In 1987 lawyers for a Black inmate on Georgia’s death row presented the Supreme Court with evidence of the extreme racial disparity in the state’s administration of the death penalty. Their client, Warren McCleskey, had been convicted of killing a white police officer. In Georgia, Black murder defendants whose victims were white were almost twenty-two times as likely to be sentenced to death as those who killed other Blacks, and more than seven times as likely as whites whose victims were Black. Given those numbers, the lawyers argued, McCleskey’s execution would violate the Constitution’s guarantee of equal protection.

The justices accepted the validity of the statistical study but blinked at accepting its implications. The vote was 5–4. There was simply “no limiting principle” to a statistical argument of this sort, Justice Lewis Powell warned in the majority opinion. “If we accepted McCleskey’s claim that racial bias has impermissibly tainted the capital sentencing decision, we could soon be faced with similar claims as to other types of penalty.” Justice William Brennan’s dissenting opinion distilled the majority’s view into an acerbic observation: “Taken on its face, such a statement seems to suggest a fear of too much justice.”

That fear has scarcely dissipated in the ensuing thirty-six years. Justice Brennan’s pungent phrase could just as well apply, for example, to a troubling decision from the Supreme Court’s most recent term. Marcus Jones, the petitioner in Jones v. Hendrix, was convicted in 2000 of violating a federal law that prohibits several categories of people—including those in the country illegally and those, like Jones, with previous felony convictions—from owning a gun. Nineteen years later, in Rehaif v. United States, the Supreme Court interpreted the statute as requiring the government to prove that the defendant knew that he was in one of the prohibited categories. The government provided no such proof in Jones’s case. With less than a third of his twenty-seven-year sentence remaining, he filed a habeas corpus petition in federal court to have his conviction set aside.

But in a majority opinion by Justice Clarence Thomas, the Court ruled against Jones, arguing that because he had filed a habeas petition challenging his conviction years earlier, the new petition was “successive” and did not qualify for the extremely narrow exception to the statutory bar against successive petitions. The vote was 6–3. Justice Ketanji Brown Jackson’s dissenting opinion, in which she found “simply no justification” for the majority’s conclusion, echoed Brennan’s. “It is quite clear,” she wrote, “that the Court’s rulings in this area of the law reflect a general ethos that convicted prisoners should not be permitted to file” motions for federal court relief “at all.” Jones remains in federal prison for a crime of which he is, as a legal matter, innocent.

The Fear of Too Much Justice could serve as the title for any number of books about the administration of criminal justice in the United States today. Both of the books under review here seek to understand why the system serves so many so poorly. The one that actually bears the title, by Stephen B. Bright and James Kwak, examines the myriad ways in which the search for justice unravels once someone is charged with a crime, beginning with the nearly unlimited discretion accorded prosecutors to shape the case and exploit the advantages they have in resources and access to information. The other, Joanna Schwartz’s Shielded: How the Police Became Untouchable, explores why the police are so rarely called to account for misconduct despite the theoretical availability of tools to punish bad actors and to provide redress to their victims.

The narrative that emerges from these two books is of a long string of promises unfulfilled by indifferent politicians or broken by ambitious prosecutors and hostile judges. That there is little overlap between the two, with their separate focuses on criminal process and civilian redress, makes the picture all the more dispiriting: this is a system that disserves the innocent and the guilty alike.

Had Stephen Bright chosen to tell the stories in The Fear of Too Much Justice in the first person, he would have had an ample supply to choose from. He is a well-known and much-honored capital defense lawyer who has argued and won four death penalty cases at the Supreme Court. In fact, without identifying himself as a participant, he discusses one of his notable victories, a 2016 case that resulted in the overturning of a Georgia death sentence due to the prosecution’s removal of every Black potential juror from the murder trial of a Black defendant. The vote was 7–1, with Chief Justice John Roberts writing for the majority and Justice Thomas the only dissenter. Bright has also taught as a visiting lecturer at numerous law schools, including Yale, where his coauthor, James Kwak, a former law professor himself, was his student.


The racial gamesmanship that is common to jury selection is one of the many problems The Fear of Too Much Justice identifies. Given that 90 percent of all criminal cases are resolved by guilty pleas and therefore never go to trial, it may not even be the worst problem. But it is emblematic of a major theme of the book: the divergence between how the criminal justice system looks on paper and how it works in practice. Bright won the Georgia jury case on the basis of the Supreme Court’s ruling thirty years earlier in Batson v. Kentucky, which required prosecutors to explain their use of peremptory jury strikes if the defense lawyer objected that they were being deployed for reasons of race. Batson empowered the trial judge to disallow a strike for which the prosecution could not provide a plausible nonracial reason.

The decision was hailed at the time as an important breakthrough for racial justice in the legal system, but as Bright and Kwak explain, “the rule declared in Batson simply does not work.” Prosecutors are often specifically trained in how to satisfy judges with seemingly race-neutral reasons for removing Black potential jurors. A list provided at a Texas trial-skills course for prosecutors included such all-purpose reasons as “didn’t speak,” “very vocal,” “angry,” “expressionless,” and “no religious preference.” A California prosecutors’ manual lists seventy-seven such reasons.

And judges, especially elected judges, have proved unduly persuadable by prosecutors’ explanations. Bright and Kwak observe that “many elected judges may find it politically impossible to rule that a prosecutor intentionally discriminated and then asserted a false reason for the strike.” Bright’s client prevailed in the Supreme Court because the prosecution’s notes in the case, made available through Georgia’s Open Records Act, made the strategy for removing Black jurors indisputably clear—if not to the judges in Georgia, then at least to seven Supreme Court justices.

Another landmark Supreme Court decision, Brady v. Maryland (1963), has also fallen far short of its promise. Brady requires prosecutors to disclose, in advance of trial, evidence favorable to the defense. And yet the National Registry of Exonerations, in a shocking report issued in 2020,

found that failure to disclose exculpatory evidence played a role in the convictions of more than a thousand people who have since been exonerated—44 percent of all the exonerations from 1989 to 2019.

How can there be such a disparity between theory and practice? According to Bright and Kwak, the answer lies in “the awesome powers of the prosecution.” Although the phrase “adversary system” suggests a level playing field for the two sides in a criminal case, the system is fundamentally asymmetrical. Prosecutors, in addition to having greater access to information from law enforcement and more resources for investigating and consulting with experts, also make the charging decisions. They can decide between a misdemeanor and a felony, and they may “overcharge” to increase their plea-bargaining leverage.

No other country elects its prosecutors. In 2019 there were 2,442 elected prosecutors in the United States, 95 percent of whom were white and 73 percent of whom were white men. Of course, as Chesa Boudin learned in a 2022 recall vote less than three years after being elected San Francisco’s district attorney as an insurgent, prosecutors can lose their office when the public turns against them. Politics is never absent from the criminal justice equation.

It is decidedly not absent from the civil justice system either. Joanna Schwartz’s Shielded covers less familiar ground, yet its resonance with The Fear of Too Much Justice is striking. It is in part the story of how the legal doctrine of qualified immunity rose to subvert the promise of an old civil rights law that enables people to sue the police for violating their constitutional rights. Once again, rights on paper prove to be only that.

In its benign origins, qualified immunity simply meant that law enforcement officials shouldn’t be liable for mistakes made in good faith. But decades of Supreme Court decisions have converted a commonsense rule into a powerful doctrine that deprives people injured by police misconduct of recourse unless they can point to a court decision that on nearly identical facts found the police to have violated someone’s constitutional rights.

Shielded opens with a story that captures the doctrine’s perverse effects. Onree Norris, a seventy-eight-year-old Black man, was sitting quietly in his bedroom in Henry County, Georgia, when members of a county sheriff’s task force, using flash grenades and battering rams, burst in and pointed assault rifles at him. When he was slow to obey orders to get on the ground, they grabbed him and threw him down. It wasn’t until the agents had him outside his house in handcuffs that they realized they had meant to look at a different address for a suspected drug dealer, at which point “they turned their cameras off one by one.”


When the county brought no charges against the officers, Norris filed suit, seeking compensation for what he had suffered. Not only had the officers barged into the wrong house, causing substantial damage, but they lacked a warrant to enter any house. Clearly, their actions amounted to an unreasonable search, in violation of the Fourth Amendment. But no one who has followed the saga of qualified immunity would be surprised to learn that a judge threw the case out for lack of a precedent that precisely defined the misconduct about which Norris complained and that found, on precisely those facts, a constitutional violation. That is the essence of qualified immunity.

Schwartz is a law professor at UCLA who began her career at a small civil rights law firm in New York that specialized in suing the police. She expected that such cases would be hard to win, and they were, but what surprised her was how little effect the lawsuits had when they did succeed. After all, the assumption was that private suits for damages would, when successful, impart powerful lessons to the police, deterring future misconduct in a way that criminal cases, given their extreme rarity, never could. (Schwartz notes that police officers are criminally charged in less than 2 percent of fatal shootings and are convicted in less than one third of those few cases.)

But what if the deterrent effect proved so powerful as to threaten to render the police, plagued by damage awards and looking over their shoulders for the next lawsuit, unable to do their jobs? Schwartz maintains that while the concern that “public safety will be imperiled by too much oversight” has always accompanied the desire to hold the police accountable, it is now nearly an unquestioned assumption that lawsuits against the police exact too high a price. In other words, here, too, is a fear of too much justice.

But this assumption, Schwartz contends, is a myth that has distorted the civil justice system by persuading judges of the need to insulate the police from accountability. “Myths about the dangers of making it too easy to sue police have made a mess of our system,” she writes. Qualified immunity and the many other obstacles Schwartz describes doom most meritorious lawsuits. But even those that succeed have little impact, because cities typically indemnify their police forces; individual officers usually pay nothing out of pocket in the rare case that ends in a judgment against them, even when judges award punitive damages. Schwartz studied misconduct settlements and judgments in eighty-one jurisdictions over a period of six years. Plaintiffs received $735 million, of which individual officers paid just 0.02 percent, while governments paid the rest.

The decision to indemnify is rooted in politics: elected officials don’t want to seem antipolice. After the murder of George Floyd by Minneapolis police in 2020, Colorado enacted a law authorizing local governments to require their police officers to contribute up to $25,000 to a judgment if they were found to have acted in bad faith. One Denver suburb, Greenwood Village, then passed a resolution promising that it would never make a bad-faith finding. The mayor explained that the village did not want “our police officers to worry about us throwing them under the bus.” In some places, individual officers are not even told that they have been sued. Nothing about a suit for damages appears in the officer’s personnel file.

The Supreme Court is Schwartz’s primary focus. In her account, it is the Court that gave and the Court that took away. A landmark decision in the Warren Court’s criminal procedure and civil rights revolutions was Monroe v. Pape (1961), which interpreted Section 1983 of the Civil Rights Act of 1871 as authorizing individual lawsuits against law enforcement officers for violating constitutional rights. But the Court, beset by criticism that it had gone too far, soon began to close the door it had opened. In 1967 Chief Justice Earl Warren wrote the majority opinion in Pierson v. Ray that established a limited “good-faith” immunity for police officers who believed they were acting with legal authority. The good-faith defense gradually, maybe inevitably, morphed into a broader defense that regarded an officer’s subjective intent as irrelevant. Instead, the Court held in Harlow v. Fitzgerald (1982), officers were entitled to immunity as long as their actions did not violate “clearly established law.”

The shift away from intent proved much more important than it appeared at first. When intent was at issue, plaintiffs were entitled to discovery to try to show what the officer knew and believed. Once intent was irrelevant, judges could dismiss cases before discovery as long as the defense could show that there was no clearly established law on the books that applied to the situation. Subsequent Supreme Court decisions have raised the “clearly established” hurdle to mean not just a legal principle but a published court opinion in a case with virtually identical facts. “For more than five decades,” Schwartz reports,

the Supreme Court has repeatedly strengthened qualified immunity’s protections, describing each additional layer of defense in increasingly terrified tones as necessary to protect officers from the unyielding power of civil rights lawsuits.

The fear of too much justice, yet again.

There seems to be a growing appreciation in the legal world, on the right as well as the left, that qualified immunity has gone too far. In a separate opinion in a 2017 case, Justice Thomas called for reconsidering the doctrine, criticizing it as a judicial invention unrelated to the understanding of official immunity that existed in 1871, when Congress enacted Section 1983. The Institute for Justice, a libertarian organization that appears regularly before the Supreme Court, publishes a weekly online newsletter called Short Circuit that highlights particularly troubling qualified-immunity decisions in the lower federal courts. I have read the newsletter for years and am often left shaking my head at the contortions judges go through in order to avoid holding the police to account.

An example, picked almost at random from a recent newsletter, is a July 19 ruling from the federal appeals court in New York affirming a grant of qualified immunity for a New York police officer who, for no apparent reason, pointed a gun at a driver he had stopped for making an illegal lane change. The driver, who sued the officer for excessive force, had stopped his car and pulled over within seconds, providing no indication that he posed a threat. The appeals court ruled for the officer in Cerisier v. City of New York on the grounds that there was no “clearly established law” that defined the officer’s behavior as improper.

Cases directly challenging qualified immunity are regularly reaching the Court now. It’s possible to imagine that the justices might undertake to decide one in the near future, and that a cross-ideological coalition of justices could rewrite the immunity rules. Such a top-down reform in fact might be easier to achieve than some other proposals offered by the authors of both these books, which, though well considered, nonetheless depend on the willingness of state and local governments to spend the money necessary to address systemic problems.

For example, sixty years after Gideon v. Wainwright established the obligation to provide a lawyer for a defendant who can’t afford one, it is widely conceded that Gideon is a “promise broken,” as Bright and Kwak put it. Part of the issue is money, of course—Tennessee has a cap of $3,000 for the defense of a noncapital felony charge—but Bright and Kwak argue that what is really lacking is the motivation to construct a better system. “Many of the major players in the criminal legal system like things just the way they are,” they write. “Many programs for providing representation are dependent on judges, county commissions, governors, and others whose primary interest is the fast and inexpensive processing of cases.”

Recently ProPublica published an article about a reform effort in Mississippi, the uncertain fate of which completely validates Bright and Kwak’s assessment. This past April the state adopted a new rule for its courts, requiring lawyers to be appointed to represent indigent defendants during the crucial period between arrest and indictment. The rule is aimed at the typical situation in which the lawyer appointed for the defendant’s initial court appearance exits the case after that appearance, and another lawyer is not appointed until the defendant has been indicted. In Mississippi, this lawyerless gap can last years, during which the defendant may remain in jail, without resources or the ability to organize a defense or negotiate a plea bargain. When the new rule went into effect, courts in many of the state’s counties had made no arrangements for compliance. “There’s really not a plan,” one county judge told ProPublica.

I might earlier have regarded the Mississippi situation as an unfortunate one-off in a state hardly known for a robust commitment to criminal justice. But not having to think about the problems in the administration of both criminal and civil justice on a systemic level is a luxury that, having read these books, I no longer have. A line toward the end of one of them might have served so well for the other that I won’t identify which it comes from: “Because of the fear of too much justice, we have been willing to accept too little.”