This might be a surprise to some people, considering my chosen career and how I have been portrayed at times in the media, but I consider myself a civil libertarian. I am wary of governmental power and overreach. I believe citizens should never be subjected to harassment or unlawful invasions of privacy. I cherish the U.S. Constitution. In every job I’ve ever held in the public arena, from Marine Corps officer to police commissioner, I have sworn to “serve, protect and defend the Constitution of the United States of America against all enemies, foreign and domestic.” That, to me, is a sacred duty, and I have always fulfilled it to the absolute best of my ability. I don’t believe in cutting corners. Not in my personal life. Not professionally. And certainly not with the law. I am also a lawyer.
Which brings me, of course, to the policing tactic that is sometimes wrongly labeled “stop and frisk.”
One of my biggest frustrations as police commissioner was how my views on this subject were so badly misconstrued. Some of this was just a matter of public confusion. Phrases fly around. People pay half attention. Being “stopped and frisked” by a police officer certainly doesn’t sound like any fun. But much of this distortion, I came to believe, was purposeful. It occurred, I am convinced, for cynical, political reasons by politicians and activists who wanted to advance their own ambitions and didn’t care about the facts—or the life-or-death consequences to the public.
I could handle the public relations heat. My poll numbers stayed high. So did the NYPD’s. In January of 2013, the start of my twelfth year as police commissioner, as the so-called stop-and-frisk debate was really heating up, the department had a 70 percent approval rating, according to the widely respected Quinnipiac University Polling Institute. My personal approval rating was even higher at 75 percent. “Whether Democratic or Republican, white or nonwhite, male or female, wealthy or not, New York City voters overwhelmingly like Police Commissioner Raymond W. Kelly and would prefer that he remain in the job under the next mayor,” wrote New York Times reporter Wendy Ruderman in her piece on the Quinnipiac poll. Those numbers, in a city as complex and contentious as New York and after eleven-plus years on the job, were extremely gratifying. What bothered me—what still bothers me—is that the stop-and-frisk controversy managed to undermine a valuable, appropriate, and legal—let me emphasize legal—tool of modern law enforcement, one that had helped to save literally thousands of innocent lives.
You want the police to stop and question people who are behaving suspiciously on the street. That’s what we pay them for. Even the politicians who complain most loudly about “stop and frisk” want it. You’ll notice they almost never demand that we quit stopping people behaving suspiciously. They call for “a more appropriate number of stops” without ever detailing how many that might be. The practice must be event driven, not numbers driven. To anyone concerned about crime today, ending these street stops would make no sense at all.
The reason is obvious: If a police officer stops someone acting suspiciously, someone whose behavior suggests the person is planning to commit a crime, we will have prevented that crime. There will be one less crime victim as a result. It is really no more complicated than that—at least, it shouldn’t be.
Now, what do I mean by acting suspiciously? Here is a classic example, one that actually plays out every day in New York City and in other cities and towns everywhere. A person is walking up and down the street and attempting to open car doors, seeing if any of them is unlocked. Now, it is theoretically possible that this person owns every car on the street. It’s possible he is an undercover block-patrol volunteer out on his daily rounds. I suppose that has happened once or twice in the history of the universe. But it is extremely unlikely. It is far more likely that the individual is looking for a car to steal or an unlocked vehicle with something valuable inside. At the NYPD, we would call this suspicious behavior, the same phrase the U.S. Supreme Court has used in decisions resoundingly approving of the practice. Under our guidelines, the officer should stop that person and ask a few questions. If there is some reason to believe the person may be armed or otherwise dangerous, a light pat-down is also permitted for safety’s sake.
Keep in mind, trying car door handles is not illegal, even on cars that don’t belong to you. You won’t be arrested for that alone. But it is suspicious behavior, warranting an additional inquiry from the police. It’s exactly the kind of activity an engaged police officer should confront.
Respectfully. Professionally. Firmly.
If the individual has no gun and no drugs and offers no other evidence of committing a crime, there is no basis for arrest. That person will be sent promptly on his way. But aren’t you glad the police officer didn’t stand by and do nothing? I think most people are.
This is what New York City public advocate Bill de Blasio, the New York Civil Liberties Union, the Center for Constitutional Rights, and various other litigators and critics couldn’t or wouldn’t recognize: They insisted that in cases like this, perfectly innocent people were stopped and harassed for no good reason. I believe differently. I believe the police just prevented a crime. Lawfully. Appropriately. Doing exactly what we should want them to do.
I don’t like to question anyone’s motives. I believe most people try to do what they think is right. But these street stops have been so profoundly mischaracterized, I also believe there has to be something political behind the attacks.
* * *
The concept behind these street stops goes all the way back to English common law, the very basis of the American legal system. Back then, it was the constable who would see something suspicious and could intervene. In our country, the U.S. Supreme Court has validated the practice repeatedly, most prominently in the landmark 1968 case Terry v. Ohio, and it is enshrined in state and federal law. John W. Terry and two friends were pacing back and forth outside a downtown jewelry store as if “casing a job, a stick-up,” when a veteran Cleveland detective confronted them. Asked to identify themselves, the men “mumbled something.” In a brief pat-down, the detective felt a pistol in Terry’s overcoat. Once the case got to court, Terry’s lawyer argued the gun should be suppressed as the product of an illegal search. The court’s eight-to-one majority, with only Justice William Douglas dissenting, was clear and unequivocal: The Fourth Amendment to the United States Constitution, which prohibits unreasonable search and seizure, is not violated when a police officer briefly stops and questions someone—so long as the officer has a “reasonable suspicion” that the person has committed, is committing, or is about to commit a crime. The pat-down during such a stop must be limited—not a full-on body search but a quick check of the outer clothing to be sure the individual isn’t carrying a weapon and everyone is safe. Terry v. Ohio is why in many states these are called “Terry stops.” In some way, shape, or form, stop-question-and-frisk is authorized in every state in the country and used by virtually every police department. It’s what cops do when they are doing their job.
Let me clear up a couple of other facts about street stops. At the NYPD, there is no street-stop unit or street-stop program or Operation Street Stop. There never has been. Briefly stopping certain people, questioning them, and patting them down when appropriate is something all officers are expected to do. It is what proactive policing is all about, getting cops to get out of their cars, engage with the public, and be police. It is as fundamental a part of the job as knowing how to pull over a speeding vehicle or how to place a pair of handcuffs around a suspect’s wrists. In my time as commissioner—and in other administrations, I presume—there were never any quotas or specific numerical targets for street stops, despite occasional claims to the contrary from the unions. The officers were simply expected to stay active and do the job they were hired to—all parts of the job, including engaging lawfully with people acting suspiciously on the street.
There is no need to exaggerate the importance of this one tactic. No one in policing considers it the be-all-and-end-all of crime fighting. But in conjunction with a variety of other methods and strategies, it has helped to drive crime down in New York City and to make the streets safer for everyone. It’s hard to prove a negative, so we may never know for certain how many weapons weren’t carried, how many muggings didn’t occur, how many shootings or burglaries were avoided, how many people weren’t killed because of those street stops. But we know this: In the twelve years of the Bloomberg administration, the city had nearly 9,500 fewer murders than in the twelve previous years, and the vast majority of the lives saved were those of young minority men, if history is any guide. Street stops certainly weren’t the only reason for the steep decline, but they just as certainly played a role. In the final year of the Bloomberg administration murders and shootings were at record lows.
This didn’t mean we wouldn’t get sued, of course. In this litigious era, the city of New York constantly gets sued, the police department especially. The first big legal challenge to the street stops came in 1999, nearly three years before I returned as commissioner. That case, Daniels v. City of New York, was brought by one of the groups that sues the city often, the Center for Constitutional Rights. The matter was assigned to U.S. district judge Shira Scheindlin, who was nominated by President Bill Clinton to fill a slot at Manhattan’s federal court that was vacated when Louis Freeh left to become FBI director. Lawyers for the group argued that under the Giuliani administration, the department’s street stops amounted to illegal racial profiling, since the majority of the people stopped were either African American or Latino, a far higher percentage than in the population at large. They also complained that the NYPD kept inadequate records of these stops.
The suit came at a dicey moment for the Giuliani administration, right at the peak of protests over the fatal police shooting of Amadou Diallo, the unarmed African immigrant street peddler shot forty-one times outside his apartment in the Soundview section of the Bronx. The lawyers sought to channel the Diallo outrage into their own big victory in court, but that never happened. The Daniels case didn’t ever go to trial. It inched along on Judge Scheindlin’s lengthy calendar as the Giuliani administration ended and the Bloomberg administration came in. The city law department, led by corporation counsel Michael Cardozo, settled the case at the end of 2003, promising that the NYPD would keep clearer records and make sure people weren’t being stopped based on their race. The city also agreed to conduct regular audits of the street stops, reporting how closely the law was being followed.
We lived under that agreement for the next several years, following the procedures as we had promised to. People were stopped because of suspicious behavior—not based on race, ethnicity, national origin, or any other similar factors. We made sure the stops were being recorded. This was certainly not a practice that had been adhered to in the past. After every stop, the officers were required to fill out a UF-250, the Stop, Question and Frisk Report Worksheet, noting why the stop was made. The revised form was a product of the Daniels case with input from all sides, including the judge. The information was input into department computers after every shift.
But no legal fight is ever really over in New York. So it was with the Daniels case. The plaintiffs asked Judge Scheindlin to extend the settlement period beyond its 2008 expiration date. She declined but came up with a novel way of keeping the issue in her court. She invited the plaintiffs to bring another lawsuit and mark it as related to Daniels. That way, she would get to hear that one too.
I didn’t realize such a thing was allowed. I thought cases were supposed to be assigned to judges at random. But that’s what Judge Scheindlin did.
Suddenly, the Center for Constitutional Rights was back with a whole new set of plaintiffs and the same old set of arguments against the NYPD street stops. They put their complaint again in starkly racial terms and purported to speak for all minority New Yorkers. The new case was called Floyd v. City of New York. This time, I was among the defendants, along with Mayor Bloomberg. The plaintiffs were several black and Latino people, including an African American man named David Floyd, who claimed he had been stopped and searched without any reasonable suspicion that he was committing, had committed, or was about to commit a crime.
On February 27, 2008, Floyd and another man were outside a building in the Bronx, trying to unlock a basement apartment. As they tried one key after another, three police officers walked up, asked the two men what they were doing, asked to see identification, and conducted a light pat-down.
The officers testified later that they thought the two men were committing a burglary, that there had been a pattern of burglaries in the area at that time of day, and that Floyd’s behavior made them suspicious. They had filled out the appropriate UF-250, noting that the suspected crime was burglary.
In a preliminary ruling on August 31, 2011, Judge Scheindlin found that the police officers did, in fact, have reason to be suspicious and therefore the stop was legal, but that wasn’t the end of the case. She said the plaintiffs’ class-action suit could continue anyway. “There is a triable issue of fact as to whether the NYPD leadership has been deliberately indifferent to the need to train, monitor, supervise, and discipline its officers adequately in order to prevent a widespread pattern of suspicionless and race-based stops.”
That was ridiculous. Our street stops weren’t race-based. They were based on the presence of suspicious activity that was closely connected to crime. If there were more stops in some neighborhoods than others—and this was always the case—that was because those neighborhoods had more criminal activity.
* * *
The number of stops had been growing. At least, the numbers being counted had, though it was hard to say how much was the result of more activity and how much was the result of better record keeping.
In 2002, the first year of the Bloomberg administration, the police department recorded 97,296 street stops, though in those pre-Daniels days, certainly many stops went unrecorded. When I was making street stops as a young patrol officer, we only filled out a form if there was a problem or a complaint. Oversight and review were far less stringent in those days.
In 2008, the year Floyd was filed, the street-stop number crossed half a million. The peak year was 2011, with 685,724 stops. Overall, in about 12 percent of the cases, people were charged with some offense or crime. In other cases, people were sent on their way. Weapons were found in about 2 percent of the stops. You can’t measure the amount of crime that wasn’t committed.
But here’s some context: the NYPD has twenty-three million citizen contacts a year, twelve million calls to the 911 system, four hundred thousand arrests, five hundred thousand summonses. Is 685,274 street stops a big number in a city with record-high population of 8.4 million people? Compared to everything else the department does, not really. It’s more like business as usual. Here’s what that number adds up to: less than one stop per patrol officer per week, and less than one limited pat-down every two weeks. With averages like those, it’s hard to call the practice an egregious overuse.
On a per capita basis, the NYPD wasn’t stopping people dramatically more often than other big-city police departments. When you normed for population, the number of stops was actually lower than in Philadelphia or Baltimore. Chicago’s stop-question-and-frisk rate was four times New York City’s. If you factored in the relative size of the cities, the Los Angeles police stopped the equivalent of 560,000 a year in New York City numbers.
But that didn’t mean someone couldn’t turn this into a political issue, even beyond the dispute in federal court. Soon enough, someone did.
* * *
In early June 2012, a group of black and Latino lawmakers met with attorneys at the Justice Department in Washington, complaining about stop-and-frisk. A big protest was planned by Al Sharpton’s National Action Network for Father’s Day, June 17, in which several thousand people, including NAACP president Benjamin Todd Jealous and hip-hop executive Kevin Liles were expected to march down Fifth Avenue from Harlem to the mayor’s town house on East Seventy-Ninth Street. The mayor wanted to get in front of that. So the Sunday before, he rode out to Brownsville, Brooklyn, to address the congregation at the First Baptist Full Gospel Church. I went along, as did schools chancellor Dennis Walcott and two deputy mayors, Howard Wolfson and Caswell Holloway. That was a large entourage for a church drop-in.
Though far safer than it had been a decade earlier, Brownsville still had some of the highest violent-crime numbers in the city—and therefore also one of the highest numbers of street stops. It was a perfect neighborhood for a review of our crime-fighting techniques, their effectiveness, and their fairness.
Bloomberg was eloquent.
“We are not going to walk away from a strategy that we know saves lives,” he said from the pulpit that Sunday as the people listened in rapt attention. “At the same time, we owe it to New Yorkers to ensure that stops are properly conducted and carried out in a respectful way.”
Racial profiling was banned at the New York City Police Department, he assured the congregation. “We will not tolerate it.” But any effort to fight crime, he said, must go where the criminals are—regardless of where that is. It made no sense, he said, to insist that the people the police encounter precisely mirror the population at large. Women make up slightly more than half the city’s population, the mayor pointed out. Does that mean half the people stopped for suspicious behavior have to be women? That made no sense at all. “If we stopped people based on census numbers, we would stop many fewer criminals, recover many fewer weapons, and allow many more violent crimes to take place,” he told the congregation.
“We will not do that,” he vowed. “We will not bury our heads in the sand. By making it ‘too hot to carry,’ the NYPD is preventing guns from being carried on our streets. That is our real goal—preventing violence before it occurs, not responding to the victims after the fact.”
I couldn’t have said it better myself.
* * *
Judge Scheindlin conducted a two-month bench trial—no jury—in the spring of 2013, already into our final year in office. From the questions she asked, the evidence she admitted, and that which she kept out, it was very clear to me that she was going to rule against the city, no matter how the evidence came in.
She took extensive testimony from a professor at Columbia Law School, Jeffrey Fagan, purporting to show that the stops were flat-out racist. Nonsensically, his criteria for determining whether racial profiling was happening revolved around census data. Taking his argument to its logical conclusion, half of the stops would be of females. His position involved something called “indirect racial profiling,” which I am still not sure anyone understands.
He and his research assistants had reviewed something like four million police stops over the years. They contended that about 6 percent of those—just 6 percent—were based on questionable circumstances. In other words, 94 percent of the police stops met Constitutional muster. For the case, they chose nineteen examples, presumably the most egregious they could find, ten of which were ultimately found to have been acceptable under the Constitution and the law.
The lawyers for the city countered that the police department had been living fully within the limits established by Terry v. Ohio, the Daniels settlement, and state and federal law, calling the street stops perfectly legal and routine—not remotely racial profiling.
Some of the points I wanted to see pushed were never brought to the fore, and others were kept out of evidence by Judge Scheindlin. Was it really plausible that the most diverse police department on earth, with officers hailing from 106 different countries and representing every imaginable race, would engage in a massive conspiracy to conduct street stops to deny minorities their Constitutional rights? How would such a policy be disseminated? How would it be enforced? How could it possibly be kept secret? In fact, we had done exactly what we said we were doing. We went where the crime was, whatever color the perpetrators turned out to be.
At the police-officer rank, the department was already “majority minority.” The recent academy classes were solidly so. The number of African Americans had dipped slightly, but so had the percentage of black residents citywide. The numbers of Latinos, Asians, and other nonwhites in the department were way up. This was an extraordinarily diverse police force, serving an extraordinarily diverse city—and proud of it. Was this department really engaged in wholesale racial profiling, making street-level decisions based on factors other than the demands of fighting crime?
I thought it was important to bring some objective measurement to the issue. Otherwise, the debate would be an endless exchange of conflicting slogans and little else. Without reliable data, who could ever know: Were the NYPD street stops built on prejudice, profiling, and preconception—or simply the realities of confronting actual street crime?
I asked the New York City Police Foundation to hire the RAND Corporation to study this in 2006. The consultants’ first order of business: finding the right metric to compare the stops against. What’s the proper benchmark for determining if police stops are fair?
Again it was clear why census data wasn’t the right comparison. With slightly more than half the population being female and the vast majority of street crime being committed by young men, no one could expect crime and criminals to mirror the population at large. The RAND researchers rejected arrest reports as well. The department controls those statistics by deciding whom to arrest. Arrest numbers could be as biased as the street stops.
So what did that leave?
In their report, “Analysis of Racial Disparities in the New York City Police Department’s Stop, Question, and Frisk Practices,” the researchers laid out a very sensible idea. They said we should use the descriptions of suspects provided to police by the victims of violent crime. More than any other measure, that should mirror the population that is actually committing crimes. The government doesn’t create those numbers. The police don’t craft those descriptions. Crime victims do. According to that data, the perpetrator descriptions over time in New York City are about 69 percent black. On average, 53 percent of the people stopped by the NYPD have been black. By that data-driven analysis, we were proportionally understopping African Americans—not overstopping them. To quote the RAND report: “We found that black pedestrians were stopped at a rate that is 20 to 30 percent lower than their representation in crime-suspect descriptions.” The judge would not admit the report into evidence.