Free Speech and Obedience to Law

Frederick Schauer*

I.

Several generations ago Alexander Meiklejohn asserted that among the virtues of a regime of freedom of speech was its connection with the obligation to obey the law.[1] More specifically, Meiklejohn maintained not only that the right to voice disagreements with laws was a morally and politically necessary condition of compelling people to obey laws with which they disagreed, but seemed to imply as well that people would in fact be more inclined to obey those laws when they were given the opportunity to object than would be the case were their dissenting voices to be stifled by official action.

The relationship between democratic legitimacy and freedom of speech has subsequently been the subject of analyses offered by Ronald Dworkin,[2] by Robert Post,[3] and, most recently, by James Weinstein in this Symposium[4] and elsewhere.[5] Weinstein in particular advances our understanding of the issue by drawing on the venerable distinction between normative legitimacy and descriptive (or sociological) legitimacy.[6] Normative legitimacy is a process-based philosophical idea,[7] and designates or describes those forms of governmental organization and governmental action that are right, or just, as matter of political philosophy. The idea is normative and not empirical, and a government or its actions are legitimate insofar as they are democratic, or egalitarian, or deliberative, or in some other way built on normatively desirable foundations. More specifically, normative legitimacy typically is taken as referring to the conditions permitting the political state to justifiably demand obedience from its citizens, and thus to impose its laws on those who refuse to obey.[8]

If we understand normative legitimacy as a fundamentally non-consequentialist and non-instrumental idea, and if we understand it as focused primarily on procedure in the broadest sense of that word, then we can say that a normatively desirable form of governmental organization is to be preferred independent of the value of the consequences that may flow from adopting it. We might believe, for example, that majoritarian democracy is a good in itself, and that it has moral merits as a form of decision-making independent of whether it produces better policies, or more citizen happiness, or more truth, or anything else.[9] And thus we might believe, as Weinstein plainly does believe, that allowing people to object to policies with which they disagree is a necessary component of normative legitimacy, and thus of the warrant of the state to enforce its directives by coercive means. From this perspective, normative legitimacy stands in no need of consequential, instrumental, or empirical justification. It is simply a matter of first-order political morality. In offering this claim, Weinstein seems largely correct, even though it might be plausible to argue that a democracy exists when people have the right to vote for their representatives, or have the right to vote on matters of policy, and that, especially in a representative rather than direct democracy, the right of the citizen to speak out is not a necessary condition of democratic legitimacy itself as long as the citizen can be part of choosing those who will represent her.[10]

Although there are substantial difficulties with treating representation as a sufficient condition for either democracy or legitimacy, I mention that position here only to highlight the fact that tying freedom of speech to normative legitimacy needs some argument, and that a strong and continuous right to freedom of speech is not entailed by the very idea of democracy, at least as long as the idea of representative democracy is not an oxymoron. Still, it seems difficult to imagine a process of selecting representatives or policies that is not crucially facilitated by direct citizen speech, and it seems even more difficult to imagine government by the people that does not permit those people to participate in policy-making outside of the episodic process of voting.[11] As a result, the connection between freedom of (political[12]) speech and democratic decision-making emerges as a close one, entailed by a wide range of views about simply what it is governmental decisions to be made by as well as for the citizenry. Insofar as democracy is itself desirable, therefore, its components are necessarily so, and thus the argument is not so much that freedom of speech is a consequence of democracy or a facilitator of democracy as it is that freedom of speech is arguably simply part of the definition of what democracy and democratic legitimacy just are. It is not clear to me that translating this basic proposition of normative political philosophy into the language of “legitimacy” adds very much, but this is largely a terminological quibble. The basic point, regardless of the language we use to describe it, is that freedom of political speech is normatively justifiable as a necessary component of a normatively justifiable form of governance in which citizens have substantial input by voting and otherwise into the decisions that will affect them and that will control their activities.

II.

Although both this Symposium and much of the discussion of legitimacy takes place in the context of debates over laws prohibiting so-called hate speech, I will say very little about hate speech here. Rather, I will address more generally a range of issues that are (contingently) implicated by some of the issues in the hate speech debate, but my focus will be on the general issues and not on hate speech. One reason for skirting the hate speech controversy is that the issues of legitimacy, and of the relationship between freedom of speech and political legitimacy, are far broader than the question of hate speech in particular. But although the issue of hate speech is but one instantiation of the topic of free speech generally and of free speech and legitimacy more particularly, the hate speech question has both a high degree of political salience and a substantial amount of ideological valence. Consequently there exists some risk that the analysis of interesting and important questions about the relationship between political legitimacy and freedom of speech will be both crowded out and distorted by discussing them in the context of the contentious subject of hate speech. We know from the literature on the availability heuristic that that which is most salient may distort our appreciation of the full range of issues that some principle or decision may encompass,[13] and we know from the research on motivated reasoning that an antecedent preference for a an outcome may influence the analysis of the considerations that would lead to accepting or rejecting that outcome.[14] Accordingly, it seems preferable to discuss the broader philosophical and empirical questions about free speech and legitimacy in a discursive environment in which the participating discussants have few ideological or political priors, or at least are thinking about a wide range of applications, but the hate speech debate is decidedly not such an environment. And thus because I find the issue of the relationship between freedom of speech and political legitimacy independently important and interesting, I prefer to avoid rather than embrace, at least here, the distractions of the hate speech controversy.[15]

In addition, discussions of hate speech necessarily presuppose empirical as well as philosophical questions about the consequences of such speech. Tellingly, Dworkin, Weinstein, and many other opponents of hate speech laws describe the object of such laws as the desire on the part of government to suppress that with which it disagrees or that with which people take offense.[16] By contrast, most proponents of hate speech laws describe the object of such laws as the prevention of harm.[17] In the face of this important terminological and substantive difference about how to characterize the consequences of racial insults, endorsements of sexual violence, and Holocaust denials, for example, an initial and necessary question is thus the very nature of such consequences, a question whose empirical dimensions cannot be avoided.[18]

Consider, for example, Weinstein’s claim that imposing tax laws on those who disagree with the taxes is legitimate only if those who object to such laws have the opportunity to articulate their objections.[19] But suppose, quite realistically,[20] that there exists a highly secret algorithm employed by the Internal Revenue Service to determine whom to audit, the possession of which would enable taxpayers with almost complete certainty to avoid detection for plainly unlawful tax avoidance. Under such circumstances, would a prohibition on disclosing the algorithm—a prohibition on aiding and abetting unlawful tax avoidance—eliminate or diminish the obligation of those who disagreed with the tax to comply?[21] If not, then the only relevant distinction between this example and the hate speech scenario is an alleged closer empirical connection between this form of aiding and abetting tax avoidance and the act of actual tax avoidance than there is between facilitating, causing or constituting racial or ethnic or religious or sexual orientation discrimination by words and the actual practice of such discrimination. But whether there is such a difference is exactly the matter in issue, and thus it is question-begging to assume at the outset that hate speech is neither harmful in itself or of lesser consequence than the verbal aiding and abetting of conventional criminal activity. If hate speech just is an act of racial (or other unlawful) discrimination,[22] or if it is as causally connected with unlawful acts of discrimination as publishing the IRS algorithm is with unlawful tax evasion, then it is hard to see why the demands of legitimacy require permitting the objector to engage in the very practice that is prohibited, which is the point of the tax evasion example. Now it may be that there are important differences between the two examples, but those differences need to be explained and justified, empirically as well as conceptually and normatively. And thus the tax evasion example serves as a caution against simply assuming at the outset that hate speech is not itself an act of discrimination or is not as closely causally connected with acts of discrimination as disclosure of the IRS algorithm is with tax evasion.[23] And because this is not the forum for delving into these difficult empirical questions, I prefer to leave hate speech aside rather than to assume the answers to empirical questions about which the conceptual framing is complex and the data are at best inconclusive.

III.

Specific questions about hate speech aside, therefore, the question – or at least one question, and the one I address here – is whether the possession of a right to object to a regulatory law with which one disagrees will increase the likelihood that the disagreeing citizen will comply with the law despite her disagreement. This question is closely connected with the question of descriptive legitimacy as Weinstein articulates it, but is not identical to it, for reasons that are worth closer inspection.

To repeat what was noted above, we are no longer in the realm of normative legitimacy. At least for the sake of argument, I will accept that allowing citizens to express their individual (as opposed to doing so through their representatives) disagreement with a law is a necessary condition for that law’s normative legitimacy, and thus for the citizen’s obligation to comply with that law just because it is a law. Such an assumption presupposes, pace the philosophical anarchists,[24] that there are at least some laws that should be obeyed even in the face of first-order (or content-based) disagreement just because those laws are law. And if there are such laws—if the very fact of law independent of the content of the law provides a reason to obey it[25]—then the normative legitimacy question is whether the opportunity to object is a member of the set of necessary conditions for that law to generate an obligation to comply on the part of the subject who believes such a law morally or otherwise mistaken.[26] My assumption here is that freedom publicly to object is indeed one of those conditions, although, as discussed above, it is hardly self-evident that such an assumption is warranted.

With that assumption in hand, we can then turn from normative to descriptive legitimacy. But the question of descriptive legitimacy is not as straightforward as some, including Weinstein here, believe. Specifically, the descriptive question on which I focus is whether allowing objectors to object will increase the likelihood of compliance with the laws to which they object. Will those who disagree with a law find compliance “more acceptable” than they would absent an opportunity to participate in a decision that turned out to go against their preferences or their own considered judgments?[27] But although this looks like one question, it is in fact three questions, and it will be valuable to attempt to distinguish them.

The first question is the “wholesale” question whether people who live in a society in which dissent is permitted are more likely to believe that they have a moral obligation to obey the law than are people who live in a society in which dissent is restricted, controlling for all other possible differences. As with most other questions of comparative law, comparative political theory, and comparative public opinion, the presence of countless variables makes anything even approaching a conclusive answer virtually impossible. Still, and relying most on the research done by Tom Tyler and his collaborators on the relationship between a person’s sense of governmental legitimacy and her inclinations to obey laws and officials,[28] we might conclude that the answer to this question is in the affirmative. Although Tyler’s indicia of governmental legitimacy do not explicitly include the kind of opportunity for public dissent that is at the heart of the question on the table in this Symposium and this Article, it is a reasonable extrapolation from the indicia that Tyler does include that people who live in a society in which the laws are made publicly and democratically, and in which people have the opportunity publicly to dissent from those laws with which they disagree, will be more likely to believe that there is a general obligation to obey laws with which they disagree than those who live in a society in which such earmarks of legitimacy are absent.

Let us further assume an affirmative answer to this wholesale question. That is, let us assume that belief in a general obligation to obey the law will increase when people are allowed to object to the laws with which they disagree. But even if this is so, further questions remain. The second question, therefore, is whether people who believe in a general—abstract, or wholesale—obligation to obey the law will feel a “retail” obligation to obey specific laws with which they disagree—specific laws whose content they find objectionable. And here the reliance on Tyler, including Weinstein’s, becomes more problematic. Because all but one of Tyler’s survey questions fail to distinguish between laws with which people agree from laws with which they disagree, he is able to conclude that the chief factor in determining belief in an obligation to obey a specific law is the moral content of that law.[29] The moral desirability of a law is undoubtedly an important factor in explaining why people might believe that they should obey a law that disadvantages them personally, but it says little about people’s beliefs about laws they find mistaken, whether on moral or policy grounds.[30] In the context of compliance issues generally, this might be only a minor problem, as there is good reason to be interested in the sanction-independent grounds for compliance with personally disadvantageous laws that people find generally desirable, including many of the laws about, for example, taxes and driving. But the free speech question arises explicitly in the context of disagreements about the content of our laws, and the claim I am addressing is that those who object to a law—and not just those who are disadvantaged by a law to which they do not in the abstract object—will be more likely to comply when they have been afforded an opportunity to object to the passage or enforcement of that law. And on this question, the conflation of the moral and the legal is of determinative importance. If people profess to be willing without the necessity of sanction to comply with laws whose content they find morally agreeable, then it is hard to see why giving them an opportunity to object to such laws will have any effect on the degree of their willingness to comply.

Although this problem affects and infects the ability to infer a belief in the willingness to comply with an objectionable law from a willingness to comply with laws that subjects find morally desirable, Tyler’s research does include one question focused on the professed willingness of people to comply with laws whose content they find objectionable.[31] And here he similarly finds that people claim to be more willing to comply with such objectionable (to them) laws if they have been made under conditions of procedural legitimacy.[32] Again. Tyler does not explicitly include freedom to object publicly within his set of conditions, but he comes close by including citizen input within his definition of legitimacy, and thus Weinstein’s extrapolation seems entirely reasonable. Accordingly, we might conclude from Weinstein’s gloss on Tyler’s research that people profess to have a greater willingness to comply with laws with which they disagree when they have been given an opportunity to express that disagreement publicly than when such disagreement is in some way restricted or prohibited. This is the question of free speech and legitimacy as applied to particular laws, and now we are much closer to addressing the precise question that I consider here, that Meiklejohn, Dworkin, Post, and others have addressed implicitly, and that Weinstein here addresses most directly and explicitly. Thus, it appears from Tyler’s research to be the case that giving people an opportunity to object to laws with which they disagree will not only increase their belief in the legitimacy of the system as a whole, and not only increase their belief in the legitimacy of laws that that system has made, but will also increase their belief in an obligation to obey even those particular laws they find objectionable.[33]

It is at this point, however, that Tyler’s survey research, and Weinstein’s reliance on it, stops short of answering the ultimate question—will people who claim to believe in the obligation to obey laws with which they disagree, and who in fact believe that they have an obligation to obey such laws, actually obey such laws. And here we must confront the distinction between what people believe they ought to do and what they actually do. I believe, for example, that I ought to lose weight, but in the face of bacon-inspired temptation my genuine beliefs give way. Similarly, we ought to be interested in whether people who believe they ought to follow laws with which they disagree will in fact follow them when the “temptation” to disobey is present, a temptation that will occur with considerable frequency within the set of laws that people find objectionable. If we are now focused on laws with which people disagree, it should come as little surprise that such laws will often present for people situations in which the law says one thing and their own all-things-except-the-law-considered judgment says something else. There is less research on this question than we might wish, and this is not the occasion to delve into it, in part because I have done so previously and elsewhere.[34] But even if people believe that there is an obligation to obey those laws with which they disagree, it should be no more surprising that this abstract belief is under-reflected in actual practice than there is that the abstract belief in the desirability of weight loss is under-reflected in actual practice.

This is not to say that the claims about the relationship between an opportunity to object and the incidence of compliance have been falsified. It is only to say that they have not been established, and that the research that might be understood to establish them turns out not to do so. The question of descriptive or sociological legitimacy is important, but we must distinguish a description of belief from a description of behavior. Even if it is the case that freedom of speech increases the degree of belief in the legitimacy of law, and belief in the obligation to obey disagreeable laws, on the question of whether freedom of speech increases actual compliance with laws with which people disagree, the best we can do is simply to say that we do not know.

IV.

The relationship between a regime of freedom of speech and political legitimacy is important, even more important—or at least more pervasive—than the question of hate speech. The larger issue is the soundness of what we might label the “due process” argument for freedom of speech. Just as most developed legal traditions consider it a requirement of procedural justice that those who are to be punished or otherwise sanctioned by the legal system have an opportunity to speak on their own behalf prior to the imposition of the sanction—we call it procedural due process in the United States, and the English refer to audi alteram partem (hear the other side) as a fundamental principal of natural justice[35]—so too might we generalize from this to the possibility that something analogous to due process grounds the right to speak out against even a general law before one is bound by the force of the state to obey it.

In considering the legitimacy based due process argument for freedom of speech, Weinstein has valuably advanced the inquiry by urging us in considering this question to distinguish between normative and descriptive/sociological legitimacy. But once we do address the vital topic of descriptive legitimacy, we see that there is a difference between the empirical fact of a belief in legitimacy and the empirical fact of the behavior that might flow from such a belief. Because theorists from Meiklejohn to Weinstein have supplemented their broader normative claims with more or less specific empirical claims that one justification for freedom of speech is that it will increase the degree of sanction-independent legal compliance with laws with which people disagree, this Symposium has seemed the ideal occasion to address that particular claim. The claim is interesting and important, and, if sound, would provide another foundation for a democracy-focused account of freedom of speech. But at the moment we have little basis for believing it sound, even if it were good were it to be so.

         *    David and Mary Harrison Distinguished Professor of Law, University of Virginia.

        [1].    Alexander Meiklejohn, Free Speech and Its Relation to Self-Government 10–11 (1948). See also Alexander Meiklejohn, Political Freedom: The Constitutional Powers of the People 100 (1960).

        [2].    Ronald Dworkin, Foreword, in Extreme Speech and Democracy vii (Ivan Hare & James Weinstein eds., 2009). Dworkin’s views on democratic legitimacy are elaborated in Justice for Hedgehogs (2011).

        [3].    Robert C. Post, Citizens Divided: Campaign Finance Reform and the Constitution (2014); Robert C. Post, Constitutional Domains: Democracy, Community, Management 280–88 (1995); Robert C. Post, The Constitutional Status of Commercial Speech, 48 UCLA L. Rev. 1 (2000).

        [4].    James Weinstein, Hate Speech Bans, Democracy, and Political Legitimacy, 32 Const. Comment. 527 (2017).

        [5].    James Weinstein, Free Speech and Political Legitimacy: A Response to Ed Baker, 27 Const. Comment. 361 (2011).

        [6].    An excellent overview of the issues and the literature is Fabienne Peter, Politcal Legitimacy, Stanford Encyclopedia of Philosophy, https://plato.stanford.edu/‌entries/‌legitimacy (2016). See also The Psychology of Legitimacy: Emerging Perspectives on Ideology, Justice, and Intergroup Relations (John T. Jost & Brenda Major eds., 2001). In the legal literature, valuable contributions include Richard H. Fallon, Jr., Legitimacy and the Constitution, 118 Harv. L. Rev. 1787 (2005); Michael L. Wells, “Sociological Legitimacy” in Supreme Court Opinions, 64 Wash. & Lee L. Rev. 1011 (2007).

        [7].    I say “process-based” to distinguish a notion of legitimacy that is largely or entirely dependent on the substantive rightness of an outcome (see, e.g., Philip Pettit, Judging Justice: An Introduction to Contemporary Political Philosophy 75-103 (1980)) from a notion that focuses on the way in which outcomes are generated. Because it is not clear that the former notion adds anything to the idea of substantive rightness (or justice), I understand legitimacy to refer to the way in which outcomes are produced, thus leaving conceptual (and empirical) space for legitimately produced but substantively wrong outcomes.

        [8].    See Stephen Guest, Ronald Dworkin 105 (3d ed. 2013); Tom R. Tyler, Psychology and the Law, in The Oxford Handbook of Law and Politics 711, 716–17 (Keith E. Whittington et al. eds., 2008).

        [9].    On the distinction between the instrumentally and the intrinsically valuable as arguments for democratic decision-making (and, implicitly, for any other form of governmental organization), see William N. Nelson, On Justifying Democracy (1980).

      [10].    Of some relevance here is the distinction between trustee and delegate models of representation. See Suzanne Dovi, The Good Representatative (2007); Hanna Pitkin, The Concept of Representation (1967); Jane Mansbridge, Rethinking Representation, 97 Am. Pol. Sci. Rev. 515 (2003). Under a delegate model, representatives are expected to represent the actual wishes of their constituents, but under a trustee model the representative is to advance her constituents’ best interests, independent of their actual expressed wishes, just as the trustee of an estate should aim to maximize the value of the estate in the interest of the beneficiaries but need not take instructions from the beneficiaries as to how to do so. The value of freedom of citizen communication outside of the electoral process is thus more closely connected with a delegate than a trustee model, and arguments for free speech founded on continuous communication between citizens and their representatives thus implicitly incorporate much of the delegate model of representation. See Frederick Schauer, Constitutions of Hope and Fear, 124 Yale L.J. 528 (2014) (book review).

      [11].    See Cass R. Sunstein, Democracy and the Problem of Free Speech 241-43 (1995).

      [12].    For Meiklejohn, the argument for freedom of speech as a consequence of compelling people to obey governmental decisions with which they disagree produced the conclusion that only political speech – speech directed to questions of governmental organization or policy – was protected by the First Amendment. Meiklejohn, supra note 1. See also Alexander Meiklejohn, The First Amendment is an Absolute, 1961 Sup. Ct. Rev. 245; A considerable amount of subsequent commentary has engaged with Meiklejohn on this question, often arguing that a great deal of art and literature, among other topics and genres, is also relevant to public policy, causing Meiklejohn’s distinction between the political and the personal to collapse. See, e.g., Harry Kelven, Jr., The New York Times Case: A Note on the “Central Meaning of the First Amendment,” 1964 Sup. Ct. Rev. 191; Robert C. Post, The Constitutional Concept of Public Discourse: Outrageous Opinion, Democratic Deliberation, and Hustler Magazine v. Falwell, 103 Harv. L. Rev. 601 (1990). This is an interesting and important debate, but its relevance to the relationship between legitimacy and obedience is at best tangential, and so I will do no more than note it here.

      [13].    The seminal source on the availability heuristic is Amos Tversky & Daniel Kahneman, Availability: A Heuristic for Judging Frequency and Probability, 5 Cog. Psych. 207 (1973). See also Shelley E. Taylor, The Availability Bias in Social Perception and Interaction, in Judgment under Uncertainty: Heuristics and Biases 190 (Daniel Kahneman et al. eds., 1982).

      [14].    See Ziva Kunda, The Case for Motivated Reasoning, 108 Psych. Bull. 489 (1990). See also Peter H. Ditto et al., Motivated Moral Reasoning, 50 Psych. Learning & Motivation 307 (2009); Keith E. Stanovich et al., Myside Bias, Rational Thinking, and Intelligence, 22 Current Directions in Psych. Sci. 259 (2013).

      [15].    My reluctance to situate the legitimacy question within the concrete, controversial, political, and highly salient questions about hate speech is also part of a larger concern about the risks of making (or arguing about) general policies or general principles in the looming shadow of immediate and important problems. See Frederick Schauer, Do Cases Make Bad Law,? 73 U. Chi. L. Rev. 883 (2006); Frederick Schauer & Richard Zeckhauser, The Trouble with Cases, in Regulation vs. Litigation: Perspectives from Economics and Law 45 (Daniel P. Kessler ed., 2011).

      [16].    Weinstein, supra note 4, at 527.

      [17].    See, for example, most of the contribution in Speech and Harm: Controversies Over Free Speech (Ishani Maitra & Mary Kate McGowan eds., 2012). On harm and free speech more generally, see Frederick Schauer, Harm(s) and the First Amendment, 2011 Sup. Ct. Rev. 81.

      [18].    To observe that there are empirical questions about the nature of the harms caused by hate speech is not to deny that there are also conceptual questions. For example, the emotional and psychological consequences of a race-based insult directed at an individual are different in kind (putting aside questions of degree) from race-based incitements to violence, discrimination, and other forms of independently unlawful activity. See Frederick Schauer, The Phenomenology of Speech and Harm, 103 Ethics 635 (1993); Schauer, supra note 17.

      [19].    Weinstein, supra note 4, at 530.

      [20].    See Frederick Schauer, Profiles, Probabilities, and Stereotypes 159–67 (2003).

      [21].    For purposes of the example, nothing turns on whether the prohibition targets only IRS employees or others who might be in possession of such information. As a matter of existing First Amendment doctrine, publication of unlawfully obtained information by one who was not a participant in the original illegality is protected. See Bartnicki v. Vopper, 532 U.S. 514 (2001); Landmark Communications, Inc. v. Virginia, 435 U.S. 829 (1978); New York Times Co. v. United States, 403 U.S. 713 (1971) (per curiam). But even if the prohibition in my hypothetical example extended to everyone in possession of the algorithm, it is difficult to see how this would affect the obligation of citizens to pay their taxes.

      [22].    On the possibility that hate speech is constitutive rather than causal of discrimination, see Rae Langton, Sexual Solipsism: Philosophical Essays on Pornography and Objectification (2009); Catharine MacKinnon, Only Words (1993). The claim is controversial, but does bear an interesting affinity with the conclusions about discrimination in Brown v. Board of Education, 347 U.S. 483 (1954). See Charles R. Lawrence, If He Hollers Let Him Go: Regulating Racist Speech on Campus, 1990 Duke L.J. 431.

      [23].    Indeed, although it is often stated that government is suppressing opinions with which it disagrees, actual instances of suppression justified on such grounds are virtually non-existent. When government suppresses, it almost always claims that it is doing so to prevent some harm, and not just to suppress disagreement. Government may frequently be mistaken in that assessment, and it is important to discuss the soundness (or not) of such claims. But to assume at the outset that the basis for suppression is disagreement is to frame the issue in such a way as to avoid precisely such discussion.

      [24].    See, e.g., Chaim Gans, Philosophical Anarchism and Political Disobedience (1992); Robert Paul Wolff, In Defense of Anarchism (1970); A. John Simmons, Moral Principles and Political Obligations (1979); A. John Simmons, Philosophical Anarchism, in For and Against the State: New Philosophical Readings 19 (John T. Sanders & Jan Narveson eds., 1996); M.B.E. Smith, Is There a Prima Facie Obligation to Obey the Law?, 82 Yale L.J. 950 (1973).

      [25].    As maintained by a long line of thinkers, from Socrates to Locke to Rawls and beyond. See William Edmundson, State of the Art: The Duty to Obey the Law, 10 Legal Theory 215 (2004); Leslie Green, Law and Obligations, in The Oxford Handbook of Jurisprudence and Philosophy of Law 514 (Jules Coleman & Scott Shapiro eds., 2002); George Klosko, The Moral Obligation to Obey the Law, in The Routledge Companion to the Philosophy of Law 511 (2012).

      [26].    To put the same issue in different words, the question is about the identity of the properties that an official directive (perhaps calling it a “law” is question-begging) most possess in order for it to ground a content-independent (prima facie) obligation of compliance compliance. It is worth emphasizing that even the philosophical anarchist believes that there is an obligation to refrain from murder, sexual assault, theft, dangerous driving, and the like, but believes that this obligation arises from those acts being morally wrong independent of their legal prohibition. As a result, it is useful in isolating the issue to bear in mind the image of the person who believes that the law’s prohibitions or mandates are mistaken. If there is an obligation to obey the law qua law, then the subject who believes a law to be mistaken, whether as a matter of morality or policy or something else, would still have a (sanction-independent as well as content-independent) obligation to comply.

      [27].    On the “more acceptable” framing, which is somewhat ambiguous about whether it refers to what people will think and what people will actually do, see Amy Gutmann & Dennis Thompson, Why Deliberative Democracy 10 (2004).

      [28].    See especially Tom R. Tyler. Why People Obey the Law (2d ed., 2006). See also Tyler, supra note 8; Tom R. Tyler, Compliance with Intellectual Property Laws: A Psychological Perspective, 29 NYU J. Int’l Law & Politics 29 (1997); Tom R. Tyler, Beyond Self-Interest: Why People Obey Laws and Accept Judicial Decisions, The Responsive Community 44 (Fall 1998). Of special relevance is Tom R. Tyler, Understanding the Force of Law, 51 Tulsa L. Rev. 507 (2015), in which Tyler responds to the criticisms of some of his conclusions that I offered in Frederick Schauer, The Force of Law (2015), and which I draw on here. This is not the place to continue my debate with Tyler, but it would be wrong to deny the existence of that debate or his disagreements—some conceptual and some empirical—with some of the arguments that I offer here.

      [29].    See Tyler, Compliance with Intellectual Property Laws, supra note 28, at 224; Tyler, Beyond Self-Interest, supra note 28, at 45.

      [30].    Tyler’s research is explicitly and carefully criticized on precisely these grounds in Leslie Green, Who Believes in Political Obligation?, in For and Against the State 1, 10–24 (John T. Sanders & Jan Narveson eds., 1996).

      [31].    Tyler, supra note 28, at 46.

      [32].    With respect to this question, Tyler does not appear to exclude the fear of sanctions, and thus the answers to this question may not be as supportive of his general program and conclusions as he believes. But this is a side issue on the free speech question, so I will say no more about it here.

      [33].    Defenders of (some) hate-speech regulations could (and do) argue that restrictions on invective, epithets, insults, and the like would not interfere with the ability of people to object to laws in more civil terms. This is a more than plausible response to Weinstein and others, but a plausible rejoinder is that the response presupposes too much of a distinction between the propositional content of an objection and the style or language in which the objection is couched. Cf. Cohen v. California, 403 U.S. 15 (1971). Resolving this genuine dispute is not germane to my principal point in this Article, so I will do no more than note it here, and note that a Cohen-inspired conflation between content and style may need more justification and explanation than is typically found in much of the literature objecting to hate-speech laws.

      [34].    See Schauer, supra note 28, at 57–74, 197–205.

      [35].    See Frederick Schauer, English Natural Justice and American Due Process, 18 Wm. & Mary L. Rev. 47 (1976).