Summary fro mm article

1 of 998 DOCUMENTS

Copyright (c) 2002 Wayne State University Law School The Wayne Law Review

Fall, 2002

48 Wayne L. Rev. 1071

LENGTH: 12056 words

FOUNDATIONS OF A THEORY OF HATE SPEECH

NAME: J. Angelo Corlett +Robert Francescotti ++

BIO:

+ Professor of Philosophy and Ethics, San Diego State University. Ph.D. in Philosophy, University of Arizona. Editor-in-Chief, The Journal of Ethics: An International Philosophical Review.

++ Associate Professor of Philosophy, San Diego State University. Ph.D. in Philosophy, Syracuse University.

TEXT: [*1071]

Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof; or abridging the freedom of speech, or of the press; or of the right of the people peaceably to assemble, and to petition the Government for a redress of grievances. n1

All persons born or naturalized in the United States, and subject to the jurisdiction thereof, are citizens of the United States and of the State wherein they reside. No State shall make or enforce any law which shall abridge the privileges or immunities of citizens of the United States; nor shall any State deprive any person of life, liberty, or property, without due process of law; nor deny to any person within its jurisdiction the equal protection of the laws. n2

[*1072]

I. Introduction

Much has been written in recent years on hate speech, n3 focusing mostly on whether or not, in the context of United States constitutional law, it ought to be prohibited. Indeed, these discussions have taken at least two main turns: one toward hate speech in general and the other toward hate speech on university campuses, n4 each concentrating mainly, though inexclusively, on racist and/or sexist hate speech. n5 Many believe that the problem of hate speech is, among other things, one of balancing the constitutional value of freedom of expression guaranteed by the First Amendment, on the one hand, with the value of equal protection guaranteed by the Fourteenth Amendment, on the other. n6 As Owen Fiss avers, "The regulation of hate speech, . . . forces the legal system to choose between transcendent commitments-liberty and equality-and yet the Constitution provides no guidance as to how that [*1073] choice should be made." n7 If Thomas Scanlon is correct in arguing that "[f]reedom of expression . . . is an instance of a more general problem about the nature and status of rights, n8 then hate speech, as an instance of freedom of expression, also concerns the nature and status of rights.

We do not attempt to say which Amendment (the First or the Fourteenth) ought to be lexically prior to the other, for that would require the statement and defense of a full-blown theory of justice, n9 which is beyond the scope of this project. This is not to deny either that equality is a compelling interest of the state, n10 or that the answer to the problem of hate speech and freedom of expression more generally requires us to balance competing interests or goods, n11 or that the interests at stake are those of individuals and those of the community. n12 Nor is it a denial of the significance of the concept of equality for the general debate over hate speech and freedom of expression. In fact, if Fiss is correct that "we have no principled way of resolving the conflict between liberty and equality," n13 then arguments between critical race theorists and their detractors about whether or not considerations of equality trump those of freedom may in the end beg questions on either side of the debate. Thus it is important to explore avenues in which freedom of expression and equal protection do not have to be pitted [*1074] against one another.

Philosophically speaking, there are several competing positions on the regulation of hate speech. One is that hate

Philosophically speaking, there are several competing positions on the regulation of hate speech. One is that hate speech ought not to be regulated because it is protected by the First Amendment and that only actions that harm qualify for legal regulation. This view seeks to draw a clear distinction between speech and action in its seeking to protect individual freedom of expression at (seemingly and sometimes unfortunately) any cost. Construing freedom of expression as an absolute right, the view held by many U.S. citizens when issues of freedom of expression arise, is perhaps meant to be captured by the adage: "Sticks and stones may break my bones, but words will never hurt me." Another position is that hate speech ought to be outlawed because it harms, and that political liberalism is problematic insofar as it respects individual freedom of expression over equal protection. This second view sees freedom of expression as a prima facie right, which can be overridden by certain considerations. The first of these positions on hate speech is what we shall refer to as "First Amendment Absolutism," while the second is articulated by some critical race theorists such as Kimberle Crenshaw, Richard Delgado, Charles Lawrence and Mari Matsuda. n14

We will argue that First Amendment Absolutism is wrong to assume that words cannot constitute actions that can harm. n15 Insofar as words can and do wound others, they ought to be regulated since they fall under one or more existing categories of unprotected expression. However, some critical race theorists are incorrect to think that political liberalism is problematic in that it cannot accommodate the regulation of hate speech. n16 Our view, unlike some others, does not assume that all instances of hate speech harm, or that all speech that harms ought to be regulated. We will use John Stuart Mill's Harm Principle to argue, on behalf of liberalism, in favor of the regulation of at least some hate speech that harms. n17 In the context of his defense of freedom of expression, Mill articulates his Harm Principle (from On Liberty) as follows: "the only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent [*1075] harm to others." n18 We interpret Mill's claim to mean that the prevention of harm to others is both a reason, and the only good reason to restrict behavior, making harm (and, in some cases, either the attempt or intent thereof) a necessary condition for legal prohibition of behavior. But is harm a sufficient condition for legal prohibition?

David Lyons refers to the Harm Principle as the "Principle of Liberty:"

The Principle of Liberty permits some "trade-offs," but it never sanctions the imposition of burdens on some for the sake of others' positive benefits. No benefits beyond harm-prevention can justify coercion under the Principle of Liberty. The trade-offs it allows are these: loss of liberty (plus some incidental harm by way of side effects or social costs of enforcement) in order to prevent or eliminate greater harm to others. n19

Of Mill's discussion of justice in Utilitarianism, n20 Lyons writes:

[H]e says there that "Justice is a name for certain classes of moral rules which concern the essentials of human well-being more nearly, and are therefore of more absolute obligation, than any other rules for the guidance of life." Mill plainly believes that injustice and unfairness threaten the most vital of human interests, and that the protection of those interests is the rationale behind the corresponding principles. He thus regards considerations of justice and fairness as rooted in, not as independent of, harm-prevention. More important, Mill understands these principles as exerting independent weight in moral reasoning, relative to welfare arguments. n21

If Lyons' interpretation of the Harm (or Liberty) Principle is plausible, [*1076] then that principle is designed to protect certain vital interests. And it is sufficient to justify legal prohibition of certain harms to the extent that (i) those vital interests are severely threatened by such harms, and (ii) the agent is morally responsible for the harm produced (a discussion of (ii) follows later in the paper). n22

In light of the Harm Principle, we argue that, while First Amendment Absolutism is incorrect to favor freedom of speech as a putatively absolute right over equal protection (because the protection of even freedom of speech that harms is arbitrary), some critical race theorists are incorrect to think that political liberalism lacks the conceptual framework to regulate expressions that harm (such as certain forms of hate speech). n23

There is another attempt to show how liberalism can in some way prohibit hate speech. Concerning hate speech on university campuses, Andrew Altman argues that the liberal principle of "viewpoint neutrality" can be used to regulate some, but not all, hate speech. n24 "The liberal principle of viewpoint-neutrality holds that those in authority should not be permitted to limit speech on the ground that it expresses a viewpoint that is wrong, evil, or otherwise deficient." n25 But Altman argues that insofar as some hate speech subordinates in a morally illicit way, n26 it violates viewpoint neutrality. n27 So, he concludes, "[T]here is a defensible liberal middle ground between those who oppose all campus hate-speech regulation and those who favor the sweeping regulation of [*1077] such speech." n28 But we shall argue that it is the liberal Harm Principle (in addition to the principle of viewpoint neutrality) that allows liberalism to regulate hate speech that harms or sets back vital interests as defined by Lyons. n29 Such interests are those that are basic to a person's self- respect, dignity, and autonomy. These would include, but are not limited to, our interest in not being treated in racist or sexist ways.

The law already finds some kinds of speech and expression unprotected by the First Amendment. And Joel Feinberg argues that Mill's Harm Principle could be used to justify restrictions on expressions which constitute

defamation or "malicious truth," invasions of privacy, speech which causes panic, provokes retaliatory violence, or incites others to insurrection. n30 Should hate speech be permitted in a politically liberal society such as the U.S.? Or, is hate speech legitimately subsumable under some existing category of unprotected expression? n31 Or, should there exist a separate category of unprotected expression under which hate speech falls?

We understand that discussions of hate speech (as with speech in general) in philosophy and legal studies have for the most part ignored the importance of analyzing philosophically the nature of hate speech. Even when a few such discussions have attempted to define the nature of hate speech, they have for a variety of reasons failed in doing so. n32 After we enumerate the desiderata for an adequate theory of hate speech, we provide a plausible conceptual analysis of hate speech that [*1078] can be used for purposes of law. n33 Subsequently, we address some of the more plausible arguments against the prohibition of hate speech in a liberal society such as the United States. n34 In so doing, we conclude that the liberal Harm Principle may be used to justify the legal regulation of some hate speech that harms. n35 This implies that liberals who are First Amendment Absolutists are wrong in arguing that the individual's use of hate speech is protected by the absolute right to freedom of expression. n36 On the other hand, critical race theorists need not argue that the liberal framework is impoverished because it cannot accommodate the legal restriction of hate speech.

In order to determine whether or not hate speech ought to be protected by the First Amendment, we first need to know what hate speech is. But to know the nature of hate speech requires an understanding of the various kinds of hate speech. We address these issues in Section III. As a preliminary task, it is important to consider some of the desiderata for an adequate theory of hate speech.

II. Some Desiderata for an Adequate Theory of Hate Speech

A first desideratum of a theory of hate speech is that it define the nature of hate speech, for without a working definition of "hate speech," there is a strong likelihood of equivocation in arguing about it. This is the Desideratum of Definition. It should be used to judge the adequacy of a theory of hate speech. For a theory of hate speech is only as good as the definition of "hate speech" that it employs.

A second desideratum of a theory of hate speech is that it distinguish and address separately the question of the nature of hate speech from the questions of its legal and moral statuses. This distinction minimizes the chances of defining hate speech to bias unwarrantedly questions of its legality or [*1079] morality. This is the Desideratum of Objective Analysis.

A third desideratum of a theory of hate speech, we argue, is that it respect the complexities of competing legal and moral principles that undergird the law. Important considerations both for and against the outlawing of hate speech must be juxtaposed to one another, providing an honest and complicated web of philosophical considerations that ought to pull us in different directions, based on various constitutional and extra-constitutional considerations. This is the Desideratum of Theoretical Adequacy, which holds that an adequate theory of hate speech not run afoul of the most plausible legal, moral, psychological, and political principles. It should be recognized that a theory of hate speech rests on the plausibility of the moral and legal principles that ground it.

A fourth desideratum of an adequate theory of hate speech is that it take seriously the practical nuances and difficulties in the law. Such a theory ought not to run counter to basic legal principles against, say, ex post facto law, or cause more confusion than clarity in the law. This is the Desideratum of Legal Practicality. After all, a theory of hate speech ought not to result in a recommendation to the law that is impossible for an agent to perform, violating the "ought implies can" dictum. Nor should it recommend to the law that which is morally repugnant.

A fifth desideratum is that a theory of hate speech include an account of how offenses to anti-hate speech tort and/ or criminal legislation can and should, or cannot be or should not, be handled by tort and/or criminal law. This is the Desideratum of Adjudication. It might turn out that anti-hate speech legislation is only implementable in tort law, or criminal law, or neither, or both. In either case, it is desired that a comprehensive theory of hate speech explain the extent, if any, to which anti-hate speech laws are a matter of tort law, criminal law, neither, or both. For instance, while Delgado proffers the remedy for racial hate speech harms in terms of civil court action in tort law, n37 Matsuda argues that a narrow class of racial hate speech that causes the most serious harm should be handled in a criminal justice setting. n38

A theory of hate speech is adequate to the extent to which it satisfies each of these desiderata. In our attempt to define the nature of [*1080] hate speech in general, we address concerns related to the first two desiderata, and to some extent, the third and fourth. We do not address the Desideratum of Adjudication in that ours is a philosophical analysis of the nature of hate speech, rather than an investigation into the problems of how U.S. law ought to handle such cases. Our conceptual "foundations of a theory of hate speech" are meant to serve as part of the philosophical basis of an analysis of how various legal systems might implement sound thinking about hate speech.

III. The Nature of Hate Speech

While there are many characterizations of the nature of campus hate speech, analyses of the nature of hate speech in general are notably absent from the philosophical literature. Consistent with the Desideratum of Objective Analysis,

we insist on a distinction between the question of the nature of hate speech, on the one hand, and of the moral and legal justificatory statuses of hate speech, on the other. Discussions of hate speech must not unwarrantedly conflate these questions, causing confusion by illicitly sneaking into the definition of "hate speech" a notion that begs the question either for or against its justification. With this in mind, we shall try to formulate a neutral definition.

In a recent article, Susan Brison defined "hate speech" as "speech that vilifies individuals or groups on the basis of such characteristics as race, sex, ethnicity, religion, and sexual orientation, which (1) constitutes face-to-face vilification, (2) creates a hostile or intimidating environment, or (3) is a kind of group libel." n39 Since conditions (1)-(3) are disjoined, no one of them is meant to be a necessary condition. n40 All that is required is that at least one of the three conditions is satisfied (and not necessarily the same one) on any particular occasion of hate speech. n41 However, even this weaker requirement is too strong. Suppose that an European American motorist becomes angry with a Latino motorist and mutters to herself, "Learn how to drive, you stupid spic!" Suppose, moreover, that no one hears this remark (she is alone in her [*1081] car with the windows rolled up) and no one notices her angry gestures. In this case, the remark is not face-to-face, it does not create a hostile or intimidating environment, and it does not qualify as group libel. But it does seem to count as hate speech. n42

It might be suggested that while a particular instance of hate speech might fail to create a hostile and intimidating social environment, it is part of the concept of hate speech that it tends to create a hostile and intimidating social environment. This important fact about hate speech, it might be argued, is precisely what makes it so hateful, and if so, then a suitably weakened version of (2) should be considered a necessary condition for hate speech.

But recall the Desideratum of Objective Analysis. An analysis of hate speech should not beg any important questions about its legal and moral status. In particular, we should not assume, in the very definition of "hate speech," that it warrants legal restriction, and we would come very close to making this assumption if we defined hate speech as tending to create a hostile social environment. While it may be true (and is true, as we argue later) n43 that hate speech often has significant negative consequences, the truth of this claim does not follow from the very definition of hate speech. Whether the claim is true is an empirical question.

Suppose that we eliminate (1)-(3) and define hate speech simply as speech that vilifies on the basis of such characteristics as race, sex, ethnicity, religion, and sexual orientation. While this is an improvement, a worry remains. The phrase "such characteristics as" allows that hate speech might be based on characteristics other than those explicitly mentioned. For example, it might be argued that language minimizing the rights of non-human animals creates and sustains social attitudes and practices that are hostile and cruel toward the animals. So we may wish to include the category, species membership, to allow that certain forms of "speciesist" language qualify as hate speech. n44 And if we allow [*1082] that negative remarks based on membership in a religious group might count as hate speech, we should also allow that negative remarks based on one's membership in a political group could qualify as hate speech. But, then, what type of group is not included? Might there be hate speech on the basis of one's fashion or automobile preferences?

It might be argued that the group characteristic must be one on the basis of which people already discriminate in regular and significant ways. While it is plausible to suppose that non-human animals have been systematically dominated by humans, no one can plausibly argue that such widespread oppression has occurred on the basis of the type of car one likes to drive.

One problem with this suggestion is that it fails to anticipate forms of oppression that have not yet occurred. Suppose that at some future time, a large portion of the population comes to believe that people with protruding navels are intellectually and morally inferior to those who lack this characteristic; as a result, those with protruding navels become systematically oppressed in many significant ways. Although this situation is highly unlikely to occur, the point remains that if it were to occur, the term "hate speech" would seem to apply to the negative remarks directed toward the "outies." So perhaps our definition should allow that hate speech might be based on membership in any type of group.

The definition should also make clear that hate speech need not take linguistic form. Recall, for example, the Stanford "Beethoven" incident. n45 At a party, two white students were arguing with a black student about whether Beethoven was of African decent (the white students vigorously denied that he was). n46 The next day, a defaced poster of Beethoven was found on the door of the black student's dorm room; the picture of Beethoven was given various "Sambo" like features, e.g., wild curly hair, big lips, and red eyes. n47 This is an instance of hate speech, though no words were used. Presumably, Brison is using the word "speech" broadly enough to include the defaced Beethoven [*1083] poster, but to make it perfectly clear that the defacing qualifies as hate speech, we suggest the following definition: Hate speech consists in symbolism, linguistic or otherwise, that expresses intense antipathy toward some group or toward an individual on the basis of membership in some group.

Hate speech includes as targets individuals, on the one hand, and collectives, on the other. Furthermore, the perpetrators of hate speech may be individuals or groups. This presents at least four kinds of hate speech against persons: (i) an individual's hate speech directed at an individual; (ii) an individual's hate speech directed at a group; (iii) a group's hate speech directed at an individual; and (iv) a group's hate speech directed at a group. An unfortunately

common example of hate speech by an individual that is directed at another individual might be when an individual refers to a particular Latino as a "worthless spic," a particular gay person as a "filthy fudge-packer," or a particular woman as a "stupid fucking bitch." An example of an individual's hate speech directed at a group would be when a person refers to a particular group in a hateful way: "Those goddamned gooks!" But groups also engage in such activities, most notably, perhaps, the KKK and neo-nazi groups as they publish and utter various racist expressions against a number of individuals because they are, say, African Americans, Native Americans, Jewish Americans, Latinos, Asian Americans, etc. Typically, however, such groups express hatred toward other groups as they speak, not of individuals within such groups, but of the groups themselves. These utterances are instances of hate speech to the extent that they express intense antipathy toward their targets.

Another distinction is between self-hate speech and hate speech that targets others. The hate speech literature has devoted its attention to examinations of hate speech that harms others, typically those belonging to groups to which the perpetrator does not belong. But hate speech can be an expression of individual self- hatred, as when an African American uses "nigger" to refer to herself in a demeaning and derogatory way. Or, it might obtain as an instance of collective self-hatred when an African American refers to other African Americans as "niggers." Although the term "nigger" is often used by some African Americans as a category of affection (e.g., "He's my nigga!"), this would not preclude the possibility of those African Americans who would use the term, perhaps even unintentionally, as an expression of self- hatred. One needs to ask oneself if such African Americans would ever refer to themselves as "white trash" if they were part of the privileged classes of [*1084] European American citizens. If not, then there is some reason to wonder about the extent to which some such African Americans, in using "nigger" to refer to themselves and other members of the African American communities, are engaged in intentional or unintentional expressions of self- hatred. Given that one can cause harm to oneself, n48 or to a group to which one belongs, it is helpful that the definition of "hate speech" reflect this fact. Our definition of "hate speech" does not rule out that an expression of hate speech might be the result of self-hatred, whereas (1) of Brison's definition seems to not make room for expressions of "self- hate speech." n49 This is because, on a strong construal of "face-to-face" in (1) of Brison's definition, expressions of self-hate speech make no sense in that hate speech can be directed only toward another. On a weaker reading, (1) requires only that the target be aware of the speaker's utterances. This weaker condition is often satisfied in cases of self-hate speech. However, it need not be satisfied since one can be unaware of the fact that one is engaged in self-hate expression.

It is also important to bear in mind that in the typical case of genuine hate speech, it is the hateful use, not the hateful mention, of terms that is crucial. One might say something such as "Never use the word 'nigger' because it is insulting to and disrespectful of black folk." There seems to be nothing wrong with mentioning the word in this way, especially in the context of teaching others to not use the term. But using the word "nigger" involves one's believing in some meaningful sense that the term actually applies, in some or all of its negative connotations, to some African American(s). Thus, referring to someone as a "nigger" (use) is a racial insult, and if it expresses intense antipathy, it would constitute a form of racist hate speech. So generally, hate speech involves the use, not the mention, of terms. But this need not always be the case, as in instances where one says, in a hateful way, "You know, the word 'nigger' really applies to you!" (where "you" is meant to refer to an African American person). This would constitute racist use and mention of the vitriolic term. n50

[*1085]

Our definition says nothing about how hate speech affects the audience. For instance, our definition does not require (for the reasons mentioned earlier) that hate speech create a hostile and intimidating social environment. n51 But, should the definition at least require that others (intended or unintended audiences) recognize the hatefulness of the remark, whether or not it has any negative effect upon them? In describing the nature of an illocutionary act, J. L. Austin writes:

I cannot be said to have warned an audience unless it hears what I say and takes what I say in a certain sense. An effect must be achieved on the audience if the illocutionary act is to be carried out. . . . Generally the effect amounts to bringing about the understanding of the meaning and of the force of the locution. So the performance of an illocutionary act involves the securing of uptake. n52

However, even a speech act theorist can, and should, reject the idea that illocutionary acts must secure uptake. As Peter F. Strawson notes, "a man may, for example, actually have made such and such a bequest, or gift, even if no one ever reads his will or instrument of gift." n53 In this example, there is no audience; so there is no audience with respect to which the speech act secures uptake. And even if an audience were present, the audience need not realize either the content of what was said (let alone that the content qualifies as intensely antipathic), or the illocutionary force of the utterance. Suppose the speaker uses a sentence to make a statement (illocutionary force) to the effect that gays are immoral (content). If the audience does not understand what "gay" means in this context, then they will not recognize the content of what was said. An audience member might not speak the language at all, in which case, she might not even recognize that a statement was made. Even when we hear our native language, it is not always clear to us whether the speaker has made a statement, asked a question, or issued a command. Thus, our definition, like Brison's, does not require that the [*1086] hateful speech act be recognized by the audience as having either the content or the

illocutionary force that it has. n54

Are there any requirements regarding the state of mind of the speaker? For example, must the hateful speech act be performed intentionally? There are two separate claims to consider here: whether

(a) one can utter a sentence with content C without intending to utter a sentence with C, and

(b) one can utter a sentence that expresses intense antipathy without intending to utter something that expresses intense antipathy.

To show that proposition (a) is true, consider a simple slip of the tongue, where the speaker accidentally utters a sentence that he did not intend to utter. One might argue that, in the case of hate speech, slips of the tongue are really "Freudian" slips-i.e., the speaker is expressing an unconscious intention. But we can also imagine that the slip is one the speaker in no way intends to make. For example, consider a woman who does not speak Spanish but wishes to compliment a Spanish- speaking waiter. After just learning how to say " Tu eres magnifico, Paco!" ("You are magnificent, Paco!"), she instead utters the sentence " Tu eres un menso puto!" ("You are a stupid fucker!"). She did not intend (consciously or unconsciously) to utter a sentence with that content.

But is proposition (b) true? Does it make sense to say that an utterance expresses a certain emotion (such as antipathy) if the speaker lacks the emotion on that occasion of utterance? To answer this question, it is necessary to distinguish between

(i) uttering a sentence that happens to have hateful content, and

(ii) uttering hatefully a sentence with such-and-such content.

Whether a sentence has hateful content ultimately depends on the typical intentions of those who use that sentence. It would be inappropriate (if [*1087] not false) to say that a sentence has hateful content if speakers seldom have hateful intentions when uttering the sentence. But this does not entail that on every occasion of utterance the speaker has a hateful intention. As an analogy, consider encouragement: one can utter encouraging words without a desire to encourage (e.g., one might not realize that the words one utters are encouraging). Likewise, one can utter words that express hatefulness without having any hateful feelings, such as in the case of the Spanish waiter. The sentence " Tu eres un menso puto!" has hateful content (i.e., it is a sentence that is generally uttered with intense antipathy), but since the speaker does not realize it has hateful content, she does not utter it in hatred. We might even come to learn about the expressive power of words in our native language. This may be a process of self-discovery. Since we can be ignorant of our emotions, we might come to learn that the emotions responsible for some of our prior utterances qualify as hateful emotions. n55 Or it may simply be a matter of learning that the words we used were not the best ones to use to reveal our true emotions. Thus, although action (ii) depends on the speaker's intentions, action (i) does not necessarily incorporate action (ii). n56

Since this analysis requires only that hate speech have intensely antipathetic content, it does not require that hate speech have hateful emotions. Of course, when determining the moral and legal ramifications of a particular utterance of hate speech, the intentions and emotions of the speaker are likely to be an important consideration. n57 But, if hateful emotions are included as a necessary condition for speech to be characterized as hate speech, infusing the subjectivity of emotions will limit the potential to conduct objective analysis.

[*1088]

According to the analysis offered herein, hate speech can take racist, sexist, and speciesist forms. It might also be based on membership in a political or religious group. In fact, it often happens that those on either side of an ethical issue express hateful things toward one another. n58 Given the wide variety of forms of hate speech, not all forms will necessarily bear the same moral and legal status. For example, some instances of hate speech may be morally/legally permissible by the standards of a particular contemporary society. Of course, as Kent Greenawalt recognizes, "drawing the line between rough but acceptable common discourse and what is beyond acceptability is extraordinarily difficult." n59 The following discussion proposes criteria to make this difficult task more manageable.

IV. Should Hate Speech Be Prohibited by Law?

Some critical race theorists n60 have argued that certain words, particularly many racist/sexist words, can "wound" or harm, and that this is sufficient reason to consider them unprotected by the First Amendment. n61 Hate speech tends to encourage fear of violence based on historic violence against certain groups in U.S. society. n62 Moreover, hate speech often produces in its targets such feelings as isolation and a loss of self-confidence. n63 The harm need not be psychological. We believe that the social scientific evidence that some racist/sexist expressions often can and do harm is beyond reasonable dispute. n64 So, on the basis of Mill's Harm Principle, we suggest what can aptly be termed the "Argument from Harm." Hate speech ought to be prohibited to the extent that it significantly harms others.

But what exactly is a harm? Feinberg has analyzed the concept of [*1089] "harms as setbacks to interest[s]," n65 where the illegitimate setting back of an interest or the setting back of a legitimate interest is a bad thing. Lyons

articulates the notion of a harm thusly:

Harms thus concern interests that are readily appreciated; most if not all of these are, at bottom, common to all persons. They are not to be understood in terms of mere existing preferences but rather as conditions that must be satisfied if one is to live well as a human being; they include physical necessities, personal security, social freedom (from oppressive custom as well as others' interference), and a variety of experiences and opportunities for self- development. To the extent that one is denied or deprived of such conditions, one suffers what Mill counts as "harm." n66

Within this understanding of harm, racist hate speech harms to the extent that it sets back a person's legitimate vital interest in receiving a fair opportunity to be educated, to vote, and to carry on her daily activities without harassment from others. Social freedom from harassment is violated in some cases of hate speech, while in other cases one's development as, for example, a graduate student is hindered unfairly by racial slurs directed at that student's ethnic group. The same might be said for certain instances of sexist hate speech that harm. These kinds of hate speech ought to be prohibited to the extent that instances of them set back the legitimate vital interests of persons whose rights to freedom and fair equality of opportunity are violated by the expressions of hatred.

It is important to note that the harm being considered is not restricted to that suffered by the target of the hate speech. If a hateful remark is made against an individual or group of individuals on the basis of race, the harm to be considered also includes that produced in other members of her race and in the community at large. Thus, the Argument from Harm is based on considerations that motivate the more specific "Argument from Public Harms." To the extent that there exists a public good in a respected and healthy social environment, [*1090] certain harmful expressions toward humans or non-humans ought to be deemed unprotected by legislation. (Even accepting the proposition that, as argued above, hate speech can be directed at non-human animals, it would still be an open question as to whether hate speech directed at such non-human entities could constitute a genuine harm. And it is helpful to heed Feinberg's point that public harms need not be harms because of the perceived intrinsic wrongness of the act in question. n67 Instead, such wrongs need only be wrong "for the public" or extrinsically wrong).

It is also important to note that the Argument from Harm does not require abandoning liberalism, for liberalism has the conceptual and constitutional wherewithal to accommodate the restriction of harmful words. n68 While critical race theorists might challenge liberalism's degree of devotion to "viewpoint-neutrality" or "reasonable pluralism" in a democracy, n69 they need not challenge cultural, religious, or political pluralism itself. John Rawls has convincingly argued that reasonable pluralism in a "realistic utopia" would have limits to its toleration of the values of "outlaw states" and the same would hold true for outlaw groups within a liberal society. n70

Recall that Mill's Harm Principle disallows exercises of freedom that harm others in certain ways. n71 And if, as Frederick Schauer argues, "there is no non-question-begging way of basing a principle of free speech on the descriptive claim that the category of speech is less harmful as a category than the category of conduct," n72 then speech that [*1091] harms should not be protected by the First Amendment, just as harmful actions do not have constitutional protection. This holds true unless there is a plausible reason why-even assuming that there exists a clear distinction between words and actions n73-harmful actions are to be regulated by criminal statutes while harmful words ought not to be outlawed.

Arguably, many forms of hate speech fall under one or more of the existing categories of constitutionally unprotected expression. There are arguments above and beyond the idea that hate speech causes psychological harm to "minorities," justifying its legal prohibition under the Equal Protection Clause of the Fourteenth Amendment. For example, certain instances of hate speech ought to be regulated to the extent that they invade privacy, are obscene or perjurous, cause panic, provoke retaliatory violence, or incite others to insurrection. These are harms in a robust sense, consistent with Feinberg's construal of Mill's conception of harm as a setback of a legitimate interest, n74 and with Lyons' notion of harm as the setback of a vital interest. n75 So, the significance of our liberal Argument from Harm is that free speech absolutists who oppose restricting hate speech are placed in an uncomfortable position in that their own liberalism permits the suppression of hate speech that harms, while some critical race theorists find it controversial that liberalism can indeed make room for their basic claim.

However, the fact that an instance of speech harms is not, by itself, [*1092] a sufficient reason to condemn and punish it within the legal system. The harm must qualify as significant and unjust; mild unpleasantries do not warrant legal sanction. The degree and type of harm produced by a certain instance of hate speech is an empirical question, the answer to which might be revealed by evidence from the social and natural sciences.

Also, to qualify for legal sanction, the speech act must be performed intentionally, knowingly, and voluntarily. n76 Consider a case where a U.S. citizen is approached by a new visitor to the U.S., Vincenzo, who, being from a rural town in Italy, knows very little English. The U.S. citizen befriends him, and, thinking deviously, decides to play a trick on him. In answer to his question about how to ask women of the U.S. out for dates, the U.S. citizen teaches him to say, in pretty clear English: "You are a stupid fucking bitch of a whore, and I hope you die!" Vincenzo then approaches a woman to whom he is attracted and says: "You are a stupid fucking bitch of a whore, and I hope you die!" The woman,

of course, slaps him and calls for the authorities, leaving the man wondering why U.S. women are so harsh in response to his well-intentioned advances. This example demonstrates that speaker intention and knowledge are crucial for moral and legal responsibility concerning the use of hate speech. Not only did our friend Vincenzo not intend to say hateful things to the woman, but he had no knowledge at all that he was saying something that could very well be interpreted as being hateful. Indeed, his intention was to compliment her. Since he neither knew what he was saying in English nor intended to utter hateful words, we would hardly say he was morally or legally responsible for his comment, even if it resulted in harming the woman. Furthermore, even if Vincenzo knew that what he was saying to the woman was hateful, he would not be responsible for resultant harm to her if he were forced to utter those words. So, the following are among the necessary conditions for the legal liability of some instance, S, of hate speech: the agent intends to engage in speech that has antipathic content, she knows that S expresses antipathy, and she engages in S voluntarily.

But these conditions are not jointly sufficient. Calling someone a "stupid yuppie!" usually does not cause significant harm. But, suppose that, unbeknownst to the speaker, the target of the remark is so emotionally fragile that the remark has very damaging psychological [*1093] consequences. Although the remark produces significant harm, and is made knowingly, intentionally, and voluntary, it seems that the remark does not merit legal sanction. What makes the remark legally permissible can be explained in terms of the "reasonable person" standard; a reasonable person would not expect the remark to have significantly harmful consequences. So an instance of hate speech is legally liable only if the speaker intentionally, voluntarily, and knowingly utters a remark that a reasonable person could expect to produce significant and unjust harm given the circumstances in which the hate speech is expressed. n77 It seems that hate speech uttered or expressed by a person privately n78 and only to herself does not count as restricted speech (normatively speaking), for the simple fact that a reasonable person would not expect it to produce significant harm. However, the more public the forum, the more reasonable it is to expect significant harm.

Perhaps another condition is required. Intuitively, it would appear that the hatred of a criminal by the survivor of rape is morally justified. Even if it were the case that the victim forgave her rapist, it would hardly follow that third parties must forgive him, or ought not to hate him (assumed here is the claim that rape is not only a crime against the victim, but against society as well). One might argue that hate has an appropriate place in our moral dispositions and attitudes. n79 However, [*1094] hatred has a place in our moral dispositions and attitudes only when it is deserved. n80 That a person or group does something unjust to harm others in a serious way makes her or them a prima facie legitimate target for hatred by others. Generally, the more severe one's wrongdoing toward others, the more severe she makes herself a legitimate object of others' hatred. This assumes, of course, that the wrongdoer is at fault in acting intentionally, knowingly and voluntarily, i.e., is a responsible (culpable) agent. n81

Now if it is plausible to argue that hatred is sometimes warranted on moral grounds, then perhaps the law ought to make room for cases of harmful hate speech in such instances. A woman who confronts her rapist with the words, "I hate you, you goddamned bastard, and hope you burn in hell!" is hardly unjustified on moral grounds, and no sensible legal system would want to punish her for what seems to be a natural and justified response to her perpetrator. n82 And this would hold true so long as the psychological harm suffered by the rapist is not greater than the harms suffered by the rapist's victim (assumed here is a plausible principle of proportional punishment). Or suppose the public responds in a hateful manner to a duly convicted child molester who is moving into its neighborhood upon early release from prison due to some technicality. One might argue that this hateful response is also justified.

The distinction between deserved and undeserved harm explains what makes racism and sexism morally abhorrent. It is that such harmful expressions are morally arbitrary in how they treat or portray [*1095] their targets. When harmful hate speech arbitrarily singles out the target based on such characteristics as race and gender, then the expressions harm in an undeserved way. And if someone is significantly harmed in an undeserving way, she is harmed in a way that is not supported by the balance of human reason.

In light of these points, we offer the following analysis of legal responsibility for harms of hate speech. Suppose that an agent engages in some significantly harmful hate speech act, S. Then she is legally responsible for S to the extent that:

(1) she intends to engage in speech that has intensely antipathic content, she knows that S expresses intense antipathy, and she engages in S voluntarily;

(2) a reasonable person could expect S to produce significant harm; and

(3) the harm caused by S is unwarranted.

"Unwarranted" reflects the idea that a court would find the exercise of hate speech unjustified, say, because the target of the expression was undeserving or innocent. "To the extent that" and "significant" allow that such responsibility admits of degrees.

Some, like Frederick Schauer, have argued that a theory of speech, such as one of hate speech, must take into account the fact that even though speech and action can be equally harmful, there is special reason to doubt the viability of anti-hate speech legislation. n83 One such reason is to guard against the state's self- serving tendency to repress expressions that are critical of state policies. n84 The state may do this in either of two ways. First, the state might

exercise the harm of intolerance. n85 Secondly, the state might well mistakenly identify which forms of hate speech should be prevented. n86

Regarding Schauer's second point, his position against the regulation of hate speech presupposes that we cannot distinguish different kinds of hate speech (justified or unjustified). Of course, harm [*1096] itself is not a sufficient reason to regulate hate speech. Perhaps this is one reason why Schauer does not make the Harm Principle foundational to his view of hate speech legislation. However, it would appear that whatever else a theory of hate speech does, it ought to be congruent with the Harm Principle, that is, unless it also provides arguments which would undermine the plausibility of political liberalism itself. So, there should be a non-arbitrary distinction between hate speech that is harmful and unjustified and that which is harmful and justified. And we have provided that distinction. The second part of Schauer's main concern with the regulation of hate speech, then, can be satisfied by the use of our analysis.

The first part of Schauer's concern about the prohibition of hate speech-that the state might exercise intolerance in the process-seems to constitute a slippery slope fallacy. In fairness to Schauer's rather eloquent defense of the use of some slippery slope argumentation in legal contexts, n87 the fact that the state might sometimes be wrong in the way it carries out the law on hate speech regulation hardly justifies refusing to make legitimate law against hate speech that harms. Thus, we contend that Schauer's use of slippery slope argumentation in this context is dubious. If Schauer's concern about state intolerance were taken seriously, then it would seem to suggest the absurd claim that all of law ought to be eliminated because of the fear of state intolerance and abuse of implementation. But, that the state sometimes is wrong in its conviction of accused persons hardly means we ought to give up on prosecuting crimes, or on punishing them. Abuse of the application of law in no way justifies not attempting to enact and enforce good laws.

In part, our rationale for the regulation of hate speech is consequentialist. However, David Lewis uses consequentialist considerations to argue for the opposite conclusion. n88 Writing as if Mill were the founder of the concept of freedom of expression, n89 Lewis [*1097] argues that considerations of social utility favor toleration of expression, as Mill would have it. n90 Granted, considerations of social utility require a significant degree of tolerance regarding expression. However, even a consequentialist can and should place value on human dignity, respect and rights. A society that permits toleration to run roughshod over human dignity, respect and rights to equal protection deceives itself into thinking that undeserved abusive expression is worth the cost of denying many persons what is basic and decent for peace and tranquility. Since social utility consists partly in peace and tranquility, even the consequentialist should agree that toleration of expression has its limits. n91

[*1098]

While consequentialist considerations motivate the argument from harm, there are deontological reasons in support of it as well. Unrestricted tolerance of expression threatens human dignity, respect and the right to equal protection. But, it is arguable that these qualities are intrinsically worthwhile. If this were true, then there would be a good reason (which has nothing to do with social utility) to impose some limits on toleration of expression. So, the argument from harm can be strongly motivated on either consequentialist or deontological grounds.

V. Conclusion

In sum, freedom of expression is a good thing, and it ought to be protected to some extent and within reason by the First Amendment. But as Ronald Dworkin argues,

The First Amendment, . . . cannot be applied to concrete cases except by assigning some overall point or purpose to the amendment's abstract guarantee of "freedom of speech or of the press." . . . Contemporary lawyers and judges must try to find a political justification of the First Amendment that fits most past constitutional practice, including past decisions of the Supreme Court, and also provides a compelling reason why we should grant freedom of speech such a special and privileged place among our liberties. n92

[*1099]

If Karl Marx is correct about the importance of freedom of expression in general, n93 and if Dworkin is right about the nature of the First Amendment, then before us lies the Herculean task of finding normative principles within and external to the law which would ground the law coherently, while at the same time not alienating what we should think about the First Amendment from the remainder of the body of law itself. In so doing, it is not necessary to divine the original intent of the framers and ratifiers of the First Amendment. n94

In setting forth the foundations of a theory of hate speech, we have articulated some desiderata for an adequate theory of hate speech. Subsequent to this, we defined the nature of hate speech in an inclusive but non-question- begging way. Then we considered some arguments for hate speech regulation on both utilitarian and deontological grounds. In the end, though, what counts as harmful hate speech is an empirical matter. Thus, what has been argued for herein must be supplemented with experimental evidence (say, from experimental social psychology) for what science deems in its best lights as harmful hate speech, and just how harmful it is. Such evidence, along with the reasonable person standard in law, can serve well as guides to the extent to which harmful hate speech ought to be prohibited.

First Amendment absolutists who are also liberals wrongfully ignore [*1100] the harm principle which does not

First Amendment absolutists who are also liberals wrongfully ignore [*1100] the harm principle which does not favor the protection of harmful words over harmful deeds. The harm principle provides liberalism with a theoretical framework by which to regulate harmful deeds and words, and this point needs to be recognized by many liberals in their defenses of the protection of hate speech from legal regulation, as well as by some critical race theorists in their criticisms of liberalism. Moreover, the categories of unprotected expression seem to accommodate the forms that harmful hate speech might take. So there is already a political and legal basis for the regulation of at least certain instances of harmful hate speech.

What we have not set out to do is to even outline an account of remedies for wrongful uses of hate speech. This would involve, among other things, providing a justification for which kinds of law (e.g., tort and/or criminal) should regulate instances of harmful hate speech. Such accounts have been given, both by some critical race theorists and by others. n95 Although it is incumbent on us to devise such an account in order to complete this theory of hate speech, we leave that task for another occasion. Our aim here was to articulate an important part of the philosophical groundwork for a theory of hate speech.

Legal Topics:

For related research and practice materials, see the following legal topics: Constitutional LawBill of RightsFundamental FreedomsFreedom of SpeechScope of FreedomConstitutional LawSubstantive Due ProcessCitizenshipConstitutional LawEqual ProtectionScope of Protection

FOOTNOTES:

n1 U.S. Const. amend. I.

n2 U.S. Const. amend. XIV.

n3 Not all scholars use this category to refer to the kinds of expression that are the concern of this project. Some refer to such expressions as racist speech, sexist speech, or more generally as "biased" speech. See Frederick M. Lawrence, Punishing Hate (1999). Still others refer to such expressions as "subordinating speech." See Joan C. Callahan, Speech that Harms: The Case of Lesbian Families, in On Feminist Ethics & Politics 244, 244-45 (Claudia Card ed., 1999).

n4 For a discussion of hate speech on university campuses, see Andrew Altman, Liberalism and Campus Hate Speech: A Philosophical Examination, 103 Ethics 302 (1993); Timothy Shiell, Campus Hate Speech on Trial (1998).

n5 Examples of those who discuss hate speech in terms of racist hate speech include critical race theorists such as Mari J. Matsuda. See Words that Wound (Mari J. Matsuda et al. eds., 1993). Examples of those who construe hate speech primarily, though not exclusively, in terms of gender include Catherine MacKinnon, Only Words (1993). See generally Daniel Jacobson, Freedom of Speech Acts? A Response to Langton, 24 Phil. & Pub. Aff. 64 (1995); Rea Langton, Speech Acts and Unspeakable Acts, 22 Phil. & Pub. Aff. 293 (1993); Clyde E. Willis, The Phenomenology of Pornography: A Commentary on Catherine MacKinnon's Only Words, 16 Law & Phil. 177 (1997) (providing commentary on MacKinnon).

n6 This ought not to surprise us in light of Thomas Scanlon's insight that "theoretical bases of the doctrine of freedom of expression are multiple and diverse." Thomas Scanlon, A Theory of Freedom of Expression, in The Philosophy of Law 156 (Ronald Dworkin ed., 1977). Thomas Grey argues that the problem of hate speech can be seen as a conflict between civil rights and civil liberties perspectives in U.S. society. See generally Thomas C. Grey, Civil Rights vs. Civil Liberties, 8 Soc. Phil. & Pol'y 81 (1991); Thomas C. Grey, Discriminatory Harassment and Free Speech, 14 Harv. J.L. & Pub. Pol'y 157 (1991).

n7 Owen Fiss, The Irony of Free Speech 13 (1996).

n8 Thomas Scanlon, Freedom of Expression and Categories of Expression, in Pornography and Censorship 139 (David Copp & Susan Wendell eds., 1983).

n9 The most notable instances of such theories include Jan Narveson, The Libertarian Idea (1989); Robert Nozick, Anarchy, State, and Utopia (1974); John Rawls, A Theory of Justice (3d ed. 1999); John Rawls, Political Liberalism (1993); John Rawls: Collected Papers

(Samuel Freeman ed., 1999); John Rawls, The Law of Peoples (1999); Michael Walzer, Spheres of Justice (1983); Robert Paul Wolff, In Defense of Anarchism (1970). For discussions of the respective political philosophies of Robert Nozick and John Rawls, see Equality and Liberty: Analyzing Rawls and Nozick (J. Angelo Corlett ed., 1991); John Rawls' Theory of Social Justice (H. Gene Blocker & Elizabeth Smith eds., 1981); Reading Rawls (Norman Daniels ed., Stanford University Press 1989) (1975); Reading Nozick (Jeffrey Paul ed., 1981); Thomas Pogge, Realizing Rawls (1989).

n10 MacKinnon, supra note 5, at 107. This claim is affirmed in Kent Greenawalt, Fighting Words (1995). See Cynthia A. Stark, The Words We Love to Hate, 16 Law & Phil. 107 (1997) (reviewing Greenawalt).

n11 See Scanlon, A Theory of Freedom of Expression, supra note 6, at 168.

n12 See Fiss, supra note 7, at 3.

n13 Id. at 15.

n14 See Words that Wound, supra note 5.

n15 See infra Part IV.

n16 See Words that Wound, supra note 5.

n17 See infra Part IV.

n18 John Stuart Mill, On Liberty 9 (Elizabeth Rapaport ed., Hackett Pub. Co. 1978) (1859).

n19 David Lyons, Liberty and Harm to Others, in Mill's On Liberty 132 (Gerald Dworkin ed., 1997) [hereinafter Lyons, Liberty and Harm]. For further discussion of Mill on liberty, rights and welfare, see David Lyons, Rights, Welfare, and Mill's Moral Theory (1994).

n20 Mill, supra note 18, ch. 1 (written in 1863).

n21 Lyons, Liberty and Harm, supra note 19, at 134.

n22 However, we concur that mere offenses, no matter how profound in nature and effect, are insufficient reasons for legal interference. For a discussion of offenses and how they differ from harms, see Joel Feinberg, Offense to Others (1985). For a discussion of profound offenses, see id. at 137-66.

n23 See Words that Wound, supra note 5.

n24 Altman, supra note 4, at 315-16.

n25 Id. at 304.

n26 Martin Golding noted: Acts which morally subordinate are those which treat a person as though his or her life is inherently less

valuable, and his or her interests inherently less important, than the lives and interests of some reference group. Such words as "kike," "faggot," "nigger," and "spic" are typically used for such linguistic acts of subordination; they are the conventional, verbal instruments of subordination. They carry roughly the same social meaning that spitting on someone carries in our culture. Martin Golding, Free Speech on Campus 111 (Steven M. Cahn ed., 2000).

n27 Altman, supra note 4, at 304.

n28 Id. at 317.

n29 See infra Part IV.

n30 See generally Joel Feinberg, Limits to the Free Expression of Opinion, in Philosophy of Law 262 (Joel Feinberg & Hyman Gross eds., 5th ed. 1995). One should, however, bear in mind that although the U.S. Supreme Court has created these categories of unprotected expression, "[w]hat the Court has not done is to give a clear sense of the unifying factors that justify the creation of these categories." Cass Sunstein, The Partial Constitution 243 (1993).

n31 It has been argued that hate speech which violates privacy or incites violence is unprotected by law. See Edward J. Eberle, Hate Speech, Offensive Speech, and Public Discourse in America, 29 Wake Forest L. Rev. 1135, 1205 (1994) (citing Cohen v. California, 403 U.S. 15, 20-21 (1971)).

n32 See Susan J. Brison, The Autonomy Defense of Free Speech, 108 Ethics 312 (1998) [hereinafter Brison, Autonomy Defense], whose analysis shall be discussed in Part III. See also Cass R. Sunstein, Preferences and Politics, 20 Phil. & Pub. Affairs 31 (1991).

n33 See infra Part III.

n34 See discussion infra pp. 1095-97.

n35 See discussion infra pp. 1098-1100.

n36 Anthony Lewis rejects an absolutist view of the First Amendment on freedom of expression for at least two reasons. First, insofar as such absolutism is grounded in constitutional originalism, it does not enjoy the support of the framers and ratifiers of the Constitution themselves. Secondly, First Amendment absolutism runs counter to basic legal prohibitions against extortion and perjury. Anthony Lewis, Make No Law 47-48 (1991).

n37 Richard Delgado, Words that Wound: A Tort Action for Racial Insults, Epithets, and Name Calling, in Words that Wound, supra note 5, at 89-90.

n38 Mari J. Matsuda, Public Response to Racist Speech: Considering the Victim's Story, in Words that Wound, supra note 5, at 17.

n39 Brison, Autonomy Defense, supra note 32, at 313. For an alternative autonomy-based view of why defamatory speech, including hate speech, ought to be regulated, see Filimon Peonidis, Freedom of Expression, Autonomy, and Defamation, 17 Law & Phil. 1 (1998).

n40 Brison, Autonomy Defense, supra note 32, at 314.

n41 Id.

n42 Note that, in certain contexts, a remark might qualify as group libel even if it is directed toward a specific individual. Suppose that a

n42 Note that, in certain contexts, a remark might qualify as group libel even if it is directed toward a specific individual. Suppose that a teacher vilifies a particular student on the basis of his race, and does so in front of the entire class. This might be considered group libel. It would be quite a stretch, however, to classify the remark of the angry motorist described above as group libel.

n43 See infra Parts III and IV.

n44 Although the term "speciesist" was made most popular by Peter Singer, Animal Liberation (2d ed. 1990), Singer credits Richard Ryder with the term, id. at 6 n.4. An example of philosophical speciesism is Rene Descartes. See Immanuel Kant, Lectures on Ethics 239-41 (Louis Infield trans., Benjamin Nelson ed., Harper & Row 1963) (1930); John Marshall, Descartes's Moral Theory (1998).

n45 Charles Lawrence III, If He Hollers Let Him Go: Regulating Racist Speech on Campus, in Words that Wound, supra note 5, at 55.

n46 Words that Wound, supra note 5, at 8.

n47 Id.

n48 Joel Feinberg, Harm to Self (1986).

n49 See supra text accompanying note 39.

n50 For an analysis of racist language, see J. Angelo Corlett, Analyzing Racism, 12 Pub. Aff. Q. 23, 32-33 (1998), J. Angelo Corlett, Race, Racism, and Reparations (2003).

n51 See supra pp. 1080-81.

n52 J. L. Austin, How To Do Things With Words 115-16 (J.O. Urmson ed., 1962) (emphasis in original).

n53 Peter F. Strawson, Intention and Convention in Speech Acts, in Readings in the Philosophy of Language 599, 606 (Jay F. Rosenberg & Charles Travis eds., 1971).

n54 Brison, Autonomy Defense, supra note 32, at 313. Hate speech often takes the form of a statement (e.g., "Gays are immoral!"), but the illocutionary force of a hate speech act might also be interrogative ("Why must gays engage in such immoral behavior?"), or even imperative ("Stop the immoral behavior!" uttered, say, at a gay pride parade).

n55 For some compelling examples of being mistaken about one's emotions, see Robert C. Roberts, Feeling One's Emotions and Knowing Oneself, 77 Phil. Stud. 319 (1995).

n56 These points call into question the claim that "[h]ate speech is self-consciously directed toward an issue of public concern; it is conspicuously and intentionally political in nature." Sunstein, Preferences and Politics, supra note 32, at 31. If the intention to speak with hateful content is not necessary for hate speech, then it is hard to see how the intention to speak with political content would be necessary for hate speech (even assuming that hate speech is political in nature). And if the intention is not necessary, then an awareness of the intention is not necessary; thus, by this logic, the utterance need not be self-consciously directed toward anything.

n57 " The intent requirement is needed to accommodate cases in which an epithet or slur is not used with any intent to treat the addressee as a moral subordinate." Altman, supra note 4, at 314.

n58 For instance: "Those fucking utilitarians!"

n58 For instance: "Those fucking utilitarians!"

n59 Kent Greenawalt, Speech, Crime, and the Uses of Language 143 (1989).

n60 For a discussion of critical race theory, which considers racism as "normal, not aberrant, in American society," see the Introduction to Critical Race Theory: The Cutting Edge xvi (Richard Delgado & Jean Stefancic eds., 2d ed. 2000).

n61 Words that Wound, supra note 5.

n62 Id.

n63 Id.

n64 See, e.g., Delgado, supra note 37; Matsuda, supra note 38; Susan Brison, Speech, Harm, and the Mind-Body Problem in First Amendment Jurisprudence, 4 Legal Theory 39 (1998).

n65 Joel Feinberg, Harm to Others: The Moral Limits of the Criminal Law 31-64 (1984) [hereinafter Feinberg, Moral Limits].

n66 Lyons, Liberty and Harm, supra note 19, at 129-30.

n67 Feinberg, Moral Limits, supra note 65, at 230.

n68 For example, recall Altman's point that since some hate speech subordinates in a morally illicit way, it violates the liberal principle of viewpoint neutrality. Altman, supra note 4.

n69 See Rawls, A Theory of Justice, supra note 9; Rawls, Political Liberalism, supra note 9; John Rawls: Collected Papers, supra note 9; Rawls, The Law of Peoples, supra note 9. For a communitarian critique of political liberalism's commitment to pluralism, see Will Kymlicka, Liberalism, Community, and Culture (1989); Will Kymlicka, Multicultural Citizenship: A Liberal Theory of Minority Rights (1995); The Rights of Minority Cultures (Will Kymlicka ed., 1995).

n70 See Rawls, Political Liberalism, supra note 9; Rawls, The Law of Peoples, supra note 9.

n71 Mill, supra note 18 and accompanying text. For a discussion of Mill's Harm Principle, see Feinberg, Moral Limits, supra note 65.

n72 Frederick Schauer, The Phenomenology of Speech and Harm, 103 Ethics 635, 652 (1993).

n73 This analysis of hateful utterances as speech acts argues that these utterances qualify as conduct as well as speech. The proposition is not simply that such utterances are perlocutionary acts (i.e., have an effect on the audience), but that the utterance has an effect by virtue of the stylistic aspects that give the utterance its force and content. For a detailed illustration, see Paul Berckmans, The Semantics of Symbolic Speech, 16 L. & Phil. 145 (1997), and the analysis of pornographic speech in Langton, supra note 5. There is also the proposition, noted earlier, that hate speech need not take the form of words; pictorial representations and even bodily movements can qualify as hateful speech in the model proposed herein. These two propositions support the view, held by a host of philosophers, that the hard distinction between words and deeds is to a large extent artificial and in many cases groundless. This collapsing of the distinction between language and conduct provides substantive arguments to those who oppose First Amendment absolutists.

n74 Feinberg, Moral Limits, supra note 65, at 31-64.

n74 Feinberg, Moral Limits, supra note 65, at 31-64.

n75 Lyons, Liberty and Harm, supra note 19.

n76 See sources cited infra note 81 and accompanying text.

n77 While those like Matsuda argue that it is the victim of hate speech who is in the best position to measure harm, still, although this might be true in cases of mentally competent victims, it would not follow that the victim is necessarily in the best position to judge whether or not the harm was imposed wrongfully. See Matsuda, supra note 38. Moreover, a victim of hate speech might only be an effective judge of the resultant emotional feeling of harm, rather than of actual harm from hate speech beyond transient emotional feelings. For these reasons, a reasonable person standard ought to be used to determine, however imperfectly, the extent to which hate speech harms on a case-by-case basis. This analysis leaves open the question as to precisely who qualifies as a reasonable person for such purposes.

n78 This observation suggests that the right to freedom of expression is not unrelated, at least in some cases, to the right to privacy. For philosophical discussions of the right to privacy, see J. Angelo Corlett, The Nature and Value of the Moral Right to Privacy, 16 Pub. Aff. Q. 329 (2002); Judith Wagner DeCew, In Pursuit of Privacy: Law, Ethics, and the Rise of Technology (1997); Julie C. Inness, Privacy, Intimacy and Isolation (1992); Philosophical Dimensions of Privacy: An Anthology (Ferdinand Schoeman ed., 1984).

n79 For a discussion of retributivist attitudes, see Peter F. Strawson, Freedom and Resentment, in Perspectives on Moral Responsibility 45 (John Martin Fischer & Mark Ravizza eds., 1993).

n80 For discussions of the concept of desert, see George Sher, Desert (1987); John Braithwaite & Philip Pettit, Not Just Deserts (1990); What Do We Deserve?: A Reader on Justice and Desert (Louis P. Pojman & Owen McLeod eds., 1999).

n81 For discussions of moral and/or legal responsibility, see J. Angelo Corlett, Responsibility and Punishment (2001); Joel Feinberg, Doing and Deserving: Essays in the Theory of Responsibility (1970); Moral Responsibility (John Martin Fischer ed., 1986); Perspectives on Moral Responsibility (John Martin Fischer & Mark Ravizza eds., 1993); Harry Frankfurt, The Importance of What We Care About (1988); H.L.A. Hart, Punishment and Responsibility (1968); H.L. A. Hart & A.M. Honore, Causation in the Law (1959).

n82 A legal system, however, might want to prohibit such expressions during certain phases of the defendant's trial so as not to unduly influence the jury, or even the judge.

n83 See Frederick Schauer, Free Speech: A Philosophical Enquiry 12, 68, 83, 106 (1982). For a review of Schauer's book, see David A. J. Richards, Toleration and Free Speech, 17 Phil. & Pub. Aff. 323 (1988).

n84 Schauer, supra note 83.

n85 Id.

n86 Id.

n87 See Frederick Schauer, Slippery Slopes, 99 Harv. L. Rev. 361 (1985).

n88 David Lewis, Mill and Milquetoast, 67 Australasian J. of Phil. 152 (1989).

n89 Historically and philosophically speaking, many philosophers and legal scholars believe that Mill, in On Liberty, supra note 14, articulated an account of freedom of speech and expression that has served as the foundation of the predominant U.S. attitude toward the

First Amendment and the right to freedom of expression. For example, in articulating some of the "arch- defenders of free speech," one author makes no mention at all of Karl Marx. See Jacobson, supra note 5, at 67-68. Moreover, in his discussion of freedom of expression, Frederick Schauer makes reference to Mill, but never to Marx. See Frederick Schauer, The First Amendment as Ideology, 33 Wm. & Mary L. Rev. 853 (1992). But in 1842, Karl Marx argued that: As soon as one facet of freedom is repudiated, freedom itself is repudiated, and it can lead only a mere semblance of life, since afterwards it is pure chance which object unfreedom takes over as the dominant power. Unfreedom is the rule and freedom the exception of chance and caprice. Karl Marx, On Freedom of the Press and Censorship 46 (Saul K. Padover trans. & ed., 4 The Karl Marx Library 1974). Of freedom of the press in particular, Marx writes "[t]he essence of a free press is the characterful, reasonable, ethical essence of freedom. The character of a censored press is the characterless ogre of unfreedom; it is a civilized monster, a perfumed abortion." Id. at 26. Not only, then, does Marx express his unambiguous support of freedom of expression in publication, he condemns any attempt of a government to suppress it or to limit it in any way. And Marx's words are not the rantings of an opinionist with merely emotive content. For as a philosopher, he wants to consider rationally the putative justifications for censorship: "we must above all examine whether censorship is in its essence a good means." Id. at 28. His conclusion is that censorship of the press is but a police measure that does not even achieve what it wants to achieve: "[t]he censorship is thus no law but a police measure, but it is itself a bad police measure, because it does not achieve what it wants and it does not want what it achieves." Id. at 31. This is because "censorship is a constant attack on the rights of private persons and even more so on ideas." Id. at 34. Marx derives this inference from the premise that freedom in general is a good thing, and a good thing to protect: "[i]f freedom in general is justified, it goes without saying that a facet of freedom is the more justified the greater the splendor and the development of essence that freedom has won in it."Id. at 39. Although Mill's defense of freedom of expression differs from Marx's in certain respects, it would be incorrect to suppose that it is Mill who first argued in favor of the right to freedom of expression.

n90 Lewis, supra note 88, at 171.

n91 For a discussion of toleration, see Michael Walzer, On Toleration (1997). Specific consequentialist arguments against the restriction of hate speech include the "pressure valve" argument (the idea that hate speech offers racists a safe way to blow off steam), the idea that exposure to hate speech allows others to see the absurdity of the ideas expressed, and the suspicion that the authorities will apply rules regarding hate speech against minorities. But see, Richard Delgado & David H. Yun, Pressure Valves and Bloodied Chickens: An Analysis of Paternalistic Objections to Hate Speech Regulation, 82 Cal. L. Rev. 871 (1994) (countering these "paternalistic" objections). There is also the non-consequentialist argument from autonomy, which is successfully refuted in Brison, Autonomy Defense, supra note 32. See also Peonidus' discussion of autonomy and defamation in general. Peonidus, supra note 39. Finally, there are a variety of practical concerns regarding the legal restriction of hate speech. One wonders how to handle problematic, borderline cases (e.g., scientific claims about racial inferiority, the case of the angry nationalist, and wordless hate speech), and there are also concerns about fraudulent claims and measuring/ appropriating damages. These worries have been aptly addressed by Matsuda, supra note 38, and Delgado, supra note 37.

n92 Ronald Dworkin, Freedom's Law 199 (1996). For an articulation of Dworkin's philosophy of law, see Ronald Dworkin, Taking Rights Seriously (1978); Ronald Dworkin, A Matter of Principle (1985); Ronald Dworkin, Law's Empire (1986); Ronald Dworkin, Life's Dominion (1993). For a discussion of Dworkin's theory of legal interpretation, see Ronald Dworkin and Contemporary Jurisprudence (Marshall Cohen ed., 1983); J. Angelo Corlett, Dworkin's Empire Strikes Back!, 21 Stat. L. Rev. 41 (2000). Dworkin's words about interpreting the First Amendment's right to freedom of expression are reminiscent of Marx's (anti- legal positivist) claim that "[t]he judge is also part of the law. If laws were self-applying, then courts would be superfluous. . . . [T]he judge has a duty to interpret the law in an individual case as he understands it after conscientious examination. . . ." Marx, supra note 89, at 32-33.

n93 See discussion, supra note 89.

n94 Discussions of original intent in interpreting the Constitution (i.e., constitutional originalism) include Gregory Bassham ,Original Intent and the Constitution (1992), Sunstein, supra note 30, at 232-56; Robert Bork, The Tempting of America (1990); Lewis, supra note 36; Charles W. Mills, Blackness Visible: Essays on Philosophy and Race (1998); Original Intent and Framers of the Constitution: A Disputed Question (Harry V. Jaffa ed., 1994); David Lyons, Constitutional Interpretation and Original Meaning, 4 Soc. Phil. & Pol'y 75 (1986); Original Intent and Legal Interpretation, 24 Austl. J. Legal Phil. 1 (1999).

n95 Diana Tietjens Meyers, Rights in Collision: A Non- Punitive Compensatory Remedy for Abusive Speech, 18 Law & Phil. 203 (1995). For a commentary on Meyers' view, see Thomas W. Peard, Diana Tietjens Meyers's Remedy for Abusive Speech: Objections, 18 Law & Phil. 1 (1999).