For most of us, shooting and killing an unarmed man in broad daylight on a public street, as Darren Wilson did to Michael Brown, would result in almost certain arrest on the spot and swift prosecution. But the law treats police officers differently. In recognition of the nature and unique demands of the work they do, police are given more leeway to use deadly force, sometimes even against a person who is unarmed. The leeway is broad enough that most instances of police use of deadly force against an unarmed civilian are handled much as Wilson’s was—that is, police are called to the scene, but there is not an immediate arrest.
Of course, just because police typically do not immediately arrest a fellow officer who has shot and killed someone does not mean the shooting is not a crime. Moreover, although whether to arrest at the scene is largely up to the police, whether a formal charge will be brought and the matter pursued criminally will be determined, not by the police, but by a prosecutor. Thus, when Darren Wilson shot and killed Michael Brown, it fell to the chief prosecutor for the jurisdiction where the shooting took place, St. Louis County Prosecuting Attorney Robert P. McCulloch, to decide what criminal charges, if any, would be brought against Wilson.
Missouri law gave McCulloch three options: (1) he could decide on his own to bring charges against Wilson;1 (2) he could decide on his own that no charges would be brought;2 or (3) he could decide not to make the determination himself and instead put the matter in the hands of a grand jury.3 Within days of the shooting, McCulloch announced that he was going to go the grand jury route.
The Grand Jury Proceedings
The grand jury that heard the Wilson case had already been serving and hearing routine cases presented by prosecutors from McCulloch’s office for nearly four months by the time the presentation of evidence in the Wilson case got underway.4 McCulloch’s opening words to the grand jurors took note of their prior experience: “I want to tell you how this is going to proceed. Obviously, it is going to be different from a lot of the other cases that you’ve heard … during your term.”5 And indeed it was very different—not just from this particular grand jury’s other cases, but from how state grand jury proceedings are ordinarily handled.6
Differences from the Grand Jury “Norm”
One aspect of the Wilson proceedings that differed markedly from the usual grand jury proceeding was the scope of the prosecution’s presentation, in terms of the number of witnesses and volume of evidence presented and, as a result, the length of time it took to complete it. In a typical case, the prosecution calls one or two witnesses at most and presents few if any exhibits. Each witness gives a bare-bones account of the key facts of the case—the facts that, in the prosecution’s view, are sufficient to establish the requisite probable cause to believe that a crime was committed and that the suspect (often called the “target”) committed it.7 As a result, whole cases are often presented in well under an hour.8 In the Wilson case, by contrast, the grand jury met for about 70 hours over a three-month period, on 25 separate days, during which time the prosecution called 60 witnesses, played hours upon hours of video and audio recordings, and presented hundreds of photographs, maps, diagrams, reports, and other exhibits.9
Another major difference from the usual was that the prosecution purported to take a “neutral” stance as to whether any charges should be brought in the case. Ordinarily, a prosecutor presents a grand jury with proposed charges, in the form of an indictment; the grand jury decides if the charges are supported by probable cause; and, if so, they vote to return the indictment or bring the proposed charges (also called returning “a true bill”). If a prosecutor is not convinced that charges should be brought, then the prosecutor simply does not present the matter to a grand jury.10
Two other differences, both also significant departures from the grand jury norm, worked in tandem. First, the “target”—here, Darren Wilson—testified before the grand jury and gave his version of the relevant events; and second, through Wilson and other witnesses, the prosecution presented extensive testimony in support of Wilson’s claimed legal justifications for killing Brown. For a multitude of reasons, the target of an investigation rarely testifies before the grand jury. In Missouri, as in most jurisdictions, there is no right to do so, even if one is the grand jury’s target.11 If a target wants to testify (assuming they are even aware of the investigation), the most they can do is ask. Even if the prosecution is amenable, defense attorneys ordinarily counsel their clients against testifying, on the theory that the legal risks associated with doing so are simply too great.12
As rare as it is for a target to testify before the grand jury, it is even more unusual for the prosecution to present evidence favorable to the target, especially evidence that, if believed, would establish a complete defense to any possible charge. Although prosecutors in possession of such evidence do have a constitutionally imposed duty to disclose it to the defense for use at trial,13 in most jurisdictions, including Missouri, there is no comparable duty to present it to the grand jury.14 This is in keeping with the widely accepted view that the purpose of a grand jury is “not to determine guilt or innocence, but to assess whether there is adequate basis for bringing a criminal charge.”15
Cause for Concern
Of course, just because the grand jury presentation in the Wilson case differed markedly from the usual grand jury presentation does not mean there was anything wrong with it. For other reasons, however, several aspects of the presentation were quite troubling.
McCulloch’s Cryptic Message
Prosecuting Attorney McCulloch’s opening words to the grand jury about how the case was “obviously … going to be different from” the grand jury’s previous cases got the proceedings off to a questionable start.16 Recall that this grand jury had been hearing cases for months. Although the facts of each case were different, the prosecution’s approach in all of them was basically the same. Now here was the prosecuting attorney himself, telling the grand jurors that this next case would “obviously” be “different.” But what did he mean that it was going to be “different?” And in what sense was the difference “obvious?” More importantly, what message did it send to the grand jurors that the prosecution viewed it that way?
Prosecution-Created Confusion
McCulloch’s cryptic message aside, the grand jurors’ task was also complicated by certain of the prosecution’s choices as to what would and would not be presented to the grand jury—choices that made the presentation far more muddled, unfocused, and difficult to follow than it would have been had the prosecutors handled it in their usual way. McCulloch made the plan for the presentation clear at the outset: “Absolutely everything will be presented to the grand jury, every scrap of paper that we have, every photograph that was taken, every bit of physical evidence that has been gathered, every video clip, anything that we can get.”17 Left unsaid was an important corollary: presenting “[a]bsolutely everything” to the grand jury meant there would not be any of the sort of filtering that prosecutors usually do in order to help grand juries make sense of the cases they hear.
As a result of the decision to present “absolutely everything,” the grand jury heard testimony that no prosecutor who was actually trying to help them do their job—that is, trying to help them determine whether there was probable cause to believe that Darren Wilson had committed a crime when he killed Michael Brown—would ever present. With the law enforcement witnesses who investigated the shooting, for example, the prosecutors elicited testimony, not just about the shooting and the surrounding circumstances, but also about the general step-by-step procedures and protocols (sometimes described in excruciating and mind-numbing detail) that they customarily followed in their jobs.18 Even when prosecutors did elicit testimony about the case, their presentation was unfocused. In one instance they had a witness separately show the grand jury and testify about every one of the 161 crime scene photos the witness had taken. It was almost as though the prosecutors had thought to themselves, why select only some of the photos to show to the grand jury—for example, the clear ones, or the important ones, or the most helpful ones—when we can just have them look at all 161 photos instead?19 The prosecutors went so far with their no-filtering stance that even when they knew witnesses were deliberately lying, they presented the perjured testimony to the grand jury anyway.20
Not everything that made the grand jury presentation so confusing was attributable to the prosecution’s everything-but-the-kitchen-sink approach; prosecutors also created confusion about the applicable law. One of the most glaring instances occurred shortly before Darren Wilson began testifying, when prosecutors gave the grand jurors copies of the Missouri statute that specifies when police are justified in using deadly force to make an arrest.21 A portion of the statute purports to authorize conduct that the U.S. Supreme Court ruled decades ago is unconstitutional.22 Although prosecutors eventually told the grand jurors that there was a problem with the statute, they waited for over two months to mention it.23 For that entire two-month period, the grand jurors were left to believe that the problematic statute stated the rule for when Wilson could use deadly force to arrest Brown. Even when the prosecutors finally told the grand jury about the statute issue, their explanation of the problem and how the grand jury should deal with it was confusing: they simply distributed their own statement of the law to replace the original handout and told the grand jurors, “So the statute [we] gave you [before], if you want to fold that in half just so that you know don’t necessarily rely on that because there is a portion of that that doesn’t comply with the law.”24
The prosecutors also created confusion about the law by waiting until the last day of the proceedings to give the grand jurors the Missouri homicide statute.25 The statute was essential: it specified the various charges to be considered, and set forth the legal requirements governing the probable cause determinations that the grand jury needed to make for each one in deciding whether to indict Wilson. It also could have helped the grand jurors throughout the proceedings in determining what information was relevant and thus what questions to ask the witnesses when they testified. The prosecutors clearly understood how much more difficult the grand jury’s job would be without the statute, and early on they assured the grand jurors that a copy would be provided:
[N]ormally when we’ve charged somebody with an offense, you have the charge in front of you, and you don’t have that in this case…. [T]hat kind of leaves you not sure how you are supposed to look at this evidence. So after this morning session, [we] will sit down and … come up with statutes for you on the various degrees of homicide and … some other relevant statutes on the use of … deadly force … and possibly self-defense, so you will have [those] by … next time. We’ll have that for you so you can kind of at least understand the law as you are hearing this evidence.26
Although prosecutors did provide a copy of the use-of-deadly-force statute in time for the grand jurors to be able to “understand the law as [they were] hearing the evidence,”27 not so with the homicide statute. Thus, they informed the grand jury of a defense that Wilson might be able to raise before informing them of any crimes he might have committed.
Anti-Indictment Bias
Quite apart from the problems discussed thus far, the grand jury presentation was seriously compromised because it was driven from start to finish by a Darren-Wilson-should-not-be-indicted bias so pervasive and powerful that a “no true bill” vote was inevitable. To be clear, this is not a judgment about the actual intent of the prosecutors who made the presentation, nor is it a judgment about the correctness of the grand jury’s decision not to indict. Rather, the point is simply that whether consciously or unconsciously, the prosecutors conducted the grand jury proceedings in a way that made fair consideration of the option of returning an indictment against Darren Wilson virtually impossible.
The prosecution’s decision to grant Wilson’s request to appear before the grand jury benefited him greatly, but even after deciding that much, prosecutors still had complete control over when in the proceedings he would testify and what the grand jurors would see and hear before they heard from him. Most prosecutors who know that a grand jury target is going to testify would postpone the testimony until at or near the end of the proceedings, on the theory that hearing other (and often conflicting) accounts beforehand puts the grand jury and the prosecutor in a better position to scrutinize the testimony critically and with appropriate skepticism, and then question the target accordingly. Here, though, prosecutors took a different tack.
Wilson was allowed to testify relatively early in the proceedings—he was ninth of the 60 witnesses presented by the prosecution—and the proceedings leading up to his appearance could not have set the stage for his testimony any better had he planned them himself. Seven of the eight witnesses who testified ahead of him, although not themselves present for any of the events surrounding the shooting, either had taken a statement from Wilson about those events or had spoken to someone else who did, and prosecutors had all seven of them recount for the grand jurors what Wilson had said. As a result, by the time Wilson testified, the grand jury had repeatedly heard all or parts of his version of the events leading up to the shooting—including, for example, Wilson’s claims: (1) that Michael Brown had “assaulted and attempted to kill him” when Wilson was just sitting in his car;28 (2) that when Brown first started running away Wilson chased him because “he knew Brown would assault another responding officer or witness” and he “did not want Brown to cause injury or death to anyone else;”29 and (3) that when Brown suddenly turned around and charged directly at Wilson, Wilson thought his life was in danger and so felt he had no choice but to kill Brown.30
The grand jurors did hear one account that differed from Wilson’s before Wilson testified—namely, that of Dorian Johnson, the young man who had been with Michael Brown in the hours leading up to the shooting—but prosecutors took various measures to blunt the impact of his testimony. Chief among these measures was the playing of and questioning Johnson about a surveillance video that was at least somewhat unfavorable to Johnson and highly unfavorable to Brown.31 The video, taken on the morning of the shooting at a local convenience store known as the Ferguson Market, showed Brown taking items without paying and pushing a store clerk in the process32—an incident that came to play a prominent role in Wilson’s account of the events leading to his shooting of Brown.33
Also by the time Wilson testified, the grand jurors had heard favorable things about him and the opposite about Michael Brown. Wilson, they heard, “always ha[d] a smile on his face,” and was “very easy going” “a good officer,” and not one to “go … look[ing] for trouble” or start a fight with a suspect.34 Concerning Brown, by contrast, in addition to hearing repeated references to Wilson’s account of his aggressive conduct leading up to the shooting; the grand jurors were told that he came from a neighborhood that was “known for violence, guns, gangs and drugs;”35 they watched the Ferguson Market video; and they heard one of the prosecutors characterize Brown’s behavior in the video as “brash,” “threatening,” and “intimidating.”36 Then, having painted these starkly contrasting pictures of Wilson and Brown, shortly before Wilson testified, prosecutors distributed copies of the previously discussed Missouri use-of-deadly-force statute.37 From Wilson’s perspective, coming as it did shortly before he began his testimony about his own use of deadly force against Brown, the timing of the distribution could not have been better. Granted, the grand jurors were told months later that there was a problem with the statute, but as far as they knew when they heard from Wilson, the statute appeared to set forth a complete defense that Wilson could rely on to defeat any criminal charge the grand jury might vote to bring in connection with Wilson’s use of deadly force against Brown.
With the stage thus set, enter Wilson. At the risk of stating the obvious, no witness had a bigger stake in the outcome of these grand jury proceedings than he did. As a practical matter, his testimony raised two questions for the grand jurors: first, was his version of events believable; and second, even assuming he was telling the truth about what happened, was his conduct reasonable?38 Although the prosecutors presumably had their own views on both questions, as they themselves understood, legally, the grand jurors’ views were the only ones that mattered.39 Thus, it was up to the prosecutors to put aside any views of their own and conduct an examination of Wilson aimed at helping the grand jurors answer the believability and reasonableness questions for themselves. Certainly, any prosecutor would appreciate that to examine Wilson effectively they would need to scrutinize his testimony carefully and actively challenge what he said; otherwise, as one of these very prosecutors put it, “you don’t get to the truth.”40 At almost every turn, however, the prosecutors simply failed to do their job. Two examples, each applicable to one of the questions the grand jury needed to resolve, will help make this point.
On the believability question, Wilson claimed that he knew about the earlier Ferguson Market stealing incident before he encountered Michael Brown and Dorian Johnson on the day of the shooting. This was an integral part of his version of events.41 Although he said it was the two men walking in the middle of the street that had initially drawn his attention, Wilson described how that initial encounter had morphed into something more serious when he noticed that Brown was carrying cigarillos (the items reportedly taken from the Ferguson Market) and that Johnson was wearing a black shirt (clothing reportedly worn by one of the suspects), and it “clicked for [him]” that “these [we]re the two from the stealing.”42
But what if it did not happen that way? More specifically, what if Wilson truly did not know anything about the stealing incident when he encountered Brown and Johnson? Certainly, there was some basis for thinking he might not have. The very first person Wilson spoke to after the shooting, a Ferguson police sergeant who was Wilson’s supervisor at the time, testified that Wilson told him then and in subsequent conversations that he (Wilson) knew nothing about the Ferguson Market incident when he spotted the two men, and that everything had unfolded from their refusal to walk on the sidewalk.43 The prosecutors seemed surprised by this testimony and inquired again several times, but the sergeant was emphatic: Wilson said he did not know anything about the stealing incident.44 That is a significant discrepancy: either the sergeant was mistaken and Wilson really did mention having realized the connection to the stealing incident, or Wilson’s story changed. Either way, once the sergeant testified that Wilson did not know about the incident, the prosecutors owed it to the grand jury to try to “get to the truth” by challenging Wilson’s subsequent testimony that he did know about it. But that never happened; in fact, the prosecutors never even mentioned the discrepancy to Wilson.
Regarding the reasonableness of Wilson’s conduct, Wilson’s decision that he “had to kill [Brown]”45 was an obvious topic for inquiry. There were many missed opportunities, but one in particular came relatively early in Wilson’s testimony, when one of the prosecutors questioned him about the various weapons he had with him on the day of the shooting and asked specifically about a Taser:
Q. Did you carry a Taser?
A. No.
Q. Why not?
A. I normally don’t carry a Taser. We only have a select amount. Usually there is one available, but I usually elect not to carry one. It is not the most comfortable thing. They are very large, I don’t have a lot of room in the front for it to be positioned.
Q. Had you been trained on how to use a Taser?
A. Yes, ma’am.
* * *
Q. You prefer not to have a Taser?
A. Correct.46
Here again, the prosecutors owed it to the grand jury to do more than they did. Instead of simply confirming Wilson’s preference (“You prefer not to have a Taser?”), for example, they might have asked questions aimed at establishing (1) that Wilson could have been carrying a Taser that day but decided not to; (2) that a Taser enables a police officer to subdue a dangerous person without having to use deadly force like a gun; (3) that not carrying a Taser increases the risk of having to use deadly force to subdue someone; and (4) that Wilson understood all of this, yet opted not to carry a Taser anyway, mainly for reasons of personal comfort.
Likewise, consider Wilson’s subsequent testimony about Brown hitting him in the face through Wilson’s open car window when Wilson was seated in his car. Wilson said that from his perspective, that assault by Brown made Wilson’s gun a legitimate “deadly force option”; in other words, he felt that Brown’s hitting him in the face was enough to justify shooting Brown.47 The prosecutor’s response? “Okay, all right. So then you go to the [police] station?”48
Where was the scrutiny? Why not establish through Wilson the relatively minor nature of his facial injuries as a result of the hitting,49 and then challenge the reasonableness of his judgment that the hitting justified deadly force? And why not also ask Wilson whether he would have reached the same conclusion about the appropriateness of using his gun if he had been carrying a Taser at the time? Might the grand jurors have wondered, if Wilson had been carrying a Taser, maybe Michael Brown would be alive today?50
There were numerous instances of similar prosecutorial ball-dropping throughout the proceedings, and of course we all know what happened in the end: on November 24, 2014, nearly four months after Darren Wilson shot and killed Michael Brown, prosecuting Attorney McCulloch called a nighttime press conference to announce that the grand jury had voted not to indict Wilson. Later that night and over the next few weeks, McCulloch released literally thousands of pages of grand jury documents and transcripts—an almost unheard-of action, but one he had said he would take for the sake of “transparen[cy]” in the event that Wilson was not indicted.51 The release of the transcripts, in turn, brought to light what had gone on behind the closed doors of the grand jury room, including the problems previously discussed.
Ignoring Race (the Elephant in the Grand Jury Room)
In addition to the problems just noted, there was one other problem that was painfully evident from the grand jury transcripts: race mattered in this case, and yet its role was never even acknowledged in the grand jury, let alone discussed. As evidence for the proposition that race mattered here, we have, inter alia, the March 2015 Department of Justice report on the Ferguson Police Department,52 the available empirical evidence (scant as it is) about police shootings of unarmed African American men and boys,53 a large body of social science research on the subject of implicit (unconscious) racial bias,54 and, most importantly, accounts from African Americans about the experience of being Black in this country55 and, even more to the point, accounts from African American men about their interactions with police.56
The Justice Department report paints a picture of how the Ferguson police were policing Ferguson on August 9, 2014, the day Michael Brown was killed. And Darren Wilson’s grand jury testimony reveals the lens through which Wilson was viewing Michael Brown that day. Although prosecutors did not question Wilson directly about his racial attitudes, research concerning the phenomenon of implicit bias suggests that his answers probably would not have been as informative on the point as other aspects of what he said.57
Implicit bias refers to the negative attitudes and stereotypes that influence our perceptions, decisions, and actions at the subconscious level, without any intent or awareness on our part.58 With implicit racial bias in particular, studies suggest that most of us are influenced at some level by messages we constantly receive in various ways that stereotype Black men as violent, dangerous, and inclined toward criminality.59 According to one study, just over 40 percent of White Americans believe that “many” or “almost all” Black men—men like Michael Brown—are violent.60 Studies also indicate that views about the dangerousness of Black men are often associated with views that Black people are superhuman, nonhuman, and/or less susceptible to pain—all of which, taken together, serve to increase the perceived danger, which in turn, warrants greater fear.61
Before the grand jury, Wilson testified that at one point he tried to grab Brown, [but] “the only way I can describe it is I felt like a five-year-old holding onto Hulk Hogan.”62 A short time later, according to Wilson, Brown “looked up at me and had the most intense aggressive face. The only way I can describe it, it looks like a demon, that’s how angry he looked.”63 After Wilson fired his first round of shots, Brown “made like a grunting, like aggravated sound.”64 When Wilson fired another round, “it looked like he was almost bulking up to run through the shots, like it was making him mad that I’m shooting at him. And the face that he had was looking straight through me, like I wasn’t even there, I wasn’t even anything in his way.”65
Wilson also testified that after Brown hit him in the face twice, “I felt that another one of those punches in my face could knock me out or worse. I mean it was, he’s obviously bigger than I was and stronger and the, I’ve already taken two to the face and I didn’t think I would, the third one could be fatal if he hit me right.”66 Bearing in mind the stereotype of the superhuman Black man, what is significant about this testimony is how poorly it fits with the nature and extent of Wilson’s actual injuries as a result of his encounter with Brown: some scratches on his neck and a contusion on the right side of his face67—not nothing, to be sure, but arguably also not enough to reasonably fear that a third punch might kill him.
The prosecutors in some ways contributed to Wilson’s portrayal of Brown as out of control and frightening by suggesting that Brown’s prior marijuana use (as shown by a toxicology report) might have caused his extreme aggression toward Wilson (as described by Wilson). They had one witness testify that paranoia, hallucinations, and even psychosis are possible from marijuana use, and that there was no way to know from testing whether Michael Brown was experiencing any of those in the time just before Wilson shot him.68 They also questioned several witnesses about a highly concentrated and potent marijuana product known as “wax,” and about Michael Brown’s possible use of it (although their basis for asking is unclear, as there seems to have been no evidence that Brown had ever used it and actually some evidence suggesting that he had never heard of it).69
The reason all of this matters is that the reasonableness of Darren Wilson’s assessment of Brown’s dangerousness is a factor that the grand jury needed to consider in determining whether Wilson should be criminally charged for shooting Brown.70 Indeed, the reasonableness of police assessments of danger, and hence their fear, is an issue in many police use-of-force cases.71 Unless we are going to say that it is always reasonable for police officers to fear Black men because Black men are dangerous (which makes race a proxy for danger and an automatic trigger for the lawful use of deadly force), grand jurors will sometimes need to consider the role of race in an officer’s assessment of danger in determining whether the assessment was reasonable.72 Prosecutors can play a key role in all of this. Conversations about race can be difficult, especially among strangers. Prosecutors, by first informing themselves and then educating grand jurors about implicit bias, by their questioning of witnesses, and by their decisions regarding what witnesses to call are in a position to help grand jurors discuss issues of race openly and honestly.73 In the Darren Wilson case, unfortunately, that did not happen; in fact, in some instances, even if unconsciously, the prosecutors did just the opposite.
The prosecution’s marijuana theory, which lent support to Wilson’s negative portrayal of Brown, is one example of this, but some of the clearest examples occurred in connection with the questioning of various witnesses about Canfield, the overwhelmingly African American apartment complex where Michael Brown was shot and killed. The very first witness who testified, an investigator for the medical examiner’s office who was called to the scene of the shooting shortly after it occurred, was asked whether he was frightened when he was at Canfield.74 Another witness, a Ferguson police sergeant, was asked, “Didn’t people at Canfield hate police?” The witness responded that it was a “business relationship.”75 Apparently not satisfied with the witness’s answer, the prosecutor pressed further: “So there was no understanding that the residents just hated the police, it wasn’t like that?”76 (Interestingly, there was no comparable inquiry concerning how the police felt about the Canfield residents.)
When Wilson testified, the prosecutor raised the subject again:
Q. Did you guys have a volatile, well, how can I put this. Did you not really get along well with the folks that lived in [Canfield]?
A. It is an antipolice area for sure.
Q. And when you say antipolice, tell me more?
A. There’s a lot of gangs that reside or associate with that area. There’s a lot of violence in that area, there’s a lot of gun activity, drug activity, it is just not a very well-liked community. That community doesn’t like the police.77
One might ask, what were the prosecutors trying to convey to the grand jurors through this line of questioning? More importantly, what would the grand jurors have understood it to mean?
The Decision to Use a Grand Jury
With so much to criticize about the grand jury presentation in this case it is well to remember that under Missouri law, Prosecuting Attorney McCulloch did not have to use a grand jury at all; rather, he could have made the charging decision on his own. If he decided not to bring charges, he simply could have announced his decision; if he decided to go forward with charges, he could have brought them himself by filing a complaint, presenting evidence at a preliminary hearing, and, if the judge made the necessary probable cause determination, filing an information against Wilson and proceeding to trial.78 The significance of the choice McCulloch made—presenting the case to a grand jury instead of to a judge at a preliminary hearing—would be difficult to overstate. A preliminary hearing is an adversary proceeding (both prosecution and defense are represented), held before a judge in a courtroom that is open to the public, at the conclusion of which the judge decides, based on the evidence presented, whether the probable cause standard has been met. A grand jury proceeding, by contrast, is nonadversarial (prosecutor only; no defense counsel, no judge), it is held in secret, and the decision maker is a group of lay people who, not being law-trained and having neither the authority nor the means to do their own investigating, are completely dependent on the prosecutor. When McCulloch chose to use the grand jury for Wilson, he was opting for a process that he knew would ensure secrecy, near-complete prosecutorial control, and a setting in which Wilson could safely present his version of what happened, including his claimed justifications for shooting Brown, free from cross-examination and any real-time public or media scrutiny. McCulloch also knew that the grand jury process would offer him political cover because, strictly speaking, the final decision would be the grand jury’s and not his own. And although his after-the-fact release of grand jury documents was unusual, it did nothing to alter the essential nature of the secret, prosecutor-controlled proceedings that resulted in the decision not to bring criminal charges against Wilson for killing Michael Brown.
Lessons Learned?
The St. Louis County criminal justice system’s treatment of the Wilson case followed what has become a depressingly familiar pattern: a police officer (usually White79), while acting in the course of the officer’s police duties, kills or seriously injures an unarmed Black man or boy; the local prosecutor puts responsibility for deciding whether criminal charges should be brought against the officer in the hands of a local grand jury; and, after a secret, prosecution-controlled presentation of evidence, the grand jury votes not to indict.80 Recently, however, the Wilson case and a string of other well-publicized police killings have put a spotlight on the virtual immunity from prosecution that police seem to have been granted in these cases.81 Public awareness of the pattern has prompted widespread skepticism about the fairness and impartiality of the system that has been producing it,82 and that in turn has sparked calls for change in how the cases are handled. The calls for change have focused on two points in particular: (1) the people who are making all the key charging-related decisions in these cases—namely, local prosecutors—work hand-in-hand with the police officers whose use of force is ostensibly being examined for possible prosecution;83 and (2) the presentations to the grand jury (handled by these same local prosecutors) are made in secret, out of public view.84
Some have argued that oversight of these cases should be taken out of the hands of local prosecutors and shifted to a prosecuting authority that does not have such close ties with local police.85 The theory is that an inherent conflict of interest exists whenever local prosecutors are called on to decide whether to bring criminal charges against the very police officers whose cooperation, help, and goodwill the prosecutors depend on in order to do their work. The prosecutors who presented the Wilson case to the grand jury were laboring under just this sort of conflict, and while we cannot know for sure, the likelihood that it played a part in the bias they showed throughout the presentation—including, above all, their kid-glove treatment of Wilson—seems inescapable.
Others, focusing on the use of grand juries and the secrecy that attends them, have urged the use of a less secretive and more open charging process, if not in all cases, then at least in police lethal force cases.86 Secrecy breeds suspicion, all the more when it conceals something that seems so clearly the public’s business as the determination at issue in these cases—that is, whether a public servant who has used lethal force to kill an unarmed member of the very public supposedly being served at the time should be charged with a crime. Particularly for those who have little or no confidence in the fairness of the criminal process to begin with, chief among them the Black men and boys who are so disproportionately the victims when police kill, the secrecy only serves to confirm their sense that the fix is in.
Less often mentioned in discussions of how to reform the charging process in police lethal force cases is dealing with the subject of race when the person the police have killed is a Black man. For years, it has been standard operating procedure for prosecutors to ignore race in these cases, as though race were irrelevant. The Wilson case is a perfect example. But if we did not know better before, surely we do now. Instead of ignoring race, as was done in the Wilson case, there is a need for measures designed to help the various decision makers in the charging process address the special issues of race that these cases present. Whatever else the measures may entail, at a minimum, they should include familiarizing everyone—preliminary hearing judges, grand jurors where grand juries are used, and above all, prosecutors—with the phenomenon of implicit racial bias, and encouraging them to hold honest conversations about race.
Notes
1. Missouri permits felony charges to be brought by grand jury indictment or by complaint and information, and allows prosecutors to choose which way to proceed. See MO. REV. STAT. § 545.010. When a prosecutor chooses the latter, the process begins with the filing of a written complaint, MO. R. CRIM. P. 22.01, after which a hearing is held to determine if there are sufficient grounds (probable cause) to hold a trial on the charges, MO. R. CRIM. P. 22.09(a). If probable cause is found, in order to proceed with the prosecution, the prosecutor must file a formal written charge, called an “information.” MO. R. CRIM. P. 23.03. Both the initial complaint and the subsequent information are signed by, and subject to the sole control of, the prosecutor.
Not all prosecutors have the option of bringing charges on their own. In federal criminal proceedings, for example, the Fifth Amendment requires that felony charges be brought by grand jury indictment. U.S. CONST. amend. V. Although that right does not apply in the states, Hurtado v. California, 110 U.S. 516, 538 (1884), nearly half the states have similar requirements under their own constitutions and/or laws. See Susan Brenner & Lori Shaw, Grand Jury Functions, FED. GRAND JURY WEBSITE, http://campus.udayton.edu/~grandjur/stategj/funcsgj.htm.
2. American prosecutors have “virtually unlimited discretion not to proceed with a case.” Gerard E. Lynch, Prosecution: Prosecutorial Discretion, in 3 ENCYCLOPEDIA OF CRIME & JUSTICE 1248 (Joshua Dressler ed., 2d ed. 2002).
3. MO. REV. STAT. §§ 545.010, 545.031; MO. CONST. art. I, § 17. A grand jury is a panel of citizens convened to determine whether evidence presented by a prosecutor is sufficient to bring criminal charges against the supposed perpetrator and hold a trial. Grand jury proceedings are held in secret without a judge and without defense counsel; only the prosecutor, the witnesses chosen by the prosecutor, and the grand jurors (and sometimes a stenographer) are permitted to attend. See MO. REV. STAT. §§ 540.130, 540.320. This is to be distinguished from a trial jury, which comes into play later in the process, only after criminal charges are brought, whether by a grand jury or by a prosecutor acting alone.
Some favored yet another option, which would have involved replacing McCulloch with a special prosecutor, whether by having him voluntarily recuse himself or by having Missouri Governor Jay Nixon remove him involuntarily. Those advocating this option questioned McCulloch’s ability to handle the case against Wilson impartially, based in part on his close family ties to local police, but even more on his record of failing to prosecute police in similar cases. See Jaeah Lee, Ferguson Cop Darren Wilson Is Just the Latest to Go Unprosecuted for a Fatal Shooting, MOTHER JONES (Nov. 24, 2014), http://www.motherjones.com/politics/2014/11/darren-wilson-grand-jury-decision-ferguson-police-prosecutions. Ultimately, however, efforts to get a special prosecutor appointed were unsuccessful: the situation did not fit Missouri’s requirements for replacing the regular prosecutor, see MO. REV. STAT. § 56.110; Governor Nixon declined to get involved, see Jo Mannies, Nixon Sticking with McCulloch, Who Has No Plans to Step Out of Ferguson Case, ST. LOUIS PUB. RADIO (Aug. 19, 2014), http://news.stlpublicradio.org/post/nixon-sticking-mcculloch-who-has-no-plans-step-out-ferguson-case; and McCulloch rejected calls for him to recuse himself voluntarily, Interview by McGraw Milhaven with Robert McCulloch, Prosecuting Attorney for St. Louis County, MO, KTRS.COM (Aug. 20, 2014), http://www.ktrs.com/mcculloch-tells-nixon-to-man-up-and-make-decision/.
4. See Christine Byers, Grand Jury Now Has until January to Decide Whether to Charge Ferguson Officer, STLTODAY.COM (Sept. 16, 2014), http://www.stltoday.com/news/local/crime-and-courts/grand-jury-now-has-until-january-to-decide-whether-to/article_aa4111fc-2952-54c9-8316-76c4867dea48.html.
5. Transcript of: Grand Jury vol. 1, 5 (2014) [hereinafter Tr. vol. __, __ (2014)], http://apps.stlpublicradio.org/ferguson-project/evidence.html.
6. Allegations about some of these differences and how noticeable they were to the grand jurors were central to an unusual lawsuit brought by one of the grand jurors after the proceedings in the Wilson case had concluded. See Complaint ¶¶ 19–22, Grand Juror Doe v. McCulloch, No. 4:15-cv-00006, 2015 WL 47623 (E.D. Mo. Jan. 5, 2015), dismissed on abstention grounds, 2015 WL 2092492 (E.D. Mo. May 5, 2015). Brought first in federal court, dismissed there “to allow Missouri courts to address … [the unsettled state law issues raised by the lawsuit],” Grand Juror Doe v. McCulloch, 2015 WL 2092492, at 1 (E.D. Mo. May 5, 2015), and then brought in state court, see Grand Juror Doe v. McCulloch, 15SL-CC01891 (Mo. Cir. Ct. 21st Cir. June 2, 2015), the suit (still pending as of this writing) seeks release from the secrecy obligations normally imposed on grand jurors under state law to allow the plaintiff to speak out about her or his experience as a grand juror and participate in public discussions about issues in the case, especially issues involving race relations, see Compl. ¶¶ 1, 3, 4, 33, Grand Juror Doe v. McCulloch, 15SL-CC01891 (Mo Cir. Ct. 21st Cir. June 2, 2015).
7. See 19 MISSOURI PRACTICE, CRIMINAL PRACTICE & PROCEDURE § 12:2 (3d ed. 2013).
8. See Tr. vol. 2, 7 (2014).
9. Robert McCulloch, Prosecuting Att’y for St. Louis Cnty., MO, Missouri Grand Jury Decision Announcement (Nov. 24, 2014), C-SPAN, http://www.c-span.org/video/?322925-1/ferguson-missouri-grand-jury-decision-announcement.
10. Although some commentators posited that the prosecution’s neutral stance was appropriate because the Wilson grand jury was not serving as a “regular (or indicting) grand jury,” but as an “investigative grand jury,” see, e.g., Matt Hodapp & Dan Margolies, Attorneys in Missouri Debate Role of Grand Jury, KCUR.ORG (Dec. 5, 2014), http://kcur.org/post/attorneys-missouri-debate-role-grand-jury, Missouri law does not recognize this distinction or even mention anything about different types of grand juries, “investigative” or otherwise, see MO. REV. STAT. § 545.031; MO. CONST. art. I, § 17.
11. See 19 MISSOURI PRACTICE: CRIMINAL PRACTICE & PROCEDURE § 12:8 (3d ed. 2013).
12. See William Glaberson, New Trend before Grand Juries: Meet the Accused, N.Y. TIMES, June 20, 2004, http://www.nytimes.com/2004/06/20/nyregion/new-trend-before-grand-juries-meet-the-accused.html (noting as one of the main risks “the potential use of grand jury testimony to poke holes in a defendant’s account of events at trial”).
13. Brady v. Maryland, 373 U.S. 83, 87 (1963).
14. State v. Easter, 661 S.W.2d 644, 645 (Mo. Ct. App. 1983).
15. United States v. Williams, 504 U.S. 36, 51 (1992).
16. See supra text accompanying note 5.
17. Robert McCulloch, Prosecuting Att’y for St. Louis Cnty., MO, Press Conference, KSDK.COM (Aug. 13, 2014), http://www.ksdk.com/story/news/local/2014/08/13/st-louis-county-bob-mcculloch-press-conference-ferguson-shooting/14019205.
18. See, e.g., Tr. vol. 2, 21–37 (2014).
19. Id. at 61–175 (2014).
20. Interview by McGraw Milhaven with Robert McCulloch, Prosecuting Att’y for St. Louis Cnty., MO, KTRS.COM (Dec. 19, 2014), http://www.ktrs.com/st-louis-prosecuting-attorney-bob-mcculloch-breaks-silence/.
21. See Tr. vol. 5, 5 (2014). Darren Wilson testified that he shot Brown while trying to arrest him, see id. at 232, thus making the statute, MO REV. STAT. § 563.046 (1979), potentially relevant.
22. The problematic portion is MO REV. STAT. § 563.046 (3)(2)(a) (1979), which says that police are justified in using deadly force to arrest a fleeing felon even if the person fleeing is unarmed and seemingly not otherwise dangerous. This contravenes the Court’s decision in Tennessee v. Garner, 471 U.S. 1 (1985), which struck down as unconstitutional a similar Tennessee statute. Id. at 10.
23. Prosecutors distributed the statute on September 15, see Tr. vol. 5, 5 (2014), but did not mention the problem until November 21, see Tr. vol. 24, 134 (2014).
24. Tr. vol. 24, 135 (2014).
25. See id. at 131–33 (2014).
26. Tr. vol. 2, 14–15 (2014). The assurance came on Day 2 of the proceedings; the homicide statute was not distributed until Day 25. See Tr. vol. 24, 132–33 (2014).
27. See supra text accompanying note 21. Of course, the understanding provided by the statute was itself problematic. See supra note 22 and accompanying text.
28. Tr. vol. 5, 31 (2014).
29. Id. at 33–34.
30. Id. at 164.
31. Tr. vol. 4, 13, 31–37 (2014). Other measures included impeaching Johnson with prior inconsistent statements, see id. at 5, and questioning him about past encounters with the law, see id. at 171–76.
32. See id. at 31–37.
33. See infra text accompanying notes 41 & 42.
34. Tr. vol. 5, 74 (2014).
35. Id. at 170.
36. Tr. vol. 4, 84–86 (2014).
37. See supra text accompanying note 21.
38. The believability question is one that would apply to any witness. The reasonableness question ties in with certain of the potential charges against Wilson (specifically, both voluntary and involuntary manslaughter, see MO REV. STAT. §§ 565.023.1, 565.024.1), and with both affirmative defenses that prosecutors told the grand jurors were applicable, law enforcement use of force to arrest, see id. § 563.046, and self-defense, see id. § 563.031.1.
39. See Tr. vol. 24, 142 (2014).
40. Id. at 141 (quoting in full: “We want you to understand as attorneys it is our job to challenge witnesses’ statements and that sometimes, you know, you don’t get to the truth unless you challenge a witness statement.”).
41. Tr. vol. 5, 202, 209 (2014).
42. Id. at 206–07, 209.
43. See id. at 52–53, 57–58, 31.
44. See id. at 52–53, 57–58
45. Id. at 236.
46. Id. at 205–06.
47. See id. at 236–37.
48. Id. at 237.
49. Wilson’s injuries consisted of some scratches on his neck and bruising and swelling of his right jaw, resulting in jaw pain, for which an anti-inflammatory and ice were prescribed. See Tr. vol. 22, 77, 89–90 (2014).
50. In fairness, one of the prosecutors did initiate one reasonableness-related inquiry, concerning why Wilson did not remain in his car and wait for backup instead of trying to handle Brown on his own, see id. vol. 5, 261 (2014), but after a few questions on the subject the other prosecutor intervened with a different line of questioning and the matter was dropped, see id. at 261–62.
51. Id. vol. 1, 21 (2014). Whether McCulloch’s release of the documents was legal is far from clear—it relied on a highly questionable interpretation of Missouri’s Sunshine (open government) Law, MO. REV. STAT. §§ 610.020 et seq. (1973)—but no one came forward to object so the issue was never litigated.
52. CIVIL RIGHTS DIV., U.S. DEP’T OF JUSTICE, INVESTIGATION OF THE FERGUSON POLICE DEPARTMENT (Mar. 4, 2015), http://www.justice.gov/sites/default/files/opa/press-releases/attachments/2015/03/04/ferguson_police_department_report.pdf (finding, inter alia, that “Ferguson’s approach to law enforcement both reflects and reinforces racial bias, including stereotyping[,]” … that this “disproportionately … [harms] African Americans, and [that] there is evidence that this is due in part to intentional discrimination on the basis of race,” id. at 4).
53. Data on deaths and serious injuries caused by police use of deadly force is woefully incomplete. See Carl Balik, A New Estimate of Killings by Police Is Way Higher—And Still Too Low, FIVETHIR-TYEIGHT.COM (Mar. 6, 2015), http://fivethirtyeight.com/features/a-new-estimate-of-killings-by-police-is-way-higher-and-still-too-low/ (discussing various databases and their shortcomings). Mindful of the limitations, researchers and statisticians have nonetheless found what they believe to be reliable evidence of racial disparities. All agree, for example, that Black people are more likely than White people to be shot and killed by police, with estimates of the disparity ranging from three times more likely, see, e.g., Mapping Police Violence, http://mappingpoliceviolence.org/ [hereinafter Mapping Police Violence 2014 Study]; Kimberly Kindy (reported by Julie Tate, Jennifer Jenkins, Steven Rich, Keith L. Alexander & Wesley Lowery), Fatal Police Shootings in 2015 Approaching 400 Nationwide, WASH. POST, May 30, 2015, http://www.washingtonpost.com/national/fatal-police-shootings-in-2015-approaching-400-nationwide/2015/05/30/d322256a-058e-11e5-a428-c984eb077d4e_story.html [hereinafter Wash. Post 2015 Study] to as high as 21 times more likely, see Ryan Gabrielson, Ryan Grochowski Jones, & Eric Sagara, Deadly Force, in Black and White, PROPUBLICA (Oct. 10, 2014; updated Dec. 23, 2014), http://www.propublica.org/article/deadly-force-in-black-and-white. Two that have looked specifically at situations where police on duty have shot and killed unarmed individuals, as in the Wilson case, have found even more pronounced disparities: just over one-half of all of these involved Black victims according to one of the studies, see Mapping Police Violence 2014 Study; two-thirds according to the other, see Wash. Post 2015 Study.
54. See, e.g., Kirwan Inst. for the Study of Race & Ethnicity, State of the Science: Implicit Bias Review 2015, http://kirwaninstitute.osu.edu/implicit-bias-review/, and sources cited therein.
55. See, e.g., Kimberly Norwood, Why I Fear for My Sons, CNN (Aug. 25, 2014), http://www.cnn.com/2014/08/25/opinion/norwood-ferguson-sons-brown-police/; Arienne Thompson, Voices: The Exhausting Task of Being Black in America, USA TODAY Dec. 4, 2014, http://www.usatoday.com/story/life/people/2014/12/04/the-exhausting-task-of-being-black-in-america/19894223/.
56. See, e.g., Stan Chu Ilo, Being a Black Male in America: Racism and the Police, HUFFINGTON POST, BLACKVOICES (Apr. 10, 2015), http://www.huffingtonpost.com/stan-chu-ilo/being-a-black-male-in-ame_b_7035468.html.
57. This is not to rule out the possibility of explicit (conscious) racial bias, but such views are sufficiently unacceptable under current social norms that a person holding them is unlikely to admit it, at least when talking to strangers.
58. See KIRWAN INST. FOR THE STUDY OF RACE & ETHNICITY, STATE OF THE SCIENCE: IMPLICIT BIAS REVIEW 2013, at 7, http://kirwaninstitute.osu.edu/initiatives/implicit-bias-review/.
59. See, e.g., Jennifer L. Eberhardt et al., Seeing Black: Race, Crime, and Visual Processing, 2004 J. PERSONALITY & SOC. PSYCHOL. 876.
60. See John Sides, Trayvon Martin and the Burden of Being a Black Male, MONKEY CAGE (July 15, 2013), http://themonkeycage.org/2013/07/15/trayvon-martin-and-the-burden-of-being-a-black-male/ (discussing study conducted with Ismail White). By comparison, only about 12 percent of Whites believe that “many” or “almost all” White men are violent. Id.
61. Adam Waytz, Kelly Marie Hoffman & Sophie Trawalter, A Superhumanization Bias in Whites’ Perceptions of Blacks, 6 SOC. PSYCH. & PERSONALITY SCI. 352 (2015).
62. Tr. vol. 5, 212 (2014). (emphasis added)
63. Id. at 224–25. (emphasis added)
64. Id. at 227. (emphasis added)
65. Id. at 228. (emphasis added)
66. Id. at 216 (2014).
67. See supra note 49.
68. See Tr. vol. 19, 66–67, 70, 74 (2014).
69. See, e.g., Tr. vol. 11, 100–04 (2014); Tr. vol. 12, 170–71 (2014); Tr. vol. 13, 99 (2014); Tr. vol. 19, 70–74 (2014).
70. See supra note 38.
71. See Carol D. Leonnig, Current Law Gives Police Wide Latitude to Use Deadly Force, WASH. POST, Aug. 28, 2014, http://www.washingtonpost.com/politics/current-law-gives-police-wide-latitude-to-use-deadly-force/2014/08/28/768090c4-2d64-11e4-994d-202962a9150c_story.html.
72. See Falguni A. Sheth, Shoot First, Ask Later: Why the Concept of “Reasonable Fear” Is Anything but Reasonable, SALON (Sept. 6, 2014), http://www.salon.com/2014/09/06/shoot_first_ask_later_why_the_concept_of_reasonable_fear_is_anything_but_reasonable/.
73. See Jerry Kang et al., Implicit Bias in the Courtroom, 59 UCLA L. REV. 1124, 1181–84 (2012) (discussing possible methods of educating trial jurors about implicit bias). Prosecutors, of course, are as susceptible to implicit biases as the rest of us. See Robert J. Smith & Justin D. Levinson, The Impact of Implicit Bias on the Exercise of Prosecutorial Discretion, 35 SEATTLE U. L. R. 795 (2012). Lest there be concern that talking about race in this context might be counterproductive, research on the subject suggests otherwise. See Kang et al., supra, at 1184 (addressing a similar concern about trial jurors, and discussing research suggesting that when jurors are encouraged to discuss race, “it is precisely this greater degree of discussion, and even confrontation, that can potentially decrease the amount of biased decision-making”). For a discussion of the research, see Alexander M. Czopp et al., Standing up for a Change: Reducing Bias through Interpersonal Confrontation, 90 J. PERSONALITY & SOC. PSYCHOL. 784 (2006).
74. Tr. vol. 1, 43 (2014).
75. Tr. vol. 5, 50 (2014).
76. Id. at 51.
77. Id. at 238.
78. See supra note 1.
79. Subconscious biases that cause people to associate Black men with violence and criminality are not the exclusive province of Whites. See Theodore R. Johnson, Black-on-Black Racism: The Hazards of Implicit Bias, ATLANTIC (Dec 26, 2014), http://www.theatlantic.com/politics/archive/2014/12/black-on-black-racism-the-hazards-of-implicit-bias/384028/.
80. Sometimes prosecutors in police killing cases have decided on their own not to bring charges, thus skipping the grand jury step, but those cases still are similar enough to fit the pattern because (1) as with the grand jury cases, there was no public hearing into the circumstances of the killing; and (2) the cases reached the same “no prosecution” result.
81. Other highly publicized cases occurring at around the same time included the killings of Eric Garner (in New York), Tamir Rice (in Ohio), Walter Scott (in South Carolina), and Eric Harris (in Oklahoma). See Nicholas Quah & Laura E. Davis, Here’s a Timeline of Unarmed Black People Killed by Police over Past Year, BUZZFEED (May 1, 2015), http://www.buzzfeed.com/nicholasquah/heres-a-timeline-of-unarmed-black-men-killed-by-police-over#.oiV7XowlX.
82. See Susan Page, Whites and Blacks Question Police Accountability, USA TODAY, Aug. 25, 2014, http://www.usatoday.com/story/news/nation/2014/08/25/usa-today-pew-poll-police-tactics-military-equipment/14561633/.
83. See, e.g., Walter Katz, Commentary: Enhancing Accountability and Trust with Independent Investigations of Police Lethal Force, 128 HARV. L. REV. F. 235 (Apr. 10, 2015).
84. See, e.g., Melody Gutierrez, Calls Grow to Eliminate Grand Juries’ Secrecy in Police Killings, SFGATE (Dec. 14, 2014), http://www.sfgate.com/crime/article/Calls-grow-to-eliminate-grand-juries-secrecy-5956945.php.
85. See, e.g., Katz, supra note 83.
86. See, e.g., Gutierrez, supra note 84. This could be accomplished, even in jurisdictions where indictment by a grand jury is sometimes required, by prohibiting the use of a grand jury as the initial step in the charging process, and instead requiring that prosecutors either decide on their own not to bring charges or start the process with a complaint and a preliminary hearing in open court. See supra note 1.
* As many are aware, several months after the grand jury referred to in the title to this chapter voted not to indict Darren Wilson for killing Michael Brown, the U.S. Department of Justice [DOJ] issued a report on the federal criminal investigation regarding the same incident. See U.S. DEP’T OF JUSTICE, REPORT REGARDING THE CRIMINAL INVESTIGATION INTO THE SHOOTING DEATH OF MICHAEL BROWN BY FERGUSON, MISSOURI POLICE OFFICER DARREN WILSON (Mar. 4, 2015), http://www.justice.gov/sites/default/files/usao-mdpa/legacy/2015/03/18/DOJ%20Report%20on%20Shooting%20of%20Michael%20Brown.pdf. The report concluded, inter alia, that “Wilson’s actions [did] not constitute prosecutable violations under the federal criminal civil rights statute.” Id. at 5. Although the DOJ report shed important light on what happened between Wilson and Brown in the moments leading up to the shooting, it did not address the state criminal process in the Wilson case, which is what this chapter is about.
** I am grateful to Beverly Beimdiek, Kathleen Cash, Caterina DiTraglia, Ellen Goldwasser, Peter Joy and Ann Shields for their comments and suggestions; to Brian Hall, Nick Papadimitriou and Jenny Terrell for their research assistance; and to my dear friend Kimberly Norwood for inviting me to participate in this important project.