State of the Debate: Do Hate Speech Laws Work?
Discussions of hate speech laws often crop up in debates around free speech in America. In one sense, this is odd because America has no such laws, nor is it likely to have them, at least not any time soon. In another sense, discussion of hate speech law is rhetorically useful — they help an author make a broader point about free speech. Consider the following argument from an op-ed writer to the American Conservative:
Since the Great Recession, the default response to political crises on both sides of the Atlantic has been a feverish search for scapegoats and superficial fixes. Everything from social media to Russian meddling has been blamed for populist spasms against immigration, demographic change, and economic stagnation. These explanations are attractive because they absolve our leadership class of responsibility while suggesting easy fixes to thorny political issues. Block a few Russian bots and viral videos, so the argument goes, and we can get back to business as usual. The left-wing fixation on banning Donald Trump from Twitter is merely the latest in this desperate search for a shortcut back to normality. European hate speech laws are instructive because they expose the fallacy of this line of thinking. Restricting speech in the United States is unlikely to solve any of our political problems, just as speech restrictions in the United Kingdom and Germany have not magically erased European divisions over immigration or the EU. Before the United States abrogates the First Amendment, we should ask ourselves what hate speech laws would actually accomplish. The answer, if Europe is any indication, is “not much.”1
Here the author is using hate speech laws in Europe not merely to make the argument that such laws themselves are wrong, but also to make a much more core argument that censorship of any kind does not work. Such laws would only be “superficial fixes,” whether targeting hate speech, online misinformation, or political figures. Instead, “real” fixes are needed that address something other than speech. As to what constitutes a real fix, that is unclear.
Hate speech laws are not merely ineffective and “superficial” however; they are actively counterproductive. Take the following argument from the libertarian website Reason:
The most baffling thing about the people—mostly liberals—who push for laws against "hate speech" is their apparent inability to imagine these bans backfiring. In their zeal to punish those who spread sexist, racist, transphobic, or otherwise unfashionable speech, they too often ignore the ways tools of censorship—including hate speech laws—are used to suppress religious, social, sexual, and political minorities around the world... Before liberals rush to endorse laws that would give them even more leeway to punish speech, they should pause to ponder whether they might ever feel the need to criticize authority in "uncivil" terms themselves.2
Ultimately, the article argues, it is liberals and minorities who will suffer the most if hate speech laws are enacted, not those who use or endorse hate speech. Glenn Greenwald makes a very similar argument in an op-ed for The Intercept:"Many Americans who long for Europe’s hate speech restrictions assume that those laws are used to outlaw and punish expression of the bigoted ideas they most hate: racism, homophobia, Islamophobia, misogyny. Often, such laws are used that way. There are numerous cases in western Europe and Canada of far-right extremists being arrested, fined, or even jailed for publicly spouting that type of overt bigotry. But hate speech restrictions are used in those countries to suppress, outlaw, and punish more than far-right bigotry. Those laws have frequently been used to constrain and sanction a wide range of political views that many left-wing censorship advocates would never dream could be deemed “hateful,” and even against opinions which many of them likely share."3
Greenwald goes on to make another point that is rhetorically very common when discussing hate speech laws — they actually empower hateful groups. He writes:“In particular, the assumption that censorship will weaken hateful groups and make them go away is completely backward. Nothing strengthens hate groups more than censoring them, as it turns them into free speech martyrs, feeds their sense of grievance, and forces them to seek out more destructive means of activism.”
These three arguments are core to rhetorical opposition to hate speech laws in the United States: (1) that hate speech laws simply do not work; (2) that such laws are likely to be or often are used against marginalized groups and/or the left-wing; and (3) that the enforcement of such laws actually empowers speakers of hate speech by making them martyrs.
In an earlier post, I made a similar argument. I concluded that post, saying that, “it remains true, however, that the broad protections offered by a century of legal battles around the First Amendment have been useful to anti-racist activists. And the evidence indicates that the price of enduring racist speech is outweighed by this protection.”4 My intention in making this argument was twofold: first to challenge the idea that broad protections for free speech are incompatible with goals of social justice (a key theme I've returned to over and over again), and second to disentangle stronger argument in favor of broad free speech protections (such as those above) from weaker ones (which dismiss social justice goals altogether, argue that harms from hate speech are imagined rather than real, or rely on ad hominems attacking supporters of such laws). However, I'm not the kind to ever really be satisfied with an argument and I think it's worth reassessing my reasoning.
One issue that each of the arguments cited here has, mine included, is that they don't really address any specific argument made by someone in favor of hate speech laws. The article from the American Conservative begins by acknowledging opposing views taken by editorialists in the New York Times and Washington Post but the author doesn't really address their arguments. For one thing, the NYT articles doesn't actually make an explicit case for any hate speech law, but proposes other changes. The WaPo article, on the other hand, doesn't have a concrete proposal for a hate speech law, but says rather:
"Let the debate begin. Hate speech has a less violent, but nearly as damaging, impact in another way: It diminishes tolerance. It enables discrimination. Isn’t that, by definition, speech that undermines the values that the First Amendment was designed to protect: fairness, due process, equality before the law? Why shouldn’t the states experiment with their own version of hate speech statutes to penalize speech that deliberately insults people based on religion, race, ethnicity and sexual orientation?"
This is a bit more tepid of a claim. Certainly, his claim would be bolstered by evidence of the efficacy of hate speech laws, but the absence of such evidence (in an editorial with limited space) is hardly a catastrophe for his point. Moreover, both the NYT and WaPo articles make significant points that the response in the American Conservative doesn't address. What would the author say is the solution to hate speech online, as brought up in the NYT article? What response do they have to the critiques raised against the marketplace of ideas in the WaPo article?
The Reason article doesn't even make a pretense of seriously addressing arguments against it. It starts by accusing proponents of hate speech laws of lacking imagination, of not even considering the negative consequences of hate speech legislation. Of course, that the Reason article cites no supporters of hate speech laws, who have — as we shall see — absolutely considered that point, and that the article furthermore never address the potential or real positives of such legislation is somehow not a problem. The highly charged emotional language of the article also undermines it. The author uses words like “zeal,” “baffling,” “inability,” “rush,” and “punish,” to describe imagined (since no real opponent is actually mentioned or debated) opponents as foolish and inept, while also using euphemistic language, like “unfashionable” (rather than hateful) and “unpopular” (rather than hateful and false) to refer to the hate speech they oppose. It's a nice rhetorical trick, but mocking one's opponents, or in this case a strawman of one's opponents, is unlikely to persuade them of your point of view. It's also, ironically, not particularly reasonable.
Glenn Greenwald's article fares no better. At no point does it actually directly address any arguments made by supporters of hate speech laws. It does cursorily reference two op-eds — one a Medium post from someone with 1.6 thousand followers, the other an op-ed in the Los Angeles Times — but he never contends directly with the points they made. Instead, Greenwald contends with ideas like the following
“Many Americans who long for Europe’s hate speech restrictions assume that those laws are used to outlaw and punish expression of the bigoted ideas they most hate: racism, homophobia, Islamophobia, misogyny. Often, such laws are used that way...But hate speech restrictions are used in those countries to suppress, outlaw, and punish more than far-right bigotry.”
This implies, just as with the American Conservative and Reasonarticles, that supporters of hate speech laws have never contended with the idea that such laws might be used against marginalized groups, and indeed are making groundless assumptions rather than stating facts. The problem here is that, once again, no actual supporter of hate speech laws is cited as having the supposedly naïve view that hate speech laws never backfire. The two op-eds Greenwald referenced earlier make absolutely no mention of Europe whatsoever: the Medium post is focused on Twitter, and the LA Times article focuses on the harms caused by hate speech in America. These actual points go unaddressed. At no point does Greenwald actually contend with the harms caused by hate speech that the LA Times article brings up.
I think it is worth actually contending with the arguments made by supporters of hate speech laws. America is alone among modern, western democracies in not having such laws. Not only Europe, but Australia, New Zealand, and all of central and south America have some form of hate speech laws.5 It is possible that everyone else is wrong, and that only America has the right approach, i.e. no hate speech laws whatsoever. After all, the popularity of an idea is not necessarily proportional to its truth (which cuts against the marketplace of ideas, by the way). Still, one cannot reasonably assume that having no hate speech laws is the only obviously correct approach if such an approach is only obvious to one country, and then not even obvious to all of the people in it.
The three articles I cited do provide some evidence for the potential problems and pitfalls of hate speech laws — evidence we will consider here — but yet, despite these potential problems, countries have maintained, or even strengthened their hate speech laws over time. This fact should strike us as very odd, indeed. If hate speech laws don't work, if they lead to more and more censorship, and if they just empower users of hate speech, then why do so many countries have such laws? We could chalk it up to mass psychosis, or conspiracy theories about liberal elites, but I think you'll agree that these arguments would be rather weak or purely ad hominem. Perhaps we could chalk it up to America “leading the way” on this issue, in the same way that, say, the Netherlands or Portugal have on drug liberalization, or New Zealand did on recognizing women's suffrage. On this hypothesis, perhaps America has truly got the right answer on this policy question and everyone else will ultimately catch up to us as they come to learn the error of their ways. It is also possible, and worth considering, that America might just be wrong. In either case, supporters of broad free speech protection in the public sphere (as opposed to academic sphere which is much better about this), would be better off actually addressing the real concerns and arguments that proponents of hate speech laws make, and taking a calmer and more reasonable approach to both the costs and the benefits of such laws.
So, let's do that.
We will dive into the three core arguments we saw before: (1) that hate speech laws simply do not work; (2) that such laws are likely to be or often are used against marginalized groups and/or the left-wing; and (3) that the enforcement of such laws actually empowers speakers of hate speech by making them martyrs.
Hate Speech Laws Don't Work
There is a lot to consider here, but the main question I'm interested in is: What would it mean for hate speech laws to work? To answer this, we will look at some common reasons presented for why hate speech laws don't work, and we can infer from that what it would mean for them to work.No one can Define Hate Speech
One of the main arguments against hate speech laws is that hate speech is difficult or impossible to define. What is offensive is subjective, the argument goes, so there can be no clear definition of hate speech and therefore it would not be clear what a law that punishes hate speech would actually be punishing. In other words, hate speech laws can't work, because no one can agree on what hate speech actually is. There is no doubt that hate speech is difficult to define. Even the United Nations, in their Strategy and Plan of Action on Hate Speech say as much. The authors write, "there is no international legal definition of hate speech, and the characterization of what is 'hateful' is controversial and disputed."6 Yet, at the same time, they provide a working definition that emphasizes communicated attacks on a person or group based on identity characteristics. The definition isn't perfect, no doubt, but it strikes me as an exaggeration to say that such a definition is just as useless as having no clear, agreed upon definition at all. It should also be kept in mind that our legal system engages in the consideration of difficult and hard to define concepts on a daily basis. Questions such as: does this statute violate separation of church and states? How should we interpret the separation of powers between the branches? What constitutes incitement? and so on are questions that certainly do not have a completely agreed upon answer. Even the basic terms like "incitement" or "separation" are debated hotly. If legislatures or courts could not deal with concepts that are hard to define, they could not deal with very much at all. It does not seem to me that defining hate speech, as hard as it might be to do that, is particularly more difficult to define, or more subjective than, any other significant legal concept we already contend with. It seems, then, that hate speech can be defined at least as well as other difficult concepts. This makes it difficult to shoot down hate speech laws on these grounds.Hate Speech Laws Don't work because Hate Speech is Just Words.
An additional consideration here, is the idea that hate speech laws can't work because they target hate speech, and hate speech is ultimately harmless -- words are just words after all. It would be worth considering this, then. First, let's provide some context. Does anyone actually think that hate speech is a problem (beyond, of course, nefarious social justice warriors)? Indeed, there are people that do. In fact, a study published in Advances in Political Psychology summarized research on hate speech as follows: "Hate Speech has severe consequences for human intergroup relations and should be considered a large-scale societal issue that deteriorates living quality, increases aggression, and affects mental health and well-being of minorities. At the same time, hate speech is common in political discourse and its presence in social media has increased."7 Hate speech deteriorated the well-being of minorities, resulting, in one study in higher suicide rates among those targeted in addition to higher levels of depression and PTSD. Exposure to hate speech increases desensitization to further hate speech as well. At the same time, hate speech "can lead to outgroup avoidance and discrimination" I'll defer to the research here, which clearly shows that hate speech has a significant negative impact, that it is more than just words.Hate Speech laws will Backfire and Be Used against Minorities they are meant to protect
One study in hate speech in Australia makes no mention of any reports or cases where a minority member was targeted by a white Australian alleging “reverse racism” or something to that effect.8
In another sense, the question about who is targeted by hate speech laws makes an unwarranted assumption, an assumption that was not clear to me, and I imagine not clear to other Americans. it assumes that hate speech laws can only be criminal statutes whose enforcement from beginning to end is entirely done by the government, perhaps through a public prosecutor's office. That's actually not how many hate speech laws work. Many laws are civil rather than criminal, meaning that they may involve compensation paid to victims but not jail time for the offender, and many laws only initiate proceedings at the behest of harmed parties. That is to say, in many cases the government doesn't decide who is on trial, but the victim. This is all to say that not all hate speech laws are equal in how much power, and how much discretion to use that power, they give to governments.
Hate Speech Laws will Just Make Martyrs out of those Being Censored
That same study from Australia found no such martyrdom effect. The authors highlight the example of Andrew Bolt, who shot to national prominence on the basis of his punishment for racist hate speech in an article he wrote for the Herald and Weekly Times. However, this was the only case, in over twenty years, that the authors found could constitute a martyrdom effect for those punished for violating hate speech laws. They conclude that no such martyrdom effect was detectable, in part because the vast majority — 98% — of cases were resolved behind closed doors.
A study on hate speech laws in Latin America found some examples of the martyrdom effect, but largely this occurred when such laws were on the books but not rigorously enforced, leading to a sort of show trial. 9 A study looking at anti-immigration parties in Europe did find that there was sometimes a small positive effect for those parties after leaders were prosecuted for hate speech, although the effect was not strong, only occurred when there was intensive media coverage, and seemed to mobilize already anti-immigrant voters to vote rather than actually persuade anyone else to vote for anti-immigrant parties.10
What the evidence suggests then is not that hate speech laws and persecutions necessarily, inevitably, or even mostly lead to masses of right-wing martyrs using the trial to become even more powerful. Instead, it suggests that widespread media attention, hate speech prosecutions or not, is a more important factor.
That shouldn't come as much surprise. Plenty of groups experience widespread censorship without a martyrdom effect, nor do people concerned about such an effect talk about it in terms of the groups most likely to suffer under censorship — marginalized groups. When Black Lives Matter protestors are threatened by the police, when state governments pass laws making it harder to protest environmental or racial injustice, or when states prevent teachers from telling kids that is perfectly fine to be queer, no one brings up the martyrdom effect — at least not that I've been able to find. Why? Because the groups most likely to be targeted by such censorship are also those groups least able to fight back (though they can and do) because they have the least access to mainstream media.
Conclusion
Hate speech laws are hardly the boogeyman that free speech (near-)absolutists make them out to be. Research has shown that concerns around their misuse and potential to backfire are not completely groundless, but also that, in practice, hate speech laws do not tend to actually chill speech, be used to target already marginalized groups, or produce right-wing martyrs (absent widespread media attention). They can also provide some benefits such as making protected groups feel safer and more resilient, providing pathways for groups to redress grievances legally, and reducing the use of vitriolic language in public discourse. However, as a strategy to reduce prejudice, they don't seem particularly effective and can introduce additional complications such as requiring victims to carry the burden of pushing their own cases.So where does that leave us?
Assessing the real (rather than imagined or hypothesized) costs of hate speech laws, and the real (rather than imagined or hypothesized) benefits, it seems to me that the potential benefits outweigh the potential costs. I can understand why so many countries around the world have adopted them. However, they still seem to me like the wrong strategy because they aren't particularly effective at doing the one thing that really matters — reducing actual prejudice and instances of prejudicial hate speech. The same legal resources, activist energy, and so on that could be spent dealing with hate speech laws would be better spent developing other strategies. We already have far more effective strategies for dealing with prejudice than hate speech laws, strategies that would matter more and face less resistance than passing hate speech laws would.
So to answer the question: Do hate speech laws work? yes, but only a little, and far less than other strategies we already have at our disposal.
https://reason.com/2015/12/08/how-hate-speech-laws-actually/
https://write.as/langueverte/state-of-the-debate-should-free-speech-protect-racist-hate-speech
Hate Speech Epidemic. The Dynamic Effects of Derogatory Language on Intergroup Relations and Political Radicalization from Advances in Political Science
I appear to have lost the article title for this link. I'll try and find it again later.
Hate Speech and the Language of Racism in Latin America: A Lens for Reconsidering Global Hate Speech Restrictions and Legislation Models
Prosecuted, yet popular? Hate speech prosecution of anti‑immigration politicians in the news and electoral support