Racial Profiling: Pro and Con

Eugene Volokh has a characteristically interesting post, on racial profiling, taking issue with an OpinionJournal criticism of the judge in lawsuit filed by an Arab-American passenger removed from a United Airlines flight “because the crew wasn’t comfortable with him aboard.”

Eugene points out that “the proper role of racial profiling in fighting terrorism is an interesting an difficult one,” but that in any event criticism of the judge is not called for. The problem, insofar as there is one, is with the applicable statute, 49 U.S.C. § 40127(a), which provides that an “air carrier . . . may not subject a person in air transportation to discrimination on the basis of race, color, national origin, religion, sex, or ancestry.” The statute does not, Eugene notes, have exceptions for emergencies or post 9/11 concerns about Arab-appearing passengers. In short, the text — until and unless it is liberally construed — is clear.

I think the problems raised by racial profiling are indeed difficult, and that we haven’t dealt very well with them. Like many others, I’ve been struck by the inconsistency of liberals who oppose racial profiling when done by police but applaud it when done by admissions officers or employers. This is not so surprising for a movement whose intellectuals tend to reject principles on principle, but in my view conservatives haven’t been much more coherent. Most conservatives these days are committed to color-blindness — not just in practice but in principle — and yet most also believe it is foolish not to subject Arabs/Muslims to added scrutiny in boarding airlines, etc. This inconsistency deserves more careful attention than it has received.

My very first post on this blog dealt, at least in passing, with racial profiling. In fact, one of the issues that compelled me to start the blog (other than daughter Jessie’s insistence) was dissatisfaction with a discussion of profiling by Mickey Kaus on kausfiles, and frustration that my email to him pointing out the error of his ways had been ignored. I intended to begin the blog with that issue, but got sidetracked by more current (by a week or so) concerns and never got back to it. Now I can. (All frustrated letter writers should start their own blog!)

Kaus wrote in his kausfiles of 6/12/2002 (I don’t know how to find/link his archived material, if indeed it is archived) that “even anti-liberals” opposed profiling, as if this were somehow unexpected, and he cited a comment by President Clinton in his well-known 1997 White House meeting with critics of affirmative action implying that conservatives were inconsistent on the matter. Here is my email to him of the same day:

Your comments about “even” the neo-cons etc. opposing racial profiling is thoughtful and “even”-handed, as usual, but wrong. You write that “if even anti-liberals were adamantly opposing profiling, it’s unfair to blame that opposition on liberals.” That is true as far as it goes, but your “even” implies there is something out of character or unexpected about “anti-liberal” opposition to racial profiling, an implication you repeat two sentences later when you explain the FBI’s behavior by noting that “Nobody was going to come forward, even from the right, to defend them if they got caught in a flight-school profiling scandal.”

Both the neo- and the paleo- cons are perfectly consistent in their principled opposition to racial profiling, wherever it occurs. It is, and for a generation now has been, the liberals who are inconsistent and even incoherent on the subject. When liberals approve of benefits or burdens being bestowed on the basis of race, they call it affirmative action. When they disapprove, they call it racial profiling.

Sometimes they confuse even themselves and call them both affirmative action. Thus, in his White House meeting with affirmative action critics that you cite, President Clinton actually described racial profiling by police as “a race-based public policy.” In fact, in the very paragraph you quote, you substitute ellipses for Clinton’s most telling comment, given below in bold:

CLINTON — [O]ver half of the cocaine in the country comes across the Mexican border. So, all right, fast forward. What do you do if you’re a local police officer with a drug problem? That’s what this whole profiling is about — [inaudible] people who are Hispanic if they’re driving through town. That’s an affirmative action program. That’s a race-based affirmative action program. [White House meeting was 12/19/97. Transcript was at www.whitehouse.gov, but I don’t know where it is, if anywhere, now.] [JSR, 3/19/17: Transcript is here.]

Perhaps 9/11 should force conservatives to compromise their principled opposition to all racial profiling, but at least they have a principle to compromise. That’s more than can be said for the liberals who oppose “race-based affirmative action” when done by the police but support it when done by admissions officers and employers.

I sent a copy of my Kaus email to some friends, hoping they could clear up my increasing discomfort about the inconsistency in the conservative position. One of them, a well-known scholar/critic of affirmative action (whose name I’m withholding because it was a private email not intended for distribution), replied:

It also strikes me that one can distinguish singling out a category of people for special scrutiny from judging people by different standards. I don’t know if anyone is saying that Arab males shouldn’t be allowed on planes–only that they be given a closer examination than other folks. The standard for denying them the right to fly ought to be the same one we use for anyone. Isn’t this different from saying that an LSAT of 160 and a GPA is good enough for admission to UVA Law School if you’re black but that others need much better credentials?

This is helpful — it is striving toward maintaining a single standard — but I still didn’t, and don’t, find it altogether satisfactory. I replied as follows (what good is having your own blog if you can’t quote yourself ad nauseum?):

In one sense this approach assimilates the profiling debate into the debate over disparate impact–applying a general standard of potential dangerousness to all travelers would leave little old ladies “underrepresented” and Arab males “overrepresented” in the pool of targets.

Still, I’m not sure this resolves our dilemma completely, at least not insofar as our general (?) standard of dangerousness contains religion/ethnicity as one of its components (as it must, it seems to me, to be realistic).

The nub of the issue, as usual, is whether/when it is legitimate for the state to be conscious or take account of race. One could say, as you suggest, that doing so is only proscribed when it leads to discrimination, and that added scrutiny of passengers involves no discrimination. This line may become quite fine, however, when applied to police surveillance, infiltration, etc.

Or one could apply an updated version of the old “bona fide occupational qualification” (bfoq) exception to Title VII — just as almost everyone would regard it as legitimate to take race into account in deciding which police officer to place undercover in a black gang, it’s hard to see how you can protect against young, male, Arab/Muslim terrorists without looking carefully at young, male Arab/Muslims, infiltrating Mosques, etc.

I think this is about where I am at present, but it does make me a bit uncomfortable because, by personality or temperament or ideology, I like firm principles, bright lines, clear rules, etc., and this is a breach in the dike of principle. In effect it allows that we CAN take race and religion into account if the reason is good enough. It leaves us with something like the high scrutiny (which I always thought was rather dumb)/compelling state interest/no alternative method standard that courts have articlulated before in this area. Since preferentialists have had some success making those very arguments with regard to preferential admission, I’ve always thought that “standard” too low, or porous, or something. But now I’m not sure we can do much better.

We seem to be balancing precariously on a point surrounded in every direction by slippery slopes. In this regard, in preparing for my letter to Mickey Kaus I re-read the transcript of [the affirmative action critics’] meeting with Clinton that he cited, and I was struck by some comments of Rep. Canady that call into serious question what I just said above:

MR. CANADY: Let me give you another example that’s directly related to the sort of thing you’re talking about. There are some people that contend that you have to take race into account in undercover police work. The theory is that you need black people to serve as undercover agents, and that’s the only effective way that can be carried out. Well, you know, it’s very interesting that the Drug Enforcement Administration actually had a policy of doing that. And the black drug enforcement agent sued the Drug Enforcement Agency over that very policy because it was discriminating against them. It was putting them in contexts where they were at greater risk, and it was limiting their opportunities for advancement. So what may start out as making some sense from one perspective can end up harming the very people that supposedly will benefit.

Of course, placing black agents in black gangs was not a policy designed to help the black agents but to penetrate the gangs, and I think is presumptively justified. Still, once this is accepted it’s harder to resist the Glenn Loury arguments about the necessity of black prison guards, the Gore arguments for black police that he made in the meeting, and even the legion of deans who argue that the fate of the republic rests on having a “critical mass” of minority lawyers.

I hate to leave this issue unresolved to my own satisfaction, but this is as far as I can take it at the moment.

For me this issue is still unresolved, and I’d appreciate some of you resolving it for me.

Say What? (19)

  1. Brendan A. Maher October 15, 2002 at 3:19 pm | | Reply

    You state “It leaves us with something like the high scrutiny (which I always thought was rather dumb)/compelling state interest/no alternative method standard that courts have articlulated before in this area. Since preferentialists have had some success making those very arguments with regard to preferential admission, I’ve always thought that “standard” too low, or porous, or something. But now I’m not sure we can do much better.”

    I’m not sure I can “resolve” this issue for you (or for anyone), but I do think that the “strict scrutiny” test, properly applied, can provide a principled basis for opposing affirmative action in admissions while supporting, to some extent, “profiling” in law enforcement.

    It is important to remember that, media hyperbole notwithstanding, “discrimination” on the basis of race is not ALWAYS illegal or “invidious.” Rather, to put the applicable test in everyday terms, if the government is going to discriminate on the basis of race, it better have a damn good reason to do so.

    In my opinion, when the test is properly and objectively applied to both race based admissions and racial profiling, the results are consistent. Although you correctly note that the “preferentialists” have had some success in arguing that “diversity” is a sufficiently compelling interest to allow race-based discrimination in admissions, their arguments are hollow, easily refuted and not supported by any real evidence. More importantly, the tide has started to turn in the courts, thanks to organizations like the Center for Individual Rights. Simply stated, there is no objectively quantifiable evidence that “diversity” is a compelling government interest and that discriminating AGAINST more qualified white or Asian applicants is the least restrictive means of serving that interest. In other words, race based affirmative action fails both prongs of the test.

    In stark contrast, racial profiling in law enforcement, when backed by actual evidentiary “profiles” of a certain type of criminal or crime, will almost always meet the “strict scrutiny” test.

    Limiting the discussion to the “War on Terror” yields the cleanest application of the test. First, it is unlikely that even the most left-wing politicians would argue that preventing another terrorist incident like 9/11 is not a “compelling government interest” (indeed, it can be plausibly argued that protecting the country is the MOST

    compelling government interest). Thus, the first prong of the “strict scrutiny” test is easily met.

    The question then becomes, “Is singling out young, male Arab-looking individuals the ‘least restrictive’ method of serving the ‘compelling government interest’ of preventing additional terrorist attacks?” Absent any additional factual information, the a principled “color blind” conservative would answer “no.”

    Fortunately, however, we have plenty of factual information to work with in the terrorism context. As we all know, virtually every one of the 9/11 hijackers was a young, male of Arab descent. Moreover, we know that al Qaeda planned and supported the attacks, and we know that, John Walker Lindh notwithstanding, virtually all members of al Qaeda are males of Arab descent. We also know that elderly white women, middle aged black men and teenage Hispanic girls have not been implicated in any terrorist plots and have not flown passenger jets into buildings.

    Thus, based on the factual information available, a plausible argument can be made that singling out young Arab-looking males is the “least restrictive” means of serving the “compelling government interest” of national security.

    It’s not perfect, of course, but I do think the test allows conservatives to be intellectually consistent.

  2. Kaimi Wenger October 15, 2002 at 4:43 pm | | Reply

    Nice post. And an interesting reply. I was actually thinking along similar lines yet coming to an opposite conclusion.

    The basic premise is the same: Categorizing by race is generally proscribed, but may be allowed where it is the only way to fix an important problem (I’m avoiding the term “strict scrutiny” here because that’s somewhat of a muddle recently, but it’s similar to what a principled strict scrutiny test would look like).

    Under that test, affirmative action as a general proposition passes because it is the only way to remedy harm done along racial lines, caused by centuries of state-sponsored discrimination. (Not all affirmative action programs might be adequately designed for this purpose, but ones which are could pass).

    Racial profiling, on the other hand, does not pass. It is use of racial categories in law enforcement, and other means could be used for the same result (better law enforcement). Adding more police, better technology, better techniques, and so on could provide the added protection from crime that racial profiling might claim to add. Since an alternative route exists, racial profiling is verboten.

  3. John Rosenberg October 15, 2002 at 9:57 pm | | Reply

    I’ve been driving most of the day, and am both bushed and on a borrowed computer. But I didn’t want to let the above two comments go unremarked. Or to put that another way, I find them both remarkably perceptive, and hence all the more fascinating that they both come, so reasonably, to opposite conclusions. I suspect I’ll have more to say on them both shortly, but right now I don’t know what it will be. Thus for now I’ll sign off simply by thanking both commenters for unusually thoughtful comments.

  4. Brendan A. Maher October 16, 2002 at 1:49 pm | | Reply

    A few more thoughts in response to Mr. Wenger’s post:

    (1) “Remedying past discrimination” is, in theory, a valid reason for using race-based affirmative action. In my opinion, however, not a single eighteen year old student who is applying for college today (i.e. who was born in, say, 1984-85) can advance that theory with a straight face. It has been illegal to discriminate on the basis of race in admissions (and in just about everything else) since the late 1960s. There is not a student alive who can claim that his or her admission to a particular university is necessary to remedy past discrimination. Nor is there a college or university that can plausibly claim that it NEEDS to discriminate NOW, to make up for its pre-1964 discrimination.

    Indeed, I have not seen a college or university use the “remedying past discrimination” justification for some time now. Instead, as I noted in my earlier post, the “compelling government interest” du jour is “diversity”, i.e. that a multi-colored student body is “necessary” to quality learning. The problem with this is that, to my knowledge, there is simply NO evidence WHATSOEVER that “diversity” of skin color adds anything to the higher education process (if such empirical evidence does exist, I’d love to see it). Moreover, the argument is undercut by its proponents’ insistence that Asians, Indians, Pakistanis and other similarly high-achieving minorities don’t count in the diversity calculus.

    So, I stand by my earlier assertion: There is no compelling government interest in using race as factor (whether overt or hidden) in the college or university admissions process.

    (2) With respect to racial profiling, the test is not whether there are ANY less restrictive means of achieving the compelling government interest, but whether the means chosen, set against the particular factual background at issue, are the LEAST RESTRICTIVE for achieving the government’s goal.

    While it is certainly true that, in a perfect world, we would have all the police and federal agents needed to effectively combat terrorism, and all of those police and agents would have the best training, weapons and other tools available, that is simply not the case. Police and federal agents are stretched thin, and there is not enough time, money or equipment to search every passenger on a plane or every car that crosses our borders.

    In light of that reality, the question is whether the police are entitled to use statistical probabilities to improve their odds of serving the compelling government interest of protecting American citizens from further terror attacks. In my view, this is a much closer call than using race as a factor in college admissions, and I believe that, under the right set of facts, the government could/should win a “racial profiling” challenge.

    Finally, the whole academic discussion of racial profiling in this context seems determined to avoid any use of common sense. It simply does not make any sense at all to use a random number system that results in strip searches of sitting Congressmen and old ladies at airports while allowing a certain number of young, Arab-looking males to pass through with minimal scrutiny.

    As a white male of noticeably Irish descent, I don’t pretend to know what it’s like to be singled out solely because of one’s skin color. I like to think, however, that if the Irish Republican Army was responsible for mass murder on U.S. soil, I wouldn’t object to being singled out for “special scrutiny” at airports.

  5. John Rosenberg October 16, 2002 at 10:55 pm | | Reply

    Once again I’m at the end of a long day of driviing (from Wilmington to Boston, through high wind and heavy rain all day), but wanted to post another “Amen,” or at least thanks, to Mr. Maher for his thoughtful comment. I think Kaimi Wenger is right in, by implication at least, endorsing compensation as a stronger argument for preferences than diversity. I suspect the reason the universities haven’t been arguing it lately is not because the moral appeal is weak but because the courts have pretty well slammed the door on that justification, while not having done so with diversity (yet?). That said, and following Mr. Maher’s point about current 18 yr olds, the weakest aspect of the compensation justification is its assumption that the harm done to blacks in the past can in fact be compensated by remedies given to other blacks today. This assumption in turn must be based on some notion of group rights, which sits uncomfortably if at all in our individual rights tradition.

  6. Kaimi Wenger October 17, 2002 at 7:55 pm | | Reply


    I have a new reply which I posted on my blog.

  7. John Rosenberg October 17, 2002 at 10:40 pm | | Reply

    Kaimi’s comment, referred to above, can be found on his blog at:


    Since Kaimi does not have comments implemented on his (otherwise!) excellent blog, I will comment on the above post here. Obviously, you should read Kaimi before proceeding.

    …but the harm [of past discrimination] has not been cured….

    No, it hasn’t, but it does not follow that perpetuating the primary cause of that harm, awarding benefits and burdens on the basis of race, is the best or even an appropriate method of curing it. There is also the often overlooked matter of identifying precisely the nature of the harm of past discrimination. It was not simply — and maybe not even primarily — that it denied a group of people benefits they otherwise would, and should, have enjoyed. Perhaps even more fundamental was that it violated our core value that every individual should be treated without regard to race, religion, national origin. That damage was inflicted on our whole society, not just the racial victims. It cannot be cured by perpetuating it.

    Consider the following thought experiment: For generations Jews were excluded from or their admission severely restricted at Ivy League and other elite colleges and professional schools. As a result, for many years there were fewer, say, Jewish doctors and lawyers than there would have been absent the discrimination. Is it appropriate now, or would it have been appropriate in, say, 1940 or 1950 or 1960 for Ivy League professional schools to give preferences in admissions to Jews? I don’t think so.

    If more black poverty is thought to be the primary result of past discrimination, then implement policies to combat poverty. But there is no group right not to be “overrepresented” among the poor, and so the aim of anti-poverty programs should not be to equalize poverty rates among different groups.

    Again, consider: in a free society, there will be a hierarchy of success. If discrimination on the basis of race were not wrong per se, on its face, there would be nothing wrong with those clustered at the bottom being members of one racial or ethnic group. But we do think that’s wrong, and it’s because of the core value mentioned above. Violate the core value, and you’ve undermined the very principle that makes the status quo you want to correct wrong.

    For example, many universities have preferences in place for alumni relatives.

    I have written on several occasions about what, in a recent example, I called “The Fallacy of Fungible Discriminations.”

    See here and here.

    I would think that diversity is a good thing…

    I do too, but skin color is not a good proxy for it. It is also overrated, in my opinion, as a justification for discriminatory admissions policies at universities that contain, say, engineering, math, veterinary science, etc., where even legitimate intellectual diversity would seem to be of limited value.

    … where a racial categorization could lead to removal of personal freedom [i.e., racial profiling], a higher standard should be used.

    Interesting point, but racial profiling itself does not produce a loss of personal freedom (except the temporary loss of freedom of movement caused by the added scrutiny at a boarding gate or being pulled over or stopped on the street by a cop). What causes the loss of personal freedom is what results from the added scrutiny entailed by racial profiling. The question is whether the added scrutiny of racial profiling is the equivalent of an illegal search, where its fruits should be suppressed, or a warranted (because the individual matched the profile) search. For what it’s worth, I myself am ambivalent here.

    Finally, I second Kaimi’s point about the civility of this discussion, which is a welcome relief. For a similar one, I recommend the too-numerous-to-link exchanges early last summer I had over disparate impact with Garret Moritz of gTexts.

  8. Lindsey October 31, 2002 at 9:54 am | | Reply

    I am doing a school project at my high school. I need to do a report supporting racial profiling, while still showing the negative side to it as well. if anyone has any information, could you please email me.

    Thank you

  9. Bill Jumbo December 6, 2002 at 3:21 pm | | Reply

    your site is fucking junk. I did not get one fucking piece of info that I needed so you can all go to hell

  10. Lance Heidn December 6, 2002 at 3:24 pm | | Reply

    Your site is shit. You are all a bunch of butt fucks. I hope you all rot in hell and next time give me some better fucking info, you fucking fucks

  11. Jacob December 11, 2002 at 10:15 pm | | Reply

    Guys, you idiots comin on here and starting shit is stupid. You don’t need to come on the site and do that. It isn’t very cool.

  12. Ed December 27, 2002 at 6:48 am | | Reply

    Ignore the last three posts, those people are idiots. This site is very interesting, and both sides of “racial profiling” make good points…

  13. Scott December 27, 2002 at 6:56 am | | Reply

    I agree with Ed. You can argue both sides of “racial profiling” and any good thing can be bad if you abuse it.

    I personally don’t see anything wrong with the police using race as a factor in keeping America safe. However, I do see a problem when a university or employer discriminates against a higher qualified individual based solely on the color of his or her skin, and then tries to call it “affirmitive action.”

  14. BOB January 29, 2003 at 11:01 am | | Reply

    i read some of the quotes. As i before a studen doing a report on the pro’s. I to would like to know “who” is saying the it is justified. Personally I do think that it is justified to a degree. IT IS UNFAIR, but some what justifiable to the exstent of providing some level of confort and safty.

    As in the Middle East, they have not got a whole lot secess in stopping the waves of terror. So do we racial profile or wait to die

  15. Ryan December 11, 2003 at 6:58 am | | Reply

    I would like to say that Brendan Maher is as on point as one can get. I am currently writing a paper at my very liberal college (U of Mich) supporting the practices of racial / religous / ethnic profiling. I couldn’t agree more with you that racial profiling passes both prongs of the test the way you have stated it. However have not ever heard about factual information being used. My understanding was that a law or procedure including a racial classification is narrowly tailored to serve that purpose only if there is no race-neutral alternative that can do the job equally as well. That would lead me to say that racial profiling fails the test. But with your interpretation that the test is not whether there are ANY less restrictive means of achieving the compelling government interest, but whether the means chosen, set against the particular factual background at issue, are the LEAST RESTRICTIVE for achieving the government’s goal, I am left agreeing.

    Can you clear this up?

  16. Anonymous March 2, 2005 at 11:23 am | | Reply


  17. Anonymous March 2, 2005 at 11:30 am | | Reply


  18. Anonymous March 2, 2005 at 11:31 am | | Reply


  19. […] are called upon to assign an undercover officer to a black or white gang (something I discussed here, here, and […]

Say What?