Thursday, April 3, 2014

Racial vilification, free speech, 18C and John Stuart Mill’s On Liberty

What could a work, written a century and a half ago on the opposite side of the globe, discussing neither racism, vilification or human rights, hope to contribute to Australia’s debate about changes to its Racial Discrimination Act?
Quite a lot, I think.

Friends of free speech perennially invoke Mill’s most famous work, On Liberty, because its arguments call into question the wisdom of allowing feelings of offence to curtail vigorous debate, and because they invite us to reflect upon what happens to opinions and individuals silenced by the state.

Our shared temptation to silence the wrong and immoral

Mill cautions us about using our feelings of moral offence and outrage as reasons to stifle other’s thought and action.

“The disposition of mankind, whether as rulers or as fellow citizens, to impose their own opinions and inclinations as a rule of conduct on others is so energetically supported by some of the best and by some of the worst feelings incident to human nature that it is hardly ever kept under restraint by anything but want of power...”

We are all of us tempted, Mill observed, to wield the power of the law to ensure others acquiesce to our desires about what they can do and say. Allowing feelings of offence to trigger legal prohibitions amounts to doing just this. But a pluralistic society, populated by diverse individuals, houses many different perspectives, all with different views on what must be held sacred. Such differences inevitably result in feelings of offence when others fail to respect what we hold dear. As a result, offence-giving is a natural part of a genuinely pluralist society. To suppress offence amounts to suppressing diversity.  

In his defence of diversity and individuality, Mill argued a pluralist society would need to draw a strong distinction between moral condemnation and legal prohibitions. Responses of moral condemnation and outrage must be protected—these follow from listener’s own liberties of thought and speech. I am free to be appalled by what you say—and to voice my shock. But all sides in a diverse society need to resist calling on the law to coercively prohibit immoral actions, and immoral speech in particular, lest they stifle others’ diversity, feed their own intolerance and their belief that their certainty counts as absolute certainty, and—in the final analysis—license the use of a weapon they might eventually have turned back upon themselves.

Silencing an opinion does not make it disappear

Mills’ signature contribution to free speech debates lies in asking us to seriously consider what happens to the opinion and the person silenced by the state. Consider a speaker who wants to express an opinion about some matter of public policy or interest, but is silenced by the force of the law. Let’s call our speaker here, ‘Chitra’ (the name might help remind us that the law, in its infinite majesty, does not only silence (powerful, influential, popular) members of the majority, but all speakers, including those who may be disadvantaged, marginalized or vulnerable in their own ways).First and foremost, Mill stressed, we need to remember that silencing Chitra’s opinion does not make her opinion, or any of the reasons she had for believing it, disappear. The state merely uses the force of law to stop Chitra publicly giving voice to what she is already thinking.

Two possibilities result. First, and most likely, Chitra holds firm to her opinion. After all, she has not been presented with any reasons to change her mind. Chitra has not heard the arguments that could be voiced in response to it—arguments coming from those who have different and perhaps more informed perspectives on the matter, and perhaps even arguments coming from those who Chitra respects. As such, her opinion remains, and it festers. Chitra whispers it only to confidants and perhaps even bears it with a secret pride as a badge of her free-thinking ways. She cannot voice her unlawful view publically, but its presence can still leak out in other ways—such as when she enters the privacy of the ballot box.

Alternatively, Chitra might change her mind. But she does not change her mind because she has been presented with reasons bearing directly on the reasonableness of her opinion. To the contrary, Chitra cannot be confronted with such reasons until she is allowed to voice the opinion, as strongly and doggedly as she can, and lay down her specific reasons for holding it. Instead, she changes her mind purely on the basis of the state’s authoritative say-so. She acquiesces that her thoughts are so repugnant society cannot abide their being spoken—though she cannot apprehend why this is so, or where the error in her thinking lies. Mill asks whether we really desire this result—whether we wanted to stifle Chitra’s integrity of character and the confidence to follow her own train of thought. For Chitra’s mind has not been persuaded. It has been pacified, tamed, subdued.

Mill also asks us to think about Chitra’s opinion itself, and whether society in general wouldn’t be better served having such opinions aired, rather than suppressed. Even if Chitra cannot be convinced of her errors, third parties witnessing her spirited defence of the opinion, and others’ spirited attacks on it, can judge for themselves, satisfied they have heard both sides of the story. Indeed, even those welded to the view that Chitra attacks can benefit from being confronted with her arguments. “He who knows only his own side of the case,” Mill argued, “knows little of that.” Chitra’s opponents learn more about their own view, and must search out and reflect on the most important reasons for their holding it, as they publicly refute her stance. Whether Chitra’s deviant view proves correct, partly correct, mostly false or even demonstrably false, Mill contends, society gains by hearing it and appreciating the responses marshalled against it.

Australia in the 21st Century

All that said, you might yet wonder whether any of this really bears on Australia’s debate on Sections 18C and 18D of its Racial Discrimination Act? After all, Mill focuses on the presentation of genuinely held opinions, not acts of direct vilification and intentional harm. Indeed, we might wonder how Mill himself would respond if we stressed that many minorities support the existing law. Far from entrenching his feared ‘tyranny of the majority’, minorities largely welcome the law and its protections. And we might wonder, too, if Mill would reconsider if we impressed upon him the genocidal horrors wrought by racism throughout the Twentieth Century, and the legacy of racist oppression still infecting our societies today.

I suspect Mill would be sympathetic to these concerns, but remain cautious about the extent to which we should see them reflected in law. (I suspect the author of The Subjection of Women would also wonder why sexual discrimination was not equally prohibited.) With Mill’s arguments in mind, several specific councils to the contemporary debate about Sections 18C and 18D of the Act rear their heads.

First, despite all the harms of racism noted above, we still have reason to be careful about opening this door and incorporating subjective feelings of offence into the laws’ operation. (18C at present does this by requiring an objective test about subjective feelings: it asks whether the act is reasonably likely to offend another person or group.) Human nature tempts every one of us to use the law to make others behave as we morally want, with unavoidable risks to freedom and diversity. We must bear in mind that genuine tolerance requires more than respecting difference: it requires respecting differences we hate—differences we think we should hate.

Second, we must be honest about the free speech costs of such laws. Let’s grant that in the starkest cases of vilification (such as all-too-commonly witnessed at football grounds and on public transport), where the speaker directly addresses his victim, he does not intend to assert an opinion, and aims only at immediately harming another person. Stifling such speech carries few free-speech costs in Mill’s terms. But many other opinions can be aired in political and social discussions that bitingly offend certain groups, and yet are genuinely believed and voiced as opinions worthy of consideration. If we opt for suppression, then we must not pretend there are no free-speech issues at stake.

Third, if we must limit speech, let’s do it as surgically as possible, and ensure the general citizenry know exactly where the specific limitations lie. Consider two potential limitations on speech. A first limitation would prohibit only speech aimed to offend—that is, speech deliberately addressed to, and plainly aiming to denigrate, the victim. These are arguably the worst and paradigm cases of vilification, and in these situations the speaker aims to wound, not to offer an opinion.

A second limitation on speech, less surgically but still topically, would prohibit an objective type of offence-giving, rather than a subjective measure of whether a group feels offended. For example, we could protect people from hate speech by prohibiting any speech that asserts the mental or moral inferiority of those of another race (or nationality, ethnicity) or claims such people should be hated or scorned. Such a prohibition would protect people from racist speech, but not from race-related offence.

Now strictly speaking, Mill’s worries about silencing opinions apply even to cases where we are absolutely certain about the wrongness of the opinion we aim to silence. But categorically racist claims impact so heavily on people still suffering from current and historical racism, that even a modern-day Millean might accede to their prohibition. After all, and despite Mill’s falliablism, the overwhelming majority of us really are absolutely certain that claims of racial superiority are utterly false and beyond the pale. It’s not that such claims put forward a perverse moral view—it’s that it’s hard to see how the view counts moral as all. It fails what is sometimes called the ‘basic moral perspective’, which involves taking a position of impartiality regarding oneself and others. Can one really be said to be contributing to public discourse if one rejects the basic moral perspective that furnishes the starting point and ground of public discourse itself?

However: we possess no similar time-tried consensus about other claims regarding race—such as who should or should not belong in a particular category, for example, or whether we should employ these categories at all (questions lying at the heart of the Eatock v. Bolt case that precipitated the change in legislation). One does not have to be beyond the moral pale to wonder about the purposes of these categories and their proper extension, especially when state policies employ these categories to distribute resources and opportunities. No doubt, people will be outraged when others argue the merits of whether they belong in one category or another. Such discussions cut to the very quick of people’s identities and self-understanding. The feelings of offence they experience are entirely legitimate. But the presence of such offence cannot budge the fact that different people possess different opinions about such categories and their proper boundaries and purposes. Silencing debate on these questions ultimately serves nobody’s purpose. (Consider the debate on gay marriage—should the current community of married persons wield absolute control over the definition of ‘married’ and who counts as ‘married’? Surely not. The debate on gay marriage impacts upon people’s identities and the social understanding of their relationships—but that is no reason to stifle the debate itself.)

What’s the upshot of all this? Namely, that legitimate debate needs to be protected even when we can predict that it will cause race-related offence (even understandable and reasonable offence). We should resist allowing subjective feelings of race-related offence to prohibit such debate.


Different sorts of offensive, racist and race-related speech
What does this mean in practical terms? Several different sorts of speech acts concern us here.

Case 1. Direct vilification: Direct vilification occurs where, a) the speaker effectively asserts that one race, colour, nation etc is by its very nature worthy of hatred, scorn or paternalistic treatment; b) there is no evidence the speaker aims to inject an opinion into a political or social debate, rather than to merely wound another; and c) the speaker actively ensures the delivery of the message to its intended target (by email, or as an online comment directed to them, e.g., or by actual physical presence). News reports regularly draw public attention to direct vilification occurring on public transport and at football grounds.

Case 2. Hate speech: Hate speech happens when the speaker effectively asserts that one race, colour, nation etc is by its very nature worthy of hatred, scorn or paternalistic treatment. But such hate speech does not amount to direct vilification because it occurs as part of a genuine debate or political standpoint (for example at a rally, march or in a publication or blog). Consider the KKK or neo-Nazis having a public rally.

Case 3. Offence taken on racial grounds (but not by reason of racism itself): In this case someone says something that offends reasonable members of a particular racial group. However, the statement does not assert (nor even imply) that the group is inherently morally inferior in any way—the statement gives offence for other reasons.

Case 3.a: Calling into question whether a person counts as a ‘genuine’ Aborigine (or ‘Australian’, or ‘black’) provides an instance of Case 3. This is the category that snared Andrew Bolt in Eatock v. Bolt. He accused certain pale-skinned people of categorizing themselves as Aboriginal in order to enhance their employment prospects. Such an accusation was bound to be offensive to the people targeted—and would doubtless have remained offensive even if it had not been possible (as it was in the event) to prove the falsity of Bolt’s specific accusations through examination of those people’s histories. But Bolt’s claims were not offensive because they constituted or implied racism or hate speech. His accusation is quite intelligible even if all racial groups are completely equal in moral status and respect. As such, Bolt’s accusation does not present as an instance of Case 2, but of Case 3: speech giving rise to reasonable, race-related offence.

Case 3.b: Holocaust denial provides another instance of Case 3. Denying the occurrence of a well-documented historical event of profound significance to a given people, ethnicity or race will doubtless provoke offence and outrage by that group. In my experience at least, Holocaust-denial seem almost always to be driven by genuinely racist or anti-Semitic beliefs. But the denial itself need involve no assertion of racial inferiority or hatred, so legislation prohibiting objective racism and hate speech (i.e. legislation targeting Case 2) will not prohibit it.

The Millean arguments supplied above provide little reason, so far as I can see, for resisting prohibitions on Direct Vilification (Case 1). Direct vilification adds nothing to public debate, and intrinsically aims to wound its victims. 

The Millean arguments do provide reason for being cautious about prohibiting Hate Speech (Case 2), as people can genuinely offer these views as inputs into the public discourse, and suppressing their views will hardly make such opinions disappear. However, the harm caused by such speech and its denial of the basic moral perspective arguably outweighs these countervailing considerations.

However, I think someone persuaded by Mill’s arguments must fundamentally reject any attempts to prohibit race-related offence as it appears in Case 3.a. Prohibiting such offence amounts to suppressing reasonable pluralism and the public contestation of standing political policies. Combatting offensive views here must be done through opposing speeches, not legal suppression.

I confess I remain unsure about cases like Case 3.b. Arguably, a legislative act could require certain minimal levels of good faith for any acts causing race-related offence, where ‘good faith’ implies, (a) that the speaker genuinely holds the views espoused; and (b) that their views accord with minimal standards for evidence and fact-checking, in accordance with community standards for accuracy in the context in which they are spoken. (Note that this was not the way ‘good faith’ was interpreted by the judge in Eatock v. Bolt.) The thought here is that complete detachment from evidence and logic may be countenanced in flippant comments online, for example, but that the community requires increasingly higher standards of accuracy in professional blogs, public speeches, opinion-pieces and journalistic reports. Since blithely denying the Holocaust amounts to complete disregard for well-known and well-documented evidence, such denial would be prohibited in these latter cases. While I think Mill would resist even these prohibitions (and I worry about them too), at least it can be said in their defence that the process of a finding of guilt would centre on the available and objective evidence for the falsity of the speaker’s view. Hence the legal process holds at least some hope for convincing the person of their error, rather than merely taming their mind into submission to the orthodox view.

Draft Amendments to the Racial Discrimination Act 

The Attorney-General’s Department released draft amendments to the Act on the 25th March 2014. As I read them, in the context of the above discussion, they would (rightly) prohibit Case 1: Direct vilification. Worryingly, though, the draft section would only prohibit direct vilification that incites hatred or causes fear. The section will not explicitly prohibit directly castigating someone as being morally or mentally inferior on the basis of their race.

More generally, because the section will not apply to public discussion, it seems to me it will allow not only both instances of Case 3, but also instances of Case 2: of flagrantly racist hate speech that is put forward in context of public discussion. The draft amendments thus take a very strong stance on free speech (stronger, indeed, than I have set out above).

I think even friends of free speech should worry that this amendment swings the pendulum too far in favor of free speech. My reading (and it is just my reading) of the broad position of the majority of Australians on the matter is that they think Case 1 Direct Vilification and Case 2: Hate speech must be prohibited. Most of them, I think, also harbor sympathy for prohibiting Case 3a: Holocaust denial.


If the amendment were legislated as drafted, I think widespread popular sentiment would mount against it, and whenever the opposition Labour Party next get into power, we can almost guarantee it would be repealed. Its replacement would probably be the restatement of the original act, which prohibits 3a, giving race-related offence. If what I have said above is on the right track, then the friend of Millean free speech will be concerned above all to protect speech of Case 3a. The best way to secure this protection in the long term seems to me to be to draft legislation that protects Case 3a, prohibits Case 1 and Case 2, and allows a nuanced position on Case 3b. In other words, direct vilification and hate speech must be prohibited, while speech giving subjective race-related offence must be protected, perhaps with a proviso that such speech lives up to certain minimal standards of fact-checking, appropriate to the context in which it is uttered.