Hate and Punishment

Attitudes and Their Manifestations

I will begin with the notion of an attitude or sentiment, as it is crucial to any LE account. Although the term is sometimes used loosely, an attitude is distinct from an emotion or passion or affect. Rather, an attitude is a way of relating to someone or something that disposes one to have different emotions in different situations, to want certain things, to focus attention on certain things, to deliberate in certain ways, and possibly to make use of evidence in certain ways (cf. Anderson & Pildes, 2000). My attitudes show how much I value someone or something. Thus, if I love someone, I will feel joy when he or she thrives and sadness if he or she suffers, I want that person to do well for his or her own sake and am committed to doing things that further his or her good, I notice things that are opportunities or threats for him or her, I take those opportunities and threats into account in practical deliberation as reasons for action, and may be more likely to believe that he or she is excellent at his or her job than the evidence warrants. Perhaps there is a distinct feeling of love that goes with the attitude, but love itself is more than that feeling.

Moreover, in the case of an attitude like love, I welcome or endorse the reactions associated with the sentiment, rather than being in their grip. In one sense of the word, I can identify with the pattern of passive responses. Only once I do so, the attitude will be genuinely mine. This does not mean that the attitude is under direct voluntary control, but neither is it purely passive, as it is sometimes portrayed. In fact, there is plausibly an asymmetry between our control over loving and not loving someone. We cannot will love into existence, but we may be able to will it out of existence, that is, if thoughts about someone give rise to no positive feeling or spontaneously focus our attention, there is nothing we can directly do to change that. That is the problem with some arranged marriages. In contrast, if I am already in a relationship and find myself spontaneously attracted to a co-worker, say, I may well be able to deliberately set aside the feelings and refocus my attention on something else, and catch myself before falling in love.

Hate, as I understand it, is in many ways the mirror image of love. If I hate someone, I want him or her to do badly, whether or not it is of instrumental benefit for me. I feel bad if the person does well, get easily angry with him or her, and may be delighted if misfortune befalls him or her. I tend to notice things that are opportunities or threats, and take those as reasons for action—though obviously in the opposite way to the case of love. In full-blown hate, I do not regard these dispositions as something I want to get rid of, but embrace them, perhaps even take pride in them. Again, I may not be able to will hate into existence, but still able to will it out of existence. Perhaps I realize that I am responding to Jewish people in the same negative way, regardless of their individual qualities. Insofar as I refuse to allow these feelings and thoughts to influence my deliberation and action, I am not a full-blown anti-semite—though of course, I am not quite innocent either, until and unless the passive responses cease.

I do not want to exaggerate the similarities between hate and love, however. One important difference between them is that hate may be entirely impersonal, as Thomas Brudholm (2010) emphasizes—the individual characteristics of the hated person need not matter, if she is hated on account of belonging to a group (p. 295). Relatedly, unlike resentment or indignation, hate does not essentially seek punishment of its target for what the person has done. There does not have to be a wrong that should or could be repaid or forgiven, although hate may of course be triggered by wrongdoing as well. As Brudholm observes, even when it is a kind of personal reactive attitude, hate targets the character of the person rather than particular actions (Brudholm, 2010).

One important consequence of the general picture of attitudes I have introduced is that unlike occurrent emotions, attitudes are not transparent. It may be that one cannot be mistaken about whether one feels joy or sadness or rage right now, but one can be mistaken about whether one loves someone—or hates someone. When having an attitude goes against social norms, false belief about it may even be the normal condition. In many circles, anti-semitism, for example, is very disreputable, so people who are prejudiced toward Jews may sincerely but falsely believe that they are not. Other people can be better judges of our attitudes than we are ourselves.

We can manifest or express our attitudes in two different ways. We can do so by way of using language or some other convention-based system of signs that is designed to convey our stance to others. I can express my love to my wife by telling her that I love her or by making her something she likes. As I use the term, the expression relation is non-causal, non-factive, and not necessarily intentional. Speech and symbols can express psychological states we do not actually have, as happens when we are insincere. They can also express attitudes we are not aware of having or do not mean to express. Otherwise psychoanalysts would go out of business.

The other way of expressing attitudes is acting in a way that is best made sense of by attributing the attitude to the agent. I will talk about enacting the attitude in this case to contrast it with symbolic expression. For example, I enact my love for someone by going out of my way to comfort him or her when the person is in need, whether or not I say anything about my feelings. I just express the love by performing the actions that it motivates me to do. Now, as I have defined enacting, it is also potentially non-factive. It may be that what makes best sense of my behavior is an attitude I do not actually have. It may be that what I do and the way I do it is such that any reasonable person, well-versed in my cultural context, would conclude that I love someone, were they to examine my behavior, when in fact I am actually entirely indifferent. Still, in my view, I have enacted love, and people who attributed the attitude to me would be justified in doing so, though mistaken about my mental states. Or, to put it differently, they would be mistaken about my actual attitudes, but not about the attitude embodied by my behavior. It is a common mistake in this context to liken or assimilate this kind of expression to linguistic meaning (see, for example, Adler, 2000; Hampton, 1992). In enactment, convention plays at best an indirect role, and its pragmatic effects are very different.

My view belongs to what we may call the externalist family, as according to it, the meaning of an action is not determined by what goes on inside the agent’s (or any particular interpreter’s) mind. Instead,

to grasp the expressive meaning of an act, we try to make sense of it by fitting it into an interpretive context . . . [A] proposed interpretation must make sense in light of the community’s other practices, its history, and shared meanings. (Anderson & Pildes, 2000, p. 1525)

Think about a case of everyday jealousy. A friend of mine is a very good singer and musician. Recently, he went to a wedding party with his wife, who urged him to grab an acoustic guitar lying around. When he reluctantly did so, some 20-something girls gathered around him, showering him with requests and adulation. He complied with the requests and welcomed the adulation, with the result that his wife subsequently accused him of flirting with the girls. My friend felt unfairly judged, as he was not in fact interested in the young women (or so he says). However, his wife had a point, too: His behavior did reasonably express an interest. Without going too deep into what makes for a reasonable interpretation, a good starting point is asking oneself about the goals that an informed and rational agent in the relevant cultural context would pursue by engaging in the observed kind of behavior. There will inevitably be indeterminacy here, but that is the kind of fact of life that must often be acknowledged in the context of law.

One relevant issue in this area will be whether groups of people, such as legislatures or courts, can have or express attitudes above and beyond the attitudes of the individuals who comprise them. The philosophical pendulum has swung in the direction of the legitimacy of attributing group attitudes (see, for example, List & Pettit, 2011; Tuomela, 2007). Following this tradition, I will take it as given that the attitudes of a collective (such as a state) are determined by the goals, assumptions, and inferences that enough of its relevant members jointly accept as the basis of collective action. As members of a group, people can take as given premises that they would not make use of in their personal practical or theoretical reasoning, which means that we can attribute to a group goals or beliefs that a minority of members, or no one, has. For example, military planners might agree to design an armament program starting from the assumption that the Chinese will have new fighter planes in 2016, even if few of them personally think they will not be ready yet.

Which attitudes should be ascribed to a group, then? I will take Blackburn’s (2010) credibility principle as a starting point:

A group may be said to have been committed to a belief (goal, principle) if there is no way—no credible way—that the group could rationally sustain their open affirmations were they not also prepared to stand by the belief (goal, principle). (p. 81)

The most straightforward way a group can commit to a goal is announcing, after going through the procedures its members consider as authoritative, that it has decided to do such-and-such. Thus, the European Central Bank (ECB) may make it known that it will lend money to Greece. But credibility also leaves room for groups having beliefs and goals (and indeed attitudes) that are not directly affirmed by any of the relevant spokespeople. The ECB might be committed to preferring a rise in unemployment to a rise in inflation, if it could not credibly deny such a preference, given the pattern of past and present decisions, even if it is never openly affirmed by anyone.

Hate Crime and the Varieties of Harm

No one, I think, would deny that some of the attitudes we have, such as hate or bias, are morally inappropriate. But should having or manifesting such attitudes be the business of the law, as advocates of hate crime and hate speech legislation want? Isn’t law concerned with our external behavior rather than motives, as Kant, among others, argued? Is it right to punish one offender more strictly than another, if the only difference between them is in the quality of their motives?

The commonly accepted starting point for answering these questions is the principle of proportionality: Deserved punishment for a crime is a function of both the degree of wrongdoing and the degree of culpability or responsibility (see, for example, Al-Hakim & Dimock, 2012). If hate crimes deserve a stricter punishment, they must either amount to a greater wrong or involve greater culpability. The latter, plausibly, is primarily determined by factors such as intentionality, presence of coercion or invincible ignorance, negligence, and so on. With respect to them, there is unlikely to be a systematic difference between hate crimes and parallel crimes. Much of the discussion about hate crimes, including this article, thus focuses on the difference in wrongdoing.

What determines the degree of wrongdoing? One simple answer is that wrongdoing is at least in part a function of the degree of harm to others that an action causes. Perhaps rape is worse than sexual harassment, because it is more harmful to the victim. Along this line, some have defended stricter punishment for hate crimes on the grounds that they cause greater harm to the victim (Iganski, 2001; Lawrence, 1999; Wisconsin v. Mitchell, 1993). For example, there is some evidence that the psychological and physical harms of hate crimes are likely to be more serious or more widespread than those of parallel crimes.

However, Heidi Hurd and Michael Moore (2004) argue convincingly that enhanced punishment for hate crimes cannot in fact be defended on the grounds that such crimes cause greater physical, psychological, collateral, or social harm than otherwise motivated crimes. This, they rightly point out, is a contingent matter. If a greed-motivated crime, say, causes the same harm, why should it be punished more lightly? Motivation here is only a poor proxy for what really makes for the degree of seriousness of crime, namely, harm (Hurd & Moore, 2004). ² Similarly, if it is morally worse to harm someone who is in a vulnerable position than someone who is not, this does not justify differential treatment of a perpetrator who targets someone vulnerable because of his or her race, for example—it is the vulnerability rather than the sort of characteristics that hate crime legislation involves that justifies the difference (Hurd & Moore, 2004). Conversely, as Anderson and Pildes (2000) point out, thick-skinned victims of hate crime may be psychologically unaffected by the hate aspect. This leads to the ridiculous consequence that “only the thin-skinned and psychologically fragile are entitled to be treated with dignity” (Anderson & Pildes, 2000, p. 1543). So it is no good to appeal to greater conventional harm to justify enhanced punishment for hate crime.

Many defenders of hate crime legislation recognize the difficulties with appealing to ordinary kinds of harm, and consequently introduce a new concept of distinctive expressive harm. This, then, provides a rationale for enhanced punishment. As Dan Kahan (2001) puts it, “By imposing greater punishment on [hate] offenders, hate crime laws say that society regards the harms they impose as different from and worse than the harms inflicted by those who assault or kill for other reasons” (p. 182). Kahan does not do much to explicate the different kind of harm—indeed, his considered view is better construed as a version of the “wrong valuation” account I discuss below. For a full discussion of what makes for expressive harm, we need to turn to Elizabeth Anderson and Richard Pildes (2000).

Anderson and Pildes’ (2000) argument proceeds as follows. They begin with the claim that “A person suffers expressive harm when she is treated according to principles that express negative or inappropriate attitudes toward her” (p. 1527). This definition does not by itself say what makes expression of inappropriate attitudes a non-psychological harm for a person. Their answer draws on a link between expression of attitudes and social relationships: “The communication of attitudes creates social relationships by establishing shared understandings of the attitudes that will govern the interactions of the parties” (p. 1528). Given this, “Communications can expressively harm people by creating or changing the social relationships in which the addressees stand to the communicator” (p. 1529), where these social relationships include “friendship and enmity, collegiality and rivalry, and superior and inferior caste status” (p. 1530). So the argument seems to have three premises:

The first premise concerns the nature of social relationships. It says that what makes us colleagues, for example, is that we mutually understand that we are to relate to each other in certain ways, and, perhaps, not in other ways—for example, we share an understanding that we will not hate or love each other. (If we do, we do not do so qua colleagues.) The second premise says that the shared understandings are created by expressing attitudes. To be sure, the understanding that governs our relations may not be the result of anything you and I specifically did. In most cases, we inherit the conception of what goes with what role from our cultural and historical background. But our expressive actions reaffirm and possibly shape the understanding we have, and perhaps the way we acquire the tradition is by way of expressive action. At least, this is not wholly implausible. The last bit, then, is that some social relationships are as such bad for us to stand in. Which relationships? Anderson and Pildes (2000) do not say, but their example, not of hate crime but of expressive harm of state action, illustrates what they mean:

Racial segregation sends the message that blacks are untouchable, a kind of social pollutant from which “pure” whites must be protected. For the communicative goal to be realized, its meaning must be acknowledged. This does not mean that the addressees must believe, approve of, or accept the message. They simply have to understand it. Once people share an understanding that segregation laws express contempt for blacks, these laws constitute blacks as an “untouchable,” stigmatized caste. (p. 1528)

So on this view, the expressive harm of hate crime, independently of its psychological effect on people, consists in its manifesting attitudes toward the victim that place the offender and the victim in an undesirable and harmful relationship in which one dominates the other. If that is the case, enhanced punishment is justified on the basis of greater wrongdoing than in the parallel crime.

As Section “Responding to Hate” will show, I find much that is congenial about this picture. However, I do not believe it will quite suffice. Some social relationships may be all about how we feel about each other: what makes us lovers is that we love each other. But other relationships involve more than a shared understanding of interaction-governing attitudes, and some of those are highly relevant in the context of hate crime. Relationships of domination and subordination and oppression, in particular, may exist in the absence of such understanding. I may think that I am the social or legal equal of someone above me in a hierarchy—and in some cases, they may think the same way—but sadly, that is not sufficient to make us equals. Relationships of relative power do not reduce to how we understand each other.

The problem for Anderson and Pildes is that to claim that simply understanding that the behavior of others expresses contempt toward me as a disabled person, for example, while nevertheless not accepting that it is in any way merited, suffices to make me stigmatized or disadvantaged makes little sense. Instead, being dominated or subordinated is a status that one may have. As I will argue in the next section, status does not supervise just on attitudes that others express toward me, but on how society, and the law in particular, responds to those attitudes. What matters for status is not just symbolic communication but concrete enactment through punishment.

Punishment, Expression, and Inviolability

If the arguments of the previous section are correct, it is not greater or different kind of harm that makes hate crime deserve enhanced punishment. A different understanding of both crime and punishment is called for. This is what LE promises. Proponents of this type of view emphasize that both crime and punishment have an attitudinal or communicative dimension. It is common for proponents to begin with the observation that merely causing harm or loss to another, even if done intentionally, does not suffice for punishable crime. Instead, we need to look at the attitudes expressed by the perpetrator. Many emphasize that these attitudes embody a mistaken evaluation of the victim’s worth. As Dan Kahan (2001) puts it,

what moves us to condemn an actor for harming another isn’t the simple perception that her actions have diminished another person’s welfare, but rather the judgment that her actions express too low a valuation of the other person’s worth relative to the actor’s own ends. (p. 181; cf. Kahan & Nussbaum, 1996, pp. 351-352)

Kahan draws on the work of Jean Hampton, who argues that treating someone in a way that is precluded by his or her value or represents him or her as less valuable than it is “diminishes” his or her value, and simultaneously represents the agent as elevated with respect to the victim (Hampton, 1992, pp. 1672-1677; cf. Murphy & Hampton, 1988, pp. 44-45, 124). Note that this is not the same as the victim feeling demeaned or insulted, or losing self-respect. Nor does it require that the perpetrator intends to diminish the value of the victim—that is not what a pickpocket has in mind. The wrongdoing involved in crime is a function of the valuation it expresses rather than harm alone. Call this the wrong valuation account of wrongdoing.

Let me be explicit that in this view, wrongdoing is a matter of the attitude enacted by the action, or what some call its social meaning. As I have endorsed an externalist construal of social meaning, this view is a target of the following objection raised by Hurd and Moore (2004):

[Externalist] construals of social meaning . . . would make a defendant more blameworthy (and thus deserving of greater punishment) not because of any fact about him or his deed; rather, he would be subjected to increased punishment because of the appearance of there being a fact about him, namely, the appearance that he possessed a hateful or bigoted motivation for his crime (regardless of whether he in fact possessed such a motivation). (p. 1107)

Hurd and Moore argue that such appearance could not possibly justify punishment, especially when false. However, their reading misconstrues the externalist view. There is a fact about the deed—for example, that it is disrespectful of someone. This fact is nothing other than the fact that it would be reasonably taken to be so by someone aware of the features of the action and context. Blameworthiness is a separate issue. For LE, what makes the action a potential reason for blame in the first place is that it (objectively) expresses bad attitude or wrong valuation. ³ Someone who performs a hateful act, for example, without realizing it is such (and thus not acting out of hate) may be excused on ordinary grounds—the act was not intentional under the description, the agent was invincibly ignorant of the social meaning of the act, and so on.

The next step is that punishment expresses what Peter Strawson (1962) labeled reactive attitudes—attitudes toward the attitudes or quality of the will that the agent has expressed by his or her action. Such attitudes can meaningfully be attributed to groups on the basis of goals and assumptions that their representatives could not credibly disavow, given their behavior. In the case of punishment, the reactive attitudes are those of the community through authoritative representatives. A classic formulation is given by Joel Feinberg’s (1965) definition of punishment, though he gives it a more conventionalist twist than necessary:

Punishment is a conventional device for the expression of attitudes of resentment and indignation, and of judgments of disapproval and reprobation, either on the part of the punishing authority himself or of those “in whose name” the punishment is inflicted. (p. 400)

For legal expressivists, expressing reactive attitudes is what makes something a punishment. In Antony Duff’s (2001) words, punishment is “a mode of moral communication with offenders that seeks to persuade them to repent their crimes, to reform themselves, and to reconcile themselves with those they have wronged” (p. 116). Some critics argue that expressing public condemnation is not a sufficient account of punishment, as punishment necessarily involves hard treatment, and resentment can be expressed without it. Nathan Hanna (2008), for example, argues that we might express our condemnation by requiring payment of court costs or compensation or by confining people in a way comparable with involuntary psychiatric treatment, and thus without intending to cause suffering (Hanna, 2009). ⁴ Call this the enactment problem: Why does the condemnation have to be enacted by hard treatment rather than merely symbolically expressed?

Putting the enactment problem aside for now, the expressive account of punishment, as such, leaves open the question of whether and how punishment is justified. The traditional justifications for punishment appeal either to the intrinsic value of the perpetrator’s suffering and the perpetrator’s deserving pain (retributivism) or to the benefits of hard treatment, such as the good social consequences of deterrence or incapacitation (consequentialism). An expressivist understanding of crime and punishment need not appeal to a new kind of justification altogether, but may instead provide a different insight into existing models (Kahan, 1996).

Jean Hampton’s take on retributivism is a good example of this. For her, crime sends a message, and punishment counters or contradicts it with its own message. Hampton’s case for this draws on the wrong valuation account of wrongdoing. The criminal’s action makes a false moral claim and denies moral reality. Retributive punishment is “the defeat of the wrongdoer at the hands of the victim . . . that symbolizes the correct relative value of wrongdoer and victim” (Murphy & Hampton, 1988, p. 125). It need not involve the infliction of pain (though that is often part of it), but rather mastering the would-be master himself to “deny the wrongdoer’s false claim to superiority and to assert the victim’s equal value” (Murphy & Hampton, 1988, p. 126). Thus, punishment can “annul the message” (Murphy & Hampton, 1988, p. 131) sent by the crime. Similarly, Kahan (1996) says, “The proper retributive punishment is the one that appropriately expresses condemnation and reaffirms the values that the wrongdoer denies” (p. 602).

I believe that Hampton’s view is on the right track. But it places too much weight on communicating a message about the relative value of the criminal and the victim, and the “evidence” that such a message supposedly provides (Hampton, 1992, p. 1676). There is more than mere communication at stake, or so I will argue. Furthermore, instead of Hampton’s wrong valuation account of wrongdoing, I adopt the more general expressive wrongdoing view: wronging others is a matter of enacting inappropriate or unfitting attitudes toward them (where inappropriateness need not be understood in terms of value).

Responding to Hate

In the previous section, I defended an expressive wrongdoing account of crime and a victim-focused legal expressivist account of the nature and justification of punishment in general. I will now argue that it provides the most promising framework for thinking about hate crime legislation.

In the expressive wrongdoing view, what makes something a crime is that it enacts attitudes that are incompatible with the legal status the victim enjoys as a citizen or occupant of a particular role. Unless society in turn expresses its rejection of such attitudes, the victim’s status really is lowered: He or she is someone to whom it is (legally) acceptable to do such things. In victim-focused LE, this is what justifies punishment, in proportion to the victim’s status and the seriousness of the violation. Ordinary crime, motivated typically by perception of self-interest or a personal grudge, poses a threat to the victim’s inviolability only as a side effect, however. The victim’s status is no part of the criminal’s concern. ⁵

Hate crime is different. The attitudes it enacts—contempt, hate, disgust—contain an evaluation of the victim, by virtue of who he or she is (and hence everyone else who shares the relevant characteristic), as inferior. Contempt, for example, presents its object as ranking low in worth as a person, by the light of some ideal the contemner (perhaps wrongly) endorses, in Michelle Mason’s (2003) apt formulation. ⁶ In that sense, at least, these attitudes aim at lowering the status of the victim and those like him or her. And if those who are in positions of authority in the society do not respond to this appropriately, the perpetrators do succeed. It does not suffice to reject the disrespect toward the victim that all crimes manifest to cancel out the hate and contempt. Indeed, failing to respond to this aspect of the crime leaves other members of the society complicit with it. So my main argument for hate crime legislation goes as follows:

The first premise delineates some of the grounds for concluding that a crime expresses a demeaning or degrading attitude toward a group in addition to whatever disrespect or disregard the base crime already manifests. If one burns my house because it is in the way of a new development one wants to build, he or she is disrespecting me; if one burns my house because he or she does not want my kind to live in the neighborhood, he or she also shows contempt and disgust toward me and my people. The second premise points to the link between imposing sanctions and the attitudes of the public this manifests. The third one links sanctions to status, as discussed above. The fourth premise draws the preliminary conclusion that the public’s attitudes manifested by the failure to sanction the hate aspect constitute the victim’s group as inferior to others. This is a strong claim, of course. But it seems to parallel the case with disrespect: Again, if we do nothing (or react only verbally) while someone is being assaulted, for example, or adopt norms that fail to direct us to do anything, we are thereby making the victim into someone de facto insignificant. If we do nothing to react to the hate or contempt aspect manifested by the hate crime, we acquiesce with the treatment of the victim as less significant than others, thereby lowering his or her de facto status.

The fifth premise says that an inferior status is bad in itself. In the real world, having such a status and being aware of it is of course also psychologically harmful. If one realizes that no one is going to do anything if one is raped or beaten, this will have effects beyond the rape and beating itself. But as Nagel (2002) and Kamm (2005) pointed out, lacking inviolability is bad in itself. To put the point differently, it amounts to being dominated by another. As Philip Pettit (1997) has famously argued, even a slave who is never actually interfered with is worse off than a free person, as the owner is always in a position to interfere with the slave’s life on arbitrary grounds.

The conclusion, then, is that equal respect for persons, the kind of attitude that any legitimate liberal state is required to manifest toward all citizens, requires that crimes that embody hate or prejudice are punished more seriously than parallel crimes that do not embody the same attitudes. Contrary to a claim popular with critics, this is not punishment for having an attitude, but for enacting it, and furthermore enacting it in a way that is independently disrespectful of others in a way that merits state intervention (and thus should be criminalized in any case). Again, wrongdoing does not consist merely in harming others, but doing so in a way that embodies a devaluing attitude. Laws such as the Wisconsin Hate Crime Statute (Wis. Stat. § 939.645) that prescribe increased penalties for intentional selection of the victim on the basis of perception of protected characteristics seem to target the right thing.

Conclusion

In this article, I have argued in defense of the view that both crime and punishment should be understood in expressive terms in general. Part of this defense was distinguishing the different ways in which attitudes can be expressed, symbolic expression and enactment. I have argued that although attitudes of hate or bias are in themselves beyond the rightful scope of legislation, those who manifest them by way of performing actions that are independently wrong and are reasonably interpreted as aiming to lower the status of their victims are morally liable to suffer enhanced punishment to protect the equal status of the victim’s group. ⁷ For us all to live in a society of equals, we must express our rejection of the rejection of equality that hate crime enacts by more than just words.