Race and Criminal Justice
The interaction between race or minority status and the criminal justice system is a particularly salient aspect of the racial problems in the United States, and it represents one of the crucial issues that must be addressed as the nation tries to deal with its racial conflicts. There is a large disproportionate representation of minorities, especially Blacks, involved in all aspects of the criminal justice system; and this disproportionality alone, regardless of its legitimacy, conveys a profound sense of unfairness to the overrepresented groups. That sense of unfair treatment is certainly reflected in many polls that show a sharp difference between minority groups and others in their assessment of the fairness of the system. By its very name, the “justice” system carries a particularly heavy burden—it must not only be fair, but it must also be seen as being fair. These concerns are undoubtedly why one of the four missions President Clinton posed to the Race Initiative Advisory Board was to “address crime and the administration of justice.”
There are clearly profound differences in the involvement of different racial groups with the criminal justice system. The differences are most stark between Blacks and non-Hispanic Whites. For example:
This critique is based on Randall Kennedy’s paper, “Racial Trends in the Administration of Criminal Justice,” prepared for the Research Conference on Racial Trends in the United States. National Research Council, Washington, D.C., October 15–16, 1998.
Blacks’ incarceration rate in 1999 was 2.8 times their rate in 1980.
Blacks’ incarceration rate in 1999 was 8.2 times the incarceration rate for non-Hispanic Whites.
Of the Black population, 1.6 percent was incarcerated in state and federal prisons in 1999 (Blumstein and Beck, 1999: Table 2).
Of Black males in their 20s, 8.3 percent are in prison in 1999 (Blumstein and Beck, 1999: Table 2).
It is likely that 28.5 percent of Black males will serve time in a state or federal prison; whereas it is likely that 4.4 percent of non-Hispanic White males will serve time in a state or federal prison (Bonczar and Beck, 1997).
Of Black males in their 20s, 30 percent were under control of the criminal justice system—prison, jail, probation, or parole (Mauer, 1990)
In some cities, the rate of Black males in their 20s under control of the criminal justice system exceeds 50 percent (Miller, 1996).
Anyone who looks at these high rates of involvement must find them most distressing, especially for the social control system that is intended to be a last resort for engendering proper behavior in a civil society. Even if they represented totally even-handed administration of justice, the high rates of intrusiveness—and especially the glaring disparities between Blacks and Whites—must raise profound concerns and an intense search for means of reducing the racial disparities. It is then critical to examine the sources of the disparities in order to know where to focus attention to alleviate the problems.
The paper by Randall Kennedy (1998) highlights a number of important examples of policies and practices that contribute to the racial disparities in the criminal justice system.
FACTORS CONTRIBUTING TO THE DISPARITIES
It is important to isolate factors contributing to the disparity, because different factors will lead to different approaches for improvement. We can identify three different contributors:
individual acts of discrimination,
policies that have differential racial effects, and
racial differences in participation in the crimes that lead to involvement with the criminal justice system.
The first and most obvious cause is individual acts of discrimination by police, prosecutors, judges, corrections agencies, or parole authorities. Discrimination is often difficult to identify in an individual case, but statistical patterns of discriminatory decision making can be ascertained. There is little doubt that any such discriminatory behavior should be highlighted, routed out, and guilty officials appropriately dealt with. There can be little doubt that such acts of discrimination occur; but the racial disparities are so great that it strains credulity to believe that in the United States, at the end of the twentieth century, the bulk of the disparities can be attributed to individual acts of discrimination.
Policies with Differential Racial Impact
The second major source of disparity is policies, reflected in statutes or formal administrative actions, that have a racially disparate consequence. For example, sentences meted out for violent crimes, like robbery or homicide, tend to be higher than those for property crimes like burglary. It would be discriminatory if White robbers received shorter sentences than Black robbers, controlling for other legitimate factors that enter into judicial decision making. There should be serious challenge if the sentence for robbery was made more severe than for burglary because Blacks were more disproportionately involved in robbery than in burglary. On the other hand, because society views robbery as a more serious offence than burglary, it is not discriminatory if the respective sentences reflect only that difference in seriousness.
Given the complexity of discerning intentions in the establishment of any policies, it will often be difficult to distinguish clear and simple reasons for any particular policy. But any policy that results in significantly different racial outcomes, that does not have a strong rational basis, must be suspect and should be challenged.
Differential Involvement in Crimes
The third consideration (aside from discrimination and policies with racial consequences) is the differential involvement in the kinds of crimes that give rise to serious punishment. To the extent that the racial disproportionality in prison is a consequence, not of discrimination by the criminal justice system, but rather because more Blacks commit the serious crimes that give rise to incarceration, then we must search outside the criminal justice system for means of changing that behavior, and recog-
nize that improvements within the criminal justice system will fail to diminish those differences.
That again raises the question of whether we punish certain crimes more severely because we consider them to be more serious, or because they are disproportionately engaged in by Blacks or other minority groups. With the exception of drug offenses—and this is an important exception because drug prisoners currently comprise 23 percent of state prisoners and 60 percent of federal prisoners (Blumstein and Beck, 1999) — there seems to be strong evidence that there is a high ordinal association between people’s perception of the seriousness of the offense and the time served for them (Blumstein and Cohen, 1980). This poses a research challenge of finding a basis for assessing the rates of commission of various crimes by members of different racial groups.
POLICE PROFILING BASED ON RACE
Kennedy highlights one important policy that is sometimes formal and more often informal—the use of race as a factor in police “profiling,” or the use of an individual’s attributes to make him a suspect of a particular violation, such as being a drug courier. Here again, there is widespread agreement that use of race alone is entirely inappropriate in any police action. The issue raised in Kennedy’s paper is whether race may be used at all, even if it is only one of a number of other factors in profiling. Obviously, the argument for including race requires the presence of a solid empirical basis for strong differences in involvement in certain crimes based on race that are not accounted for by other observable characteristics. Certainly, in the absence of a strong empirical basis, then it would be entirely inappropriate for race to be taken into consideration.
But even if there is an empirical difference, it seems entirely reasonable never to permit race to become a factor in profiling. If it were, then it would open the door to the considerable potential for enhancing racial stereotyping, which can create severe problems in a society working hard to eliminate the remnants of past racism.
Furthermore, with changing conditions, there could well be important changes in the empirical bases linking race to the crime outcome, potentially even eliminating the correlation. Because race has no theoretical meaning in this context, bringing about this change in the profiling would be most difficult in the face of the considerable inertia inhibiting revision of the factors used in profiling.
A further danger of using race in profiling is the relative difficulty in measurement of the different factors in the profile. Race is one of the most readily observed attributes of an individual, so that, even if it is not one of the most important factors in a profile, it can easily become far more
salient than other more salient attributes that are more difficult to measure. As a result, it is likely to be given far more weight in the profiling decision than would be warranted, based on its empirical support. Kennedy has noted that courts have generally upheld the use of race in profiling, but one can see strong reasons why it should be prohibited.
In this context, it is useful to note the observation by civil rights leader, Rev. Jesse Jackson, that “there is nothing more painful to me at this stage in my life, than to walk down the street and hear footsteps, and start thinking about robbery—then look around and see somebody White and feel relieved.” This provides a clear indication that, even though the use of racial information in profiling may not be inherently irrational, one can find powerful public-policy reasons for precluding its use. The pain to Rev. Jackson, and to the society more broadly, can be far more insidious than whatever crime-control benefit is attained that cannot be attained virtually as well by using other, more acceptable indicators.
Perhaps the strongest concerns about racial differences are associated with the ways in which punishment is meted out. The case is made even more conspicuous in the case of capital punishment, which, even if infrequent, is our most extreme sanction, and one that is extremely symbolic because of its connection to the abhorrent traditions of lynching.
Kennedy highlights the dilemma of the McCleskey case, in which the strong analytic work of Baldus et al. (1983) was presented to the U.S. Supreme Court. Baldus showed that decisions for capital punishment took account of the race of the victim as a salient consideration—i.e., murderers who killed White victims were sentenced to execution more often than those who killed Black victims. The Court chose not to take account of this finding of clear differential treatment based on race because they claimed there was no proof that that consideration applied in the McCleskey case.
Another important finding in the Baldus research was the recognition that the race of the offender was not significantly associated with a capital sentence. That finding is a source of some degree of comfort; failing to find that would certainly raise serious concerns about the discriminatory treatment of offenders. The joint effect of these two findings—no effect associated with the race of the offender but a racial bias (often labeled as “victim discounting”) associated with the race of the victim—raises a difficult dilemma. The great majority of murders are between people of
the same race, with the largest exception being murders committed in the course of a felony such as a robbery. If capital-punishment decisions were purged of victim discounting, that would lead to an increase in the already distressingly high racial disproportionality on death row, regardless of whether capital-punishment rates were reduced to the rates associated with Black victims (thereby freeing many White offenders from capital punishment) or increased to the level associated with White victims (thereby passing a death sentence on many more Black offenders). We have no way of knowing to what extent this dilemma entered the consideration of the Supreme Court in their review of the McCleskey case and contributed to their choice not to accommodate the claims that were raised.
Drug Sentencing Policies
Perhaps the most visible of the policies with differential racial impact is the striking disparity highlighted by Kennedy of the federal statute (Anti-Drug Abuse Act of 1988, P.L. No. 100–690, 102 Stat. 4181, 1988 [from Kennedy, 1998]) prescribing mandatory-minimum sentences for possession of cocaine with an intent to deliver. The same mandatory-minimum sentence of 10 years is prescribed for both 50 grams of crack cocaine and for 5,000 grams of powder cocaine, a 100:1 ratio. The underlying presumption is that crack cocaine is more often associated with violence than is powder cocaine. It is also the case, however, that individuals convicted of crack cocaine charges are predominantly Black (according to Kennedy, 92.6 percent compared to 4.7 percent White), while individuals convicted of powder cocaine charges are predominantly White (45.2 percent compared to 20.7 percent Black).
This difference must raise the question of whether this extreme 100:1 disparity is based on rational consideration of violence or whether it is a subterfuge for specifically being more punitive to Blacks, or perhaps some combination. The U.S. Sentencing Commission was sensitive to these concerns and several times asked for changes in this particular racially disparate policy, but Congress rejected those proposals, and so the policies still stand. This is in the face of decisions by other courts, such as the Minnesota Supreme Court, which declared that any treatment difference between crack cocaine and powder cocaine was a violation of at least the Minnesota State Constitution.
Differences in Incarceration Rate
The two previous examples—capital punishment decisions and the drug mandatory-minimum sentences—are certainly important, but they
represent a relatively small fraction of the total prison population. But the ratio of 8.2:1 between the incarceration rate for Blacks compared to non-Hispanic Whites applies to the entire prison population, and must raise important concerns. Here, it is important to be able to distinguish how much of this difference is attributable to racial differences in involvement in crime, especially the kinds of serious crime that leads to imprisonment, and how much is attributable to some mixture of discrimination and policies within the criminal justice system that have differential racial impact.
The initial challenge, then, is finding some means of estimating the involvement of different racial groups in the kinds of serious crimes that lead to incarceration. Self-report studies typically have too few of those events. The National Crime Victimization Survey (NCVS), conducted annually by the Bureau of Justice Statistics since 1973, asks victims about the race of the offender who victimized them. That information is helpful for crimes involving a personal confrontation like robbery, assault, or rape, but not very useful for property crimes where there was no direct confrontation. Hindelang (1978), in a classic study, compared the victim reports of the race of the offender in a number of cities with the race of the arrestees for robbery and aggravated assault (offenses in which the victim could observe the offender), and found a strong correspondence between the two. This suggests the reasonableness of using arrest information for the serious crimes that give rise to prison, as a proxy for race information on the offenders themselves.
I published such a study in 1982 (Blumstein, 1982) and found for two of the most serious offenses, robbery and murder, that the Black/White ratio of arrestees was close to the race ratio in prison; and so arrest accounted for about 100 percent of the prison disproportionality. As one examines the less-serious offenses, where there is more discretion at the various stages of the criminal justice system, the race ratio at arrest accounts for less of the racial disproportionality in prison. The race ratio in drug arrests in particular accounted for only about 50 percent of the prison disproportionality for drug offenses. Aggregating across the various offenses, weighting them by their relative proportion of prisoners, about 80 percent of the racial difference in prison is accounted for by the racial differences in arrest. The other 20 percent includes other factors that may be different between the races. Some may be factors that judges may legitimately take into account, such as prior record or employment status. Racial discrimination is also a potential factor. We certainly know that discrimination exists at various stages of the criminal justice system, but it would be astonishing if discrimination could be a major explanation of a difference of more than 80 percent.
In a 1993 update of that earlier study (Blumstein, 1993), the prison
disproportionality accounted for by arrest dropped from 80 percent to 76 percent. The dominant reason for that drop was the growing prevalence of drug offenders in prison. In the 1982 study, based on data from the 1970s, drug offenders accounted for only 5.7 percent of prisoners. By the 1993 update, based on a survey of prisoners in 1991, drug offenders had become 21.5 percent of all prisoners. Race differences in the arrest of drug offenders still accounted for only 50 percent of the racial disproportionality of drug prisoners, but because drug prisoners had increased so significantly, their presence diminished the ability of arrest disproportionality to explain the total prison disparity.
When we look at the individual states, we also find some striking results. The aggregate Black/White incarceration ratio for the total United States is about 7 to 1. The individual states display a considerable range, from 20.4 at one end to 4.0 at the other. The five states at one extreme are Tennessee, South Carolina, Georgia, Mississippi, and Alabama—generally southern states. The five states at the other extreme are Minnesota, Connecticut, Utah, New Jersey, and Nebraska—generally progressive states. The interesting observation is that the southern states have a low incarceration ratio and the progressive states have the high incarceration ratio. This clearly conflicts with the stereotypical presumption that the southern states, with their history of discrimination, should be at the high end rather than the low end. This difference could well be associated with different rates of offending of Blacks in the South, inasmuch as those rates could be associated, perhaps, with greater socialization through residential stability and more rural residency. Also, the typically higher incarceration rates in the southern states could result from more punishment for less-serious offenses, where racial differences are less than in the most serious offenses.
These studies of prison disproportionality show that the bulk of the disparity is attributable to factors outside the criminal justice system. That makes it particularly important to address the many other factors that contribute to differential involvement in crime, and particularly the most serious crimes. These are the issues addressed by all the other sessions of this conference.
SOME RESEARCH NEEDS
Explanation of the Factors Contributing to the Disparities
Given the importance of the problem of race and criminal justice, it is striking to note how little research attention has been paid to the fundamental task of isolating the factors contributing to the important racial
differences in the criminal justice system. Generating such empirical knowledge should be a high priority in order to help address and effectively reduce the problems of racial disparity.
Further efforts are needed to isolate the factors contributing to the radical racial differences in incarceration rates. Information about racial differences by crime type in arrest and, where possible, on victimization through the NCVS, will be important in contributing to such analyses. It is important that differences by state be explored. Other relevant factors taken into account by judges, such as type of defense counsel, prior record, employment status, and other factors, should be addressed wherever possible.
It is also important to distinguish how racial disparities differ between the adult and juvenile systems. There are strong indications that the disparities are much greater in the juvenile system, especially in distinguishing placements in private, versus public, institutions. It is important to identify the sources of those disparities and their legitimacy.
Existence and Impact of Police Profiling
A second important theme is examining the degree to which police profiling occurs. Some initial studies of this issue are already being pursued by Mastrofski et al. (1998) in two cities (St. Petersburg, Florida, and Dayton, Ohio) under a National Institute of Justice grant. In addition, there have been a variety of reports of traffic stops and drug searches carried out with a strong suggestion of racial profiling.
Role of the Drug War in Contributing to Racial Differences
The strongest candidate for a policy area that does have strong racial implications has been the efforts to pursue drug abuse through enforcement and incarceration. It would be desirable to assess the form and degree of racial impact associated with the nation’s drug enforcement efforts. Any such study should focus first on differential effects on individuals, reflecting the degree to which the racial disproportionality in prison is attributable to differential involvement in arrest for drug offenses and the racially different punishment associated with drugs. In particular, it is important to isolate the degree to which the distinctions made between powder cocaine and crack cocaine, especially in mandatory-minimum sentencing, have contributed to these differences. It would also be desirable to examine the degree to which the high incarceration rates of young Black males, many of whom are in prison on drug charges, have contributed both to significant disruption within their communities
as well as to a reduction in those communities in the deterrent effect associated with the threat of incarceration.
Interactions Between the Police and Minority Communities
A fourth major theme should pursue the issue of interactions between the police and minority communities. So often, it is the interaction between the police and minority individuals that serves to escalate the tensions and, thereby, deteriorate interaction between the two groups. A highly publicized confrontation, whether appropriate or not, gives rise to hostility by members of the group, leading to their withholding support from the police when and where they need it, and aggressive or excessively defensive action in their interactions with the police. All of this contributes to putting the police on the defensive, and results in police acting more aggressively and more defensively when they interact with members of these groups. This process of escalating conflict and tension is not easy to extinguish. It is certainly exacerbated by those individual police officers who display a persistent pattern of racism. It is in the interest of every police department to develop procedures for identifying such individuals and appropriately punishing or dismissing them from the department. Symmetrically, it is important that leadership groups in the minority communities be recruited to work with the police to help reduce these tensions.
Research in this area should attempt to measure the level of tension between the police and various minority groups, and relate that level of tension to various aspects of police operations. It should also include consideration of recent incidents of conflict between the police and members of the minority communities. It is likely that these tensions increase considerably on both sides following singular incidents of police use of force that is at least arguably excessive. The dynamics of that growth of tension as well as its subsidence can be examined through community and police surveys as well as through ethnographic studies in communities before and after such critical incidents.
Baldus, D., C.Pulaski, and G.Woodworth 1983 Comparative review of death sentences: An empirical study of the Georgia experience. Journal of Criminal Law and Criminology 74(3):661–753.
Blumstein, A. 1982 On the racial disproportionality of U.S. prison populations. Journal of Criminal Law and Criminology 73(3):1259–1281.
1993 Racial disproportionality of U.S. prison populations revisited. University of Colorado Law Review 64(3):743–760.
Blumstein, A., and A.Beck 1999 Factors contributing to the growth in U.S. prison populations. Pp. 17–61 in Crime and Justice: A Review of Research, M.Tonry and J.Petersilia, eds. Chicago: University of Chicago Press.
Blumstein, A., and J.Cohen 1980 Sentencing of convicted offenders: An analysis of the public’s view. Law and Society Review 14(2).
Bonczar, T., and A.J.Beck 1997 Lifetime likelihood of going to state or federal prison. Bureau of Justice Statistics Special Report No. 160092, March 6.
Hindelang, M. 1978 Race and involvement in common law personal crimes. American Sociological Review 43(1):93+.
Kennedy, R. 1998 Overview of racial trends in the administration of criminal justice. Paper prepared for the Research Conference on Racial Trends in the United States, Washington, D.C., October 15–16.
Mastrofski, S., R.Parks, C.DeJong, and R.Worden 1998 Race and everyday policing: A research perspective. Paper delivered at the 12th International Congress on Criminology, Seoul, Korea, August.
Mauer, M. 1990 Young Black Men and the Criminal Justice System: A Growing Problem. Washington, D.C.: The Sentencing Project.
Miller, J. 1996 Search and Destroy: African-American Males in the Criminal Justice System. Cambridge: University Press.