by

Russell Blackford

The presence of racial hate propaganda in our community and on the internet discomfits free speech advocates and severely tests the principle of free speech and expression. This underscores a dramatic change of social values that took place in the twentieth century, from a time when many other forms of speech and expression were constrained by law or popular sentiment but, as described by Samuel Walker in respect of 1920s America (Hate Speech, 1994), "unabashed racist thinking" was acceptable and frequently expressed "even among the most educated members of society." Now, by contrast, redress is demanded confidently for every insult or offence to members of racial and other minorities that have come under the protection of anti-discrimination law.

In its overall spirit, this revaluation was justified, but its manifestations include new kinds of moral rigidity and illiberalism. With its exacting standards of politeness and inoffensiveness, contemporary human rights law often conflicts with basic individual liberties such as freedom of thought, conscience, speech and expression (all of which I'll refer to loosely as "free speech"). Both the new standards and the older liberal values have something to recommend them; when they clash the intellectual and practical problems are genuinely difficult, suggesting that compromise may be necessary.

What should be done about the evil of racial hate propaganda without abandoning a commitment to free speech? A starting point is to identify the deeper considerations that underlie freedom of speech as a political principle. In part, the justification for free speech is the general libertarian presumption against coercing individuals from living how they please and doing what they want. However, a number of justifications are commonly put forward for free speech, in particular. In her judgment in R v Keegstra, a 1990 Canadian Supreme Court case on hate propaganda, Justice McLachlan provided an accessible and concise synopsis: (1) free speech promotes "the free flow of ideas essential to political democracy and democratic institutions" and limits the ability of the state to subvert other rights and freedoms; (2) it promotes a marketplace of ideas, which includes, but is not limited to, the search for truth; (3) it is intrinsically valuable as part of the self-actualisation of speakers and listeners; and (4) it is justified by the dangers for good government of allowing its suppression.

Though other justifications are sometimes adduced, such as the need to obtain information about products and services to make enlightened choices in a consumer society, those stated by Justice McLachlan cover the most persuasive considerations. She has described a complex of well-known reasons, distinguishable but overlapping, why free speech should be recognised as a basic liberty, even if, as she puts the point, "free expression may put other values at risk".

Each of these justifications can be elaborated in a variety of ways and some may need to be qualified. The first and fourth can be bracketed together as democratic justifications. They relate to aspects of free speech's political role in a liberal democracy. The third relates most closely to general libertarian values but stresses the particular importance of language, symbolism and representation for our lives and autonomy. It can be developed further by referring to the importance for individuals of communicating deeply held religious and similar beliefs and the value of creativity as expressed in literature, art and many other ways, including personal presentation or "style".

The issue here is how we should treat other individuals as moral and psychological beings. We might refer to this as the "moral" justification, if we want to distinguish between political principles and principles that relate to the moral rights of individuals and our intuitions about how people should be treated. However, it also highlights the fact that we are beings with psychological needs that involve self-expression and self-actualisation. With this in mind, and with some misgivings about the expression, I will refer to the "psychological" justification of free speech.

Alan Haworth (Free Speech, 1998), has suggested that the metaphor of a marketplace of ideas, the second justification mentioned by Justice McLachlan, is misleading. As Haworth explains, John Stuart Mill's classic defence of free speech, in On Liberty (1859), is actually phrased as a defence of "the Liberty of Thought and Discussion". Mill did not express this idea in terms of a "marketplace" and his conception was essentially that of freedom to develop and discuss ideas in the search for truth or understanding. Mill pictured society not as a marketplace of ideas, but as something like a large-scale academic seminar. A seminar, of course, is an effective forum for refining and testing ideas only because its participants can rely upon tacit standards of conduct and interaction, including some degree of mutual respect.

Another way of putting this point is that free speech of certain kinds is integral to rational inquiry. If we value this, we should also advocate the liberty to articulate potentially unpalatable ideas and unpopular social critiques. That, however, is quite a narrow concept compared with freedom of speech as commonly understood, which includes robust and even offensive kinds of interaction that would be strongly inhibited, if not actually forbidden, in an academic seminar.

Mill's argument might be termed the rationalist justification. It is very powerful as far as it goes but inevitably somewhat elitist, for relatively little speech and expression in real-world societies appeals primarily to the intellect. However, there are senses in which the rationalist justification can be extended beyond the speech of academics, scientists and other intellectuals. In one sense, it merges with the psychological justification, if it is interpreted as our individual need to pursue truth and understanding in our way own, necessarily reliant on resources available through language. In another sense, it encourages us to protect serious literature and art, especially narrative forms such as prose fiction, theatre and cinema, one function of which is to open minds by appeals to the imagination.

This analysis suggests a number of conclusions. First, there are powerful overlapping arguments for free speech as a basic political principle in any liberal democracy. Second, however, free speech is not a simple and absolute concept but a liberty that is justified by even deeper values. Third, the values implicit in the democratic, rationalist and psychological justifications for free speech will not apply equally strongly to all speech in all circumstances, and these "free speech values" may sometimes have little application at all. For example, they are not seriously at stake in arguments about purely commercial advertising.

If free speech is to be a political or constitutional principle that imposes practical restraints on the coercive power of the state, it needs to be formulated in a relatively simple and sweeping way. It cannot track the precise relevance of underlying values in every circumstance where speech might be suppressed. It follows that a constitutional restriction on state interference with free speech might give practical protection to some speech that has little to do with democratic, rationalist and psychological values. This creates a buffer zone around the more central areas where free speech values apply, and that may well be desirable. In particular circumstances, other values might be more important than free speech but any exceptions to the principle must be defined carefully; otherwise, they will soon gobble up the rule.

#

In the US, the First Amendment of the Bill of Rights has been interpreted as giving constitutional protection to a vast range of speech and other symbolic expression. Any speech of "serious literary, artistic, political or scientific value" falls within the protected zone, as stated in the key case of Miller v California (1973). Political speech, in particular, does not need to be sophisticated, or even rational, to attract constitutional protection, while "political value" simply means that some political viewpoint is being expressed. As shown by American cases that have accorded constitutional protection to the display of Nazi swastikas or Ku Klux Klan burning crosses, the viewpoint expressed may be positively malevolent.

In Australia, the constitutional situation is very different. In the absence of any express constitutional protection of free speech, it is open to the state parliaments to prohibit speech that has serious artistic or intellectual value or which is basic to everyday self-expression. The power of the federal parliament is more limited in the sense that it can legislate only on matters falling within a head of power set out in the Australian Constitution. Some federal heads of power are interpreted by the courts as "purposive", meaning that they confer a power to enact statutes for particular purposes, such as national defence. Since anything can be enacted for almost any purpose at all, or at least asserted to be, the courts require some rational connection and proportionality between the enactment and the supposed purpose. This opens the possibility that legislation attacking basic rights and liberties will sometimes be held to exceed the power of the parliament. Thus the High Court struck down a ban on the use of the everyday words "200 years" in proximity to "1788" or "1988", which was part of an overly zealous effort to promote Australia's bicentenary celebrations: see Davis v The Commonwealth (1988).

In many contexts, the sole constraint on the power of the Australian legislatures is the doctrine of an implied constitutional freedom of political speech. Developed by the High Court through the 1990s, this doctrine prevents political censorship, at least for its own sake. As formulated in Lange v Australian Broadcasting Corporation (1997), currently the Court's definitive word on the subject, the test of a federal or state law's validity is two-fold. First, does it burden free speech about government and political matters? Second, to use the Court's jargon, is it reasonably appropriate and adapted to serve a legitimate end compatible with the procedures for federal elections and constitutional referendums set out in the Constitution? If the answer to the first question is "yes" and that to the second question is "no", a statute is unconstitutional.

In Lange, the Court acknowledged that the federal legislation scrapped in Australian Capital Television v The Commonwealth (1992) was an example of a law that might properly be held unconstitutional. Though it had the legitimate end of tackling political corruption, the means were too drastic. It severely limited political advertising in a way that would have undermined the ability of lobby groups and minor parties to compete with the major parties for the hearts and minds of the electorate. A bad law went through the judicial shredder and this was cause for rejoicing, though not a proof of sound legal reasoning.

The cases show that the limited constitutional protection of free speech in Australia is an inference from those provisions in the Constitution that entrench voting procedures for the houses of federal parliament and for constitutional referendums. The reasoning is that such voting procedures presuppose a liberty for members of the electorate to articulate, receive and debate ideas that are relevant to the exercise of their democratic choice. State or federal legislation that impacts disproportionately on this liberty is thus repugnant to a presupposition underlying the Constitution's structure and operation.

This is a narrow protection based on reasoning that is neither as obviously flawed as asserted by critics of judicial activism nor entirely compelling to a sceptical mind. A future High Court judgment could overturn the free speech jurisprudence built up in cases such as Australian Capital Television and Lange, though this is unlikely in the medium term, since Lange was clearly intended to lay down a broadly acceptable compromise rule. All the same, the situation in Australian law remains unsatisfactory. A great deal of material that would be constitutionally protected in the US because of its artistic or intellectual value receives no protection in Australia. Nothing in our constitutional arrangements gives expression to the rationalist and psychological justifications for free speech. At the same time, such constitutional protection as does exist for political speech has an intellectually dubious basis.

For these reasons, I favour an express and clear protection of free speech in the Australian Constitution, perhaps along the lines of that in the Canadian Charter of Rights and Freedoms, which refers to "freedom of thought, belief, opinion and expression, including freedom of the press and other media of communication". That, however, is not the end of the story. In Canada, the rights and freedoms contained in the Charter are expressed to be "subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society." The effect of this is to hand the courts the task of developing a jurisprudence in respect of what kinds of laws "can be demonstrably justified". This is jurisprudentially legitimate, since there is an express constitutional basis for the courts' activity. It is democratically tolerable, since the task is restricted to approving legislative action that would otherwise be filtered out by the nation's constitutional arrangements. Such an approach might well be followed in Australia.

#

Free speech is not an opaque and indefeasible principle that defies analysis and overrides every other consideration that might ever be adduced. It is based on deeper values relating to how we wish our society to function and how we should treat each other as beings with psychological and moral interests. The relevance of these values will vary with circumstances.

Laws relating to child pornography and to defamation are examples of restrictions most free speech advocates consider legitimate. The reasons for suppressing child pornography are obvious enough, but the law of defamation impinges seriously on robust democratic debate and individual self-expression. Its legitimate role is based on the fact that human beings are essentially social animals. If an individual's good reputation in the society where she lives, and on which she depends economically and psychologically, is destroyed, that is a significant injury.

The law of defamation is sometimes absurdly overprotective, and I advocate narrowing its operation rather than extending it further through concepts such as group libel. This, however, does not detract from the argument that there are limits to the attacks upon individuals' reputations that should be tolerated by law. It may be easy to accept restrictions on free speech in the case of defamation law merely because the idea is so familiar. Bearing that in mind, there may be other legitimate reasons for restricting particular categories of speech. Some may not have seemed cogent at earlier stages of our society's development.

I have expressed unease about arguments for restrictions on free speech that rely on concepts such as group libel. Is this unease justifiable? Speech can destroy individuals if it destroys their individual reputations in society. For example, it would do me untold harm if I acquired a personal reputation as, say, a swindler or a child molester. Speech which attacks entire peoples, communities or other identifiable groups might seem roughly analogous to this and even more destructive. To step away from the racial context, if it were widely and genuinely believed that all lawyers are swindlers, and Lolita happens to be a lawyer, this logically entails that Lolita is a swindler.

If Lolita were believed to be a swindler solely on the ground that she is a lawyer, this might certainly demonstrate that the legal profession had fallen to an unprecedented level of disrepute. However, it would not have the same disastrous effect on her livelihood and life in society as if she were thought a swindler because of scurrilous anecdotes about her personally. In the former case, Lolita's professional rivals would gain no advantage. Nor would she be turfed from her legal career by horrified colleagues. All lawyers would have to put up with their poor reputation, work out some new public relations approaches, perhaps, and otherwise get on with it.

This example suggests that the damage done by group libels depends on context in a way that does not generally apply to the defamation of identifiable individuals. Generalised attacks on the reputation of a privileged group, such as lawyers, might have little impact at all. If a group that is already downtrodden and discriminated against is libelled, the situation might be more akin to the defamation of individuals, but even this does not necessarily follow.

Say it is alleged that people from a particular ethnic background have a propensity to commit terrorist acts, either because of some feature of their culture or because of some genetic trait or metaphysical characteristic in their "soul" or their "blood". This kind of claim can be damaging and offensive, but it will not necessarily have the same impact as if a particular individual were accused of being a terrorist. In the latter case, the individual might be shunned by people of her own ethnic background as well as by the larger society in which she lives. In the group case, it is likely that the affected individuals will at least receive the support of people from their own background, who have suffered the same generalised slur. Moreover, the slur will often seem implausible to the larger society. The claim that there is a cultural characteristic that turns all people from a particular culture into potential terrorists is difficult to credit, while claims based on genetics or wild theories about "blood" are simply bizarre.

Furthermore, there will be circumstances where particular cultures have, indeed, developed along lines that are open to criticism. For example, some cultures may be unusually sexist, racist, or inclined to glorify violence. Discussion of these issues may cause offence, but it is perfectly legitimate. Unless we embrace an irrational ethics of cultural relativism, we will see this kind of discussion as having value.

None of this is to deny that there may be contexts in which attacks on small and despised groups within a larger society will have disastrous consequences. This is especially so if what is being attacked is not some aspect of culture but the allegedly deep characteristics of populations who are stigmatised as fundamentally evil or inferior. Though I described as "bizarre" racial theories based on concepts of "blood", they have been taken all too seriously in the past. In some cases, existing resentment prejudice and discrimination may fuel, then be fueled by, irrational expressions of racial hatred, creating a downward spiral of malice and bigotry. In some specific contexts, it may be justifiable to suppress hate propaganda as a measure to combat discrimination against an especially vulnerable group. To concede all this, however, is not to concede that a robust attack on some racial, national or ethnic group should be treated as analogous to the defamation of an individual.

#

Any case for suppressing racial hate propaganda should be based on special considerations that are not analogous to ordinary principles of defamation law and that can be applied to recognisable categories of speech in limited circumstances. Here are some considerations: (1) mankind's appalling and near-universal experience of racism, which reached a crescendo of malevolence in the Nazi Holocaust; (2) the exceptionally deep and rationally justified offence caused by racial hate propaganda; (3) the obvious element of malice involved; and (4) the fact that hate propaganda is in conflict with important moral values relating to the preciousness of individuals, whatever their race.

The second point is alluded to in the decisions of anti-discrimination tribunals. For example, in Kazak v John Fairfax Publications (June 2000), the New South Wales Administrative Decisions Tribunal made much of a "right to a dignified and peaceful existence free from racist harassment and vilification". To some extent, this argument parallels more traditional attempts to protect public decency. Lecturing and writing in the 1960s, H.L.A. Hart suggested that it is legitimate for the law to forbid public displays likely to outrage sentiments that are very deep and widespread in the community. On that basis, Hart claimed to have identified an acceptable reason for prohibiting the public display of some acts whose moral acceptability (when performed in private) was disputed by no one, such as sex between a husband and wife. According to this analysis, what may be forbidden by statute is not the conduct itself but the public display.

Nearly four decades later, Hart's analysis seems quaint, especially his far-fetched example of bigamy as a form of conduct that is justifiably suppressed as a form of offensive public display. In a 1974 article, David Conway responded that all sorts of conduct or expression might be suppressed if any public display that merely offends somebody is a legitimate target. Conway referred to eating oysters or wearing pink and orange clothing. The partial retreat and partial accommodation of the 1960s sexual revolution make Conway's favourite example seem as quaint as Hart's: he asks why a nudist should be prevented from enjoying and expressing her freedom by shopping unclothed in a supermarket.

The shoppers might respond that they have a reasonable expectation that their peaceful enjoyment of ordinary public environments, such as supermarkets, will not be disturbed by such flamboyant efforts at self-actualisation. The nudist might then reply that she is not acting out of malice, while adding that her fellow shoppers' distress, if that is what they feel, is ultimately based on irrational ideas about sex and the human body that she is entitled to repudiate. There I leave the debate between Hart and Conway. In themselves, the proclivities of indiscreet bigamists or poorly socialised nudists have little relevance to Australia's current social problems.

Note, however, that Hart's argument becomes far more compelling if the public display does not express merely an individual's, perhaps eccentric or embarrassing, sense of style or taste, but something truly malevolent. Whatever might be said about the reactions to public nudity of Conway's peaceful shoppers, there is nothing irrational about experiencing deep feelings of distress, sickness, confusion, provocation and dread when confronted by racial bigotry's ugly nakedness. So much the worse if the bigotry is politically organised and theorised as the belief that a racial group to which you belong is, to borrow the language used in another case "inferior or evil by nature and . . . should be made to suffer for that reason": see Hobart Hebrew Congregation and Jeremy Jones v Olga Scully (Human Rights and Equal Opportunity Commission, September 2000).

People do have a legitimate expectation of going about their everyday activities without encountering material that rationally inspires such deep negative feelings. That expectation would be defeated by Nazi street marches in residential districts, the prominent display of anti-Semitic hate propaganda in streets, plazas, shops and markets, or the unsolicited stuffing of such material into letterboxes. In so far as US constitutional law has given protection to Nazi rallies in largely Jewish suburbs and to the use of Nazi or Klan symbols, this strikes me as unnecessarily protective of genuinely indecent public display.

#

As I've considered and researched this article, and discussed its ideas with others, I have been surprised at the frequently-expressed opinion that the state cannot be trusted with any power to suppress free speech. The view seems to be that, if the disreputable voices of neo-Nazis and Holocaust deniers can be suppressed today, those of evolutionary scientists or trade union officials might be suppressed tomorrow by some barbarous regime. In a sense, this distrust of the state may be exaggerated, since it would be far more difficult for a modern liberal democracy to suppress speech in defiance of international norms than in purported compliance with them. Nonetheless, I have become more hopeful that even educated, or "elite", opinion is suspicious of any governmental role in suppressing free speech.

A gut feeling that the state cannot be trusted with free speech issues has dominated policy in the US for several decades. Samuel Walker points out that US public policy in respect of hate propaganda moved in a different direction from that of other Western nations in the years after World War Two. He argues that this was because there was no powerful voice to counterbalance the uncompromising free speech advocacy of the American Civil Liberties Union. Civil rights activists supported unrestricted freedom of speech because of their own reliance upon emotive and aggressive speech for political mobilisation. To attack the constitutional edifice protecting offensive speech seemed counterproductive. This attitude was not challenged in the US until the emergence of the campus-based political correctness movement of the 1980s, which remains a powerful social presence but has suffered some decisive defeats in the courts over restrictive speech codes and draconian anti-pornography ordinances.

In Australia, there has been a widespread enactment of provisions in anti-discrimination statutes forbidding conduct or speech that is offensive, apparently in the weak sense of unwelcome or annoying, rather than in reference to deeply felt distress, provocation and similar emotions. "Racial vilification" laws have gone far beyond the prohibition of gross public displays of hatred or of malevolent racist messages comparable or analogous to those of the Nazis. The Kazak case shows how these laws can threaten legitimate criticism of particular cultures, societies and political leaderships.

In Kazak, a forcefully worded opinion piece by Tom Switzer, published in the Financial Review, was held to have breached a provision of the Anti-Discrimination Act 1977 (NSW) forbidding communications to the public inciting "hatred, serious contempt or ridicule" of a person or group of persons on the ground of their race. That provision, with its references to incitement and to "hatred, contempt or severe ridicule" might have been thought to prohibit only those utterances expressing a Nazi-like malevolence of purpose. However, the Switzer piece was nothing of the sort.

It comprised some 250 words of robust, opinionated criticism of "the Palestinians" for turning against the US when Iraq was bombed in 1998. Towards the end, there was an unfortunate statement that "it would appear that the Palestinians remain vicious thugs who show no serious willingness to comply with agreements". In context, however, Switzer's attack was on the Palestinian leadership, specifically Yasser Arafat, rather than on "the poor of Gaza", whom Arafat was condemned for supposedly neglecting in favour of "his military and bureaucratic elite".

Although this piece was intemperate, perhaps unfair, it is difficult to imagine the Financial Review's business-oriented readership interpreting it as a message that Palestinians are essentially subhuman and evil in the way claimed of Jews by the Nazis and their successors and sympathisers. Yet the tribunal rejected the argument that the article was directed at the leadership or administration of the Palestinian people rather than Palestinians as such. It noted that the phrase "the Palestinians" was used six times and asserted that no other synonymous or related term was used even once. By contrast, so the tribunal argued, there was one mention of "the Israelis" and one mention of "the Netanyahu Government". Actually, the article made specific mention of "Mr Yasser Arafat" and his alleged "military and bureaucratic elite", contrasted, as I have mentioned, with "the poor of Gaza". The tribunal concluded these half-baked literary critical efforts by stating, "Any incitement to hatred etc is directed towards the Palestinians as a race."

In this and other cases, anti-discrimination tribunals have given narrow interpretations to the statutory defences provided for speech of artistic, intellectual or political value. In the Scully case, which related to a woman involved in a public campaign of vicious anti-Semitic propaganda, the finding that she had acted unlawfully seems justifiable when the special considerations applying to racial hate propaganda are taken into account. Some of the material was received unwillingly in people's letterboxes. In another Human Rights and Equal Opportunity Commission case, Jeremy Jones v Fredrick Toben on behalf of the Adelaide Institute (October 2000), an order was made for removal of material from a notorious Holocaust-denial web site which used inflammatory anti-Semitic language. It is difficult to feel sympathy for the respondents in these cases. Unfortunately, the decisions went further than necessary and largely reasoned away the defences provided for more valuable kinds of speech.

In the Adelaide Institute case, the tribunal was required to apply federal legislation that included a number of defences, including that something was "said or done reasonably and in good faith ... in the course of any statement, publication, discussion or debate made or held for any genuine academic, artistic or scientific purpose or any other genuine purpose in the public interest". At first glance, this seems to provide ample protection of intellectual and political discussion, but that impression is misleading.

The tribunal could have held that speech uttered for a genuine academic, artistic or scientific purposes, or for analogous purposes that might be identified from time to time, is automatically considered to be in the public interest. On this reading, the point is to identify categories of protected speech. Such an approach to the statutory defence would protect artistically or intellectually valuable speech, not unlike the situation with US Supreme Court's First Amendment jurisprudence. However, the tribunal adopted a much narrower reading of the defence, according to which a particular instance of academic speech, for example, must not only have a genuine academic purpose but also be in the public interest. That is to say, the particular instance of speech must be proved by the speaker to be in the public interest. In practice, this might be almost impossible, once an isolated speech act has been identified as racially offensive.

To find against Toben and the Adelaide Institute, it was not necessary for such a narrow view to be taken of the defences in the federal Racial Discrimination Act since the web site concerned had no academic value in any event. However, more artistically or intellectually valuable material may be left without adequate protection in future cases. As I write, the Adelaide Institute has announced that it will defy the tribunal's orders, which suggests that the case will be reheard by the Federal Court. With respect to all concerned, I cannot imagine a greatly different outcome at Federal Court level. The material seems indefensible. However, I hope that the Court will now develop an approach to racial vilification legislation that is more protective of artistically or intellectually valuable speech.

#

There are many ways in which the law could be adapted to restrict hate propaganda without the need of racial vilification statutes. Advocacy of Nazi or similar messages of racial hate could be taken into account in general censorship and broadcasting standards. Nazis and white supremacists could be denied permits for demonstrations in some, perhaps many, locations. Public displays of racial hate material could be forbidden by the extension of ordinary laws aimed at nuisances and breaches of the peace. Truly violent or threatening acts could be punished more severely when they are motivated by racial hate.

This piecemeal approach would continue to impose significant burdens on the free speech of bigots and fanatics, and that is entirely justifiable. It is not obligatory for a liberal democracy to ensure a level playing field for every idea, however illiberal and anti-democratic, with political content. Yet a policy of imposing restrictions of a specific, contextual nature is desirable, since it would give greater respect to free speech as a basic individual liberty. There would still be ample means to deal with those categories of speech that are most deeply offensive, provocative of racial violence or likely to harm truly despised and vulnerable groups.

The fact is that broad, generally-worded racial vilification laws are now rooted like mountains in the Australian statutory landscape. By contrast, the possibility of constitutional change to enshrine free speech in a Charter of Rights and Freedoms, or similar document, is more remote than ever. In that context I propose that racial vilification laws of some kind be accepted and that we concentrate on confining them to their central purpose. If it is possible today for an organ of the state to target a legitimate, if opinionated, pieces on the Middle East conflict, we must wonder what can be made a target tomorrow. Evolutionary scientists and trade union officials may, indeed, be safe but what about a novel such as The Hand that Signed the Paper by Helen Darville (a.k.a. "Helen Demidenko")?

During the mid-1990s, as Darville's eccentricities were sniffed out by journalists and commentators with blood in their nostrils, it came to seem that her book might be motivated, in part, by anti-Semitism. Yet, despite its undoubted literary failings and historical misconceptions, The Hand that Signed the Paper is a work of some artistic value. It attempts to describe and understand the evil allure of Nazism but does not endorse what it describes. Such underlying anti-Semitism as it might express is light-years from the malevolent claims of Nazis and their sympathisers that Jews are evil and deserve to suffer or be murdered. Yet I can see no legal reason why, if its reasoning in Kazak is followed, the NSW tribunal could not make an order banning further publication or distribution in that state of The Hand that Signed the Paper. I hope it is clear that this outcome would be intolerable.

Since Australia has gone down the path of enacting specific racial vilification legislation, rather than using other laws to control racial hate propaganda in specific contexts, I put forward the following proposals to restrain the state's coercive power in this area. Some of them might require legislative amendments:

First, racial vilification statutes should be read narrowly so that they do not interfere, beyond their clear words, with the general principle of free speech. This means rejecting the view commonly expressed by courts and tribunals that human rights law is to be read expansively, even if it impinges on basic individual liberties.

Second, it should be put beyond doubt that these laws are not intended to censor publications of genuine artistic, intellectual or political value. This need not protect Holocaust denial material of a kind sufficiently vicious to attract the law's attention, since the Holocaust denial genre possesses no more value as serious history than does creation science as serious biology or geology.

Third, the courts and tribunals applying racial vilification laws should reach an understanding that the speech they are required to suppress is not just anything that might be considered offensive in the weak sense of annoying or unwelcome. Rather, the central target is the kind of hate message promulgated by neo-Nazis and Nazi sympathisers. The wider target is any other speech that is genuinely comparable or analogous to this. Courts and tribunals should be reluctant to find such comparability except in clearly deserved cases.

Fourth, the goal of racial vilification law should not be total suppression of ideas, however hateful, but the protection of sensibilities from displays and campaigns that might cause deep offence (in the sense of distress or provocation). It might not be necessary that a display take place in a public area such as a street or a shop if it is of a such a character as to come to people's attention through the mass media. It might even be enough if there were a sudden proliferation of racist web sites, or if these drew attention to themselves by widespread "spamming" of the internet. However, there must be some element of display to an unwilling public to justify the suppression of even deeply offensive speech.

In a difficult area such as this, compromise is needed to give effect to real but conflicting values. I have offered proposals in that spirit. They would compromise free speech as an absolute political principle, allowing the state considerable scope to attack deeply offensive racist messages. At the same time, they would lessen the state's current power to interfere with free speech values. If limitations such as this are rejected as too much of a muzzle on the state's jaws, we need to think harder about who, from time to time, is controlling the political agenda and what it means for basic individual liberties.

##

Russell Blackford's most recent contribution to Quadrant, "The Inward Journeys of Raimond Gaita", appeared in the September 2000 issue.