clock menu more-arrow no yes

What the Supreme Court got wrong about homicides committed by cops

Rogue officers like Derek Chauvin probably won’t be deterred by good law, but excessively vague law encourages bad behavior.

A picture of George Floyd on cyclone fencing outside municipal buildings.
Law enforcement stands guard as crews remove artwork from temporary fencing outside the Hennepin County Government Center on April 2, 2021, in Minneapolis, Minnesota. Demonstrations have been ongoing outside the Government Center as the trial of former Minneapolis police officer Derek Chauvin, who is charged with multiple counts of murder in the death of George Floyd, continues inside.
Stephen Maturen/Getty Images

On Wednesday, more than one week into the murder trial of former Minneapolis police officer Derek Chauvin for the killing of George Floyd, Chauvin’s lawyer read an excerpt from the department’s manual governing the use of force.

“The ‘reasonableness’ of a particular use of force,” the manual stated, “must be judged from the perspective of the reasonable officer on the scene, rather than with the 20/20 vision of hindsight.”

Minneapolis revised its manual after Floyd’s death to place clearer and tighter constraints on officers engaged in the use of force. But the vague rule laid out in the version of the manual that was in effect during Floyd’s fatal encounter with Chauvin is fairly typical of the guidance provided to officers in the field.

As Sgt. Jody Stiger, a member of the Los Angeles Police Department called by prosecutors in the Chauvin trial, testified, most police departments derive their policies governing the use of force from Graham v. Connor. Graham is a 1989 Supreme Court case that, in the words of scholars Osagie Obasogie and Zachary Newman, “established the modern constitutional landscape for police excessive force claims.”

The language Chauvin’s lawyer read from the police manual was lifted, word for word, from the Court’s decision in Graham.

Authored by Chief Justice William Rehnquist, one of the primary proponents of a tough-on-crime approach that often animated the Court’s decisions during his tenure, the Graham opinion warns that police accused of using excessive force often have to make difficult decisions in highly stressful situations. In determining whether an officer acted reasonably, Rehnquist wrote for his Court, “the calculus of reasonableness must embody allowance for the fact that police officers are often forced to make split-second judgments — in circumstances that are tense, uncertain, and rapidly evolving — about the amount of force that is necessary in a particular situation.”

Perhaps even more significantly, Graham left cops with little guidance on just what limits the Constitution places on use of force by police. As then-University of Virginia law professor William Stuntz wrote six years after Graham was handed down, “one searches in vain for any body of case law that gives” Graham’s vague reasonableness standard “some content.”

Yet, while some academics did criticize Graham’s approach early on, many prominent commentators outside of the academy only recently have started to think of Graham as a major wrong turn by the Supreme Court. Though three justices joined a partial dissent by Justice Harry Blackmun that criticized some parts of Rehnquist’s decision, all nine justices agreed with most of Rehnquist’s reasoning. That includes Justice Thurgood Marshall, the legendary civil rights lawyer.

But with the benefit of hindsight — and with the benefit of empirical evidence showing that clear legal rules lead to better policing — Graham now looks like a serious error by the Court. As Rachel Harmon, a law professor at the University of Virginia and author of The Law of the Police, told me in an email, “Graham offers a standard focused on judging the use of force after it has happened,” and it “offers very little guidance to officers and departments about how to use force.”

It does little, in other words, to advise police on how they can avoid conduct that might needlessly injure or kill a criminal suspect.

It’s unlikely that clearer rules would have saved George Floyd’s life. As Minneapolis Police Chief Medaria Arradondo testified at Chauvin’s trial, Chauvin “absolutely” violated department policy when he knelt on Floyd’s neck after Floyd was already subdued and handcuffed.

But clear rules can ensure that cops tossed into a dangerous and uncertain situation can fall back on those rules, rather than making a potentially deadly decision with only their fear to guide them. As law professors Brandon Garrett and Seth Stoughton wrote in a 2017 article, Graham’s “‘split-second’ approach presents obvious problems from the perspective of law enforcement supervisors, who cannot provide meaningful guidance about or oversight of how officers react in the moment in an objectively reasonable way.”

Graham was correct about one thing. Officers do sometimes find themselves in “tense, uncertain, and rapidly evolving” encounters where they have to make quick decisions about how to use force. But if we want these officers to make the right decision in these fraught moments, police departments need to provide them with clear guidance on how they should react.

And the Supreme Court’s vague “reasonableness” standard does nothing of the sort.

How clear rules can save lives

On a fall night in 1974, Officer Elton Hymon arrived at the scene of an alleged home break-in. He soon found Edward Garner, an eighth-grade boy weighing about 110 pounds, in the backyard of the home. Hymon later admitted that he was “reasonably sure” that Garner was unarmed. Yet, as Garner attempted to climb a fence at the edge of the yard, Hymon shot him in the back of the head and killed him.

Police later found a stolen purse and $10 in Garner’s possession.

The stunning thing about Garner’s death, which formed the basis of the Supreme Court’s decision in Tennessee v. Garner (1985), is that Officer Hymon had every reason to believe that he acted lawfully when he killed an unarmed 15-year-old boy who’d committed a fairly minor act of theft.

A Tennessee state law provided that, after an officer notifies a suspect of their intention to arrest the suspect, if “he either flee or forcibly resist, the officer may use all the necessary means to effect the arrest.” In other words, state law clearly permitted police to use deadly force against fleeing felony suspects.

Nor was Tennessee particularly unusual in this regard. As Justice Sandra Day O’Connor noted in her dissenting opinion in Garner, in 1985 “nearly half the States” still followed a “venerable common law rule authorizing the use of deadly force if necessary to apprehend a fleeing felon.” As a 1736 treatise described that common law rule, “it is no felony” for a law enforcement officer to slay a suspect who “shall either resist or fly before they are apprehended.”

Garner, which abandoned that common law rule in a 6-3 decision, represents a “high-water mark” in the Court’s decisions governing use of force by police, according to Garrett and Stoughton. Unlike future decisions like Graham, Garner laid down a fairly clear rule that police could follow when determining whether to use deadly force against a fleeing suspect.

Where the officer has probable cause to believe that the suspect poses a threat of serious physical harm, either to the officer or to others, it is not constitutionally unreasonable to prevent escape by using deadly force. Thus, if the suspect threatens the officer with a weapon or there is probable cause to believe that he has committed a crime involving the infliction or threatened infliction of serious physical harm, deadly force may be used if necessary to prevent escape, and if, where feasible, some warning has been given.

Under Garner, in other words, police would no longer use their own judgment to decide whether to fire on a fleeing suspect. The Court told police when they could use deadly force — if the suspect “poses a threat of serious physical harm,” if they “threaten[] the officer with a weapon,” or when the suspect “committed a crime involving the infliction or threatened infliction of serious physical harm” — and thus informed police that they could not use deadly force against other fleeing suspects.

The impact of Garner on police behavior was swift and dramatic. According to a 1994 study by criminologist Abraham Tennenbaum, homicides committed by police dropped about 16 percent in the nation as a whole after Garner was decided. In states that previously followed the unconstitutional common law rule, “the reduction was approximately twenty-four percent (23.80%).”

A more recent appeals court decision bolsters the proposition that clear legal rules are effective in reducing police violence.

In Estate of Armstrong v. Village of Pinehurst (2016), the United States Court of Appeals for the Fourth Circuit heard an allegation that police used excessive force when they repeatedly used a taser to subdue a mentally ill man, who died during his encounter with the police. Though the Fourth Circuit ruled in favor of the cops, on the theory that the officers were protected under a doctrine known as “qualified immunity,” the court also laid down several limits on the use of tasers by police.

“A police officer may only use serious injurious force, like a taser, when an objectively reasonable officer would conclude that the circumstances present a risk of immediate danger that could be mitigated by the use of force,” Judge Stephanie Thacker wrote for her court. She added that “‘physical resistance’ is not synonymous with ‘risk of immediate danger.’”

The Fourth Circuit oversees federal litigation in Maryland, Virginia, West Virginia, North Carolina, and South Carolina, and a 2017 report by Reuters found that eight major cities in those states adopted stricter policies governing the use of tasers by police in the immediate wake of the Armstrong decision. These policies proved very successful in reducing the use of tasers.

In Baltimore, police used Tasers 47 percent fewer times last year than in 2015, according to records reviewed by Reuters. Deployments fell 65 percent in Virginia Beach; 60 percent in Greensboro, North Carolina; 55 percent in Charleston, South Carolina; and 52 percent in Huntington, West Virginia. Norfolk, Virginia, saw deployments plunge 95 percent.

As Professor Harmon told me, cases like Garner and Armstrong demonstrate that “when courts provide clearer guidance, it can make a difference.” Regarding the Armstrong case, Harmon told me that she “would want to know more about what officers used instead of tasers before throwing a victory parade, but it does illustrate the power of the law, when courts actually provide specific and meaningful guidance to the police.”

The Supreme Court moved away from giving clear guidance to police after Garner

The facts of Graham v. Connor are as shocking as the facts are in Garner, even though they did not result in anyone’s death.

Dethorne Graham was a Black man and a diabetic living in Charlotte, North Carolina, in 1984, when he felt the beginning of an insulin reaction. Because such a reaction is treated with sugar, Graham asked a friend to drive him to a convenience store so he could buy some orange juice. But when they arrived at the store, there was a long line. Fearing he would not be able to buy the juice fast enough, Graham immediately left and asked his friend to take him to a friend’s house instead.

A police officer witnessed Graham’s very brief visit to the store and deemed it suspicious, because the cop pulled Graham and his friend over and would not let the two men go even after Graham’s friend explained Graham’s medical condition to the cop.

At one point, while Graham was waiting for the officer to let him go, he got out of the car, ran around it twice, and then passed out on the curb. Erratic behavior can be a symptom of a diabetic emergency, but the police apparently took Graham’s behavior as a sign of something sinister. After more officers arrived on the scene, Graham was handcuffed and forced face-down onto the car’s hood. When Graham told the police to check his wallet for a decal indicating that he is diabetic, an officer told him to “shut up.”

They eventually let him go after they received a report that Graham hadn’t done anything wrong at the convenience store.

And yet, despite these disturbing facts, the Supreme Court’s decision emphasized that police must deal with “tense, uncertain, and rapidly evolving” situations when they encounter someone like Dethorne Graham.

Graham didn’t say that there are no limits on police conduct. In addition to holding that police must behave as a “reasonable officer” would behave, the Court also listed several factors that lower courts could consider when an officer is accused of excessive force, “including the severity of the crime at issue, whether the suspect poses an immediate threat to the safety of the officers or others, and whether he is actively resisting arrest or attempting to evade arrest by flight.”

But these were simply factors that could be considered, not bright-line rules that gave clear guidance to police about what kind of conduct is permitted. And the Graham case itself suggests that these factors offer little protection for many victims of excessive force.

After all, Graham himself committed no crime. He posed no threat to anyone, and he neither resisted arrest nor attempted to flee. But the Supreme Court sent his case back down to a trial court for a second hearing, and Graham ultimately lost his case.

One possible explanation for the lopsided vote in the Graham case — again, much of the decision was unanimous — is that the Supreme Court hands down decisions that are intended to be read and applied by lawyers and judges, not by police officers.

Despite Graham’s admonition that judges should evaluate an officer’s conduct without “the 20/20 vision of hindsight,” courts are in the business of hindsight. Lawsuits, by their very nature, do not arise until after an alleged legal violation has occurred. So, when an officer is hauled into Court due to allegations of excessive force, Graham reminds judges that they will probably know more about the circumstances that led to that allegation than the officer reasonably could have known at the time.

Yet, while Graham’s holding may offer a useful reminder to judges, we also know that police departments use decisions like Graham to shape their own policies and training manuals. And the sort of open-ended legal standards that judges are accustomed to applying to individual cases do not provide adequate guidance to police officers. A vague standard may be useful for a judge with a law degree, years of legal experience, and months to study the facts of a particular case. But such standards are inadequate for a cop who, often for the first and only time in their career, is caught in a dangerous situation with their gun drawn.

Nevertheless, since Graham, the Court has only doubled down on its preference for vague, flexible standards over clear legal rules governing police. In Scott v. Harris (2007), for example, the Court ruled in favor of police officers who, during a high-speed chase, rammed a suspect’s car off the road and caused him serious injury.

Yet, rather than evaluating this case under the fairly clear rule laid out in GarnerGarner, after all, was a case about when police can use potentially deadly force against a fleeing suspect — Scott arguably abandoned Garner’s approach altogether. While the fleeing motorist’s “attempt to craft an easy-to-apply legal test in the Fourth Amendment context is admirable,” Justice Antonin Scalia wrote for the Court, “in the end we must still slosh our way through the factbound morass of ‘reasonableness.’”

“Whether or not Scott’s actions constituted application of ‘deadly force,’” Scalia added, “all that matters is whether Scott’s actions were reasonable.”

As one federal judge wrote just a few months after Scott was decided, under the Scott decision, “there is no Garner bright-line test.” There is only a vague “reasonableness” test.

One major problem with this approach is that it gives virtually no guidance to police departments when they draft their own policies guiding the use of force, and it can lead individual officers to guess what kind of behavior is acceptable if they are in a situation that might require force. As Harmon, the UVA professor, writes, the Supreme Court’s current framework “does not answer adequately the most basic questions about police uses of force: when a police officer may use force against a citizen, how much force he may use, and what kinds of force are permissible.”

Again, it’s unlikely that a more rules-based approach, like the one the Court took in Garner, could have saved George Floyd’s life. Chauvin appears to have shown such extraordinary disregard for his department’s policies that even his own police chief testified against him at his murder trial.

But clear rules can and do save lives. According to Tennenbaum’s study of Garner, that decision “reduced the total number of police homicides by approximately sixty homicides a year.”

That’s 60 people a year who would have died if the Court hadn’t given clear guidance to police officers.

Sign up for the newsletter Sign up for The Weeds

Get our essential policy newsletter delivered Fridays.