Flying while Arab: lessons from the racial profiling controversy

In the aftermath of the September 11 tragedies in New York and Washington, DC, we Americans have heard countless times that our country has "changed forever." In many ways, especially in terms of national and personal security, this is quite true. Americans have always assumed that terrorism and other violent manifestations of the world's problems did not and would never happen here, that our geographic isolation by the Atlantic and Pacific Oceans protected us. Indeed, since the Civil War, the United States has experienced no sustained violence or war on its own soil. Sadly, we know now that we are vulnerable, and that like countries all over the world, we must take steps to protect ourselves.

This is the new reality that Americans find themselves adjusting to: searches and inspections of ourselves and our belongings when we enter public buildings and areas, such as government offices, sports stadiums, and airport concourses; increased presence of law enforcement and even military personnel; enhanced police powers and curtailed civil liberties; and new powers and tactics our government will use to deal more strictly with foreigners and immigrants. While some of these changes amount to little more than inconveniences, others--particularly changes in the law that limit individual freedom while expanding government power--are in fact major changes in our way of life and the core values and meaning of American society. The U.S. Congress has already passed a sweeping piece of legislation, increasing government power over everything from wiretaps, e-mail, formerly secret grand jury information, to the detention and trial of noncitizens.

We know that the United States is a nation of immigrants--that, in many ways, immigrants built our great nation. We know that the immigrant experience has, in many ways, been at the core of the American experience, along with the experiences of African Americans liberated from slavery. The diversity and energy that immigrants have brought to our country has been, and continues to be, one of our greatest strengths. But, we also know that we have sometimes dealt harshly and unfairly with immigrants and noncitizen residents, especially in times of national emergency and crisis. Thus, it is critical that we try to understand the implications of the changes that have taken place and will continue because of the events of September 11--changes in the very idea of what America is, and in what it will be in the future.

One of these changes has been particularly noticeable--both because it represents a radical shift in what we did prior to September 11, and because it also continues a public discussion that was taking place in our country before that terrible day. Racial profiling--the use of race or ethnic appearance as a factor in deciding who merits police attention as a suspicious person--has undergone a sudden and almost complete rehabilitation. Prior to September 11, many Americans had recognized racial profiling for what it is--a form of institutional discrimination that had gone unquestioned for too long. Thirteen states had passed anti-profiling bills of one type or another, and hundreds of police departments around the country had begun to collect data on all traffic stops, in order to facilitate better, unbiased practices. On the federal level, Congressman John Conyers, Jr., of Michigan and Senator Russell Feingold of Wisconsin had introduced the End Racial Profiling Act of 2001, a bill aimed at directly confronting and reducing racially biased traffic stops through a comprehensive, management-based, carrot-and-stick approach.

September 11 dramatically recast the issue of racial profiling. Suddenly, racial profiling was not a discredited law enforcement tactic that alienated and injured citizens while it did little to combat crime and drugs; instead, it became a vital tool to assure national security, especially in airports. The public discussion regarding the targets of profiling changed too--from African Americans, Latinos, and other minorities suspected of domestic crime, especially drug crime, to Arab Americans, Muslims, and others of Middle Eastern origin, who looked like the suicidal hijackers of September 11. In some respects, this was not hard to understand. The September 11 attacks had caused catastrophic damage and loss of life among innocent civilians; people were shocked, stunned, and afraid. And they knew that all of the hijackers were Arab or Middle Eastern men carrying out the deadly threats of Osama bin Laden's al Qaeda terrorist network based in the Middle East, which of course claims Islam as its justification for the attacks and many others around the world. Therefore, many said that it just makes sense to profile people who looked Arab, Muslim, or Middle Eastern. After all, "they" were the ones who'd carried out the attacks and continued to threaten us; ignoring these facts amounted to some kind of political correctness run amok in a time of great danger.

But if the renewed respectability and use of profiling was one of the ways in which September 11 changed things, we might also notice that the "new" racial profiling demonstrated the truth of an old saw: the more things change, the more they stay the same. We should remember that racial profiling of African Americans and Latinos also originated in a war--the metaphorical "war on drugs"--and was justified with the same arguments. But even more importantly, we should learn from what we now know were the grand mistakes of profiling in the last 10 years. If we do that, we will see that using Arab or Muslim background or appearance to profile for potential terrorists will almost certainly fail--even as it damages our enforcement efforts and our capacity to collect intelligence.

History

As in almost any serious policy inquiry, a look at the history of our country can help us attain a proper perspective on how to view what we do now. Unfortunately, that history gives us reasons to feel concern at this critical juncture. Any serious appraisal of American history during some of the key periods of the 20th century would counsel an abundance of caution; when we have faced other national security crises, we have sometimes overreacted--or at the very least acted more out of emotion than was wise.

In the wake of World War I, the infamous Palmer Raids resulted in the rounding up of a considerable number of immigrants. These people were deported, often without so much as a scintilla of evidence. During World War II, tens of thousands of Japanese--immigrants and native born, citizens and legal residents--were interned in camps, their property confiscated and sold off at fire-sale prices. To its everlasting shame, the U.S. Supreme Court gave the internment of the Japanese its constitutional blessing in the infamous Korematsu case. It took the United States government decades, but eventually it apologized and paid reparations to the Japanese. And during the 1950s, the Red Scare resulted in the mining of lives and careers and the jailing of citizens, because they had had the temerity to exercise their constitutionally protected rights to free association by becoming members of the Communist Party years or even decades before.

Hopefully, we can see the common thread that runs through these now notorious examples: an apprehension of danger to the country not only from the outside but from a group of people within who are identified racially, ethnically, or politically with those thought to pose the threat, and a willingness to take measures that sweep widely through the identified group--more widely than the threat might justify. (Of course, we have also learned that these threats have been wildly exaggerated; for example, the discovery of government documents more than four decades after the internment of the Japanese showed that the government misled the courts by intentionally withholding critical information that contradicted official efforts to make the case for a sufficiently severe threat to justify the internment. (1)) The threat we face now bears many similarities: a danger from overseas posed by one group, and an identified group in the United States that has come under suspicion. All of this ought to encourage us not to leap forward with racial or ethnic profiling, but to hesitate before we do.

Categorical Thinking

We must hope that we have learned the lessons of this history--that the emotions of the moment, when we feel threatened, can cause us to damage our civil liberties and our fellow citizens, particularly our immigrant populations. And it is this legacy that should make us think now, even as we engage in a long and detailed investigation of the September 11 terror attacks. As we listen to accounts of that investigation, reports indicate that the investigation has been strongly focused on Arab Americans and Muslims. What's more, private citizens have made Middle Eastern appearance an important criterion in deciding how to react to those who look different around them. Many of these reports have involved treatment of persons of Middle Eastern descent in airports.

In itself, this is not really surprising. We face a situation in which there has been a terrorist attack by a small group of suicidal hijackers, and as far as we know, all of those involved were Arabs and Muslims and had Arabic surnames. Some or all had entered the country recently. Given the incredibly high stakes, some Americans have reacted to Middle Easterners as a group, based on their appearance. In a way, this is understandable. We seldom have much information on any of the strangers around us, so we tend to think in broad categories like race and gender. When human beings experience fear, it is a natural reaction to make judgments concerning our safety based on these broad categories, and to avoid those who arouse fear in us. This may translate easily into a type of racial and ethnic profiling, in which--as has been reported--passengers on airliners refuse to fly with other passengers who have a Middle Eastern appearance.

Use of Race and Ethnic Appearance in Law Enforcement

The far more worrying development, however, is the possibility that profiling of Arabs and Muslims will become standard procedure in law enforcement. Again, it is not hard to understand the impulse; we want to catch and stop these suicidal hijackers, every one of whom fits the description of Arab or Muslim. So we stop, question, and search more of these people because we believe it's a way to play the odds. If all the September 11 terrorists were Middle Easterners, then we get the biggest bang for the enforcement buck by questioning, searching, and screening as many Middle Easterners as possible. This should, we think, give us the best chance of finding those who helped the terrorists or those bent on creating further havoc.

But we need to be conscious of some of the things that we have learned over the last few years in the ongoing racial profiling controversy. Using race or ethnic appearance as part of a description of particular suspects may indeed help an investigation; using race or ethnic appearance as a broad predictor of who is involved in crime or terrorism will likely hurt our investigative efforts. All the evidence indicates that profiling Arab Americans or Muslims would be an ineffective waste of law enforcement resources that would damage our intelligence efforts while it compromises basic civil liberties. If we want to do everything we can to secure our country, we have to be smart about the steps we take.

As we think about the possible profiling of Arabs and Muslims, recall the arguments made for years about domestic efforts against drugs and crime. African Americans and Latinos are disproportionately involved in drug crime, proponents of profiling said; therefore concentrate on them. Many state and local police agencies, led by the federal Drug Enforcement Administration, did exactly that from the late 1980s on. We now know that police departments in many jurisdictions used racial profiling, especially in efforts to get drugs and guns off the highways and out of the cities. For example, state police in Maryland used a profile on Interstate 95 during the 1990s in an effort to apprehend drug couriers. According to data from the state police themselves, while only 17 percent of the drivers on the highway were African American, over 70 percent of those stopped and searched were black. Statistics from New Jersey, New York, and other jurisdictions showed similar patterns: the only factor that predicted who police stopped and searched was race or ethnicity. (2) No other factor--not driving behavior, not the crime rate of an area or neighborhood, and not reported crimes that involved persons of particular racial or ethnic groups--explained the outcomes that showed great racial or ethnic disproportionalities among those stopped and searched.

But as we look back, what really stands out is how ineffective this profile-based law enforcement was. If proponents of profiling were right--that police should concentrate on minorities because criminals were disproportionately minorities--focusing on "those people" should yield better returns on the investment of law enforcement resources in crime fighting than traditional policing does. In other words, using profiles that include racial and ethnic appearance should succeed more often than enforcement based on other, less sophisticated techniques. In any event, it should not succeed less often than traditional policing. But in fact, in departments that focused on African Americans, Latinos, and other minorities, the "hit rates"--the rates of searches that succeeded in finding contraband like drugs or guns--were actually lower for minorities than were the hit rates for whites, who of course were not apprehended by using a racial or ethnic profile. That's right: when police agencies used race or ethnic appearance as a factor--not as the only factor but one factor among many--they did not get the higher returns on their enforcement efforts that they were expecting. Instead, they did not do as well; their use of traditional police methods against whites did a better job than racial profiling, and did not sweep a high number of innocent people into law enforcement's net.

The reason that this happened is subtle but important: race and ethnic appearance are very poor predictors of behavior. Race and ethnicity describe people well, and there is absolutely nothing wrong with using skin color or other features to describe known suspects. But since only a very small percentage of African Americans and Latinos participate in the drug trade, race and ethnic appearance do a bad job identifying the particular African Americans and Latinos in whom police should be interested. Racial and ethnic profiling caused police to spread their enforcement activities far too widely and indiscriminately. The results of this misguided effort have been disastrous for law enforcement. This treatment has alienated African Americans, Latinos, and other minorities from the police--a critical strategic loss in the fight against crime, since police can only win this fight if they have the full cooperation and support of those they serve. And it is precisely this lesson we ought to think about now, as the cry goes up to use profiling and intensive searches against people who look Arab, Middle Eastern, or Muslim.

Profiling to Catch Terrorists

Using race, ethnic appearance, or religion as a way to decide who to regard as a potential terrorist will almost surely produce the same kinds of results: no effect on terrorist activity; many innocent people treated like suspects; damage to our enforcement and prevention efforts.

Even if the suicide hijackers of September 11 shared a particular ethnic appearance or background, subjecting all Middle Easterners to intrusive questioning, stops, or searches will have a perverse and unexpected effect: it will spread our enforcement and detection efforts over a huge pool of people who police would not otherwise think worthy of attention. The vast majority of people who look like Mohammed Atta and the other hijackers will never have anything to do with any kind of ethnic or religious extremism. Yet a profile that includes race, ethnicity, or religion may well include them, drawing them into the universe of people who law enforcement will stop, question, and search. Almost all of them will be people who would not otherwise have attracted police attention, because no other aspect of their behavior would have drawn scrutiny. Profiling will thus drain enforcement efforts and resources away from more worthy investigative efforts and tactics that focus on the close observation of behavior--like the buying of expensive one-way tickets with cash just a short time before takeoff, as some of the World Trade Center hijackers did.

This has several important implications. First, just as happened with African Americans and Latinos in the war on drags, profiling of Arabs and Muslims will be overinclusive--it will put many more under police suspicion of terrorist activity than would otherwise be warranted. Almost all of these people will be hard-working, tax-paying, law-abiding individuals. While they might understand one such stop to be a mere inconvenience that they must put up with for the sake of national security, repetition of these experiences for large numbers of people within the same ethnic groups will lead to resentment, alienation, and anger at the authorities.

Second, and perhaps more important, focusing on race and ethnicity keeps police attention on a set of surface details that tells us very little and draws officers' attention away from what is much more important and concrete: behavior. The two most important tools law enforcement agents have in preventing crime and catching criminals are observation of behavior and intelligence. As any experienced police officer knows, what's important in understanding who's up to no good is not what people look like, but what they do. Investigating people who "look suspicious" will often lead officers down the wrong path; the key to success is to observe behavior. Anyone who simply looks different may seem strange or suspicious to the untrained eye; the veteran law enforcement officer knows that suspicious behavior is what really should attract attention and investigation. Thus focusing on those who "look suspicious" will necessarily take police attention away from those who act suspicious. Even in the current climate, in which we want to do everything possible to prevent another attack and to apprehend those who destroyed the World Trade Center and damaged the Pentagon, law enforcement resources are not infinite. We Americans must make decisions on how we run our criminal investigation and prevention efforts that move us away from doing just anything, and toward doing what is most effective.

Third, if observation of suspicious behavior is one of law enforcement's two important tools, using profiles of Arabs, Muslims, and other Middle Easterners can damage our capacity to make use of the other tool: the gathering, analysis, and use of intelligence. There is nothing exotic about intelligence; it simply means information that can be useful in crime fighting. If we are concerned about terrorists of Middle Eastern origin, among the most fertile places from which to gather intelligence will be the Arab American and Muslim communities. If we adopt a security policy that stigmatizes every member of these groups in airports and other public places with intrusive stops, questioning, and searches, we will alienate them from the enforcement efforts at precisely the time we need them most. And the larger the population we subject to this treatment, the greater the total amount of damage we inflict on law-abiding persons.

And of course the profiling of Arabs and Muslims assumes that we need worry about only one type of terrorist. We must not forget that, prior to the attacks on September 11, the most deadly terrorist attack on American soil was carried out not by Middle Easterners with Arabic names and accents, but by two very average American white men: Timothy McVeigh, a U.S. Army veteran from upstate New York, and Terry Nichols, a farmer from Michigan. Yet we were smart enough in the wake of McVeigh and Nichols' crime not to call for a profile emphasizing the fact that the perpetrators were white males. The unhappy truth is that we just don't know what the next group of terrorists might look like.

Race or Ethnicity As Just One Factor Among Many?

In many discussions of profiling, the question some raise is not whether to use race or ethnic appearance, but how. Proponents and defenders of racial and ethnic profiling have argued that profiling would be both acceptable and effective if race or ethnic appearance was not the only factor that indicated suspicion, but just one factor among many. The idea is that race and ethnic appearance should never be the only factors that prompt suspicion, but could be useful if they are part of the whole picture that also includes behavior. Are there, in fact, conditions under which it might make sense to treat people differently according to their race or ethnic appearance, as long as it is just one factor among many?

Our prior experience with profiling counsels against this approach. Despite what many believe, racial profiling has almost never involved situations in which police used race as the only factor in deciding which drivers or pedestrians to stop. In fact, it would be surprising if this were ever true. Human motivation is far too complex in any given situation to be based on one fact; moreover, even the thickest, most bigoted member of a police organization would know better than to simply stop people based on race. And the numbers of drivers and pedestrians in the world would make this impossible anyway; as Justice Robert Jackson said many years ago, when he was the attorney general of the U.S., traffic laws and violators of those laws are so numerous that police must inevitably choose between violators when deciding against whom to enforce the law.

But even if race or ethnicity is just one factor among others, it still presents dangers. Using race or ethnicity for purposes other than describing a particular suspect or suspects means that we must accept that race or ethnicity can become the dominant or most important factor among all of the others. And since people remain likely to attribute suspicion to those different from themselves in the broad categorical ways discussed earlier, we end up with race or ethnicity not just as an additional, sharpening factor as we focus suspicion, but as the factor that for all practical purposes directs our actions as we decide who to stop, question, and search. This, of course, brings us back to the pillars of traditional policing: race or ethnic appearance may be a valuable descriptor, but it is not behavior. It tells us nothing about what people do or have done, and instead distracts us from observing behavior.

Second, we cannot discount the obvious skill and determination of the adversaries we face in this struggle. The September 11 attacks made dear that the al Qaeda terrorists were not wild, unguided fanatics. Rather they showed a high degree of intelligence and cunning, spotting and taking advantage of unnoticed weaknesses in our immigration and aviation security systems. They showed the ability and the patience for long-range planning and careful action, as well as strict self-discipline. All of this is, of course, in addition to a belief in their own cause so strong that they were willing to sacrifice their own lives to attain their goals. And we cannot forget that the attack on the World Trade Center on September 11 was not the first, but the second attempt to destroy those buildings; their first attempt, in 1993, was unsuccessful, and they watched, waited, and planned for eight years to try again. With enemies of such craftiness and determination, it seems extremely unlikely that they will use people for their next attack who look like exactly what we are looking for. Rather, they will shift to light-skinned people who look less like Arabs or Middle Easterners, without Arabic names, or to people who are not Middle Easterners at all, such as individuals from African nations or the Philippines. (In both places, there are significant numbers of Muslims, a small but significant number of whom have been radicalized.) This, of course, will put us back where we started, and racial or ethnic appearance will become a longest-of-long-shot, almost certainly an ineffective predictor at best, and a damaging distracting factor at worst.

Conclusion

The terrorist attacks in New York and Washington, DC, present us with many difficult choices that will test us. We will have to ask ourselves deep questions: Who are we, as a nation? What is important to us? What values lay at the core of our Constitution and our democracy? How will we find effective ways to secure ourselves without giving up what is best about our country? The proper balance between safety and civil rights will sometimes be difficult to see. But we should not simply repeat the mistakes of the past as we take on this new challenge. Only our adversaries would gain from that.

References

(1) See Korematsu v. United States 584 F. Supp. 1406 (N.D. Cal. 1984); Hirabayashi v. United States, 627 E Supp. 1445 (W.D. Wash. 1986), aff'd in part and rev'd in part, 828 F.2d 591 (9th Circuit, 1987).

(2) For Maryland numbers, see John Lamberth, testimony before the Congressional Black Caucus, 1998, accessed at www.lamberthconsulting.com/downloads/cbc_presentation.doc; see also Wilkins v. Maryland State Police, No. CCB-93-468 (order of Apr. 22, 1997) and Maryland State Conference of NAACP Branches, et al. v. Maryland Department of State Polic, et al., 72 F.Supp 2d 560 (September 1999). For New Jersey, see John Lamberth, "Revised Statistical Analysis of the Incidence of Police Stops and Arrests of Black Drivers/Travelers on the New Jersey Turnpike Between Exits or Interchanges 1 and 3 from the Years 1988 through 1991." November 1994, accessed at www.lamberthconsulting.com/research_articles.asp. For New York, see Eliot Spitzer, Attorney General of the State of New York, "The New York Police Department's 'Stop and Frisk' Practices: A Report to the People of the State of New York," 1999, accessed at www.oag.state.ny.us/press/reports/stop_frisk/stp_frsk.pdf.

DAVID HARRIS IS BALK PROFESSOR OF LAW AND VALUES AT THE UNIVERSITY OF TOLEDO COLLEGE OF LAW AND SOROS SENIOR JUSTICE FELLOW. HE IS THE AUTHOR OF PROFILES IN INJUSTICE: WHY RACIAL PROFILING CANNOT WORK, THE NEW PRESS, 2002.

COPYRIGHT 2002 U.S. Commission on Civil Rights
COPYRIGHT 2003 Gale Group