July 1996                    Vol. 8, No. 4 (G)

Human Rights Violations in the United States

RACE AND DRUG LAW ENFORCEMENT IN THE STATE OF GEORGIA


This report was originally published as a chapter in Human Rights Watch, Modern Capital of Human Rights?: Abuses in the State of Georgia (New York: Human Rights Watch, July 1996) .

I. INTRODUCTION

The impact of crime control policies on minorities is among the most important, disturbing and contentious social issues facing the United States. Overwhelming data establish the striking proportion of African-Americans entangled in the criminal justice system-on any given day one in three young black American males is either in prison or jail, on probation or parole.1 Drug laws and enforcement policies are among the most important causes of this national crisis. As one expert has noted, "Urban black Americans have borne the brunt of the War on Drugs. They have been arrested, prosecuted, convicted, and imprisoned at increasing rates since the early 1980s, and grossly out of proportion to their numbers in the general population or among drug users." 2

The national pattern of racial disproportion in the "war on drugs" is replicated in the state of Georgia.3 As we document in this report, both black and white Georgia residents use and distribute drugs, but black residents are far more likely to be arrested and incarcerated for drug offenses.4 Black residents of Georgia are arrested for all drug offenses at a rate five times greater than white residents of the state. For cocaine-related offenses, they are arrested at seventeen times the rate of whites. Blacks are imprisoned for drug offenses at twice the rate of whites and have received 98 percent of the mandatory life sentences that have been imposed for those offenses. Fifty young black men between the ages of eighteen and twenty-one have received life sentences.

The operation of the criminal justice system in Georgia is governed by state and federal law, both of which enjoin discrimination on the basis of race. International human rights law is also implicated: one of the overarching principles of international human rights is that of equality before the law.5 The International Convention on the Elimination of All Forms of Racial Discrimination (CERD), to which the United States is a signatory, is the most comprehensive international codification of the human rights principle of racial equality.6 It calls on nationalgovernments to take steps to eliminate discrimination and to seek to prohibit discrimination under the law as well as to guard against discrimination arising as a result of the law. 7

In this report we examine drug law enforcement in Georgia in light of CERD and the requirement of non-discrimination, focussing primarily on the years 1990 to 1995. Drawing on computerized statewide databases,8 we have compiled statistics on the racial dimension of arrests and imprisonment for drug offenses in Georgia that have never been published before.9 Because of the limitations in the data, however, our figures should be considered as estimates illuminating the general contours of the racial patterns in drug law enforcement. The nature of the data available to us also precludes an analysis of the role race may play in the many decision points in the criminal justice system between arrest and sentencing.10 The disparate racial impact we are able to document at the end points of the criminal justice system-arrest and incarceration-suffices, however, to raise a warning flag concerning the fairness and equity of Georgia's drug law enforcement.

II. GEORGIA DRUG LAWS

Georgia imposes criminal penalties for the unauthorized possession, manufacturing, distribution, sale and trafficking of controlled substances which, following the federal model, are placed within one of five categories or "schedules."11 Purchasing or possession of a controlled substance in schedule I, or of certain drugs in schedule II (e.g. cocaine),12 is a felony punishable by not less than two years of imprisonment and not more than fifteen. A second or subsequent conviction for possession is punishable by between five and thirty years in prison.

More serious punishment is levied on the manufacture, sale, or possession with intent to sell of drugs such as cocaine. The first offense is punishable by five to thirty years. Conviction of a second or subsequent offense has been punishable, in theory, by mandatory life imprisonment. The law also establishes a penalty of one to ten years of imprisonment for marijuana possession, distribution, sale or possession with intent to distribute or manufacture. Penalties for trafficking, ie. the production or sale of twenty-eight grams or more of controlled substances, are set according to the quantity of the drug involved. Sentences range from a minimum of five years to a maximum of thirty, in addition to fines not to exceed US one million dollars.

III. DRUG OFFENSE ARRESTS

Although by their terms Georgia's drug laws are racially neutral, the enforcement and application of these laws tell a different story. Statewide arrest figures reveal a striking disparity between the numbers of African-Americans and whites arrested for drug offenses.13 Before the so-called "war on drugs" was launched across the nation in the mid 1980s, more whites than blacks were arrested for drug offenses. By the end of the decade, the total number of arrests for drug offenses had increased dramatically and, as shown in Figure 1, the racial composition of those arrested had reversed: the number of blacks arrested for drugs was more than double that of whites. Over the decade, the annual number of white arrests increased only marginally, from 10,376 to 11,850. In contrast, the number of blacks arrested for drug offenses increased from 5,689 in 1980 to 24,512 in 1989.

The disparity in the numbers of blacks and whites arrested for drug offenses continued between 1990 and 1995. During this period, at least 200,243 persons were arrested in Georgia for the illegal possession or sale of drugs. Although blacks constitute less than one-third of the population of Georgia,14 64.2 percent of those arrested for drugs were black men and women. Only 35.6 percent were white men and women.

The best measure with which to assess the relative impact of drug arrests on blacks and whites is the ratio of arrests to population. Figure 2 shows shows the comparative ratios of arrests for drug offenses per 100,000 of the white and African-American adult populations in the years 1990-1995. In each year, blacks

were arrested at a rate five times greater than whites. Our analysis of the arrest data also reveals a strong difference by race in the number of arrests according to the drug involved. The drug of most significance is cocaine-the drug whose use fueled the "war on drugs" nationwide as well as in Georgia and the drug involved in the greatest number of arrests. As shown in Table 1, blacks constituted 83.7 percent of all the arrests in Georgia between 1990 and 1995 for possession of cocaine, and constituted 87 percent of the arrests for its sale .15

Table 1: Total Drug Offense Arrests by Race 1990-1995
  Black White
Number of Arrests Percent of Total Arrests Rate Per 100,000 Number of Arrests Percent of Total Arrests Rate Per 100,000
Cocaine            
Possession 57,701 83.67 % 4,957 11,146 16% 319
Sale 31,559 87 % 2,711 4,487 12% 128
All 89,260 85 % 7,668 15,633 14.9% 447
Marijuana            
Possession 25,350 42.7 % 2,178 33,833 57% 968
Sale 4,866 36 % 418 8,578 64% 245
All 30,216 41.5 % 2,596 42,411 58% 1,213
Other Drugs            
Possession 1,354 27.7% 116 3,532 72% 101
Sale 8,014 44.8% 689 9,823 55% 281
All 9,368 41 % 805 13,355 58.6% 382
TOTAL 128,845 64.2 % 11,069 71,399 35.6% 2,043

Source: Arrest data from Georgia Crime Information Center

Comparison of the ratio of arrests to population reveals an even starker racial discrepancy: blacks were arrested for cocaine offenses at a rate of 7,668.4 per 100,000 black adults. Whites, in contrast, were arrested at a rate of 447.2 per 100,000 white adults or one-seventeenth the rate of blacks.16 The rate of black arrests per 100,000 black adults for marijuana offenses is more than double the rate of white arrests per 100,000 white adults, even though, in absolute numbers, more whites than blacks were arrested for marijuana possession and sale.17 The dramatic difference between blacks and whites in the ratios of arrests by drug type to population is depicted in Figure 3.

Source: Arrest data from Georgia Department of Corrections. Population from 1990 Census. Rates calculated on basis of populations of black and white adults eighteen and over.

Rate of Arrests Compared to Rate of Offending Conduct

The difference between black and white arrest rates is stunning. But the greater number and rate of arrests of blacks compared to whites by themselves do not establish discrimination or unequal treatment. If blacks were arrested more frequently because they break the law more frequently, that is, if different arrest rates for blacks and whites reflected different rates of criminal conduct, then the data would not suggest discrimination in the enforcement of the drug laws. Unfortunately, there are no specific data on the number and racial composition of drug users and sellers in Georgia.18 However, anecdotal information available for Georgia and national drug surveys do establish an approximation of the racial composition of the Georgia drug market.

Police, prosecutors, defense attorneys and ethnographers in Georgia interviewed by Human Rights Watch agree that drug use in Georgia is spread across racial and socio-economic lines. Cocaine is used by both races. In its crack form, cocaine is prevalent in lower-income black communities, although white use of crack is increasing. The district attorney for Gwinnett County, for example, told Human Rights Watch that a "sting" in 1992 in which law enforcement personnel posed as crack sellers in a black neighborhood resulted in the arrest of some five dozen people in the first two hours. Two-thirds of those arrested were white.19 Powder cocaine, which is more expensive than crack, is primarily consumed by middle- and upper-income individuals, who in Georgia are primarily white.

The monitoring of drug trends by the U.S. Office of National Drug Control Policy (ONDCP) also indicates multi-racial use of drugs in Georgia. For example, in the Fall 1995 Pulse Check published by ONDCP, ethnographers reported that in Atlanta powder cocaine was used by "white snorters" and crack cocaine was used by African-Americans in their twenties. The preceding Pulse Check had summarized cocaine users in Atlanta as: "late teens, early 20s, whites; older African-Americans."20

While valuable, the anecdotal information cannot be used as a basis for comparison with arrest statistics. In the absence of Georgia specific drug possession statistics, we have utilized drug use rates taken from national household surveys to draw comparisons with Georgia arrest rates. By all accounts, drug use in Georgia does not appear to differ appreciably from national rates. 21

In an equitable criminal justice system, we would expect that racial proportions in arrest rates for possession would resemble racial proportions in drug use.22 In Georgia, however, we find a startling discrepancy. Using the most recent national rates for current illicit drug use, we estimate that in 1994, for example, at least 7,300 black Georgians per 100,000 were current users of illicit drugs compared to 6,000 whites per 100,000.23 Thus blacks apparently use drugs at a rate about 20 percent higher than whites. Yet blacks were arrested for possession of illicit drugs at a proportional rate that was 500 percent greater than whites.

As shown in Table 2, African-Americans in Georgia are also arrested at rates greatly disproportionate to their estimated share of the total drug using population. 24 In 1994, for example, although blacks constituted approximately 14 percent of all current drug users, they constituted 58 percent of persons arrested for drug possession.25 Conversely, whites represented 76 percent of the drug users in Georgia, yet they accounted for only 41 percent of those arrested. In other words, a black drug user had a much greater likelihood of being arrested for drug possession than a white drug user.

Table 2: Comparison by Race of Drug Use and Possession Arrest, 1991-1995
Year Black  White
Percent of Current 

Users

Percent of Arrests Percent of Current Users Percent of Arrests
1991 16.97% 67 % 72.4 % 32.6 %
1992 13.7 % 63.5 % 76.4 % 36.2 %
1993 13 %  61.5 % 74 % 38.1 %
1994 14 % 58.3 % 76.5 % 41.3 %
1995 N/A 59 % N/A 40 %

Source: Arrests from Georgia Crime Information Center. Figures on drug use calculated from

SAMHSA data. Data covers all illicit drugs.

The discrepancy between use and arrest rates for whites and blacks is even greater if we look at the comparative rates for cocaine. Blacks use cocaine at a rate that is two and a half times greater than the rate of whites.26 Yet blacks are arrested for cocaine possession at a rate that is fifteen times greater than whites.27 Furthermore, as shown in Table 3, black users are arrested at a rate greatly in excess of their estimated share of the total population of cocaine users, while whites, conversely, are arrested at a rate substantially less than their share of users.

Table 3: Comparison by Race of Cocaine Use and Possession Arrests
Year Black White
Percent of Total Current Cocaine Users Percent of Total Arrests Percent of Total Current Cocaine Users Percent of Total Arrests
1991 37.6 % 85.1 % 57 % 14.8 %
1992 17.6 %  84.3 % 65.9 % 15.5 %
1993 21.8 %  83.3 % 57.9 % 16.4 %
1994 22 % 79.0 % 62 %  20.8 %
1995 N/A 80.6 % N/A 19 %

Source: Arrest data from Georgia Crime information Center. Figures on drug

use calculated from SAMHSA data.

Marijuana arrests present similar racial disproportions. Between 1990 and 1995, blacks accounted for a larger percentage of the total arrests for marijuana possession than they did of the population of marijuana users. (SeeTable 4.) In addition, blacks were arrested at an annual rate of 363 per 100,000 compared to a rate of 161 per 100,000 for whites. Although more whites, in absolute numbers, were arrested than blacks, their arrest rates were not comensurate with their share of the marijuana using population.

Table 4: Comparison by Race of Rates of Marijuana Use and Possession Arrests
Year Black White
Percent of Users Percent of Arrests Percent of Users Percent of Arrests
1991 17 % 40 % 73 % 59 %
1992 18 % 38.6 % 77 % 61 %
1993 14 % 40 % 74 % 59.5 %
1994 13.7 % 42 % 76 % 57.5 %
1995 N/A 46 % N/A 53 %

Source: Arrest data from Georgia Crime Information Center.

Figures on drug use calculated from SAMHSA data.

Given the high number of arrests that are for drug sales and the more serious penalties attached to sales, we have attempted to compare the racial proportions of arrests with the racial proportions of the drug-selling population. The effort must be seen as, at best, a crude approximation, because there are no reliable analyses of the drug-selling population by race. Nevertheless, the anecdotal and statistical data that do exist indicate that the drug selling population in Georgia is more mixed racially than the population that is actually arrested by the police.

According to anecdotal information from law enforcement personnel and defense lawyers in Georgia, whites constitute a significant proportion of drug sellers. Police personnel in the Atlanta metropolitan area told Human Rights Watch that at the retail level, that is, regarding sales to individuals purchasing primarily for their own use, blacks dominate the sale of crack cocaine but both blacks and whites sell powder cocaine and marijuana to drug consumers. Methamphetamine, or "redneck cocaine," a drug whose use is growing, is sold almost entirely by whites. The ONDCP's Pulse Check also confirms that in Atlanta, at least, both whites and blacks sell drugs. In the Pulse Check published in the summer of 1995, for example, ethnographers reported that white dealers in Atlanta were selling powder cocaine and young African-Americans were selling crack.28

Data on the prevalence of drug sellers nationwide is available from SAMSHA for the three-year period 1991 to 1993.29 SAMHSA figures, based on answers to questions during their voluntary household interviews, indicate that whites may have comprised 82 percent of the total number of drug sellers nationwide, and blacks comprised 16 percent.30 Given the nature of the population surveyed by SAMHSA, these figures undoubtedly undercount theactual percentage of black sellers. Nevertheless, they suggest, at the very least, that whites constitute at least as many sellers as blacks.

If we assume, as seems reasonable, that the racial composition of the total drug-selling population in Georgia does not differ dramatically from that obtaining nationwide, then the racial breakdown of arrests for drug sales in Georgia is startling. As indicated in Table 1, in the past six years twice as many African-Americans have been arrested for drug sales as whites. Eighty-seven percent of the persons arrested for cocaine sales are black, compared to 12 percent white. Only with regard to sales of marijuana does the whites' percentage of arrests (69.9 percent) begin to resemble their estimated share of the selling population. Firm conclusions would be inappropriate given the speculative nature of the seller population data. Nevertheless, the available data does suggest that black sellers may be arrested in numbers disproportionate to their share of the drug selling population.

Why the Disparate Impact?

Arrest rates reflect both drug-market activity and the choices of police enforcing the drug laws. Taken together, the data discussed above indicate that blacks in Georgia have been arrested at rates far higher than their rate of criminal conduct. Discriminatory purpose or racial bias- conscious or unconscious31-may contribute to police drug law enforcement practices, but we have no valid means of assessing its presence or the extent of its influence. Law enforcement officials interviewed by Human Rights Watch denied their practices were racially biased. Almost every single person Human Rights Watch interviewed in Georgia, including police officials, stated that the racially skewed arrest statistics flowed from one central reality in drug law enforcement: it is easier to make drug arrests in low-income neighborhoods. According to this view, black offenders are not targeted because they are black. Rather, black offenders are arrested more frequently because the circumstances of their lives and drug transactions make them easier to arrest.

We were told that most of the drug arrests by Georgia police are of lower-level drug dealers and buyers, such as "retail" sellers and consumers, and that most of these arrests occur in low-income minority areas. Retail drug sales in these neighborhoods frequently occur on the streets and between sellers and buyers who do not know each other. That is, the transaction is public and the clientele for street sellers includes many strangers (black and white) who will walk or drive up to a seller at a known location to buy a small amount of drugs for personal consumption. Most of these sellers are black. In contrast, white drug sellers tend to sell indoors, in bar and clubs and within private homes, and to more affluent purchasers, also primarily white.

A number of tactical considerations make it easier to arrest drug offenders who engage in criminal conduct on the streets: they are easier to find and monitor (and catch on videotape). Uniformed police arrest individuals they encounter whom they see engaged in unlawful drug transactions. Undercover officers typically arrest a seller after making one or more drug purchases from that seller, and it is easier for an officer to arrange a buy from sellers accustomed to sell to strangers. "[I]n poor urban minority neighborhoods, it is easier for undercover narcotics officers to penetrate networks of friends and acquaintances than in more stable and closely knit working-class and middle-class neighborhoods. The stranger buying drugs on the urban street corner or in an alley, or overcoming local suspicions by hanging around for a few days and then buying drugs, is commonplace... Police undercover operations can succeed [in working- and middle-class neighborhoods], but they take longer, cost more and are less likely to succeed."32

In other words, blacks may be arrested more frequently because they more frequently engage in drug transactions that are easier to detect and bust.33 Faced with a choice between going after offenders who are easier (faster, cheaper) to arrest versus offenders who will take much more effort, Georgia police not surprisingly have opted for the former.

That choice-concentrating on drug offenders in low-income rather than more affluent neighborhoods-has also been politically pragmatic. As Chief Justice Robert Benham of the Georgia Supreme Court told Human Rights Watch, a concerted effort to root out drug dealing in middle-class enclaves would undoubtedly generate considerable opposition and criticism.34 In contrast, there is no "hue and cry" when the police target low income neighborhoods for drug law enforcement. Indeed, attacking drug dealing in inner city neighborhoods is supported by the neighborhood itself, the general public, the media and political leaders.

The violence, disorder, nuisance and assaults on the quality of life that often accompany public drug markets in low-income communities produce pressure on police departments to commit more resources to those neighborhoods. According to Atlanta police, for example, police departments are complaint-driven organizations.35 They receive few complaints that relatively affluent individuals are engaged in private drug transactions in a bar or office building; those transactions do not create the kind of visible public nuisance and generate the public outrage that prompts complaints. In contrast, residents of low income neighborhoods plagued by drug dealing do complain to public officials and to the police as they seek to free their streets of individuals who make it difficult for them and their children to lead safe and peaceful lives. To their voices are added those of the media, politicians and others who for many different reasons-legitimate concern, the quest for political gain, and so on-point to crime and drug dealing in low income neighborhoods and call for police crackdowns.

It may be that, on closer examination, the racial disparities in arrests are disparities of class. In Georgia, as in many states in the U.S., race and class are to a great extent conflated: a law enforcement system focused on economically disadvantaged individuals is one that more seriously affects minorities, and vice versa.36 But justice is no more served when the poor are disproportionately targeted than when one minority is. To the extent that police choose to concentrate drug law enforcement in poor and/or black neighborhoods as opposed to more affluent whiteneighborhoods, those choices raises the question of equal justice. Where the disparate racial impact is readily foreseeable, even if not expressly intended, equal rights principles are implicated.37

IV. IMPRISONMENT

The racially disparate impact of law enforcement evident from the arrest data is also reflected at the other end of the criminal justice system, in the pattern of incarceration.38 Over the last twenty-four years, between 1972 and March 1996, the state of Georgia has sent approximately 41,000 persons to prison for drug offenses.39 Some 27,657 (or 67.3 percent) of those were black. Among those incarcerated were 4,865 young adults between the ages of eighteen and twenty-one. Of these, 3,135 (or 64.4 percent) were black. As shown in Figure 4, more white offenders

were admitted than blacks from 1972 until the late 1980s. Then, with the rise in arrests of blacks for drug offenses that has accompanied the "war on drugs," the number of blacks incarcerated jumped and has since remained consistently higher than that of whites. Since 1990, as shown in Table 5, blacks have consistently accounted for more than three-quarters of persons admitted to prison for drug offenses.40

Table 5: Comparison of Whites and Blacks Admitted to Prison for Drug Offenses
Year White Black Total
Number admitted Percent of admitted Number admitted Percent of admitted
1990 1,045 25 % 3,079 74 % 4,148
1991 940 20.9 % 3,506 78.6 % 4,455
1992 981 20.3% 3,821 79.3% 4,816
1993 1,012 19.4 % 4,160 80.1 % 5,192
1994 1,192 19.8 % 4,793 79.9 % 5,992
1995 1,259 20 % 5,006 79.7 % 6,280
TOTAL 6,429 20.8 % 24,365 78.9 % 30,883

Source: Georgia Department of Corrections.

As shown in Table 6, young black adults consistently accounted for more than 84.9 percent of the admissions of all young adults for all drug offenses over the decade ending in 1995.

Table 6: Admission of Young Adults to Prison for Drug Offenses 1985-1995
  Black White
  Number  Percent Number Percent
All Drug Offenses 2,571 84.9 % 459 15.1 %
Possession 1,399 87.2 % 206 12.8 %
Sales 1,172 82.2 % 253 17.8 %

Source: Georgia Department of Corrections

The data show that between 1990 and 1995, black drug offenders were incarcerated at more than twice the rate of white drug offenders: 8.8 percent of blacks compared to 3.6 percent of whites arrested for drug offenses were ultimately admitted to prison (See Table 7). As a result, blacks constitute a growing proportion of those admitted to prison even though they account for a declining proportion of the total number of drug arrrests.41

Table 7: Comparison by Race of Drug Offender Arrests and Imprisonment, 1990-1995
Year Black White
Percent of Total Arrests Percent of Prison

Admissions

Percent of Total Arrests Percent of Prison Admissions
1990 68 % 74 % 31.7 % 25 %
1991 69.6 % 78.6 % 30 % 20.9 %
1992 64 % 79.3 % 35 % 20.3 %
1993 62.5 % 80.1 % 37 % 19.4 %
1994 60.8 % 79.9 % 38.9 % 19.8 %
1995 59.9 % 78.9 % 39.8 % 20 %

Source: Arrest data from Georgia Crime Information Center. Prison data from Georgia

Department of Corrections.

At first blush the significant difference in incarceration rates is troubling and suggests unwarranted discrimination in sentencing. Upon closer examination, however, most of the discrepancy appears in fact to originate with different incarceration rates according to the drug involved. As indicated in Table 8, on the average, 11.34%

Table 8 : Rate of Incarceration of Arrested Drug Offenders, 1990-1995
Drug Offense Average Rate of

Imprisonment

Rate of Black Arrestees Imprisoned Rate of White Arrestees Imprisoned
Cocaine 11.34% 12% 7.5%
Marijuana 2.07% 1.6% 2.3%
Other Drugs 2.04% .4% 3.1%

Source: Arrest data from Georgia Crime Information Center. Prison admissions data

from Georgia Department of Corrections. Note: average rate of imprisonment calculated

by aggregating black and white offenders and their corresponding rates.

of the persons arrested for cocaine offenses are sent to prison compared to 2.07% of those arrested for marijuana offenses.42 Many (59 percent) of the white drug offenders in the 1990-1995 period were arrested for marijuana offenses and only 21.8 percent for cocaine. By contrast, only 23 percent of black drug offenders were arrested for marijuana offenses, while 69 percent were arrested for cocaine. (See Table 1). That is, a much greater percentage of black offenders than white were arrested for the drug offenses carrying the highest imprisonment rate.

Using the average imprisonment rates by drug type to compute the expected number of offenders who would be incarcerated given the drugs for which they were arrested, we calculate that 2,923 white arrestees should have been imprisoned between 1990 and 1995. 43 In fact, 2,590-or 11 percent fewer-were sent to prison. By the samecalculations, we would expect that 10,939 of the black arrestees should have been incarcerated. In fact, 11,275 were (a difference of 3 percent).

The difference between the expected and actual number of incarcerated white and black drug offenders reflects that difference between the average rates of imprisonment in all races compared to the actual rate for each race. As shown in Table 8, black cocaine offenders were imprisoned at a rate marginally higher than the average, and white cocaine offenders were imprisoned at a rate substantially lower than the average. Black marijuana offenders were incarcerated at a rate mariginally lower and white marijuana offenders were incarcerated at a rate marginally higher than the average.

Numerous factors that legitimately influence case processing decisions and outcomes may have produced the actual rate differential. Unfortunately, computerized statewide data does not exist that would enable us to examine differences among offenders with regard to their prior criminal histories44, seriousness of arrest charges, number of counts charged, or youthful offender status eligibility. Without being able to control for these and other relevant race-neutral variables, we are not able to assess, for example, whether black cocaine offenders are incarcerated more frequently than comparably situated whites.45

Length of Sentence

Excluding life sentences, the difference between the length of sentences given white and black drug offenders was small.46 As shown in Table 9, the greatest difference was for sale of cocaine, for which blacks received a mean sentence that was two years longer than that given whites. Whites received somewhat longer sentences than blacks for the possession of narcotics. Official databases do not, however, provide that data needed to determine the extent to which race-neutral factors, such as the number of counts charged in each case or prior criminal histories, may have contributed to these sentencing disparities.

Table 9: Comparison by Race of Sentences for Drug Offenses, 1990-1995
  Sale Possession Trafficking
Cocaine Narcotic Marijuana Cocaine Narcotic Marijuana Cocaine Narcotic Marijuana
Black 5.9 4.7 3.4 3.7 2.7 3.2 13.1 - -
White 4.0 4.4 3.2 3.2 3.4 3.2 13.1 - -

Source: Georgia Department of Corrections. Where data not included, the number of admissions too small for statistical reliability.

Life Sentences

The racial disparity in life sentences imposed for drug offenses is shocking. In the past fourteen years, 560 blacks were sentenced to life in prison for drug offenses compared to 13 whites.47 That is, 97.7 percent of the life sentences for drug offenses were given to African-Americans.

Until March of 1996, drug offenders convicted a second or subsequent time of the sale of certain controlled substances, including cocaine, faced a mandatory life sentence under O.C.G.A. 16-13-30(d). Through a procedural loophole, however, what was to have been a mandatory sentence became discretionary in practice. For a defendant to receive a life sentence for a second conviction, the prosecutor had to give notice prior to trial that he or she intended to seek the enhanced punishment based on past convictions.48 In most cases, prosecutors chose not to seek the aggravated sentence.49 If the prosecutor filed the pretrial notice requesting a life sentence, the judge had no choice but to impose it if the defendant were convicted. Many prosecutors objected to the law, considering it "ham-fisted"50 and "stupid."51 Perhaps as a result, over 85 percent of those who were eligible to be sentenced to life were in fact sentenced to lesser terms. 52 But as shown in Table 10, of those who were given life, almost all were African-American.

Table 10: Life Sentences for Drug Offenses
Year Black White All
1990 44 2 46
1991 59 2 61
1992 133 3 136
1993 121 1 122
1994 124 3 127
1995 79 2 81
All 560 13 573

Source: Georgia Department of Corrections.

The disproportionate number of blacks receiving life compared to whites did not mirror the racial distribution of offenders who were eligible for a penalty of life imprisonment. Only 3 percent of the whites who were convicted a second or subsequent time of a qualifying drug offense were sentenced to life imprisonment. By contrast, 15 percent of the blacks who were eligible received life sentences. 53 In other words, a life-eligible black was five times more likely to receive a life sentence than a life-eligible white.

The injustice apparent from the racial pattern of life sentences is even greater when we look at the ages of those receiving life terms. Since 1982, fifty young adults between the ages of eighteen and twenty-one were sentenced to life . All fifty were black. The convictions which sent them to prison did not include any serious non-drug crimes. Drug sale offenses crimes sufficed within the Georgia criminal justice system to have young people at the threshold of their adult lives condemned to life imprisonment.54

According to an analysis prepared by the Georgia State Board of Pardons and Paroles in August, 1993, most of the drug offenders who received life sentences were convicted for offenses involving small amounts of drugs. That is, the law was not being used to punish serious offenders. (Indeed, those who deal in greater quantities of drugs, the traffickers, are not covered by the mandatory life sentence statute). Seventy-seven percent of the offenses leadingto the first conviction and 79 percent of the offenses leading to the second conviction involved less than one gram of a controlled substance. Sixty percent of the cases involved drug values of less than US$50.55

Application of the mandatory life sentence statute has been challenged several times as violative of the equal protection guarantees of the federal and state constitutions.56 Despite strong statistical proof of a discriminatory impact, the courts consistently refused to make a finding of unconstitutionality, citing the absence of proof that the prosecutors were motivated by a discriminatory purpose.57

The shocking racial disparity in the mandatory sentences eventually forced the Georgia legislature to act. In March 1996, the state legislature passed legislation to revise O.C.G.A. 16-13-30(d) that was supported by both the prosecutor and defense attorney associations. Under the revised statute, conviction of a second or subsequent drug sale offense, is punishable by ten to forty years or life.58 Prosecutors can recommend a life sentence, but the judge will be able to decide whether or not to impose it. Defense attorneys and civil rights activists hope the new legislation will lead to a less racially skewed pattern of sentencing because the sentencing decision will no longer be concentrated in a single decision-maker: the prosecutor can seek a life sentence, but the judge will now be able to decide whether to impose it.

Although they supported the legislative reform, prosecutors never publicly conceded that the racial pattern of life sentences reflected racial bias. On the other hand, they have never offered an explanation for how relevant race-neutral factors might have caused the dramatic racial disparity. It remains to be seen, of course, whether granting judges more leeway in sentencing second-time drug offenders will lead to a more racially equitable imposition of life sentences. The discrimination apparent in the imposition of the death penalty in Georgia suggests that more fundamental reforms are needed to ensure racial equity.59

V. CONCLUSION AND RECOMMENDATIONS

The data Human Rights Watch has compiled on drug law enforcement in Georgia, albeit necessarily incomplete, suggest a disturbing pattern of racially disparate impact. The question arises whether Georgia public officials enforce facially neutral drug laws in a discriminatory manner. Human Rights Watch recognizes that law enforcement officials face incessant and evolving challenges to help safeguard communities from crime and disorder. Of necessity, they must set priorities and make continual choices about which crimes and criminals to target and what strategies to adopt to deter crime and to bring criminals to justice. Although discretion is essential to effective performance of their duties, that discretion is not unfettered. It is limited, inter alia, by the principles of equal protection and due process contained in federal and state law and international human rights treaties.

Under federal and Georgia state constitutional law, the racially disparate enforcement of drug laws violates equal protection guarantees if it is undertaken with discriminatory intent or purpose. Contemporary racism in public institutions, however, is frequently subtle, diffuse, and systemic and less likely to be the result of the conscious prejudices of individual actors. As a result, the requirement of proof of intent has been a formidable barrier for victims of discrimination seeking judicial relief.60

International human rights law wisely does not impose the requirement of discriminatory intent. The International Convention on the Elimination of All forms of Racial Discrimination (CERD) defines discrimination as conduct that has the "purpose or effect" of restricting rights on the basis of race.61 CERD has been interpreted as requiring the elimination of practices which have an unjustifiable disparate impact upon a racial group. It proscribes, for example, race-neutral practices curtailing fundamental rights that unnecessarily create statistically significant racial disparities.62

Assessing whether the harsh impact of drug law enforcement on blacks is justified or necessary requires scrutiny of the goals of that enforcement and the methods used. Because the fundamental human right of equal protection of the law is at stake, more justification is required than, for example, the advantages to the police of following the path of least resistance-inasmuch as drug arrests are easier in certain neighborhoods which only coincidentally happen to be black. It is difficult to conceive of any justification for a pattern of life sentences in which such serious punishment is imposed almost exclusively on black offenders who are primarily small-scale, street level dealers.

In the context of growing debates nationwide over the use of the criminal law to address drug use, doubts about the fairness and justice of enforcing those laws disproportionately against minorities take on even greater significance. There are numerous policy alternatives to current patterns of criminal law enforcement that would reduce adverse racial disparities while continuing to respond to social concerns about public drug dealing and drug abuse.

Advocating specific drug policies is beyond the mandate of Human Rights Watch. As an international human rights group, however, we insist that the right to be free of discrimination cannot be sacrificed to drug control strategies. We recognize Georgia's interest in addressing the public health and social consequences of drug abuse. But the development of drug policies, including the nature and enforcement of criminal laws, must be built on the foundation of respect for racial equality. To assist Georgia to ensure that its drug policies are consistent with international human rights, we recommend:

* Georgia's public officials, lawmakers and the public at large should scrutinize the means used to enforce drug laws and when they have a disparate racial impact, assess their necessity in light of the state's drug objectives. Policies and practices that have a racially disparate impact and are not necessary to meet the state's drug control goals should be modified.

* Georgia should institute police department reporting mechanisms, for the larger cities at least, that will enable the state to monitor the racial impact of drug law enforcement choices made by the departments.

* Georgia should review the collection of data within the criminal justice system and undertake revisions regarding the design of reporting mechanisms and databases that are needed to improve the availability of accurate information relevant for research and policy analysis, including on the racial impact of drug law arrests, prosecution and sentencing decisions. At the very least, the state should ensure that data is gathered and made available that will enable assessment of whether life sentences for drug offenses are imposed in a non-discriminatory manner.

VI. ACKNOWLEDGMENTS

This report was written by Jamie Fellner, Human Rights Watch associate counsel. It was edited by Cynthia Brown and Allyson Collins and Lenee Simon provided production assistance. We are grateful to Chief Justice Robert Bentham, District Attorney Daniel Porter, District Attorney Spencer Lawton, Police Maj. William Shannon, Walter Britt, Gregory Smith and the other prosecutors, defense attorneys and police personnel in Georgia who spoke with us, providing critical insights and information. We also wish to acknowledge the generous assistance and data provided by Timothy Carr, of the Georgia Department of Corrections, and Willene White-Smith of the Crime Information Center of the Georgia Bureau of Investigations. Dr. Joseph Katz provided invaluable help with the statistical analyses that are the core of this report.

This report was produced as part of Human Rights Watch's special initiative on the human rights consequences of drug programs and policies. We are grateful to the Open Society Institute for its funding of this initiative.

***

Human Rights Watch

Human Rights Watch is a nongovernmental organization established in 1978 to monitor and promote the observance of internationally recognized human rights in Africa, the Americas, Asia, the Middle East and among the signatories of the Helsinki accords. It is supported by contributions from private individuals and foundations worldwide. It accepts no government funds, directly or indirectly. The staff includes Kenneth Roth, executive director; Cynthia Brown, program director; Holly J. Burkhalter, advocacy director; Barbara Guglielmo, finance and administration director; Robert Kimzey, publications director; Jeri Laber, special advisor; Lotte Leicht, Brussels office director; Juan Méndez, general counsel; Susan Osnos, communications director; Jemera Rone, counsel; and Joanna Weschler, United Nations representative. Robert L. Bernstein is the chair of the board and Adrian W. DeWind is vice chair.

1 See Marc Mauer and Tracy Huling, Young Black Americans and the Criminal Justice System: Five Years Later, (Washington, D.C.: The Sentencing Project, October 1995). According to their analysis, African-Americans constitute 34.7 percent of arrests for drug possession nationwide and African-Americans and Hispanics constitute almost 90 percent of drug possession offenders sentenced to state prison.

2 Michael Tonry, Malign Neglect: Race, Crime and Punishment in America, (New York: Oxford University Press, 1995), p.105. See also, Alfred Blumstein, "Racial Disproportionality of U.S. Prison Populations Revisited," 64 University of Colorado Law Review 743 ( 1993).

3 In preparing this report, Human Rights Watch conducted a series of interviews in Georgia with police officials, prosecutors, defense attorneys and the chief justice of the Georgia Supreme Court.

4 We use the term "drug" to refer to controlled substances covered by Chapter 13 of Title 16 of the criminal code of Georgia.

5 See, for example, Article 2, the Universal Declaration of Human Rights; Article 2, International Covenant on Civil and Political Rights; Article 1, American Convention on Human Rights. See generally, Warwick McKean, Equality and Discrimination under International Law, (Oxford, England: Clarendon Press, 1983).

6 CERD has been described as "the international community's only tool for combating racial discrimination which is at one and the same time universal in reach, comprehensive in scope, legally binding in character, and equipped with built-in measures of implementation." 33 UN GAOR Supp. (No.18) at 108, 109 UN Doc. A/33/18 (1978) cited in Theodor Meron, "The Meaning and Reach of the International Convention on the Elimination of All Forms of Racial Discrimination," 79 The American Journal of International Law 283 (1985).

7 U.N. Doc. CERD/C/SEWER.967 at par.32 (introductory comments of Mr. Wolfrum). In an October 27, 1995 letter to Secretary of State Warren Christopher, Human Rights Watch, the International Human Rights Law Group and the NAACP Legal Defense and Educational Fund urged the United States to address the question of racial discrimination in the enforcement of drug laws in its submission reporting on U.S. law and practice relating to race discrimination to the United Nations Committee on the Elimination of Racial Discrimination.

8 The raw arrest data utilized in this report was provided by the uniform crime reporting program of the Georgia Crime Information Center (GCIC), a division of the Georgia Bureau of Investigation. Incarceration data was provided by the Georgia Department of Corrections (GDC).

9 This report looks only at statewide aggregate data. It does not address local variations in law enforcement practices or drug markets.

10 A review of possible racial bias in different aspects of the Georgia criminal justice system was undertaken by the Georgia Supreme Court Commission on Racial and Ethnic Bias in the Court System, Let Justice Be Done: Equally, Fairly, and Impartially, (Atlanta: Administrative Office of the Courts, August 1995).

11 Official Code of Georgia Annotated (O.C.G.A.)§§16-13-24, 30 and 31 (1995).

12 Georgia law does not distinguish between forms of cocaine, e.g., crack and powder.

13 Data provided by the GCIC include the number of arrests by Georgia police, race of arrestees and drug offenses involved. Arrests by federal agents are not included. The GCIC classifies an arrest according to the most serious crime or charge. If, for example, a person is arrested possessing marijuana and trying to sell cocaine, the arrest is classified as a cocaine sale arrest.

14 According to the 1990 U.S. census, the total population of Georgia is 6,478,216. The number of people classified as white is 4,600,148 ( 71 percent of the total); the number of people classified as black is 1,746,565 ( 27 percent of total), and the number of people classified as all other races combined is 131,503 ( 2 percent of the total). In this report, we address only the impact of the criminal justice system on whites and blacks. The number of persons from other races arrested and imprisoned for drug offenses is minuscule. Neither the GCIC nor the GDC classify hispanics separately from blacks and whites.

15 GCIC figures on the number of arrests for cocaine include arrests for opium and its derivatives (e.g. heroin). Most of the arrests in this category are, however, for cocaine.

16 Rates were calculated on basis of figures for white and black adults over the age of eighteen contained in 1990 census.

17 The total number of arrests for marijuana is considerably less than for cocaine, even though one can assume that in Georgia, as in the nation, marijuana is the most widely used drug. For example, in 1994, marijuana users comprised approximately 80 percent of current (past month) drug users nationwide. The Substance Abuse and Mental Health Services Administration ( SAMHSA) of the U.S. Department of Health and Human Services, "Preliminary Estimates from the 1994 National Household Survey on Drug Abuse," (Rockville, MD: SAMHSA, September 1995), p. 20. SAMHSA conducts annual surveys of drug use based on voluntary household interviews with a nationwide statistical sample. On the basis of these surveys, SAMHSA publishes calculations of the rate or prevalence of drug use by different population categories, including by race, as well as estimates of the total numbers of drug users within those population categories.

18 A household survey of drug use in Georgia has been initiated by the Division of Mental Health, Mental Retardation and Substance Abuse of the Georgia Department of Human Resources and is scheduled to be completed in September, 1996.

19 Human Rights Watch interview, Daniel Porter, district attorney for Gwinnett County, Lawrenceville, Georgia, March 5, 1996.

20 Office of National Drug Control Policy, Pulse Check: National Trends in Drug Abuse (Washington, D.C.: Fall, 1995), p.22; Pulse Check, Summer, 1995, p.20. Pulse Check reports on illegal drug use trends based on information ONDCP obtains from police, ethnographers and epidemiologists working in the drug field. Trends in Atlanta are routinely included. The reports do not, however, provide statistical data on the total numbers or proportions of different races using controlled substances. See also, Claire Sterk-Elifson, Kathleen Dolan, "Metropolitan Atlanta Drug Abuse Trends," in Proceedings of the Community Epidemiologic Working Group (National Institute of Drug Abuse, December, 1994).

21 In the mid-1980s, federal government surveys found significant regional differences in drug use. By the mid 1990s, however, those differences, particularly with regard to cocaine use, had largely disappeared.

22 Drug use rates provide a reasonable proxy for possession rates.

23 The SAMHSA household surveys provide the most comprehensive national data on drug use, but they do not include institutionalized persons, homeless persons not living in shelters and people with less stable residences generally. In this report, Human Rights Watch has used the SAMHSA national figures for the years 1991-1993 and the preliminary estimates for 1994.

24 SAMHSA defines current users as those using drugs at least once within the month preceding the survey date. Human Rights Watch calculated the use percentages for each race from SAMHSA figures on the estimated total number of drug users and the figures for each race. The total drug-using population nationally includes other race and ethnic groups. The SAMSHA surveys also count Hispanics as a separate drug using population. We have not included their use in our calculations both because other races and Hispanics constitute less than 2 percent of the population in Georgia and because they are a small percentage of the total population of drug users nationwide.

25 SAMHSA's surveys have consistently shown that in absolute numbers, far more whites use illicit drug, including cocaine, than blacks. SAMHSA data also belies the stereotype prevalent in the U.S. media that crack users are poor African-Americans. According to the 1994 survey, for example, 292,000 whites were current users of crack cocaine compared to 161,000 blacks. SAMHSA, Population Estimates for 1994, September 1995, Table 5 B and D.

26 SAMHSA surveys indicate that between 1991 and 1994, the average percentage of blacks who were current cocaine users was 1.35 percent or 1,350 per 100,000; for whites the average was .55 percent or 550 per 100,000.

27 The average annual arrest rate for cocaine possession for blacks was 826 per 100,000 versus 53 per 100,000 for whites. Even assuming the figures on black use may differ from actual use by a factor of 100 percent, the difference between the arrest rates and the use rates for blacks would still be significant.

28 ONDCP, Pulse Check, Summer 1995.

29 Beginning in 1991, during the household survey SAMHSA asked respondents whether they had sold any illicit drugs during the preceding year. One can assume that self-reporting on illegal conduct may be conservative, and that withholding information would more prevalent with regard to drug selling.

30 Patrick Langan, "The Race Disparity in U.S. Drug Arrests," unpublished manuscript, September 21, 1995. Langan is a senior statistician with the Bureau of Justice Statistics, U.S. Department of Justice. According to Joseph Gfroerer, Chief of Prevalence Branch, Office of Applied Studies, SAMSHA, the results of the questions on drug selling, along with other questions on criminal activity, are not included in the published household survey reports, but are available from SAMSHA.

31 See Charles R. Lawrence, "The Id, the Ego, and Equal Protection: Reckoning with Unconscious Racism," 39 Stanford Law Review 317 (1987).

32 M. Tonry, Malign Neglect, p. 106. See also, Alfred Blumstein, "Youth Violence, Guns, and the Illicit-Drug Industry", 86 The Journal of Criminal Law and Criminology 10, 29 (1995). Other logistical factors may be important as well. For example, low-income purchasers of cocaine buy the drug in the cheap form of single or several hits of crack. They must engage in far more illegal transactions to satisfy their desire for drugs than middle-class consumers of powder cocaine who have theresources to buy larger and longer lasting supplies. The greater frequency of purchases may affect the arrest rates. See A. Blumstein, "Youth Violence," p. 30.

33 As one Georgia public official, who requested anonymity, succinctly explained to Human Rights Watch:"When you want to catch fish, you go where the fishing is easiest."

34 Human Rights Watch interview with Robert Benham, chief justice of the Supreme Court of Georgia, Atlanta, March 4, 1996.

35 Human Rights Watch interview with Police Maj. William Shannon, Atlanta, March 4, 1996.

36 According to the 1990 census, the per capita income of white persons in urban Georgia was three times that of blacks and twice that of blacks in rural areas. One-third of all black Georgians are below the poverty line, compared to less than 10 percent of white persons. At the other end of the income scale, approximately 55,000 black households had incomes of more than $50,000 compared to 475,000 white households. We are not aware of any statistical studies of the econmic status of people arrested and incarcerated in Georgia for drug offenses-or other crimes. The consensus, however, is that most of the drug offenders who are arrested are low income.

37 Law professor Michael Tonry argues cogently that policy makers should be held accountable morally and politically for the foreseeable racially disparate impact of the "war on drugs". He advocates the approach used, for example, in criminal law, where acting with knowledge of likely effect can be as culpable as acting with specific intent to cause that effect. Tonry, Malign Neglect, pp.4-5.

38 There are no reliable, comprehensive data available with which to evaluate the racial impact of decisions made at the numerous decision points in the criminal justice system following arrest and prior to incarceration

39 The total of 41,068 persons does not include persons of races other than white and African-American. According to statisticians with the Georgia Department of Corrections, there are extremely few inmates classified as neither white nor black.

40 Table 5 includes persons who may have been convicted of more than one drug offense and who may also have been convicted of non-drug offenses. Data from the Department of Corrections suggests that between 1990 and 1995 blacks received approximately half of the straight probation sentences that were awarded to drug offenders.

41 Incarceration rates were calculated by Human Rights Watch on the basis of Georgia Crime Information Center arrest data and Georgia Department of Corrections prison admission data. In order to permit the most acurate comparison possible with arrest data, calculation of imprisonment rates was based on prisoners with only one drug offense type for the current conviction and who were not also serving time for a non-drug felony. The actual number of inmates serving time at least in part because of one or more drug offenses is greater. If we use that larger pool of imprisoned offenders, the incarceration rate for blacks is 18.9 percent and for whites is 8.5 percent.

42 This is not surprising, as marijuana is commonly considered the least dangerous of the illicit drugs, Georgia law punishes marijuana offenses less harshly than cocaine, and prosecutors and judges reputedly are also more lenient in charging and sentencing decisions for marijuana.

43 For each drug (cocaine, marijuana and other), the average imprisonment rate was calculated by aggregating black and white offenders. The three average imprisonment rates were used to compute the expected number of white and black offenders who would be imprisoned after being arrested for a drug offense.

44 Department of Corrections data suggest there is little difference between white and black drug offenders with regard to prior incarcerations. They show that, between 1990 and 1995, 61 percent of black offenders and 67 percent of white offenders admitted to prison had not previously been incarcerated. On the other hand, 77 percent of black offenders and 39 percent of white offenders who were incarcerated for a drug offense conviction also were serving time for a non-drug felony. Department of Corrections databases do not include data that would permit us to ascertain whether drug inmates were previously convicted for other crimes, whether their history included convictions for which no prison time was served, or even how many times they had been previously incarcerated.

45 The different incarceration rates for white and black drug offenders may, of course, also reflect biases that are not specific to drug cases but which operate throughout the criminal justice system. As the Georgia Supreme Court Commission on Racial and Ethnic Bias concluded, for example,"there are still areas within the state where members of minorities, whether racial or ethnic, do not receive equal treatment from the legal system...[M]ore frequently than intentional acts, there are incidences of bias which appear to result from unintentional conduct or conduct resulting from a lack of awareness....[Moreover,] the system is biased against economically disadvantaged individuals." Georgia Supreme Court Commission, Let Justice Be Done, p. 9.

46 To try to get as accurate a picture as possible of sentences for comparable drug offenses by comparable offenders within the limitations of the available data, we have looked at the sentence length of inmates who have no current non-drug felony conviction, have no prior record of incarceration, who did not receive a life sentence, and who have only one drug offense type for the current conviction.

47 The Department of Corrections records do not specify whether a life sentence was imposed under O.C.G.A..§§ 16-13-30(d) for drug offenses or for some other offense such as murder or kidnaping. We added a non-drug offense variable to the data classification to screen out any offenders who might have been sentenced for a drug offense but who received a life sentence for a non-drug crime.

48 See Mays v. State, 262 Ga. 90 (1992). In Stephens v. State, 265 Ga. 356, 360 (1995), Justice Thompson noted in his concurring opinion that "O.C.G.A.§§ 16-13-30(d) has been converted from a mandatory life sentence statute into a statute which imposes a life sentence only in those cases in which a district attorney, in the excercise of his or her discretion, informs a defendant that the state is seeking enhanced punishment."

49 Human Rights Watch's review of the life sentences given in different judicial circuits suggests considerable variation in the practices of district attorneys, with some rarely seeking life sentences and others applying the law more consistently.

50 Human Rights Watch telephone interview with Spencer Lawton, district attorney for the eastern judicial circuit of Georgia and president of the Prosecuting Attorneys' Council, Savannah, December, 1995.

51 Human Rights Watch interview with Daniel Porter, district attorney for Gwinnett County, Lawrenceville, March 5, 1996. Porter objected particularly to fact that under the statute small dealers faced mandatory life sentences while major trafficker did not.

52 Percentage calculated from Georgia Department of Corrections data.

53 Limitations in the Department of Corrections database preclude a definitive calculation of the number of offenders who were eligible for life sentences under O.C.G.A.. §§16-13-30(d). Our calculations are based on the most current data available, through December 1995, and reflect the most careful analysis possible within the constraints imposed by the coding of relevant variables in the database. The number of persons eligible for life sentences was considered both by the Georgia Supreme Court and by Georgia's Supreme Court Commission on Racial and Ethnic Bias in the Court System. In Stephens v. State, a 1995 case, the court was presented with data indicating that a "life eligible" African-American had a one in six chance and an eligible white had a one in 167 chance of receiving a life sentence. 265 Ga.356, at 359. The data before the court, however, overcounted the number of persons "eligible" for life sentences because it included offenders convicted of marijuana offenses, although marijuana offenses are not included within the life sentence statute. The Supreme Court Commission looked at data that were broader than that considered in Stephens (it included individuals whose convictions may have included probation or split sentences, whereas the Stephens court looked only at convictions resulting in incarceration) and that covered a longer time frame. The Commission's analysis showed that 0.5 percent of the white offenders having two or more convictions for drug sales received a life sentence compared to 5.7 percent of black offenders.

54 Persons sentenced to life imprisonment for drug offenses are eligible for parole. Human Rights Watch does not have figures indicating how long, on the average, inmates sentenced to life actually serve before release on parole. Those released on parole, however, face a life-long threat of being returned to prison for any subsequent infraction or crime.

55 Lisa Reid, "Drug Offenders with Life Sentences: A Profile," a report prepared by the Georgia State Board of Pardons and Paroles in 1992 and updated as of May 4, 1994.

56 The Supreme Court Commission concluded that the statistics it had reviewed on application of life sentence statute "demonstrate that the outcome of these drug offense cases differ significantly along racial lines" and called for further study on the issue. Supreme Court Commission, Let Justice Be Done, p. 165.

57 E.g. Stephens v. State, 265 Ga. 356 (1995); Hailey v. State, 263 Ga. 210 (1993); Hall v. State, 260 Ga. 596 (1992).

58 The new legislation also expanded the number of drugs that would be covered by the statute.

59 Human Rights Watch, Modern Capital of Human Rights? Abuses in the State of Georgia (New York: Human Rights Watch, June 1996), pp.35-59.

60 See "Developments in the Law: Race and the Criminal Process," 101 Harvard Law Review 1520 (1988).

61 International Convention on the Elimination of All Forms of Racial Discrimination, Par. I, Article 1, 3. In Centre for Human Rights, Human Rights: A Compilation of International Instruments, Vol., ST/HR/1/REV.5 (New York: United Nations, 1994), p. 66. Also available at http://www.un.org/Depts/Treaty/.

62 See CERD, General Recommendation XIV(42) on article 1, paragraph 1, of the Convention, U.N. GAOR, 48th Sess., Supp. No. 18, at 176, U.N. Doc. a/48/18(1993). See also, Theodor Meron, "The Meaning and Reach of the International Convention on the Elimination of All Forms of Racial Discrimination," 79 The American Journal of International Law 283, 287-88 (1985).