Talking to Strangers: What We Should Know About the People We Don’t Know

by Malcom Gladwell

Seven years ago this past weekend, on July 17, 2014, a Staten Island man named Eric Garner was killed by police in a gruesome scene that went viral and helped launch the Black Lives Matter movement. Press reports usually say Garner was stopped on suspicion of selling cigarettes by plainclothes officers who then choked him to death, but the story I wrote about in I Can’t Breathe was both stupider and more tragic than that. Garner’s death was a confluence of a hundred terrible developments, but above all a grotesque governance failure. It was a classic example of how even the most harmless-sounding ideas can, in the hands of the wrong people, become deadly policy.

Garner’s death was accelerated by policing strategies based on the “Broken Windows” theory. Often attributed to famed Stanford researcher Philip Zimbardo, the theory’s origins really go back to 1963, when criminologist George Kelling took a job running a home for troubled youth in Lino Lakes, Minnesota. Before Kelling’s arrival, Freud-inspired clinicians at the 64-bed facility stressed observing rather than correcting the emotionally disturbed minors in their care. If a resident broke a light bulb, for instance, they would leave broken glass on the floor and just keep taking notes.

Kelling, a former parole officer, ordered staff to clean up the glass. After this and some other changes, violent incidents in the facility declined. He made the same observation most parents understand implicitly, that turning visual noise down and setting clear boundaries lowers anxiety and discourages acting out.

Nearly twenty years later, Kelling and James Q. Wilson co-authored an influential article in the Atlantic called “Broken Windows,” whose central argument was far more ambitious. Kelling and Wilson believed allowing visible signs of disorder in public invited crime. Reformers from there began encouraging a shift in emphasis from reactive policing of criminal violations to affirmative promotion of the more nebulous concept of “order,” which at first meant tackling graffiti, public drunkenness, jaywalking, and, yes, broken glass.

By 2014, police had begun to define a poorly dressed, 350-pound black man like Garner standing on a street corner as a species of visible public “disorder.” Kelling in 2015 told me he was aware as far back as 1982 that this might happen. He’d spent time with cops in South Boston, whose “idea of ‘maintaining order’ was keeping the black people out,’” he said. “So I knew that was a potential problem.”

Kelling is mentioned in Talking to Strangers, a carefully provocative 2019 book on policing by pop-wisdom king Malcom Gladwell, which I read on the anniversary of Garner’s death. It begins by recounting the infamous July, 2015 encounter between Texas traffic officer Brian Encinia and an African-American woman named Sandra Bland. Stopped for the preposterous reason that she’d failed to signal before changing lanes to accommodate the accelerating Encinia, Bland ended up being jailed after the traffic stop turned hostile. Three days later, she killed herself in custody in an incident that may have been the most disturbing of all the police misconduct cases that galvanized America during those last years of the Obama presidency.

Gladwell, who couldn’t have known he was releasing a book a year before the death of George Floyd would once again make police brutality the defining issue in American society, referred to the time between the summers of 2014 and 2015 as a “strange interlude.” Even just a few years ago, it seemed strange when America actually paid close attention to police abuse cases. Gladwell notes that the period that began with the the deaths of people like Garner and Michael Brown and ended roughly with the suicide of Bland was “when a civil rights movement, Black Lives Matter, was born.”

Of course, “we put aside these controversies after a decent interval and moved on to other things.” In the introduction Gladwell announces, “I don’t want to move on to other things,” and frames Talking to Strangers as “an attempt to understand what really happened by the side of the highway that day in rural Texas.”

The English-Canadian Gladwell may be the most bankable writer in the American publishing market. The #1 spot on the New York Times bestseller list is as much his home as the Super Bowl is for Tom Brady. There’s an intellectual drive-thru quality to his approach, which takes an idea and draws it out in bite-size chapters built around familiar pop culture episodes. He does this again in Talking to Strangers, a book about police brutality that somehow contains chapters about Amanda Knox, Bernie Madoff, Jerry Sandusky, and the TV show Friends. Gladwell makes it capital-E Easy for the medicine of thought to go down, a talent I once grumbled at, almost surely out of jealousy. I now see it’s a blessing in the United States, a country where a fair portion of the mass audience is capable of losing at tic-tac-toe.

An attempt to take on grim problems of race and violence, Talking to Strangers has more of an edge to it than Blink, “the power of thinking without thinking,” or Outliers, “the story of success.” This book about all the different ways in which strangers misunderstand one another feels like it was written as a way to nudge an increasingly polarized country to consider how things might look from another’s perspective. When he tells the story of the meeting of Montezuma and Cortes, an epic example of mixed signals that leads to one of the bloodiest wars in history, it’s hard not to feel like it’s a metaphor for Trump’s America, two camps of people in different worlds talking past one another. In particular, though, Talking to Strangers speaks to our increasingly dangerous habit of governing according to the panicked impulses of the population.

For years now, the national conversation about policing has been dominated by emotional mob reactions to pebble-bits of information on social media or snippets of video that we debate ragefully and at length, often without even a pretense of trying to learn the underlying context first. Gladwell seems to want to get underneath those reactions, and ends up laying out why knee-jerk takes often don’t work with this issue, beginning with a crucial, oft-overlooked problem that leads to many policing catastrophes: people suck a lot worse than they think at judging people they don’t know.

His thesis revolves around the research of psychologist Tim Levine, who did a series of studies involving subjects given a trivia test. Told they’d win a cash prize if they do well, they’re set up to work on the quiz with a “stranger,” actually a researcher. Midway through the test, their instructor leaves the room, and the “stranger” nudges the subject and points out that the answers have been left on the desk. Why not cheat? In a humorously depressing 30% of cases, they do.

The core of the research rests in surveys of people shown videos of the subjects interviewed after. Some of the liars are terrible. In denying they’ve cheated, they blush and stammer, and look to the side. One was asked if she thought the other subject would also deny cheating. “Probably,” she offered, with amusing hesitancy. Others are good, flatly denying cheating, not hesitating to say their quiz partner would do the same.

When shown these videos, people are terrible overall at identifying who lies and who tells the truth. The average success rate is about 54 percent, barely better than chance. But Levine, and a graduate student named Hee Sun Park, noticed something useful. Most of those surveyed did badly at identifying the liars. But they were far better than chance at identifying the truth-tellers. This is because people have what Levine called a “default to truth.” As Gladwell describes it, this means “our operating assumption is that the people we’re dealing with are honest.” We believe people, and even overlook evidence that something is amiss, when they tell us what we expect to hear.

Gladwell spends much of the book recounting myriad examples of when “default to truth” becomes a problem. A spy for Cuba has a spectacular rise within the Defense Intelligence Agency because even people trained to be on the lookout for traitors have a tendency to accept explanations for curious behavior. The infamous pedophile doctor for the U.S. gymnastics team, Larry Nasser, abused countless girls, often with parents in the room, because victims and parents alike “defaulted to truth,” i.e. they assumed his explanations about the medical necessity of his behavior were true. Bernie Madoff beat back many inquiries simply just because, as one reporter who interviewed him put it in dismissing the idea he was a scam artist, “He’s either the best actor I’ve ever seen or a total sociopath.”

Similarly — we’re getting to how this impacts law enforcement — human beings have a tendency to over-judge people who don’t conform to behavioral expectations. Gladwell brought up Friends to highlight the show’s awesomely unsubtle acting. He had a psychologist named Jennifer Fugate take an episode of Friends and score the facial expressions of the actors using FACS, or Facial Acting Coding System.

You might have seen FACS mentioned here or there; it’s a technique that measures “forty-three distinctive muscle movements” and computes them to summarize the emotion expressed: polite “Pan Am” smile, wide and genuine “Duchenne” smile, anger, confusion, etc. Each muscle movement is given an intensity score from A to E, with E being the strongest.

In a moment Gladwell picked during episode fifteen of season five of Friends, which apparently involved Ross discovering Monica and Chandler were in love, Ross’s facial expression included a string of Es — intense emotions all over! In fact, the FACS scores of all the actors in the scene corresponded more or less exactly to the script, which is why, Gladwell notes, “You can watch the scene with the sound turned off and still follow along… The actors’ performances in Friends are transparent.”

The problem is that in life, people expect others to be like Friends actors, with faces exactly matching what’s going on underneath. When that’s not the case, we either get fooled, as in the case of Madoff, or we overreact in the other direction, as in the case of Amanda Knox, whose weird outward behavior convinced everyone she’d committed murder. It turns out that we only catch liars when we’re wired like famed investigator Harry Markopoulos, whose first instinct was to suspect the worst and to continue trusting his suspicions. His doggedness led him to stay after Madoff until he was exposed.

Eventually, Gladwell comes back to a story involving Kelling. Once upon a time, famous policing theorists like Orlando “O.W.” Wilson were convinced the patrol car would completely eliminate crime. Kelling in Kansas City tested that proposition. He got a grant to divide the city into three sectors. In one, patrols remained the same, in another patrols were doubled, and in the third, patrols were cut off in all cases, with police responding only when called.

The results shocked cops everywhere: there was zero difference between the sectors. People didn’t feel safer with increased patrols. Crime didn’t go up or down. Patrol cars just didn’t matter.

The realization that old-school methods didn’t work became part of the justification for more interventionist concepts like Broken Windows. Police didn’t stop there. Gladwell recounts another influential experiment, conducted by researcher Larry Sherman, called the “Kansas City Gun Study.” Essentially, researchers picked an area in the city with some of the highest rates of gun violence and began aggressively stopping and searching cars, seizing every gun they found.

This was the automobile version of Stop and Frisk, whose ostensible aim was the seizure of guns off the streets. After a program of mass car stops, gun violence in the targeted District 144 of Kansas City dropped in half — a miracle! When the New York Times did a story about the study, Sherman’s phone “rang off the hook” as hundreds of police departments around the country wanted the secret sauce, whether they had severe gun violence problems or not. They all believed Sherman had cracked the code of policing. Patrol cars, they believed, did work after all, so long as cops were empowered to stop as many people as possible and search everyone they could, under whatever pretext was handy.

Just as Broken Windows hugely increased the number of contacts between citizens and street patrol officers, the Kansas City experiment led to a mass increase in vehicle stops. Gladwell notes the example of North Carolina going from 400,000 stops a year to 800,000. The entire United States at the time the book was published two years ago saw twenty million stops a year, an average of 55,000 per day.

A secondary consequence of all this was that police were now trained to reverse their “default to truth” and become maniacally suspicious. Every car stop became an exercise in looking for “the tiniest clues.”

Air fresheners shaped like little fir trees were a sign of a potential drug courier. Fast food remains on the floor suggested a driver reluctant to leave a vehicle containing valuable cargo. Tools in the back seat might hint at secret compartments. Is the mileage high for that model year? Is there a whole key ring in the ignition, or just one car key? Is there too little luggage, too much? Cops became trained to scan every inch of a motorist and his or her car for “curiosity ticklers,” and the visuals were just the beginning. “The officer in an investigatory stop,” Gladwell wrote:

… is instructed to drag things out as long as possible. Where you from? Where you headed? Chicago? Got family there? Where?

This is the backdrop of the Bland case. Brian Encinia was and is the poster child for the automotive stop-and-frisk policies that took America by storm. His record at the time of the Bland arrest was incredible, having already written 1,557 tickets in under a year. He’d stopped three people just in the twenty-six minutes before he stopped Bland!

He pulled Bland over on the flimsiest of pretexts, then acted by the book in working himself into a paranoid state during the course of the stop. While seated in his car and checking her info, he saw her making the dreaded “furtive movement” downward, and convinced himself she was maybe grabbing a weapon. Gladwell tells us this is why Encinia returned to her car on the driver side as opposed to the passenger side, because, “officer safety training has taught me that it was much easier for a violator to attempt to shoot me on the passenger side.”

Conventional wisdom in the wake of George Floyd’s death is that institutional racism is the sole reason black people get killed by cops. As I learned in writing I Can’t Breathe, it’s doubtless a significant part of the picture. In New York City, the fact of being young, black, and male was an explicit “curiosity tickler” for cops in one precinct, who were openly instructed to choose those men for Stop-and-Frisk searches. One whistleblower officer named Pedro Serrano recorded a precinct superior instructing the troops in how to choose targets for stops, saying, “The problem is what…? I have no problem telling you this, male blacks, fourteen to twenty, twenty one.”

The hot-tempered officer who jumped Garner, Daniel Pantaleo, had a history of bad interactions with black residents in that neighborhood. Even the dealers had businesslike relationships with most of the other area patrolmen, but Tompkinsville locals put Pantaleo on an unofficial list of badges to avoid, the way sex workers keep a bad trick list. I found one black man who was strip-searched by him in broad daylight outside a laundromat and came away convinced the man had a sexual hangup. The city was successfully sued by two other men for similar incidents involving Pantaleo, though nothing happened to him. The fact that problem officers like this almost never get seriously disciplined stinks of institutional malevolence.

Pantaleo’s individual behavior, however, was seriously compounded by policy. Garner would still be alive if not for the precinct lieutenant who drove by Tompkinsville Park on that fateful morning, saw him standing on a corner, and ordered the two plainclothes detectives to move him. The senior officer was probably inspired by pressure Staten Island police were getting from developers of the luxury condos across from the park where Garner sold smokes.

Hours later, the two police showed up, dreamed up the pretext of approaching Garner for suspected untaxed cigarette sales, and tried to arrest him, despite the fact that he’d just broken up a fight and was not selling at that time. From there, a little-publicized Kafka factor worked against Garner. The two cops had no way to prove they’d executed their task without getting a transaction on the books. As one officer put it to me, they needed a “piece of paper” to show the boss, so they couldn’t just tell Garner to walk around the block. They needed a charge. This idiocy is why Pantaleo and partner Justin D’Amico launched a lethal struggle to get the enormous, incredulously angry man in a car.

A secondary theme of Gladwell’s book, though he shows it more than elucidating it himself, is that bad herd press coverage about incidents like this often follows simple human misreads of strange people and situations.

The book’s surprising chapter five, “Case Study: The Boy in the Shower,” was probably meant to highlight how the “default to truth” problem allowed Penn State coach Jerry Sandusky to elude investigation for decades. People like famed college football coach Joe Paterno and Penn State University President Graham Spanier believed Sandusky. They “opted for the likeliest explanation — that Sandusky was who he claimed to be.”

Reading between the lines, though, it’s clear Gladwell found the case against Sandusky to be less clear than he expected. It was certainly not as open and shut as the horrific Nasser episode, which he recounted in contrast. The much-publicized “boy in the shower” story was told to Paterno by assistant coach Mike McQueary not right away as is commonly believed, but five weeks after it happened, and the prosecution seems to have misrepresented what McQueary told them (he denies seeing actual rape). There are some other problems with the way the case was reported as well, including overlooked details about contradictory testimony.

Gladwell doesn’t go so far as to declare Sandusky innocent, but he does say the story turns out to be “complicated,” far more than we’ve been led to believe, from Happy Valley to Paterno. Here, a different kind of “default to truth” has come into play, one in which we tend to believe what we’re told by a confluence of media authorities, especially when they agree on a caricatured story narrative early.

Gladwell published too early to say this, and perhaps he wouldn’t have anyway, but I will: such a caricature happened in the wake of Floyd’s death last summer. The battle cries we heard all over America for weeks and months after Floyd was killed were almost all of a single note, ranging from defund the police (“We are committed to dismantling policing as we know it,” Minneapolis City Council president Lisa Bender told CNN last year) to All Cops Are Bastards (“Four simple words are all you need to show your disdain for centuries of brutality,” wrote Vice). It became instantaneous conventional wisdom that police were just evil, the living legacy of racist slave patrols. In the year since, the debate devolved into something more like a exorcism of historical sin from the body of Derek Chauvin than a layered discussion about what’s wrong with policing.

These caricatures not only ended up boxing Democrats all over the country into a corner — “Defund” was and is a loser politically, including among minority voters — but froze out the opportunity to talk about fixable policing issues.

For instance, one big problem is that as politicians began to feel pressure to make cuts to things like mental health and schooling and economic development, they tend to fall in love with magic-bullet enforcement ideas like Broken Windows or the Kansas City gun experiment. Politicians, essentially, keep trying to rule on the cheap using cops. Police are asked to do the jobs of mental health counselors and ambulance drivers and day care center operators and fill a dozen other roles, but at the same original unit cost. Instead of investing in any of those areas in any real way, policymakers just give police more tools to stop and harass people, close their eyes, and tacitly urge them to get whatever job done by whatever means, figuring the occasional outrages will blow over.

Gladwell’s point seems to be that if you ask police to stop millions of cars and pedestrians, and instruct them to look for pretexts to conduct searches of all of them, police will override their “default to truth” and begin to see threats in innocent people everywhere. He’s trying to be understanding about scenes like the Encinia video, by asking readers to look at the policy context underneath that car stop.

The backdrop of the Ferguson, Missouri case, for instance, involved the strained finances of the city. As the Justice Department later found, “City officials routinely urge [police] to generate more revenue through enforcement,” which meant busting people not just for breaking the law but violating municipal order codes:

FPD officers are authorized to initiate charges—by issuing citations or summonses, or by making arrests—under both the municipal code and state law… [which] addresses nearly every aspect of civic life… housing violations, such as High Grass and Weeds; requirements for permits to rent an apartment or use the City’s trash service; animal control ordinances, such as Barking Dog and Dog Running at Large; and a number of other violations, such as Manner of Walking in Roadway.

The controversy that gripped America in the wake of Floyd’s death overlooked a lot of this. Individual police got most of the blame, and in some cases deserved it, but it’s politicians desperate for revenue or lower crime numbers who artificially heighten stranger contacts, jack up numbers of bogus summonses and tickets, and push people like Brian Encinia to fudge pretexts for thousands if not millions of stops and searches.

A percentage of those encounters will always go wrong, and when they do, it’s not always all about racism. It’s usually also about political stupidity, greed, and laziness, and a host of other problems our habit of reaching for simplistic explanations prevents us from understanding. Saying it’s all about race or white supremacy isn’t just inaccurate, it lets bad actors off the hook — especially city politicians and their upscale yuppie donors who vote for these interventionist policies, and are all too happy to see badge-wearing social janitors from middle-class towns in Long Island or Westchester take the rap when things go bad.

Gladwell concludes that “Sandra Bland is what happens when a society does not know how to talk to strangers,” but I think that doesn’t put it strongly enough. Bland is what happens when police spend too much time talking to strangers, and when the rest of us talk too little about why that is.