II. Free Expression: Institutional Variations

It is imperative to recognize from the outset that the abstract ideal of a system of free expression can be translated into institutional practice in a variety of ways. It is “institutional fetishism,” to borrow Roberto Unger's term, to think that there is just one natural, necessary, or correct way in which the abstract ideal can receive institutional embodiment.⁷

Unger makes the case against institutional fetishism in connection with the ideals of the market and democracy. Yet, a parallel case can be convincingly made with respect to the liberal ideal of a system of free expression. Indeed, it makes sense to think of the different possible embodiments of such an ideal as tied to alternative institutional forms of democracy. Different systems of free expression constitute, in part, different forms of democracy.

It is true that there is much more to a particular form of democracy than its specific rules regarding expressive liberty. These rules do not dictate whether a democracy has a parliamentary system or a presidential one, a system of proportional representation or a district-based, “first-past-the-post” one. Nor do the rules dictate whether a democracy operates with judicial review of legislative and executive action or without such judicial power.

However, expressive activities occur within a framework that is broader than simply the rules that are explicitly and exclusively concerned with expression. The broader framework helps to shape the kind of system of expression a society has. This framework does not merely include the law of libel, the legal rules governing the production and distribution of pornography, the law concerning racist or anti-Semitic speech, and more generally, the abstract doctrinal principles concerning the authority of government to restrict expression on account of its propositional or emotive content. The framework also includes the nature of the political party and electoral systems, the power of courts, the rules regarding private property, the nature of the mass media, and much else that shapes the ways in which persons and groups engage in communicative activities.

Accordingly, liberal systems of free expression vary in their rules regarding expression and in the broader social, legal, and political framework in which they operate. That having been said, the abstract ideal of a liberal system is not wholly malleable. There are certain paradigmatic features, and when some system departs in significant measure from these features, there must be reasons that are not only strong but also of a characteristically liberal sort to justify the departures. Otherwise, the system's claims to be a liberal one will be weak or implausible.

United States Supreme Court Justice William J. Brennan described one of the central features of a liberal system: “debate on public issues should be uninhibited, robust, and wide-open.”⁸

In fact, this robust debate consists of a number of interconnected elements: there is a broad array of political views, which are voiced in a way that is readily accessible by a typical member of the public; disagreement and discussion over the validity of political programs, proposals, and policies take place in the open, without fear of suppression by government or illegitimate coercion by private parties; criticism of existing or alleged power structures can proceed without fear of rights-violating retaliation; and association for expressive purposes can occur without similar fears.

These aspects of robust public debate require certain institutional arrangements, or are at least fostered by such arrangements, under conditions of modern social and political life. For example, in the absence of a system of relatively strong property rights, liberals find it impossible to see how adequate public debate can exist. The assertion of an inextricable connection between property and political freedom was one of the main contentions of the major figures of modern liberalism.⁹

The twentieth century saw considerable debate within the liberal camp over the proper scope, strength, and grounds of rights to property. One of the key points of contention was whether liberal socialism was a viable institutional arrangement.¹⁰

Notwithstanding these disputes, all current liberals would agree with Cass Sunstein's claim that liberal democracy “requires constitutional protections of many individual rights, including … the right to private property, for people cannot be independent citizens if their holdings are subject to unlimited government readjustment.”¹¹ Moreover, it is clear that liberal principles require a market economy and flatly rule out any form of state socialism as unjust in principle and incompatible with free speech in practice.

Although robust debate on public issues is central to a liberal system of expression, the system must also protect forms of communicative activity that are neither political nor public. Much communication does not concern questions of law or policy, and even some expression that does implicate such questions occurs in settings reasonably regarded as private. Thus, much religious, scientific, and artistic expression is not intended or understood as contributing to debate on public issues, and communication among friends, family members, and fellow worshippers is typically outside the realm of public discussion. A liberal system will carve out a broad zone of communicative freedom for such private forms of expression.

Many thinkers take the principle of ‘content neutrality’ to capture one of the key guiding thoughts behind a liberal system of expression. In rough terms, the principle prohibits restrictions on speech based on its propositional or emotive content. However, the principle has been widely misunderstood. As legal scholar James Weinstein points out, the principle cannot be reasonably formulated as “a blanket rule against content discrimination.”¹²

There are too many kinds of noncontroversial content-based restrictions on expression, from libel laws and securities regulations to copyright and antitrust rules, for a general prohibition or even a strong presumption against content restriction to be plausible. Weinstein's more reasonable alternative is a principle against the content-based restriction of those forms of communication that help to constitute “the public discourse by which we govern ourselves and through which we build our culture.”¹³

However, it is important to realize that even Weinstein's formulation of the principle of content neutrality cannot by itself capture all, or even most, of the key elements of a liberal system of expression. There is also a broad realm of private discourse that must be free from regulation, regardless of its content. Clearly, this realm does not include threats of violence or of other rights violations, but it must include the private expression of racist, sexist, or heterosexist attitudes. Moreover, even the principle of content neutrality must operate within the context of a regime of private property, multiparty politics, and the other institutions that make possible the robust public debate central to Justice Brennan's understanding of free speech. In the absence of such institutions, a liberal system of expression could not exist.

III. Skeptical Challenges

Skeptics of the liberal view, including some radical critics, will argue that ‘robustness’, ‘wide ranging’, ‘broad zone’, and similar liberal terms are largely metaphorical and emotive in content. These skeptics will claim that the terms are little more than emotionally positive labels that liberals—or anyone else for that matter—affix to the systems of which they approve. If the skeptics are right, then I have not described a distinctively liberal system of expression but have simply constructed a set of empty conceptual boxes into which virtually any ideological position can pack its favored institutions.

Accordingly, Stanley Fish contends that “abstract concepts like free speech do not have a ‘natural’ content but are filled with whatever content and direction one can manage to put into them. ‘Free speech’ is just the name we give to verbal behavior that serves substantive agendas we wish to advance…. Free speech, in short, is not an independent value but a political prize.”¹⁴

Competing ideological factions struggle over the power to decide what verbal behavior “merits” this emotionally charged term, and “there's no such thing as free speech” because any conception of free speech will presuppose some particular set of political values.

Fish contradicts his claim that the concept of free speech has no natural content in the very next sentence of the passage from which I have cited: he asserts that the concept refers to verbal behavior. The reference to such behavior surely gives the concept some natural semantic content. In fact, a more sensible reading of the concept's content would explicate it much more broadly, in terms of communicative activity.¹⁵

However, if we set this point aside and treat the claim about the absence of natural content as hyperbole, it is possible to make sense of Fish's understanding of the idea of free speech.

Fish is disaggregating the term ‘free speech’ into distinct descriptive and emotive elements. The descriptive element simply refers to verbal behavior or communicative activity more generally. The emotive element is the attitude of approval that the speaker, based on her political values, is expressing toward the protection of a certain sort of communicative activity from punishment by society.¹⁶

On Fish's account, any set of rules regulating expression can count as a system of free speech: the speaker's own political values and preferences determine what rules count as such, and the political preferences of speakers run the gamut. Certainly, the range extends well beyond the particular preferences of liberals.

The kind of emotivist analysis that I am attributing to Fish has its precedents in twentieth century metaethics and has been subject to a wide range of criticisms.¹⁷

The problem on which I intend to focus in criticizing Fish will show that there is much more descriptive content to the concept of free speech than Fish acknowledges. This criticism is not, by itself, inconsistent with an emotivist analysis that permits a fixed descriptive content. However, the criticism is inconsistent with Fish's analysis, because his account leaves the concept of free speech devoid of almost all fixed descriptive meaning.

My criticism stems from the idea that there is a kind of “folk political philosophy” whose propositions help to fix the meaning of the concept of free speech.¹⁸

These propositions are platitudinous: “Totalitarian systems are incompatible with free speech,” “Government suppression of dissenting political views typically violates free speech,” “Whether a given instance of suppressing expression is compatible with free speech depends in part on the circumstances,” “Banning works of art because the majority finds them offensive is a violation of free speech,” “Private violence against nonviolent forms of religious worship violates free speech,” “Multiparty political systems are essential to a system of free speech,” “Tolerance of conflicting opinions is essential to a system of free speech,” “Multiple, independent sources of news and information are essential to a system of free speech,” and so on.

Philosopher Michael Smith has defended an account of conceptual analysis according to which “the aim of an analysis is to give us knowledge of all and only the platitudes surrounding our use of [a] concept.”¹⁹

I regard this account as presenting an oversimplified view of the connection between conceptual meanings and platitudes. It ignores the fact that concepts have an internal structure and a place within a wider network of concepts. Thus, some platitudes are more central to the meaning of a given concept than others, and the meaning of a concept is, in part, a function of its various links to other concepts that share its semantic field. Moreover, the platitudes surrounding our use of a given concept can to some extent change.²⁰ Nonetheless, it would be unclear how one could pick out any subject to talk about—free speech, water, tigers, or whatever—unless the platitudes surrounding the subject played a significant role in defining the meaning of the corresponding concept.

Fish would be quick to point out that the platitudes of the folk political philosophy of free speech are suspiciously liberal, and liberal they are indeed. We should not be surprised: the concept of free speech was, after all, constructed by modern liberal thinkers. Fish can contest the extent to which, or the way in which, a liberal system values free speech; he can argue for some alternative to a liberal system of free expression. However, if he rejects all or most of our liberal commonplaces, he is not talking about free speech from a different political viewpoint but is simply changing the subject.

Fish is unquestionably right in thinking that speakers use ‘free speech’ in a way that expresses an attitude of approval toward certain policies and conditions, such as constitutional protection for competing political parties or economic arrangements that will provide multiple and independent sources of news and information. However, the approval is linked in a specific way to the descriptive content of the concept: speakers think that there are justificatory reasons for favoring those policies. If someone rejects those policies, then he is not operating with a nonliberal conception of free speech but is, rather, opposing the policies implicit in the very idea of free speech.

I do not claim that Fish is himself such an opponent. In fact, I doubt that he is because there is no reason to think that he would reject the liberal commonplaces, despite his initial pejorative reference to some of them as “First Amendment pieties.”²¹

Rather, he is asserting one of the liberal platitudes as an unqualified generalization and elevating it to a privileged position above all of the others, namely, the platitude that whether a given instance of speech suppression violates free speech depends on the circumstances. Fish's formulation is this: “The question of whether or not to regulate will always be a local one, and we cannot rely either on abstractions that are empty of content or filled with the content of some partisan agenda to generate a ‘principled’ answer.”²² The sound approach is “to consider in every case what is at stake and what are the risks and gains of alternative courses of action.”²³

Fish's approach may seem to throw out all of the other liberal platitudes and to claim that particular circumstances are all that count in determining whether an instance of speech is an exercise of expressive freedom. However, in the end he winds up accepting the other liberal platitudes, even though he does not think that they can generate acontextual answers to questions about particular issues of speech regulation: “First Amendment formulas … still serve a function that is not at all negligible: they slow down outcomes in an area in which fear of overhasty outcomes is justified by a long record of abuses.”²⁴

Fish's initially dismissive talk of “First Amendment pieties” has thus become a more respectful reference to “First Amendment formulas.”

In any case, it seems dubious for Fish to elevate above all other free speech platitudes the one that concerns the circumstances of speech. The other platitudes are at least as important because no account of the circumstances of speech can ignore the considerations to which the other platitudes make reference, such as the existence of a multiparty system or a totalitarian one, the toleration of religious difference or private violence against dissenting creeds, the control by government of the media or multiple and independent sources of information, and so on. Nor is there any reason to think that the relevant particular circumstances are so variable that all presumptive generalizations about free speech are to be rejected in favor of the unqualified generalization that issues of free speech are “always local.”

I do think that Fish is right to this extent: the free speech platitudes do not dictate in any simple way the answers to many of the free speech issues that confront liberal societies. I include among these issues the ones on which contemporary radicals focus: pornography regulation, sexual harassment rules, and restrictions on hate speech. Moreover, I agree with Fish that considerations of historical and social context ought to carry weight in dealing with these issues. None of these points is inconsistent with recognizing that the concept of free speech has a robust descriptive content or affirming that a liberal system of expression ought to be endorsed. Fish's efforts to show that the concept is inherently devoid of descriptive content fail. However, even if we can fix at least some of the content of the concept by reference to liberal platitudes, the question still remains whether a liberal system of expression ought to be endorsed.

IV. Marcuse and the Rejection of Liberalism

In the 1960s one of the most influential voices on the left was Herbert Marcuse (1898–1979), and he explicitly rejected a liberal system of expression in favor of what he regarded as an equality-promoting alternative. Marcuse did not defend his alternative as a permanent fixture. Rather, he believed that a liberal system could not be justified at this historical juncture because it tolerated forms of expression whose content blocked progress toward peace, equality, and effective freedom for all. The progress of history required the systematic suppression of certain ideas now so that a liberal system of free expression could be instituted at a later stage of social development.

For the advancement of freedom, peace, and equality, Marcuse advocated what he called a system of “liberating tolerance.” In his writings, he describes what he has in mind in only the sketchiest of terms, but these terms—together with the justification he offers for liberating tolerance—make it clear that he is advocating a radical alternative to a liberal system. Liberating tolerance would employ “apparently undemocratic means,” including “the withdrawal of toleration of speech and assembly from groups and movements which promote aggressive policies, armament, chauvinism, discrimination on grounds of race and religion, or which oppose the extension of public services, social security, medical care, etc.”²⁵

In order to achieve a free and democratic society, “tolerance cannot be indiscriminate and equal with respect to the contents of expression.”²⁶

The reason resides in the very framework of liberal capitalist society: “The antagonistic structure of society rigs the rules of the game. Those who stand against the system are a priori at a disadvantage, which is not removed by toleration of their ideas, speeches, and newspapers.”²⁷ To achieve true equality requires “liberating tolerance [which] would mean intolerance against movements from the Right, and toleration of movements from the Left.”²⁸ The result “would amount to an upheaval” and involve the “cancellation of the liberal creed of free and equal discussion.”²⁹

Marcuse is clear that liberating tolerance would involve sweeping restrictions on expressive liberty, and his account of the nature of post–World War II liberal society is meant to support such restrictions:

In the foregoing passage, Marcuse is turning what had been a standard liberal rule of free expression into the basis for a repressive system. The rule was formulated by U.S. Supreme Court Justice Oliver Wendell Holmes, namely, the “clear and present danger” test. For much of the twentieth century, jurists saw this test as highly protective of expression. Holmes had written, in Schenck v. United States (1919), that government could restrict speech in a way that is consistent with the First Amendment only as long as the speech posed a “clear and present danger” of causing evils that government had the right to prevent.³¹

In his opinion for the Court, Holmes drew a distinction between wartime and peacetime, suggesting that expression receives less protection during the former. The context of war can make speech with a certain content a clear and present danger, when that same speech during peacetime would pose no such threat. The Court found that Schenck's antiwar pamphlets posed a clear and present danger, having been circulated during wartime and intended to obstruct the draft. Holmes employed his famous example of falsely shouting “Fire!” in a crowded theatre to illustrate the point that circumstances can make a difference in deciding whether an instance of speech ought to be protected as “free speech” or not.

Marcuse makes two moves to turn Holmes's clear and present danger test into a justification for “an extreme suspension of the right of free speech and free assembly.” First, he goes beyond the Court's wartime exception to free speech by arguing that society is in an even more dire condition: a “state of emergency” in which it confronts “extreme danger.” Second, he rejects the notion that this state of emergency is temporary: it is the “normal state of affairs” and has become entrenched in social life. Thus, he alludes again to Holmes and writes that the clear and present danger test “seems no longer adequate to a stage where the whole society is in the situation of the theater audience when somebody cries: ‘fire.’ ”³²

The emergency condition of society consists, in part, of the oppressive structural inequalities that characterize liberal capitalism. These inequalities are dehumanizing to those who suffer under their weight, but they also implant in the individual a false consciousness about his condition: his dehumanization appears perfectly normal to him. Marcuse then suggests the appropriate remedy:

Accordingly, Marcuse defends aggressive censorship in the service of dismantling the entrenched capitalist structures of inequality. However, he does not wish to abandon the idea that there is great value in a system of free speech, provided existing inequalities have been dismantled. Indeed, Marcuse explicitly endorses Mill's argument that freedom of speech is valuable because it fosters the discovery of truth, but then Marcuse claims that liberal capitalist societies are like the societies of “barbarians” who Mill said were not yet fit to rule themselves because they were not yet capable of rational deliberation and learning from their mistakes. “Radicalism now; liberalism later,” is the essence of Marcuse's view on free speech.

Yet, the critical question that Marcuse fails adequately to address is, Who is to design and operate his system of censorship in the service of equality? Mill was certain of who should help the barbarians become civilized: Western powers such as Britain, which had achieved sufficient social maturity. Marcuse certainly did not think that the Soviet Union or Communist China was fit to play an analogous role in relation to liberal capitalist regimes. At times, he raises the possibility of a Leninist strategy of an intellectual revolutionary vanguard serving as the new ruling elite, and it seems that he was not prepared to rule out this strategy.³⁴

Indeed, it is difficult to see how the Leninist strategy could be avoided given that “the people … are indoctrinated…. To enable them to become autonomous, to find by themselves what is true and what is false for man in the existing society, they would have to be freed from the prevailing indoctrination (which is no longer recognized as indoctrination).”³⁵

Marcuse contends that “the small and powerless minorities which struggle against the false consciousness and its beneficiaries must be helped” by the establishment of liberating tolerance.³⁶ However, the vast majority in society is surely not going to help, if Marcuse's social theory is accurate, thereby leaving an enlightened Leninist vanguard as the only real alternative. Perhaps the cunning of history can come up with some other trick to break the stranglehold of liberal false consciousness and the “totalitarian democracy” that the ideology supports, but Marcuse is at a loss to describe what that trick might be.³⁷

V. Contemporary Radicals

In current debates over free speech issues, radical feminists and critical race theorists often echo Marcuse's approach, but with some important variations. Contemporary radicals tend to focus more on inequalities of race and sex than did Marcuse. In addition, they are more explicit than he was in asserting the power of communication and representation to create and sustain the oppressive hierarchies of power and privilege that they claim characterize liberal society. However, the contemporary radicals are in agreement with Marcuse in depicting society as a system of oppression and domination, and in appealing to such a depiction in order to criticize liberal institutions.

Charles Lawrence expresses a view characteristic of critical race theorists in claiming that we exist “in a society where racism is ubiquitous.”³⁸

This racism typically goes unnoticed by whites, Lawrence contends, because it is “unconscious.”³⁹ Moreover, speech plays a crucial role in creating the oppressive structures of our racist society: “all racist speech constructs the social reality that constrains the liberty of nonwhites because of their race. By limiting the life opportunities of others, this act of constructing meaning also makes racist speech conduct.”⁴⁰ Lawrence proceeds to defend campus speech codes prohibiting racist forms of expression.

Another prominent critical race theorist, Mari Matsuda, affirms “the structural reality of racism,” which consists of “the ideology of racial supremacy and the mechanisms for keeping selected victim groups in subordinated positions.”⁴¹

She claims that racist speech is among these mechanisms of subordination and argues that “a range of legal interventions, including the use of tort law and criminal law principles, is appropriate to combat racist hate propaganda.”⁴² Speech should be subject to criminalization when it conveys a message of racial inferiority against a historically oppressed group and the message is “hateful, persecutory, and degrading.”⁴³

For example, Matsuda would favor criminalizing anti-Semitic writings, including Holocaust denial literature. She would limit content-based restrictions on expression to racist speech because such speech “is best treated as a sui generis category, presenting an idea so historically untenable, so dangerous and so tied to the perpetuation of violence and degradation … that it is properly treated as outside the realm of protected discourse.”⁴⁴

Among other proposals made by critical race theorists are laws criminalizing incitement to racial hatred.”⁴⁵

In criticizing constitutional doctrine in the United States as formalistic and ahistorical, the theorists note that the United Kingdom and Canada, among other liberal democracies, already have such laws. These laws and the other proposals of the critical race theorists are seen as part of a defensible response to protect nonwhites from “racist assailants who employ words and symbols as part of an integrated arsenal of weapons of oppression and subordination.”⁴⁶

Radical feminist critics make parallel arguments when it comes to claims of sexual subordination and the alleged implications for the regulation of pornography. Thus, Catherine Itzin claims that “women are oppressed in every aspect of their public and private lives.”⁴⁷

She goes on to contend that “the oppression of women is both institutionalized and internalized.”⁴⁸ As with Marcuse's dehumanized and deluded masses, “Women are conditioned to conform to the stereotyped images of femininity and womanhood in such a way that they are often unaware that they are misrepresented and mistreated. They often, apparently willingly, agree to participate in misrepresentation and mistreatment.”⁴⁹

Itzin regards the depiction of women in pornography as playing a central role in the perpetuation of their oppression. On this basis, she defends “civil sex discrimination legislation against pornography [that] would enable women to take action on grounds of harm done to them by pornography.”⁵⁰

Itzin's position largely reflects the views of Catharine MacKinnon, who has been one of the principal architects of feminist antipornography legislation in the United States and whose legal arguments have played a crucial role in the creation of sexual harassment doctrine.⁵¹

MacKinnon describes the condition of women's sexual oppression: “Rape, battery, sexual harassment, forced prostitution…. a woman is socially defined as a person who, whether or not she is or has been, can be treated in these ways by men at any time, and little, if anything, will be done about it.”⁵² The result is that women are treated as “less than human, on the basis of sex,” and Americans live under a system of “male supremacy” whose lynchpin is the sexual abuse of women by men.⁵³

MacKinnon goes on to argue that this sexual oppression goes largely unrecognized: “[T]he force behind sexism, the subordination in gender inequality, is made invisible; dissent from it becomes inaudible as well as rare.”⁵⁴

And so “there is a belief that in this society women and men are basically equals,” and this belief is “an article of faith” even among “most women.”⁵⁵

In MacKinnon's view, any liberal remedy to women's subordination is misguided and futile from the start, especially if it depends on fighting sexist speech with the traditional liberal remedy of “more speech,” famously articulated by Justice Louis Brandeis in 1927.⁵⁶

As she sees it, “To liberals, speech must never be sacrificed to other social goals. But liberalism has never understood that the free speech of men silences the free speech of women.”⁵⁷ Liberal concepts “support power as currently organized.”⁵⁸

MacKinnon is especially critical of the liberal principle of content neutrality. She contends that the principle is invoked selectively to strike down antipornography ordinances, while it is ignored in the cases of other laws that are clearly content based, including traditional obscenity statutes. The result is that the “peculiar genius of liberalism has been to seem to accommodate fundamental challenges without fundamentally changing.”⁵⁹

The absence of fundamental changes stems from the failure of liberalism “to grasp systematically the group-based determinants of inequality, and … its relative blindness to organized power in diverse social forms, so that it often effectively sides with dominance.”⁶⁰ Nonetheless, MacKinnon writes that liberalism “might work and its terms might mean what they are supposed to mean [under] conditions of sex equality.”⁶¹

VI. Marcuse as the Model?

In their book, The Shadow University, historian Alan Charles Kors and attorney Charles Silverglate note the continuities in thinking between Marcuse and contemporary radical critics of liberalism. For example, they write that “MacKinnon's rhetoric is unmistakably Marcusean, especially the notion that liberty is a zero-sum game, and that giving it simultaneously to ‘dominant’ and ‘subordinated’ groups inevitably reinforces the power of the dominant group.”⁶²

Kors and Silverglate are especially concerned with the way in which campus speech codes and related manifestations of political correctness have squelched speech on college campuses across the United States: “Marcuse's prescriptions are the model for the assaults on free speech in today's academic world.”⁶³

There are several key similarities between Marcuse and current radical thinkers. In particular, they share the idea that the liberal principle of content neutrality should be rejected under existing conditions because (1) there are serious systemic inequalities which ought to be eliminated; (2) various forms of speech play a critical role in sustaining these inequalities, in large measure by convincing most people that the inequalities are fully justified; (3) restrictions on speech are justifiable on the ground that they can play an important role in eliminating the inequalities; and (4) a liberal system of free expression can work as advertised by liberals only after the inequalities have been eliminated and safely relegated to the past.

Notwithstanding these important similarities, the differences between Marcuse and the current radicals are just as striking. In particular, none of the current radicals is prepared to endorse the “extreme suspension” of the right of free speech and the “cancellation of the liberal creed of free and equal discussion” involved in the implementation of Marcuse's scheme of liberating tolerance. Most of the radical proposals are modest and can in fact be found in the law of many liberal democracies. Even the most aggressive of the proposals cannot be fairly equated with the “upheaval” for which Marcuse called.

Consider MacKinnon's antipornography efforts. She is critical of legal doctrine in the United States for not taking women's equality seriously, but she does not endorse the generalized system of speech control that Marcuse had in mind. She writes, “When equality is recognized as a constitutional value and mandate, the idea that some people are inferior to others on the basis of group membership is authoritatively rejected as the basis for public policy. This does not mean that ideas to the contrary cannot be debated or expressed.”⁶⁴

While Marcuse would be prepared to endorse general restrictions on the expression of anti-egalitarian ideas, MacKinnon's proposal is aimed at sexually explicit depictions of women's sexual subordination.

Moreover, MacKinnon regards Canadian law as taking women's equality seriously. Under the rulings of that nation's Supreme Court, antipornography laws have been upheld on grounds of sexual equality.⁶⁵

One can reasonably argue that Canadian constitutional doctrine rests on an inadequate understanding of liberal principles and that Canada's system of expression is not the most reasonable liberal system. However, it would be implausible to contend that Canada does not have a liberal system at all. Canada's laws and institutions form an institutional variant of a liberal system of expression. In her unrelenting criticism of liberalism, MacKinnon fails to recognize this important fact of institutional variation.

It is true that MacKinnon aims to put out of business the entire pornography industry, including the production and distribution of “soft core” pornography.⁶⁶

However, even if we agree that her favored ordinances violate any reasonable understanding of free speech principles, the slice of mass culture that the laws would affect would be quite narrow. In contrast, the “upheaval” in the treatment of speech for which Marcuse called would not leave the vast bulk of mass culture untouched. On his analysis, mass culture is what effectively indoctrinates people with the stultifying ideology of liberalism and what liberating tolerance is meant to combat. Although MacKinnon takes herself to be advocating principles deeply antagonistic to a liberal system of speech, her prescriptions are mild in comparison with those of Marcuse and, by her own admission, constitute part of the law in some existing liberal democracies.⁶⁷

In contrast to MacKinnon's explicit hostility toward liberalism, several of the leading critical race theorists take pains to argue that their proposals are compatible with a liberal system of expression. Matsuda seeks “to recognize and accommodate the civil libertarian position” by recommending restrictions on hate speech that are “consistent with [F]irst [A]mendment values.”⁶⁸

Lawrence agrees that “there are very strong reasons for protecting even racist speech” and believes that “good lawyers” can fashion speech codes capable of “limiting the harm of hate speech without opening the floodgates of censorship.”⁶⁹ In a similar vein, Richard Delgado and Jean Stefancic write that “the experiences of Canada, Denmark, France, Germany, and the Netherlands imply that limited regulation of hate speech does not invariably, or even frequently, weaken the respect accorded free speech.”⁷⁰

My point is not that the speech codes proposed by critical race theorists are the most reasonable responses to racist speech in universities or in society at large. Indeed, I regard Matsuda's suggestion that anti-Semitic literature should be criminalized as unreasonable, at least in liberal democracies outside of Germany. However, the key point to keep in mind is that Matsuda explicitly accepts the relevance of liberal principles of speech to her argument by seeking “to recognize and accommodate the civil libertarian position.”

Such an argument contrasts with that of Marcuse, who explicitly rejected any accommodation with civil liberties. He believed that civil liberties would have to wait until the day when some form of radical equality was established and that accommodating such liberties now would block progress toward that day. Similar contrasts also can be drawn between Marcuse's view and that of the other critical race theorists.

On the feminist side, Itzin's argument also contrasts with that of Marcuse. She does not contend that the protection of civil liberties must wait until sexual equality is established. Rather, she holds that “[f]rom a civil liberties perspective, the greatest freedom would be achieved [by laws that] … would make incitement to sexual hatred and violence illegal and … would enable women who could prove that they were injured by pornography to sue and to seek compensation.” In her view, “It is also possible to limit freedom of expression in the case of pornography on grounds of harm without any fundamental threat to freedom of speech.”⁷¹

Marcuse's position is that as long as oppressive inequalities exist—and both he and Itzin posit such inequalities—violations of freedom of speech are justifiable as necessary steps in the elimination of inequality. Itzin's argument only makes sense on the assumption that, just as liberals claim, free speech principles are important and cannot be justifiably ignored on the road to sexual equality.

Accordingly, contemporary radicals are prepared to argue within a framework that incorporates liberal principles of free speech and to operate politically within a liberal system of expression. The burden of their argument is that existing liberal systems of expression, especially in the United States, are insufficiently responsive to systemic and unjust inequalities based on race and sex. However, unlike Marcuse, they do not favor jettisoning the existing systems in favor of some antiliberal alternative. Marcuse's thinking may have been a model in some respects for current radicals, but it is a model whose radical elements they have refrained from endorsing. The question is, Why?

VII. Wrong, Remedy, and Paradox

There is an apparent discrepancy in contemporary radical thought. On the one side is a diagnosis of the racial and sexual wrongs of existing society and of the role that expression plays in perpetuating these wrongs. These wrongs are severe, systemic, and entrenched, amounting to oppression based on race and sex. Expression is crucial in perpetuating the wrongs: the public media, among other modes of communication, create a cultural environment that makes it difficult for people to perceive the wrongs, even as such communication shapes the attitudes and beliefs that entrench the wrongs. On the other side is the remedy proposed by the radicals: campus speech codes, legal restrictions on pornography, laws against racial and sexual incitement. The remedy seems rather inadequate to the wrong.

Thus, the radical position involves a triad that, if not strictly an inconsistent one, does raise the question of whether the position is normatively coherent. The three key theses are as follows:

1. There is an oppressive racial and social order.
2. Speech plays a substantial role in maintaining the oppression, mainly by portraying oppression as normal and natural.
3. Restrictions on the speech that helps to maintain the oppressive order should be quite limited in scope.

The apparent normative incoherence derives from the discrepancy between the magnitude of the wrong presented by (1) and (2), on the one side, and the very limited nature of the remedy proposed by (3), on the other. There are several ways of trying to make sense of this discrepancy. One possibility is to argue that any remedy involving substantially greater incursions on speech would simply not create any marginal gains in the direction of greater racial and sexual equality. This pragmatic approach would not place any limits, in advance, on permissible speech restrictions, but simply endorses whatever restrictions would have the largest egalitarian impact. The judgment would be that only very limited restrictions have beneficial effects at the margin.

It is reasonable to regard Fish as taking this approach. On his analysis, there is no value in the protection of expressive liberty aside from the consequentialist benefits as measured in terms of values other than expressive liberty. However, his analysis purchases the coherence of the radical position at the price of normative implausibility. It entails that a Marcusean-style system of speech control would be justified if only it yields marginal gains in the direction of racial and sexual equality. Even if we were to agree that significant departures from content neutrality could be justified in a limited domain of expression when the departures produce substantial gains toward sexual or racial equality, we would still be a far cry from justifying the rejection of a liberal system of expression. Indeed, MacKinnon makes precisely this kind of argument, and no contemporary radical has been prepared to go beyond it and to contend that a system of liberating tolerance would be justified if only it produces marginal egalitarian gains. The contention is not normatively plausible.

A second way of trying to render the radical position coherent is to argue that expressive liberty is inherently valuable and/or instrumentally valuable (apart from any equality-promoting consequences). Marcusean-style incursions on speech would not be justifiable on this view because they would violate the value of that liberty. On one line of argument, a liberal system of expression does a far better job of treating all persons with dignity than any alternative, including the Marcusean system. On another line of argument, a liberal system does a far better job of arriving at social, ethical, and scientific truths and of enabling persons to arrive at reasoned views and to better appreciate the significance of their own views.

Of course, the arguments that I have just sketched are quintessentially liberal ones.⁷²

Radicals may be hesitant to embrace them, but it is difficult to see what kinds of arguments other than liberal ones can be used to defend, on grounds other than purely pragmatic ones, a liberal system of speech in a society supposed to be riven by racial and sexual oppression. Indeed, the more the alleged oppression is said to be perpetuated by speech, the more weighty these liberal arguments must be in order to avoid the conclusion that a liberal system of expression ought to be jettisoned in favor of something akin to Marcuse's liberating tolerance. Aside from strictly practical considerations, there would be no good reason to withhold support from a Marcusean remedy unless one supposed that liberal considerations in favor of a system of free expression were quite weighty. If there is a great deal of public and private communication that serves to maintain oppression, then it would seem that strong reasons would be needed to defend the toleration of such communication. Liberal reasons are the only ones that could adequately play such a role.

In addition, support for a liberal system of expression in the face of an oppressive racial and sexual order sustained in substantial measure by speech would make little sense unless one also supposed that (a) the liberating power of speech could counteract its power to oppress and (b) false consciousness concerning the existence of oppression did not have a stranglehold over society. Without these suppositions, a system of free speech would be an unambiguous and unrelenting force for the maintenance of oppression. It is difficult to see why such a system should be tolerated. With the suppositions, however, we are back to Justice Brandeis's formula that the proper remedy for bad speech is good speech.

One might respond that, although the radicals' remedies are relatively modest and do not jettison the overall liberal system, the slice of speech that they do target for legal restriction is especially important in maintaining racial and sexual oppression. For example, MacKinnon maintains that pornography is especially important in maintaining the subordination of women. Presumably, the elimination of pornography would be a major blow to the system of male supremacy, and so there is no need to buy into Justice Brandeis's liberal formula.

However, the MacKinnonite response is implausible for several reasons. First, even if her ordinance were enacted and enforced, it is reasonable to expect that a considerable amount of “underground” pornography would still circulate. To paraphrase what she has written about rape, pornography would not be banned so much as regulated.⁷³

More importantly, the subordination of women, assuming it exists, is clearly overdetermined. Women were subordinated well before there was a multibillion dollar pornography industry, and in light of how women are depicted in much of the mainstream media, there would be no reason for a radical feminist to think that a dramatic difference would be made in women's status by legal restrictions on pornography. Certainly the experience in Canada, whose antipornography laws are supported by MacKinnon, does not support the idea that a substantial improvement in women's status flows from legally restricting graphic depictions of women's sexual subordination.⁷⁴

For critical race theorists, it is even more implausible to reject the potentially liberating power of speech than it would be for MacKinnon. She has a hypothesis, albeit an unsubstantiated and implausible one, concerning the essential role that pornography plays in maintaining sexual oppression. Critical race theorists favor campus speech codes and laws against the incitement of racial hatred, but none of the theorists hypothesizes that the speech that is targeted by such rules has a central role in perpetuating racial oppression. To the contrary, the theorists think that images and depictions throughout mass culture are largely responsible for the ubiquity of racism. Yet, critical race theorists reject a general ban on racist speech and, as we have seen, generally attempt to accommodate free speech concerns. This position is a coherent one if considerable weight is attached to at least some of the liberal arguments supporting free speech and to the idea that the liberating power of speech can counteract its oppressive force.

This, then, is the radical paradox: In order for the contemporary radical position on racist and sexist speech to be coherent, it must rely heavily on quintessential liberal arguments about free speech.⁷⁵

Marcuse avoided this paradox because his radicalism posited that the entire liberal system of expression was an insuperable obstacle in the way of dismantling an oppressive system. Modifying or qualifying one or another aspect of the system would be wholly insufficient. The whole system had to go if the oppression was to be undone. Accordingly, Marcuse's position on free speech was coherent with his social theory, even if the social theory was itself deeply flawed.

Current radicals want the radicalism of Marcuse's social theory without the radicalism of his political prescriptions. However, there is no plausible way to make this newer, truncated form of radicalism coherent, without conceding that liberals are right in according great weight to the value of speech.