Angela Davis - Prosecution and Race: The Power and Privilege of Discretion

October, 1998

67 Fordham L. Rev. 13

ARTICLE: PROSECUTION AND RACE: THE POWER AND PRIVILEGE OF DISCRETION

Angela J. Davis. Associate Professor of Law, American University, Washington College of Law. B.A., Howard University; J.D., Harvard Law School; former staff attorney, Deputy Director and Director of the Public Defender Service for the District of Columbia. I owe a special debt of gratitude to Professors Paul Butler, Samuel Gross, and Tracey Meares, who read and critiqued numerous drafts and spent countless hours helping me develop the ideas for this article. I also thank Professors Jamin Raskin, Ira Robbins, and Carol Steiker for their helpful comments and suggestions on various drafts and Michael Barbosa for his helpful research assistance. This article also benefited from the advice and recommendations of members of the law faculty at American University, the University of Michigan, Rutgers, and Washington and Lee. Finally, I am most grateful to Sean O'Brien, whose superb research assistance, expert editorial advice, and insightful suggestions were essential to the completion of this article.

TEXT:

[*13] 

For everyone to whom much is given, from him much will be required; and to whom much has been committed, of him they will ask the more. n1

IT was the happiest day of David McKnight's life. n2 That evening, he went to a bar in Washington, D.C., to celebrate. Mr. McKnight bought a bottle of Dom Perignon and popped it open ceremoniously. "Drinks for everybody - my treat!," he announced. "What are we celebrating?," someone asked. "I killed someone and got away with it!," replied Mr. McKnight. He had just learned that a District of Columbia grand jury had voted not to indict him for the murder of John Nguyen.

The year was 1987. I was a staff attorney at the Public Defender Service for the District of Columbia ("PDS"). The court had appointed Michele Roberts, the chief of our trial division, to represent Mr. McKnight. The case was one of the most peculiar I observed in my dozen years as a public defender in the nation's capital. n3 Two factors were noteworthy. First, someone had been brutally killed, and the grand jury, with a silent and consenting prosecuting attorney, decided that the killer should go free. Second, the accused killer was [*14] white. n4 The way the case was handled convinced me that the two factors were related.

David McKnight was a twenty-five year old white male who worked as a bartender in a restaurant in Washington, D.C. He lived in a small, one bedroom apartment that he shared with John Nguyen, a fifty-five year old Vietnamese immigrant who worked as a cook in the restaurant. Mr. Nguyen paid Mr. McKnight rent to sleep in the walk-in closet of the apartment, a space barely large enough for a small bed.

One Saturday evening, McKnight invited a woman friend to the apartment. Nguyen was in the apartment at the time, resting in his "bedroom" closet. McKnight tried to persuade the woman to spend the night with him, but she declined. After she left, Nguyen came out of the closet and teased McKnight about his failed romance. Already upset about the turn of the evening's events, McKnight became even more enraged. He attacked Nguyen with a large machete. McKnight was much taller and heavier than Nguyen, who was just over five feet tall. Nguyen was able to escape into the bathroom, but McKnight hacked the bathroom door open with the machete. He then "almost sliced [Mr. Nguyen] in half." n5 Nguyen managed to stagger out of the apartment and into the street. Both men were covered with Nguyen's blood. Ironically, the first ambulance on the scene picked up McKnight, leaving Nguyen dying in the street. A second ambulance came for Nguyen and took him to the hospital. He died later that night.

Michele Roberts was appointed to represent McKnight on the following Monday morning. n6 The United States Attorney's office charged McKnight with assault with a dangerous weapon, not murder, despite the fact that Nguyen had died the previous Saturday night. The magistrate released McKnight pending the trial of his case.

The case never went to trial. The prosecutor, who was white, called Ms. Roberts within a day or two and invited her to identify witnesses who might testify before the grand jury on behalf of McKnight. n7 The prosecutor suggested that McKnight might have a good claim of self- [*15] defense and thought there might be witnesses who could testify about Nguyen's reputation for violence and McKnight's peaceful reputation. Ms. Roberts was stunned. She had been a trial lawyer at PDS for seven years and had probably tried more homicide cases than any other lawyer in the office. As the trial chief of the office, she had supervised most of the homicide cases handled by PDS. Ms. Roberts had never before received or heard of such an offer by a prosecutor to assist a criminal defendant, especially one who may have been guilty of murder.

Ms. Roberts identified witnesses willing to testify on behalf of McKnight. Although defense attorneys are not allowed to be present during grand jury hearings, the witnesses informed her that they would testify about McKnight's good character. Several weeks later, the prosecutor informed Ms. Roberts that the grand jury voted not to indict McKnight. All charges were dismissed. Ms. Roberts was never told about the testimony before the grand jury or whether the prosecutor advocated for McKnight. n8

Did race have anything to do with the outcome of McKnight's case? My colleagues and I were convinced that it did. It seemed obvious to us that the fact that McKnight was white and his victim was a Vietnamese immigrant had everything to do with the prosecutor's unusual attitude about prosecuting the case. Most of my colleagues and I had handled cases with much stronger evidence of self defense, but never had the prosecutor offered to present this exculpatory evidence to the grand jury. These cases generally proceeded to trial.

Mr. McKnight's case is just one example of many cases I handled or observed during my dozen years as a public defender in Washington, D.C., in which it appeared that the defendant or victim was treated differently based on his race. Almost always, this disparate treatment was the result of action taken by the prosecutor at the charging, plea bargaining, trial, or sentencing stage of the case. This phenomenon was so common that the attorneys in my office assumed that if the victim in a particular case was white, the defendant would surely be treated more harshly by the prosecutor. The plea bargain would be either unattractive or nonexistent and the prosecutor would devote more attention to the case. The converse was true for the few white clients represented by the office during my twelve-year stint. They [*16] would receive favorable treatment or, like Mr. McKnight, even have their cases dismissed. n9

At every step of the criminal process, there is evidence that African Americans are not treated as well as whites - both as victims of crime and as criminal defendants. n10 And because prosecutors play such a dominant and commanding role in the criminal justice system through [*17] the exercise of broad, unchecked discretion, their role in the complexities of racial inequality in the criminal process is inextricable and profound. n11

In this article, I examine prosecutorial discretion - a major cause of racial inequality in the criminal justice system. I argue that prosecutorial discretion may instead be used to construct effective solutions to racial injustice. n12 Prosecutors, more than any other officials [*18] in the system, have the power, discretion, and responsibility to remedy the discriminatory treatment of African Americans in the criminal justice process. Through the exercise of prosecutorial discretion, prosecutors make decisions that not only often predetermine the outcome of criminal cases, but also contribute to the discriminatory treatment of African Americans as both criminal defendants and victims of crime. I suggest that this discretion, which is almost always exercised in private, gives prosecutors more power than any other criminal justice officials, n13 with practically no corresponding accountability to the public they serve. n14 Thus, I maintain that prosecutors, through their overall duty to pursue justice, have the responsibility to use their discretion to help eradicate the discriminatory treatment of African Americans in the criminal justice system.

Courts have consistently upheld and sanctioned prosecutorial discretion, and make it increasingly difficult to mount legal challenges to discretionary decisions that have a discriminatory effect on African American criminal defendants and crime victims. n15 These challenges are usually brought as selective prosecution claims under the Equal Protection Clause, requiring a nearly impossible showing that the prosecutor intentionally discriminated against the defendant or the victim. One reason this standard is so difficult to meet is that much of the discriminatory treatment of defendants and victims may be based on unconscious racism and institutional bias rather than on discriminatory intent. Another reason is the exacting legal standard for obtaining discovery of information that would help to prove discriminatory intent when it does exist. n16

In this article, I suggest a solution that would promote equal protection of the laws through the electoral process and help address the difficult legal challenges to discriminatory treatment. I propose the use of racial impact studies in prosecution offices to advance the responsible, nondiscriminatory exercise of prosecutorial discretion. n17 [*19] The crux of the racial impact studies is the collection and publication of data on the race of the defendant and the victim in each case for each category of offense, and the action taken at each step of the criminal process. This data would then be analyzed to determine if race had a statistically significant correlation with various prosecutorial decisions. The studies would serve a number of purposes. First, they would reveal whether there is disparate treatment of African American defendants or victims. n18 Second, they may reveal the discriminatory impact of race-neutral discretionary decisions and policies. Third, they would help prosecutors make informed decisions about the formulation of policies and establish standards to guide the exercise of discretion in specific cases. Finally, the publication of these studies would inform criminal defendants, crime victims, and the general public about the exercise of prosecutorial discretion and force prosecutors to be accountable for their decisions. Publication of the information would help inform the general public about prosecutorial practices so they may more effectively hold prosecutors accountable through the electoral process. Publication may also help criminal defendants alleging race-based selective prosecution to overcome the strict discovery standard set by the Supreme Court in United States v. Armstrong.

Racial impact studies may reveal illegitimate differential treatment based on unconscious racism or class bias, or legitimately different outcomes based on disproportionate offending and uninterested or uncooperative victims. The studies may also reveal that there is no differential treatment, thereby invalidating misperceptions of unfairness. In any case, the availability of the information would be invaluable in improving the overall administration of criminal justice. As Justice Brandeis said, "Sunlight is said to be the best of disinfectants." n19

Part I of this Article explains the importance and impact of the prosecution function. Part II discusses the role of race and racism in discretionary prosecutorial decisions. Part III examines the difficulty in bringing successful legal challenges to discriminatory prosecutorial decisions. Part IV suggests the use of prosecutorial discretion as a remedy for race discrimination in the criminal justice system. Finally, Part V proposes the use of racial impact studies in prosecution offices to advance the responsible, nondiscriminatory exercise of prosecutorial discretion.

[*20]

I. The Prosecution Function

Like the decisions of many officials in the criminal justice system, prosecutorial decisions often have a discriminatory effect on African Americans. The decisions and actions of prosecutors - and thus the discriminatory effect of these decisions - have greater impact and more serious consequences than those of any other criminal justice official. This great influence and its consequences stem from the extraordinary, almost unreviewable, discretion and power of prosecutors.

The Supreme Court has consistently advanced a number of reasons for deferring to the exercise of prosecutorial discretion. These reasons include: "(1) promoting prosecutorial and judicial economy and avoiding delay; (2) preventing the chilling of law enforcement; (3) avoiding the undermining of prosecutorial effectiveness; and (4) adhering to the constitutional principle of separation of powers" and deferring to the expertise of prosecutors. n20 The Court has expressed concern that allowing inquiries into a prosecutor's motives might have a chilling effect on the performance of her law enforcement duties and may undermine her effectiveness by disclosing law enforcement policies. n21 Although these concerns may have some merit, n22 they cannot be used to justify the abuse of prosecutorial power.

Despite its potential abuse, however, prosecutorial discretion is necessary. It is difficult to imagine a fair and workable system that does not include some level of measured discretion in the prosecutorial process. One easily thinks of the prototypical case of a poor man who steals a loaf of bread to feed his starving family. Few would question the propriety or fairness of a prosecutorial decision to dismiss criminal charges against this criminal defendant. Such a decision could not be made in the absence of some level of discretion. n23

The deficiency of prosecutorial discretion lies not in its existence, but in the randomness and arbitrariness of its application. Even in prosecution offices that promulgate general policies for the prosecution of criminal cases, n24 there is no effective mechanism for enforce [*21] ment or public accountability. Self-regulation by prosecution offices is largely nonexistent or ineffective, n25 and Supreme Court jurisprudence has protected prosecutors from both public and judicial scrutiny. n26

The arbitrary use of prosecutorial discretion greatly exacerbates racial disparities in the criminal process. The collection and publication of data in the form of racial impact studies would educate prosecutors and the public about the racial effects of discretionary prosecutorial decisions. Before exploring these effects, though, it is important to understand the function and significance of prosecutorial decisions in the criminal process. The remainder of this part explores the prosecutor's role in charging and plea bargaining decisions.

A. The Charging Decision

The first and "most important function exercised by a prosecutor" is the charging decision. n27 Although police officers decide whether to arrest a suspect, the prosecutor decides whether he should be formally charged with a crime and what the charge should be. n28 This decision is entirely discretionary. n29 Even if there is probable cause to believe the suspect has committed a crime, the prosecutor may decide to dismiss the case and release the suspect. n30 She may also file a charge that is either more or less serious than that recommended by the police officer, as long as there is probable cause to believe the suspect commit [*22] ted the crime. n31 Other than a constitutional challenge by a criminal defendant, n32 there is very little process for review of these decisions. n33

Some states require that felony charges be formally instituted by a grand jury, n34 but the grand jury process is controlled entirely by the prosecutor. n35 The grand jury can be as small as five members and as large as about two dozen. n36 Grand jurors hear the testimony of witnesses, ask questions, and decide whether and with what offenses the suspect should be charged. n37 The prosecutor, however, usually decides which witnesses will be called, directs the questioning of those witnesses, interprets the law, and makes a recommendation to the grand jury. n38 Neither the target of the investigation nor his counsel may be present during the grand jury proceedings, n39 nor may they or [*23] any member of the general public be informed of the substance of the witnesses' testimony or any other details of these proceedings. n40

The charging decision is one of the most important decisions a prosecutor makes. In conjunction with the plea bargaining process, n41 the charging decision almost predetermines the outcome of a criminal case, n42 because the vast majority of criminal cases result in guilty pleas or guilty verdicts. n43 The charge also often determines the sentence that the defendant will receive, particularly in federal court, where criminal sentences are governed by the federal sentencing guidelines, n44 and in state cases involving mandatory sentences. n45 Because the sentencing guidelines and mandatory sentencing laws virtually eliminate judicial discretion, n46 the prosecutor often effectively determines the defendant's sentence at the charging stage of the process, if the defendant is eventually found guilty. n47 The charging decision's effect on the outcome of a case is felt to a lesser degree in state court. Although some state courts have some form of sentencing guide [*24] lines, n48 most states give judges more discretion in determining the sentence for a convicted defendant. Nonetheless, the range of penalties is set by the initial charging decision.

The gravity of the charging decision is epitomized by a prosecutor's decision to seek the death penalty. No state's laws require that the death penalty be sought in any particular case, n49 and all thirty-eight states that currently provide for the death penalty n50 leave the decision to the discretion of the prosecutor. n51 Most state death penalty laws allow the prosecutor to seek the death penalty in cases involving specific aggravating factors. n52 The decision to seek the death penalty, like all charging decisions, can only be challenged on constitutional grounds, n53 and these challenges are extremely difficult to sustain. n54

B. Plea Bargaining

Most criminal cases end in a guilty plea. n55 The typical plea bargain arrangement involves an agreement by the prosecutor to dismiss the most serious charge or charges in exchange for the defendant's guilty plea to a less serious offense. The defendant gives up his right to a trial and avoids the possibility of being convicted of a more serious offense and being imprisoned for a longer period of time. n56 Sometimes, the plea bargain offers the possibility of avoiding imprisonment entirely. Although the judge must approve plea bargains in most jurisdictions, judges routinely approve these agreements because they [*25] expedite the process by disposing of criminal cases without the time and expense of a trial. n57

Like the charging decision, the plea bargaining process is controlled entirely by the prosecutor and decisions are entirely within her discretion. A criminal defendant cannot plead guilty to a less serious offense unless the prosecutor decides to make a plea offer. n58 While the defense attorney may attempt to negotiate the best possible offer, the decision is ultimately left to the prosecutor's discretion. n59

Although prosecutors make important, influential decisions at other stages of the criminal process, the charging and plea bargaining stages provide the most independent power and control and allow the least opportunity for counterbalancing input from the defense. Prosecutors do not control the trial stage of the process and defense attorneys are theoretically on equal footing at this stage, as judges control the pace and content of the trial through their rulings. The sentencing hearing also provides for participation by the defense attorney and is controlled by the judge, unless mandatory sentencing laws are involved. In mandatory sentencing cases, which are increasingly common, the prior charging and plea bargaining decisions of the prosecutor often determine the sentence the defendant will receive, making the input of the defense attorney and the judge almost irrelevant.

II. Racism in the Prosecutorial Process

Because prosecutors are arguably the most powerful decisionmakers in the process, their decisions potentially have the greatest discriminatory impact. No discussion of the impact of prosecutorial discretion on racial disparities in the criminal justice system would be complete, however, without a discussion of the discretionary decisions of police officers. Prosecutors typically do not become involved in a criminal case unless and until a police officer makes an arrest. n60 If race is a factor in the decision to arrest a suspect, the police officer has [*26] infused the process with a layer of racial discrimination even before the prosecutor has an opportunity to exercise her discretion. n61

Racial impact studies in prosecutor offices may reveal racial disparaties in law enforcement. These studies may demonstrate that police are arresting African Americans and presenting them for prosecution at disproportionately higher rates than their similarly situated white counterparts. If such evidence is revealed, police should be compelled to explain or remedy the disparities. n62 The available literature suggests that such racial disparities frequently exist in the arrest stage of the criminal process. n63

A. Race and Police Discretion

Police officers often act in a discriminatory manner in the performance of their official duties when they disproportionately stop, detain, and arrest African American men, n64 with or without probable cause, [*27] and with or without articulable suspicion. n65 In fact, courts have upheld race as a legitimate factor in the decision to stop and detain a suspect. n66

One factor that contributes to discrimination at the arrest stage is the discretion afforded police officers in deciding who to stop and whether to make an arrest. Despite the requirement that a police officer's decision to stop a suspect must be based on an articulable suspicion, n67 the Supreme Court has shown increasing deference to the judgment of police officers in its interpretation of this requirement. n68 The practical effect of this deference is the assimilation of police officers' subjective beliefs, biases, hunches, and prejudices into law. n69 Because police officers are not required to make an arrest when they observe conduct creating probable cause, n70 their discretion may result in the failure to detain or arrest whites who commit acts for which their African American counterparts would often be detained or arrested. n71

[*28] Police officers engage in discriminatory conduct in numerous forums. Traffic stops provide a particularly egregious example. Police officers use alleged traffic violations as an excuse for stopping and detaining motorists they suspect of other criminal conduct. Once the driver is stopped, the police officer will use the opportunity to observe the interior of the car for any items that might be in "plain view." n72 The officer may also obtain the consent of an intimidated driver to conduct a full search of the car. n73 Even if the driver refuses to give the officer consent to search the car, the police officer often may arrest the driver, rather than simply issue a citation, as long as there is probable cause to believe that a traffic violation has been committed. n74 [*29] Once an arrest is made, the officer is justified in conducting a full- fledged search of the person and the surrounding area. n75 Police officers may use minor traffic violations as an excuse to stop individuals they would otherwise have no legitimate reason to detain. These so- called "pretextual stops" allow police officers to use any traffic violation, no matter how minor, as a justification for a stop and possible arrest. n76 Because police officers do not stop motorists every time they [*30] observe a traffic violation, n77 the stops can be used in a discriminatory manner. n78 The Supreme Court has unanimously upheld the constitutionality of pretextual stops, n79 despite studies showing that they are disproportionately used to stop and detain African Americans and Latinos. n80

The racial impact of discretionary arrest decisions is particularly significant for drug offenses. Although whites use drugs in far greater numbers than African Americans, n81 African Americans comprise a disproportionate percentage of drug arrests, convictions, and imprisonments. n82 Law enforcement agents concede that drugs are used and sold in middle and upper class neighborhoods and business districts, n83 but they have focused their law enforcement efforts in urban and inner-city areas which are populated primarily by African Americans and other people of color. n84 These decisions have obvious racial effects.

[*31]

B. Race and Prosecutorial Discretion

Although some prosecutors are involved in the investigatory stage of the criminal process, most prosecutors enter the process after an arrest is made. n85 A prosecutor may not know about racial considerations in the arrest process, and in most instances, does not have jurisdiction or control over the police department or other law enforcement agencies. n86 If a prosecutor is aware of the inappropriate or illegal consideration of race at the arrest stage of the process, she may legitimately decide to exercise her discretion to decline prosecution. The consideration of race in the arrest process is usually not obvious, however, and unless a prosecutor were intentionally attempting to ferret out such decisions, she may not discover them. Additionally, because prosecutors must rely on police officers to prosecute their cases successfully, they are not motivated to confront them with accusations of racism and discrimination.

Prosecutors should bear the brunt of the remedial responsibility to eliminate racism in the criminal process, even though inappropriate or illegal considerations of race may occur at the arrest stage, often before prosecutorial participation in the process. The Supreme Court's decision in Whren v. United States n87 has created the same kind of constitutional hurdles in cases involving police officers as Armstrong v. United States has created in selective prosecution cases - namely, the virtual impossibility of proving intentional discrimination based on race. n88 Police officers, however, have less power and opportunity than prosecutors to move beyond constitutional limitations. Although police officers may and should take steps to discover and eliminate the inappropriate consideration of race, n89 their power to affect and influence the criminal process begins and ends at the arrest stage. Prosecutorial power affects every stage of the process, including the arrest stage. n90

[*32] Like police officers, prosecutors often make decisions that discriminate against African American victims and defendants. n91 These decisions may or may not be intentional or conscious. Although it may be difficult to prove intentional discrimination when it exists, unintentional discrimination poses even greater challenges. Prosecutors may not be aware that the seemingly harmless, reasonable, race-neutral decisions they make every day may have a racially discriminatory impact. This discriminatory impact may occur because of unconscious racism - a phenomenon that plays a powerful role in so many discretionary decisions in the criminal process - and because the lack of power and disadvantaged circumstances of so many African American defendants and victims make it more likely that prosecutors will treat them less well than whites.

1. Unconscious Racism

If one acknowledges that African Americans experience both disparate and discriminatory treatment in the criminal justice system, n92 the discussion ultimately turns to the issue of blame. Whose fault is it? Who has committed the invidious act or acts that have caused African [*33] Americans to experience this discriminatory treatment? It is this intent-focused analysis, sanctioned by the Supreme Court in its equal protection analysis, n93 that has stymied legal challenges to discrimination in the criminal context. n94 Instead of focusing on the harm experienced by African Americans as a result of actions by state actors, the Court has focused on whether the act itself is inherently invidious and whether the actor intended to cause the harm. n95 In addition, the Court has placed the burden of proving intent on the shoulders of the victim. n96 If the victim is unable to prove the actor's bad intent or, in certain contexts, if the actor can establish a nondiscriminatory explanation for his behavior, n97 the Court offers no remedy for the harm experienced by the victim.

The main problem with this intent-focused analysis is that it is backward-looking. Although perhaps adequate in combating straightforward and explicit discrimination as it existed in the past, it is totally deficient as a remedy for the more complex and systemic discrimination that African Americans currently experience. When state actors openly expressed their racist views, it was easy to identify and label the invidious nature of their actions. n98 But today, with some notable exceptions, n99 most racist behavior is not openly expressed. More significantly, some racist behavior is committed unconsciously, n100 and many who engage in this behavior are well-intentioned people who would be appalled by the notion that they would be seen as behaving in a racist or discriminatory manner.

Unconscious racism, although arguably less offensive than purposeful discrimination, is no less harmful. In fact, in many ways it is more perilous because it is often unrecognizable to the victim as well [*34] as the perpetrator. And the Court, by focusing on intent rather than harm, has refused to recognize, much less provide a remedy for, this most common and widespread form of racism. n101 By focusing on blame rather than injury, the Court serves to satisfy the psychological needs of the uninjured party while leaving the victim without relief.

Professor Charles Lawrence defines unconscious racism as the ideas, attitudes, and beliefs developed in American historical and cultural heritage that cause Americans unconsciously to "attach significance to an individual's race and [which] induce negative feelings and opinions about nonwhites." n102 He argues that, although America's historical experience has made racism an integral part of our culture, most people exclude it from their conscious minds because it is rejected as immoral. n103 Professor Lawrence's definition of unconscious racism provides a useful framework in which to examine the discriminatory impact of prosecutorial decisionmaking.

2. The Discriminatory Impact of Race-Neutral Prosecutorial Decisionmaking

Most prosecutors today would vehemently deny that they have ever discriminated against African American defendants or victims or that they take race into account in any way in the exercise of their prosecutorial duties. Prosecutors exercise a tremendous amount of discretion without governing rules and regulations and are not legally required to exercise their discretion in any particular way. n104 Nonetheless, many prosecutors informally consider nonracial factors in making charging and plea bargaining decisions. n105 Factors that prosecutors frequently cite as reasons for making certain charging and plea bargaining decisions are: the seriousness of the offense, the defendant's prior criminal record, the victim's interest in prosecution, the strength of the evidence, the likelihood of conviction, and the availa [*35] bility of alternative dispositions. n106 These otherwise legitimate, race- neutral factors may be permeated with unconscious racism.

The seriousness of the offense is certainly a legitimate factor in making a charging decision. A prosecutor may decide to dismiss a case involving the possession of a single marijuana cigarette while charging and vigorously prosecuting a case involving distribution of a large amount of cocaine. Few would question this decision regardless of the race of the defendants. The more difficult issue arises when two cases involving the same offense but defendants or victims of different races are charged differently. If two murder cases involving similar facts with victims of different races are charged differently, the issue of unconscious racism becomes relevant. If a defendant in a case involving a white victim is charged with capital murder while a defendant in a similar case involving a black victim is charged with second-degree murder, questions arise about the value the prosecutors unconsciously placed on the lives of the respective victims. n107 A prosecutor may unconsciously consider a case involving a white victim as more serious than a case involving a black victim. n108 This unconscious view may influence not only the charging decision, but related decisions as well.

[*36] For example, if a prosecutor deems a particular case to be more serious than others, she will tend to invest more time and resources in that case, both investigating and preparing for trial. Such an increased investment would consequently yield more evidence and stiffen prosecutorial resolve. The likelihood of conviction is also obviously increased by the additional investment in investigation. Thus, although the strength of the evidence and the likelihood of conviction are facially race-neutral factors, they may be influenced by an unconsciously racist valuation of a case involving a white victim. n109

The victim's interest in prosecution is another legitimate factor that prosecutors consider in making charging and plea bargaining decisions. If the victim of a crime informs the prosecutor that he has no interest in the prosecution of his case and no desire to see the defendant punished, the prosecutor may legitimately dismiss the case based on the victim's feelings, especially if she believes that the defendant does not pose a danger to society and there are no other legitimate reasons for pursuing the prosecution. Few would question this decision, especially if the victim of the crime considered the prosecution process too onerous and difficult.

On the other hand, should a prosecutor pursue a prosecution in a case that she would otherwise dismiss for legitimate reasons simply because the victim wants to see the defendant punished? Or should a prosecutor assume that a victim is not interested in prosecution when the victim does not appear for witness conferences or respond to a subpoena? These questions demonstrate the significance of the intersection of class and race in the criminal process. n110 They also raise fundamental questions about the duty and responsibility of the prosecutor to seek justice for all parties - defendants as well as victims - and to assure that all parties receive equal protection under the law.

The prior record of the defendant is another legitimate, seemingly race-neutral factor considered by prosecutors in the charging/plea bargaining process. Defendants with prior records are more likely to be charged and less likely to receive a favorable plea offer. Prosecutors consider both arrest and conviction records; defendants with recidivist tendencies are arguably more deserving of prosecution. Race, however, may affect the existence of a prior criminal record even in the absence of recidivist tendencies on the part of the suspect.

[*37] As previously noted, race plays a role in the decision to detain and/or arrest a suspect. n111 Some courts have even legitimized this practice. n112 In addition, policy decisions about where police officers should be deployed and what offenses they should investigate have racial ramifications. n113 The fact that a white defendant has no criminal arrest or conviction record may not be a reflection of a lack of criminality on his part. If he lives in a neighborhood or attends a school that resolves certain criminal offenses (drug use, assault, etc.) without police intervention, he may be a recidivist without a record. Likewise, a black defendant who lives in a designated "high crime" area may have been detained and arrested on numerous occasions with or without probable cause. n114 Thus, the existence or nonexistence of an arrest or conviction record may or may not reflect relative criminality in black and white defendants. A prosecutor without knowledge of or sensitivity to this issue may give prior arrests undue consideration in making charging and plea bargaining decisions.

Another factor that prosecutors sometimes consider is the availability of alternative dispositions. For some less serious offenses, prosecutors may be willing to consider dismissing a case based on the existence of alternative resolutions that serve the overall interest of justice. For example, if a defendant who has stolen is able to make restitution and the victim is satisfied with this resolution and would be burdened by numerous court appearances, dismissal of the case may be the best disposition for all parties. The dismissal would also have the added benefit of eliminating the time and expense of trying another case for the prosecutor, the defense attorney, and the court. As with all of the otherwise legitimate considerations, however, this issue also has class and racial ramifications.

The murder case discussed in the introduction to this article demonstrates the complexity of these issues. There is no way of determining whether the prosecutor in Mr. McKnight's case considered any of the traditional nonracial factors in deciding not to vigorously prosecute this murder case, or if he did, which factors he considered. n115 Even if he did consider these factors, his decisions unconsciously may have been influenced by race.

The seriousness of the alleged offense - murder - would urge prosecution. The defendant had no prior record. In a murder case this factor would, at best, suggest leniency in sentencing or a favorable plea bargain, not total dismissal of the case. As far as the victim's interest was concerned, Mr. Nguyen had no relatives in the District of Columbia willing or able to lobby the prosecutor to vigorously prosecute the [*38] case, nor did the prosecutor seek to locate interested family members. Although the strength of the government's case was unclear, the prosecutor's failure to actively investigate the case certainly would have contributed to any weaknesses in the government's evidence and, consequently, the likelihood of conviction.

The prosecutor would probably deny that race or class had anything to do with the decisions made in McKnight's case. His unconscious racial biases, however, may have played a significant role in the process. It is doubtful that the white male prosecutor empathized with the middle-aged Vietnamese immigrant; it is likely that he would identify with the defendant who was a white male college student. In the absence of family members or others to advocate on his behalf, Mr. Nguyen was almost invisible - a foreign person of color whose life had little or no value. n116

Mr. McKnight's case illustrates how the discretionary decisions of prosecutors affect both offenders and victims in the criminal process. Although the prosecutor may not act invidiously, the discriminatory effects of his unconscious, race-neutral decisions may harm both victims and offenders. n117 Current legal remedies are totally inadequate to address this harm.

III. Legal Challenges to Discriminatory Prosecutorial Decisions

It is extremely difficult, if not impossible, for either victims of crime or criminal defendants to sustain legal challenges to the type of discretionary decisions described in the previous section. First, prosecutors ordinarily are not required to justify their discretionary decisions, either orally or in writing. n118 Second, even if such records were volun [*39] tarily kept in a particular prosecutor's office, they would not be available to the general public n119 and ordinarily would not be discoverable by other parties in a legal proceeding. n120 Finally, victims are not legal parties in criminal proceedings n121 and do not have standing to challenge prosecutorial decisions in the criminal setting. n122

Criminal defendants and crime victims claiming race-based prosecutorial policies or practices typically allege a violation of the Equal Protection Clause of the United States Constitution. n123 As the following discussion illustrates, it is extremely difficult to challenge such policies and practices successfully in light of equal protection law. The available legal avenues include a federal civil rights claim under 42 U.S.C. 1983 n124 and a motion to dismiss the indictment for selective prosecution based on race filed by a criminal defendant during the pendency of his case. n125

[*40]

A. Federal Civil Rights Claims

Both victims of crimes and criminal defendants may bring federal civil rights claims against prosecutors alleging racially selective prosecution in violation of the Equal Protection Clause. n126 A victim claiming discrimination would attempt to compel a prosecutor to bring charges against a particular person or persons on the ground that the prosecutor has given preferential treatment to such persons because of their race. n127 A criminal defendant may also bring a civil rights action, claiming that the prosecutor is seeking to prosecute him because of his race. n128

Courts have been unreceptive to these claims, making it difficult even to establish standing. n129 In fact, by raising the bar on standing requirements, the courts have avoided reaching the substantive issues. To establish standing, a plaintiff must show that she has suffered a direct injury-in-fact, n130 that the injury is traceable to the alleged unlawful conduct of the prosecutor, n131 and that the injury is redressable by the relief requested. n132 This standing requirement has been particularly onerous for African Americans seeking class action relief on the ground that prosecutorial practices have discriminated against African Americans as a group. n133 Even if the plaintiff is able to establish standing, she must prove discriminatory motive on the part of the prosecutor. n134

[*41]

B. Selective Prosecution in Criminal Cases

The selective prosecution claim is more commonly raised by a criminal defendant seeking dismissal of his case on the ground that he has been singled out for prosecution in violation of the Equal Protection Clause. n135 In Oyler v. Boles, n136 the Supreme Court held that selective prosecution violates the Constitution only if it is based on an unjustifiable standard such as race, religion, or some other arbitrary classification. n137 The Court also held that the selectivity must be purposeful or intentional. n138 After Oyler, a number of circuit courts accepted evidence of disproportionate impact as adequate proof of intentional discrimination. n139

The Second Circuit's decision in United States v. Berrios reversed the trend of accepting evidence of disparate impact as proof of purposeful discrimination. n140 Mr. Berrios claimed he was singled out for prosecution because of his outspoken opposition to President Nixon. n141 The court rejected his claim, establishing a more strict two- pronged analysis. The court held that a defendant claiming selective prosecution must establish: (1) that others similarly situated have not been prosecuted; and (2) that the decision to prosecute him was discriminatory and made invidiously or in bad faith. n142 Thus, both discriminatory effect as well as purpose was required. n143

The requirement of an independent showing of discriminatory purpose was solidified by the Supreme Court's landmark equal protection decision in Washington v. Davis. n144 In Davis, African Americans claimed that a civil service examination discriminated against them on the basis of race. The Court held that they must prove discriminatory purpose independent of disproportionate impact to prove that a facially neutral law violates the Equal Protection Clause. n145 Lower [*42] courts subsequently applied the strict intent requirement in the selective prosecution context. n146

In 1985, the Supreme Court decided Wayte v. United States, n147 directly applying the strict intent standard to a selective prosecution case. In Wayte, the defendant claimed that he had been prosecuted because he had written letters to the President and other government officials informing them of his refusal to register for the draft. n148 The Court rejected his claim, holding that, even if there was evidence of discriminatory impact, Mr. Wayte must prove that the government intended to discriminate against him because of his protests. n149 The Court made it clear that a showing of discriminatory impact was not sufficient to infer a discriminatory motive. Thus, since Wayte, defendants must make a prima facie showing of discriminatory purpose and effect even to obtain an evidentiary hearing, and statistical evidence of disproportionate impact is insufficient to show discriminatory purpose. n150

C. United States v. Armstrong

In United States v. Armstrong, n151 the Supreme Court established the applicable standard for discovery in selective prosecution claims based on race. n152 The case is also significant because the Court affirmed its unwillingness to interfere with the exercise of prosecutorial discretion and sanctioned the difficult standard for a plaintiff or criminal defendant who alleges selective prosecution. n153 Furthermore, the Court created near-impossible constitutional limitations in the discovery stage of selective prosecution claims. n154 Armstrong affirms the Court's failure to provide meaningful judicial relief for race-based selective prosecution and establishes the need for creative non-judicial remedies.

[*43] In Armstrong, nine black defendants in Los Angeles were charged in federal court with conspiring to distribute and conspiring to possess with intent to distribute more than fifty grams of cocaine base (crack). They were also charged with various firearms offenses. The defendants filed a motion to dismiss the indictment for selective prosecution based on race. They claimed that the United States Attorney prosecuted virtually all African Americans charged with crack offenses in federal court, leaving all white crack defendants to be prosecuted in state court. Their claim was based on the fact that the federal law penalizes crack trafficking much more harshly than the California state law. n155

The defendants filed a discovery motion to obtain information in support of their claim. The information requested included the United States Attorney's criteria for deciding whether to bring charges in federal court and the number and racial identity of all defendants charged with crack offenses in both federal court and state court. The district court granted the motion. n156 The Court of Appeals for the Ninth Circuit affirmed the district court after a rehearing en banc, holding that a "colorable basis" for selective prosecution entitles a defendant to discovery and that a defendant is not required to demonstrate that the government has failed to prosecute other similarly situated defendants. n157

The Supreme Court reversed the Ninth Circuit decision, holding that, in order to establish entitlement to discovery in selective prosecution cases based on race, a defendant must produce credible evidence that similarly situated defendants of other races could have been prosecuted, but were not. n158 The Court held that the defendants in Armstrong did not meet this threshold. Before establishing the standard for discovery in selective prosecution cases based on race, the Armstrong Court reviewed and reaffirmed the requirements for prevailing on the merits in such cases. Citing Oyler v. Boles n159 and Wayte v. United States, n160 the Court reiterated the equal protection standard applicable in selective prosecution claims. The Court noted that the claimant must show discriminatory effect and purpose, and explained that, to establish discriminatory effect, the claimant must show that "similarly situated individuals of a different race were not prosecuted." n161

[*44] Obviously anticipating the difficulty in meeting this standard, the Court proclaimed that the "similarly situated" requirement does not make it impossible to prevail in these cases. n162 In support of this statement, the Court cited the only case in which it upheld a claim of race- based selective prosecution - Yick Wo v. Hopkins. n163 The plaintiff in Yick Wo was able to prove that similarly situated laundry operators of another race could have been prosecuted, but were not. n164 The Court's citation of Yick Wo is significant primarily because the case was decided 112 years ago. The Court obviously failed to recognize that 1886 race remedies are not applicable to 1998 race problems. The nature of racism and discrimination has changed significantly since 1886. n165 Racism and discrimination are not always overtly displayed or even intentional. n166 Although Justice Brennan's dissent in McCleskey v. Kemp n167 and Justice Marshall's concurrence in Batson v. Kentucky [*45] n168 acknowledged and explained unconscious racism n169 and its effect on prosecutors, the Armstrong opinion chose to ignore the subject.

The Court's citation of Yick Wo is also problematic because it was decided long before Wayte v. United States. Wayte imposed the strict intent requirement in selective prosecution cases - a requirement that the plaintiff in Yick Wo did not have to fulfill. Furthermore, the fact that no civil plaintiff or criminal defendant in a race-based selective prosecution case has prevailed in the Supreme Court in 112 years suggests the scant likelihood of satisfying the "similarly situated" requirement. n170

After reviewing the requirements for proving a selective prosecution claim based on race, the Court established the showing necessary to obtain discovery in support of such claims. To obtain discovery, a defendant must "show that the Government declined to prosecute similarly situated suspects of other races." n171 This showing is amazingly similar to the proof necessary to establish the "discriminatory effect" prong of the ultimate claim. n172 To prevail on the merits, in addition to establishing discriminatory intent, a defendant must prove that "similarly situated individuals of a different race were not prosecuted." n173 Other language in the opinion differentiates the standards only slightly. The Court stated that to obtain discovery, a defendant must produce "some evidence that similarly situated [individuals] could have been prosecuted but were not." n174 It also stated that the defendant must make a "credible showing of different treatment of similarly situated persons." n175

The Court failed to explain precisely the difference between proving a failure to prosecute similarly situated individuals of another race and producing "some evidence" or a "credible showing" of the same. [*46] One would assume that more proof is necessary to prevail on the merits than to obtain discovery, but how much more? And how much is needed to cross the discovery threshold? One white suspected drug dealer? Two? The Court was equally vague on what constitutes "credible" evidence. The sworn affidavit of a member of the bar did not suffice in Armstrong, n176 yet the Court expressed no similar concerns about the sworn affidavits of law enforcement agents.

The Court was particularly critical of the conclusion of the court of appeals that "people of all races commit all types of crimes." n177 The Court purported to dispute this conclusion by citing statistics of the United States Sentencing Commission. Once again, the Court's evidence in support of its conclusions failed to prove its point. As proof that people of certain races commit certain types of crimes, the Court cited the Sentencing Commission's statistics which show that "more than 90% of the persons sentenced in 1994 for crack cocaine trafficking were black ... ; 93.4% of convicted LSD dealers were white ... ; and 91% of those convicted for pornography or prostitution were white ...." n178 Thus the Court inappropriately cited conviction and sentencing statistics as proof of who commits these crimes.

The very reason for bringing a selective prosecution claim is to show that there are people committing the crimes who are not being prosecuted, convicted, and sentenced. As Justice Stevens pointed out in his dissent:

The presumption that some whites are prosecuted in state court is not "contradicted" by the statistics the majority cites, which show only that high percentages of blacks are convicted of certain federal crimes, while high percentages of whites are convicted of other federal crimes ... Those figures are entirely consistent with the allegation of selective prosecution. The relevant comparison, rather, would be with the percentages of blacks and whites who commit those crimes. n179

This flaw in the Court's reasoning is troubling on several levels. First, it could illustrate a basic failure to recognize that conviction and sentencing statistics do not accurately reflect crime rates. Second, it could reflect the Court's unwillingness to question the prosecutorial decisions of the Executive branch. Language in the Court's opinion suggests that, at a minimum, the latter view is true. Throughout the opinion, the Court extolled the importance of the prosecutor's broad discretion to enforce the laws n180 and the "presumption of regularity" [*47] supporting prosecutorial decisions. n181 The Court affirmed judicial deference to prosecutorial decisions, noting the competence of prosecutors in these areas and the concern that a court's examination of the basis of a prosecutor's decision would threaten to chill law enforcement n182 and "undermine prosecutorial effectiveness by revealing the Government's enforcement policy." n183

Armstrong leaves the ordinary criminal defendant with little hope that he might ever prevail on a race-based selective prosecution claim and even less guidance on how he might do so. Even if the Court had more precisely explained the quality and quantity of evidence necessary to cross the discovery threshold, it is doubtful that most criminal defendants would be able to meet that standard. Most criminal defendants are indigent n184 and thus incapable of hiring lawyers and experts to conduct the type of investigation and reports apparently necessary to obtain discovery. Indeed, the evidence presented by the federal defender in Armstrong was probably more than most overburdened court appointed attorneys would be able to produce. n185

Armstrong also affirms the inadequacy of current legal remedies to correct the discriminatory effect of prosecutorial decisions. Armstrong addresses the standard for discovery and affirms the standard for prevailing on the merits in race-based selective prosecution cases. The failure to prosecute similarly situated persons of another race only establishes discriminatory effect. n186 The defendant must also prove that the prosecutor was motivated by a discriminatory purpose. n187 The Armstrong Court does not address the amount of proof necessary for a showing of discriminatory purpose. One could only imagine that such proof would be impossible short of a frank admission of prejudice by a prosecutor or discovery of clearly prejudicial documents. n188 Given the nature of modern-day racism and discrimi [*48] nation, it is highly improbable that such evidence could be produced. n189

The Court's decision in McCleskey v. Kemp, n190 decided almost ten years prior to Armstrong, confirms the improbability that the Court will accept even highly incriminating statistical evidence as sufficient to prove intentional behavior. Warren McCleskey was an African American man who used a statistical study to challenge the imposition of the death penalty in his case after he was convicted of killing a white police officer in the State of Georgia. n191 The study, conducted by Professor David Baldus, provided stark statistical evidence of race discrimination in the implementation of the death penalty in Georgia. Professor Baldus examined thousands of murder cases in Georgia and the numerous variables that might influence the imposition of the death penalty. n192 Baldus took into consideration hundreds of such nonracial variables, including age, education, prior criminal record, and the strength of the evidence, and concluded that in Georgia, defendants who killed whites were 4.3 times more likely to receive the death penalty than defendants who killed blacks. n193

Mr. McCleskey claimed that his race and the race of his victim had played an unconstitutionally impermissible role in the decision to sentence him to death. n194 Mr. McCleskey's claim was ultimately rejected by the Supreme Court in a five to four decision. The Court held that the Baldus study did not prove that criminal justice officials in Mr. McCleskey's case, including prosecutors, had intentionally discriminated on the basis of race and that the Court would require "exceptionally clear proof" before inferring that a sentencing authority had abused its discretion. n195 The Court stated that the Baldus study did [*49] not meet this standard, n196 despite that many distinguished researchers have acknowledged the accuracy and legitimacy of the study's methodology. n197

As in Armstrong, the McCleskey Court extolled the virtues of prosecutorial discretion. n198 Given the sophistication of the Baldus study and the Court's deference to prosecutorial discretion, McCleskey and Armstrong lend little hope to criminal defendants seeking to present statistical evidence of discriminatory treatment by prosecutors. n199 The three separate dissenting opinions in McCleskey thoroughly discuss the evidence of racial discrimination in the implementation of the death penalty. n200 Even Justice Scalia, who joined the majority opinion in McCleskey, acknowledged the existence of race discrimination in the criminal process. In a memorandum to [*50] the other justices before the opinion was published, Justice Scalia wrote:

And I do not share the view, implicit in the opinion, that an effect of racial factors upon sentencing, if it could only be shown by sufficiently strong statistical evidence, would require reversal. Since it is my view that the unconscious operation of irrational sympathies and antipathies, including racial, upon jury decisions and (hence) prosecutorial decisions is real, acknowledged in the decisions of this court, and ineradicable, I cannot honestly say that all I need is more proof. n201

These cases and Justice Scalia's acknowledgement of the effect of unconscious racism on prosecutorial decisions suggest that prosecutors must be held accountable for the exercise of their discretion through nonjudicial means.

IV. Prosecutorial Discretion - Power, Privilege, and Ethical Responsibility

Given the inadequacy of current legal remedies to combat race discrimination in the criminal justice system, the Court's recent affirmation of broad prosecutorial discretion, and the high legal barriers erected to discourage selective prosecution claims, other remedies must be constructed and implemented. Officials in all three branches of government certainly have the responsibility to seek remedies for disparities and discrimination in the criminal justice system. Likewise, criminal justice officials (law enforcement officers, defense attorneys, prosecutors, judges, probation and parole officers, and corrections officials) should all have an interest in, and the responsibility for, eliminating discrimination within the system. Prosecutors, however, are uniquely positioned and empowered to remedy these injustices most effectively and efficiently.

As members of the Executive branch and highly specialized officials in the criminal justice system, prosecutors are in the unique position to use their discretion to eliminate many of the racial disparities in the criminal justice system. n202 No other official is empowered to effect such change. The Court's recent reaffirmation of prosecutorial power in Armstrong should encourage prosecutors to use it not only as a shield against claims of wrongdoing, but as a weapon against any [*51] wrongdoing in the criminal justice system, including race discrimination. n203

The elimination of race discrimination is totally consistent with the role of the prosecutor. It is the responsibility of the prosecutor to seek justice, not simply to win convictions. n204 As the Supreme Court noted in Brady v. Maryland:

Society wins not only when the guilty are convicted but when criminal trials are fair; our system of the administration of justice suffers when any accused is treated unfairly. An inscription on the walls of the Department of Justice states the proposition candidly for the federal domain: "The United States wins its point whenever justice is done its citizens in the courts." n205

The duty to seek justice is not limited to the prosecutor's responsibilities in individual cases, but also applies to the administration of justice in the criminal justice system as a whole. In fact, the prosecutor's duties include the oversight function of insuring the fairness and efficiency of the criminal justice system. n206 Those duties should include recognizing injustice in the system and initiating corrective measures. n207

The prosecutor's duties and responsibilities to the criminal justice system as a whole stem from her dual role as an advocate for the gov [*52] ernment and as an administrator of justice. n208 As administrator of justice, the prosecutor represents the interests of society as a whole, including the interests of the defendant as a member of society. n209 The prosecutor often experiences tension and conflict in her attempt to implement these dual roles. n210

The prosecutor's duty to combat discrimination in the criminal justice system exemplifies the conflict in the implementation of these dual responsibilities. On the one hand, the prosecutor has a duty as both an advocate for the government and a representative of society to protect the community through the enforcement of the criminal laws. On the other hand, she has a responsibility to ensure that victims and defendants in the criminal justice system are treated fairly, equitably, and in a nondiscriminatory manner. The latter responsibility is owed to individual victims and defendants and to society as a whole; society has an interest in a fair and nondiscriminatory criminal justice system. The two responsibilities do not present an insurmountable conflict. The simple answer is that the prosecutor must protect the community through the fair, equitable, and nondiscriminatory enforcement of the laws.

The difficulties arise not in the conceptualization of these responsibilities but in their implementation. The facts of United States v. Armstrong illustrate these difficulties. If the defendants in Armstrong are guilty of cocaine trafficking and firearms possession, the government [*53] and society have an interest in their prosecution and punishment. n211 The government and society, however, also have an interest in preventing both selective prosecution and harsher treatment of black cocaine traffickers. If a prosecutor becomes aware of criminal behavior by African Americans, how does she assure that they are not being singled out and treated more harshly than white offenders? Must she seek out similarly situated white offenders and prosecute them in proportionate numbers? Should she decline to prosecute black offenders until a proportionate number of similar white offenders are arrested? n212 These alternatives, although perhaps controversial, would be well within the lawful exercise of the prosecutor's discretion.

Prosecutors currently use their discretion not only in the prosecution of individual cases, but also in the implementation of general policies. n213 United States Attorneys or State's Attorneys may implement formal or informal general policies regarding the prosecution of certain types of cases. For example, a chief prosecutor may have a policy of prosecuting all cases involving firearms possession, domestic violence, or some other criminal behavior that may be particularly widespread in his community. Likewise, a prosecutor may institute a policy of dismissing cases involving small quantities of marijuana or of seeking a probationary sentence for the defendants in these cases. Each of these policies may reflect the attitudes, values and priorities of the prosecutor and/or the community about certain criminal behavior.

The interest and responsibility of both the prosecutor and the community in the nondiscriminatory treatment of defendants and victims in the criminal justice system should also be the basis for the implementation of general prosecutorial policies. As with most criminal justice policies, they would reflect a balancing of all of the interests at stake. The goal of the prosecutor's office should be the implementation of policies in a manner consistent with the overall administration of justice. n214

[*54]

V. Racial Impact Studies and the Remedial Use of Prosecutorial Discretion

Not every disparity is evidence of discrimination. Since many legitimate factors affect prosecutorial decisions, it may be appropriate to treat victims and defendants differently, even in similar cases. n215 A prerequisite to eliminating race discrimination in the criminal process is the determination of whether the dissimilar treatment of similarly situated people is based on race rather than some legitimate reason. Whether the treatment is intentional or purposeful should not matter - the goal should be elimination of harm. Thus, the first step is the implementation of racial impact studies designed to reveal racially discriminatory treatment. The second step is the publication of these studies so victims of discrimination and the general public may act to eradicate undesired policies and practices.

A. Racial Impact Studies

Racial impact studies would involve the collection of data on the race of the defendant and victim for each category of offense and the status of the case at each step of the prosecutorial process. For example, in each case involving an arrest for possession of cocaine, the prosecutor would document the race of the defendant, the defendant's criminal history, the initial charging decision, each plea offer made, accepted, or rejected, and the sentence advocated by the prosecutor. If relevant, the prosecutor should also document whether and how a decision was made to charge in federal versus state court and whether a departure from the sentencing guidelines was sought.

The statistics would be collected for each type of offense so that an appropriate statistical analysis comparing the disposition of the cases of white and African American defendants and victims could be done. These studies would not only be helpful in determining whether defendants of color receive harsher treatment for the same criminal behavior, but in cases involving victims, they would also demonstrate whether cases involving white victims were prosecuted more vigorously than cases involving African American victims. n216 The data would also indicate whether similarly situated defendants and victims of different races are treated the same at each step of the process. n217 [*55] Are defendants in cases involving white victims initially charged with the same offense as similarly situated defendants in cases involving black victims? Do they receive the same plea offers? Are the same sentences advocated at the sentencing hearing?

The data may help to reveal the extent to which whites are being arrested and presented for prosecution by law enforcement officers. If the majority of the cases in any particular category of offense involve African American defendants, the prosecutor should investigate further to determine whether African Americans comprise a majority of the population in that jurisdiction. If they do, the data would not necessarily indicate the selective detention and prosecution of African Americans. If African Americans do not comprise a majority of the population, further investigation would certainly be warranted, particularly if there is a considerable difference between the arrest rates and the African American population. The further investigation should attempt to determine whether African Americans commit the crime in question at greater rates than whites. In the absence of credible evidence that they do, the prosecutor should presume that no one particular race is inherently more likely to commit certain types of crimes. n218

Significant conclusions could not be reached from the simple collection of data without the appropriate statistical analysis. The Baldus study used in McCleskey v. Kemp exemplifies the model statistical analysis of this type of data. n219 Widely acclaimed as one of the most thorough and statistically sound analyses of sentencing, n220 the Baldus study examined thousands of murder cases over a seven-year period and took into account thirty-nine nonracial variables most likely to influence sentencing patterns in Georgia before reaching the conclusion that the race of the victim had a statistically significant correlation with the imposition of the death penalty. n221

Similar studies in prosecutors' offices would determine whether racial disparities exist in the prosecution of all types of cases and whether the disparities are statistically significant. A Baldus-type study which takes those factors into account would be essential to the [*56] credibility of the evaluation n222 because there are so many legitimate, nonracial factors that may be considered in prosecutorial decisions. n223 This type of evaluation would determine whether race is the determinative factor.

B. Publication of the Racial Impact Studies

The racial impact studies will indicate whether, and to what extent, disparate treatment of similarly situated victims and defendants is based on race. How, then, might these studies be used to help eradicate race-based disparities? How these studies are used depends, to a large degree, on whether they are made public. If the results of the studies are revealed only to the prosecutor, she would have no legal obligation to take any action. The difficult discovery standard established in Armstrong would rarely require a prosecutor to turn over a racial impact study in a selective prosecution case. n224 Similarly, difficult standing requirements in civil selective prosecution cases n225 make it unlikely that such cases will be brought, much less reach the discovery stage.

If the studies are revealed only to the prosecutor, she would be free either to do nothing or to take some action to eliminate the disparities. The Rules of Professional Responsibility suggest that the prosecutor would have an ethical obligation to take steps to remedy the disparities. n226 Prosecutors willing to voluntarily fulfill this obligation would face the difficult task of establishing and implementing some workable remedy. Prosecutors who do not view the elimination of racial disparities as a priority would do nothing. n227

Publication of the studies to the general public would be an important first step that might serve as a catalyst for developing workable remedies. The public could hold the prosecutor accountable through the electoral process by requiring that the disparities either be explained or remedied. In addition, litigants in selective prosecution actions could attempt to force similar action.

[*57]

1. The Electoral Process

Forty-three states hold popular elections for Attorney General. n228 At the county and municipal level, more than ninety-five percent of the chief prosecutors are elected. n229 These positions are highly political, and candidates usually campaign on general crime themes, not on specific proposals about how they plan to exercise their prosecutorial power. n230 Prosecutors are usually elected in the same general elections as other public officials. The state and county prosecutors hire assistant district attorneys to handle the caseloads of their offices.

Federal prosecutors are appointed, but their selection is also political. The President of the United States appoints the Attorney General who oversees the entire Justice Department. n231 The President also appoints a United States Attorney for each of the federal judicial districts. n232 The Attorney General may appoint additional Assistant United States Attorneys for any of the districts. n233 The Attorney General and each United States Attorney must be confirmed by the United States Senate. n234 Thus, the selection and confirmation of the Attorney General and the United States Attorneys are greatly influenced by the political party of the President and a majority of the Senate. Theoretically, the confirmation hearings provide an opportunity to inform the public of the practices and policies of a particular prosecutor since the hearings are open to members of the public, who may express their views by writing or calling their senators.

Ironically, the current system of choosing state and local prosecutors through the electoral process was established for the purpose of holding prosecutors accountable to the people they serve. The elected [*58] prosecutor emerged during the rise of Jeffersonian democracy in the 1820s, when the system of popularly elected officials was adopted. n235 No longer beholden to the governor or the court, n236 the prosecutor was deemed accountable to this amorphous body called "the people," specifically his constituents. Of course, the actions and decisions of the prosecutor were not generally a matter of public record, so the people's ability to hold the prosecutor accountable was quite limited. Nonetheless, the ballot box was seen as the most democratic and effective mechanism for achieving this goal. n237

The public's access to information about prosecutorial decisions has not expanded since the 1820s. The electorate has very little information about a prosecutor's specific charging and plea bargaining practices or how he plans to exercise his discretion before electing him to office, or, in the case of appointed prosecutors, before commenting on his appointment. n238 Elected prosecutors typically run on very general " [*59] tough on crime" themes with no information about specific office policies. n239 Certainly issues concerning race, such as strategies for preventing selective prosecution or other types of discriminatory treatment, are rarely discussed. n240 Because of the paucity of such relevant information, the Jeffersonian democratic ideals that inspired the first elected prosecutors in the 1820s have never been achieved. n241 Although the electorate can and does vote prosecutors out of office, it is not making these decisions in a fully informed manner.

The publication of racial impact studies would inform the public about the possible discriminatory effects of prosecution policies and practices. Such studies would force a public debate about racial disparities and compel prosecutors to be truly accountable to their constituents. Prosecutors could do this by either establishing policies and practices to help eliminate the disparities or by explaining that there are legitimate, race-neutral reasons for such disparities. n242 If the pub [*60] lic was not satisfied with the results of the study, the efforts to eliminate the disparities, or the prosecutor's explanation for disparities, it could then remove the prosecutor from office through the electoral process. n243 The public debate would also help the prosecutor to establish workable remedial policies and practices. Thus, public access to the studies would motivate prosecutors to correct inequities and help to make the electoral process a more meaningful check on unacceptable prosecutorial practices.

2. Selective Prosecution Claims

Publication of racial impact studies would also allow for their use in selective prosecution claims brought by either defendants or crime victims. To the extent that such studies reveal that a prosecutor used his dismissal power in cases involving white defendants rather than black defendants, a black defendant claiming selective prosecution would be able to prove that similarly situated white defendants could have been prosecuted, but were not. n244 Conversely, if a prosecutor dismisses cases involving black victims more than those with white victims, black victims claiming selective prosecution in civil cases would be able to prove that the cases of similarly situated white victims were prosecuted when cases of black victims were not prosecuted. In other words, the racial impact studies would obviate the need for discovery, thereby making it unnecessary to meet the impossible standard for discovery established in the Armstrong case.

The racial impact studies would not provide sufficient evidence to meet the standard of proof for prevailing on the merits in selective prosecution cases. Presumably, the studies would be sufficient to prove disparate impact if they reveal disparate dismissal rates based on race. They undoubtedly would be insufficient to prove the necessary discriminatory intent, however, in light of the Court's rejection of the Baldus study in McCleskey. n245

[*61] At least one scholar has suggested that the legal process achieved through selective prosecution hearings serves a useful purpose despite the improbability of success on the merits. Professor Anne Bowen Poulin argues that courts should continue to consider claims of selective prosecution despite the restrictions imposed since Armstrong because "even if the court ultimately denied relief, the exposure of disparate treatment through legal process may effect some reduction in improper selective prosecution as the government and the public respond to reduce or eliminate improper disparity." n246 The use of racial impact studies in selective prosecution hearings would certainly help to achieve this goal.

C. Enforcement Through Legislation

National and state legislation should be enacted to require the use of racial impact reports in prosecution offices because reliance on voluntary efforts may not produce significant results. Few elected prosecutors are motivated to campaign on themes involving the promotion of racial equality. n247 Like most politicians, prosecutors view "tough on crime" themes as the most effective tools to assure re-election. n248 Legislation requiring the production and publication of racial impact studies would give prosecutors the necessary political cover. If the studies were required by law, no prosecutor could be accused of being " [*62] soft on crime" for focusing her attention on these issues. Furthermore, since the legislation would not require that prosecutors take any particular action, but would require only the collection and publication of information, it would not be subject to the criticisms raised by the Supreme Court in Wayte v. United States. n249

D. The Impact of the Publication of Racial Impact Studies

The publication of racial impact studies should have a significant impact on prosecutors, police, and the general public. Although it is difficult to determine what that impact would be, the goal would be to deter policies and practices that have a discriminatory impact on African Americans and other people of color, and encourage the development of policies and practices that would further the equitable treatment of defendants and victims of crime, regardless of race. This section will explore possible responses to the proposal.

1. The Response of Prosecutors

Prosecutors might object to racial impact studies and/or the publication of the studies as a significant interference with their law enforcement duties. Criticisms would undoubtedly include the Supreme Court's reasons for deferring to prosecutorial discretion in Wayte v. United States: "Examining the basis of a prosecution delays the criminal proceeding, threatens to chill law enforcement by subjecting the prosecutor's motives and decisionmaking to outside inquiry, and may undermine prosecutorial effectiveness by revealing the Government's enforcement policy." n250

Prosecutors understandably would be concerned about the time and resources necessary to implement racial impact studies. The prosecutor's primary function is law enforcement; n251 any undertaking which substantially interferes with that responsibility would be subject to legitimate criticism. If the collection of data were a tedious process that substantially interfered with the performance of important prosecutorial duties, most prosecutors would object to the studies. Prosecutors, however, could collect the relevant information in an efficient, non-intrusive manner. Prosecutorial offices could create forms with checklists on which the prosecutors could quickly and easily note [*63] the relevant information. n252 Most prosecutors routinely make written entries in case files whenever an action is taken in a particular case. These forms or checklists could be kept in the same case file and would involve no more time than the routine case file entries. The only difference would be the type of information and the format for its collection.

Time is not the only relevant factor. Few prosecutor offices would have the expertise or resources to perform the necessary statistical analysis of the collected data. Social scientists or other researchers with expertise in this field would have to be hired, and few prosecutor offices have the financial resources for such an investment. Although prosecutor offices tend to have more financial resources than defender offices, n253 many prosecutor offices lack the resources to adequately perform their basic prosecutorial responsibilities efficiently and effectively. n254

One possible solution to the resource problem may be the volunteer efforts of local colleges and universities. Criminology and criminal justice departments may be willing to conduct such research and would provide a wealth of resources through the use of graduate students from various departments. The studies would provide a great public service as well as a rich academic experience for professors, scholars, and students. Use of university resources would also give the project the necessary objectivity that would be lacking if the project were conducted by the prosecutors themselves.

Prosecutors may claim that the publication of the studies may chill law enforcement by subjecting the prosecutor's motives to outside inquiry. This argument suggests that prosecutors may be hesitant to prosecute certain cases if they believe that members of the public, criminal defendants, or victims will question their decisions. Thus, some criminal activity will not be prosecuted.

[*64] The goal of the publication of the studies is not to chill appropriate and fair law enforcement, but to totally eliminate unfair, discriminatory law enforcement. To the extent that law enforcement tactics or prosecutorial policies discriminate based on race, they should not merely be chilled - they should be entirely eliminated and replaced with tactics that enforce the law fairly and impartially. The studies, and the knowledge that they will be published, should cause prosecutors to be more careful and meticulous in making decisions. They should motivate prosecutors to assure that similarly situated victims and defendants are treated equitably.

The Supreme Court's concern that judicial interference with prosecutorial discretion would undermine prosecutorial effectiveness by revealing the government's enforcement policy cannot apply to the publication of racial impact studies. That argument suggests that if the public is aware of how and under what circumstances cases are prosecuted, they will adjust their behavior to avoid prosecution. For example, if the public is aware that a prosecutor has a policy of only prosecuting cases involving more than five grams of cocaine, dealers and users will only distribute or possess quantities less than five grams. The publication of racial impact studies should not raise this concern because the studies would not reveal specific law enforcement policies. The information would be limited to racial and other demographic data.

2. The Response of Law Enforcement Officers

The studies may reveal that police or other law enforcement officers are disproportionately arresting African American criminal defendants. n255 Such information may prompt an inquiry of the police chief or federal law enforcement agent and subject him to the same accountability as the prosecutor. The police would be compelled to explain racially disparate arrest patterns. If the public were not satisfied with the explanation, the chief may be required to develop policies designed to eliminate the unacceptable practices.

If the studies implicate police in discriminatory behavior, they may cause tension between prosecutors and police. Prosecutors must rely on the cooperation of police officers and other law enforcement officers long after they complete the arrest process. Police officers continue to investigate cases during the grand jury process and help prosecutors prepare for trials and other hearings. The testimony of arresting officers and other law enforcement personnel is almost always needed in grand jury and trial proceedings. Preparation for trial and other proceedings would certainly be more difficult for prosecutors with hostile, uncooperative law enforcement officers.

[*65] By declining to prosecute cases where they suspect police misconduct, prosecutors may deter discriminatory law enforcement. Such action would compel police to develop policies to ferret out discriminatory practices and promote the fair and equitable enforcement of the law. n256 Ideally, prosecutors and police would work together to establish these policies in their respective offices. Such collaborative efforts would have to be voluntary, as police and prosecutors are generally governed by different political entities. n257

3. The Response of the General Public

One of the primary goals of the publication of racial impact studies is to inspire the electorate to hold prosecutors accountable for their actions. The public would require prosecutors to explain statistics which suggest discriminatory practices and would compel them to remedy the disparities if they found the explanations to be unsatisfactory. Prosecutors who failed to remedy unacceptable disparities would be voted out of office.

One possible criticism of this proposal is its possible ineffectiveness in communities in which African Americans are a small minority of the population. One might argue that such minority communities may not be able to effectively influence prosecutorial decisions because of their lack of political power. Such criticisms assume that white voters do not care about issues of racial disparity and discrimination which harm African Americans. If such a view proved to be true in certain communities, the widespread publication of such information might serve to stigmatize those communities and indirectly force a change in policies or practices. n258

[*66] Even if the studies were published in communities with a politically powerful African American community, the proposal assumes that these communities would view the racial disparities as harmful. Professor Randall Kennedy, however, has a different view. He is a proponent of laws and practices which provide more law enforcement resources to African American communities, even those laws or practices may have a racially discriminatory impact on African American criminal defendants. He argues that such law enforcement only harms African Americans who violate the law while benefiting law-abiding African American citizens. n259

There is much evidence suggesting that Professor Kennedy's view is not the prevailing sentiment in most African American communities. n260 African Americans led the fight to change federal cocaine sentencing laws which discriminate against them. n261 Numerous civil rights organizations, including the National Association for the Advancement of Colored People, the National Rainbow Coalition, the National Council of Negro Women, the Southern Christian Leadership Conference, and the National Political Congress of Black Women fought to eliminate the sentencing disparities which discriminate against African Americans. n262 The National Black Police Organization, the Progressive Baptist Convention, and the National Black Caucus of State Legislators - African American organizations which represent vastly different constituencies - also lobbied to eliminate the discriminatory aspects of the law. n263

Despite the efforts of these groups, the recommendation of the United States Sentencing Commission, n264 and the opposition of Con [*67] gressional Black Caucus, n265 Congress passed legislation maintaining the disparity in sentencing between crack and powder cocaine offenses. n266 President Clinton refused to veto Congress's action. n267

The response of widely divergent segments of the African American community to the discriminatory federal cocaine laws suggests that African Americans will be troubled by studies revealing that law enforcement and prosecutorial practices have a racially discriminatory impact. That experience also suggests that the African American community will take action to assure that discriminatory policies and practices are eliminated. Although the efforts in opposition to the cocaine laws were not totally successful, they did prompt some action. In 1997, Attorney General Janet Reno recommended that the federal cocaine laws be amended to reduce the disparity from 100:1 to 10:1. n268 Although this compromise did not achieve the stated goal, it demonstrates how political action can hold elected officials accountable.

Conclusion

The biblical quote cited at the beginning of this article has a simple meaning: From whom much is given, much is expected. Prosecutors have been given more power and discretion than any other criminal justice official, so they have a greater ability to affect change where it is needed. The Supreme Court has not required much of prosecutors, but the Court's standards should serve as a floor rather than a ceiling, as a base rather than a goal. Prosecutors can, and should, seek to eliminate racial disparities regardless of blame and intent. They can, and should, set a higher standard of performance for themselves than the requirements set by the Supreme Court. They have both the power and privilege to do so. The publication of racial impact studies will help to assure that they do.

Legal Topics:

For related research and practice materials, see the following legal topics:

Criminal Law & ProcedureCriminal OffensesWeaponsGeneral OverviewCriminal Law & ProcedureGuilty PleasGeneral OverviewCriminal Law & ProcedureDefensesSelf-Defense

FOOTNOTES:

n1. Luke 12:48 (King James).

n2. The names of the defendant and decedent are fictitious.

n3. The case was peculiar, not because it was an aberration, but because it was the most egregious example of a phenomenon that I observed throughout my career as a public defender.

n4. Very few whites were charged with crimes in the Superior Court of the District of Columbia. I was a public defender for twelve years and carried a full caseload for six years. I represented only two white defendants during those years. The experiences of my colleagues were very similar.

n5. Telephone Interview with Michele Roberts, partner, Rochon, Roberts, and Stern (Dec. 18, 1997).

n6. The court determined that Mr. McKnight was indigent and eligible for a court- appointed attorney.

n7. In the District of Columbia, as in federal court, felony cases are presented to a grand jury by the prosecutor. The grand jury hears evidence and decides whether there is probable cause to believe the defendant committed the crime. If it finds probable cause, the grand jury issues a formal charging document called an indictment, the case is assigned to a judge, and a trial date is set. If the grand jury does not issue an indictment, the case is dismissed. The grand jury hearings are secret proceedings conducted by the prosecutor. Neither defense attorneys nor any member of the public are allowed to be present. See infra notes 34-48 and accompanying text (discussing the grand jury and the prosecution function).

n8. Ms. Roberts refused to reveal the name of the prosecutor. She indicated that the prosecutor was still handling cases in the United States Attorney's Office for the District of Columbia. Ms. Roberts stated that if she revealed the prosecutor's name, this author's attempts to interview him, and the possibility that his name might be published would ruin her relationship with him, and thus her ability to obtain information and negotiate plea bargains for current and future clients. Telephone Interview with Michele Roberts, partner, Rochon, Roberts, and Stern (March 29, 1998).

n9. There was a running joke in the office that any lawyer who could not get a good deal for a white client should turn in his license to practice law.

n10. For a discussion of race discrimination in the criminal justice system, see generally Randall Kennedy, Race, Crime and the Law (1997) (exploring the history of race discrimination in the criminal justice system); Coramae Richey Mann, Unequal Justice: A Question of Color (1993) (studying discrimination in the criminal justice system against African Americans, Asian Americans, Hispanic Americans, and Native Americans); Jerome G. Miller, Search and Destroy (1996) (theorizing that so many persons have been run through jails and prisons that the violent ethos of the correctional facility has increasingly come to shape behavior on the streets and undermine respect for the law); Katheryn K. Russell, The Color of Crime (1998) (noting that racism continues to undermine society's criminal justice system and skews the public's perception of its black citizens and crime); Samuel Walker et al., The Color of Justice: Race, Ethnicity, and Crime in America (1996) (examining the racial and ethnic discrimination and victimization of minorities in the criminal justice system); Robert D. Crutchfield, et al., Analytical and Aggregation Biases in Analyses of Imprisonment: Reconciling Discrepancies in Studies of Racial Disparity, 31 J. Res. Crime & Delinq. 166 (1994) (same); Angela J. Davis, Benign Neglect of Racism in the Criminal Justice System, 94 Mich. L. Rev. 1660 (1996) (criticizing Tonry, infra, for trivializing the role of racial bias in the overrepresentation of African American men in the criminal justice system); Christopher Johns, The Color of Justice, Ariz. Republic, July 4, 1993, at C1 (finding discrimination against minorities in arrest rates, prosecutorial discretion, and sentencing). But see Michael Tonry, Malign Neglect: Race, Crime, and Punishment in America 49 (1995) (arguing that, with the exception of drug offenses, higher representation of African American men in the criminal justice system is the result of disproportionate offending, not racial bias by police and other criminal justice officials); William Wilbanks, The Myth of a Racist Criminal Justice System (1987) (acknowledging racism in the criminal justice system but concluding that the bulk of racial disproportionality in incarceration reflects greater involvement by African Americans in serious crimes like homicide and robbery); Alfred Blumstein, Racial Disproportionality of U.S. Prison Populations Revisited, 64 U. Colo. L. Rev. 743 (1993) (same).

The author's decision to focus on discrimination against African Americans in the criminal justice system is in no way meant to minimize the significance of similar treatment of Latinos, Asian Americans, and Native Americans. The focus on African Americans stems from a recognition of the extent to which the treatment of African Americans, more than other racial and ethnic groups, has dominated the development of criminal law and procedure in American history. See A. Leon Higginbotham, Jr., In the Matter of Color: Race and the American Legal Process, The Colonial Period (1978) (tracing the black experience in colonial America); Kennedy, supra, at xii. But see Margaret E. Montoya, Of "Subtle Prejudices," White Supremacy, and Affirmative Action: A Reply to Paul Butler, 68 U. Colo. L. Rev. 891, 924 (1997) (criticizing Professor Paul Butler for focusing on the African American experience to the exclusion of Latinos). In addition, although there is a dearth of empirical research on the disparate treatment of African Americans in the criminal justice system, there is even less data on the treatment of Latinos, Asian Americans and Native Americans. Federal statistics include American Indians, Alaska Natives, Asians, and Pacific Islanders in one category. See Darrell K. Gilliard & Allen J. Beck, U.S. Dep't of Justice, Prison and Jail Inmates at MidYear 1996, at 6 tbl.7 (1997).

n11. This author's focus on prosecutorial discretion does not suggest that prosecutors are the only, or even the primary, cause of discrimination in the criminal justice system. The causes of, and possible solutions for, race discrimination in the criminal justice system are as varied and complex as the causes of, and possible solutions for, discrimination in society as a whole. See Jennifer L. Hochschild, Facing Up To The American Dream: Race, Class & The Soul Of The Nation (1995) (arguing that the American dream is in danger of being abandoned because of critical racial and class factors); Douglas S. Massey & Nancy A. Denton, American Apartheid: Segregation and the Making of the Underclass (1993) (examining how racial and residential segregation aid in the perpetuation of black poverty, crime, and social disorder); Racial Discrimination in the United States (Thomas F. Pettigrew ed., 1975); Charles W. Bowser, Race Relations in the 1980s: The Case of the United States, 15 Black Stud. 307 (1985).

Nor does this author suggest that discrimination by criminal justice officials is the only cause of racial disparity in the criminal justice system. Larger societal phenomena such as poverty, white supremacy, and decisions about which behaviors are criminalized all contribute to the disproportionate involvement of African Americans and other people of color in the criminal justice system. See generally Marc Mauer and Tracey Huling, Young Black Americans and the Criminal Justice System: Five Years Later (1995) (reporting that the percentage of African American young men under control of the criminal justice system has risen from one-fourth to one-third, and discussing possible reasons); Tonry, supra note 10 (noting that blacks' presence in the criminal justice system is greatly out of proportion to their presence in the general population); Paul Butler, Affirmative Action and the Criminal Law, 68 U. Colo. L. Rev. 841 (1997) (noting that over half of all men and almost half of the women in prisons are black).

n12. This article will focus on race discrimination, but class and socio-economic status are both relevant to the treatment of criminal defendants and victims of crime. From the police, prosecutors, lawyers, judges, and other officials who run the system to the defendants and victims who are drawn into it unwillingly, the behavior and treatment of almost everyone involved in the criminal justice process reflect the complex social mechanisms of race and class. For a discussion of the relevance of both race and socio-economic status to discriminatory behavior by criminal justice officials, see Derrick Bell, And We Are Not Saved: The Elusive Quest for Racial Justice 5 (1987) (noting that "race-related disadvantages" are "now as likely to be a result as much of social class as of color"); Tonry, supra note 10, at 62-63 (discussing critiques of criminal justice statistics as reflecting racial and class bias); Angela J. Davis, It's Class and Race, Wash. Post, Nov. 25, 1995, at A19 (arguing that both race and class discrimination exist in the criminal justice system).

It is often difficult to discern whether discrimination is based on race, class, or both. First, the number of Americans living in poverty are disproportionately African American, and African Americans are disproportionately poor. See Sheldon Danziger & Peter Gottschalk, America Unequal 73-74 (1995) (discussing data showing disproportionate rate of African American poverty). Second, because discrimination is often unintentional or unconscious, and thus unacknowledged, its origin is sometimes difficult to discern. See infra Part II.B.1. It appears that Mr. McKnight's disparate treatment, discussed at the beginning of this section, was clearly based on race, since he was indigent like most of the criminal defendants in the District of Columbia. Other examples of disparate treatment, like the discrimination alleged in United States v. Armstrong, infra Part III.C, also clearly appear to be based on race rather than class. Race and class discrimination, however, are not always so distinct or discernable.

n13. See Charles P. Bubany & Frank F. Skillern, Taming the Dragon: An Administrative Law for Prosecutorial Decision Making, 13 Am. Crim. L. Rev. 473, 477 (1976) (describing the prosecutor as "the single most powerful figure in the administration of criminal justice"); Bennett L. Gershman, The New Prosecutors, 53 U. Pitt. L. Rev. 393, 448 (1992) ("The American prosecutor, owing to a variety of social and political factors, has emerged as the most pervasive and dominant force in criminal justice.").

n14. See infra Part III; see also Robert L. Misner, Recasting Prosecutorial Discretion, 86 J. Crim. L. & Criminology 717, 722 (1996) (arguing that "the exercise of prosecutorial discretion" should be tied to "the availability of prison resources").

n15. See infra Part III.

n16. See infra Part V (discussing United States v. Armstrong, 517 U.S. 456 (1996)).

n17. A number of scholars have discussed the advantages and disadvantages of prosecutorial discretion and have proposed strategies for limiting or guiding its use. See infra Part V.A.

n18. Disparate impact occurs when policies or laws affect groups of people differently. Disparate impact may or may not reflect discriminatory intent. See Tonry, supra note 10, at 81-123 (criticizing defenses of anti-drug policies that point to the lack of observable discriminatory intent but ignore disparate impact of such policies on African Americans).

n19. Louis D. Brandeis, Other People's Money, and How the Bankers Use It 62 (1914).

n20. See Robert Heller, Selective Prosecution and the Federalization of Criminal Law: The Need for Meaningful Judicial Review of Prosecutorial Discretion, 145 U. Pa. L. Rev. 1309, 1326 (1997); see also infra Part III (discussing these issues).

n21. See United States v. Armstrong, 517 U.S. 456, 462 (1996) (citation omitted).

n22. For criticisms of the arguments in support of judicial deference to prosecutorial discretion, see Kenneth Culp Davis, Discretionary Justice: A Preliminary Inquiry 210 (1969); Heller, supra note 20, at 1328-41.

n23. See Davis, supra note 22, at 9-12 (citing examples of discretionary decisions made by prosecutors and other government administrators); Abraham S. Goldstein, The Passive Judiciary: Prosecutorial Discretion and the Guilty Plea 12-24 (1981) (discussing the prosecutor's discretion not to press criminal charges against a defendant).

n24. See James Vorenberg, Decent Restraint of Prosecutorial Power, 94 Harv. L. Rev. 1521, 1543 (1981) (discussing federal and state policies on prosecutorial discretion).

n25. See id. at 1544-45 (discussing limited effectiveness of self-regulation by prosecutors in ensuring fairness to defendants). For a proposal to improve prosecutors' accountability to the public, see Anne Bowen Poulin, Prosecutorial Discretion and Selective Prosecution: Enforcing Protection After United States v. Armstrong, 34 Am. Crim. L. Rev. 1071, 1119-22 (1997) (advocating mandatory record-keeping by prosecutors on cases prosecuted and not prosecuted and police stops and arrests, to aid in evaluating selective prosecution claims and to encourage better self-regulation by prosecutors).

n26. See infra Part III.B.

n27. See Kenneth J. Melilli, Prosecutorial Discretion in an Adversary System, 1992 B.Y.U. L. Rev. 669, 671.

n28. See Bubany & Skillern, supra note 13, at 476-77 (noting the prosecutor's uncontrolled decision making power).

n29. See id. at 476 (noting that the prosecutor "has the freedom to decide which cases he will prosecute"); Vorenberg, supra note 24, at 1524-32 (criticizing broad prosecutorial discretion in deciding whether to prosecute); James Vorenberg, Narrowing the Discretion of Criminal Justice Officials, 1976 Duke L.J. 651, 678 ("The prosecutor's decision whether and what to charge is the broadest discretionary power in criminal administration.").

n30. See Bubany & Skillern, supra note 13, at 478 (describing prosecutor's power to decline to charge a suspect with a crime).

n31. See 2 Wayne R. LaFave & Jerold H. Israel, Criminal Procedure 13.1(a), (e), 13.2(a) (1984); Charles H. Whitebread & Christopher Slobogin, Criminal Procedure 21.01 (3d ed. 1993) (discussing dominant role of prosecutor in charging decision); see also Tracey L. Meares, Rewards For Good Behavior: Influencing Prosecutorial Discretion and Conduct With Financial Incentives, 64 Fordham L. Rev. 851 (1995) (proposing financial rewards to prosecutors who limit their charging and plea bargaining discretion by charging only those defendants with offenses they believe they can prove at trial).

n32. See infra Part III (discussing selective prosecution challenges).

n33. See Whitebread & Slobogin, supra note 31, 23.06 (discussing weak checks on prosecutor's discretion once there is sufficient evidence to prosecute).

n34. The grand jury was originally conceived in the American colonies as a popular check on the British government's power to prosecute. See Whitebread & Slobogin, supra note 31, 23.01. The Fifth Amendment's requirement of grand jury indictment for a "capital, or otherwise infamous crime" was held by the Supreme Court not to be essential to due process, and hence not applicable to the states through the Fourteenth Amendment. Hurtado v. California, 110 U.S. 516, 538 (1884).

Sixteen states require prosecution by grand jury indictment for all felonies. See Whitebread & Slobogin, supra note 31, 23.01 n.6. Up to thirty states allow felony prosecutions to be brought by "information." A typical information procedure is to require a magistrate's finding of probable cause following a preliminary examination. The remaining states require prosecution by indictment only for capital offenses. See id.

n35. See Andrew D. Leipold, Why Grand Juries Do Not (And Cannot) Protect the Accused, 80 Cornell L. Rev. 260, 266 n.24 (1995); see also Vorenberg, supra note 24, at 1537-38 (discussing how grand jury proceedings are "orchestrated" by the prosecutor).

n36. See Whitebread & Slobogin, supra note 31, 23.02(b).

n37. See id. 23.05; see also Amy Bushyeager & Maria N. Nikiforos, Twenty Sixth Annual Review of Criminal Procedure: Grand Jury, 85 Geo. L.J. 1002, 1002-04, 1012- 13 (1997) (discussing size and power of grand juries in federal system).

n38. See Whitebread & Slobogin, supra note 31, 23.05(d) (discussing power of prosecutor in grand jury proceedings).

n39. See Bushyeager & Nikiforos, supra note 37, at 1020 (discussing grand jury secrecy). In order to protect the secrecy of the investigation, "few states give the target of a grand jury investigation the right to appear before the grand jury." See Whitebread & Slobogin, supra note 31, 23.04(a).

n40. See Mark Kadish, Behind the Locked Door of an American Grand Jury: Its History, Its Secrecy, and Its Process, 24 Fla. St. U. L. Rev. 1, 12-22 (1997) (examining grand jury secrecy).

n41. See infra notes 55-59 and accompanying text (discussing plea bargaining).

n42. Cf. Vorenberg, supra note 24, at 1525-26 (discussing the impact of the charging decision on a suspect's contact with the criminal justice system).

n43. See 2 LaFave & Israel, supra note 31, 20.1 (discussing preponderance of cases resulting in guilty pleas).

n44. See Lisa M. Farabee, Disparate Departures Under the Federal Sentencing Guidelines: A Tale of Two Districts, 30 Conn. L. Rev. 569 (1998) (analyzing federal sentencing guidelines); Daniel J. Freed, Federal Sentencing in the Wake of the Guidelines: Unacceptable Limits on the Discretion of Sentencers, 101 Yale L.J. 1681, 1752-53 (1992) (criticizing the Guidelines' limiting of judicial discretion in sentencing); Cynthia K.Y. Lee, A New "Sliding Scale of Deference" Approach to Abuse of Discretion: Appellate Review of District Court Departures Under the Federal Sentencing Guidelines, 35 Am. Crim. L. Rev. 1, 54 (1998) (arguing that appellate courts retain considerable power to check district court departures from the Guidelines despite Koon v. United States's purported reigning in of appellate courts' power of review).

n45. "Three strikes" laws, a popular form of mandatory sentencing, require long sentences for repeat offenders. For example, California's three strikes law imposes a sentence of 25 years to life in prison for an offender convicted of a felony following two prior convictions for serious crimes. Cal. Penal Code 1170.1 (West 1998). The law also mandates doubling a sentence on the second "strike," requires consecutive sentences for multiple counts, and limits "good time" credits. Id.; see Peter Greenwood et al., Estimated Benefits and Costs of California's New Mandatory-Sentencing Law, in Three Strikes and You're Out: Vengeance as Public Policy 53, 53-54 (David Shichor & Dale K. Sechrest eds. 1996).

n46. See Kate Stith & Jose A. Cabranes, Judging Under the Federal Sentencing Guidelines, 91 Nw. U. L. Rev. 1247, 1254 (1997) (explaining how under the Guidelines judges are "required to follow complex and abstract rules and to make minute arithmetic calculations in order to arrive at a sentence").

n47. See Stephen J. Schulhofer & Ilene H. Nagel, Plea Negotiations Under the Federal Sentencing Guidelines: Guidelines Circumvention and Its Dynamics in the Post- Mistretta Period, 91 Nw. U. L. Rev. 1284, 1284 n.3 (1997) (discussing the claims of many judges that the Guidelines have transferred sentencing discretion from the court to the prosecutor).

n48. See, e.g., id. at 1287 (discussing Minnesota's sentencing guidelines).

n49. See Wayne R. LaFave & Austin W. Scott, Jr., Criminal Law 2.14(f), at 178-79 (2d ed. 1986) (discussing a succession of Supreme Court cases striking down state mandatory death penalty statutes as unconstitutional).

n50. See Kimberly Woolley, Constitutional Interpretations of the Antiterrorism Act's Habeas Corpus Provisions, 66 Geo. Wash. L. Rev. 414, 433 (1998) (citing statistics).

n51. See Thomas Johnson, When Prosecutors Seek the Death Penalty, 22 Am. J. Crim. L. 280, 280 (1994).

n52. See, e.g., LaFave & Scott, supra note 49, at 178 (discussing statutory scheme upheld in Gregg v. Georgia, 428 U.S. 153 (1976)).

n53. See infra notes 190-201 and accompanying text for a discussion of McClesky v. Kemp, 481 U.S. 279 (1987) (holding that a defendant must prove a Fourteenth Amendment violation in his capital sentence by showing purposeful discrimination in his case, not racial disparities in sentencing generally). A proposed Racial Justice Act would permit defendants to challenge death sentences based on racial disparities, but the legislation has never been passed by Congress. See Paul Schoeman, Note, Easing the Fear of Too Much Justice: A Compromise Proposal to Revise the Racial Justice Act, 30 Harv. C.R.-C.L. L. Rev. 543, 551-52 (1995).

n54. For a discussion of legal challenges to discriminatory prosecutorial decisions, see infra Part III.

n55. See Rebecca Hollander-Blumoff, Getting to "Guilty": Plea Bargaining as Negotiation, 2 Harv. Negotiation L. Rev. 115, 117 n.7 (1997) (discussing studies showing that as many as ninety percent of criminal cases end in guilty pleas).

n56. See 2 LaFave & Israel, supra note 31, 20.1 (discussing plea negotiation system).

n57. In some jurisdictions, the judge participates in the plea negotiation process and will inform the defendant what the sentence will be prior to the acceptance of the plea. See 2 LaFave & Israel, supra note 31, 20.3(d) (discussing judicial involvement in plea negotiations in some jurisdictions).

n58. See id. 20.3(c) (discussing prosecutor's discretion to refuse to engage in plea bargaining).

n59. Of course, a defendant may choose to plead guilty to all charges without an agreement to dismiss the most serious charge or charges, but typically there are no advantages to choosing this option unless the defendant has reason to believe that the judge will impose a lenient sentence in consideration of the guilty plea.

n60. But see Peter J. Henning, Testing the Limits of Investigating and Prosecuting White Collar Crime: How Far Will the Courts Allow Prosecutors to Go?, 54 U. Pitt. L. Rev. 405 (1993) (describing cases in the federal system in which prosecutors participate in the investigation of the case before an arrest is made); Lisa F. Salvatore, United States v. Hammad: Encouraging Ethical Conduct of Prosecutors During Pre- Indictment Investigations, 56 Brook. L. Rev. 577 (1990) (describing circumstances in which prosecutors may participate in pre-indictment investigations).

n61. See Poulin, supra note 25, at 1084 ("If the police improperly exercise their discretion, the initial discrimination skews the decision to prosecute even if the prosecutor later culls out some of the referred cases.").

n62. For a discussion of the effect of racial impact studies on police, see infra Part V.D.

n63. See Noryal Morris, Race and Crime: What Evidence is There that Race Influences Results in the Criminal Justice System?, 72 Judicature 111, 112 (1988); Joseph F. Sheley, Structural Influences on the Problem of Race, Crime, and Criminal Justice Discrimination, 67 Tul. L. Rev. 2273, 2275-76 (1993); Alan J. Tomkins et al., Subtle Discrimination in Juvenile Justice Decisionmaking: Social Scientific Perspectives and Explanations, 29 Creighton L. Rev. 1619, 1632 (1996).

n64. See generally Brief for Petitioners at 23-27, Whren v. United States, 517 U.S. 806 (1996) (No. 95-5841) (discussing both statistical and anecdotal evidence of disproportionate police stops of African American men); Angela J. Davis, Race, Cops, and Traffic Stops, 51 U. Miami L. Rev. 425, 427 (1997) [hereinafter Davis, Race, Cops, and Traffic Stops] (arguing that Whren leaves victims of pretextual traffic stops without a viable legal remedy); David A. Harris, "Driving While Black" and All Other Traffic Offenses: The Supreme Court and Pretextual Traffic Stops, 87 J. Crim. L. & Criminology 544, 546 (1997) [hereinafter Harris, Driving While Black] (arguing that Whren put the Court's imprimatur on longstanding police practice of stopping African American drivers "just to see what officers can find."); David A. Harris, Factors for Reasonable Suspicion: When Black and Poor Means Stopped and Frisked, 69 Ind. L.J. 659 (1994) [hereinafter Harris, Factors for Reasonable Suspicion] (arguing that a substantial body of law now allows police officers to stop an individual based on presence in a high crime area and evasive behavior - two factors which affect African Americans and Latinos more frequently than whites); Sheri Lynn Johnson, Race and the Decision to Detain a Suspect, 93 Yale L.J. 214, 214 (1983) (discussing the impact of race on police stops and detentions).

For examples of discriminatory stops, see Michael A. Fletcher, Driven to Extremes, Wash. Post, Mar. 29, 1996, at A1. For statistical analyses of police decisions to stop motorists, see Jeff Brazil & Steve Berry, Color of Driver is Key to Stops in I-95 Videos, Orlando Sentinel Trib., Aug. 23, 1992, at A1; Paul Butler, Encounters with the Police on my Street, Legal Times, Nov. 10, 1997, at 23; Henry Pierson Curtis, Statistics Show Pattern of Discrimination, Orlando Sentinel Trib., Aug. 23, 1992, at A11; John Lamberth, Driving While Black, Wash. Post, Aug. 16, 1998, at C1; Paul W. Valentine, Md. State Police Still Target Black Motorists, ACLU Says, Wash. Post, Nov. 15, 1996, at A1.

n65. A police officer may stop and frisk suspects without probable cause if he has a reasonable and articulable suspicion that "criminal activity may be afoot and that the persons with whom he is dealing may be armed and presently dangerous." Terry v. Ohio, 392 U.S. 1, 30 (1968). The Court has defined reasonable suspicion as "specific and articulable facts which, taken together with rational inferences from those facts, reasonably warrant that intrusion." Id. at 21. Police officers may arrest suspects for a felony offense in a public place without a warrant if they have probable cause to believe a crime has been committed. United States v. Watson, 423 U.S. 411, 421-23 (1976). If the crime is a misdemeanor, the officer may make an arrest only if the crime was committed in her presence. Id. at 422.

n66. See United States v. Weaver, 966 F.2d 391, 392 (8th Cir. 1992) (explaining that factors that created reasonable suspicion included fact that defendant was a "'roughly dressed' young black male"); see also United States v. Brignoni-Ponce, 422 U.S. 873, 886-87 (1975) (stating that a person's "Mexican appearance" in an area near the U.S.- Mexico border is a relevant factor in creating reasonable suspicion that the person is an illegal alien, but "standing alone it does not justify stopping all Mexican-Americans to ask if they are aliens"); United States v. Harvey, 16 F.3d 109, 113 (6th Cir. 1994) (Keith, J., dissenting) (noting that the majority upheld a pretextual traffic stop despite the police officer's admission that he was motivated to make stop based on fact that "there were three young black male occupants in an old vehicle").

n67. See supra note 65.

n68. See Harris, Factors for Reasonable Suspicion, supra note 64, at 660-69 (1994) (suggesting that the Court has relaxed the Terry standard in deference to law enforcement concerns).

n69. See United States v. Robinson, 414 U.S. 218, 242 (1973) (Marshall, J., dissenting) ("The majority's fear of overruling the 'quick ad hoc judgment' of the police officer is thus inconsistent with the very function of the [Fourth] Amendment - to ensure that the quick ad hoc judgments of police officers are subject to review and control by the judiciary.").

n70. Practically speaking, a police officer's decision not to stop, detain, or arrest is unreviewable. See Stephen A. Saltzburg & Daniel J. Capra, American Criminal Procedure 681 (5th ed. 1996).

n71. See Courtland Milloy, Unequal Justice in P.G.?, Wash. Post, Feb. 25, 1996, at B1 (describing an event in Prince George's County, Maryland, where white officers observed three white adults smoking crack cocaine in a car with a baby and neither made arrests nor filed charges).

n72. The criteria that generally guide "plain view" searches are set forth in Coolidge v. New Hampshire, 403 U.S. 443 (1971).

It is well established that under certain circumstances the police may seize evidence in plain view without a warrant ...

...

What the 'plain view' cases have in common is that the police officer in each of them had a prior justification for an intrusion in the course of which he came ... across a piece of evidence incriminating the accused. The doctrine serves to supplement the prior justification [for conducting a search or seizure] - whether it be a warrant for another object, hot pursuit, search incident to lawful arrest, or some other legitimate reason for being present unconnected with a search directed against the accused - and permits the warrantless seizure. Of course, the extension of the original justification is legitimate only where it is immediately apparent to the police that they have evidence before them; the 'plain view' doctrine may not be used to extend a general exploratory search from one object to another until something incriminating at last emerges.

Id. at 465-66. The Court removed the requirement that evidence seized in plain view be discovered inadvertently in Horton v. California, 496 U.S. 128, 139-42 (1990).

n73. A traffic stop gives a police officer the opportunity to request permission to search a person's car. The police officer may legally request such permission to search, whether or not he decides to issue a traffic ticket. Although the officer must have reasonable suspicion or probable cause to stop or arrest a motorist, once the reason for the stop or arrest is completed, the officer has the right to continue to speak with him. Even if a police officer does not have probable cause or reasonable suspicion to believe that a crime has been or is about to be committed, he nonetheless has the right to approach an individual and speak with him. See Florida v. Bostick, 501 U.S. 429 (1991). The Supreme Court has held that if the individual does not want to talk to the officer, he has the right to decline and walk away, and the exercise of those rights may not be used as the sole basis for probable cause or even reasonable suspicion. See Florida v. Royer, 460 U.S. 491, 497-98 (1983). But police need not necessarily inform a person of his right to refuse to consent to a search. See Schneckloth v. Bustamonte, 412 U.S. 218, 248-49 (1973). Nor must police inform "a lawfully seized defendant ... that he is free to go before his consent to [a] search will be ... recognized as voluntary." Ohio v. Robinette, 519 U.S. 33, 35 (1996).

n74. See Robinson, 414 U.S. at 248 (Marshall, J., dissenting). In Robinson, the defendant was arrested pursuant to a regulation that required the arrest of a person operating a motor vehicle after revocation of an operator's permit. Justice Marshall wrote in dissent:

Although, in this particular case, Officer Jenks was required by police department regulations to make an in-custody arrest rather than to issue a citation, in most jurisdictions and for most traffic offenses the determination of whether to issue a citation or effect a full arrest is discretionary with the officer. There is always the possibility that a police officer, lacking probable cause to obtain a search warrant, will use a traffic arrest as a pretext to conduct a search.

Id. (Marshall, J., dissenting); see also Gustafson v. Florida, 414 U.S. 260, 266 (1973) (holding that a police officer may search a subject's person incident to a lawful arrest). In Gustafson, the police officer was not required by regulation to make a custodial arrest. Id. at 263. Justice Stewart wrote in concurrence: "It seems to me that a persuasive claim might have been made in this case that the custodial arrest of the petitioner for a minor traffic offense violated his rights under the Fourth and Fourteenth Amendments. But no such claim has been made." Id. at 266-67 (Stewart, J., concurring).

n75. If the police officer decides to place the motorist under arrest, he then has the legal authority to conduct a full-blown search incident to the arrest of the motorist and the area around him. See New York v. Belton, 453 U.S. 454 (1981); Chimel v. California, 395 U.S. 752, 762-63 (1969). Any evidence or contraband found on the motorist may be used as the basis for probable cause to arrest him for other crimes. See Whren v. United States, 517 U.S. 806, 812-14 (1996); Robinson, 414 U.S. at 266.

n76. The Supreme Court has held that a police officer may stop a vehicle if he has probable cause to believe a traffic violation has occurred, even if the officer uses the traffic stop as a pretext to investigate suspicion of criminal activity not supported by probable cause. See Whren, 517 U.S. at 812-14. In criticizing the standard for pretextual stops adopted in Whren, one court has noted: "Given the 'multitude of applicable traffic and equipment regulations' in any jurisdiction, upholding a stop on the basis of a regulation seldom enforced opens the door to ... arbitrary exercise of police discretion ...." United States v. Botero-Ospina, 71 F.3d 783, 790 (10th Cir. 1995) (Seymour, C.J., dissenting) (citation omitted). LaFave notes that the Whren standard has "conferred upon the police virtual carte blanche to stop people because of the color of their skin or for any other arbitrary reason." 1 Wayne R. LaFave, Search & Seizure 1.4(e), at 123 (3d ed. 1996) (footnote omitted). He also states that:

Given the pervasiveness of ... minor offenses and the ease with which law enforcement agents may uncover them in the conduct of virtually everyone ... there exists "a power that places the liberty of every man in the hands of every petty officer," precisely the kind of arbitrary authority which gave rise to the Fourth Amendment.

Id. 1.4(e), at 121-22. For frank admissions by police of how they use traffic violations as a pretext to investigate hunches not supported by reasonable suspicion, see Brief for Petitioners at 21, Whren (No. 95-5841).

For examples of pretextual stops upheld by courts under the standard endorsed by Whren, see United States v. Harvey, 16 F.3d 109, 113 (6th Cir. 1994) (upholding a stop in which the arresting officer testified that he stopped the car in part because "there were three young black male occupants in an old vehicle") (Keith, J., dissenting); and United States v. Roberson, 6 F.3d 1088, 1092 (5th Cir. 1993) (upholding a stop in which a state trooper known "for patrolling the Fourth Amendment's outer frontier" stopped a vehicle with black occupants for changing lanes without signaling on an open stretch of highway). Whren invalidated cases holding that a traffic stop passed Fourth Amendment scrutiny only if a reasonable officer would have made the stop. 517 U.S. at 339. Such cases include United States v. Laymon, 730 F. Supp. 332 (D. Colo. 1990), where an undercover officer followed a car because it had out-of-state plates, waiting for the vehicle to commit a traffic infraction. Whren, 517 U.S. at 339. When the car's right tire finally touched the white shoulder stripe, the officer pulled the car over for "weaving." Id. at 333. Using the test of whether a reasonable officer would have made the stop, Laymon held the stop unreasonable. Id. Under Whren, such a stop is perfectly legal. Id. at 339.

n77. For example, 70-80 percent of motorists exceed the posted speed limit on interstate highways, yet police observe a "traditional five mph enforcement 'tolerance' in most states." See Brief for Petitioners at 19, Whren (No. 95-5841) (citing U.S. Dept. of Transportation statistics).

n78. For a discussion of the discriminatory nature of pretextual traffic stops, see Harris, Driving While Black, supra note 64, which argues that Whren approves racially discriminatory traffic stops. See also Davis, Race, Cops and Traffic Stops, supra note 64 (arguing that Whren left people of color without an adequate remedy for discriminatory police stops).

n79. See supra note 76.

n80. There are numerous other examples of the disproportionate use of pretextual traffic stops against African Americans. Several class-action lawsuits filed by black motorists against city and county officials have resulted in out-of-court settlements requiring monitoring of police traffic stops. See Russell, supra note 10, at 40-43 (discussing lawsuits). Discovery materials in many cases reveal that police often are shameless in their use of pretextual stops to target black motorists. See Brief for Petitioners at 25-29, Whren (No. 95-5841) (discussing discovery materials); see also Harris, Driving While Black, supra note 64 (discussing cases which document the use of pretextual traffic stops against African Americans and Latinos).

n81. Although African Americans comprise only 13% of drug users, they make up 35% of those arrested for drug possession, 55% of those convicted of drug possession, and 74% of those imprisoned for drug possession. See Mauer & Huling, supra note 11, at 12 (discussing statistics).

n82. See id.

n83. United States Sentencing Commission, Special Report to Congress: Cocaine and Federal Sentencing Policy 70 (1995) [hereinafter Sentencing Commission].

n84. Law enforcement officials cite reasons other than race for this focus on making arrests in urban areas. See Tonry, supra note 10, at 106 (noting that some law enforcement officials have cited the ease of infiltrating and making arrests in the visible, open air drug markets of urban areas as compared to drug use and trafficking in more affluent areas); see also Sentencing Commission, supra note 83, at 70 ("Although there is a reasonably clear idea of who sells cocaine in the street and in crack and shooting houses, there is less awareness of how cocaine is sold in the suburbs, in upper-class neighborhoods, and to business people."). For discussion of debate about whether or not this focus on law enforcement in communities of color is harmful, see infra Part V.D.2.

n85. See supra note 28 and accompanying text.

n86. See Melilli, supra note 27, at 676 ("Police decisions not to investigate or not to arrest will ordinarily receive no review by the prosecutor's office."); see also supra note 76 and accompanying text (describing lack of judicial review of pretextual traffic stops made by police officers).

n87. 517 U.S. 806 (1996).

n88. See infra notes 171-85 and accompanying text.

n89. See Sean Hecker, Race and Pretextual Traffic Stops: An Expanded Role for Civilian Review Boards, 28 Colum. Hum. Rts. L. Rev. 551, 561-62 (1997) (discussing agreement by Maryland State Police not to use racial profiles in drug enforcement, which was part of a settlement of a suit brought by African American motorist who was illegally detained because of his race).

n90. Two legal scholars noted:

The decisions [the prosecutor] makes influence and often determine the disposition in all cases brought to him by the police. The prosecutor's decisions also significantly affect the arrest practices of the police, the volume of cases in the courts, and the number of offenders referred to the correctional system. Thus, the prosecutor is in the most favorable position to bring about needed coordination among the various law enforcement and correctional agencies in the community.

Charles W. Thomas & W. Anthony Fitch, Prosecutorial Decision Making, 13 Am. Crim. L. Rev. 507, 509 n.14 (1976) (quoting President's Commission on Law Enforcement and Administration of Justice, Task Force Report: The Courts 72 (1967)); see also Paul Butler, The Home Front: Eric Holder's Reefer Madness, Legal Times, Jan. 20, 1997, at 25 (criticizing how U.S. Attorney Eric H. Holder, Jr. "has directed his assistants to vigorously prosecute people who possess small amounts of marijuana"); Philip P. Pan & Robert E. Pierre, For Police, a Dead End: D.C., Pr. George's Have This in Common: Too Many Homicides, Wash. Post, Jan. 12, 1997, at B1 (discussing how former D.C. U.S. Attorney Eric H. Holder, Jr.'s new get tough policy of prosecuting cases involving any amount of marijuana has led police to stricter enforcement of marijuana laws and increased arrests); Robert E. Pierre, Marijuana's Violent Side: As Use of Drug Rises, So Do Slayings in D.C., Wash. Post, Sept. 9, 1996, at A1 (discussing how U.S. Attorney Eric H. Holder, Jr. is "considering not only prosecuting more marijuana cases but also asking the D.C., Council to enact stiffer penalties for the sale and use of marijuana").

n91. See William Bowers et al., Legal Homicide: Death as Punishment in America 1864-1982 (1980); Developments in the Law: Race and the Criminal Process, 101 Harv. L. Rev. 1472, 1525 (1988) [hereinafter Developments in the Law]; Larry Michael Fehr, Racial and Ethnic Disparities in Prosecution and Sentencing: Empirical Research of the Washington State Minority and Justice Commission, 32 Gonz. L. Rev. 577, 581-86 (1997); Gary D. LaFree, The Effect of Sexual Stratification by Race on Official Reactions to Rape, 45 Am. Soc. Rev. 842, 852-53 (1990); Michael L. Radelet & Glenn L. Pierce, Race and Prosecutorial Discretion in Homicide Cases, 19 L. & Soc'y Rev. 587, 615-19 (1985); Cassia Spohn et al., The Impact of Ethnicity and Gender of Defendants on the Decision to Reject or Dismiss Felony Charges, 25 Criminology 175 (1987).

n92. For the distinction between disparate and discriminatory treatment, see infra Part III.B.

n93. For a discussion of the equal protection analysis, see infra Part III.A-B.

n94. For a discussion of the inadequacy of current legal remedies for discriminatory prosecutorial behavior, see infra Part III.

n95. See infra Part III.

n96. See infra Part III.

n97. See Batson v. Kentucky, 476 U.S. 79 (1986) (holding that a prosecutor must establish a race-neutral reason for using a peremptory strike against African American member of jury pool).

n98. A good example is the blatant racism in the Scottsboro Boys case. See Dan T. Carter, Scottsboro: A Tragedy of the American South (rev. ed. 1979); James Goodman, Stories of Scottsboro (1994); see also infra Part III (discussing Yick Wo v. Hopkins, 118 U.S. 356 (1886)).

n99. See, e.g., Jim Newton & Bill Boyarsky, Witnesses Tell Jury of Fuhrman's Racial Epithets, L.A. Times, Sept. 6, 1995, at A1 (describing witnesses' accounts of racial epithets by police officer who was the key witness in O.J. Simpson's murder trial).

n100. See Sheri Lynn Johnson, Unconscious Racism and the Criminal Law, 73 Cornell L. Rev. 1016 (1988) (exploring unconscious racism in criminal law decisions); Naomi Wolf, The Racism of Well-Meaning White People, in Skin Deep: Black Women and White Women Write About Race 37 (Marita Golden & Susan Richards Shreve, eds., 1995); Edward Patrick Boyle, Note, It's Not Easy Bein' Green: The Psychology of Racism, Environmental Discrimination, and the Argument for Modernizing Equal Protection Analysis, 46 Vand. L. Rev. 937, 939 (1993) (describing unconscious or "aversive" racism).

n101. But see infra notes 170-71 and accompanying text (discussing Justice Brennan's dissent in McCleskey v. Kemp, 481 U.S. 279 (1987), and Justice Marshall's concurrence in Batson v. Kentucky, 476 U.S. 79 (1986)).

n102. Charles H. Lawrence III, The Id, the Ego, and Equal Protection: Reckoning with Unconscious Racism, 39 Stan. L. Rev. 317, 322 (1987).

n103. See id. at 322-23.

n104. William T. Pizzi, Understanding Prosecutorial Discretion in the United States: The Limits of Comparative Criminal Procedure as an Instrument of Reform, 54 Ohio St. L.J. 1325, 1344 (1993) (describing informal guidelines and standards within prosecutors offices).

n105. The Department of Justice Manual states that "the U.S. Attorney is authorized to decline prosecution in any case referred directly to him/her by an agency unless a statute provides otherwise. Whenever a case is closed without prosecution, the U.S. Attorney's files should reflect the action taken and the reason for it." Department of Justice Manual 9-2.020 (1993-94 Supp.). The manual also has more detailed guidelines on when to conduct a federal prosecution after a defendant has been acquitted in a state court. See id. 9-2.142.

n106. These factors are frequently noted informally in plea negotiations and less frequently in defense of legal challenges alleging discriminatory prosecutorial behavior. See Developments in the Law, supra note 91, at 1523; see also infra Part III (discussing legal challenges to discriminatory prosecutorial decisions). Prosecutors are not routinely required to justify or give reasons for charging or other discretionary decisions. See Wayte v. United States, 470 U.S. 598, 607 (1985) ("The decision to prosecute is particularly ill-suited to judicial review."); Vorenberg, supra note 24, at 1542 (arguing that task of showing discrimination by prosecutor is made more difficult by lack of requirement to state reasons for charging decision). But see Wash. Rev. Code Ann. 9.94A.430-460 (West 1998) (setting forth statutory filing and disposition standards to guide prosecutorial discretion in charging and plea bargaining decisions). See generally Michelle A. Gail, Twenty-Sixth Annual Review of Criminal Procedure: Prosecutorial Discretion, 85 Geo. L.J. 983 (1997) (describing breadth of prosecutorial discretion). Washington is the only state with such guidelines. See Fehr, supra note 91, at 581. The guidelines are voluntary and do not create enforceable legal rights. Id.

n107. See infra notes 188-94 and accompanying text for discussion of the Baldus study which proved that murder cases in the state of Georgia from 1973-79 involving white victims were three to four times more likely to be prosecuted as death penalty cases than cases involving black victims. See also Kennedy, supra note 10, at 331 nn.51 & 52 (discussing the Baldus study and others showing race disparities in capital sentencing); Developments in the Law, supra note 91, at 1525-27 (discussing studies showing discrimination in prosecutor's charging decision based on race of victim and race of offender); Radelet & Pierce, supra note 91 (discussing relationships between prosecutorial classifications and victims' and defendants' race); Saundra Torry, ABA Endorses Moratorium on Capital Punishment, Wash. Post, Feb. 2, 1997, at A4 (discussing the ABA call for a moratorium on the death penalty until more study is done on role of race and economics in capital sentencing). See generally Stephen L. Carter, When Victims Happen to Be Black, 97 Yale L.J. 420 (1988) (discussing assumptions about race in the public's perception of victims and victimizers).

n108. See Radelet & Pierce, supra note 91, at 616 ("Regardless of the race of the defendants, prosecutors may consider white victims more credible than black victims or their troubles more worthy of full prosecution." (citing Martha A. Myers and John Hagan, Private and Public Trouble: Prosecutors and the Allocation of Court Resources, 26 Soc. Probs. 439, 447 (1979))).

n109. See supra note 103.

n110. Although the economic status of offenders and victims may appear to be the determinative factor in these decisions, because the African American population is disproportionately poor, any class-based discussion inevitably entails a discussion of race. Moreover, if the behavior is unintentional and unconscious, it is difficult to discern whether it is based on class, race, or both. See supra note 10.

n111. See supra notes 64-65 and accompanying text.

n112. See supra note 66 and accompanying text.

n113. See supra note 84 and accompanying text.

n114. See supra note 84 and accompanying text.

n115. See supra Introduction.

n116. Justice Brennan's dissent in McCleskey v. Kemp discusses this valuation of human beings in the context of the decision to seek the death penalty:

Decisions influenced by race rest in part on a categorical assessment of the worth of human beings according to color, insensitive to whatever qualities the individuals in question may possess. Enhanced willingness to impose the death sentence on black defendants, or diminished willingness to render such a sentence when blacks are victims, reflects a devaluation of the lives of black persons.

481 U.S. 279, 336 (1987) (Brennan, J., dissenting).

n117. As one scholar notes:

The prosecutor's freedom from client control gives rise to vast discretion. That, in turn, creates a risk that prejudice or self-interest will govern her decisions. She may arbitrarily favor one defendant, or type of defendant, over others. Alternatively, because her success is measured by her conviction rate, she may be tempted to ignore the rights of defendants, victims, or the community in order to obtain pleas or guilty verdicts.

Fred C. Zacharias, Structuring the Ethics of Prosecutorial Trial Practice: Can Prosecutors Do Justice?, 44 Vand. L. Rev. 45, 58-59 (1991).

n118. See Kenneth Culp Davis, Discretionary Justice: A Preliminary Inquiry 190 (1969) (noting "the prevailing assumption that decisions to prosecute or not to prosecute never call for reasoned opinions"); Paul E. Dow, Discretionary Justice: A Critical Inquiry 125 (1981) ("There are no records kept concerning the prosecutor's motives for instituting litigation.").

n119. See Vorenberg, supra note 24, at 1565 (noting that these records, if they exist at all, are confidential).

n120. Prosecutors may be required to justify certain discretionary decisions in response to a legal challenge by a criminal defendant such as a motion to dismiss an indictment for selective prosecution. See infra Part III.B. But see infra Part III.C (discussing the Supreme Court's decision in United States v. Armstrong, 517 U.S. 456 (1996), requiring criminal defendants claiming selective prosecution to make a very detailed showing of discriminatory impact before receiving discovery of internal prosecutorial records).

n121. See Wayne R. LaFave & Jerold H. Israel, Criminal Procedure 13.3, at 629-30 (2d ed. 1992) (noting the "generally accepted view" that the prosecution function should be performed by a public prosecutor rather than a private party, and limiting a private party's role to seeking writ of mandamus by a court to compel a prosecutor to initiate a prosecution - a request rarely granted by courts).

n122. But see S.J. Res. 6, 105th Cong. (1997), which would give victims of violent crime the right:

(1) to be informed of his alleged attacker's trial; (2) to be present at all stages of the trial; (3) to be heard at sentencing; (4) to object to plea bargains or release from custody; (5) to be given notice of any release or escape; and (6) to receive full restitution from the convicted offender.

Monroe Leigh, Witness Anonymity is Inconsistent with Due Process, 91 Am. J. Int'l L. 80, 83 (1997) (summarizing proposed Victims' Rights Amendment). For articles in support of a Victims' Rights Amendment, see Richard Barajas & Scott Alexander Nelson, The Proposed Crime Victims' Federal Constitutional Amendment: Working Toward a Proper Balance, 49 Baylor L. Rev. 1 (1997); Richard M. Romley, Letter to the Editor, Constitutional Rights for Victims: Another Perspective, Prosecutor, Jan.- Feb. 1997, at 7. For arguments against a Victims' Rights Amendment, see William W. Taylor, III., Victims' Rights & the Constitution: Proceed with Caution, Prosecutor, Mar.-Apr. 1997, at 12; see also Robert Fichenberg, The Controversial Victims' Rights Amendment, Prosecutor, Sept.-Oct. 1996, at 38.

n123. "No state shall ... deny to any person within its jurisdiction the equal protection of the laws." U.S. Const. amend. XIV, 1.

n124. 42 U.S.C. 1983 (1994).

n125. See Developments in the Law, supra note 91, at 1537-49 (discussing the selective prosecution doctrine).

n126. Such claims may be brought under 42 U.S.C. 1981-83, 1985 and under the U.S. Constitution. Criminal defendants would bring these claims as separate civil actions.

n127. See NAACP v. Levi, 418 F. Supp. 1109, 1115-17 (D.D.C. 1976) (finding denial of plaintiff's civil rights in failure of federal officers to adequately investigate shooting of plaintiff's husband by state police officers); Developments in the Law, supra note 91, at 1532-35 (discussing federal civil rights suits).

n128. See Developments in the Law, supra note 91, at 1532-35 (describing the process of plaintiffs bringing "federal civil rights suits against prosecutors for allegedly engaging in racially selective prosecution"). Criminal defendants would raise such claims in a separate civil action.

n129. See id. at 1533 (explaining why "courts have indicated a clear lack of receptivity to these suits").

n130. See Warth v. Seldin, 422 U.S. 490, 501 (1975).

n131. See Village of Arlington Heights v. Metropolitan Hous. Dev. Corp., 429 U.S. 252, 261 (1977).

n132. See Simon v. Eastern Ky. Welfare Rts. Org., 426 U.S. 26, 38 (1976).

n133. For example, in O'Shea v. Littleton, African American citizens brought a class action suit against a prosecutor, a magistrate and a judge, alleging discriminatory treatment in bond-setting, sentencing, and jury fee practices. 414 U.S. 488 (1974). The Court dismissed the claim, holding that the plaintiffs did not have standing because they were not current victims of specific discrimination. See id. at 495-96.

n134. See Washington v. Davis, 426 U.S. 229, 238-48 (1976). Although the discriminatory intent requirement was developed in criminal cases alleging selective prosecution, the Court has applied this requirement in the civil context as well. See Developments in the Law, supra note 91, at 1542 (describing "an impossibly high burden of proof on defendants attempting to raise claims of selective prosecution in the racial context").

n135. See Developments in the Law, supra note 91, at 1535-36 (discussing selective prosecution claim).

n136. 368 U.S. 448 (1962).

n137. See id. at 456.

n138. See id.

n139. For detailed discussion of circuit cases which allowed evidence of disproportionate impact as proof of intentional discrimination, see Developments in the Law, supra note 91, at 1536-39.

n140. 501 F.2d 1207 (2d Cir. 1974).

n141. See id. at 1209.

n142. See id. at 1211.

n143. Several circuit courts applied the Berrios two-pronged analysis, holding that evidence of disproportionate impact satisfied the discriminatory effect prong but not the discriminatory purpose prong. See, e.g., United States v. Ojala, 544 F.2d 940, 943- 45 (8th Cir. 1976) (rejecting a claim that the appellant was impermissibly selected for prosecution after he announced his purposeful refusal to comply with IRS filing requirements); United States v. Hazel, 696 F.2d 473, 475 (6th Cir. 1983) (holding that defendants singled out for "great notoriety of their protests" failed to prove impermissible motive for prosecutor).

n144. 426 U.S. 229 (1976).

n145. See id. at 237-39.

n146. See, e.g., Delaware v. Holloway, 460 A.2d 976, 978-80 (Del. Super. Ct. 1983) (holding that the defendant must prove intentional and purposeful discrimination in tax violation charges); State v. Russell, 343 N.W.2d 36, 37 (Minn. 1984) (holding that the defendants charged with theft who claimed that they were victims of racially discriminatory enforcement failed to meet their burden of establishing a prima facie case of racially discriminatory impact and discriminatory intent); State v. Bird Head, 285 N.W.2d 698, 702 (Neb. 1979) (noting that Native American defendants must allege and prove deliberate selective process of enforcement based on race); State v. Sheedy 480 A.2d 887, 888 (N.H. 1984) (rejecting a claim of selective prosecution where a defendant convicted of "wilfully intercepting" telephone conversations had not established intentional or purposeful discrimination).

n147. 470 U.S. 598 (1985).

n148. See id. at 601, 604.

n149. See id. at 608-10.

n150. See Developments in the Law, supra note 91, at 1541-42 (discussing the impact of Wayte).

n151. 517 U.S. 456 (1996).

n152. See id. at 467-68.

n153. See id. at 464-65.

n154. See id. at 465.

n155. Cal. Health & Safety Code 11351-11351.5 (West 1991) (establishing a two- to-five year sentence for distribution of cocaine); see United States v. Armstrong, 48 F.3d 1508, 1510-12 (9th Cir. 1995) (en banc), rev'd, 517 U.S. 456 (1996).

n156. Armstrong, 517 U.S. at 462.

n157. See Armstrong, 48 F.3d at 1513-14.

n158. See Armstrong, 517 U.S. at 466-67.

n159. 368 U.S. 448, 456 (1962).

n160. 470 U.S. 598, 608 (1985).

n161. Armstrong, 517 U.S. at 466-67.

n162. Id.

n163. See id. (citing Yick Wo v. Hopkins, 118 U.S. 356 (1886)).

n164. See Armstrong, 517 U.S. at 467.

n165. The thinly veiled racial animus of the respondents in Yick Wo is less prevalent in case law a century later. Racial prejudice today is more likely to be unconscious. Charles Lawrence approaches unconscious racism from a Freudian perspective. He notes that "the human mind defends itself against the discomfort of guilt by denying or refusing to recognize those ideas, wishes, and beliefs that conflict with what the individual has learned is good [and] right." Lawrence, supra note 102, at 322. Lawrence then concludes: "While our historical experience has made racism an integral part of our culture, our society has more recently embraced an ideal that rejects racism as immoral. When an individual experiences conflict between racist ideas and the societal ethic that condemns those ideas, the mind excludes his racism from consciousness." Id. at 322-23 (footnote omitted).

Adding lessons from cognitive psychology's study of unstated cultural beliefs, Lawrence writes:

The individual is unaware ... that the ubiquitous presence of a cultural stereotype has influenced her perception that blacks are lazy or unintelligent. Because racism is so deeply ingrained in our culture, it is likely to be transmitted by tacit understandings: Even if a child is not told that blacks are inferior, he learns that lesson by observing the behavior of others. These tacit understandings, because they have never been articulated, are less likely to be experienced at a conscious level.

Id. at 323. Applying his theory of unconscious racism to Equal Protection jurisprudence, Lawrence continues:

Requiring proof of conscious or intentional motivation as a prerequisite to constitutional recognition that a decision is race-dependent ignores much of what we understand about how the human mind works. It also disregards both the irrationality of racism and the profound effect that the history of American race relations has had on the individual and collective unconscious.

Id. (footnotes omitted).

n166. See supra Part II.B.1 (discussing unconscious racism).

n167. Justice Brennan argued that:

The discretion afforded prosecutors and jurors in the Georgia capital sentencing system creates such opportunities. No guidelines govern prosecutorial decisions to seek the death penalty, and Georgia provides juries with no list of aggravating and mitigating factors, nor any standard for balancing them against one another. Once a jury identifies one aggravating factor, it has complete discretion in choosing life or death, and need not articulate its basis for selecting life imprisonment. The Georgia sentencing system therefore provides considerable opportunity for racial considerations, however subtle and unconscious, to influence charging and sentencing decisions.

McCleskey v. Kemp, 481 U.S. 279, 333-34 (1987) (Brennan, J., dissenting) (footnote omitted).

n168. Justice Marshall argued that:

A prosecutor's own conscious or unconscious racism may lead him easily to the conclusion that a prospective black juror is 'sullen,' or 'distant,' a characterization that would not have come to his mind if a white juror had acted identically. A judge's own conscious or unconscious racism may lead him to accept such an explanation as well supported.

476 U.S. 79, 106 (1986) (Marshall, J., concurring).

n169. See supra Part II.B.1.

n170. Review of Supreme Court cases citing Yick Wo.

n171. United States v. Armstrong, 517 U.S. 456, 458 (1996).

n172. Id. at 465.

n173. Id.

n174. Id. at 469.

n175. Id. at 470.

n176. The Court deemed the affidavit to be "hearsay" and "personal conclusions based on anecdotal evidence." Id.

n177. Id. at 469 (quoting United States v. Armstrong, 48 F.3d 1508, 1516-17 (9th Cir. 1995)).

n178. Id. at 469 (citations omitted).

n179. Id. at 482 (Stevens, J., dissenting).

n180. See id. at 464.

n181. Id. (quoting United States v. Chemical Found. Inc., 272 U.S. 1, 14-15 (1926)).

n182. See id. at 465 (quoting Wayte v. United States, 470 U.S. 598, 607 (1985)).

n183. Id. (quoting Wayte v. United States, 470 U.S. 598, 607 (1985)).

n184. See Francis D. Doucette, Non-Appointment of Counsel in Indigent Criminal Cases: A Case Study, 31 New Eng. L. Rev. 495, 511 (1997) ("It also overlooks the unavoidable fact that the majority of criminal defendants, misdemeanor and felony alike, are indigent and likely to remain so."); see also Norman Lefstein, Criminal Defense Services for the Poor 3 n.4 (1982) (describing how of seven million felony and non-traffic misdemeanor arrests in 1971, approximately 3.4 million required appointed counsel).

n185. See Lefstein, supra note 184, at 3; see also Richard Berk & Alec Campbell, Preliminary Data on Race and Crack Charging Practices in Los Angeles, 6 Fed. Sentencing Rep. 37 (1993) (reporting on a study of state and federal charging practices in Los Angeles and finding that 222 crack defendants in state prosecutions were white, compared to 4410 black defendants; there were no white crack defendants in federal court, and 36 of 43 federal crack defendants were black).

n186. Armstrong, 517 U.S. at 465.

n187. Id.

n188. For a discussion of the difficulty of proving intentional discrimination by a prosecutor, see Donald G. Gifford, Equal Protection and the Prosecutor's Charging Decision: Enforcing an Ideal, 49 Geo. Wash. L. Rev. 659, 692-698 (1981) (discussing difficulty in using statistical data to establish prima facie case of prosecutorial bias); Vorenberg, supra note 24, at 1542 (noting near impossibility of proving a prosecutor's discriminatory motive); Tracey L. McCain, Note, The Interplay of Editorial and Prosecutorial Discretion in the Perpetuation of Racism in the Criminal Justice System, 25 Colum. J.L. & Soc. Probs. 601, 639 (1992) (noting the lack of recordkeeping of reasons for charging decisions as undercutting task of proving selective prosecution). But see Michael Selmi, Proving Intentional Discrimination: The Reality of Supreme Court Rhetoric, 86 Geo. L.J. 279 (1997) (arguing that intentional discrimination only requires proof of differential treatment, not proof of animus or illicit motive).

n189. In McCleskey v. Kemp, 481 U.S. 279 (1987), the defendant produced statistical evidence of the discriminatory effect of prosecutorial decisions to seek the death penalty, but such evidence was deemed insufficient to prove purposeful discrimination against Mr. McCleskey by prosecutors in the state of Georgia. See infra notes 219-27 and accompanying text.

n190. 481 U.S. 279 (1987).

n191. See id.

n192. See id. at 287-89; Kennedy, supra note 10, at 328-32 (discussing the Baldus study).

n193. See McClesky, 481 U.S. at 286-89.

n194. See id. at 286.

n195. See id. at 297.

n196. See id.

n197. See Kennedy, supra note 10, at 330-31 (discussing analysis of the Baldus study).

n198. "The policy considerations behind a prosecutor's traditionally 'wide discretion' suggest the impropriety of our requiring prosecutors to defend their decisions to seek death penalties, 'often years after they were made.'" McClesky, 481 U.S. at 296.

n199. But see Batson v. Kentucky, 476 U.S. 79 (1986). In Batson, the Supreme Court banned racially discriminatory peremptory challenges and established a process that requires the prosecutor to give a race neutral explanation for peremptory challenges after the defendant has made out a prima facie case that the prosecutor is using his strikes in a racially discriminatory manner. In his concurring opinion, Justice Marshall discussed the role of unconscious racism in prosecutorial decisions and suggested that prosecutors will always be able to come up with some race neutral reason to mask either unconscious or intentional discrimination. Cases decided since Batson seem to confirm Justice Marshall's prediction. See, e.g., United States v. Forbes, 819 F.2d 1006, 1010 (5th Cir. 1987) (finding a challenge to have been racially neutral when prosecutor "sensed by [the juror's] posture and demeanor that she was hostile to being in court"); Wallace v. State, 530 So. 2d 849 (Ala. Crim. App. 1987) (dismissing a peremptory challenge by prosecutor who removed a young black woman juror because she was a homemaker and lacked knowledge of what life was like out on the street, a middle-aged black woman because she appeared to be same age as defendants' parents, a young black man because he had a beard, and a middle-aged black man because he was unemployed); Charles J. Ogletree, Just Say No!: A Proposal to Eliminate Racially Discriminatory Uses of Peremptory Challenges, 31 Am. Crim. L. Rev. 1099, 1107-12 (1994) (describing courts that frequently accept race-neutral explanations that appear to be made on subconsciously racial grounds).

n200. Justice Brennan stated that "close analysis of the Baldus study, however, in light of both statistical principles and human experience, reveals that the risk that race influenced McCleskey's sentence is intolerable by any imaginable standard." McCleskey, 481 U.S. at 325 (Brennan, J., dissenting). In his dissent, Justice Blackmun noted that "the most persuasive evidence ... is McCleskey's proof that the race of the victim is more important in explaining the imposition of a death sentence than is the factor whether the defendant was a prime mover in the homicide." Id. at 355 (Blackmun, J., dissenting). Justice Stevens added that "the studies demonstrate a strong probability that McCleskey's sentencing jury ... was influenced by the fact that McCleskey is black and his victim was white, and that this same outrage would not have been generated if he had killed a member of his own race." Id. at 366 (Stevens, J., dissenting) (citation omitted); see also Kennedy, supra note 10, at 333-35 (discussing the dissenting opinions).

n201. Justice Scalia's Memorandum to the Conference, McClesky (No. 84-6811) (found among Justice Marshall's papers after his death).

n202. The Model Code of Professional Responsibility states that:

The responsibility of a public prosecutor differs from that of the usual advocate; his duty is to seek justice, not merely to convict. This special duty exists because: (1) the prosecutor represents the sovereign and therefore should use restraint in the discretionary exercise of governmental powers, such as in the selection of cases to prosecute.

Model Code of Professional Responsibility EC 7-13 (1981).

n203. See generally Meares, supra note 31 (proposing the use of financial incentives to encourage a higher standard of ethical behavior by prosecutors than required by the Constitution).

n204. See Berger v. United States, 295 U.S. 78, 88 (1935); Model Code of Professional Responsibility EC 7-13 ("The responsibility of a public prosecutor differs from that of the usual advocate; his duty is to seek justice, not merely to convict."); Model Rules of Professional Conduct Rule 3.8 cmt. (1983) (describing a prosecutor's responsibilities "to see that [a] defendant is accorded procedural justice and that guilt is decided upon the basis of sufficient evidence"); Standards Relating to the Admin. of Criminal Justice Standard 3-1.2(c) (1992) ("The duty of the prosecutor is to seek justice, not merely to convict."). For a critique of the Model Code's "do justice" standard, see Zacharias, supra note 117, at 48 (explaining that the 'do justice' standard establishes no identifiable norm and its vagueness leaves prosecutors with only their individual sense of morality to determine just conduct).

n205. 373 U.S. 83, 87 (1963) (citing Judge Simon E. Sobeloff, Address at the Judicial Conference of the Fourth Circuit (June 29, 1954)).

n206. Zacharias, supra note 117, at 57 (citing Carol A. Corrigan, On Prosecutorial Ethics, 13 Hastings Const. L.Q. 537, 538-39 (1986)).

n207. See Model Code of Professional Responsibility EC 8-1; see also id. EC 8-9 ("The advancement of our legal system is of vital importance in maintaining the rule of law and in facilitating orderly changes; therefore, lawyers should encourage, and should aid in making, needed changes and improvements."); Model Rules of Professional Conduct pmbl. ("A lawyer should be mindful of deficiencies in the administration of justice ...."); Standards Relating to the Admin. of Criminal Justice Standard 3-1.2(d) ("It is an important function of the prosecutor to seek to reform and improve the administration of criminal justice. When inadequacies or injustices in the substantive or procedural law come to the prosecutor's attention, he or she should stimulate efforts for remedial action.").

n208. See Model Rules of Professional Conduct Rule 3.8 (outlining special responsibilities of a prosecutor); Standards Relating to the Admin. of Criminal Justice Standard 3-1.1 (describing the function of the standards as a guide to professional conduct and performance); see also Melilli, supra note 27, at 697 (describing prosecutor's dual role as an advocate and as a minister of justice); Vorenberg, supra note 24, at 1524 (discussing how prosecutors shift from an adversarial role to deciding guilt and punishment). A California court described the prosecutor's role:

It is true that a public prosecutor, as representative of the People, must satisfy additional standards of conduct by reason of his position as the officer who possesses the power and authority to speak for the State. In practical effect the public prosecutor functions in a dual capacity - as both agent and principal, as both attorney and client. Because he exercises a dual function, the prosecutor possesses additional responsibilities and becomes subject to broader duties than does defense counsel, who only exercises the one function of agent-attorney.

People v. Kelley, 142 Cal. Rptr. 457, 466 (Ct. App. 1977).

n209. David M. Nissman & Ed Hagen, The Prosecution Function 7 (1982).

n210. See Stanley Z. Fisher, In Search of the Virtuous Prosecutor: A Conceptual Framework, 15 Am. J. Crim. L. 197, 217 (1988) (discussing prosecutor's "quasi-judicial" role for fairness and justice, and her "zealous advocate" role to maximize the number of convictions and severity of sentences imposed); Vorenberg, supra note 24, at 1557 (noting that a prosecutor's commitment to impartiality is inconsistent with the adversarial role); see also Mellili, supra note 27, at 698 ("The prosecutor is left with an ongoing schizophrenia, acting simultaneously as an advocate and a minister of justice. As a result, the prosecutor is faced with a dilemma. He or she must determine which role will take priority.").

n211. But see Paul Butler, Racially Based Jury Nullification: Black Power in the Criminal Justice System, 105 Yale L.J. 677 (1995) (arguing that the high number of black men in the criminal justice system is the result of societal racism, and advocating jury nullification by black jurors when African Americans are on trial for certain nonviolent felonies such as drug possession).

n212. See Butler, supra note 11, at 877 (arguing that African Americans should be arrested and sentenced to prison for drug possession only in proportion to involvement in those crimes and that by the year 2000, every jurisdiction in the United States should maintain a prison population which reflects racial demographics of that jurisdiction).

n213. See Vorenberg, supra note 24, at 1523-37 (discussing extent of prosecutorial discretion).

n214. The Supreme Court addressed the prosecutor's responsibility to balance the interests of law enforcement and protecting the rights of the accused in Berger v. United States, 295 U.S. 78 (1935).

He may prosecute with earnestness and vigor - indeed, he should do so. But, while he may strike hard blows, he is not at liberty to strike foul ones. It is as much his duty to refrain from improper methods calculated to produce a wrongful conviction as it is to use every legitimate means to bring about a just one.

Id. at 88.

n215. For a discussion of the legitimate factors considered in prosecutorial decision- making, see supra notes 105-06 and accompanying text.

n216. See supra Part III.B (discussing selective prosecution).

n217. For purposes of the report, "similarly situated" defendants would have committed the same criminal act and have similar criminal histories. "Similarly situated" victims would have the same level of interest in prosecution and similar criminal histories. Other characteristics of either the defendant or the victim (wealth, education, jury appeal, etc.) should not be relevant to the prosecutor's calculus, as they would involve discriminatory treatment based on subjective, inappropriate criteria.

n218. The prosecutor should not use conviction and sentencing rates as evidence of criminality, as the Supreme Court did in Armstrong. As Justice Stevens noted in his dissent in Armstrong, conviction and sentencing rates only reflect the number of individuals prosecuted and sentenced for certain crimes, not necessarily the number of individuals who committed these crimes. See supra notes 177-79 and accompanying text.

n219. See supra notes 192-99 and accompanying text.

n220. See Kennedy, supra note 10, at 330-31 (showing how the Baldus study has received high praise from many distinguished researchers).

n221. Id. at 329.

n222. See Developments in the Law, supra note 91, at 1529 n.31 (explaining multiple regression analysis); see also Franklin M. Fisher, Multiple Regression in Legal Proceedings, 80 Colum. L. Rev. 702 (1980) (discussing the concepts and proper use of multiple regression analysis in legal proceedings).

n223. See supra notes 105-06 and accompanying text.

n224. See supra note 158 and accompanying text.

n225. See supra notes 137-52 and accompanying text.

n226. See supra notes 202-08 and accompanying text.

n227. But see Poulin, supra note 25, at 1122-24 (suggesting that prosecutors should respond to claims of selective prosecution by conducting internal investigations and engaging in other forms of self-regulation).

n228. In Alaska, Hawaii, New Hampshire, New Jersey and Wyoming, the governor appoints the Attorney General. The legislature selects the Attorney General in Maine, and the state supreme court selects the Attorney General in Tennessee. See Bill Isaeff, Qualifications, Selection, and Term, in State Attorneys General: Powers and Responsibilities 15, 15 (Lynne M. Ross ed., 1990).

n229. See Misner, supra note 14, at 734.

n230. Recent examples of district attorney and attorney general races featuring "tough on crime" campaign themes with little detail on office policies include Suffolk County, N.Y., see Rick Brand, Democrats Bank on Anti-Catterson Theme, Newsday (Suffolk Ed.), June 5, 1997, at A34, Jefferson Parish, La., see Drew Broach, Jeff DA Candidates Spent Big, Owe Big, Times-Picayune, Jan. 16, 1997, at B1, Buffalo, N.Y., see Robert J. McCarthy, In Presidential Year, Two House Races Hold Local Interest, Buffalo News, Oct. 30, 1996, at 1F, Pennsylvania, see Peter J. Shelly, Fisher Exaggerates His Experience, Kohn Claims, Pittsburgh Post-Gazette, Oct. 26, 1996, at C1, Albuquerque, N.M., see Arley Sanchez, DA Faces Ex-Cop in Election, Albuquerque J., Sept. 26, 1996, at 1, and Baton Rouge, La., see Angela Simoneaux, DA Candidates for Crime Prevention, Baton Rouge Advocate, Aug. 30, 1996, at 1B. See also Misner, supra note 14, at 717 (noting difficulty for electorate in scrutinizing prosecutors' decisions).

n231. 28 U.S.C. 503 (1994).

n232. Id. 541(a).

n233. Id. 542(a).

n234. Id. 503, 541(a).

n235. See Joan E. Jacoby, The American Prosecutor: A Search of Identity 22 (1980); Abraham S. Goldstein, Prosecution: History of the Public Prosecutor, in 3 Encyclopedia of Crime and Justice 1287 (Sanford H. Kadish ed., 1983).

n236. In 1643, Virginia became the first colony to appoint a public prosecutor - the Attorney General. The Virginia system was modeled after the early English model as other colonies modeled their public prosecution systems after the native European countries of their early settlers. Early public prosecutors were appointed by either the court or the governor. They had little independence and discretion and were required to consult with the court or governor before making decisions. See Jacoby, supra note 235, at 11-21; see also Juan Cardenas, The Crime Victim in the Prosecutorial Process, 9 Harv. J.L. & Pub. Pol'y 357, 366-71 (1986) (surveying history of public prosecutor); Goldstein, supra note 235, at 1287 (describing development of public prosecutor).

n237. See Jacoby, supra note 235, at 22; Goldstein, supra note 235, at 1287. In the 1920s, crime commissions were formed in a number of states to examine the status of the criminal justice system and its ability to manage the post-World War I rise in crime. Most of these commissions conducted a close examination of the role of the prosecutor and were shocked by the extent of his power and discretion. A report by the National Commission on Law Observance and Enforcement noted: "In every way the Prosecutor has more power over the administration of justice than the judges, with much less public appreciation of his power. We have been jealous of the power of the trial judge, but careless of the continual growth of the power of the prosecuting attorney." Jacoby, supra note 235, at 28. Commissions formed in Georgia, New York, Illinois, Minnesota, Pennsylvania, and California made similar observations about the power of the prosecutor. See id. at 30.

The most well-known of the crime commissions of this era was the Wickersham Commission, a national commission formed to study the status of the criminal justice system. See id. at 31. The commission included a number of prominent legal scholars of the day, including Roscoe Pound of the Harvard Law School. Like virtually all of the state crime commissions, the Wickersham Commission was highly critical of the role of the prosecutor, particularly the absence of a meaningful check on prosecutorial power and discretion. See id. It noted that the popular election of prosecutors provided neither an adequate check on this power nor the best qualified candidates for the position. See id. The Commission also greatly criticized the plea bargaining power of prosecutors. See id. It recommended a number of reforms, including the establishment of a state director of public prosecutions with secure tenure to control the prosecutorial process in a systemized fashion. See id.

n238. For example, one commentator noted:

The reality is that nearly all ... decisions to prosecute or not to prosecute ... and nearly all his reasons for decisions are carefully kept secret, so that review by the electorate is nonexistent except for the occasional case that happens to be publicized. The plain fact is that more than nine-tenths of local prosecutors' decisions are supervised or reviewed by no one.

See Kenneth Culp Davis, supra note 118, at 207-08; see also Donald G. Gifford, Meaningful Reform of Plea Bargaining: The Control of Prosecutorial Discretion, 1983 U. Ill. L. Rev. 37, 54 (discussing the public's lack of access to information about plea bargaining); Daniel C. Richman, Old Chief v. United States: Stipulating Away Prosecutorial Accountability?, 83 Va. L. Rev. 939, 963 (1997) ("Even direct elections are not likely to prove an effective means of giving prosecutors guidance as to a community's enforcement priorities or of holding them accountable for the discretionary decisions that they have already made.").

n239. In fact, Daniel Richman writes:

Many elections for chief prosecutor are not even contested. Those that are may be fought on whether a specific type of crime should be prosecuted, whether a murderer deserves execution, or on the loss of a high-profile case, as well as on an office's overall win-loss record. Individual referenda on the broad range of discretionary choices that every prosecutor makes are unlikely, indeed utterly impossible.

Richman, supra note 238, at 963-64 (citations omitted); see also Misner, supra note 14, at 772-73 (noting that barely twelve percent of prosecutors' offices have written guidelines, leaving the public with little basis for judging prosecutors' effectiveness).

n240. Some prosecutors have campaigned on domestic violence issues, but few have openly discussed race issues. For examples of district attorneys campaigning against domestic violence, see Robert Greene, Garcetti, Lynch Debate District Attorney's Job in Opening Shot of Campaign, Metropolitan News-Enterprise, July 11, 1996, at 3; Brian Maffly, Abuse: The Cycle Is Vicious; Convictions Are Few in Domestic Cases, Salt Lake Trib., Sept. 30, 1996, at D1. One district attorney who openly talks about race, in addition to domestic violence, is San Francisco's Terence Hallinan. See infra note 250; see also DA Pledges Protection in Same-Sex Violence, S.F. Chron., May 23, 1997, at A24 (discussing Hallinan's commitment to legal protection for victims of domestic violence in same-sex communities).

n241. See supra note 235 and accompanying text.

n242. The publication of the studies might compel prosecutors to explain or justify their decisions, as the Supreme Court required in Batson v. Kentucky. 476 US. 79, 97 (1986). In Batson, the Court held that prosecutors could not use their peremptory strikes to eliminate African American jurors from a petit jury without providing a race-neutral reason for doing so. Id. at 89. Of course, such an exercise in the context of racial impact studies would be subject to the same criticisms as Batson. See supra note 168 and accompanying text.

n243. The use of racial stereotypes by prosecutors received rare public attention in April 1997 during the election campaign for Philadelphia D.A. Lynn Abraham, the incumbent, released a 10-year-old training video for prosecutors featuring her opponent, Jack McMahon. In the video, McMahon, who was then a Philadelphia prosecutor, made a presentation on how to pick juries that was laced with crude stereotypes. McMahon advised rookie prosecutors to avoid picking African Americans from low- income areas, as well as "young black women, teachers, doctors, social workers and smart people." He added that the aim of jury selection was not to ensure a fair trial but to gain a conviction. See Linda Loyd, D.A. Defends Her Release of Videotape, Phila. Inquirer, Apr. 4, 1997, at A1.

n244. See, e.g., United States v. Armstrong, 517 U.S. 456, 465 (1996) (requiring that the defendant produce some evidence that similarly situated defendants of other races could have been prosecuted but were not).

n245. See supra notes 189-201 and accompanying text.

n246. Poulin, supra note 25, at 1090. Professor Poulin maintains that this type of "soft enforcement" would allow the public to receive information that "may prompt a demand for more careful exercise of prosecutorial discretion." Id.

n247. In San Francisco, District Attorney Terence Hallinan has championed community prosecution. He has ordered Assistant D.A.'s in his office to visit city neighborhoods wearing jackets emblazoned with the words "Community District Attorney." Maura Dolan, A Liberal Lays Down the Law in S.F., L.A. Times, Apr. 5, 1997, at A1. Arguing that a prosecutor's job goes beyond simply prosecuting, Hallinan says that prosecutors must go to "high-crime neighborhoods to 'get to know the people and tell them how they can help their police and district attorneys. We have to break down the barriers of mistrust between the minorities and the criminal justice system.'" William Claiborne, San Francisco Prosecutor Tries 'Something Different': Crusader Applies Liberal Traditions to New Duties, Wash. Post, Feb. 20, 1996, at A3. Hallinan has spurred efforts to involve gang members in community activities, told A.D.A's to seek mentoring programs for many drug offenders instead of jail sentences, and refused to enforce California's stringent three-strikes-and-you're-out law against nonviolent repeat offenders. Id. He justifies community punishment for drug possession cases so that court dockets can be cleared for major drug traffickers and other serious offenders. Id. "Let's get the other junk out of the courtroom, the simple possessions and the kid on the street selling a rock or two of crack cocaine. The courts are so cluttered with these cases that when you get real [sic] serious crimes, it's a year or more before you can bring them to trial. It's crazy," says Hallinan. Id. He also opposes imprisoning large numbers of small-time drug dealers because of "the disparate impact it has on minority communities." Id.

n248. See Tonry, supra note 10, at 179-80 (discussing how politicians see "tough on crime" themes as a route to re-election). Elected judges sometimes are also tempted to appear tough on crime. See Stephen B. Bright & Patricia J. Keenan, Judges and the Politics of Death: Deciding Between the Bill of Rights and the Next Election in Capital Cases, 75 B.U. L. Rev. 759, 776-92 (1995).

n249. 470 U.S. 598 (1985); see supra notes 198-200 and accompanying text. In fact, such recordkeeping is already required by various civil rights and environmental laws. Cf. J. Gordon Arbuckle et al., Environmental Law Handbook 3.1 (11th ed. 1991) (discussing pervasiveness of reporting requirements in environmental statutes); Harold S. Lewis, Jr., Civil Rights and Employment Discrimination Law 5.9 (1997) (discussing employer recordkeeping requirements under Title VII of the Civil Rights Act of 1964).

n250. 470 U.S. at 607.

n251. See id.

n252. See Thomas and Fitch, supra note 90, at 523-24 (advocating that prosecutors use forms and checklists to record the basis of their charging decisions).

n253. See Bureau of Justice Statistics, U.S. Dep't of Justice, Sourcebook of Criminal Justice Statistics 13 tbl.1.11 (1995) (displaying the Federal Criminal Justice Budget for the fiscal years 1995 (actual) and 1996-2002 (estimated)); see also Kenneth B. Nunn, The Trial as Text: Allegory, Myth and Symbol in the Adversarial Criminal Process - A Critique of the Role of the Public Defender and a Proposal for Reform, 32 Am. Crim. L. Rev. 743, 802-13 (1995) (describing the funding disparity between prosecutors and public defenders); Gerald Lefcourt et al., Justice That Makes Sense, Champion, Dec. 1997, at 6, 6 (discussing how public defender services have been drastically reduced while funding for federal law enforcement, prosecution and prison construction has grown dramatically).

n254. Robert L. Spangenberg & Tessa J. Schwartz, The Indigent Defense Crisis is Chronic: Balanced Allocation of Resources is Needed to End the Constitutional Crisis, Crim. Just., Summer 1994, at 13, 13 (discussing the lack of public resources devoted to the criminal justice system and arguing that instead of putting more police on the street and building more prisons, more resources should be devoted to the prosecution, the courts, and public defense).

n255. See supra notes 91-103 and accompanying text.

n256. For a bill introduced by Congressman John Conyers that would require police to keep records of the race of persons detained in traffic stops to determine if police are using discriminatory tactics, see H.R. 118, 105th Cong. (1998).

n257. State and local prosecutors are generally independent, elected officials, see supra note 236, while police chiefs are often appointed by the mayor or county executive. In the federal system, United States Attorneys and the chiefs of law enforcement agencies such as the FBI, CIA and the DEA are all under the jurisdiction of the Attorney General of the United States. Thus, the Attorney General could order such collaborative efforts in her discretion.

n258. Simi Valley, California, has gained a worldwide reputation for racism since the city served as the site of the Rodney King beating trial. One city resident, on a medical mission in the jungles of Brazil, was being introduced to villagers through an interpreter. When asked where he was from, the man replied, "Simi Valley." A boy then jumped up and said, "Rodney King! Rodney King!" Mack Reed, Negative Image Created by Trial Haunts Simi Valley, L.A. Times, Apr. 28, 1997, at A3. City leaders have struggled mightily to repair the city's image, with very little success. Id.

The Arizona legislature's rejection of the Martin Luther King, Jr. holiday in 1990 led to a wave of 166 convention cancellations in the state and the moving of Super Bowl XXVII to California. Voters approved the holiday in 1992. Mike Mulligan, Second Effort for Arizona; Three Years Later, State Gets Its Super Bowl, Chicago Sun- Times, Jan. 22, 1996, at 70.

n259. See Randall Kennedy, The State, Criminal Law, and Racial Discrimination: A Comment, 107 Harv. L. Rev. 1255, 1256 (1994) ("The main problem confronting black communities in the United States is not excessive policing and invidious punishment but rather a failure of the state to provide black communities with the equal protection of the laws.").

n260. Professor Kennedy's defense of laws and enforcement practices which disproportionately impact African American communities has also been widely criticized by legal scholars. See Paul Butler, (Color) Blind Faith: The Tragedy of Race, Crime, and the Law, 111 Harv. L. Rev. 1270 (1998); David Cole, The Paradox of Race and Crime: A Comment on Randall Kennedy's "Politics of Distinction", 83 Geo. L.J. 2547 (1995); Kim Taylor-Thompson, The Politics of Common Ground, 111 Harv. L. Rev. 1306 (1998).

n261. The Anti-Drug Abuse Act of 1988, Pub. L. No. 100-690, 6371, 102 Stat. 4181, 4370 (codified at 21 U.S.C. 844(a)), established penalties for the possession and distribution of crack cocaine that were 20 times more severe than the penalties for the same amount of powder cocaine. Although 52% of reported crack users in 1991 were white, 84.5% of defendants convicted of possession of cocaine in 1993 were black. See National Institute on Drug Abuse, National Household Survey on Drug Abuse: Population Estimates 1991, at 38 tbl.5-B (1993).

n262. See Nkechi Taifa, Cracked Justice: A Critical Examination of Cocaine Sentencing, 27 U. West L.A. L. Rev. 107 (1996).

n263. Id.

n264. See Amendments to the Sentencing Guidelines for United States Courts, 60 Fed. Reg. 25,074 (proposed May 10, 1995).

n265. See Taifa, supra note 262, at 113.

n266. Id.

n267. Id.

n268. Peter Baker, Clinton Would Cut Disparity in Some Cocaine Sentences, Wash. Post, July 23, 1997, at A2.