State of the Debate: The Color of the Law


Randall Kennedy, Race, Crime, and the Law (Pantheon, 1997).

Craig Reinarman and Harry G. Levine, eds., Crack in America: Demon Drugs and Social Justice (University of California Press, 1997).

Michael Tonry, Malign Neglect: Race, Crime and Punishment in America (Oxford, 1995).

Franklin E. Zimring and Gordon Hawkins, Crime is Not the Problem: Lethal Violence in America (Oxford, 1997).

Is America now finally ready for the message of Randall Kennedy's new book—that in enforcing the criminal laws, the courts and the police should never be allowed to make judgments based on race? Eliminating racial bias from law enforcement was one of the great objectives of the civil rights movement of the 1960s. But it was one of the movement's historical misfortunes that it preceded a major national upswing in rates of violent crime. In a culture where race and crime have deep historical connections, rising crime rates raised social anxieties about race. As public opinion increasingly favored harsh crime policies, it became harder to get a hearing for fair treatment of suspects and offenders of any kind. Now, with crime rates easing, might Americans be more receptive to real racial equality in the criminal justice system?

It is hard to exaggerate how powerfully the law has shaped the life chances of Americans of African heritage, and in ways that we scarcely think of today. For example, criminal statutes enforced in 39 states, even in the North and West, long made it a felony for blacks (and other nonwhites) to marry whites. In fact, these "anti-miscegenation" laws remained in effect in 13 states until the United States Supreme Court declared them to be unconstitutional in 1968. According to the dominant beliefs about race, the children of "mixed" marriages would be defective. Nevertheless, black women were frequently taken or raped by white men who were rarely, if ever, punished. The children of such encounters were designated "Negro." This label barred them from inheriting their fathers' property. The laws were also intended to prevent black men from having consensual sex with white women under any circumstances, including marriage; they implied that no rational, adult, white woman would agree to have sex with a black man. Any breaking of the sex—color line taboo between a black man and white woman could be—and in the peculiar logic of the deep South should be—considered the moral equivalent of rape, even if blessed by the sacrament of marriage.

In the context of such racial theorizing, accusations of rape made by white women against black men were rarely disbelieved. Such accusations were likely to draw violent reprisals from white vigilantes, who went unpunished for crimes including whipping, torture, burning, and eventually hanging—or "lynching"—their victim (the "strange fruit" of Lillian Smith's acclaimed novel). Southern court records show that when a black man was accused of murdering a white man, he was usually not lynched but given a trial, and if found guilty, executed by the state. The accusation of rape, by contrast, was more likely to evoke the hot-blooded savagery of a lynching.

The institutions of southern justice—police and courts—typically ignored the crimes committed in the lynching. Southern blacks passed around stories that became legends about sex, terror, and the meaninglessness of the official legal order. Lynching maintained the caste superiority of whites and the bloody etiquette of crossracial sex. It also undermined any trust Americans of African descent might have had in the legal order.

Subscribe to The American Prospect

This ignominious history is the necessary preamble to any understanding of race and crime today. "Nothing has more eroded confidence in the criminal justice system than the long history of willful refusals to punish white anti-black vigilantes," writes Randall Kennedy, professor at Harvard Law School and a former law clerk to Justice Thurgood Marshall, in his thoughtful and provocative book.

Yet Kennedy does not come across as a reflexive radical on the complex and polarizing issue of race and crime. Like his mentor, Justice Marshall, he does not excuse "thuggery" perpetrated by blacks. And like Jesse Jackson, he recognizes that disproportionate criminality by young black males leads to understandable fears among potential victims, whether black or white. Kennedy's fair-mindedness on race and crime is evident from his discussion of the beating of Rodney King. The case was more complicated, he points out, than is generally recognized by those familiar only with the portion of the infamous videotape shown on television. At the Simi Valley trial, defense attorneys focused the jury's attention on King's behavior leading to the beating. He was, after all, drunk, driving at high speed, and resisting arrest. Some use of escalated force was probably justified against him, although not the 54 powerful blows that were actually inflicted.

Kennedy also recognizes how African Americans are doubly victimized by crime. "One out of every twenty-one black men can expect to be murdered," he writes, "a death rate double that of American servicemen in World War II." More recent crime statistics show that even with violent crime falling nationwide, not much has changed for African Americans. Homicide victimization rates for black men and women continue to be higher than for other segments of the population. Black males were about eight to nine times more likely than white males to commit a homicide during 1996; and most of them killed other blacks. In 1996, about nine of every ten murders involved victims and offenders of the same race (when the race of the offender was known).

In this connection, Franklin E. Zimring and Gordon Hawkins's Crime is Not the Problem: Lethal Violence in America offers an exceptionally perceptive analysis of African-American violence. Zimring and Hawkins show that America does not have higher rates of property crime (burglary, theft, and other property offenses) than other industrialized nations. According to their analysis, residents of Sydney and Los Angeles have about the same chances of being burglarized. Americans, however, are more likely to be killed, assaulted, and raped. The United States still leads in violent crimes, especially lethal violence, although with homicide rates dropping here, the difference between the United States and other advanced societies has diminished.

Zimring and Hawkins argue that contrary to widespread perceptions, American violence is not "solely, or mainly, or even distinctively, a black problem." In part, they say, this is because American blacks tend to reside in places where social conditions precipitate the greatest violence by all races, and partly because tendencies to lethal violence seem to be endemic in the United States. If all homicides by blacks were cosmetically removed from America's crime statistics, Zimring and Hawkins argue, the U.S. homicide rate "would still be a statistical 'outlier,' far beyond the experience of other industrial democracies."

Yet American crime rates are changing so rapidly that books published in 1997 citing statistics from the early 1990s are already dated. Nationally, the FBI reported that the number of homicides fell 9 percent in 1996 from the previous year, hitting the lowest level since 1969. Overall, violent crime dropped in 1996 in the United States to levels not seen since 1973. Violent crime has dropped even among juveniles since its peak in 1994, and for the first time in years, firearm homicides by and against black youth are declining. Declining crime rates could ease fears of crime committed by young black males, who are responsible for most of the violent crime committed by blacks.


In this new climate, might race become less of a factor in the administration of justice in America? Kennedy devotes much of his book to the argument that it is "never appropriate" for police to use color as a proxy for criminality, and he points to several important areas where, in fact, the police do. For example, in a number of state and federal cases, the courts have permitted police to stop and question someone who is "out of place" in a white neighborhood, or who fits a profile of a "typical" drug courier. Police can also consider race in deciding whether to subject drivers to questioning or search at border checkpoints. Kennedy deplores the legal doctrines that "permit police to treat blackness as a mark of increased risk of criminality" because, he says, "they generate large pools of distrust, anger and discord."

However admirable this "color-blind" position may be, is it realistic to expect the police and other Americans to adopt it? Although American violence is surely not, as Zimring and Hawkins point out, "distinctively a black problem," it is nevertheless disproportionately a black problem. Perhaps the police are more likely to arrest violent black criminals than violent white ones. But homicide statistics—the most reliable statistics on crime—suggest that police are arresting whites and blacks for violent crime with some evenhandedness.

I was once retained as an expert witness by attorneys for a police department that was sued under the Federal Civil Rights Statutes for maintaining a "pattern and practice" of racial discrimination. The plaintiffs made a prima facie case by alleging, quite correctly, that although slightly less than 50 percent of the population was black, more than 80 percent of the arrests were of black males. But most calls to the police for help, it turned out, were from the black community, and more than 90 percent of the alleged perpetrators were described as young black males. Moreover, although blacks constituted less than 50 percent of the general population, more than 70 percent of the youth population was black. Based on these figures, the prima facie case disappeared and a federal court dismissed the allegations.

Given the reality of disproportionate black crime, will court decisions actually affect police conduct? Will police simply not list race when it actually was a factor in stopping and questioning someone who fits a profile or appears suspicious to a police officer—even if legal doctrine says they may not? Consider the following case. A police detective sees two men, Chilton and Terry, walking up and down a street apparently "casing" a jewelry store. Lawyers familiar with the landmark 1968 case of Terry v. Ohio know the rest of the story. The officer questions the men, decides that a crime is "afoot," proceeds to pat them down, discovers guns, and arrests them, along with a third man. Detective Martin McFadden testifies that he had been patrolling in plainclothes at 2:30 in the afternoon, in an area of downtown Cleveland that had been his beat for 30 years. He says he saw something odd about these men. "Now, in this case when I looked over they didn't look right to me at the time."

The Warren Court, while understanding that Detective McFadden had less than "probable cause" to conduct his search, deferred to the practical needs of policing and permitted the limited "pat" search of Chilton and Terry and the seizure of their weapons, a major doctrinal shift in the law governing when police can lawfully "stop and frisk" suspects. Not mentioned in the Supreme Court decision is that the suspects were black teenagers, as the NAACP Legal Defense Fund brief pointed out at the time. But was McFadden necessarily a racist? After all, Terry and Chilton were behaving suspiciously, and they turned out to be armed and evidently were about to commit a crime. Yet is it credible to believe that McFadden didn't notice their skin color—not simply their behavior—as part of what made them not "look right" to him at the time? More troubling, is it possible for a police officer not to factor skin color into his or her perceptions of not "looking right" in a society where skin color is so salient and where young black males commit crimes disproportionately?

These are unsettling questions, especially for those like Kennedy (and me) who would prefer to erase skin color as a legitimate indicator of anything. When police are permitted to use race as "an indicia of suspicion," Kennedy argues, "courts also derogate from the idea that individuals should be judged on the basis of their own, particular conduct." As normative aspirations go, Kennedy's is commendable. It is a standard to which we and the courts should aspire. But I expect that in the real world of disproportionate black criminality (as well as racism), the police will continue to use race as an indicator, as McFadden must surely have done. And like McFadden—especially if courts say they cannot use race as an indicator—police won't report that they did, and will testify that what they saw was solely behavior that didn't look right.




Kennedy's insistence that skin color be irrelevant in the processing of those accused of crime extends as well to jury selection. Not until 1986 did the Supreme Court hold that the Equal Protection clause prohibits prosecutors from using peremptory challenges to exclude blacks from juries; since then it has outlawed racially based peremptory challenges for the defense as well. Judges can exercise more authority over attorneys in a courtroom than they can over police on the street. Nevertheless, Kennedy is skeptical that courts can actually keep prosecutors or defense attorneys from using peremptory challenges to shape the racial composition of the jury. Like Justice Marshall, he thus favors eliminating these challenges altogether, arguing that it is probably the only way to restrain attorneys who use race as a criterion in jury selection.

Other components of the criminal justice system are more amenable to doctrinal authority. The Supreme Court could, if it chose, abolish the penalty of death, but the present conservative Court is not about to do that. Kennedy is masterful in describing the doctrinal zigs and zags of death penalty jurisprudence, and offers an especially careful and knowledgeable analysis of the statistical data on race and execution.

In the most recent major Supreme Court case, McCleskey v. Kemp, the defense introduced a complex statistical study that showed that when murder victims were white in Georgia, perpetrators (black or white) were four times more likely to be condemned to death than when victims were black. The Court conceded that the system was skewed against black victims, but it pointed out that since most black victims had black killers, "4 percent of the black defendants received the death penalty, as opposed to 7 percent of the white defendants." Was bias involved? Not according to the Court: "apparent disparities in sentencing are an inevitable part of our criminal justice system." Accordingly, the question was whether officials had discriminated against McCleskey, who was black, in this case. The Court ruled that they had not. As Kennedy recognizes, it would be a gruesome kind of affirmative action that sought to reduce racial discrepancies in capital punishment by "leveling up" and executing more blacks who murder black victims.

Kennedy's discussion of racial fairness in the administration of the death penalty is careful, knowledgeable, and nuanced, but his own position on the larger question—support or opposition—is relegated to a footnote. I thought this a mistake, since what he says makes much sense and deserves the kind of careful elaboration he gives to the question of racial fairness. Kennedy doesn't regard capital punishment as "unconstitutional per se," but opposes it partly because he fears mistakes and partly because he deplores "the lethal, collective, bureaucratic anger that the state displays when it puts a person to death."

But something else is hinted at in the footnote, a change of heart from fervent abolitionist to mild opponent. Kennedy writes that when he clerked for Justice Marshall he was forced to "constantly read about the horrible crimes perpetrated by murderers sentenced to death." Evidently, the brutality of the murderers and the pain of the victims cooled his abolitionist fervor. It is not easy to develop a purely "rational" position on capital punishment, although Justice Harry Blackmun's argument—that the death penalty cannot be administered fairly—comes closest.

Punishment is a difficult issue, especially when one can predict that a particular race or class will be disproportionately affected. Some African-American legal scholars, notably Paul Butler in the Yale Law Journal, have advocated that black jurors nullify the evidence in cases where the black defendants are charged with what he describes as "nonviolent, malum prohibitum of fenses, including victimless crimes like narcotics offenses." Effectively, Butler wants blacks to use the jury system to pardon other blacks.

Kennedy will have none of this. Agreeing that African Americans have often been treated unjustly, he rebuts Butler point by point, arguing that Butler bases his position on a one-dimensional vision that overlooks how powerfully effective the justice system can be when allowed to run its course. Yes, Kennedy acknowledges, the prosecution of the Scottsboro boys was "horrible, and racially motivated," but both state and federal authorities intervened "in an extraordinary fashion," ultimately preventing their execution. Similarly, Rodney King's victimization was later followed by the imprisonment of the perpetrators of the brutality in a federal civil rights prosecution. Jury nullification, Kennedy insists, will scarcely advance the cause of broad social reform that Butler advocates. Those who engage in nullification will have to say that they ignored the evidence for a larger cause, and few jurors are willing to do that. The jurors in the O. J. Simpson case, for example, did not admit to nullification of the evidence.


But are the penalties for some offenses, particularly drug offenses, too draconian? No criminologist has been a more knowledgeable and articulate critic of the War on Drugs than Michael Tonry, a law professor at the University of Minnesota.

In his book Malign Neglect, Tonry is sharply critical of the punitiveness of crime policy since the 1980s. The most extreme example of a law discriminating against African-American males is the federal law that subjects a person selling five grams of crack cocaine, a relatively trifling amount, to a five-year minimum penalty, while imposing the same penalty on those who sell 500 grams of cocaine powder, an amount larger than what most street dealers possess.

As Tonry points out, the color line is drawn sharply here. In 1995, 88.4 percent of those convicted in federal courts of selling crack cocaine were black, while only 4.5 percent were white. As a result, the average sentence served by black prisoners in federal prison (71 months) was 41 percent longer than the average served by whites (50 months). In the early 1980s, in comparison, the average time served by blacks was comparable to that of whites.

This is not to say that crack and powder cocaine are the same: crack cocaine is powder cocaine dissolved in water, with baking soda added, then heated and dried into hard, smokable pellets. Street samples range from 10 percent to 40 percent cocaine by weight. If we're punishing by weight, an equal amount of powder might be more deserving of punishment because of its potential to become crack. It is easy to cook up crack from cocaine—in effect, if someone has cocaine, with little knowledge or effort he can have crack. The law is equivalent, if eggs were illegal, to punishing the possession of omelets 100 times more severely than the possession of raw eggs. Cocaine and crack are not the same, but they are not so different pharmacologically as to justify vast differences in punishment.

Defenders of elevated penalties for crack point to other considerations. Powder cocaine is sold in larger, more expensive amounts and it is sold behind closed doors. It is consequently harder to catch those who sell it than those who sell crack, who can be found mostly in "crack houses" known to neighbors and the police, or in the streets. Powder cocaine is the drug of the affluent, while crack is the drug of the poor. And because crack is sold more openly, it is more threatening to community safety and cohesion. More significantly, as John P. Morgan and Lynne Zimmer report in an excellent new paperback, Crack in America, edited by Craig Reinarman and Harry Levine, crack has been made available to "just those parts of the population who are most vulnerable to the abuse of any drug."

It is this feature of crack that has led many African-American politicians, as well as Randall Kennedy, to support the distinction in penalties. "Surely," Kennedy writes, "it would be just and sensible for a government to punish more severely a person knowingly distributing a poison in a low-priced (say $5) container as opposed to a high-priced (say $50) container even if the poison in the two containers was otherwise identical."

But would we punish more severely the persons selling the $5 containers to street buyers if we learned they were young, black, and poor, while the more affluent $50 sellers could afford to deal behind closed doors where they cut it up into $5 containers to be sold to the street sellers? Why should we punish retailers more severely than wholesalers? After all, we demand capital punishment for major drug traffickers.

According to National Institute of Drug Abuse surveys, more blacks (3.4 percent) than whites (1.6 percent) report having used crack at least once, but 83 percent of those sentenced for crack offenses are black. Cocaine is surely a dangerous drug, in any form. The gross disparity in penalties between crack and powder seems, however, far less justified than Kennedy suggests. And Kennedy recognizes that it makes no sense to level the penalties by raising the punishment for those who sell powder cocaine.

At the conclusion of the book, in the very last paragraph, Kennedy backs off. He says he doesn't "endorse" the crack-powder differential. "Even if these policies are misguided," he concludes, "being mistaken is different from being racist, and the difference is one that greatly matters." Does Kennedy mean to suggest that we should censure only the explicit use of race in crime enforcement but excuse disparate and punitive impacts as long as they result from good intentions?

Consider the history of the death penalty for rape. In a Virginia case discussed by Ken nedy, the defendant's attorneys showed that between 1908 and 1949, 45 black men, but not a single white man, had been put to death for rape. In Coker v. Georgia, the 1976 case in which the Supreme Court prohibited the imposition of the death penalty for rape, the Court selected a case involving a white victim and a white defendant, thus largely avoiding the discussion of racial disparity in sentencing. The death penalty, it held, is so "excessive" for rape that it violates the Eighth Amendment's "cruel and unusual punishments" prohibition. Kennedy notes, however, that racial disparity in rape sentencing—but not the death penalty—still exists in many places, strongly suggesting that disparity is an important measure of racism.

Kennedy would argue that the disparity in punishment of rape is clearly footprinted in a history of racism, while the crack-powder sentencing disparity was not based on racial motives. That may be true. But in a society with a history of slavery and racial discrimination, can we ignore disparate racial impacts when we are considering fairness? Perhaps someday, when equality is more of a reality. But, at present, race remains such a conspicuous factor in crime and crime policy that we cannot fail to notice sharp differences in the fate of blacks and whites. As we approach the millennium, the color line of punishment—especially in the War on Drugs—is still all too evident.

You may also like