 4 (2003).

(165) JOSEPH RAZ, THE MORALITY OF FREEDOM 156, 204 (1986).

(166) SUNSTEIN, supra note 10, at 138.

(167) Id. at 143.

(168) Id. at 176.


(170) Fallon, supra note 29, at 880.

(171) See IMMANUEL KANT, CRITIQUE OF PURE REASON 689 (Paul Guyer & Allen W. Wood trans. & eds., 1998) (discussing the difference between logical and moral certainty) [hereinafter PURE REASON].

(172) Id. at 603.

(173) Constitutive principles are actually true as a result of the nature of their object. A rule of geometry, for example, is true by virtue of the geometric shape it describes, and it will held true in the sensible world. Id. at 297-98. For discussion of the differences between regulative and constitutive principles, see IMMANUEL KANT, THE CRITIQUE OF JUDGMENT 261. 284-85, 306 (J.H. Bernard trans., 2000) [hereinafter JUDGMENT].

(174) See THOMAS HOBBES, LEVIATHAN 183-84 (Penguin Books 1985) (16513 (arguing that natural equality renders the state of nature unstable).

(175) See, e.g., JOHN RAWLS, POLITICAL LIBERALISM 18-19, 81 (1993) (arguing that citizens are free by virtue of "their two moral powers (a capacity for a sense of justice and for a conception of the good)"); JOHN RAWLS, A THEORY OF JUSTICE 329 (1971) (distinguishing moral equality and equal dignity, on the one hand, and excellence and equal value on the other).

(176) This is not, of course, to suggest that the moral principles of equality and autonomy do not place certain demands on states with respect to actual conditions. See infra Part VI.

(177) Joshua Cohen, Freedom of Expression, 22 PHIL. & PUB. AFF. 207, 222 (1993).

(178) Id.

(179) See MILL, supra note 11, at 73.

(180) John Rawls, Kantian Constructivism in Moral Theory, 9 3. PHIL. 545 (1980).

(181) See JUDGMENT, supra note 173, at 318.

(182) See IMMANUEL KANT, CRITIQUE OF PRACTICAL REASON 43 (Mary Gregor trans., 1997); PURE REASON, supra note 171, at 678-79.

(183) See PURE REASON, supra note 171, at 610.

(184) Id. at 608-09.

(185) Id. at 609.

(186) Id. at 608.

(187) See supra Part III.

(188) Strauss, supra note 99, at 364.

(189) Id. at 361.

(190) Fiss, supra note 10, at 1415.

(191) See supra Part IV.

(192) This view about institutional competence or authority has no necessary analytic relationship with (e.g., it neither implies, nor depends on nor follows from) an autonomy-based theory of free speech. In most cases, the government's affirmative duty to promote deliberative autonomy is akin to a Kantian imperfect duty in that there is "playroom (latitude) for free choice in following (complying with) the law." IMMANUEL KANT, THE METAPHYSICS OF MORALS 153 (Mary Gregor trans. & ed., 1996). One might argue that as a matter of institutional competence the legislature is better suited to choosing among alternative means of complying with the principle of deliberative autonomy, together with other political obligations, and that courts are better suited to adjudicating the binary question of whether the specific course of action the legislature has adopted falls within the "playroom" available to the state. Alternatively, one might argue that the legislature is simply the only branch authorized to make the choices involved in enacting affirmative government programs. In neither case does a theory of free speech purport to dictate the institutional division of labor. The view that courts should leave affirmative action to the legislature is informed by independent assumptions about institutional competence or authority. A full discussion of the debate about the appropriate division of labor between the judicial, executive, and legislative branches is beyond the scope of this article.

(193) See, e.g., Burr Neuborne, Toward a Democracy. Centered Reading of the First Amendment, 93 NW. U. L. REV. 1055 (1999).

(194) Strauss, supra note 99, at 341.

Aditi Bagchi, J.D., Yale Law School, 2003; M.Sc., Oxford University, 2000; A.B., Harvard College, 1999. Many thanks to Bruce Ackerman, Jack Balkin and Owen Fiss for ideas and suggestions.
COPYRIGHT 2005 Albany Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2005 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Bagchi, Aditi
Publication:Albany Law Review
Date:Sep 22, 2005
Previous Article:Balancing civil liberties and homeland security: does the USA patriot act avoid justice Robert H. Jackson's "suicide pact"?
Next Article:The governor - from figurehead to Prime Minister: a historical study of the New York state constitution and the shift of basic power to the Chief...

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters