a. Can hate speech deprive one of dignity?

The first critique articulated by Simpson questions the degree to which hate speech has the power to truly deprive its targets of dignity and perpetuate social hierarchy. Waldron, Simpson argues, has an obligation to show that hate speech “diminishes [targets’] status in some discernible sense, beyond merely being a manifestation of the speaker’s view of them as inferior or second-class people.”^{Footnote 21} Waldron must show that hate speech does not merely reflect the speaker’s endorsement of racial hierarchy, but actively perpetuates this hierarchy, for instance by driving others to treat members of marginalized racial groups with less respect. By Simpson’s lights, Waldron fails to do so. It is not clear how the utterance of a slur, for instance, would convince others to view its targets as inferior or accord them lower social and political status. This is especially true in a community in which there’s already a formal and informal endorsement of the ideals of racial equality and dignity, as is the case in most Western countries, including the U.S. Given that our laws guarantee equal respect and overwhelming majorities agree that people shouldn’t be treated differently based on their race, hate speech seems more likely to garner derision and disgust than uptake. There is, according to Simpson, a missing causal claim in Waldron’s argument regarding the ways that hate speech motivates continued prejudice. To justify his thesis that concerns over dignity outweigh free speech interests in the case of hate speech, Waldron must offer a persuasive account linking hate speech and racial hierarchy.

b. Do hate speech regulations target the right thing?

It might be thought that the primary deleterious effect of racist hate speech is not to spread racism within a society, but rather to harm what Simpson calls victims’ “felt sense of security.”^{Footnote 22} On this interpretation, hate speech does not necessarily motivate others to treat members of marginalized racial groups as inferior and unworthy of respect, but rather reveals to victims that there are those who already view and will treat them in this way. It does not so much harm targets’ dignity as show them that their dignity is already under threat. Even if this is the case, however, Simpson doubts that hate speech regulation will do much good. For here it is prejudiced attitudes, not hate speech itself, which seem to be the problem. It is people’s hearts and minds that deny members of marginalized races their dignity, and hate speech merely reveals this harsh reality to its targets. Indeed, if there are those in society who possess such views, it might actually be better if members of marginalized groups hear their expression. For seeing or hearing this hate at least alerts targets to those who wish them harm and enables opponents of racism to work to root out such views.

a. Racist hate speech as constituting harm

Throughout the hate speech literature, proponents of hate speech regulation frequently emphasize that hate speech constitutes rather than merely causes harm. What they mean by this is that it is not merely how others react to hate speech that does harm, but that hate speech by its very utterance does harm. To use a term from philosophy of language, harm is part of the illocutionary force of hate speech, not merely a perlocutionary effect. One of the more persuasive versions of this argument comes from Charles Lawrence, who argues that hate speech is an act of discrimination that excludes members of targeted groups by its very utterance.^{Footnote 23}

There are, I think, two ways that racist hate speech functions as discrimination on Lawrence’s view. First, hate speech discriminates directly in that it marks its targets as second-class citizens and thereby contributes to their being exactly that. Lawrence likens hate speech to segregation, which on his interpretation of Brown v. Board of Education Footnote ²⁴ was ruled “unconstitutional primarily because of the message [it] conveys—the message that Black children are an untouchable caste, unfit to be educated with white children.”^{Footnote 25} Even if you don’t agree with his reading of Brown, Lawrence here hits on something peculiar about utterances regarding fellow citizens’ social and political status: they are at least partly self-validating. If a white man calls a group of Black people a racial slur and tells them they are unworthy of participating in social and political life, he contributes to the diminishment of Black people’s social and political status. These utterances do not merely tell their targets that they’re unequal, but actually partly constitute this inequality, since inequality is partially grounded in the stated attitudes and conduct of members of society. Of course, such utterances can’t guarantee that targets don’t participate in certain formal parts of political life like voting, but they can erode those informal aspects of socio-political status that Karst describes, undermining their treatment as respected members of society. As an analogy, my guarantee to my class that all will be treated as equally valued participants would mean little if I allowed members of the class to denigrate others because of their membership in a racial group. Such treatment would directly undermine this equality.

Racist hate speech therefore creates a kind of hole in the fabric of equal dignity ostensibly ensured by liberal societies. When someone utters hate speech, they make it the case that their target is denied equal status and recognition respect. Intrinsic to the speech act is treatment of its target as an inferior, unworthy of same the status as other members of society. In this way, each utterance of hate speech chips away at the dignity of members of targeted groups.

One might object that a single speaker has limited power to erode their target’s dignity, presenting a version of what Ishani Maitra calls the Authority Objection.Footnote ²⁶ Average hate speakers lack both formal authority to decide on individuals’ status and informal authority over others’ treatment of marginalized group members, and so they can only make it the case that they themselves deny a target equal status. This action, it might seem, has a minimal effect on the status of targets, as all others in society may continue to accord them equal dignity.

There are two ways of responding to this challenge: the first is to show that everyday hate speakers in fact possess some kind of authority relevant to denigrating targets’ dignity, while the second is to deny that hate speakers require authority to significantly diminish targets’ dignity. I think there is a good case to be made for both of these positions, and so I will introduce them in turn. I will here focus on the Authority Objection as it pertains to constituting discrimination by denigrating targets’ dignity, but much of what I say here will apply mutatis mutandis to my argument in the next section regarding the enactment of discriminatory norms.

The first response addresses the question of what grounds a speaker’s authority to denigrate individuals’ status. It is clear that some have this authority: Ishani Maitra offers the example of a legislator in apartheid South Africa asserting (under the right legislative conditions) “Blacks are no longer permitted to vote,” thereby denying Black South Africans social and political equality.Footnote ²⁷ Most kinds of hate speech, however, are not uttered by people occupying formal positions of authority. And yet, holding a formal position is not the only way to possess authority. Maitra introduces two other ways that a speaker might have authority: via derived authority and via licensing.

Someone obtains derived authority when a possessor of positional authority confers it upon them. Crucially, this transfer of authority need not be explicit, but may occur by omission on the part of the initial authority. Maitra offers the case of the Bossy Student: a teacher wants her students to complete a project that requires each performing a different task, but before she can divide up the tasks, one student, Arlo, takes it upon himself to assign tasks himself in full view of the teacher.Footnote ²⁸ The teacher doesn’t intervene and thereby confers authority upon Arlo.

Similarly, while most hate speakers do not have formal positional authority, they may possess derived authority. When it comes to determining individuals’ dignity, one relevant authority is the government, which decides the formal elements of individuals’ social and political status by conferring their basic and political rights and deciding how to weigh their interests. In virtue of failing to prohibit or sanction hate speech, the government plausibly gives hate speakers de facto derived authority. No one with formal authority challenges hate speakers’ assaults upon dignity, and the government thereby cedes the authority to decide on individuals’ social position.Footnote ²⁹ In neither the Bossy Student nor the hate speech cases must the formal authority endorse the actions of the derived authority—the teacher may roll her eyes and the government may quietly deplore hate speakers’ actions—but by failing to interrupt the derived authority’s actions, they cede control.

A speaker can also obtain authority via licensing, which differs from derived authority in that it does not require the complicity of any formal authority. In cases of licensing, individuals implicitly grant someone authority over themselves. Maitra presents the case of the Hike Organizer as an example of licensing: a group of friends is planning a hiking trip, but no one expresses strong preferences about the details.Footnote ³⁰ As a result, the group is taking so long to iron out the logistics together that it seems like they will never finish organizing. One of the members, Andy, decides to take the organizing into his own hands, and none of the others intervene when he presents his plan and doles out responsibilities. In virtue of failing to express any objections, the other members of the group surrender to Andy their power to decide on the trip.

In the case of hate speech, hearers can license a hate speaker with their silence. Individuals’ social and political status, as I’ve argued, is not merely a function of the government’s actions, but also of the words and actions of other members of society. As a result, just as the silence of the other members of the hiking trip grants Andy authority to determine the details of the trip, the silence of onlookers can grant hate speakers the authority to determine targets’ social and political status. Their silence concedes to the position the hate speaker assigns the target, showing that they will not reject the social ranking proposed by the speaker. In effect then, audience members turn over their say in the target’s status to the hate speaker.Footnote ³¹ And again, hearers need not agree with the hate speaker’s sentiments in order to confer authority. Just as other participants in the hiking trip might quietly object to many of the details and still turn over their power to decide, audience members might disagree with the speaker and still turn over their say in the target’s status.

Both of these types of authority arise in cases in which speakers lack authority in virtue of their social position. However, it is also reasonable to think that in many cases of hate speech, speakers in fact possess a kind of informal positional authority. Given the dominant social position of mostly white hate speakers, they possess a kind of de facto authority over the position of members of marginalized racial groups.Footnote ³² Operating in the background is a structure of norms that confers on the average white person the power to at least partly determine the dignity of non-white members of society. At the very least, their opinions about targets’ social status seem to carry more weight than those of the average non-white person.

I have so far argued that hate speakers possess a kind of authority that confers upon their utterances significant influence over targets’ dignity. However, I also doubt that hate speakers require such authority in order to significantly affect the status of members of marginalized racial groups. Let us grant, for the sake of argument, that hate speakers lack derived authority from the government and licensed authority from others to decide individuals’ status. It is still not the case that the function of hate speech is merely that one person denies members of the targeted group their equal status. First, a single utterance of hate speech interacts with targets’ knowledge of the discrimination that persists in the rest of society. In a 2019 survey, three-quarters of Black and Asian Americans, as well as 58 percent of Latin Americans, said they have experienced discrimination or have been treated unfairly because of their race or ethnicity at least from time to time.Footnote ³³ As a result, targets’ understanding of hate speech is not that one person denies them equal status; rather, it reinforces their sense that many in society think them inferior and unworthy and will treat them accordingly. It provides further evidence that they are second-class citizens, erasing any doubts they might have had about how others will view and treat them. It might encourage targets to re-interpret questionable interactions they’ve had in the past, confirming their worst suspicions about the racist nature of those encounters.

One might, however, respond that this argument only establishes that single instances of hate speech can make targets think or understand that they lack equal status in the rest of society, not actually constitute this denigration of their dignity. The line between making individuals understand they lack equal status and denying them equal status is not entirely clear, but for the sake of argument we might even grant that taken in isolation, a single utterance of hate speech itself has a limited effect on the dignity of marginalized groups. However, taken together, utterances of hate speech in a society comprehensively endanger the dignity of their targets. The same 2019 survey suggests that exposure to racist language is not a marginal experience for members of marginalized racial groups, but a regular feature of their lives. 61 percent of Asians, 52 percent of Blacks, and 46 percent of Latin Americans in the U.S. say they have been subjected to slurs or jokes because of their race. Moreover, 65 percent of Americans say it has become more common for people to express racist or racially insensitive views since former U.S. President Trump was elected. This sentiment is particularly strong among Black and Latin Americans, around 75 percent of whom say the expression of racist views has become more common.Footnote ³⁴ Given the predominance of racist expression, we have good reason to think that the aggregate effect of hate speech on targets’ status is significant: it is no longer one individual who denies members of a marginalized racial group their equal status, but many. This treatment clearly threatens the place of targeted individuals in society, as they are no longer truly accorded equal status by others. It would seem foolish to withhold regulation of hate speech because individual instances have a limited effect on dignity, while taken together they have a significant effect. It is this aggregate effect on targets’ dignity that regulation seeks to stem.

Finally, as I’ll detail in the next section, individual utterances of hate speech can spur the expression of additional derogatory utterances, compounding illocutionary harms with damaging perlocutionary effects.

Consider an analogy to racial discrimination by a restaurant. Arguably, if a single restaurant doesn’t allow in Black folks, this has a minimal effect on the place of Black people in society. It’s only one restaurant, after all, and Black residents are free to go to any other restaurant in the area. And yet, we certainly wouldn’t take this to be a reason not to prohibit such racial discrimination. While any individual instance of discrimination might not have a great effect on the place of marginalized groups in society, the sum of such actions certainly does. Moreover, individual instances of racial discrimination tend to embolden others to follow suit.

So much for the Authority Objection. I have so far explained one way that hate speech constitutes discrimination for Lawrence, chipping away at targets’ status by treating them as second-class citizens. The other way that hate speech can be discriminatory is that it functions as exclusionary conduct. When someone utters racist hate speech in a place of work, a school, or a public space, they ensure that the targets of their speech are unwelcome in those spaces. Their words may not literally prevent Black, Latin American, Muslim, or Asian people from entering, but they have a chilling effect, signaling that members of these groups will be treated with hostility and even violence if they enter. In so doing, they deny members of marginalized racial groups equal access to employment, education, and public spaces. This reasoning has found expression in local public accommodations laws, which prohibit hate speech in public spaces in order to ensure that people of all races truly enjoy equal access.

If hate speech does function in these ways as an act of discrimination, it clearly contributes to racial hierarchy. For understood through this framework, hate speech denies members of targeted groups their entitlement to be treated as respected members of society and equal access to some of the most important public goods enjoyed by citizens. It denies to targets the full privileges of citizenship and thereby impugns their dignity. This function of hate speech also shows that Simpson’s understanding of how hate speech can contribute to hierarchy is unjustifiably narrow. It’s not merely that hate speech persuades others to be racist, but that by its very utterance it constitutes and guarantees inequality. Thus, with respect to Simpson’s mandate that we show how hate speech causes racial hierarchies, we can do him one better: it doesn’t merely cause such hierarchies but manifests them.

These functions of racist hate speech ground Waldron’s claim that the harms of such speech outweigh the free speech interests it promotes. Given that such speech diminishes targets’ status and excludes them from certain privileges of citizenship, the harms outlined in Section 4 are manifested. The threats to targets’ social and political equality and other basic rights outweigh the free speech interests implicated by hate speakers’ utterances.

b. Enacting discriminatory norms

The other way that racist hate speech perpetuates racial hierarchies is by enacting racist norms that others then follow to harm targeted groups. I argue by enacting such norms, hate speech constitutes the harm of discrimination, and by galvanizing others to follow these norms, it also causes harm by encouraging racist treatment. Harm therefore resides as part of both the illocutionary force and the perlocutionary effect of hate speech.

To make my case, I will rely on Mary Kate McGowan’s compelling account of how utterances of racist hate speech function as ‘covert exercitives,’ a more general parallel of the phenomena of conversational exercitives.^{Footnote 35} Conversational exercitives are speech acts that enact norms in particular conversations by changing the conversational score, a concept borrowed from David Lewis.Footnote ³⁶ The score is a record of all conversational contributions and affects the moves participants are licensed to make. So, for instance, if I bring up the book On Liberty in the course of a conversation, this licenses the participants to identify On Liberty as the referent of the demonstrative ‘the book’ until a new book becomes salient. Crucially for hate speech, McGowan argues that this phenomenon generalizes to other-norm governed activities, like social interactions or workplace behavior, meaning that any move in these activities enacts certain norms by changing the score. Hate speech in particular has as its primary function enacting discriminatory norms in social situations that others then follow in order to harm marginalized groups. For instance, if a white person yells a racial slur on a public bus, they enact norms licensing participants to make targets feel unwelcome or unsafe, which if followed, do obvious harm to people in targeted groups. By enacting norms calling for racist discrimination, such utterances clearly perpetuate racial hierarchies.

Two features of covert exercitives make this categorization particularly appropriate for hate speech. For one, uttering a covert exercitive and thereby successfully enacting a norm does not require the conscious intent to do so. When one refers to On Liberty, one does not usually intend to raise it to salience, at least in the sense that this is not one’s conscious purpose for uttering the name. As McGowan puts it, such intentions are at the very least on the low end of our manifestness spectrum, meaning they are less conscious and less explicit than for instance my intention to refer to On Liberty.^{Footnote 37} The same goes for hate speech: one can enact discriminatory norms without consciously intending to do so, and indeed many who utter racial insults claim that they are merely “speaking their mind” or “telling it like it is.”

Secondly, one does not require any formal authority to successfully enact a norm via a covert exercitive. Mirroring my response to the Authority Objection in the last section, there are two ways of parsing this: either that speakers possess a kind of derived or licensed authority to enact norms in conversational contexts, or that they do not require any authority at all to do so. On the former interpretation, the phenomena of derived authority, licensing, and unequal power relations from the last section apply equally well to enacting norms. Speakers may gain authority to enact norms in conversational contexts due to the silence of a formal authority or other conversational participants,Footnote ³⁸ or because of their disproportionate power. Or, one might think that we do not require any authority to enact norms in conversational settings, given how ubiquitous and easy it is to enact norms in conversation. One does not seem to need any authority to make On Liberty salient after referring to it in a sentence, and likewise one does not need it to enact discriminatory norms via hate speech. We enact norms in conversational settings all the time without realizing it and without possessing any special authority to do so. Racist hate speech is just another instance of this familiar phenomenon, albeit a particularly harmful instance, since the norms it enacts are norms of discrimination.

Whichever interpretation one prefers, the important thing to note is that speakers need not occupy a position of power to enact a norm in a conversational setting. These two characteristics distinguish covert exercitives from standard exercitives: speech acts uttered by those in some formal authority position that enact norms, such as a demand from an executive that all employees wear collared shirts. Such utterances not only must be spoken by an authority figure in order to enact the relevant norms, but also generally require a manifest intention on the part of that figure. Hate speech is not like this. To enact a discriminatory norm and cause others to follow that norm, one need not intend to do so nor possess any particular authority. Just as when one raises some object to salience, by uttering hate speech one adds a norm to the conversational score.

McGowan argues that by acting as covert exercitives, hate speech constitutes harm. However, she defines ‘constituting harm’ in a strange way, as causing harm by enacting norms that prescribe harm.^{Footnote 39} On first glance, this might appear to be merely an effort to define away the problems that arise when one calls for regulating speech that causes harm, particularly the worry that such regulations might apply to anything that produces some subjective offense. As McGowan acknowledges, “as many who are reluctant to embrace further speech regulations are prone to point out, just about anything can be offensive to someone or other.”^{Footnote 40}

Yet I think there’s more to her insistence that hate speech constitutes harm than a mere definitional trick. As I said, the harm hate speech produces by enacting norms resides in both its illocution and perlocution, and my interpretation of McGowan locates the harm of hate speech in these two different places. First, hate speech directly harms its targets purely in virtue of creating discriminatory norms, because these norms themselves compromise the equal status of targets and exclude them from public spaces, workplaces, educational institutions, and beyond. By locating harm directly in the norms created, McGowan effectively offers another mechanism by which hate speech functions in the two ways Lawrence thinks it does. When a speaker enacts in a conversational setting the norm that members of targeted groups ought to be discriminated against, this diminishes targets’ status. It is now a local norm that targets do not get equal treatment, that their interests do not receive equal weight, and therefore that they do not possess equal social and political status. The integration of this norm into the conversational score is a manifestation of targets’ diminished dignity. Moreover, these norms exclude targets from public accommodations. Norms calling for exclusion do not merely galvanize others to exclude, but themselves communicate to targets that they are unwelcome. It would be odd to say, analogously, that a restaurant posting a ‘Whites Only’ sign merely causes discrimination or exclusion, as if these were disparate effects at the end of a causal chain. Rather, the restaurant simply discriminates and excludes, harming non-white members of society. The norms themselves make members of targeted groups unwelcome.

The way that racist hate speech enacts discriminatory norms gives us further reason to reject the contention that hate speech merely expresses the opinion of one individual and has minimal effect on the dignity of marginalized groups. These norms constitute a kind of non-material ‘Whites Only’ sign that exists in the conversational score rather than on the wall. If a white person on a public bus hung a literal ‘Whites Only’ sign, this would not only express their opinion, but deny non-white folks equal status and access to public accommodations. Similarly, the discriminatory norms enacted by hate speech not only tell targets that one person deems them to be second-class citizens, but serve as a mechanism that denies them their dignity and access and makes them second-class citizens. Members of marginalized groups know they are unwelcome and that they will be subject to abuse and perhaps even violence if they violate the norms.

The other way that hate speech leads to harm for McGowan is causal: by marking racial discrimination as acceptable or obligatory, the norms hate speech enacts cause others to discriminate against its targets. In this way, hate speech in effect causes harm by constituting harm, compounding its illocutionary discrimination with damaging perlocutionary effects. This function of hate speech again reflects the way that hate speech harms beyond merely expressing one person’s opinion, motivating uptake of discriminatory behaviors by others in society. Moreover, this causal account also avoids the objection from subjective offense, as it focuses only on speech that produces a particular type of harm in a particular way, i.e. speech that causes discrimination by enacting discriminatory norms.

Some might think that the causal part of this theory offers an implausible account of the responses hate speech actually tends to elicit. Like Simpson, one might argue that hate speech is much more likely to inspire derision and criticism than uptake and support. Yet this seems to overlook the group dynamics often involved in racial insults. One need look no further than the ‘counter-protests’ that have recently sprung up all over the U.S. in response to Black Lives Matter protests. These counter-protests tend to start with older white people heckling protestors with phrases like “All Lives Matter” or “Blue Lives Matter,” but as they’ve gone on, many protests have devolved into expressions of racial insults and violence. In one instance, a group of older white men started screaming “White Power” at passing protestors,^{Footnote 41} and in another, a group of men surrounded and then punched a young woman carrying a “Black Lives Matter” sign.^{Footnote 42} It seems perfectly plausible that the phenomenon McGowan describes has been at work in these and other instances: counter-protestors are emboldened by the speech of those around them, gaining a sense that racial insults and violence are acceptable in this context. These hecklers feed on each other, becoming comfortable with increasingly vile racial language, until uttering hate speech, treating marginalized groups as inferior and unworthy of respect, and even performing violence becomes socially allowable, if not expected.

The U.S has also seen the phenomenon McGowan highlights at play on a much broader, national scale. Over the last several years, there has been much discussion over the racist rhetoric employed by former President Donald Trump and concern that such rhetoric has led to more racist discourse among the masses. Recent scholarship indicates that there is truth to this suspicion. A 2018 study from political scientist Brian Schaffner shows evidence for what he calls the “Trump Effect”: exposure to President Trump’s prejudiced statements during the 2016 campaign made people more likely to write offensive things both about groups Trump has targeted and other identity groups.^{Footnote 43} A 2020 study from a group of political scientists reveals a similar trend, showing that in the absence of prejudiced speech from elites, prejudiced citizens tend to keep quiet, but when elites express prejudice, prejudiced citizens are more likely to follow suit.^{Footnote 44} Both of these studies highlight the causal effects of hate speech that McGowan discusses, i.e. that it makes expressions of racial animus seem socially acceptable or even obligatory, motivating others to utter racial insults and treat members of marginalized groups with less respect.

One might object that this process does not actually perpetuate racial hierarchies because it merely leads those who are already prejudiced to express that prejudice, rather than driving prejudiced attitudes. Even if this were true, the constitutive element of McGowan’s account shows that an increase in racist speech can itself further malign the dignity of targeted groups, even without an increase in racist attitudes. By diminishing their status and excluding marginalized racial groups from the privileges of full citizenship, hate speech itself contributes to racial hierarchy. Yet there’s reason to think that the norms enacted by racist hate speech can also motivate racist attitudes, because an environment in which such speech abounds presents racism as a socially acceptable option. As Waldron puts it, advocates of hate speech regulation fear that as a result of persistent racial insults, “ordinary people will think and act on the assumption that the place of minority members in ordinary life is up for grabs.”^{Footnote 45} We can imagine young people in such an environment, still in the process of developing their views about the world, coming to see racism not as deplorable, but as a live option. Social pressure certainly has an effect on the views we embrace, and if people see that others in their community express racist attitudes, they’ll be more likely to think such attitudes are acceptable and give them serious consideration. Racist expression may also harden the views of those who already have racist proclivities. Living in a community in which racist expression is commonplace will lead one to be more complacent in one’s racist views, while consistent criticism of such views will force one to grapple with and question them.

Attention to the ways that racist hate speech enacts discriminatory norms that others follow therefore gives us further reason to accept Waldron’s argument that it denigrates targets’ status and ought to be regulated. By enacting such norms, hate speech excludes members of targeted groups and encourages others to treat them with dehumanizing animus and disregard, producing status harms that outweigh the relevant free speech interests. Again, I do not doubt that such speech implicates speakers’ expressive, deliberative, or political interests and that regulation is therefore subject to a higher standard of evidence, but contend that the harm it does outweighs those interests and meets that standard. For, as I outlined in Section 4, this harm violates the guarantee that all members of society enter social and political life as equals, arguably the fundamental requirement of liberal politics, and threatens targets’ other basic rights.

a. Hate Speech and the Deliberative Interest

Proponents of the CLS approach contend that hate speech does not promote what Joshua Cohen calls the deliberative interest: the interest people have in accessing valuable information that can help them lead better lives and help others do the same. Lawrence argues, “The racial invective is experienced as a blow, not a proffered idea” and concludes that hate speech ought to be considered the “functional equivalent of fighting words.”^{Footnote 46} Delgado echoes this sentiment when he says that racial insults “are not intended to inform or convince the listener.”^{Footnote 47} The idea is that racist hate speech has the intent and impact of injuring its victims and promoting fear, intolerance, and violence, rather than sharing an idea or motivating useful discourse. In its strong form, this argument insists that hate speech lacks what I’ll call intellectual content, i.e. an articulated idea or argument, but is rather akin to a “blow,” as Lawrence describes it. Yelling a racial slur at someone does not express an idea and is unlikely to spark a conversation, let alone a thoughtful and productive one. On the contrary, such speech is more likely to silence and drive away its targets.

Yet the position that racist hate speech lacks intellectual content is inconsistent with critical legal theorists’ identification of the harm in hate speech, because these thinkers insist that it is the message conveyed by hate speech that makes it so pernicious. As one example of hate speech, Lawrence, Matsuda, Delgado, and Crenshaw discuss a dispute among white students and a Black student at Stanford University over Beethoven’s heritage, after which a group of white students posted in image of Beethoven with Sambo-like features on the Black student’s door. Regarding this incident, the authors explain, “The message said, ‘This is you. This is you and all of your African-American brothers and sisters. You are all Sambos. It’s a joke to think you could ever be a Beethoven. It’s ridiculous to believe that you could ever be anything other than a caricature of real genius.’”^{Footnote 48} It is clear in this case that the authors think it is the idea expressed, the idea that Black people are inferior and incapable of achieving the feats of genius achieved by white people, that makes this speech so harmful. The speech here is not merely experienced as “a blow” as Lawrence suggests, but as an idea.

This example is not an aberration in the CLS account, but rather reflects its central idea. Delgado stresses that it is the internalization of the message of inferiority expressed by hate speech that is so damaging to its victims, as many “come to believe the frequent accusations that they are lazy, ignorant, dirty, and superstitious.”^{Footnote 49} He cites psychologist Kenneth Clark, who explains that, “Human beings … whose daily experience tells them that almost nowhere in society are they respected and granted the ordinary dignity and courtesy accorded to others will, as a matter of course, begin to doubt their own worth.”^{Footnote 50} It is the idea it expresses that makes hate speech so vicious, the idea that members of marginalized racial groups are inferior and unworthy of equal status. As Henry Louis Gates notes, this means that racist hate speech must have intellectual and political content. “[I]f Delgado and his fellow contributors have a central message to impart,” Gates explains, “it’s that racial insults are profoundly political.”^{Footnote 51} It is the disturbing political message expressed by hate speech that makes it more emotionally and psychologically affecting than a scream or incitement to violence. If it truly lacked intellectual content, hate speech would lack the poignancy that makes it such a compelling candidate for regulation.

Of course, there may be some expressions that incite fear and make targets feel unwelcome that do not express a political message. Screaming inarticulately at someone on the subway or blocking the path of people of color as they walk into a restaurant may stoke fear or incite violence without expressing a clear political idea. Such expressions exclude not by expressing a message of inferiority, but rather by threatening their targets directly. Yet Matsuda, Delgado, and Lawrence are not exclusively concerned with this narrow range of expressions. Rather, they focus particularly on those utterances that do express a message of inferiority, that exclude by telling targets they are unworthy of equal recognition. The prior category of expressions—those that threaten rather than express ideas—would already enjoy regulation under prohibitions on incitement to violence or direct threats. The racial nature of such utterances seems almost superfluous, for it is the threatening and dangerous environment they create that is crucial, not the message expressed. Indeed, if such expressions are to lack intellectual content, their racial content has to be superfluous. Critical race theorists like Matsuda, Delgado, and Lawrence are rightly concerned with something entirely different, a category of expressions that perpetuates racial hierarchy and prejudice by promoting a message of inferiority. Here the messages’ intellectual content is exactly what makes them so harmful. Indeed, Gates argues that the harmfulness of hate speech is often inversely related to its vulgarity and thoughtlessness. The sterile assertion that one is inferior and unworthy of equal status in society is often much more devastating than the angry use of a racial slur.^{Footnote 52}

One might try to sidestep these objections by presenting a weaker version of the view that hate speech has no essential connection to the acquisition of truth. This version says that it’s not that such speech does not express an idea, but rather that it’s not intended to—and is unlikely to—spark productive conversation. Lawrence at one point seems to endorse this weaker view when he argues that in the wake of hate speech, “it is unlikely that dialogue will follow. Racial insults are undeserving of first amendment protection because the perpetrator’s intention is not to discover truth or initiate dialogue, but to injure the victim.”^{Footnote 53} It is not that such speech lacks any intellectual content, but rather that neither the intent nor impact of the speech is to spark deliberation that’s likely to uncover the truth.

And yet, that one thinks such speech unlikely to lead to the acquisition of truth doesn’t seem sufficient for it being uncovered by a free speech principle. If I utter to a man on the street, “The frogs on the moon are triangles, not circles,” this seems much more likely to inspire befuddlement than productive conversation, and yet we would certainly not say such speech should be uncovered by a free speech principle. Of course, such an utterance is also unlikely to do as much harm as hate speech, but the question of harm is not relevant to whether or not an utterance is covered, but only to whether if, uncovered, it ought to be regulated. The reason we would extend coverage to my utterance on the street even though it seems to lack deliberative value is that it is not our place to judge in advance which ideas are likely to promote the acquisition of truth. This is at the very core of the free speech justification: we must tolerate all ideas because we don’t know in advance which are true, and even false or absurd ideas may contribute to knowledge by provoking response. By silencing even false opinions, society loses, according to Mill, “the clearer perception and livelier impression of truth, produced by its collision with error.”^{Footnote 54} This does not mean that no speech can be regulated, but rather that all speech which expresses an idea is relevant to the deliberative interest.

The efforts of philosophers of language to characterize the linguistic function of hate speech fare no better in showing it lacks intellectual content. In an earlier work, Maitra and McGowan argue that, like other types of speech uncovered by a plausible free speech principle, expressions of hate speech are “significantly obligation-enacting utterances.”^{Footnote 55} This means that such speech creates obligations of considerable consequence, in that they are not easily met or there are serious repercussions for not meeting them. Maitra and McGowan argue that such utterances are uncovered because, while the primary value of speech is in its deliberative function, the principal effect of significantly obligation-enacting utterances is to make changes in the world, i.e. creating obligations. Other examples include speech which initiates or dissolves contracts, conspires to commit a crime, or falsely advertises—none of which receive the heightened protections reserved for expression.Footnote ⁵⁶ Racist hate speech is significantly obligation-enacting by Maitra and McGowan’s lights because it enacts obligations to do something illegal, i.e. to deny targeted groups equal access to public spaces, and because there are social costs for refusing to fulfill the obligation.^{Footnote 57}

The problem with Maitra and McGowan’s assertion that utterances of racist hate speech ought to be uncovered because they are “significantly obligation-enacting utterances” is that many such utterances have intellectual and political content. Take for instance Martin Luther King’s famous assertion in his Letter from a Birmingham Jail that “[o]ne has a moral responsibility to disobey unjust laws.”^{Footnote 58} This is clearly a significantly obligation-enacting utterance as it asserts an obligation to break the law, which Maitra and McGowan explicitly identify as a significant obligation. And yet, it also clearly has deliberative value, helping people work out how they ought to act when faced with institutionalized injustice.^{Footnote 59} It is therefore clear that not all significantly obligation-enacting utterances lack expressive or deliberative value, and therefore that not all such utterances should be uncovered by the FSP.

b. Hate Speech and the Expressive Interest

Critical legal theorists also reject the connection of hate speech to what Joshua Cohen calls the ‘expressive interest’: a “direct interest in articulating thoughts, attitudes, and feelings on matters of personal or broader human concern.”Footnote ⁶⁰ Cohen’s invocation of the expressive interest is founded on the intuition that part of living a good life includes the ability to bear witness to what one perceives as true and valuable. As Delgado puts it, individuals have a right “to develop their full potentials as members of the human community.”^{Footnote 61} Yet Delgado also maintains that expressing hate speech plays no central role in promoting this interest. Rather, he argues that bigotry instead stifles one’s moral and social growth, precluding one from achieving higher values like justice or inclusion. Moreover, he does not see racial insults primarily as vehicles for self-expression, but rather as attempts to injure or insult. Just as the intent and impact of hate speech is not to promote discussion, neither is it to express oneself.^{Footnote 62}

This reasoning is flawed on two fronts. For one, the expressive interest does not only extend to expressions likely to promote one’s intellectual or moral development. Rather, as Cohen argues, the relevance of the expressive interest depends on an individual’s subjective stance towards the topic they wish to address.^{Footnote 63} What is important is that one takes oneself to have a strong obligation to express some view or sees the issue as important to justice or human welfare. The expressive interest concerns individuals’ ability to bear witness. There may be plenty of views that individuals see as important to express that we from the outside might view as objectively unimportant or even deplorable and dangerous. This seems to be the case with hate speech: although we may judge the ideas expressed to be objectively deplorable, racists see them as crucially important to share. As Jeremy Waldron argues, the ideas that purveyors of racist hate speech express “are the very messages, out of all things a person could express, which matter most to them.”^{Footnote 64} Because of their stance towards such issues, the expressive interest is implicated in hate speech, even if such expression is unlikely to promote speakers’ intellectual or moral development and instead likely to stunt it.

The other problem with Delgado’s argument is the impropriety of making government the arbiter of what views promote intellectual or moral development. It would be odd, for instance, if protections did not extend to reality TV or romantic comedies because the government deemed them of little intellectual value, but did extend to documentaries or historical dramas. Monitoring the quality of our expression and extending protections only to that expression deemed to possess intellectual or moral value simply does not seem like a proper role for government.

Protecting the dignity of its citizens from racist abuse is, on the other hand, a proper role of government. Of course, allowing a government to determine what expressions denigrate targets is not without its dangers. It is certainly possible that a government would abuse this power in order to limit expression it doesn’t like. Yet the justification of government actions here matters: while the government does not have the authority to determine what expression has intellectual value, it does have the authority to preserve dignity. Risks of abuse are tolerable when they are a side effect of a government doing its legitimate duty to protect the equal status of citizens, but not when they are a product of an already questionable practice. Moreover, allowing a legislature to limit expression that threatens dignity grants the government a more circumscribed power and creates fewer opportunities for abuse. While a government might deem any expression it doesn’t like to have no intellectual value, there are fewer candidates for speech that erodes dignity. For instance, such regulations could only extend to direct attacks on people, not challenges to policies or ideas.