Jacob Mchangama’s essay is both thoughtful and thought-provoking. The first few paragraphs give us a sort of historical overview of free speech in the world since the days of Athenian democracy and the Roman republic. I am not sure how much guidance we can draw from this or how much we should want to. But we should certainly be interested in the present-day study whose results Mchangama reports in the body of his essay—a study he conducted with the help of political scientist Rasmus Fonnesbaek Andersen. As I understand it, the study shows that there is a negative relationship between free speech protections and social conflict: the more free speech there is in a society, the less conflict there is likely to be.
If this is right, then restrictions on hate speech cannot be justified in the way they are usually intended to be justified. Countries like the United Kingdom enacted racial hatred legislation in the mid-1960s because they feared the growth of intercommunal hostility between native Britons and new immigrant groups. They worried about their social environment being polluted by deliberate attempts to stir up racial hatred. Countries like Germany enacted hate speech legislation because of their experience in the 1930s and 40s of extreme hostility of ordinary Germans towards Jews especially and other minorities, and they did not want to countenance even the possibility of the resurgence of this sort of hatred.
But Mchangama and Andersen say they have found that in fact untrammeled hate speech does not exacerbate conflict of this kind. Defenders of hate speech regulation (like me) ought to sit up and take notice. For the study seems to suggest that the democracies of the world would be better off—or at least no worse off—so far as conflict-diminution is concerned if they were to dismantle the hate speech restrictions many of them have had in place for decades.
Full estimation of the value and implications of this study must await disclosure of the parameters of the investigators’ research and, in particular, the exact questions and comparisons they essayed. Here are some questions we should want to have answered.
First, what democracies were studied? Among advanced liberal democracies, only the United States does not enjoy the benefit, such as it is, of legal restrictions on hate speech. Was it just a case of comparing the United States with the rest—with Australia, Canada, France, Germany, Israel, New Zealand, Norway, Sweden, the United Kingdom, and so on? Hate speech regulations in these countries vary in their provisions, but in all of them it is an offense to disseminate material calculated to stir up racial, ethnic, and in some cases religious conflict. Of course, all these countries would say also that despite these laws they are still very strongly committed to the principle of free speech. They think they can reconcile hate speech laws with the protection of free speech just as the United States thinks it can reconcile child pornography laws or defamation laws with the requirements of the First Amendment. It would be good to know how the countries mentioned above (with the exception of the United States) are coded in Mchangama and Andersen’s study of the effects of free speech protections. They believe in free speech, but, like everyone else, they too believe this freedom has certain limits.
The question also needs to be answered, because the essay presents the study as sometimes being about political freedom (in general) and sometimes about free speech in particular. Many of the countries I have mentioned will say that they are at least as free, politically, as the United States; some would say emphatically that they are much more so. So what is it that is supposed to prevent social conflict: political freedom in general or the absence of hate speech laws in particular? And what happens when these two factors pull in opposite directions—when a country such as, let’s say New Zealand, is politically much freer than the united States even though it limits free speech with modest restrictions on the dissemination of racial hatred?
Secondly—and assuming we are talking specifically about hate speech restrictions—how sensitive is the Mchangama and Andersen study to the precise drafting of hate speech laws? Many such laws provide safe havens for the free expression of views which if aired publicly in an abusive or threatening manner would attract a legal prohibition. Also, recent legislation in the UK on religious hate speech included an explicit free speech exception:
Section 29J (of the Public Order Act): Protection of freedom of expression. Nothing in this Part shall be read or given effect in a way which prohibits or restricts discussion, criticism or expressions of antipathy, dislike, ridicule, insult or abuse of particular religions or the beliefs or practices of their adherents, or of any other belief system or the beliefs or practices of its adherents, or proselytising or urging adherents of a different religion or belief system to cease practising their religion or belief system.
The British legislation prohibits the stirring up of hate against adherents of a given religion but it doesn’t prohibit criticism of the religion itself. Should this be coded, for the purposes of the Mchangama/Andersen study, as an instance of legislation protecting freedom of speech or as legislation limiting freedom of speech?
Thirdly, exactly what comparisons were made? The one that I have heard most commonly suggested is that racial antipathy is no worse in countries without hate speech laws than it is in countries that have them. But that is not the relevant comparison. Countries start from different baselines of racial antipathy. The United States for example, with its legacy of slavery, segregation, and racial terror has always had more of a race problem than (say) New Zealand. Does untrammeled free speech in the United States make American racial conflict worse or better; or does it have no effect at all? For other democracies, have hate speech laws improved matters within the countries in which they have been enacted and enforced? These are enormously difficult questions to answer because, looking back, they require us to entertain counterfactuals. We have to ask how much racial conflict there would have been, in the UK for example, if hate speech laws had not been enacted. And how are we to determine that? I suspect Mchangama and Andersen did not even attempt to answer such questions. How could they? Yet these are exactly the questions which, ex ante, it was the responsibility of the relevant governments to answer at the time they enacted these laws. They had to ask themselves up front what would be the effects of this legislation or what would be the effects of failing to enact it.
Fourthly—and this is a question whose importance Mchangama acknowledges in his essay—how far are the effects of hate speech restrictions undermined in (say) Canada, the European Union, and the Antipodes by the external effects of American free speech laws? Free speech in America is heard in Europe or Australia through the Internet, and requirements that (say) German social media take down hate speech postings originating in the United States can only partly address the problem. I acknowledge that it is unclear how this consideration should weigh in the discussion. On the one hand, it seems to load the dice against the effectiveness of hate speech laws. On the other hand, perhaps we should be more cautious about enacting such laws knowing that they will have to operate in this environment.
Finally, to what extent should a study like this concentrate on legislation alone, as opposed to legislation and the social sanctions that limit free speech in a given society? The United States does not countenance explicit hate speech laws, but recent reactions to Roseanne Barr’s racist outburst show that social sanctions on the matter are alive and well; her highly successful television show was cancelled by ABC within 24 hours of her racist tweet and she is now virtually unemployable in American television. Similar de facto restrictions on hate speech operate pervasively in numerous corporate and educational contexts. There is no doubt that this has a chilling effect on free speech. In fact, John Stuart Mill, an author quoted by Mchangama at the beginning of his essay, believed that such social restrictions on speech were much more important than legal restrictions. In Chapter Two of On Liberty, he observed that “[f]or a long time past, the chief mischief of the legal penalties is that they strengthen the social stigma.” It is that social stigma, he said, “which is really effective.” So, again, we have to ask: for the purposes of a study like the Mchangama/Andersen study, do we look only at laws or do we look also at the overall coerciveness or freedom of the social atmosphere, in figuring out how restricted or unrestricted speech affects social conflict?
I ask these questions not in any destructive (or defensive) spirit, but because I believe that the phenomena addressed in these studies are subtle and complex. We do ourselves no favors by jumping to broad-brush conclusions in the light of poorly designed studies. Now I don’t think for a minute that the Mchangama/ Andersen study falls into this category. But it would have been good to have had much more detail on all of these points, even at the cost of a little less at the beginning about Demosthenes and Marc Antony.