Is the Death Penalty Racially Discriminatory?: NO
by: Stanley Rothman and Stephen Powers
EXECUTION BY QUOTA?
On March 17,1994, the House Judiciary Committee voted to incorporate the Racial Justice Act into this year's Omnibus Crime Control Bill. The Act essentially would create quotas for the administration of the death penalty, under the assumption that the penalty is applied in a manner discriminatory to black Americans. While the legislation has been opposed by House Republicans, one should not, given the temper of the times and the mood of Congress, discount the possibility that it eventually will become law.
The Racial Justice Act would prohibit "the imposition or execution of the death penalty in a racially discriminatory pattern." Further, the Act provides that to establish a prima facie showing of discrimination:
it shall suffice that death sentences are being imposed or executed... upon persons of one race with a frequency that is disproportionate to their representation among the numbers of persons arrested for, charged with, or convicted of, death eligible crimes....
THE CONTROVERSY
The employment of the death penalty as the ultimate criminal sanction has been the subject of enormous debate. Execution has been challenged not only on moral and religious grounds, but more recently on constitutional grounds as a violation of the Eighth Amendment's protection against cruel and unusual punishment. Opponents of the death penalty contend that it is employed so arbitrarily as to amount to a game of state-sponsored Russian roulette. While the Supreme Court has not ruled capital punishment to be unconstitutional, in 1972 it held that the death penalty was unconstitutional as then practiced, finding evidence of arbitrariness sufficient to require that states overhaul death sentencing procedures.
One of the most controversial aspects of the arbitrariness claim is the charge leveled by numerous activists and social scientists that the death penalty has been applied in a manner unfair to blacks. In Furman vs. Georgia (1972, several members of the Court observed that racial discrimination had produced different patterns of sentencing and rates of execution for blacks and whites. Indeed, numerous studies of the late 1800s and early 1900s have found that blacks were executed in disproportionate numbers, particularly when the victims of their crimes were white.
The apparently discriminatory impact of capital punishment has not gone unnoticed in Congress. In fact, one aim of the Racial Justice Act is to circumvent prior federal court decisions which have held that statistical research does not provide sufficient evidence of "discriminatory intent" to trigger Fourteenth Amendment protection. The Act states that "it shall not be necessary to show discriminatory motive, intent, or purpose on the part of any individual or institution."
If the Racial Justice Act becomes law, state and federal authorities will have to demonstrate that any racial disparities in sentencing are "clearly and convincingly" explained by non-racial factors. Given the high cost of litigation and likely delays, as well as the difficulty of proving non-discrimination when sentencing is based partly on factors not easily subjected to statistical analysis (how does one quantify the "heinousness" of a crime?), states could be forced to abandon death sentences against some black defendants, irrespective of the merits of the cases. If this occurred, sooner or later it might also be an easy matter for white defendants to show discriminatory sentencing under the same law. The death penalty would be effectively eliminated.
But is death sentencing truly discriminatory? The truth is complicated by a number of factors that opponents of the death penalty have tended to discount or ignore. There appear to be legitimate reasons for racially disparate sentencing. Indeed, a number of social scientists have argued that racial prejudice is not a significant determinant of execution rates. These social scientists have demonstrated that when a number of legal factors are taken into account, the relationship between a defendant's race and the likelihood of execution tends to disappear. Why, we must ask, in spite of the questionable validity of the discrimination thesis, does the death penalty continue to be assailed as one of the most repugnant manifestations of American racism?
PAST STUDIES
Before the Supreme Court's decision in Furman, a majority of death penalty studies had reported that discrimination against black defendants was substantial, particularly in cases of rape and in the South. Certainly there was ample historical precedent. By law, black slaves were subject to the death penalty for numerous crimes for which whites received much more lenient sentences. In 1848, for example, Virginia enacted a statute which required that blacks be executed for any crime for which whites might receive three years' imprisonment (Dike, 1981). The evidence for discriminatory death sentencing through the nineteenth and early twentieth centuries, particularly in the South, seems incontrovertible.
Even when discriminatory sentencing was not actually prescribed by law, statistical studies show that before the 1950s black offenders were much more likely than whites to be executed for murder. In 1930, H. C. Brearley reported that in South Carolina, from 1920 to 1926, blacks accused of murder were twice as likely as whites to be convicted. And during the period from 1915 to 1927, blacks were more than three times as likely as whites to be executed. Numerous researchers reported similar findings from the 1930s through the late 1960s.
Nevertheless, by the mid-1970s studies were uncovering methodological problems with some of the earlier research. In 1974, John Hagan reanalyzed a number of studies involving capital sentencing and found that most of the studies had confused correlation and causation. When Hagan controlled for prior record and type of offense, he found that the influence of race dropped dramatically. He concluded that:
knowing the race of the offender... increases the accuracy of predicting judicial disposition by 1.5 percent. The causal importance of even this minimum relationship, however, is called into doubt by the single study controlling simultaneously for charge, and related "third" variables.
In a similar vein, a 1981 study by Gary Kleck found that between 1929 and 1966 the rate of execution for blacks (9.7 per 1,000 murders) was slightly lower than that for whites (10.4 per 1,000 murders). Indeed, Kleck made another important observation: "in the recent past, outside of the South, white execution risk has been substantially higher than the nonwhite risk, a fact which apparently has gone unnoticed in the literature."
In the past, Kleck suggested, black criminals may have been treated relatively leniently because their crimes against other blacks were not taken seriously, or perhaps because of white paternalism (blacks frequently were viewed as less culpable for their actions). In more recent times, Kleck suggested, the relatively lenient treatment of blacks may be due to the attempts of judges to compensate for what they perceive as institutional racism, or to make up for their own unconscious racism. Kleck also noted that his own figures do not take into account prior sentencing records and that, since in other studies this factor has tended to suppress racial disparities, his own findings probably understate the higher execution risk for whites. Overall, although Kleck found evidence of discrimination in certain historical periods (especially in the South), and for particular classes of crime such as rape, he found no evidence of system-wide discrimination in the imposition of the death penalty beyond the 1950s.
Since the publication of Kleck's study, many other studies have appeared that also find white defendants to be at greater risk in murder cases than black defendants, even in the South, though one or two of the studies have attempted to explain the findings away in a manner described below (Baldus et. al., 1990; Ekland-Olsen, 1988; Gross, 1985; Katz, 1989; Nakel and Hardy, 1987; Pesternoster, 1984; and Radelet, 1981).
With some research indicating that discriminatory sentencing of black offenders was confined to the South and probably had ended by the 1950s, sociologists began to search for more subtle evidence of discrimination. As far back as the 1930s, a handful of studies had reported that blacks who killed whites were more likely to receive the death penalty than blacks who killed other blacks, or than whites who killed members of either race. One of the first of these studies hypothesized that, as a subjugated race, blacks were "treated with undue severity" Johnson, 1941). The author of the study found that in a sample drawn from parts of three southern states, black offenders were significantly more likely to be executed when their victims were white than black. Only 64 percent of blacks sentenced to death for killing other blacks were executed, whereas 81 percent of blacks who killed whites were put to death.
The researcher argued that blacks who killed other blacks were treated leniently because they were viewed by authorities as childish and not fully culpable for their actions. Yet when blacks killed members of the dominant racial caste, they were punished especially severely, to keep them in their place. Without considering the influence of factors other than race, the author concluded that the data "point toward a partial confirmation of our hypothesis." This failure to consider alternative explanations is characteristic of the research prior to the 1960s.
However, some later and better constructed studies have reached similar conclusions. One study, published in 1983, found that the offender-victim racial combination was at least as significant as predictive factor in death sentencing as any other legal variable (e.g., contemporaneous felony, multiple victims).
A number of studies have attributed this seeming racism to prosecutorial discretion. A recent study in Kentucky, for example, suggested that prosecutors tend to view cases in which blacks kill whites as more serious than other types of cases. The researchers were cautious in attempting to explain why this might be, and pointed out that factors beyond the scope of their analysis might have been influential. Nevertheless, their study clearly implies that race continues to be a significant and obviously illegitimate factor in death sentencing (Keil and Vito, 1991).
Still other studies have found that social class is an important factor. A few notable studies have shown that once one controls for the offender's social class, race becomes an insignificant predictor. For example, a 1969 study by Charles Judson and others found that race was not a statistically significant determinant of death sentencing. If anything, Judson's statistics suggest that whites were more likely to receive death sentences than blacks (48 percent of whites received death sentences, and 40 percent of blacks). But when Judson and his colleagues controlled for various crime-related variables, the influence of race disappeared. The socio-economic status of the offender, however, did seem to be important. Of course the substitution of one extra-legal variable for another does not justify differential sentencing, but it does suggest that our knowledge of the factors involved in sentencing is very limited. Indeed, the number of variables that can be shown to influence sentencing seems sometimes to be limited only by the ingenuity of the researchers involved.
A 1983 report of the Panel on Sentencing Research, commissioned by the prestigious National Research Council (NRC), concluded that even among the more sophisticated studies which found discrimination in cases with black offenders and white victims, race was a relatively weak predictive variable. The NRC panel cautioned that the "validity of statistical inferences about the determinants of sentences depends crucially on the methodological rigor with which the effects were estimated.... [and] the findings presented here are weighed in light of potentially serious methodological flaws in research" (Blumstein et al., 1983).
Other critics point out that key legal variables, such as prior record and seriousness of offense, have been difficult for researchers to document and even more difficult to quantify. Other legally relevant factors, such as degree of criminal intent, frequently have been overlooked. The fundamental problem with studies of the relationship between race and the death penalty is that they fail to establish convincing causal explanations. In fact, most studies demonstrate that numerous variables influence capital sentencing. For all we know, many other influential variables maybe as yet untested. Some may be unquantifiable. On the basis of the available research, one simply cannot conclude that racial discrepancies are a function of racism.
Reviewing the history of research on race and sentencing generally, William Wilbanks found at least seven different "models of method and interpretation" in the literature (Wilbanks, 1987). Despite the wide variation, he contends that some general observations on race and sentencing are possible. Among them, he includes the following: Racial discrimination in sentencing has declined over time . . . The black/white variation in sentences is generally reduced to near zero when several legal variables are introduced as controls.... The race effect, even before controls, is not substantially significant, in that the predictive power of race is quite low.... Most sentencing studies have a large residual variation, suggesting that the models used did not fit the actual decision making of judges....
These observations have proved to be especially applicable to research on discrimination and the death penalty.
WHO KILLS WHOM AND WHY
The vast majority of murderers who receive the death penalty are involved in intra-racial offenses- that is, in cases of whites killing whites or blacks killing blacks. Most analysts agree that between 92 and 97 percent of homicides are intraracial. In the much smaller number of cases in which blacks kill whites, the circumstances surrounding the crimes appear to be substantially different. (The number of cases in which whites kill blacks is usually too small to be factored into analyses.)
Black on black homicides are most likely to occur during altercations between persons who know one another. On the other hand, black on white homicides (and to a somewhat lesser extent, white on white) are often committed during the course of a felony or by a multiple offender. In fact, these are examples of aggravating conditions that the Supreme Court has held to be valid criteria in determining sentence severity. Yet while judges and juries take these factors into account, sociological studies often do not. Lest one think the motivation of judges and juries is racism, these factors are given consideration in societies all over the world, whatever their racial composition. They are seen universally as both fair and conducive to public order.
The key issue, then, is whether blacks convicted of killing whites are more likely to be executed because of the racial identity of their victims or because of qualitative differences in the nature of their crimes. In fact, the latter is clearly the case and would appear to explain much of the racial disparity in death sentencing.
THE McCLESKEY CASE
One of the most effective challenges to the claim of racial discrimination actually arose in a court case that supporters of the discrimination thesis had hoped would prove their point. In the 1980s, the National Association for the Advancement of Colored People (NAACP) funded a major study of the effect of race on criminal sentencing. The study, directed by university professors David Baldus, Charles Pulaski, and C. George Woodworth, gained notoriety when it was used in the defense of Warren McCleskey, a black man sentenced to death for the shooting of a white police officer in Georgia. Defense attorneys relied on the Baldus study to substantiate their claim of systemic discrimination against black defendants. The study showed that in cases of mid-range aggravation, blacks who killed whites were more likely to receive the death penalty than whites who killed whites. (In cases of low and high aggravation, the study found race to be an insignificant factor.) The authors of the study argued that racial bias occurred because prosecutors and juries were prejudiced.
The attorneys prosecuting McCleskey countered by hiring an expert methodologist, Joseph Katz, who analyzed the NAACP study and found a number of conceptual and methodological problems. For one, it turned out that police reports often did not include some of the case circumstances that were supposed to have been weighted in the study. In these instances, the researchers recorded that the circumstances were not present, when, in fact, that was not possible to determine. Katz also pointed out that the researchers had not accounted satisfactorily for the fact that black offender-white victim homicides were often quite different from intra-racial homicides. Katz showed that black on white murders tended to be the most aggravated of all, and frequently were combined with armed robbery, as McCleskey's was. Katz also testified that by Baldus's own measures, McCleskey's was not a mid-range case but a highly aggravated one, and that in such cases the death penalty was as likely to be applied to whites as blacks.
The Supreme Court ended up rejecting the McCleskey defense, and ruled that statistical models alone do not provide sufficient evidence of discrimination. Later, Katz testified before the Senate Judiciary Committee, and offered further evidence of the differences between homicides in which blacks kill blacks and blacks kill whites. Katz reported that the reason why 11 percent of blacks who killed whites in Georgia received the death penalty as opposed to only 1 percent of blacks who killed blacks was that the killings of whites more often involved armed robbery (67 percent of the black on white cases, compared with only 7 percent of the black on black cases). In addition, black on white murders more frequently involved kidnapping and rape, mutilations, execution style murders, tortures, and beatings. These are all aggravating circumstances that increase the likelihood of a death sentence.
By contrast, 73 percent of the black victim homicides were precipitated by a dispute or fight, circumstances viewed by the courts as mitigating. Katz also observed that 95 percent of black victim homicides were committed by black offenders, and that there were so few white on black cases that no distinctive homicide pattern could even be ascertained. Among the fewer than thirty Georgia cases identified by Katz as white on black, mitigating circumstances seemed to outweigh aggravating. These crimes rarely involved a contemporaneous felony and often were precipitated by a fight. This pattern may or may not hold outside of Georgia, but to date there has been no detailed national study of white on black crime. (Research has also shown that death sentences are especially likely in cases in which police officers are killed in the line of duty, and that 85 percent of police officers killed are white.)
As pointed out earlier, some findings suggest that blacks may actually be treated more leniently than whites. Analysts at the Bureau of Justice Statistics have pointed out that the percentage of inmates on death row who are black (42 percent) is lower than the percentage of criminals charged with murder or nonnegligent manslaughter who are black (48 percent). If the legal system still discriminates against blacks, why do they make up a higher percentage of those charged with murder than those executed for murder?
Some critics reply that the police may be more likely to arrest and charge blacks than whites. Yet we have found few data that support this assertion. In fact, Patrick Langan, a senior statistician at the Bureau of Justice Statistics, investigated the possibility of such discrimination and found little evidence of it. Langan based his research on victims' reports of the race of offenders, and found that blacks were sentenced at rates similar to those who would expect given the reports of victims (Langan, 1985). Obviously, this kind of research could not be conducted for murder cases (because the victims are dead) but the research suggests that the discriminatory arrest argument is highly problematic.
In the federal courts, the discrimination argument has found little support. In a number of cases, judges have concluded that the evidence of systemic bias is extremely weak. Rather than order an overhaul of the legal system on the basis of highly problematic and conflicting social science research, judges have preferred to adjudicate discriminatory sentencing claims on a case by case basis. The preferred corrective has been procedural reforms. A number of states have adopted clearer sentencing standards, various provisions to remove extra-legal influences, and the judicial review of death sentences.
IDEOLOGY PREVAILS
Why, then, have some researchers continued to find evidence of racial discrimination? One possible explanation is that while the sociologists who design death penalty studies are most interested in and competent to measure such variables as the demographic characteristics of groups, these sociologists are ill equipped to assess the importance of the legal variables that influence the operation of the criminal justice system. In the past, researchers did not bother to control for even the most obvious of legal variables.
Yet despite the crudeness of their methods, sociologists have concluded confidently that racism in the legal system is rampant. Mindful of the history of racial discrimination in capital cases, sociologists perhaps are predisposed to conclude that discrimination persists today. It seems obvious.
An additional difficulty with many of the sociologists is that their assumptions concerning discrimination are often overly idealistic-for example, the belief that extra-legal variables must be entirely absent from the criminal justice system for it to be legitimate, and the assumption that complete objectivity is even possible. Taken to their logical extremes, these kinds of utopian beliefs would require us to condemn virtually every legal system in the history of the world. At best, legal systems are imperfect institutions, reflecting community standards of fairness and objectivity. The jury system and judicial discretion are indispensable instruments of social justice, which permit broad principles to be tailored to the particulars of each case. Without these instruments, and the attendant margin of error or abuse that all free exercises of judgment hazard, the legal system would be doomed either to excessive punishments or to a forbearance that placed innocent individuals at great risk.
While there is justification for the claim that discriminatory capital sentencing and execution occurred in the past, the charge that they persist today lacks support. The best available evidence indicates that disproportionate numbers of blacks commit murder, and that in those cases in which the victims are white the crimes generally are aggravated. That is why blacks are over represented on death row.
POLITICS AND THE DEATH PENALTY
Clearly there are reasons other than statistical analysis for the continued belief that the legal system discriminates against black defendants. Those who oppose the death penalty on principle, for example, tend to incorporate the discrimination argument into their litany of protest. These critics perceive capital punishment as a vestige of an outmoded, barbaric, and irrational penal code. Black elites, meanwhile, often perceive discrimination in places others do not. They are joined by members of the white cultural establishment, who are quick to sympathize with those who allege racial unfairness.
This may sound like a harsh indictment, but how else are we to explain the facts? For decades, those who argued that the death penalty was administered in a biased manner maintained that the fact that more blacks were executed than whites revealed a lack of concern for black lives. When this argument became untenable when it became clear that white murderers were actually more likely to be executed than black murders- these same critics turned to other, equally unsatisfactory arguments. Now, however, they reject the implication of their previous view- that the execution of a larger percentage of whites than blacks must reveal a lack of concern for white lives. The only issue now is the race of the criminal's victim. These critics rationalize their position, but, we submit, their stance can be explained only by a need to find racism everywhere. One is reminded of the world in Aesop's fable. The wolf insisted that the lamb was injuring him, and was quick to change his story each time the lamb pointed out the factual errors in his claims. Finally, the wolf killed and ate the lamb anyway, proving that desire can overcome the failure of rationalization.
AND SO
If the controversy over racial discrimination and the death penalty turned on the merits of the research, politicians would have to concede that death penalty discrimination has been virtually eliminated. Alas, the news media have done little to clarify matters. Most reporting on the issue is inaccurate. An article that appeared in the New York Times on April 21, 1994, is typical. The article concluded as follows:
That some bias occurs is not much at issue. Many studies show that juries mete out the death penalty to black and other minority defendants in a disproportionate number of murder cases, particularly when the victims are white and especially in states and counties that have a history of racial problems.
In fact, as we have shown, these comments are patently false. Of course, many key questions remain. Is the death penalty arbitrary, given that only a fraction of those eligible are ever executed? Is it barbaric? Is it ineffective as a deterrent? If the answers to these questions are affirmative, two remedies are available: the death penalty can be abolished or subjected to further reform. But whatever society decides, such a decision should not be based on unsubstantiated charges of racial discrimination.
POSTSCRIPT
Is the Death Penalty Racially Discriminatory?
At the beginning of their selection, Rothman and Powers discuss the Racial Justice Act of the Omnibus Crime Control Bill, which would establish quotas for the administration of the death penalty to prevent discrimination in its use. This act, however, was defeated by the U.S. Senate. The White House maintains an "aggressively neutral" stance on the issue.
In the 1958 Supreme Court case Trop v. Dulles, Chief Justice Earl Warren predicted that "evolving standards of decency" would eventually result in the elimination of capital punishment.[The truth of this last sentence is doubtful; Trop did not involve a death penalty case, so it was unlikely that a Supreme Court Chief Justice would have uttered any words involving a death sentence. Trop in volved a loss of United States citizenship.] In 1972 Furman v. Georgia effectively halted executions in the United States. As a result, many were convinced that capital punishment had finally ended in America for both blacks and whites. However, many states worked hard to put together specific guidelines for carrying out the death penalty in a consistent and fair manner. Reflecting this effort, Gregg v. Georgia (1976) which tested a new Georgia statute that would reduce the incidence of discrimination in applying the death penalty resulted in the Court's approval. Opponents of executions then contended in McCleskey v. Kemp (1987) that statistical analysis proved that blacks were discriminated against. The Supreme Court rejected this, saying there was no proof that others who "may be similarly situated did not receive the death penalty."
Defenders of the Court, such as Rothman and Powers, say that it may be true that blacks who kill whites are more likely to be sentenced to death than whites who kill blacks (only a few have been executed since 1976) or blacks who kill other blacks. This, however, has little to do with discrimination. Instead, interracial killing is likely to be more heinous than intraracial because the latter usually involves a previous relationship between the murderer and the victim. An intraracial crime is often one of anger (e.g., domestic assault) that does not entail other felonies, such as robbery, or premeditation.
Recent articles on the death penalty include "Should Capital Punishment Be Abolished?" Jet (February 13, 1995) and J. Barnes, "Capital Punishment Gridlock?" Investor's Business Daily (April 11, 1995). For a more comprehensive treatment of the issue of race and capital punishment, see C. Mann's Unequal Justice (Indiana University Press, 1993). One of the most recent studies of lynchings is "A Festival of Violence: An Analysis of Southern Lynchings" by S. Tolnay and E. Beck (University of Illinois Press, 1995). A very different kind of study is reported by S. Burgins in "Jurors Ignore, Misunderstand Instructions: The Result. . . Is a Bias in Favor of the Death Penalty," ABA Journal (May 1995).