As you are aware, I am a near-absolutist when it comes to free speech. I believe both that (a) people should be able to speak their minds, and (b) any censor will ultimately act to promote their own viewpoint and silence their opposition. More to the point, the censor will always eventually be someone who protects the interests of the powerful, ultimately censoring the truth if truth runs contrary to those interests.
Historically, liberals and the left have been the advocates for and guardians of free speech and expression. This has been in doubt for a while, and in the past couple of years, so-called “liberals”1 and elements of the left have turned against free speech. Matt Taibbi recently wrote an essay on this here. I have written a bit about this development here, here, here, and here. Glenn Greenwald has some words about it here.
The people on the left who have turned against free speech for the most part are concerned about the category they call ‘hate speech’. “Hate speech” is not, for this kind of leftist, a category of speech that should be covered by free speech. I have already touched on this point in my essay about the Meriwether case, but I want to say a bit more on the subject.
What Constitutes Hate Speech?
In asking this question, we hit a genuine problem: there is no real consensus of what constitutes hate speech. The Stanford Encyclopedia of Philosophy provides a good primer on the dispute over how to define ‘hate speech’. (Wikipedia has the shorter version.) A brief skim of the Stanford Encyclopedia entry shows that there is no clear agreement even for the basis of the category of hate speech. Does speech become hate speech because its content incites hatred for a group (the content theory)? Or does speech become hate speech when it undermines the social standing of the target group (the dignity theory)? Still another school of thought is that speech becomes hate speech when it is used to undermine a group’s social standing (the performance theory)?
These are important but subtle distinctions. The Stanford Encyclopedia gives a good example. Take the proposition “Only citizens of the United States should live in the United States.” Under both the content and the dignity theories, the proposition will not be hate speech. The phrase does not incite hatred for any particular group, rather it is about who (citizens) should be permitted to do something (reside in the United States). Thus, it is not hate speech under the content theory. Nor does the proposition on its face undermine the dignity or social standing of any particular group.2 However, under the performative theory, this proposition could be used to denigrate groups even though it is facially neutral. For example, the speaker might deploy this proposition when talking about Afghani, Hispanic, African, or Haitian refugees, but not use it when talking about Ukrainian refugees.3 In which case, what counts is that it is being used as hate speech even though facially neutral.
Now consider your local skinhead shouting the n-word in the face of a Black person. Is this hate speech? Under the dignitary and performative theories it would be. Brazenly yelling a racial epithet in someone’s face is an assault on their social standing both facially and in use. But under the content theory this is not clear cut. The skinhead in our example is certainly engaging in racist speech, but is he inciting hatred against Black people or is he just making a racist ass of himself? If your gut tells you that the skinhead is engaging in hate speech here, then your conception of hate speech is not well-defined by the content theory.
So hate speech is a combination of the speech prohibited under the content, dignity, and performative theories? Looks like hate speech is not so easy to define.
Someone who supports bans on hate speech might be inclined at this point to throw up their hands and say this problem is just academic. However, free speech is a fundamental human right. If someone want to create a whole category of speech that is not protected by the right to free speech, then what they are really asking for is a limited infringement of that right. Thus, it is incumbent upon that person to give at least a somewhat coherent definition of that speech that is not protected. One suspects that a proponent of hate speech, because it is so ill-defined, will ultimately fall back on the old “I know it when I see it” canard; i.e. they want to ban what they don’t like. Thus the problem is not just academic.
But lets press on despite these problems.
We see, then, that “hate speech” is an ill-defined category. This means that when two different people are talking about “hate speech” they could be talking about two materially different things. Ill-defined categories like this encourage arbitrary and heavy-handed censorship.
I suspect that when most people are thinking about hate speech they mostly have a content theory in mind, so I will continue this essay with that in mind. Going forward then, when we talk about ‘hate speech’ we will mean speech that is used to incite hatred of a group. But keep in mind, this may not be what someone is talking about when they say ‘hate speech’ in another forum.
With this provisional definition in mind, we run into the next problem: what constitutes inciting hatred? Some cases are easy. Nazi propaganda against Jews, Ku Klux Klan propaganda against Black people, and radio RTLM broadcasts promoting Tutsi genocide in Rwanda are all clear examples of inciting hatred. In each of those cases, preexisting prejudices and hatreds were inflamed into mass murder. But these would be the easy cases.
Suppose instead a US politician says, “we must reduce the amount of immigrants and refugees we take in because more people in the workforce drives down wages.” Is this hate speech under the content theory?
The first thing to note is the statement contains both a fact claim and a value claim. The fact claim is that more immigrants increases the size of the workforce and a larger workforce drives down wages. This is a complex question beyond the scope of this essay, but it is either true or false and so a question of fact. The value claim comes in the call to reduce immigration numbers to protect wages; i.e. the speaker says we should value keeping wages high over ease of immigration.
With that in mind, where is the hate speech located in the statement “we must reduce the amount of immigrants and refugees we take in because more people in the workforce drives down wages?” Is it found in the fact claim, the value claim, or both?
Suppose our anti-hate speech crusader, let’s call him Bob, says it’s in the fact claim. Then Bob takes the position that pointing out what is true or false can incite hatred against a group. As absurd as this seems, there is some truth to this. If a politician points out to their constituents that there wages have been going down because immigrants have expanded the workforce beyond the number of jobs available, this may lead to anger in the native population against the immigrant population. Certainly this has happened in the past. So let’s admit that the content of a statement of fact may in fact incite hatred.4
Let us turn to the value claim. Is advocating for prioritizing higher wages over more immigration is inciting hatred towards immigrants. Bob could argue that setting up a dichotomy between higher wages and more immigration will incite hatred against immigrants by directly pitting native self-interest against the presence of immigrants. That is, if merely stating that immigration drives wages down can incite hatred, then directly pitting the interests of natives against those of immigrants is even more likely to incite hatred.
Thus Bob can reasonably argue that either or both of the fact and value claims could incite hatred. This means that under the content theory, the statement can be considered hate speech.
Incitement: An Interlude
One objection that the free speech advocate, let’s call her Miranda, might make in reply is that the idea of incitement is nonsense. Miranda might ask Bob, “Well Bob, I’ve seen right wing propaganda, and never felt the need to start a riot. Does right wing propaganda have that effect on you?” Miranda is reminding Bob and the rest of us that if we believe we have agency - control over our own actions - then we cannot lay the blame for our actions on the words of others. After all, she points out, she is capable of self-control. Is Bob saying other people are not?
While I am sympathetic to this argument, it flies in the face of reality. There is a difference between two individuals arguing philosophically at a coffee shop and a talented orator whipping an already heated crowd into a frenzy. In the first case, you have two individuals arguing in the abstract. Even if things get heated, they can take a step back, agree to disagree, and change the subject. The second case can turn into a riot even though no individual in the crowd came intending to riot. Crowds can be more prone to violence than individuals. However, the word ‘can’ is key here so don’t get too carried away. (This is a point to which we will return.)
But even outside of a crowd, individuals can be provoked to anger, and anger can lead to violence. We are not the rational beings Miranda imagines us to be, and Bob should respond that, yes, under the right circumstances he can be provoked into smacking somebody. So can Miranda. Let us accept that incitement is a thing that happens.
On to Part 2
So far we have seen that so-called hate speech is ill-defined, but that people who talk about hate speech have raised some legitimate concerns. In part two, I will discuss why, despite these concerns, we must reject carving out a hate speech exception to free speech. Please join me in about a week for the next part.
As I have said before, free speech is a first-order value and tenet of classical liberalism, such that one cannot be a liberal and be against free speech. It would be like being the Pope but not being Catholic.
One could argue, I suppose, that by calling for the exclusion of the group of non-US citizens from residence in the US is an affront to the dignity of the group consisting of non-US citizens. I would argue that the group comprising non-US citizens (i.e. the vast majority of the world) do not constitute a group except under the most formal definition.
I will leave it as an exercise for the reader to determine the difference between Afghani, Hispanic, African, and Haitian refugees on the one hand and Ukrainian refugees on the other.
A more astute student of rhetoric may point out that it is not really the statement of fact that is inciting the hatred. Rather is is the unstated middle term, “Your wages have dropped” which leads to the conclusion “Immigrants are why your wages have dropped.” Thus, it is not the content of the fact claim that incites hatred but the concomitant reasoning process it triggers. I think this is correct. However if the concomitant reasoning process is there, and making the fact claim is the event that then incites hatred, then the distinction is ultimately irrelevant.