The Profile

In the comments to this post, Mark Olson and I had a long debate on (among other things) whether racial profiling was just. He said that the only relevant consideration was whether the profile was accurate--that is, does it accurately describe (and thus catch) the people most likely to be committing crimes. I argued that this seems to overlook that criminal organizations can create "anti-profiles" in response to profiling, making the locus of criminal activity simply shift from group to group. In addition, even if the original instinct is right, I gave five additional reasons why I think racial profiling is wrong:

1) I think the distributional burdens outweigh the prospective gains: treating innocent Blacks as if they were criminal is a big harm, and I don’t think it’s outweighed simply by catching more (Black) criminals.

2) It’s self-reinforcing: profiling means we’ll catch more Black criminals vis-a-vis White ones, which will enhance the perceived disparity in crime rates, which will justify more profiling.

3) It tends to exceed its own justificatory boundaries: The rates at which Black people are profiled (i.e., more likely to be pulled over than Whites) tend to exceed the level that they are more likely to commit crimes. Racial profiling lends itself to an exaggeration of the perception of Black criminality, leading to the program exploding beyond any reasonable link to the actual disparities in crime rates.

4) It makes it easier for Whites to commit crimes when police resources are specifically (and publicly) diverted away from looking at them, so the overall anti-crime effect may be marginal.

5) It’s self-defeating: Police profiling of Blacks makes Blacks (rightfully) view the police as their enemy, fueling urban tensions and making it more difficult for serious crimes to be solved.

While doing research on an unrelated matter, I found at least some evidence to support point three, which is the most direct response to Mark's claim. Clarence Lusane points out that "Federal surveys show that blacks use drugs at the same rate as whites, and there is no evidence that blacks are more involved in drug trafficking than whites, yet . . . . in at least 30 major cities, blacks are ten times more likely to be arrested for drugs than whites." [Clarence Lusane, In Perpetual Motion: The Continuing Significance of Race and America's Drug Crisis, 1994 U. Chi. Legal F. 83, 146]. Admittedly, this information is somewhat dated, but I've seen and heard no evidence that relative Black drug use (or trafficking) has increased since 1994, or that their arrest rates have gone down.

But one thing both of us assumed is that the "profile" is a stable subject. That is, when the police say they searched or arrested someone because he or she "fit the profile", they are talking about some reasonably static nexus of characteristics and behaviors. A significant portion of Mark's defense of profiling is dependent on this, because he tries to dodge the distributional justice concerns I raise by simply arguing that people can act differently (when I raised the wealthy Black man who gets pulled over in his sports car whenever he doesn't wear nice clothes, Mark said the guy should either wear a suit or buy a Prius). But this post by Columbia Law Professor Michael Dorf seems to put a dagger in that assumption. Far from being a meticulously constructed simulacrum of criminality, some judges have observed that "profile" can be a shorthand for literally any behavior.

[A] canvass of numerous cases reveals the drug courier profile's "chameleon-like way of adapting to any particular set of observations." United States v. Sokolow, 831 F.2d 1413, 1418 (9th Cir.1987), rev'd, 490 U.S. 1, 109 S.Ct. 1581, 104 L.Ed.2d 1 (1989):

Arrived late at night United States v. Nurse, 916 F.2d 20, 24 (D.C.Cir.1990).

Arrived early in the morning United States v. Reid, 448 U.S. 438, 441, 100 S.Ct. 2752, 2754, 65 L.Ed.2d 890 (1980); United States v. Millan, 912 F.2d 1014, 1017 (8th Cir.1990).

One of first to deplane United States v. Millan, 912 F.2d at 1015; United States v. Moore, 675 F.2d 802, 803 (6th Cir.1982), cert. denied, 460 U.S. 1068, 103 S.Ct. 1521, 75 L.Ed.2d 945 (1983).

One of last to deplane United States v. Mendenhall, 446 U.S. 544, 547 n. 1, 100 S.Ct. 1870, 1873 n. 1, 64 L.Ed.2d 497 (1980); United States v. Sterling, 909 F.2d 1078, 1079 (7th Cir.1990); United States v. White, 890 F.2d 1413, 1414 (8th Cir.1989), cert. denied, 498 U.S. 825, 111 S.Ct. 77, 112 L.Ed.2d 50 (1990).

Deplaned in the middle United States v. Buenaventura-Ariza, 615 F.2d 29, 31 (2d Cir.1980).

Used a one-way ticket United States v. Johnson, 910 F.2d 1506 (7th Cir.1990), cert. denied, 498 U.S. 1051, 111 S.Ct. 764, 112 L.Ed.2d 783 (1991); United States v. Colyer, 878 F.2d 469, 471 (D.C.Cir.1989); United States v. Sullivan, 625 F.2d 9, 12 (4th Cir.1980).

Used a round-trip ticket United States v. Craemer, 555 F.2d 594, 595 (6th Cir.1977).

Carried brand-new luggage United States v. Taylor, 917 F.2d at 1403; United States v. Sullivan, 625 F.2d at 12.

Carried a small gym bag United States v. Sanford, 658 F.2d 342, 343 (5th Cir.1981), cert. denied, 455 U.S. 991, 102 S.Ct. 1618, 71 L.Ed.2d 852 (1982).

Travelled alone United States v. White, 890 F.2d at 1415; United States v. Smith, 574 F.2d 882, 883 (6th Cir.1978).

Travelled with a companion United States v. Garcia, 905 F.2d 557, 559 (1st Cir.), cert. denied, 498 U.S. 986, 111 S.Ct. 522, 112 L.Ed.2d 533 (1990); United States v. Fry, 622 F.2d 1218, 1219 (5th Cir.1980).

Acted too nervous United States v. Montilla, 928 F.2d 583, 585 (2d Cir.1991); United States v. Cooke, 915 F.2d 250, 251 (6th Cir.1990).

Acted too calm United States v. McKines, 933 F.2d 1412 (8th Cir.1991); United States v. Himmelwright, 551 F.2d 991, 992 (5th Cir.), cert. denied, 434 U.S. 902, 98 S.Ct. 298, 54 L.Ed.2d 189 (1977).

Wore expensive clothing and gold jewelry United States v. Chambers, 918 F.2d 1455, 1462 (9th Cir.1990).

Dressed in black corduroys, white pullover shirt, loafers without socks United States v. McKines, supra.

Dressed in dark slacks, work shirt, and hat United States v. Taylor, 917 F.2d at 1403.

Dressed in brown leather aviator jacket, gold chain, hair down to shoulders United States v. Millan, 912 F.2d at 1015.

Dressed in loose-fitting sweatshirt and denim jacket United States v. Flowers, 909 F.2d 145, 146 (6th Cir.1990).

Walked rapidly through airport United States v. Millan, 912 F.2d at 1017; United States v. Rose, 889 F.2d 1490, 1491 (6th Cir.1989).

Walked aimlessly through airport United States v. Gomez-Norena, 908 F.2d 497, 497 (9th Cir.1990), cert. denied, 498 U.S. 947, 111 S.Ct. 363, 112 L.Ed.2d 326 (1991).

Flew in to Washington National Airport on the LaGuardia Shuttle United States v. Powell, 886 F.2d 81, 82 (4th Cir.1989), cert. denied, 493 U.S. 1084, 110 S.Ct. 1144, 107 L.Ed.2d 1049 (1990).

Had a white handkerchief in his hand United States v. Garcia, 848 F.2d 58, 59 (4th Cir.), cert. denied, 488 U.S. 957, 109 S.Ct. 395, 102 L.Ed.2d 384 (1988).

In our "Looking-Glass" world of drug enforcement, the DEA apparently seeks "to be master" by having "drug courier profile" mean, like a word means to Humpty Dumpty, "just what I choose it to mean--neither more nor less." [United States v. Hooper, 935 F.2d 484 (1991) (Pratt, J., dissenting)]

In the same case, the DEA also justified its search by saying that the suspect came from a "source city"--which, under questioning, was conceded to be "virtually any city with a major airport." In other words, the "profile" of a drug courier can be asserted to include any behavior, indicating that the entire discourse is pre-textual.

Dissenting in U.S. v. Sokolow, 490 U.S. 1, 13-14 (1989), Justice Thurgood Marshall raised similar concerns, and added that reliance on formulaic profiles could dull an officer's instincts for identifying truly suspicious behavior--an ability that undergirds the entire line of cases proceeding from Terry v. Ohio that allows officers to conduct searches in the absence of evidence that would normally rise to the level of "probable cause."

In my view, a law enforcement officer's mechanistic application of a formula of personal and behavioral traits in deciding whom to detain can only dull the officer's ability and determination to make sensitive and fact-specific inferences “in light of his experience,” Terry [v. Ohio], supra, at 27, 88 S.Ct., at 1883, particularly in ambiguous or borderline cases. Reflexive reliance on a profile of drug courier characteristics runs a far greater risk than does ordinary, case-by-case police work of subjecting innocent individuals to unwarranted police harassment and detention.

In this way, profiling not only raises the risk of innocents being subject to police harassment and detention, but also makes the police less effective and thus all of us less safe.