Justice Paper 2 – Discrimination, Random Searches and Profiling (Fall 2006)
January 1st, 2003Is it wrong for law enforcement officials to use racial, ethnic, or religious profiling in deciding whom to search in airports, train stations, subways?
The use of racial, ethnic or religious profiling by law enforcement officials has been much maligned by offended moralists who usually argue that policies informed by such profiling are morally wrong because they are unfairly discriminatory and distastefully reminiscent of racism. I argue that this assessment is both correct and yet irrelevant. By appealing to the cognitive inevitability and practical usefulness of discrimination, and with reference to communitarian assumptions, I will construct the case in favor of discriminatory practices in general. Building upon this argument, I will submit that the moral acceptability of profiling in particular is conditional upon the empirical effectiveness of policies based on profiling methods as well as the actual content and execution of these policies. I then discuss why this analysis holds true under a broad range of different moral frameworks including those of Aristotle, Mill, Locke and Rawls. At this point, I will refute the claims of absolutist proponents of equality as incoherent and untenable before closing with a further defense of profiling against Kantian logic.
Defenders of profiling sometimes cite its almost actuarial nature as making it exempt from moral considerations. In a wider sense, those who argue that profiling is not a moral question but rather an amoral question of statistics are not wrong, although the point can be more strongly made. If we understand “discriminatory” to mean “having a belief formed before case-specific facts are known”, it would certainly appear discriminatory for law enforcement agents or individuals in general to operate based on a set of generalized assumptions, which by definition are not facts and may be falsified by as yet unknown information. In this case, the assumption is that certain identifiable traits (race, ethnicity, religion) are correlated with a marginally higher probability that the bearer of those traits is a suicide bomber. Furthermore, to the extent that this assumption is not based on the individual person but on generalized, probabilistic views about a class of individuals (which may be quite broadly defined), it is difficult to avoid the charge of “unfairness”, since feeling aggrieved and indignant is a common reaction to being incorrectly associated with something as unpleasant as terrorism. But just saying that an assumption is discriminatory or even unfair is not enough to make it morally wrong. The first important fact to remember is that the most important assumptions[1] about other people are always discriminatory (whether positive or negative). Secondly, and more importantly, generalized assumptions are indispensable to human action, which are naturally colored by the assumptions that underlie them. No individual has even the remotest possibility of having enough time or cognitive resources to absorb or analyze all the information needed to behave in a fully objective, rational way towards other people. We assume things about other people all the time, often based on our own experiences and just as frequently based on how these other people look, talk and dress. To say that assumptions are discriminatory is to miss the point[2]. We possess and construct an array of constantly shifting assumptions about other people which are both generally true and also critically useful for humans trying to navigate everyday social life or order society.
The idea that assumptions about others are a necessary fact of life is not a novel concept to communitarians, who argue that people can (and should in some cases) be classified and categorized differentially based on their life histories and circumstances. In practice, these classifications are not static, one-dimensional or scrupulously defined. Obviously everyone has a myriad of dynamic associations and shifting characteristics which could be the basis of assumptions by others or even (as communitarians further reason) potentially conflicting moral obligations. Pushed to its logical limits, there is no escaping discriminatory profiling from a cognitive standpoint, which is an important realization for moral philosophers since something that is inevitable (as well as functionally necessary) can hardly be characterized as morally “wrong” or “unjust”[3], and even if we could agree on a situation where something inevitable was “unjust”, that would be a hollow conclusion since there would be nothing to be done about it.
Of course, stereotypes and assumptions can be inaccurate and not all stereotypes and assumptions are equal, nor should be treated with equal respect. From a moral standpoint, both how these assumptions are made and how they are acted upon in practice make all the difference between right and wrong. This idea of conditionality is actually fairly uncontroversial[4]: an assumption shared by moral philosophers with viewpoints as different as Aristotle, Mill, Locke and Rawls is that circumstances and intentions matter when it comes to judging an action or an institutional policy to be right or wrong. For Aristotle, the telos or purpose of the searching of bags by law enforcement officials would be of primary importance. If the primary purpose of the bag searches in question was to adhere to the principle of equal treatment of all individuals, then profiling would clearly be counterproductive and thus wrong from an Aristotelian perspective. If however the foremost intended role of the law enforcement officials and the bag-search policy is to efficiently and effectively prevent a suicide bombing, and if using profiling methods based on race or religion best achieves this purpose, then using these methods would be good or virtuous (and therefore right) in an Aristotelian sense. Clearly, the strength of this argument in favor of racial profiling will be closely coupled with the empirical effectiveness of profiling strategies. Since there would be no practical reason to enact or discuss profiling if it was demonstrably useless, we shall assume for the moment that profiling strategies are well-reasoned[5] and are effective in enabling law enforcement officials to more effectively and efficiently catch suicide bombers[6].
Aside from the empirical effectiveness of profiling, another practical but morally relevant piece of information is how exactly the bag-search policy would be articulated and carried out. Would there be disclaimers to carefully disassociate profiling from any official pejorative judgment against entire categories of people? Would the law enforcement officials treat the affected individuals with respect and courtesy? How inconvenient or embarrassing would the search procedures actually be for travelers? In the context of a given city or even an entire country, we can mostly agree that preventing a suicide bomb attack would maximize general utility within the society; additionally, it is largely true that misfortune strikes hardest on the least advantaged in society because they have fewer resources with which to cope or recover. With these two simple lines of logic we can deduce that if done in such a way as to minimize the inconvenience and embarrassment of those being searched, effectively preventing[7] a suicide bombing through searches guided by judicious profiling methods would satisfy both utilitarian thinkers as well as the Rawlsian Difference Principle, where social inequalities must be to the ultimate advantage of the least favored individuals in society. Moreover, building upon this line of reasoning, it would be natural that a policy that evidently maximizes utility while bettering the lot of the worse-off in society would gain broad popular support. From a Lockean paradigm of democratic society, this popular support would provide the critical element of consent that sets an important boundary on what the state can rightly do. In other words, if profiling turned out to be an effective tool for crafting thoughtful policies that would maximize social utility while also helping the least well-off and preserving individual dignity (thus garnering popular democratic support), there would be little reason to consider the use of profiling morally wrong.
Naturally, the moral opposition to profiling does not always stop with the simple complaint that it constitutes unfair discrimination[8]. It may be argued that having law enforcement officers treating people differently based on their outward appearance (race, ethnicity, religion) is a fundamental breach of the principle of equality and of individual rights and liberties (including those enshrined in the US Constitution).
As reasonable as this may seem at first, the objection based on equality and rights falters on two different grounds, one more practical and the other more philosophical. From a practical standpoint, the fact that individuals are undeniably unequal along almost infinite dimensions (culture, psyche, physicality, gender, age, occupation, marital status etc.) means that any attempt to treat everyone equally will quickly encounter the tension between the equally worthy ideals of equality of treatment and equality of outcome.[9] It is precisely because of the differences between individuals that equal treatment will not result in equal outcomes and conversely that equal outcomes can only be created through unequal treatment[10]. As such, it is at least an open question whether a call for “equality” can be logically coherent and achievable in practice[11]. Additionally, it is clear that society accepts all sorts of conditionally or selectively unequal treatment as just for exactly this reason. For example, only adults are allowed to vote, soldiers are beholden to military courts without juries (as opposed to civilian courts), citizens pay a progressive, graduated income tax, Indian reservations are exempt from many laws, there are mandatory provisions for the handicapped and so on. While there are diverse moral principles that can be cited to justify these instances of unequal treatment, the important point to note is that equality is a nebulous concept that in practice is hard to pin down to either equal treatment or equality of outcome, and ultimately the realization of equality is largely conditional and tempered by all sorts of practical considerations, just as I have argued is the correct way to morally evaluate profiling.
Tackling the more purely philosophical argument about the equal rights and liberties of citizens, it is again insufficient to simply say that profiling is a violation of individual rights. What exactly are these rights, where do these rights come from and why are they inviolable? As Locke would argue, individuals are bearers both of natural rights as well as of conventional rights shaped by popular consent in the formation of a state. At this point certain natural rights such as the right to revenge are “lost” or (more accurately) transformed and delegated to the state, for instance as the monopoly on coercion and the right to carry out retributive punishment. At the same time, the conventional rights are exactly those that are contingent both upon majority democratic support (however defined) as well as the exact situation of that society including its worldview, level of development, technology, demographics etc. In short, we are back to Locke’s insistence that democratic consent is the basis of the state’s rightful powers as well as the conventional rights of the citizen. Do citizens have the conventional right not to have their bags or persons searched when they use public transport or enter an airport, assuming that this is not a natural right? For Locke, this is a question for the people to answer through the democratic process, and I argue that this decision would likely turn on the practical effectiveness, description and execution of the proposed policy.
Beyond unfair discrimination, violation of rights to liberty and equal treatment, a final class of objections to the acceptability of profiling (with a very different philosophical pedigree) is that people should be treated as individuals and ends in themselves, and that it is wrong to force people to shoulder differential burdens only because they are perceived to belong to a certain community. This line of reasoning draws strongly upon Kantian philosophy, which envisions a level of homogeneity among people (the possession and practice of Reason) as well as a certain level of autonomy[12] that taken together call for all individuals to be treated with equal respect (as deserving moral agents) and as ends in themselves. From this view, search policies based on profiling supposedly only treat people as means towards the ends of the law enforcement officials without considering the individual’s feelings and humanity, and also robs individuals of their autonomy by forcing them to assume a certain (racial, ethnic, religious) identity.
I have already argued that treating others purely objectively without recourse to simplifying assumptions is merely a fanciful ideal, and may well be a dangerous delusion because it obscures the need to recognize and justify the inevitable assumptions. Kant was wrong to assume that the possession and practice of Reason would lead everyone to a universal morality, as the long-standing lack of agreement among moralists demonstrates. Additionally, the claim that search policies based on profiling will in fact treat persons merely as means to an end is clearly one that requires empirical support.[13] Finally, the argument that search policies based on profiling force individuals to bear the burdens of a certain identity rests on the mistaken assumptions that such policies necessarily define our identities, and that the burden of such search policies is unacceptably onerous, beyond what we can be reasonably asked to bear. Clearly, we must conclude that these are conditional arguments dependent on the specifics of the policy and its execution. There is no inherent moral reason to reject racial, ethnical or religious profiling.