Convinced that freedom of religion and belief should...contribute to the attainment of the goals of world peace, social justice and friendship among peoples and to the elimination of ideologies or practices of...racial discrimination...,

Resolved to adopt all necessary measures for the speedy elimination of such intolerance and all its forms and manifestations and to prevent and combat discrimination on the grounds of religion or belief.... (Resolution Adopted by the General Assembly on November 25, 1981 [on the Report of the Third Committee (A/36/684] 36/54).)

First, tolerance and nondiscrimination as they bear on religion are related in a similar way to race, language, gender, etc.

Second, tolerance and nondiscrimination, while interconnected, are nevertheless distinct principles that stand in some tension with each other. Third, "the goals of world peace, social justice and friendship among peoples" are best advanced by acknowledging the interconnection of religious, racial, and other forms of tolerance and nondiscrimination, and by dealing with them accordingly.

a. The Meaning of Tolerance. Unlike the principle of nondiscrimination, which (as we shall see) is very carefully explicated in the human rights instruments, tolerance is not so defined. We are therefore left more or less to our own devices in trying to understand it.

It is my observation that however important tolerance may seem to be in the human rights field and elsewhere, we are not, when we think about it, awfully clear what it means, or whether it is, afterall, such a good idea. You sometimes hear, for example, that tolerance is really a rather flabby, innocuous attitude. The American literary critic, Stanley Fish cites what he calls "Fish's first law of tolerance-dynamics": Tolerance, he says, "is exercised in an inverse proportion to there being anything at stake." (There's No Such Thing as Free Speech and It's a Good Thing, Too. (Oxford University Press, 1994), p. 217.) On Fish's law, tolerance is the same thing as indifference. In the same spirit, Richard Rorty takes religious tolerance to mean "the willingness of religious groups to take part in discussions without dragging religion into it." ("Interview," The Times Literary Supplement, June 24, 1994, p. 14.) Somewhat anomalously, religious tolerance, for Rorty, amounts to forgetting about religion altogether!

Or, you may hear that tolerance is not relevant to present-day circumstances, because it belongs to a by-gone age when, at best, religions treated each other with disdainful indulgence, rather than full acceptance and respect. James Madison and Thomas Paine held such a view based on their unflattering assessment of European arrangements of the period whereby a dominant, established religion extended deviant believers minimal freedom by sufferance, not equal freedom by right. Echoing Madison and Paine, Stephen Carter makes the same point in his recent book, The Culture of Disbelief. Carter contends that tolerance is a thoroughly dispensable idea since it is very different from the attitude of equal respect that is called for at present. To tolerate another view is to look down upon it, to patronize the person holding it, rather than to accept the view as having merit. (The Culture of Disbelief: How American Law and Politics Trivialize Religious Devotion (Basic Books, 1993), pp. 92-96.)

But not only is the word belittled. It is also used inconsistently. One author starts out defining tolerance as the acceptance of different points of view "without interference or disapproval." But, a few pages later she suggests that to tolerate another view may in fact include disapproving of it. (Martha Minow, "Putting Up and Putting Down: Tolerance Reconsidered," Osgoode Hall Law Journal 28,2 ((1990), pp. 414 and 422.)

Despite all the condescension and confusion, there are good reasons for retaining the word tolerance. Properly understood, it is a coherent and powerful notion, which expresses in a unique way some very important ideas. If we do a little careful thinking, we can see why it occupies such an indispensable place in human rights discussion, as well as why it keeps showing up as a reference point in resolving conflicts around the world. These reasons are, as I shall try to show, linked to a comprehensive and nuanced understanding of the notion.

We do seem to use the word in several different ways. Stanley Fish and Richard Rorty are probably right that one of those ways is, on the surface anyway, close to the attitude of indifference. To describe modern Americans as tolerant about religion could be interpreted to mean that the theological differences among, say, Episcopalians, Presbyterians and Baptists are not at present of much real interest to anyone, and that these groups get on harmoniously at present in part because they simply ignore what distinguishes them one from the other.

However, that way of putting it ignores the fact that the differences were not always of such little consequence. During the middle of the eighteenth century in Virginia, people were routinely flogged and imprisoned over the theological differences of those very three Protestant groups. Indeed, that pattern of ferocious intolerance was the background of Thomas Jefferson's Statute for Religious Freedom, passed by the Virginia legislature in 1786 in order to establish the guarantees of free exercise and nondiscrimination in regard to religious affairs, and thereby to end the religious conflict that disrupted life in Virginia until that time.

However indifferent some of our present-day attitudes toward religious distinctions may be, religious tolerance in this country was hard won. We benefit today from the contribution of people like Thomas Jefferson, who taught us that it is better to endure religious and other fundamental differences than to stifle or suppress them by force. Though we are not always as aware of it as we should be, that conviction has become an indelible part of American subconsciousness.

For if tolerance can mean indifference, it can also mean, as Jefferson understood, "bearing with" or "suffering." The primary meanings of tolerance in the Oxford English Dictionary are "to endure," or "to sustain pain or hardship," "without interference or molestation." On this understanding, tolerance has to include the possibility of disapproval and basic disagreement. It hurts, and sometimes hurts severely, to confront views one finds deviant or objectionable, or to hear criticism of deeply-held beliefs. There is a strong temptation under such circumstances to want to retaliate by paying the opponent back, or by stifling the objectionable views—with force, if necessary. The tolerant person "suffers" or "endures" or "bears with" precisely by restraining rather than releasing the impulse to punish or muzzle the opponent by forcible interference.

It is especially this second meaning—tolerance in the sense of enduring or bearing with—that is particularly pertinent to the task of resolving conflict. It is because tolerance has that meaning that people in Sudan or Sri Lanka or Tibet or Ukraine keep coming back to it as a basis for peace. Unless and until groups of people can learn to accept the pain or hardship of restraining rather than releasing the impulse to suppress and stifle those whose beliefs or characteristics are regarded as deviant or inferior, to give them "free space" and "equal time," there can, apparently, be no real peace.

Besides these two meanings of tolerance, there is also a third we need to mention. If tolerance can mean indifference or bearing with pain or hardship, it can also mean, in the elegant words of the OED, "catholicity of spirit." In this sense, a tolerant person is one who is open-minded, who welcomes diversity. Tolerance according to this meaning does not imply total agreement with or acceptance of other beliefs. If there is complete agreement, the whole idea of tolerance disappears. Rather, it implies that one can learn something from different beliefs without embracing them completely, and it also implies that even where pointed disagreement and disapproval continue, the process of dispute and argument among competing ideas serves everyone together. The pain associated with confronting objectionable or deviant beliefs, or with "suffering" criticism, is of great benefit because in the process views get sharpened, and a person becomes more honest and more self-critical, all in a nonviolent rather than a violent setting. As with physical exercise, there is gain from pain.

The Dalai Lama asserts that tolerance "can be learned only from one's enemy. It cannot be learned from your guru." (Tenzin Gyatso, The Dalai Lama at Harvard, ed. and trans. Jeffrey Hopkins (Snow Lion Publications, 1988), p. 185.) That means that occasions of conflict can be turned to advantage by regarding them as opportunities for mastering the "pain or hardship" associated with disagreement and dispute. Enduring the pain rather than yielding to the temptation to retaliate violently is a virtue that can be learned only by confronting and testing the temptation. Viewed in this way, conflicts over beliefs become, so to speak, "training exercises."

It is interesting that tolero, the Latin root of our English word, means both "to bear, endure, suffer" and "to support, sustain, nourish, protect." The connection between these two groups of words should now be apparent. The strategy of bearing with, enduring, suffering the pain or hardship associated with disagreement and dispute is itself a fundamental way human beings have of supporting, sustaining, nourishing, and protecting one another. By this means, they sublimate hostility and contain violence, thereby profiting from rather than succumbing to conflict. In the matter of tolerance, there is indeed gain from pain—at the deepest levels of human life and experience.

We are now in a position to summarize and condense our findings with respect to the three meanings of tolerance. To tolerate is to respond to a set of beliefs or practices initially regarded as deviant or objectionable, without forcible interference, and

Two fixed characteristics unite all three of the definitions. One is the idea of "pain or hardship"—of "suffering" or "bearing with"—that is closely associated with the idea. In its most important sense, to tolerate is to suffer. That is why tolerance involves a response to beliefs or practices initially regarded as deviant or objectionable. If no tension, no conflict, no "pain or hardship," were encountered, there could be no tolerance. That is even true of the third definition, which involves "a certain respect" for deviant or objectionable views.

The second shared characteristic is forgoing forcible interference. To tolerate is, at a minimum, to resist a temptation to interfere with or to try to influence or suppress the beliefs and practices of others by using force. It is, in essence, "to leave the offending beliefs and practices alone," despite an inclination to act otherwise. In that sense, tolerance includes forbearance.

The crucial variable among the three meanings is the way the offending beliefs and practices are regarded. On the first definition, disapproval is reduced or eliminated, though there must be some detectable residue of awareness that the deviant beliefs and practices were once strongly disapproved of, and that that disapproval has now receded. On the second, disapproval remains. On the third, conflict among beliefs is perceived as having a certain benefit. It is assumed that there is something estimable, something edifying, either in a part of the deviant beliefs themselves, or in the process of give-and-take that occurs among advocates of conflicting ideas, however much disagreement remains. This is what I call "sublimated disapproval," involving a certain respect.

But if the common features are important so are the differences. In particular, the distinction between the second and third meanings calls for special comment. We frequently hear people criticized for being intolerant, not because they have physically assaulted someone they disagree with, but because they have reacted to that person dismissively or disrespectfully. Such a criticism presupposes the third meaning which defines tolerance in terms of an attitude of respect in face of disagreement. On the second meaning, someone is tolerant simply by refraining from responding forcibly to deviant ideas. However, the standard is higher for the third meaning of tolerance. It is not enough to refrain from forcible interference. In addition, one must also exhibit a set of virtues associated with considering opposing views to be worthy of respect, such as practicing attentiveness, fairness and honesty in responding to them, along, no doubt, with a certain degree of emotional restraint.

We shall suggest, as we go, that while the first meaning of tolerance is not without its uses, a human rights approach has an especially strong stake in the second and third meanings, even though we must not lose sight of the distinction between them.

b. Tolerance and Nondiscrimination: Their Interdependence. The interconnection of religious and other kinds of tolerance can best be seen by drawing attention to the intimate relation between tolerance and nondiscrimination that is assumed in the human rights literature. In the Declaration and elsewhere, the principle of nondiscrimination conditions or sets limits to the idea of tolerance. As we mentioned, nondiscrimination is carefully defined. It refers to overt action of a specific kind, namely: "any distinction, exclusion, restriction or preference based on religion or belief and having as its purpose or as its effect nullification or impairment of the recognition, enjoyment or exercise of human rights and fundamental freedoms on a equal basis." (Declaration against Intolerance and Discrimination, art. 2.2.) Any violation of this principle is at the same time a direct violation of the principle of tolerance. Discrimination is thus a crucial index of intolerance.

In the first place, this has an obvious implication for explicitly religious belief. If, for example, we find a religious group, such as the Baha'is in Iran, being singled out, publicly disadvantaged and subjected to imprisonment and other forms of persecution by their government because of their beliefs and practices, that will clearly count as a violation of the principle of tolerance. In addition, if we encounter certain forms of religious expression, such Sinhala Buddhist nationalism in Sri Lanka, which justifies racial and linguistic discrimination against the Tamil minority on religious and cultural grounds, that, too, will constitute a violation of the principle of tolerance. Finally, if there is evidence that a given religion, such as some forms of Islam or Christianity, seek, on scriptural grounds, to deny women public access to educational or employment opportunities, or seek to infringe other human rights protections for women, that also amounts to a violation of the principle of tolerance.

It is not, in other words, only discrimination in which individuals or groups are singled out for abuse or arbitrary treatment because of their religious beliefs and practices that counts as a violation of the principle of tolerance. Any form of discrimination—racial, linguistic, gender-based, etc.—which is justified on religious grounds counts as a violation.

But the Declaration also contains a broader implication for nonreligious belief. The addition of the words, "or belief," to the title and the contents of the Declaration was intended by the drafters to include explicitly nonreligious convictions. I have elsewhere proposed that we should refer to the beliefs covered by the Declaration as "fundamental beliefs," thereby using an appropriately inclusive term. Accordingly, all fundamental beliefs—rather than just religious ones—that function to support discrimination of any kind (racial, linguistic, gender, etc.) would be affected by the Declaration and other pertinent human rights provisions in the same way that explicitly religious beliefs are affected.

That is, if individuals or groups espousing racial or gender discrimination were singled out for mistreatment or were publicly disadvantaged by a government just for holding or admitting to such a view, that would count as a violation of the principle of tolerance. Conversely, "any advocacy of national, racial or religious hatred that constitutes incitement to discrimination...," which means discriminatory action, would also count as a violation (though there is more to it than this, as we shall see in the next section.)

We may summarize our conclusions so far concerning the "indivisibility of tolerance."

First, the principle of tolerance encompasses all fundamental beliefs regarding discrimination, religious or not. Under the Declaration and related human rights provisions, beliefs about race, gender, national origin, language, etc. would be subject to exactly the same protections and prohibitions as would religious beliefs. Accordingly, freedom from religious intolerance and discrimination is interconnected with a similar human right, namely, freedom from racial and other forms of intolerance and discrimination.

Second, the principle of nondiscrimination conditions or limits the idea of tolerance in a profound way. In the light of what we have said about the connection between nondiscrimination and tolerance, we may now reconstruct our definitions of tolerance in the following way: According to a human rights approach, to tolerate is to respond to a set of beliefs or practices initially regarded as deviant or objectionable, without discrimination, and

So long as one does not discriminate, which is to say, does not act so as to nullify or impair the human rights of an individual or group because of religion, race, language, gender, etc., one may be tolerant toward the beliefs and practices of that individual or group in one of three ways. One may come to reduce disapproval and learn to "live and let live"; one may continue staunchly to disapprove of those beliefs and practices; or one may sublimate disapproval and develop a kind of respect for them.

a. Freedom vs. Equality. Put simply, tolerance expresses the principle of freedom, and nondiscrimination expresses the principle of equality.

To be tolerant is to honor a right to liberty. It is, as we have seen, the right not to be interfered with in matters of conscience and fundamental belief, and, therefore, to be able to manifest such beliefs, within limits, by freely uttering, disseminating, and putting them into practice. Even though the concept of tolerance means at least three different things, all three rest, in various ways, on the idea of "suffering with" beliefs and practices taken initially to be deviant or objectionable, rather than attempting to stifle or punish them by force. Disagreement and disapproval continue to obtain, whether by diminishing over time, by remaining firm and explicit, or by being sublimated through a commitment to the merits of diversity and disagreement. In all instances, tolerance amounts to the impulse to bear with the freedom of others to express and manifest their fundamental beliefs, despite a degree of disagreement and disapproval.

To observe the right of nondiscrimination is, straightforwardly, to respect the principle of equality. It is, in the words of the instruments, to disallow "any distinction, exclusion or preference" that would serve to nullify or impair "the exercise of human rights and fundamental freedoms on an equal basis." (Article 2.2 of the Declaration against Intolerance; emphasis added.)

We have seen that the principles of tolerance and nondiscrimination work together. Nondiscrimination qualifies or restricts tolerance so that the expression, dissemination, and practice of fundamental beliefs is to be tolerated to the extent such behavior does not "incite to discrimination," which means it does not lead to overt action that treats people in an illicitly unequal way. Tolerance must be nondiscriminatory, or, to say the same thing another way, freedom must be equal. The right to express and exercise fundamental beliefs cannot be allowed to defeat the right to equal protection. Conversely, of course, the right to equal protection cannot be allowed to defeat the right to freedom of expression and exercise.

Incidentally, because the principle of equality is understood in the documents to qualify tolerance in this way, it would be incorrect to confuse the idea of tolerance from a human rights point of view with the idea as it was historically understood. The objections to the concept raised by Madison and Paine, and reaffirmed above by Stephen Carter, do not apply to a human rights understanding of tolerance, since, in the documents, the idea is now explicitly governed by the principle of equal freedom.

All this is clear enough. Nevertheless, the connection between tolerance and nondiscrimination, or freedom and equality, is not without certain "problems and tensions." One set of problems is posed by the international instruments themselves. Article 20.2 of the Covenant on Civil and Political Rights does not limit itself to prohibiting "any advocacy of national, racial or religious hatred that constitutes incitement to discrimination," but also explicitly prohibits the "incitement to hostility." Similarly, Article 4.a. of the Convention on the Elimination of All Forms of Racial Discrimination stipulates that States Parties shall declare as "an offence punishable by law" not only incitement to racial discrimination and acts of violence, but also "all dissemination of ideas based on racial superiority or hatred,...and also the provision of any assistance to racist activities, including the financing thereof...."

Passages like these have stimulated among government officials and legal and other scholars intense international controversy over the meaning and defensibility of human rights language. They raise the difficulty of trying to impose legal restrictions on the expressing and harboring of beliefs and attitudes, rather than just on certain proscribed kinds of overt action. In that way they appear to challenge the protection of the free expression and exercise of beliefs guaranteed by the right to tolerance.

Terms like "hostility" or "hatred," as well as the phrase, "ideas based on hatred," refer to internal convictions and attitudes or inward dispositions and emotions that are notoriously difficult to define for legal purposes, let alone to control by legal means. The same is true of the intention to punish by law "all dissemination of ideas based on racial superiority or hatred." Moreover, the reference to prohibiting the dissemination of ideas of racial superiority raises the question of whether ideas of superiority in regard to religion or gender or ethnic group are also prohibited by human rights standards. Is, for example, a religious group that publishes and distributes literature claiming its religion to be superior to all others liable to punishment?

Reduced to their core, the arguments for and against the restrictions mentioned in existing human rights documents are these: Those who favor placing legal restrictions on belief and attitude, and who thus advocate laws against "hate speech" and group defamation or libel, contend that without such restrictions, free expression will undermine equal protection. To be subjected to communications (spoken, written or symbolic) that insult a racial, religious, or ethnic group, "whether by suggesting that they are inferior in some respect or by indicating that they are despised or not welcome for any other reason," is, it is said, to be unfairly discriminated against in violation of human rights. (Eric Neisser, "Hate Speech in the New South Africa: Constitutional Considerations for a Land Recovering from Decades of Racial Repression and Violence," South African Journal on Human Rights, p. 337.) According to this argument, excessive freedom generates inequality. (See, for example, Elizabeth F. Defeis, "Freedom of Speech and International Norms: A Response to Hate Speech," Stanford Journal of International Law (1992), who holds that "the First Amendment absolutist approach" to freedom of expression on the part especially of American jurists "has failed to accommodate equality and non-discrimination rights" (p. 59).)

It is contended that slurs and insults based on racial, religious, or ethnic membership are typically more injurious than similar comments of a more individual or "personal" nature. That is because race, religion, or ethnic origin is usually regarded as more central to one's identity than purely individual attributes. It is also because such group membership is very often ascribed rather than achieved, meaning that it is something over which an individual has very little control, and is therefore something for which one is unfairly blamed. Furthermore, it is argued, such comments frequently serve to reinforce discriminatory patterns of exclusion and mistreatment, and thereby increase the victim's sense of subordination and powerlessness.

For their part, the opponents of restriction object that by expanding the regulations on free expression in the name of equality there arises an ominous and potentially uncontrollable threat to the rights of free expression and tolerance, a threat that in addition may well double back and undermine the conditions of equality as well. That is mainly because of the difficulty of framing coherent, consistent, and reliable laws that are capable of governing attitudes or communications disconnected from overt action.

b. Hate Speech Laws: Reasons in Favor. The advocates of restriction are able to find support in the international human rights documents, and in the practice of states. Together with the articles cited above—particularly article 20.2 of the Convention on Civil and Political Rights and 4.a of the Convention on Racial Discrimination, articles 18.3 of the Civil and Political Rights Convention as well as articles 29 and 30 of the Universal Declaration are often invoked. The two additional articles from the Civil and Political Rights Covenant represent "limitations clauses," respectively, on the right of free exercise of thought, conscience, and religion or belief, and the right of free expression. The exercise of religion or belief may be restricted so long as the restrictions are legally authorized, and are "necessary to protect public safety, order, health or morals or the fundamental rights and freedoms of others."

Similarly, restrictions on expression are permissible if they are legally authorized and are necessary for "the protection of the rights and reputations of others," and "of national security, or of public order, or public health or morals." Article 29 of the Universal Declaration adds to the idea of permissible limitations the need to ensure "the general welfare in a democratic society," and according to article 30, it is allowable to restrict any action that threatens the rights and freedoms of others.

One basic argument of those who favor regulating religious, racial, and other forms of expression that "incite hostility" or that disseminate ideas based on racial and perhaps other kinds of hatred and claims to superiority, is that democratic society is especially vulnerable to such language and ideas. According to one observer with first-hand experience of the rise of European fascism in the thirties, democratic institutions, with their permissive emphasis on tolerance, are easy prey for antidemocratic groups. "Democracy and democratic tolerance have been used for their own destruction." (Karl Lowenstein, "Militant Democracy and Fundamental Rights," I and II; American Political Science Review 31 (June 1937) and (August 1937), cited in Samuel Walker, Hate Speech: The History of an American Controversy (University of Nebraska Press, 1994), p. 46.) If the objective is to protect the system of equal freedom associated with constitutional democracy and the guarantee of equal rights for all citizens, then it will be necessary out of self-defense to employ certain antidemocratic measures, such as stringent restrictions on freedom of religion and expression, in order to defeat antidemocratic forces.

Such thinking was reflected in a series of laws adopted by European countries in the thirties, sharply limiting freedom of speech, press, and assembly in face of the growing threat of fascism. Most of these anti-fascist laws were very broadly drawn to outlaw propaganda and group activities taken by the state to threaten public order and security.

Since World War II, and "undoubtedly as a consequence of the tragic lessons" of that event, there has been a further proliferation of national laws against hate speech and group libel. (Natan Lerner, "Group Libel Revisited," Israel Yearbook on Human Rights 17 (1987), p. 192.) In 1966, the Consultative Assembly of the Council of Europe drafted model legislation "making it an offense publicly to call for or incite to hatred, intolerance, discrimination, or violence against persons or groups of persons distinguished by their color, race, ethnic or national origin, or religion, or insulting such groups by holding them in contempt or slandering them on account of their distinguishing particularities." (Lerner, "Group Libel Revisited," p. 192.) While no convention against incitement to hatred has yet been adopted, the large number of current national laws embody in various ways the different aspects of this model law.

In response to its growing racial troubles, England, for example, adopted in 1965 a Race Relations Act, prohibiting utterances or publications "likely to stir up hatred...on grounds of color, race, or ethnic or national origins," and then further tightened restrictions in the Public Order Act of 1986. Together with other anti-hate laws, Germany prohibits statements denying that the Holocaust occurred. France passed a law against racial defamation and insult in 1972, and Belgium followed suit in 1986. The countries have come to outlaw utterances or acts that mock, slander, insult, threaten, or in other ways attack a group of persons on the basis of their nationality, color, race, or religion, as does the Netherlands. Similar laws can be found Poland, Romania, Australia, India, Italy, Greece, Austria, a number of Latin American countries, and other countries. (Lerner, "Group Libel Revisited," pp. 192-193.) Moreover, the European Human Rights Court has generally upheld such laws, so long as they are necessary to protect a democratic society, and are proportionate to the threat.

Interestingly, in 1990, the Conference on Security and Cooperation in Europe issued what was known as the Copenhagen Document, which reflected recent developments in Europe, including "hate campaigns against certain minority groups, increased anti-Semitic harassment, and acts of...racial violence...." (Defeis, "Freedom of Speech and International Norms," pp. 122-123.) Article 40.2 explicitly urges governments to commit themselves "to take appropriate and proportionate measures to protect persons or groups who may be subject to threats or acts of discrimination, hostility or violence as a result of their racial, ethnic, cultural, linguistic, or religious identity...."

Canada and Hungary present recent examples of efforts to enforce laws against hate speech and group libel. In 1993, the Canadian Supreme Court, in face of a guarantee of free expression in the Canadian Charter of Rights and Freedoms, upheld the conviction of a school teacher who was accused of communicating anti-Semitic statements to his students. (The Queen v. Keegstra [1993] 3 S.C.R. 697, 756 (Can.)), cited in Defeis, "Freedom of Speech and International Norms," pp. 72-73.) The teacher was judged to have violated the Criminal Code by intentionally inciting hatred against a particular group. The Court ruled that the anti-Semitic message of the accused, to the effect that "members of identifiable groups are not to be given equal standing in society, and are not human beings equally deserving of concern, respect and consideration," is "directly counter to the values central to a free and democratic society....[I]n restricting the promotion of hatred Parliament is therefore seeking to bolster the notion of mutual respect necessary in a nation which venerates the equality of persons." (Cited in Defeis, "Freedom of Speech and International Norms," pp. 72-73.)

The Hungarian Criminal Code, as revised in 1989, penalizes the incitement to hatred against "the Hungarian nation or any nationality, any people, faith or race or single groups of population...." (Andras Sajo, "Hate Speech for Hostile Hungarians," East European Constitutional Review (Spring 1994), pp. 83-84.) In response to a number of cases brought under this law, the Constitutional Court upheld its constitutionality. It took a broad view, arguing that incitement to hatred is prohibitable even if the act of incitement does not constitute "a clear and present danger," that is, a substantial and imminent threat of violence or severe social disruption. Citing recent European experience, the Court held that the promulgation of ideas of racial superiority and hatred do often disturb the peace and undermine social harmony, and for that reason should not be protected. Beyond that, whether such incitement happens in a given instance to imperil domestic tranquillity or not, the advocacy of such ideas by definition offends "a sacrosanct constitutional value," namely, the basic principles, no doubt including the principle of equality, on which democracy rests. (Hajo, "Hate Speech for Hostile Hungarians," pp. 84-85).

c. Hate Speech Laws: Reasons Against. If hate speech and group libel laws have attracted extensive support, they have also been sharply criticized. In 1965, the Columbian representative to the UN registered a familiar complaint. Such laws are "a throwback to the past" because they place an arbitrary restriction upon freedom.

[P]unishing ideas, whatever they may be, [he said] is to aid and abet tyranny, and leads to the abuse of power....As far as we are concerned and as far as democracy is concerned, ideas should be fought with ideas and reasons; theories must be refuted by arguments and not by the scaffold, prison, exile, confiscation or fines. (U.N. Doc. A/PV.1406, at 42-43 (1965); cited in Natan Lerner, "Incitement in the Racial Convention; Reach and Shortcomings of Article 4," Israel Yearbook on Human Rights, 22 (1993), p. 5.)

Moreover, on a more recent accounting, the overall results of these laws are deemed "not encouraging." (Eric Neisser, "Hate Speech in the New South Africa," pp. 348-349.)

The experience in England,...and...Germany, where racist speech has long been banned, indicated that such bans will not prevent hateful invective, development of hate organizations, or, more importantly, hate crimes....[I]n England, after 25 years, the laws criminalizing incitement of racial hatred have had little effect on the national Front and other neo-Nazi groups. Likewise, although the laws in Germany have reduced or dismantled some particular hate groups, they plainly have failed to stop the growth of...skinheads or the recent vicious crimes against immigrants. One possibility, of course, is that public prosecution or other repression of racist groups may give them visibility and the attractiveness of forbidden fruit and thereby enhance their recruitment ability.

Even more distressing...is the fact that prosecutorial discretion has often been used to prosecute minorities and other victims of racism, rather than to protect such victims from further insult. In England, for example, the first individuals prosecuted under the Race Relations Act were black power leaders, and the law ever since has been used more often to curb the speech of blacks, trade unionists, and anti-nuclear activists, than to limit the expression of racists. In the ultimate irony, the English statute that was intended to restrain the neo-Nazi National Front has banned expression by the Anti-Nazi League. Similarly, when the British National Union of Students adopted a resolution to prevent openly racist organizations from speaking on college campuses, the first victims were Israelis and Zionists, based on the UN resolution equating Zionism with racism.

That such laws are typically ineffective and self-defeating, as well as the subject of arbitrary enforcement, has also been concluded in a survey of European hate speech laws published in 1992 by Article 19, the London-based free-speech monitoring group, as well as in recent comments on the effects of Canadian hate laws. (Sandra Coliver, ed., Striking a Balance: Hate Speech, Freedom of Expression and Non-discrimination (London: Article 19, 1992), esp. Coliver, "Hate Speech Laws: Do They Work?" pp. 363-374); Stefan Braun, "Can Hate Laws Stop Hate Speech?" Moment (August 1993).)

The American experience with laws restricting speech, including hate speech and group libel, yields strong grounds for thinking twice about such restrictions. During World War I, the United States Congress sharply narrowed the range of free expression by passing the Espionage and Sedition and the Alien Acts against what were perceived at the time to be severe threats to the national interest. These laws authorized government officials to interdict publications and to prosecute individuals for making utterances that exhibited a bad-tendency-born-of-bad-intent, as the test came to be known. Instead of having to prove that spoken or written words directly cause punishable action, statements need only be shown to have an "intent" to injure the state in some way, and a "tendency" in that direction. (Zechariah Chafee, Jr., Free Speech in the United States (Harvard University Press: 1941), pp. 50ff.)

The manifest vagueness of such statutes led to what has been described as the "massive suppression" of civil liberties, during and after World War I, and as something which "did lasting damage." (Samuel Walker, Hate Speech, p 48.) That included the elimination, effectively, of the Socialist party and the Industrial Workers of the World, an anti-capitalist labor organization widely considered to be a conduit of foreign and subversive ideas. Having made a thousand or so arrests and elicited some five hundred indictments, the government's case against the IWW rested essentially on the bad-intent-and-bad-tendency test, since it was impossible in all but a very few cases to prove that members had committed overt acts of violence. As an exhaustive Justice Department review of IWW literature admitted, violent resistance was at best "hinted at" rather than expressly advocated. (William Preston, Aliens and Dissenters: Federal Suppression of Radicals, 1903-1933 (Harvard University Press: 1963), p. 101.)

Nor did things end there. "The furies of repression continued through the 1920s and into the 1930s, crippling both the labor movement and the political Left. It also established the principle of guilt by association such that anyone who defended the rights of dissidents was immediately branded 'un-American.'" (Walker, Hate Speech, p. 48.)

Experiments with hate speech laws need to be understood in this context. In 1934, the New Jersey legislature adopted an anti-race-hatred law after several violent confrontations had taken place between Nazi and anti-Nazi groups. The law imposed criminal penalties for disseminating "propaganda or statements creating or tending to create prejudice, hostility, hatred, ridicule, disgrace or contempt of people...by reason of their race, color or creed or manner of worship." (Walker, Hate Speech, p. 55; emphasis added.)

In reality, the law was very rarely enforced, and no constitutional test of it occurred until 1940 when a case reached the New Jersey supreme court involving the conviction of August Klapprott and several other members of a German-American group for disseminating race hate propaganda. In State v. Klapprott (22 A. 2d 877 (1941)), the court overturned the conviction for violating constitutional guarantees of free speech, particularly because of the vagueness of the law, and proceeded to strike down New Jersey's race hate law.

According to the court, terms like "hatred," "abuse," and "hostility" were given no precise definition, and were thus susceptible to the most arbitrary interpretation. Speech might be regulated, the court went on, only if it represents a "clear and present danger" to the state, invoking Justice Holmes's famous test laid down in Schenck v. U.S. (249 US 47 (1919)). That test moved the standards of criminal speech away from indeterminate questions of the "intent" and "tendency" of words, and toward a strict consideration of whether or not words directly incite overt violations.

In judging the law unconstitutional, the court invoked Cantwell v. Connecticut (310 U.S. 296 (1940)) in which the U.S. Supreme Court had overturned the conviction of a Jehovah's Witness for using a portable record player to broadcast attacks on Roman Catholicism in a Catholic neighborhood. The court ruled that the attacks, while clearly offensive and abusive to Catholic ears, did not represent a clear and present danger or an overt breach of the peace.

In the realm of religious faith, and in that of political belief, sharp differences arise. In both fields the tenets of one man may seem the rankest error to his neighbor. To persuade others to his own point of view, the pleader, as we know, at times, resorts to exaggeration, to vilification of men who have been, or are, prominent in church or state, and even to false statement. But the people of this nation have ordained in the light of history, that, in spite of the probability of excesses and abuses, these liberties are, in the long view, essential to enlightened opinion and right conduct on the part of the citizens of a democracy....There are limits to the exercise of these liberties...[namely,] the coercive activities of those who in the delusion of racial or religious conceit would incite violence and breaches of the peace....These and other transgressions of those limits the States appropriately may punish."

A year later, the Supreme Court appeared to pull back from the permissive position taken in Cantwell. In Chaplinsky v. New Hampshire (315 U.S. 568 (1942)), the court upheld a conviction against another Jehovah's Witness for calling a police officer "a God-damned racketeer" and "a damned Fascist." There are, said the court, "certain well-defined and narrowly limited classes of speech, the prevention and punishment of which have never been thought to raise any Constitutional problem." These include two kinds of "fighting words": words that "tend to incite an immediate breach of the peace," and words that "by their very utterance inflict injury." (Walker, Hate Speech, p. 71.)

However, neither of these categories survived without later modification. Breach of the peace as a reason for restricting speech seemed self-evident enough, but there were complications even there. In Teminiello v. Chicago (337 U.S. 1 (1949)), the Supreme Court reversed the conviction of a defrocked priest, whose abusive language provoked a crowd gathered outside the hall where he was speaking to go on a rampage causing property damage in the immediate area. The implication of the decision was that under some circumstances it may be the audience, not the speaker, who should be held liable for the outbreak of damage or violence. Otherwise, speech would invariably be held hostage to the whims and prejudices of a given audience, and thereby constitute a kind of "heckler's veto." In his opinion, Justice Douglas argued as much. Free speech, he said, will by its nature provoke dispute, and that result should be welcomed, not resisted. "[Free speech] may indeed best serve its high purpose when it induces a condition of unrest, creates dissatisfaction with conditions as they are or even stirs people to anger." (Cited by Walker, Hate Speech, p. 106.)

This particular modification of the breach of peace test would have the weightiest possible impact on decisions handed down in the 1960s involving civil rights and anti-Vietnam protests. In the case of civil rights, demonstrations and public speeches by minority leaders and their followers were consistently perceived as threatening and offensive to majority attitudes, institutions, and laws. However nonviolent such protests might be, by their nature they "stirred people to anger," "induced a condition of unrest," and were intended to "create dissatisfaction with conditions as they are."

Should the threat of unruliness and hostile action be permitted to silence such expression? U.S. courts tended to think not. In fact, a march in 1966 through Cicero, Illinois undertaken by Martin Luther King, Jr. and his followers in face of severe racial tensions and potential disruption would be cited ten years later in support of the famous decision to permit an American Nazi group to demonstrate in Skokie, Illinois. "Allowing community sensibilities to veto political expression would block Nazis in one context but also stifle civil rights groups in others, both North and South." (See Walker, Hate Speech, pp. 107-8.)

The Supreme Court temporarily toyed with the idea of group libel as a basis for regulating speech in Beauharnais v. Illinois (343 U.S. 250, 282 (1952)). That case represented a challenge to an existing but seldom enforced group libel law passed by the Illinois legislature in 1917, making it illegal "to manufacture, sell, or offer for sale, advertise or publish, present or exhibit in any public place...[anything that] portrays depravity, criminality, unchastity, or lack of virtue of a class of citizens, of any race, color, creed, or religion," when such material would expose any individual member "to contempt, derision, or obloquy or which is productive of breach of the peace or riots." (Walker, Hate Speech, p. 93). Joseph Beauharnais, president of the White Circle League of America, was convicted under the Illinois law of disseminating racist literature that was unquestionably contemptuous of all African Americans, and portrayed them as criminal by nature.

Writing for the five-to-four majority, Justice Frankfurter supported Beauharnais's conviction by drawing on the two "fighting words" tests (breach of peace and injurious utterances) laid out in Chaplinsky. Legislative majorities concerned to deter offensive utterances and the possibility of strife associated with racial conflict might reasonably be allowed to restrict speech. However, the majority position did not go unchallenged. The dissenting justices criticized the opinion for ignoring both the vagueness of the original law and its chilling implications. What may be used today to curb the venomous outbursts of white racists, warned Justice Douglas, may tomorrow be turned against "a Negro...for denouncing lynch law in heated terms." (Walker, Hate Speech, pp. 95-97.)

The dissenters prevailed in the long run. Group libel law lost its appeal as a remedy for racial and religious intolerance in U.S. law. One important reason was precisely the concern voiced by Justice Douglas. Civil rights groups, including organizations like the American Jewish Congress, who had once been sympathetic, came to recognize how such laws might actually be used to obstruct the cause of racial justice by inhibiting free and open criticism. In promoting racial equality, advocates would find it hard to avoid uttering harsh and pointed verbal attacks against the record of "white society." But such attacks could very easily be shown to constitute a direct violation of group libel law, since they attribute a "lack of virtue" (if not "depravity, criminality, and unchastity") to a certain "class of citizens," in a way that appears to expose members of that class "to contempt, derision, or obloquy." "[T]he success if not the very survival of civil rights activity depended on the protection of provocative, sometimes offensive, and occasionally even hateful speech. For the powerless and the excluded, speech was often the only resource available." (Walker, Hate Speech, p. 160.)

In response to growing expressions of overt racism on a number of American college campuses, the question of regulating hate speech revived in the 1980s and early 90s in a sharp controversy over campus speech codes. Advocates reactivated the ideas of group libel and "fighting words"—understood especially as injurious utterance, and added to them the claim that an absolute commitment to freedom of speech, including racial epithets and insults, contributes to inequality and discrimination, particularly of minorities. It was also suggested that the arguments in favor of protecting the workplace by restricting certain abusive forms of sexually-oriented speech might be applied to race relations in colleges and universities.

There followed a proliferation of campus speech codes. Some, like Stanford University's, were narrowly drawn with an eye toward the First Amendment. Others were much more expansive. The University of Connecticut code prohibited "inappropriately directed laughter [and] inconsiderate jokes," as well as the "conspicuous exclusion" of individuals from conversation. (Walker, Hate Speech, p. 133.) The University of Michigan code prohibited all speech, including speech in the classroom, that "stigmatizes or victimizes an individual on the basis of race, ethnicity, religion, sex, sexual orientation, creed, national origin, ancestry, age, marital status, handicap, or Vietnam-era veteran status." (Julie Gannon Shoop, "Freedom Versus Equality: Battle over Hate Speech," Trial (January 1991, p. 12.)

When challenged in court, the Michigan and Wisconsin codes were both struck down. (Doe v. University of Michigan 271 F. 852 (E.D. Mich.1989); UWM Post v. Board of Regents of the University of Wisconsin 774 F. Supp. 1163 (E.D. Wis. 1991).) The Michigan code was deemed overbroad and unconstitutionally vague. Likewise, the Wisconsin code was judged to be "unduly vague because it left unclear whether speech, to be violative, must actually demean the listener or merely be intended to do so." It was also called "overbroad because it targeted otherwise constitutionally protected speech," and "did not restrict only 'fighting words' because it penalized statements not likely to provoke a violent response." (Thomas A. Schweitzer, "Hate Speech on Campus and the First Amendment: Can They Be Reconciled?" Connecticut Law Review 27:493 (1995), pp. 503-504.)

"The weight of legal opinion is that both the Michigan and Wisconsin cases were correctly decided, and that it is exceedingly difficult to draft a university hate speech code that can pass constitutional muster." (Schweitzer, "Hate Speech on Campus," p. 504.) For one thing, during the time the Michigan code was enforced, not a single case of white racist speech was punished, whereas more than twenty African Americans were charged by whites with violating the code. (Henry Louis Gates, Jr., "Let Them Talk," New Republic (September 20,27, 1993), p. 44.) Beyond that, widely publicized examples of the difficulty of classifying epithets, such as the University of Pennsylvania case in which the words "water buffalo" were mistakenly taken to convey racial inferiority when they were actually meant to censor offensive behavior, have dramatized the problems of regulating speech.

The Stanford code is widely heralded as a carefully crafted code that avoids the objections to the Michigan and Wisconsin codes. It is, in the words of its author, "dramatically narrower" than other codes, limiting restrictions to "commonly recognized epithets," which, it is argued, fall within the tradition of unprotected "fighting words." Moreover, the remarks must convey "direct and visceral hatred or contempt" on the basis of race, sex, color, handicap, religion, sexual orientation or national and ethnic origin that is "addressed directly to the individual or individuals whom it insults or stigmatizes." (Gates, "Let Them Talk," p. 45.) "General statements about a minority group would not be punishable." (Julie Gannon Shoop, "Freedom Versus Equality, p. 13.)

There are, however, several problems even with the Stanford code. A primary difficulty is that the code would not cover racial or other group-related insults so long as they are civilly expressed. Because "the polite putdown can be much more devastating than...a racial epithet," an important—perhaps the most important—kind of abusive speech is unaffected by the code. "In American society today, the real power commanded by racism is likely to vary inversely with the vulgarity with which it is expressed." (Gates, "Let Them Talk," p. 45.) Moreover, such neglect may well lead to an unanticipated form of discrimination. "Banning epithets, but not skillful rhetorical skewerings, would essentially and unjustifiably discriminate against low-brow forms of expression." (Larry Alexander, "Banning Hate Speech and the Sticks and Stones Defense," Constitutional Commentary 13:71 (1996), p. 78.)

There is, undoubtedly, widespread agreement that the idea of tolerance should be regulated by the principle of equal freedom. It would obviously contradict human rights standards to permit individuals or groups to take advantage of the right to tolerance—namely, the freedom to express and practice fundamental beliefs—in such a way as unduly to interfere with or to frustrate the similar right to tolerance of other individuals and groups.

As our exposition of the debate over hate speech laws has made plain, however, differences arise regarding what constitutes undue interference with and frustration of the rights of others. Differences arise, in other words, over the question of the proper limits to tolerance. Do those limits extend exclusively to the punishment of overt action that is discriminatory or violent, or should they also properly include the regulation of expression that is disconnected from such action, and instead directly affects only inner attitudes and sentiments?

Enough has been said to show the reason why the problem is so urgent and of such great concern, as well as why it causes so much perplexity. Aggressive intolerance, or the unrelenting expression of racial, religious, and ethnic superiority, has in recent experience repeatedly and unforgettably represented a potential threat of the severest kind to domestic and international peace. Individual states, as well as the international community, are understandably concerned about limiting the effects of such bigotry. At the same time, the record of European and North American countries, and, in particular, the United States, powerfully illustrate the huge obstacles and liabilities that stand in the way of attempting to regulate racial, religious, ethnic and other comparable forms of speech when they are disconnected from punishable action.

For reasons we have reviewed, the effort to frame coherent, consistent, and reliable laws in this area is an unenviable task. Not only is it hard to avoid being vague and overbroad. Even more frustrating from the point of view of combatting discrimination is the difficulty of designing laws that do not themselves have the effect of contributing to or reinforcing discrimination and inequality. Since a fundamental objective of "hate speech" laws is advancing the cause of equality and minimizing the ill-effects of untrammeled free expression, it is especially ironic that these laws should again and again, and in a variety of different ways, be shown to work out against the principle of nondiscrimination!

Whether it is possible to regulate by law inner attitudes and sentiments like hatred and hostility without inconsistency and self-contradiction is open to question. Whether, that is, hate speech laws finally advance the implementation of tolerance and nondiscrimination, and, hence, serve the ultimate cause of peace these principles were designed to secure is, given the evidence, doubtful.

What is, however, not doubtful is that the principles of tolerance and nondiscrimination, taken together, and understood inclusively as applying to "race, color, sex, language, religion,...national and social origin," etc., correlate positively with peace and with the reduction of violence. In the "institutionalized democracies," where tolerance and nondiscrimination are relatively well-implemented, "the calculus of communal action...favors protest over rebellion,...and the opportunity structure...provides incentives for [nonviolent] protest and disincentives for rebellion," according to a recent authoritative survey. (Ted Robert Gurr, et al., Minorities at Risk: A Global View of Ethnopolitical Conflicts (USIP Press, 1993), p. 137.)

It follows that in those societies in which there is wide compliance with the norms of tolerance in a number of different areas—in religion, race, gender, language, there will be a higher degree of social tranquility and nonviolent communication. In societies like Sri Lanka and Sudan, religion, race, and language become intermingled, and the reduction of tension in one of the areas will presumably mean reduction of tension in the other areas as well. (See, for example, David Little, Sri Lanka: The Invention of Enmity (USIP Press, 1995), for an elaboration of the interconnection of religion, language, and race in the Sri Lankan ethnic conflict.)

Policies designed to advance the cause of tolerance and nondiscrimination should, broadly speaking, be conceived of in a two-pronged way. One prong is obviously to remove the barriers of discrimination from public life, starting with legal and political reform. Also involved, however, is the subject of compensatory adjustments such as affirmative action programs and other measures aimed at rectifying severe economic and social inequality. (See Barbara R. Bergmann, In Defense of Affirmative Action (Harper Collins, 1996) for a persuasive vindication of the idea of affirmative action with interesting international implications.) In keeping with the concept of nondiscrimination contained in the international documents, this policy prong primarily concerns overt action. The object is to ensure that citizens act in conformity with the principle of equal protection under the law.

All the respectable peace proposals in places like Sri Lanka, Sudan, and Tibet, for example, specifically include provisions for one formula or other of equal treatment as between the majority and minority populations who are in conflict with each other. All of the proposals contain provisions for expanded forms of cultural, linguistic, religious, and political autonomy or self-direction within a broader federal structure that are intended to weaken the patterns of discrimination and majority domination characteristic of such cases.

The second policy prong, of course, concerns the cultivation of tolerance. According to the human rights norms, as we have seen, tolerance presupposes the principle of nondiscrimination, but it is not the same thing. Nondiscrimination involves overt action, whereas tolerance includes matters of inner attitude and outlook. The coercive mechanisms of government are relevant to tolerance by enforcing nondiscriminatory behavior, but it is unlikely that coercion and law alone can promote the attitude of tolerance, as our discussion of the perplexities of hate speech laws showed. Policies designed to promote tolerance, therefore, will have to include more than just behavioral reform.

Policies aimed at cultivating tolerance are undoubtedly also a matter of education and inspiration. These activities are partly the responsibility of the government, but they are also the responsibility of the organizations and associations that make up "civil society"—religious bodies, civic groups, voluntary associations, so on. How all this is to be worked out is a topic for separate consideration.

What ought to be included in such consideration is the contribution to peace and stability of the three different types of tolerance mentioned earlier. It seems likely that a relatively high incidence of the third type of tolerance—"sublimated disapproval involving certain respect" for deviant beliefs—would be required for achieving enduring peace in a pluralistic society. That would involve considerable common commitment to the norms of nondiscrimination that undergird tolerance, together with the associated norms of free expression, freedom of belief and conscience, respect for a diversity of ideas in regard to religion, race, gender, etc.

It is, at the same time, highly unlikely that the first and second kinds of tolerance can be left out of account altogther. Refraining from interfering in the beliefs of others while simultaneously detesting those beliefs is no doubt a rudimentary form of tolerance, but it is indispensable to peace. Getting people to restrain their natural tendency to strike back and hurt or stifle those with whom they sharply disagree is an essential part of the cultivation of tolerance. In a number of conflicted societies, to achieve even this second kind of tolerance would be a huge step forward. Moreover, it will remain as an unavoidable fact of human life that certain beliefs will inevitably be considered so objectionable to some elements of society that in respect to those beliefs at least, noninterference accompanied by strong, sustained disapproval is the most that can be hoped for.

Even the first kind of tolerance—noninterference with diminished disapproval—may have its uses. One author argues convincingly that in the case of indigenous populations, noninterference accompanied by a strong sense of majority disapproval will make it very hard for that way of life to endure at all. (Minow, "Putting Up and Putting Down," pp. 422-423.) Whether widespread passive acceptance of deviant beliefs and practices can sustain a tranquil society without the third kind of tolerance—sublimated disapproval based on an active appreciation of the benefits of diversity— remains to be investigated.