Racial Profiling and a Reasonable Sense of Inferior Political Status.

Racial profiling of black and “AMEMSA” (Arab, Middle Eastern, Muslim, and South Asian) individuals is a central topic of political discussion in the United States. In this article, I will present an argument—“the political argument”—that vindicates the critics of racial profiling of black and AMEMSA people.[ 2] I make an original case that racial profiling creates an unjust relationship between black and AMEMSA people and their state, because it gives them a reasonable sense of inferior political status.[ 3] I will argue that, in place of a policy of profiling these groups, we should have randomized screening (along with a clear prohibition on profiling them).

My argument will be conditional: under present circumstances, profiling these groups would be unjust. I will also give some reasons for thinking that changing these background conditions in ways that would make profiling acceptable is unfeasible. But for people who are more optimistic about what is feasible, my argument will also provide guidance on what it would take to make profiling acceptable.

The case I make is distinctive in several respects. First, it identifies a special set of political harms that can be created by profilingand thus provides especially compelling reasons for thinking that profiling is unjust, as opposed to merely bad in some respect. Second, it provides an extensive response to concerns about whether people's psychological responses to profiling are reasonable, especially when profiling is based on genuine differences in crime rates between groups. Third, it offers a new case for thinking that, while racial profiling of black and AMEMSA people is unjust, profiling of other groups may be permissible, depending on their background relationship with the state.

The article proceeds as follows. In the next section, I will explain what racial profiling is, along with the randomized alternative, and then (Section ) I will present the standard argument in favor of profiling. I will then (Sections and ) present my political argument against racial profiling of black and AMEMSA people. I address some objections in Section , and Section VI concludes.

One final note: my primary focus here is on two specific case‐studies: racial profiling of black and AMEMSA people in the United States. These examples are of obvious contemporary importance and are also the central cases discussed in the literature. Furthermore, as I show, close attention to particular social and historical context is essential when considering profiling, and comparing the two cases helps to illustrate how quite different contextual factors can be relevant: it is especially interesting to see that even a group that doesn't face the same systematic disadvantages as black communities has, all the same, features that tell against profiling. While my argument will build on specifics of the two cases, I will also offer a framework along the way that has more general application for other groups and other countries.

WHAT IS RACIAL PROFILING, AND WHAT IS THE ALTERNATIVE?

Racial profiling consists in the authorities screening for potential criminals in a way that “relies on the race, ethnicity, or national origin and not merely on the behavior of an individual.” [ 4] It involves a search for any one of a large number of people involved in criminal behavior (or some broad category of crime), and is thus distinct from what is often called “suspect selection,” where the police are looking for the perpetrator of some particular, say, robbery, generally on the basis of witness reports.

Here are some familiar examples of racial profiling of black and AMEMSA people. One is the “pretext stop,” where black drivers are pulled aside by the police, ostensibly because of minor traffic violations, but in reality as a means of searching them for contraband. A 2007 study by the US Bureau of Justice Statistics found black drivers were significantly more likely to report being searched by the police during a traffic stop than white drivers (9.5 percent as compared with 3.6 percent).[ 5] Another example of racial profiling is focusing counter‐terrorism investigations on AMEMSAs. For instance, David A. Harris estimates that in the months following the 9/11 attacks, the Department of Justice “undertook the ‘voluntary’ questioning of 8,000 young Arab men who were, according to the U.S. attorney general, not suspects.” [ 6]

We can distinguish two kinds of racial profiling. First, there is de jure racial profiling, where the state adopts an official policy of profiling. Second, there is de facto racial profiling, which involves the police profiling people on the basis of their race, even though the state has not approved their doing so. I will focus mainly on the permissibility of de jure profiling, but I will also argue for a strict prohibition on de facto profiling.

What is the alternative to racial profiling? The policy I will focus on is one of randomized screening (combined with a prohibition on profiling).[ 7] While racial profiling creates a situation in which black and AMEMSA people are searched at a higher rate than members of other groups, selecting people at random for screening ensures that a black or AMEMSA person's chance of being searched is equal to that of a member of any other group. A widely discussed example, which has been implemented in some jurisdictions, is roadblocks, where the police set up an unannounced checkpoint and stop all drivers who pass through it. Alternatively, the police might stop, say, every third driver passing through the checkpoint. Another example is having unannounced checkpoints at airports that stop each passenger passing through a particular area.

What is it to prohibit racial profiling of a group? I will assume two components. First, the state restricts de facto profiling, announcing and enforcing a ban on individual officers screening based on race. Second, central branches of government—those with overriding legal authority—ban subsidiary branches from engaging in de jure profiling: from having a policy of profiling that group. In the United States, this would take the form of a federal prohibition.

THE CASE FOR RACIAL PROFILING

Before presenting the political argument against profiling black and AMEMSA individuals, I will briefly explain the basic case in favor of profiling, which will be useful to have in mind later.[ 8] This case assumes that there is currently a statistical correlation in our society between membership in particular racialgroups and committing certain kinds of crime—for instance, between being black and engaging in gun violence, or between being an AMEMSA and committing acts of terrorism. Under this assumption, profiling will allow the police to apprehend a larger number of criminals.[ 9] This is supposed to reduce the harms of crime, both by stopping crimes already in process and by deterring potential criminals, since it raises the expected costs of criminality. It may also be desirable in itself to catch and punish a greater number of criminals, if we wish to ensure that wrongdoers receive retribution.

An immediate problem with this argument is the difficulty in establishing that there are statistical correlations between racial group membership and likelihood of committing certain crimes. The main source of evidence for these correlations is previous arrest rates. But if, as is very plausible, the police have already been engaged in profiling—searching some groups at a higher rate than others—then arrest rate differences cannot be used to directly infer an underlying correlation between group membership and criminal behavior. Given past profiling, the arrest rates will be skewed, showing more arrests for people in the profiled groups, even if they do not in fact commit crimes at a higher rate.

Suppose, however, that the state can find good evidence for the underlying crime rates and that these suggest targeting blacks and AMEMSAs would increase the “hit rate” at which police searches successfully expose crimes. I will call this “rational profiling.” [ 10] Would this be unjust? The question is of obvious interest for those who think genuine correlations have in fact been established[ 11] despite the difficulties mentioned above, but it is also relevant to skeptics. Even if it is difficult to know what the underlying crime rate is, it is still worth considering whether the state should be spending more time working out how to discover it. And even for those skeptical about whether profiling blacks and AMEMSAs would reduce crime, considering rational profiling is a useful heuristic for assessing whether there are significant moral problems with profiling that go beyond the possibility that it would be ineffective at exposing crimes.

A REASONABLE SENSE OF INFERIOR POLITICAL STATUS

In this section, I will begin to build the political argument for prohibiting racial profiling (including rational racial profiling) of black and AMEMSA people and instead adopting randomized screening. To get the argument going, I will need to first defend some theoretical claims. In this section, I will argue that it is unjust for some members of a political community to have a reasonable sense of inferior political status. I'll start with an explanation of what that sense is and why it is relevant to justice, and then I'll discuss why that sense must be “reasonable.”

A SENSE OF INFERIOR POLITICAL STATUS: WHAT IT IS AND WHY IT MATTERS

Justice requires that the state give equal weight to the rights and interests of each member of the political community. It also plausibly requires the state to do a sufficient amount to secure for each member the confidence that her rights and interests are being given equal weight. The state must take steps to ensure that people do not have (what I will call) a “sense of inferior political status.” We can define a sense of inferior political status as the perception or belief that the state substantially discounts the rights and interests of members of one's group relative to the rights and interests of members of other groups. For our purposes, “substantially” could mean across a wide range of policy areas, or within areas that are especially central to people's lives, or both. (These conditions will be given more illustration in Section , when I consider how racial profiling affects people's interactions with various branches of government.)

I will return later in this section to the issue of just how much the state must do to ensure that people do not develop a sense of inferior political status. First, let me address why we might care about this sense at all. I hope you'll agree that it is simply plausible on its face that it is a concern of justice if people have a sense of inferior political status; but let me now offer two arguments for this view: a civic inclusion argument and an alienation argument.

Civic Inclusion

Liberal philosophy recognizes the importance of not only securing various formal legal rights for each—such as rights to run for political office, serve on juries, and practice religion—but also the means to make effective use of these formal rights. It is a problem of justice if people lack the resources to have a genuine political voice or the education needed to play an active role on a jury. Their ability to make use of their formal rights also depends on certain psychological resources. As Rawls points out, a diminished sense of self‐worth saps people's motivation to make use of their formal legal rights: without confidence that their projects are worth pursuing, they will be discouraged from making use of personal liberties that would enable them to, for instance, engage in artistic expression.[ 12]

I suggest that having a sense of inferior political status also affects people's ability to make use of their formal rights, in particular those that involve interacting with the government: they are discouraged from engaging with a state that takes their rights and interests less seriously. Empirical studies demonstrate this effect. For example, blacks and AMEMSAs who lack trust in the police and criminal justice system are less likely to report crimes committed against them.[ 13] Having a sense of inferior political status may well also lead to a diminished sense of self‐worth and affect people's ability to make use of their formal rights by that route also.

Alienation

According to a long tradition in political philosophy, inspired especially by Rousseau, justice requires that each member of a political community be able to see the state as her agent, acting on her behalf.[ 14] And this is possible only if the state is seen as acting primarily out of a concern for the common good—for the interests and rights of all members of the political community—rather than mainly a sectarian concern for the interests and rights of some particular group. When this condition is violated, individuals see state regulation as something done primarily to them rather than for them. For instance, they might see the police as an entity that aims mainly to restrict their lives for the sake of others, rather than to provide them (in addition to others) with security. Let us say that such individuals are “alienated” from their state.

People with a sense of inferior political status are alienated from their state: they view it as a force that substantially discounts their rights and interests for the sake of other members of the political community. Thus, it is plausibly unjust in itself for people to have a sense of inferior political status, independently of its effects on their civic participation.

THE LIMITS OF THE STATE'S RESPONSIBILITY

Does the mere fact that someone feels a sense of inferior political status give her a complaint of injustice? Plausibly not.[ 15] Some Christian groups in the United States claim to be politically subjugated—even though their interests are generally well represented in the political sphere—because they are unable to gain pride of place for Christian symbols in public spaces, including courthouses, state capitols, and so on. These individuals feel a sense of inferior political status, but surely this doesn't amount to an injustice.[ 16]

Why is there no injustice here? A plausible explanation is that it is unreasonable for these people to think that they have a second‐class status in the political community. The fact that the government is unwilling to do more for your group than it is willing to do for others is not a good basis for thinking that you have an inferior status in the political community.

Reasonableness, for our purposes, can be defined as follows: someone is being unreasonable if she lacks sufficient grounds to justify her belief (in the familiar epistemic sense of justification). The state does not have to prevent people from simply having a sense of inferior political status: their sense might be completely baseless. Rather, the state just needs to ensure that they don't have good grounds for such a belief. This means considering the evidence available to each of her status in the political community.[ 17] The full explanation for why the Christian groups mentioned above have no complaint of injustice, even though they have a sense of inferior political status, is this: their evidence does not support their view that they have an inferior political status, thus it is unreasonable for them to have a sense of inferior political status, thus they have no complaint of injustice.

Roughly this standard has already been frequently appealed to, at least implicitly, in a wide range of debates. Consider, for instance, discussions of segregation. A common argument against “separate but equal” facilities was that they induced in black people a sense of being a lower caste in the eyes of the state. Defenders of these facilities argued that this was “not by reason of anything found in the act, but solely because the colored race chooses to put that construction upon it”: their evidence, the argument went, supported the view that the state wanted the races separate but not the view that the state deemed one race any better than the other.[ 18] Critics of segregation pointed out that the evidence available to black people went far beyond the mere existence of separate facilities and the quality of those facilities. For instance, the fact that segregation was more common in areas where there was more explicit prejudice against blacks was good evidence that segregation was instituted to protect white people from having to associate with blacks (and not vice versa).[ 19] In sum, the parties to this debate granted that blacks had a complaint only if it was reasonable for them to acquire a sense of inferior political status. The two sides just disagreed over whether it was in fact reasonable for blacks to acquire this sense.

One final theoretical observation before we move on: I have argued that it is unjust for members of a political community to have a reasonable sense of inferior political status, and that whether it would be reasonable for someone to have this sense depends on the evidence available to her. This means that in a racially divided society, where different groups have very different interactions with other people and with the state, we must pay attention to differences in the evidence available to members of each group. Whether black and AMEMSA individuals have a complaint of injustice will turn on the evidence available to members of those groups of their status in the political community.[ 20]

THE POLITICAL ARGUMENT AGAINST RACIAL PROFILING

We saw in the previous section that it is unjust if some members of society have a reasonable sense of inferior political status. In this section, I will argue that it would be reasonable for black and AMEMSA people to develop a sense of inferior political status in light of being profiled.

Let me first explain some background reasons blacks and AMEMSAs have for distrusting the state, especially the branches that deal with crime, and then I will explain why against this background it would be reasonable for blacks and AMEMSAs to develop a sense of inferior political status in light of being profiled, even when profiling is rational.

BACKGROUND DISTRUST

Black people have multiple and strong reasons to distrust the police and criminal justice system more generally. For example, they have been especially subject to police brutality and harassment and given more severe punishments for the same crimes.[ 21] These facts give blacks reason to be wary of biases on the part of individual police officers, prosecutors, and so on, but they also reveal major political failures: a lack of policy changes that would serve black interests better, such as introducing more community policing programs, changing sentencing guidelines, and so on. Moreover, one main driving force of federal criminal justice policy has been racial stereotyping. Politicians in the latter half of the twentieth century drew heavily on racialized images of inner‐city dysfunction to create support for the major alterations to criminal justice policy made in that period, such as longer sentences and more limited parole for drug‐related crimes.[ 22] Indeed there is a plausible case that the racialization of crime in public discourse was itself the result of an earlier conscious effort by conservative white political actors—who had been set back by the civil rights movement—to create public anxieties about black criminality, and thereby prevent increasing racial integration and diminish the credibility of black leaders.[ 23]

Blacks also have significant independent reasons for distrusting policy making in other areas. Take, for instance, education and housing. While de jure (legally required) segregation has been abolished for some time, de facto segregation in schools and housing remains extremely common and is accompanied by substantial disadvantage in black neighborhoods.[ 24] Even controlling for differences in income, black children are more likely to go to schools that are under‐resourced and to live in areas with low property values and poor‐quality public services. These patterns are in place partly because of the actions of private individuals—such as choices about where to live—but they are also a product of political decisions, including limited commitment to both integrative efforts, such as school busing, and redistributive efforts, such as funding schools out of state or federal taxes rather than local property taxes.[ 25]

The situation of AMEMSAs is importantly different from that of blacks. Historically, they have not typically been “economically or geographically separated from the general American public.” [ 26] But the treatment of AMEMSAs in the United States has changed dramatically since 9/11 in ways that give them good reasons to distrust the state, including, most centrally, the police and security agencies. For example, the FBI has frequently abused the trust of AMEMSA communities by using “community outreach” events to secretly spy on mosques and individuals.[ 27] Similarly, the NYPD engaged in widespread warrantless wire‐tapping of AMEMSA families.[ 28] These policies and behaviors seem to have been driven less by any increase in terrorism since 9/11 and more by increased public stereotyping and fear of AMEMSAs.[ 29] Previous treatment of groups associated in the public consciousness with national security risk also gives AMEMSAs reason to be concerned. An obvious example is the internment and exclusion of Japanese Americans on the West Coast during the Second World War, implemented mainly in order to satisfy popular anxieties at the time rather than in response to any real threat of espionage.[ 30] AMEMSAs are unlikely to suffer such an extreme fate, but the example does suggest that fears about national security can lead to significant mistreatment of minorities.[ 31]

The relationship between AMEMSAs and security forces has also given them justification for distrust of other branches of government. The threat of surveillance is a strong deterrent to engaging in political activism or even voicing controversial political views to friends, so AMEMSAs have had good reason for feeling shut out from the political process. In so far as AMEMSAs are blocked from having a strong political voice, that gives them reason to expect that the state will often not be responsive to their interests. The association of AMEMSAs with terrorism has also affected public attitudes to their participation in government in ways that give AMEMSAs good reason to fear discrimination when they seek political involvement. For instance, by 2014 only 30 percent of white Americans were confident that an Arab American would be suitable for any important governmental role.[ 32]

WHAT CAN BE REASONABLY INFERRED FROM RACIAL PROFILING?

Considered against this background, what can blacks and AMEMSAs reasonably infer from racial profiling (including rational profiling)? I will consider two possibilities and argue that each is plausible: ( 1) that profiling is based on racial stereotyping, and ( 2) that it reflects a general willingness to impose substantial costs on blacks and AMEMSAs by using their race as (what I'll call) a “negative proxy.”

Profiling and Stereotyping

Although explicit racial prejudice and beliefs about a genetic hierarchy of races have declined somewhat, racial stereotypes are prevalent, including negative stereotypes about blacks and AMEMSAs.[ 33] Following Lawrence Blum, we can define stereotypes as “false or misleading associations between a group and an attribute that are held by their subjects in a rigid manner, resistant to counterevidence.” [ 34] Thus, blacks and AMEMSAs who believe that they are being profiled due to racial stereotypes think that the state is acting on ill‐founded associations between those groups and criminality.

Are there good grounds for this inference? Historically, racial profiling has often reflected racial stereotyping. Blacks and AMEMSAs can thus reasonably believe that a current policy of racial profiling is likewise so based. Second, it is especially hard for the government to “prove” that it is relying on accurate statistical generalizations rather than stereotypes. This is because, as we saw earlier, the main evidence for those statistics is arrest rates, and there is no straightforward way to infer the underlying crime rate from those rates. So a government that claims to be acting on accurate crime statistics would have to ask blacks and AMEMSAs to take this in significant part on trust: trust that can reasonably be withheld. Thus, even if the government was in fact relying on accurate crime statistics—engaging in rational profiling—blacks and AMEMSAs could reasonably assume that it was not.

Profiling and the Use of Race as a Negative Proxy

What if it is unreasonable for blacks and AMEMSAs to infer reliance on racial stereotypes from profiling that is in fact rational? There would, I will now argue, still be an important problem with profiling blacks and AMEMSAs. Let us say that the government uses race as a “negative proxy” when it uses race as a stand‐in for some negative trait (in our case, criminal behavior). Racial profiling essentially involves using race as a negative proxy and doing so in ways that disproportionately disadvantage members of the targeted group: blacks and AMEMSAs will be subject to more searches than members of other groups.

In addition to reliance on stereotypes, criminal justice policy has exhibited a willingness to use the race of blacks and AMEMSAs as a negative proxy in ways that differentially and substantially disadvantage members of those groups. Consider, for instance, investigations of AMEMSAs designed to identify terrorists. Plausibly, in some of these cases government officials at least believed that they were acting on accurate generalizations. Yet their actions still exhibited a lack of regard for the interests of AMEMSAS, given that those policies involved heavy invasions of privacy, restricting freedom of movement, and so on. AMEMSAs thus have grounds for concern that even if the government were to limit itself to acting on accurate generalizations, it would still do so without proper regard for their interests.

Defenders of profiling can say that it shows only that the state is generally willing to tolerate some degree of unfairness—some members of society having to bear differential burdens—in order to secure important public goods. As long as the unfairness involves the imposition of a discrete and relatively small cost—the “inconvenience” of subjection to brief extra screening at the airport, for instance—then it is not unjust to impose those costs. We might compare profiling, on this way of thinking, to subjecting some people to quarantine in order to prevent the spread of a deadly disease. There is unfairness in their having to bear these costs, but it is not sufficiently grave to amount to an injustice if the quarantine is brief.[ 35]

The problem with this suggestion is that the state cannot credibly assure blacks and AMEMSAs that a particular instance of profiling is simply a “one‐off” imposition of a discrete cost. In context, blacks and AMEMSAs can reasonably see any particular instance of racial profiling as part of an established pattern of the state being willing to use their race as a negative proxy in ways that disadvantage them—the cumulative effects of which have been substantial. They can thus reasonably take racial profiling to be a sign that the state will continue to use their race as a negative proxy without proper regard for their interests.[ 36]

By contrast, banning racial profiling of blacks and AMEMSAs sends a strong signal that the state has decided to altogether rule out using their race as a negative proxy. Given that the core argument for profiling requires using race as a negative proxy, banning profiling is an especially visible commitment to not using their race as such a proxy. It is thus an important means of building trust with black and AMEMSA people that they are not going to be continually burdened by the general use of their race as a negative proxy.

To summarize, even if racially profiling blacks and AMEMSAs is effective, and even if the state can show that it is not relying on racial stereotypes, there is still an important objection to profiling blacks and AMEMSAs.

THE POLICE AND THE STATE

I have so far argued that racial profiling of blacks and AMEMSAs would give them good grounds for thinking that criminal justice policy reflects a lack of regard for their interests. Would this really give them good grounds for inferring that they have an inferior political status? It might be said that, while they could reasonably infer that the state is discounting their interests somewhat in some particular policy areas, this doesn't mean that they could reasonably take themselves to have an inferior political status overall, which involves a substantial discounting of their interests by the state. I will now argue that it would be reasonable for blacks and AMEMSAs to take the more general view that their overall political status is one of inferiority. I will offer two reasons for this. First, that criminal justice policy plays an especially important role in the lives of blacks and AMEMSAs, and so to be treated unequally in this area heavily alters their overall status with respect to the state. Second, that unequal treatment in the area of criminal justice policy gives them reasons to infer that they will be treated unequally in other areas of policy.

Criminal justice policy plays an especially crucial role in the lives of black and AMEMSA people. The police and criminal justice system are a main source of contact with the government for many blacks. It is true of poor Americans generally that their “most visible and direct contact with the government may be with a police officer rather than a welfare counselor,” and this effect is especially pronounced in poor black communities.[ 37] Many blacks have had, or at least have family members who have had, previous arrests or convictions—black men have on average been incarcerated for over three years by the age of 35—and this means continuing interactions with parole officers, and so on.[ 38] A policy of racial profiling sustains this dynamic by ensuring that blacks will be disproportionately subject to police searches and disproportionately convicted for the same crimes. And interactions with the criminal justice system have had a very significant impact on black interests by, for example, increasing unemployment in black neighborhoods due to high levels of incarceration and decreasing black political voice through felon disenfranchisement laws.

Given that the criminal justice system has been a constant and impactful presence in black lives, and to the extent that they can reasonably view that system as serving other people's interests over theirs, they can reasonably develop “a ‘sense of the state’ [that] is one of control, hierarchy, and arbitrary power”: Lerman and Weaver's description of how negative interactions with the criminal justice system in fact affect black people's perception of their overall political standing.[ 39]

AMEMSA people have also often had, though for different reasons, especially substantial contact with the police, as well as security agencies. A core part of counter‐terrorism policy has been based on the idea that cooperation from the AMEMSA community is necessary to prevent violent extremism. This has meant, for example, that the FBI has frequently sent agents to mosques to speak with imams and to the congregation.[ 40] There has also been substantial undisclosed (though eventually discovered) contact between AMEMSAs and the police/security agencies: for example, the FBI has also sent secret agents to mosques, Arab community gatherings, and so on.[ 41] Distrust of the police and security forces has had a large impact on the interests of AMEMSAs, creating anxiety, limiting their ability to rely on the police to protect them (including against increasing levels of hate crimes), and chilling their exercise of freedoms of religion, speech, and association.

Putting all of these observations together, it seems justifiable for blacks and AMEMSAs to give special weight to their relationship with the criminal justice branch of the state in assessments of their overall political status, because that branch plays a special role in their lives, forming a central point of contact with the state and having a major impact on their interests. I'll now consider how the relationship blacks and AMEMSAs have with criminal justice branches of government affects their relationship with other branches of government.

Policing and criminal justice is ultimately controlled by central government bodies. Although individual police forces develop distinctive policies and cultures, major decisions—including decisions about whether to endorse or prohibit profiling—are still made by centralized institutions. Indeed, criminal justice issues—especially as they relate to black communities—have been a core focus of federal government attention in the last fifty years.[ 42] Counter‐terrorism efforts have also been a key policy area for the federal government since 9/11. Thus, if the federal government endorses or allows racial profiling in the criminal justice system, there are good reasons for concern that racial stereotypes and the use of race as a negative proxy (even where this use is harmful) will occur in other contexts. Conversely, a centralized prohibition on profiling is an important way for the federal government to signal that it will not allow stereotyping of blacks and AMEMSAs to seep into government action at any level, and to signal that it will not allow their race to be used as a negative proxy in any context. It is important, for instance, for blacks to be assured that their race will not be used as a proxy for criminality in decisions about housing, and for AMEMSAs to be assured that they won't be stereotyped as terrorists in government hiring decisions.

These considerations might be defeated if blacks and AMEMSAs had strong independent assurance that their interests will be given due weight in policy decisions outside of the criminal justice context. But we saw earlier that they do not have strong assurance of this kind: they in fact have grounds for suspicion that these other branches of government will not give their interests due weight.

THE ROLE OF CONTEXT AND IMPLICATIONS FOR OTHER GROUPS

I have emphasized that my argument against racial profiling assumes various background conditions, such as a historically poor relationship with the police on the part of blacks, and so on. So someone might insist that this means the real reason for any feelings of inferior political status is just these background conditions and not racial profiling itself.[ 43] But this underestimates the role that profiling itself can play. The relevant background conditions affect the interpretation of racial profiling people can reasonably come to. But once they have adopted that interpretation, it can certainly affect their relationship with the state. To see this point, consider segregation. As we saw earlier, it was only within a broader context of racial inequality that segregation signaled contempt towards blacks. Yet there is no doubt that this message was highly significant for the relationship between blacks and the state. What we see, and can expect to see, from racial profiling is a vicious cycle where distrust feeds into the interpretation of profiling, which itself fuels distrust, and so on.[ 44]

What about profiling of groups other than blacks and AMEMSAs? A key part of the argument of this article is that because of their background relationship with the criminal justice system and the state as a whole, it is reasonable for racially profiled blacks and AMEMSAs to develop a sense of inferior political status. This suggests that the political argument does not rule out other forms of profiling, including racial profiling. Suppose, for instance, that the government focused its searches for snipers on white men (based on statistical evidence about the characteristics of snipers). White men have no substantial history of having their interests discounted in the formulation of criminal justice policy or in other policy areas. Indeed, history shows their interests being mainly given additional weight. It would thus not be reasonable for them to take profiling as a sign that they have inferior political status. By contrast, racialprofiling of Latinxs in immigration enforcement may well be unjust, according to the political argument, given the substantial history of Latinxs being disadvantaged due to stereotypes of criminality.[ 45] The key general point is that any policy involving profiling must be evaluated with close attention to historical and social context.[ 46]

OBJECTIONS

THE COSTS OF CRIME

I have argued that racial profiling of black and AMEMSA people is unjust even if it is rational (based on accurate statistics that show disparate crime rates between racial groups). Supposing that such profiling catches more criminals, couldn't it be justified overall? For some crimes, such as petty theft, the gains of more effective enforcement may be relatively small, while for other crimes, such as terrorism, increased enforcement might decrease people's chances of suffering serious bodily injury and even death (and from a retributive standpoint, punishing the latter crimes might be especially important). Can profiling be rejected if and when it has such gains and, in particular, are the justice considerations I have raised sufficiently important to weigh against those gains?

Denying people a sufficiently good basis for believing that they are equal political members is, I have argued, a serious injustice. So we should weigh potential security gains here just as we weigh public goods in other contexts where requirements of justice—such as basic liberties of speech and religion—are at stake. In these contexts, we standardly insist that two tests be met:

The benefits themselves meet a very high bar, high enough to count as a “compelling state interest.” [ 47] For instance, restrictions on speech for the sake of security must serve to prevent an imminent, highly likely, and very destructive threat.

and

There are no other means of achieving the same benefits without creating similar injustice. (The “least restrictive means” test.) For example, suppose that there is a problem of voter fraud and two possible ways to solve it. One method is to require people to show an identification card, while a more expensive method uses a computerized system to verify an individual's identity. Now, suppose that many poorer people do not have the identification cards. It seems we must use the computerized method, even if it is significantly more expensive, because otherwise many individuals will face the injustice of not having an effective right to vote.

This is not the place for a detailed empirical investigation of the potential benefits of profiling (though see the brief overview in Section ). My argument at least puts the burden of proof on defenders of profiling to demonstrate that any benefits meet the high bar set by condition (a) above. And those defenders generally concede that requirement (b) is not met: they accept that there are typically alternative means of getting any potential security benefits that might come from profiling, including the policy that I have proposed: increased randomized screening.[ 49] Defenders of profiling object that these alternative measures would produce more inconvenience for all.[ 50] But though they do impose costs, these measures would not be comparably unjust, since they would not have the impact on people's sense of their political status that racial profiling of blacks and AMEMSAs does. Thus, there is a strong presumption in favor of increasing these inconveniences rather than engaging in profiling.

Isn't there some point at which the cost of securing justice would be so high that we would have to make a sacrifice? I think so. But, as the voter fraud example indicates, and as test (a) again suggests, the costs would have to be very high. The onus is on defenders of profiling to show that the costs of increased randomized screening are so high that an unjust method of reducing crime should be chosen over an equally effective just one. Mere additional inconvenience for all seems unlikely to suffice. Moreover, it is especially hard for defenders of profiling to claim both that being screened is a small enough burden that imposing it disproportionally on blacks and AMEMSAs is not deeply unfair and that increasing searches somewhat for all would be a massive burden.

REFORMING THE PRACTICE OF PROFILING

According to the second objection, it is possible to make profiling of blacks and AMEMSAs compatible with justice by changing other elements of policing. The idea is to adopt profiling, while at the same time introducing other means of assuring blacks and AMEMSAs that their interests are being given due weight by the state.

The most familiar version of this proposal focuses on the manner in which police stops are carried out.[ 51] Profiling must be accompanied by strict prohibitions on police abuse of various kinds, ranging from demeaning comments made during stops to severe police brutality. As we will see in my response to the next objection, I do think the manner in which stops are carried out is important. However, I do not think that changing these conditions would be sufficient to render racial profiling compatible with justice. Here are two reasons why.

First, it does not seem possible for the state to have a policy of profiling of blacks and AMEMSAs while at the same time providing them with strong assurance that they will not be subject to police abuse. Profiling has historically been closely correlated with police abuse. Moreover, blacks and AMEMSAs can reasonably assume that profiling contributes to police abuse in so far as police officers (even if mistakenly) take it to license harsh treatment of those groups. So they can reasonably be skeptical about the feasibility of profiling without police abuse.[ 52]

Second, even if blacks and AMEMSAs could be reassured that there would be no police abuse involved, they would still have reasons to take racial profilingas a sign that their interests are not being given due weight. As we saw earlier, there are multiple reasons—the prevalence of police abuse being only one—why blacks and AMEMSAs can reasonably distrust the motives behind criminal justice policy. So even if steps were taken to curb police abuse, blacks and AMEMSAs would be likely still to reasonably infer that profiling reflects reliance on racial stereotypes or at least a willingness to use their race as a negative proxy in ways that cause substantial disadvantage.

Now, I concede that if the state were to secure—by providing widespread evidence of equal treatment—substantial background trust on the part of blacks and AMEMSAs, the case against racial profiling of blacks and AMEMSAs would be much weaker. The more their relationship with the police and the state as a whole looks like the relationship between white men and the state, the more plausible it is to think that profiling them is compatible with justice. This does not seem feasible in the short run, however. And, plausibly, one important step in starting to secure that background trust would be having a strict prohibition on profiling these groups: we saw earlier that a prohibition of this kind is an important means of signaling that the state is not going to allow black and AMEMSA interests to be substantially set back by racial stereotyping or the use of their race as a negative proxy.

OVERDETERMINATION

The previous objection asked whether we could have profiling while rendering the signal it sends to blacks and AMEMSAs relatively benign. The “overdetermination” objection we will now look at says that, given background conditions, the decision whether or not to profile will have no effect on what black and AMEMSA people can reasonably infer from the practice of police searches.

According to this objection, blacks can interpret the presence of police stops of any kind—even randomized stops—as a sign of their inferior political status. (The objection is most plausible as applied to blacks than to AMEMSAs, so I will focus on that case.) This suggests a reductio of my position: I have argued that a policy that would generate a reasonable sense of inferior political status is unjust and thus should not be implemented. But if blacks can reasonably take any police searches as a sign of inferior political status, then my argument seems to suggest that any such stops would be unjust and must be prohibited. But surely, the objection goes, that cannot be correct: surely there should be some searches of black drivers and AMEMSA flyers.

I suggest that a policy of randomized screening combined with curbs on police abuse could be used to create a situation in which black and AMEMSA people should accept those stops without developing a sense of inferior political status. Randomized stops spread the costs of police searches equally across the population and thus ought not to signal any willingness on the part of the state to specially burden blacks or AMEMSAs. Steps would, of course, have to be taken to make the process transparent, so that black and AMEMSA people have reason to believe that the stops are genuinely randomized. I cannot canvass all possible institutional mechanisms for doing this, but here are two reasons for optimism. First, the stops themselves can be relatively transparent at the point of implementation: for example, it can be straightforwardly observed whether each driver (or every third driver) is being stopped at a roadblock. Of course, there is still the possibility that the roadblocks are being strategically located outside of places that are especially frequented by black people. But this too could be monitored somewhat by onlookers, and—this is my second major point—it is comparatively easy to establish whether the overall policy is creating randomized outcomes. All we need to compare is the rate at which members of certain groups are being stopped with the proportion of those people within the population. We do not need to establish what the underlying crime rate is, as we would need to do in order to show that profiling is rational. Thus, it is easier to provide assurance to blacks and AMEMSAs that profiling is randomized than that it is rational.

What if blacks and AMEMSAs would have good reasons to infer a lesser political status from being searched, even when the searches are at random and police abuse is curbed? Would it follow from my argument that the police should never stop a black driver or an AMEMSA flyer? It would not. My response to the Costs of Crime objection noted that an unjust policy can be acceptable overall if (a) it is necessary to prevent threats that are very substantial, and (b) there is no alternative that would be more just. Plausibly, having some police stops of blacks and AMEMSAs meets these conditions: it is necessary to prevent grave threats of murder, theft, and so on, and there is no readily available more just alternative means of preventing these threats.[ 53]

Someone might still say: if, given the background conditions, blacks and AMEMSAs will still have sufficient reasons to develop a sense of inferior political status whichever criminal justice policies are adopted today, then wouldn't it be worth profiling anyway (assuming that profiling is rational)? I have given reasons above for doubting that the situation is this dire. But if it is, the right response isn't to profile and thus exacerbate the reasons blacks and AMEMSAs have for distrusting the state; but rather to start the process of providing reassurance that they will be treated justly. A strict prohibition on racial profiling, combined with the randomized alternative, is an important step in that direction.

CONCLUSION

These are my findings about racial profiling of black and AMEMSA people in the United States. Whether or not it is an effective way to limit crime, there is at least one serious moral problem with profiling, even rational profiling, of these groups—a wrong that is done to all members of these groups, even those individuals who are not subject to substantial screening themselves. Profiling causes black and AMEMSA individuals to reasonably develop a sense of inferior political status, and this is unjust because it causes civic exclusion and alienation from the state. To justify profiling, someone would have to show both that it brings with it very significant security gains and that there are no alternative—less unjust—means of securing the same gains. But it will almost never be the case that there is no alternative. In particular, the state can almost always use randomized screening instead. And thus, given that a less unjust alternative exists, racial profiling is not justified.

Why is it reasonable for racially profiled black and AMEMSA people to develop a sense of inferior political status? Because, given recent and historical interactions with the state, they can reasonably interpret profiling as signaling a lack of regard for their interests: either a willingness to rely on racialstereotypes about them or a willingness to use their race as a negative proxy in ways that cause substantial disadvantage. By contrast, white men, for instance, could not reasonably interpret profiling of their group as a sign that the state gives their interests less weight. More generally, I have argued that whether profiling of a particular group is ruled out by the political argument depends substantially on background historical and social context.

Black and AMEMSA lives matter, but do they matter to the state as much as white lives? Racial profiling gives these groups good reasons to be doubtful, and this makes it unjust.

Notes

1 For comments I am extremely grateful to Michael Blake, Lawrence Blum, David Boonin, Adam Cox, Helena de Bres, Branden Fitelson, Helen Frowe, Stephen Gardiner, Robert Goodin (as editor), Sally Haslanger, Chris Heathwood, Aziz Huq, Kristi Olson, Ronni Gura Sadovsky, Ron Sandler, Raul Saucedo, T. M. Scanlon, Tommie Shelby, Susanna Siegel, Joseph Raz, and two very constructive anonymous reviewers. Many thanks also to audiences at Columbia Law School; Harvard, Northeastern, UCL, and Wellesley philosophy departments; and the Stanford MAP (Minorities and Philosophy) group.

2 I cannot assess here all of the alternative arguments against profiling, but for a helpful discussion, see Kasper Lippert‐Rasmussen, Born Free and Equal (Oxford: Oxford University Press, 2013), ch. 11.

3 My argument thus differs from those that focus on other psychological harms of profiling, such as humiliation, and those that focus on indirect harms of profiling, which occur via what profiling expresses to onlookers. I cannot give a full overview of the literature here, but see e.g. Annabelle Lever, “Why racialprofiling is hard to justify: a response to Risse and Zeckhauser,” Philosophy and Public Affairs, 33 (2005), 94–110; Paul Bou‐Habib, “Racial profiling and background injustice,” Journal of Ethics, 15 (2011), 33–46. Randall Kennedy's Race, Crime, and the Law (New York: Pantheon Books, 1997) and Naomi Zack's White Privilege and Black Rights: The Injustice of U.S. Police Racial Profiling and Homicide (Lanham: Rowman and Littlefield, 2015) offer the closest argument to mine, but I give a much more extensive justification. I set aside Deborah Hellman's views in “Racial profiling and the meaning of racialcategories,” Contemporary Debates in Applied Ethics, 2nd edn, ed. Andrew L. Cohen and Christopher Heath Wellman (Malden, MA: Wiley‐Blackwell, 2013) because of deeper theoretical differences. I address these in my book manuscript “Discrimination, Inclusion, and Social Progress.”

4 Mathias Risse and Richard Zeckhauser, “Racial profiling,” Philosophy and Public Affairs, 32 (2004), 131–70, at p. 136.

5 M. R. Durose, E. L. Smith, and P. A. Langan, Bureau of Justice Statistics Special Report: Contacts between Police and the Public, 2005 (Washington, DC: US Department of Justice, 2007).

6 David A. Harris, “Racial profiling revisited: ‘just common sense’ in the fight against terror?” , Criminal Justice, 17 (2002), 36–41, at p. 59. See also Jack Glaser, Suspect Race: Causes and Consequences of Racial Profiling (New York: Oxford University Press, 2014), chs 2 and 6.

7 For more details about this proposal and evidence of its effectiveness, see Bernard E. Harcourt and Tracey L. Meares, “Randomization and the Fourth Amendment,” University of Chicago Law Review, 78 (2011), 810–77.

8 See e.g. Michael Levin, “Responses to race differences in crime,” Journal of Social Philosophy, 23 (1992), 5–29; Risse and Zeckhauser, “Racial profiling,” sects 1–3.

9 I ignore here a further assumption: that the profiled group's elasticity of offending to policing is greater than or equal to that of the non‐profiled group.

10 David Boonin's term in Should Race Matter? Unusual Answers to the Usual Questions (New York: Cambridge University Press, 2011), ch. 10.

11 Risse and Zeckhauser, “Racial profiling,” p.132, think these correlations are at least plausible.

12 John Rawls, A Theory of Justice, rev. edn (Cambridge, MA: Harvard University Press, 1999), pp. 155–7, 197–200, 386–91.

13 See Glaser, Suspect Race, ch. 5; Tom Tyler, “Policing in black and white: ethnic group differences in trust and confidence in the police,” Police Quarterly, 8 (2005), 322–42; and Aziz Huq, Tom Tyler, and Stephen Schulhofer, “Why does the public cooperate with law enforcement? The influence of the purposes and targets of policing,” Psychology, Public Policy, and Law, 17 (2011), 419–50.

14 Jean‐Jacques Rousseau, The Social Contract, Rousseau, The Social Contract and the Discourses, trans G. D. H. Cole (New York: Everyman Library, 1931).

15 For similar points in the context of racial profiling and discrimination generally, see, for instance, Boonin, Should Race Matter?, ch. 11; and Sophia Moreau, “What is discrimination?” , Philosophy and Public Affairs, 143 (2010), 143–79.

16 Christopher Eisgruber and Lawrence Sager similarly warn against the “tyranny of the squeamish” in Religious Freedom and the Constitution (Cambridge, MA: Harvard University Press, 2007), p. 127.

17 There is of course a debate in epistemology about whether evidence is purely mental. For our purposes, it seems appropriate to assume that evidence is in fact “out in the world,” independent of any particular agent's mental states, since it seems appropriate for the state to be concerned with such publicly available items, rather than individual people's mental states. But my argument should go through even on the mentalistic view.

18 Plessy v. Ferguson, 163 U.S. 537 (1896), at p. 163.

19 “[G]enerally speaking, segregation is the pattern of law in communities where the extralegal patterns of discrimination against Negroes are the tightest, where Negroes are subjected to the strictest codes of ‘unwritten law’ as to job opportunities, social intercourse, patterns of housing, going to the back door, being called by the first name, saying ‘Sir,’ and all the rest of the whole sorry business.” ; Charles L. Black, “The lawfulness of the segregation decisions,” Yale Law Journal, 69 (1960), 421–30, at p. 425.

20 It is especially important not to simply rely on the assessments of whites because, as Renée Bolinger points out, those assessments are likely to be influenced by (often implicit) bias. Whites may well, for instance, be less attentive to facts that suggest substantial racial inequalities. See Renée Bolinger, “Reasonable mistakes and regulative norms: racial bias in defensive harm,” Journal of Political Philosophy, 25 (2017), 196–217.

21 See Rod K. Brunson, “‘Police don't like black people’: African American young men's accumulated police experiences,” Criminology and Public Policy, 6 (2007), 71–102; United States Sentencing Commission, Fifteen Years of Guideline Sentencing: An Assessment of How Well the Federal Criminal Justice System Is Achieving the Goals of Sentencing Reform, www.ussc.gov/sites/default/files/pdf/research-and-publications/research-projects-and-surveys/miscellaneous/15-year-study/15%5fyear%5fstudy%5ffull.pdf>, ch. 4.

22 See, for instance, Joe Soss, Richard C. Fording, and Sanford Schram, Disciplining the Poor: Neoliberal Paternalism and the Persistent Power of Race (Chicago: University of Chicago Press, 2011), pp. 63–5; Vesla Weaver, “Frontlash: race and the development of punitive crime policy,” Studies in American Political Development, 21 (2007), 230–65.

23 Weaver, “Frontlash.”

24 For instance, in the Northeastern US, “65% of black students go to school almost exclusively without white peers”; “Students' return to school is marred by renewed segregation across US,” Guardian, September 7, 2015. These predominantly black schools are systematically less well‐funded; Center for American Progress, “Students of color still receiving unequal education,” www.americanprogress.org/issues/education/news/2012/08/22/32862/students‐of‐color‐still‐receiving‐unequal‐education/>.

25 There is a debate about whether the best policy for tackling racial inequality focuses on integration or redistribution (or some combination). My point is that since neither approach has really been tried, blacks can justifiably infer that the state has not been giving their interests equal weight. For discussion of the two approaches, see Tommie Shelby, “Inequality, integration, and imperatives of justice: a review essay,” Philosophy and Public Affairs, 42 (2014), 253–85.

26 Amaney Jamal, “Muslim Americans: enriching or depleting American democracy?” , Religion and Democracy in the United States: Danger or Opportunity?, ed. A. Wolfe and I. Katznelson (Princeton: Princeton University Press, 2010), pp. 89–113.

27 As reported here:< http://usnews.nbcnews.com/%5fnews/2012/03/29/10907668-aclu-fbi-mosque-outreach-program-used-to-spy-on-muslims>. See also David A. Harris, “Law enforcement and intelligence gathering in Muslim and immigrant communities after 9/11,” New York University Review of Law and Social Change, 34 (2010), 123–90.

28 See, for instance, Mapping Muslims: NYPD Spying and Its Impact on Muslim Americans, a report issued by the Creating Law Enforcement Accountability and Responsibility (CLEAR) Project, CUNY School of Law, <http://www.law.cuny.edu/academics/clinics/immigration/clear/Mapping-Muslims.pdf>.

29 Aziz Huq, “The uses of religious identity, practice, and dogma in ‘soft’ and ‘hard’ counterterrorism,” Security and Human Rights, ed. Liora Lazarus and Benjamin Goold (Oxford: Hart, 2017).

30 See Geoffrey Stone, Perilous Times: Free Speech in Wartime from the Sedition Act of 1798 to the War on Terrorism (New York: Norton, 2004).

31 For evidence of such fears, see Arab American Institute, American Attitudes toward Arabs and Muslims, < http://www.aaiusa.org/american-attitudes-toward-arabs-and-muslims-2014>.

32 Ibid.

33 Lawrence Bobo, “Racial attitudes and relations at the close of the twentieth century,” America Becoming: Racial Trends and Their Consequences, ed. N. J. Smelser, W. J. Wilson, and F. Mitchel (Washington, DC: National Academy Press, 2001).

34 Lawrence Blum, “Stereotypes and stereotyping: a moral analysis,” Philosophical Papers, 33 (2004), 251–89. But see Erin Beeghly, “What is a stereotype? What is stereotyping?” , Hypatia, 30 (2015), 675–91, for an alternative approach.

35 This example appears in Risse and Zeckhauser's “Racial profiling,” pp. 157–8, though here I set aside some details of their discussion.

36 Perhaps this effect can be countered by compensating profiled people. If so, my argument applies just to uncompensated profiling.

37 Amy Lerman and Vesla Weaver, Arresting Citizenship (Chicago: University of Chicago Press, 2014), p. 97.

38 Ibid., p. 99.

39 Ibid., p. 112.

40 Mapping Muslims: NYPD Spying and Its Impact on Muslim Americans.

41 Ibid.

42 Weaver, “Frontlash.”

43 Risse and Zeckhauser make a related suggestion in “Racial profiling.”

44 The possibility of cycles of distrust is also noted by Lever, “Why racial profiling is hard to justify.”

45 See, for instance, Douglas S. Massey and Karen Pren, “Origins of the new Latino underclass,” Race and Social Problems, 4 (2012), 5–17, at p. 6: “political and bureaucratic entrepreneurs … demonize[d] Hispanic immigrants by constructing a ‘Latino threat narrative.’”

46 What if profiling one historically under‐served group would have security benefits for another? Or if—as Risse and Zeckhauser claim is the case for blacks—profiling members of one historically under‐served group would benefit members of the same group? Even if creating security for historically under‐served groups is especially important—perhaps it even supports their sense of equal political status—my view still recommends creating it through means other than profiling, such as randomized screening. (The claim that profiling blacks is a useful means of creating security for black communities is disputed by, for instance, Aziz Huq, “The consequences of disparate policing: evaluating stop‐and‐frisk as a modality of urban policing,” Minnesota Law Review, 101 [2017], 2397–480.)

47 See e.g. Austin v. Michigan Chamber of Commerce, 494 U.S. 652 (1990), at pp. 652, 655.

48 Alan O. Sykes, “The least restrictive means,” University of Chicago Law Review, 70 (2003), 403–19.

49 See e.g. Boonin's qualified defense in Should Race Matter?, p. 596. For evidence of the effectiveness of increased randomized screening compared with profiling and other methods, see Harcourt and Mears, “Randomization and the Fourth Amendment.”

50 Risse and Zeckhauser, “Racial profiling,” p. 162.

51 See e.g. Arthur Applbaum, “Racial generalizations, police discretion and Bayesian contractualism,” Handled with Discretion: Ethical Issues in Police Decision Making, ed. J. Kleinig (New York: Rowman and Littlefield, 1996); Risse and Zeckhauser, “Racial profiling.”

52 Risse and Zeckhauser mention feasibility constraints, but do not emphasize how fatal they are for the approach under discussion.

53 What if there is a remaining alternative that would provide greater reassurance for blacks and AMEMSAs: “reverse profiling,” where members of other groups are stopped at a higher rate? Isn't it an implausible consequence of my view that it could recommend reverse profiling? For the reasons explained in the text, I think it is unlikely that reverse profiling is required by my view, given the availability of randomized screening. But I also suggest that reverse profiling is less disturbing than it sounds: we might compare it to affirmative action, where additional burdens are placed on innocent whites for the sake of greater racial justice. While affirmative action is controversial, surely it isn't obviously unjust.

DMU Timestamp: February 07, 2020 23:04