Making The World A Safer Place

All of us want to feel safe in our daily lives. Yet when we think about crime, our first response is often a blanket approach: find the bad guys, and punish them. But what if there were another way? This week on the show, researchers Sara Heller and Chris Blattman explore how technology and psychology can be used to radically transform our approach to crime.

For more on how psychological research applies to the criminal justice system, download our episode on false confessions.

Additional Resources

Book:

Why We Fight: The Roots of War and the Paths to Peace, by Christopher Blattman, 2022.

Research:

Machine Learning Can Predict Shooting Victimization Well Enough to Help Prevent It,” by Sara B. Heller et. al, National Bureau of Economic Research, 2022.

Cognitive Behavior Therapy Reduces Crime and Violence over 10 Years: Experimental Evidence,” by Christopher Blattman et. al, National Bureau of Economic Research, 2022.

The Promise and Pitfalls of Conflict Prediction: Evidence from Colombia and Indonesia,” by Samuel Bazzi et. al, The Review of Economics and Statistics, 2022.

Thinking, Fast and Slow? Some Field Experiments to Reduce Crime and Dropout in Chicago,” by Sarah B. Heller et. al, The Quarterly Journal of Economics, 2017.

Summer Jobs Reduce Violence Among Disadvantaged Youth,” by Sarah B. Heller, Science, 2014.

Social Networks and the Risk of Gunshot Injury,” by Andrew Papachristos, Anthony Braga, David Hureau, Journal of Urban Health, 2012.

The transcript below may be for an earlier version of this episode. Our transcripts are provided by various partners and may contain errors or deviate slightly from the audio.

Shankar Vedantam:

This is Hidden Brain, I'm Shankar Vedantam. One of the first television crime dramas in the United States debuted in 1958.

Audio/video:

The same guys, wearing rubber Halloween masks. One of them with a sawed-off shotgun. They've knocked off 16 pawn shops, liquor stores...

Shankar Vedantam:

Naked City told stories of detectives in New York's 65th precinct. Each episode was a new case.

Audio/video:

I'm not saying the money he drifted off with is yours, but it's somebody's and it's good hard cash.

Shankar Vedantam:

The premise of the show? There are good guys, and there are bad guys.

Audio/video:

Let me tell you something, mister. We've got you nailed pretty good.

Shankar Vedantam:

At the end of every episode, the narrator declared:

Audio/video:

There are eight million stories in the Naked City. This has been one of them.

Shankar Vedantam:

Sixty years later, crime dramas haven't changed much. There are still good guys and bad guys, but the bad guys are now savage, remorseless. Beyond redemption.

Audio/video:

He slits their throat so they can't scream. What the hell's wrong with this guy?

Thinks he's Jack the Ripper.

Shankar Vedantam:

The way we think about crime dramas is very similar to the way we think about crime. Homicide, robberies and other kinds of violent crime have been spiking in many big cities in the United States in recent years. Coming after a long period of declining crime rates, this has caused people in many areas to demand solutions, to demand that the good guys catch the bad guys. But it turns out there are some profound mistakes we are making in thinking about crime, both in police dramas and in actual policing.

Sara Heller:

Why are people in this situation? Why are they making these choices? And importantly, what can we do as a society to make things better?

Shankar Vedantam:

This week on Hidden Brain, a subtle insight that could radically transform our approach to crime.

In 2016, Sara Heller began teaching a class on criminology. She was looking for a way to teach her students about the economics of crime. She turned to a respected source.

Audio/video:

Where's the boy, String.

DeAngelo, shut your mouth.

Shankar Vedantam:

It was the hit HBO series, The Wire.

Audio/video:

(singing)

Sara Heller:

The Wire is the story of Baltimore street crime. The drug gangs who are working there and the police were who are trying to stop them and the way that it affects the citizenry of Baltimore.

Shankar Vedantam:

If you haven't seen the show, we are going to have spoilers. Season three of The Wire features growing tensions between two leaders of a drug gang. One wants to run the business like a business. The other says, "You don't get it. You're not a businessman, you're a criminal. Embrace it."

Sara Heller:

Yeah, the whole season or for even three seasons, you've gotten to know Stringer Bell, who's one of the heads of the gang. He's also interested in trying to have some legal businesses and trying to develop a real estate career and balancing those two things. There's been growing and growing disagreement between him and the other person who leads the gang with him, Avon Barksdale.

Audio/video:

You know what? I look at you these days, you know what I see? I see a man without a country.

Sara Heller:

Avon wants to think about gangs and territory and building up as a gang leader and he disagrees with Stringer's desire to be more of a legal player and to minimize the violence.

Audio/video:

Why? Because I don't shoot up a block indiscriminate. I ain't hard enough?

Shankar Vedantam:

In a climactic scene, Avon Barksdale betrays his friend and partner. Stringer Bell finds himself in a building, unarmed and alone with an exceptionally violent criminal who holds a grudge against him. Stringer tries to reason his way out of the trap, but it doesn't work.

Audio/video:

It seems like I can't say nothing to change your minds. Well, get on with it.

Shankar Vedantam:

What was your students' reactions when you showed them the season? I mean, it must have been an engaging way to start a class that looked at the economics of crime and drug markets.

Sara Heller:

I think in some respect it blows their mind partly because it was the '90s and they don't know what the nineties looked like. That's a learning experience too. But I think they learned to think about things not through the lens of politics or their prior beliefs about what they think crime is like or what they think policing is like. But they learned to think about why are people in this situation? Why are they making these choices? And how can we shape policy to improve the situation?

Part of its use is that it helps students become social scientists and think about why do things work the way that they do? And importantly, why do people participate in crime? Why are they willing to put themselves at sometimes mortal risk for this kind of lifestyle? And how does policing work and what are police trying to do and how well do they do it? What's effective and what's not effective? And importantly, what can we do as a society to make things better?

Shankar Vedantam:

I want to go back a second to that scene at the end of scene of season three of The Wire. Because of course there's a second subtext that runs through that scene that is again, I think, implicit in our understanding of how crime operates and what the police do. Which is that you have something happen, someone gets shot, Stringer Bell's lying dead on the ground, and then the police come in, basically try and figure out what happened and try and solve the crime. Of course, that's the model that we have is that the only way in which we can intervene is once we know when something is wrong. And we have to investigate what happened and try and find the people who did the bad thing and bring them to justice.

Can you just talk about that idea that in some ways we have a fundamental model of how crime is supposed to operate that almost begins at the point that the trigger is pulled.

Sara Heller:

I think that's right. I think part of the reason we might have that model in our head is because we are thinking about solving crimes. We want to wait to the point where something has gone wrong to then think about who did it, which is necessarily retrospective.

Shankar Vedantam:

Intervening after the trigger is pulled means police are playing catch up. Increasingly, some police departments, with the help of social scientists, have tried to be more proactive. They've found that if you collect large amounts of police data and crunch it algorithmically, you can start to predict future criminal activity. In a previous episode, we explored how gun violence spreads through networks of people like an infectious disease. It's a very specific kind of infection.

Here's how sociologist Andrew Papachristos described it to me.

Andrew Papachristos:

It tends to be very specific behaviors, risky behaviors, that put you in these networks. And in some ways it becomes much more like the spread of diseases through needle sharing or unprotected sex, rather than catching a bullet from somebody sneezing.

Sara Heller:

There's lots of evidence that violence is very concentrated, both among people and among geography. There's a bunch of analytic work trying to figure out which blocks of which neighborhoods are likely to have the most gun violence. There it's a little bit easier to think about how do we put police resources where the most crime is happening? There's lots of evidence that that is successful and especially with violent crime, it does respond to increased policing.

Shankar Vedantam:

But relying on police data to build a model of future criminal activity has two big problems. First, if you care about civil liberties, it could make you uneasy. If I think you are likely to commit a crime in the future, I might be tempted to monitor where you go and what you do, even though you haven't done anything wrong. There's a second big problem with building predictive algorithms from police data.

Sara Heller:

There's a lot of evidence that police data are not a perfect measure of underlying behavior. They're a combination of individuals' behavior and police decision making. Who they decide to stop, who they decide to arrest if you're using arrest records. I think what the evidence suggests is that a combination of police officers' implicit bias and sometimes outright racism generates a set of arrests that don't mean the same thing for the same people. That they might be potentially, if police are more likely to arrest young black men, then it's going to look like young black men are riskier just because of police decision making and not because of actual risk level. It's these kinds of biases and algorithms that I think have become a really important central part of the conversation about how we should be using prediction methods.

Shankar Vedantam:

Many critics have said police data might be very good at predicting future police behavior. Prior arrests the police make are a strong predictor of future arrests. But does this data really paint an accurate picture of offenders, the people actually committing crimes? Sara, who now teaches at the University of Michigan, said police data cannot be simply taken to be an unbiased account of criminal behavior. Rather, it's an account of criminal behavior as observed by and reported to the police.

Sara Heller:

But then you have to stop and look at that and think, "How do I know if this is a good prediction?" You might be very good at predicting police behavior, but in fact you have no idea whether you're actually predicting actual offending behavior. There's just no solution to that because we don't have other good measures of who's actually pulling the trigger.

Shankar Vedantam:

We have a lot of data from police arrest records and other police information. But as you say, when two people call the police to investigate different incidents, even if the incidents are more or less identical, the police response to those incidents might not be identical. And if they're not identical, then the data that goes into the police system is going to be biased, not by the difference between the underlying incidents, but the difference in the police response to those incidents.

I want you to talk a moment here about the insight that you had, which is that if we stipulate that predicting future crimes is difficult because of these biases that are implicit in the police records, what is the solution to that? Because of course, we are reliant on police data in terms of figuring out what happens. They're the people who we call when something happens, when there's a crime that takes place. So if you have this flawed data, how do you make more sense of it for it to become more useful?

Sara Heller:

The realization that we had is that arrests for shooting and homicide are not our only measure of gun violence in the police data. We can also measure victimization.

Shankar Vedantam:

Victimization. Go back to that scene at the end of season three of The Wire. Once the police find someone has been killed, they start to investigate. They carry out forensic tests, they conduct interviews, they make arrests. All of this information goes into the police database. If you are worried about police impartiality, you might worry about the integrity of this data. But there is one aspect of the data that is virtually certain to be uncontaminated by bias. It's the fact that Stringer Bell is lying on the ground dead.

Sara Heller:

What we have is a measure of whether someone is shot. Almost all cases where someone is shot end up in police records because people seek medical treatment and the medical staff are required to tell the police.

Shankar Vedantam:

In other words, let's say there's someone who gets shot on a street corner. It's possible that police have some leads on a suspect. It's possible that they find a suspect. It's possible they found the right suspect. It's also possible police bias is involved in identifying the suspect or tracking down the person. But the one thing we absolutely know for sure, is we know the person lying on the street in a pool of blood. We know who that person is and we know that's happened.

Sara Heller:

That's right. I think people might worry that in some cases if you're shot, you would be afraid to go to the police. You might not want to report it to the police. But even if the police aren't there, even if you're not shot so badly that you're lying on the sidewalk, but you can walk away, you are very likely to go to the hospital to get medical treatment. Almost everybody seeks medical treatment even if they're not willing to call the police. Once they do that, you still show up in the police data.

Shankar Vedantam:

Sara and her colleagues asked themselves a question. Police data about previous arrests were being used to identify people at high risk of future arrest. Could the same data identify people at high risk of becoming future victims? They got access to data from the Chicago Police Department over a 20-year period. The data included information about police arrests for minor crimes like vandalism, to major crimes like shootings or robbery. Sara and her colleagues built a machine learning model that crunched the data looking for patterns. It found a match. By looking at prior arrests and other police data, the model pinpointed people at high risk of becoming future shooting victims. How accurate was this prediction? Sara picked out a sample of more than 300,000 people from the police database and tracked them over the next 18 months to see if the prediction was accurate.

Sara Heller:

It did startlingly well, or it did shockingly well. Where if you take those 330,000 people, we have some predicted probability that each person is going to get shot in the next 18 months. Line them up in order of risk level and look just at the 500 people with the highest predicted risk, so the very smallest group with the very highest risk. And it turns out that of those 500 people with the highest predicted risk, 13% of them are shot in the next 18 months. That's just a shocking risk of something as serious as a shooting victimization. Think about being in a crowded movie theater, about 500 people, and I tell you, "Thirteen percent of that group is going to be shot in the next year and a half." That's an extraordinary risk of potentially lethal violence. It's a rate that's 130 times higher than the average Chicagoan. We can find a relatively small group of people that are just at an extraordinarily high risk of being shot.

Shankar Vedantam:

I mean, it almost seems spooky to be able to do that, to be able to say, "I can tell you in the next 18 months who's going to get shot in a city of three million people." I mean, that's almost scary, spooky.

Sara Heller:

It is a little bit disturbing. I mean, imagine someone coming to you and saying, "You have more than a one in 10 chance of being shot in the next 18 months." It might in fact motivate you to participate in some of these programs that Chicago is now trying to offer. It's certainly not perfect prediction. A 13% rate means there's also 87% of the people who are in that group who are not shot. It's not a perfect prediction, but the fact that we can find such a small group at such an extraordinarily high risk really highlights, I think, the fact that we should be doing more to help them. We know who they are. We don't know who everybody is, but we know a small set of people at such high risk that we really should be spending more to try to help keep them safe.

Shankar Vedantam:

There is a huge obstacle in doing that. I'll explain more in a moment, but I want to stop and underscore Sara's subtle insight. Using police records to predict who is going to be involved in crime in the future runs the risk of the dystopian response you see in movies like Minority Report, where police go after criminals before they have committed crimes. It raises troubling questions about civil liberties, bias and surveillance. But it becomes much less problematic to use the police data not to paint a picture of would-be shooters, but a picture of would-be shooting victims. If I told you that you were at risk of doing something bad in the future, you might bristle. But if I told you that you were at risk of something bad happening to you in the future, you might listen. It's what doctors do for patients every day. Why would it be hard to apply Sara's insight to protect shooting victims? For that, you have to ask yourself a question. Why would police data on past arrests be so effective at predicting which people will get shot in the next 18 months?

Sara Heller:

In criminology, there's lots of evidence that victims and offenders tend to be the same people. This is certainly not always true. You can think of a lot of examples where it's not true. Domestic violence victims for instance or bystanders. But it is the case that the kinds of risky behaviors that lead you to be at risk of being shot may also be the same kinds of risky behaviors that make you likely to be an offender or be involved in other crime.

Shankar Vedantam:

So the big obstacle to doing anything useful with Sara's insight is us. When we see Stringer Bell lying on the ground, most of us don't say, "What a tragedy." We say, "What did he expect? If you run a gang, order hits on other people, someone is going to have you in their crosshairs." Sara says the sheer cost of crime and the benefits of preventing it call for an overhaul in our thinking.

Sara Heller:

I think policy makers have an innate understanding that prevention might be better than remediation. That if you can keep something that's very socially costly from happening, we should try to do that. That's not necessarily the model where you just wait till something happens and punish it and then hope that changes future behavior. If what you're trying to do is prevent crime, then you want to think about all of the ways in which you can prevent it. That's not just stopping someone from pulling a trigger. It might also be stopping someone from being in a position where they would be at risk of being shot.

Shankar Vedantam:

I'm wondering if one of the important insights from this work, Sara, is the importance of thinking about people in some ways, first and foremost, in terms of their risk of victimization, instead of thinking about them first and foremost in terms of their risk for becoming perpetrators. There is an overlap between those two worlds, as we discussed. But if we think about you as a potential victim, in some ways I think it changes the orientation with which people approach how they can think of you. The conventional model really is we think that someone's a threat. How do we put a fence around them to keep them from harming other people? The model that I hear you suggesting is we fear that someone's going to become a victim. How do we put a safety net under them to keep them from falling?

Sara Heller:

I think probably we should just have you write our papers from now on because I think that's a very eloquent way to put it. That if you think who's at risk of being a victim, it humanizes people. We want to stop all sorts of victimization and that's true in terms of the way we allocate medical care. We try to prevent harm from heart attacks and from depression and suicide and we try to prevent harm from homicide as well. I think coming at it from that perspective, I hope will motivate people to take prevention more seriously. To not just see the problem of gun violence as, "There are bad people out there doing bad things," but to think about how can we help keep some of the most vulnerable members of our society safe?

Shankar Vedantam:

There is a crucial mind shift that is demanded by this work. When we see Stringer Bell lying in a pool of blood, Sara asks that our first reaction not be, "Well, he was a criminal, he had it coming." Instead, she's suggesting we say, "This death, like any other shooting death, is a tragedy. How do we minimize such tragedies?" She's suggesting that instead of watching out for people like Stringer Bell, we start to look out for them. Now this can sound naive and idealistic. Some people may bristle at the idea that we should go out of our way to save the lives of people who have been previously linked to violent crime. But it turns out, if you start to think of offenders in terms of their propensity to become victims instead of their propensity to commit future crimes, you can end up having a profound impact on both those things. That's when we come back.

You're listening to Hidden Brain, I'm Shankar Vedantam.

This is Hidden Brain, I'm Shankar Vedantam. In 2004, Chris Blattman was having an interesting chat with a man in Nairobi whom he just met.

Chris Blattman:

I was about 30 years old, I was a economics graduate student. I was working in Nairobi in Kenya. One day I was sitting for lunch in a little cafe and a man strikes me up in conversation, super engaging. "What am I doing? What am I working on?" As I'm turned towards him and away from my backpack, his pal grabs it and walks off with my laptop. Fortunately, not with my passport.

Shankar Vedantam:

When did you notice the laptop was gone, Chris?

Chris Blattman:

Probably it was only sometime afterwards when I got up to leave and wondered where my backpack was. I spoke to the police, there was a cop there, and he was like, "Oh, did somebody strike you up in conversation?" And I said, "Yes." He's like, "Yeah, that's what happened, kid."

Shankar Vedantam:

Chris turned to local internet cafes to do his work and to stay in touch with colleagues, but the internet speed in these cafes was painfully slow.

Chris Blattman:

It would be about 10 minutes for each Hotmail message to load up and so it was customary to talk to the person next to you and just you ended up in a lot of idle chatter over your hour or two in the internet cafe. That was just the culture of the time. I purposefully sat next to a woman my age and we ended up striking up a conversation. She was a psychologist and a humanitarian worker who had been working in Northern Uganda.

Shankar Vedantam:

The chance meeting, occasioned by the stolen laptop, turned out to be important.

Chris Blattman:

We wrote our dissertations together. We got our first jobs at Yale together, we wrote papers together. We have had two children together, more importantly, and we've now been married for 15 years.

Shankar Vedantam:

Did you ever send a thank you note to the guy who stole your laptop?

Chris Blattman:

No, if I ever meet him, I will give him a grateful hug.

Shankar Vedantam:

After Chris finished his PhD, he began working at a think tank in Washington, D.C. One day a colleague asked him out to lunch. He wanted Chris to meet someone he had been working with in Liberia, a man named Johnson Bohr.

Chris Blattman:

Johnson is this big, smiling, meaty man. The one thing I knew about West Africans is that the one thing he probably had not had in a week or so was a huge mountain of rice. So we went to a Pan Asian restaurant and that was how we bonded. He was like, "Finally, someone has offered me more rice than I can eat," which is the secret to a Liberian's heart, it turns out.

Shankar Vedantam:

Chris forgot about Johnson after that lunch. But sometime later, Chris found himself in Liberia. The country was recovering from civil war. In 1989, a warlord named Charles Taylor invaded the country and seized power. What followed was years of uprising and violence until Taylor was finally exiled to Nigeria in 2003.

Chris Blattman:

When I arrived in early 2008, the country was in year five of peace. It had just had a democratic election that brought to power Africa's first female president. There were 15,000 peacekeeping troops, so it was a time of hope, but the place was decimated. When I was driving in from the airport, you would see just burned-out hulks of buildings everywhere. The Ministry of Defense was a blackened concrete pile. What few apartment buildings there were, were totally burned out. There were no walls, it was just this concrete skeleton and people had set up this 12-story tarp city on various floors of these buildings.

Shankar Vedantam:

Can you paint me a picture of what the state of human development among people, especially young people in Liberia?

Chris Blattman:

This was a country that had been decimated by 14 years of political instability and war. Anybody who could get out, got out. They went to Ghana, they went to the United States, they went somewhere. Anyone with an ounce of education or money did so. The people who stayed behind were all displaced from their farm, from their apartment, and they'd lost everything they had. If you were a young man and you didn't get out, there was a good chance you were pulled into one armed group or the other.

Shankar Vedantam:

As the war ended, many of the people who had left the country came pouring back in. They were starting businesses, investing in the economy, getting into politics. But the young men and women who were left behind during the war still felt left behind after it was over. They were not included in this new wave of development. Instead, they found themselves returning to the life they knew.

Chris Blattman:

These were the guys who would sell drugs or pick pocket. There was a growing amount of armed robbery, home invasions and muggings. If you were the president or if you were the UN peacekeeping force, these guys were your number one concern because these were going to be the next insurgents or they might be the next mercenaries for the civil war in the country next door.

Shankar Vedantam:

Chris started working at a reintegration center where young men and women at high risk for violence were learning to farm. He ran into his old D.C. acquaintance Johnson Bohr. Johnson had been working with people displaced by the conflict and was helping out at the center. One morning, Chris awoke to shouts.

Chris Blattman:

I can hear this rumbling and tumult and we walk from our little hut and past the main building. As we turn the corner, the noise is getting louder and louder and all 400 former combatants were striking, yelling, protesting in the yard. The food wasn't ready that day. The rubber boots hadn't shown up. All these little grievances spilled into what essentially became a riot. There was Johnson standing in the middle of them saying," We have talked about this. Get your people back to their rooms. We have ways to deal with these disputes. This is not how we do it. Fix this because this is going to go wrong." And he got them back to the negotiating table just by singular force of will, when everybody else on that campus had fled. This guy I'd met briefly in Washington, all of a sudden had something magical about him where he could operate and do things and had a charisma and a bravery and a talent for this, that I just wanted to get to know him better.

Shankar Vedantam:

Chris started following Johnson around Monrovia, the capital.

Chris Blattman:

Monrovia is a city built in a mix of jungle and swamp. It's like being in Florida at the height of summer. We're wandering the red light market to try to find all of these young men living on the fringes, doing these illegal activities, trying to understand how all that works. It's called the red light district because it was the site of Liberia's first traffic light, not for the normal reason that we have red light districts, but it was also that kind of red light district. It was everything.

Shankar Vedantam:

The streets were lined with vendors selling fuel, used clothes, beauty products, cheap electronics. Here and there, men loitered around. Johnson took Chris into small shacks, drug dens where they met former combatants engaged in illegal activity.

Chris Blattman:

We'd sit there and we'd just chat with these guys and we'd talk business. "How do you make money? How many customers do you get? How did you end up doing this? What's next?" I do remember coming out of one of these bigger drug dens and you're blinking in the sunlight. And as we are coming out, this guy comes to us from across the street and he was shining shoes. He gives Johnson a big hug and they're talking and catching up. I said, "Oh, how do you guys know one another?" He said, "Well, I used to be like them," and he points across the street to the drug den we just left, "and then I went through Johnson's program."

Shankar Vedantam:

As the days went by, Chris met more and more men who approached Johnson to thank him for helping them turn their lives around. Chris got curious. What was going on? They sat down together at a cafe to talk. Johnson told Chris he was inviting groups of young people on the fringes of Monrovia to attend a series of meetings.

Chris Blattman:

Day one, the first thing they would always do is they'd talk about how you present yourself, because a lot of these guys are not showered. They don't take care of themselves. They're wearing tattered clothing, they don't look respectable, they look tough.

Shankar Vedantam:

Johnson would introduce a special guest.

Chris Blattman:

He said, "If you want, we've got a guy here who can give you a haircut." And by that I mean he had a chair, a mirror and a razor blade, and you'd shed your facial hair or the fuzz or whatever you have that signals that you're not a part of normal society can go. You're not compelled to, it's just an option. And that was their big event, is maybe half the guys he said would do that and the others would sit back and say, "I'm not so sure about this." It was just about let's take that first baby step.

On day two, some of the guys who did get their haircut come back and they've also gotten a polo shirt and maybe a pair of khaki pants and they're wearing shoes instead of flip flops. The clothes aren't as dirty and they're not as patched and they're certainly not sending that badass signal. And they're invited to talk about what happened the last day or two. What was your experience? Most of them are like, "It was amazing. Nobody recognized me. And when they did, they were laughing. And I went into this store and nobody kicked me out." Most of them just talked about how amazing this was.

Shankar Vedantam:

Johnson told Chris that in subsequent meetings he would talk to the young men about dealing with everyday violence: what to do when something bad happened around them.

Chris Blattman:

They were finding themselves in a brawl once or twice a week. Their disputes would turn into yelling matches. Someone might get stabbed once in a while. Just managing these everyday interactions without violence is just hard and it seems inescapable. We all have these relationships in our life. Maybe it's your son or your mom or something where you can just lose your temper six seconds into the dispute. You don't do that with everybody. You can be totally calm and cool-headed with 49 other people that day and then just lose it like that. We all have relationships like that and the problem was that theirs just didn't devolve into yelling. When they lost it, somebody could get hurt.

Shankar Vedantam:

Johnson would ask the men to think about how their emotions flared up in these moments. What did it feel like to get angry?

Chris Blattman:

Then people stand up and then they tell their story. Here's this time when I exploded and we lost our tempers and somebody got stabbed, and who else has a story like that? Then they'd also talk about, okay, what do you do in that circumstance? How do you actually avoid that? How do you manage these hostile social situations? How do you manage that one relationship that goes from zero to 60 in an instant?

Shankar Vedantam:

At some point soon after this, you go back and have a conversation with your wife and you tell her a little bit about what Johnson is doing. She's of course a psychologist. Describe to me what you told her Johnson was doing and how that conversation unfolded, Chris.

Chris Blattman:

I said, "Do you remember that guy Johnson Bohr? We just wrote down everything he did in his program. Take a look at this. What does this look like to you?" She paged through and she read it and she said, "Huh, this looks like cognitive behavioral therapy."

Shankar Vedantam:

You're listening to Hidden Brain, I'm Shankar Vedantam.

This is Hidden Brain, I'm Shankar Vedantam. Several years ago, the economist Chris Blattman, now at the University of Chicago, met a charismatic Liberian. Johnson Bohr was running a shoestring program of his own invention in the poorest areas of Monrovia. He was trying to divert troubled young men and women from a life of crime. Johnson didn't have a name for his program, but it was a form of cognitive behavioral therapy. Developed in the 1960s, CBT, as it's known, suggests that by paying attention to how we act and what we do, we can subtly change how we think and what we feel. At the core of the program is the idea of practice. If you practice behaving differently, your outlook, thoughts and emotions will follow.

As an economist, Chris wanted to know if Johnson's brief eight-week program had any effects. He launched an experiment tracking about a thousand young men in Liberia who were at high risk for crime and violence. Half were selected at random to go through Johnson's program. Chris remembers the setting of the first session. It was held in one of those buildings he'd seen coming in from the airport when he arrived in the country.

Chris Blattman:

It was a six story skyscraper that was half built when the war broke out and at that point was just a concrete shell. Concrete floors, stairwell, and then open to the air. We were on the third floor and we had 20 scarred and half broken plastic chairs in a circle and Johnson and his assistants would stand in the middle and try to lead these guys in discussion.

Shankar Vedantam:

Chris could tell that many of the young men at the meeting were there because of Johnson, but that didn't mean they were converts.

Chris Blattman:

About half of them are slouched in their chair, their arms are crossed, they're looking very skeptically. They're also cautious because they're surrounded by a bunch of guys like themselves, which is almost never a good situation. It's a circumstance where they also have to be cautious. Maybe they'd been in a fight with that guy the previous week. One guy was a real holdout, very hostile and tough in his demeanor. Like, "Just don't mess with me," slouched, arms crossed, frowning. Still there though. Still coming for the first day, still coming for the next few days after that. Not engaging, not when people are given an opportunity to stand up and testify or tell some story or relate. Was not engaged, but was just checking it out, sitting at the back of the room.

Then one day he came back, but of his own volition, he had gotten this haircut and trimmed his beard and was dressed nicely. The only reason I knew it was him was because he sat down in the same chair that he was always in. And he was brought into the circle and he was engaged, but nobody made a big deal out of it as if they knew that that was the right thing to do. It was really remarkable to see that change day-by-day and how they tried to present themselves to the world and how that could change their own behavior.

Shankar Vedantam:

The experiment divided a thousand young men into two groups. After half received Johnson's program for eight weeks, the larger group was divided at random into two new groups. The first received a cash gift of $200, the other did not. Effectively, the experiment created four groups. Men who received cash and therapy, men who received cash alone, men who received therapy alone, and men who received neither. The initial results were promising. Chris decided to test something much more ambitious. Would there be any benefits after 10 years? He asked a large group of experts what to expect.

Chris Blattman:

Most people thought that therapy alone would have no lasting effects whatsoever and cash and therapy, the vast majority thought that the effects might be zero or at least deeply dissipate. Almost no one predicted that they'd persist in any kind of strength.

Shankar Vedantam:

Chris recently finished tracking the men over 10 years. There were modest long-term benefits for the men who received therapy alone. For the men who received cash only, it was a whole different story.

Chris Blattman:

For the men who received cash only, they changed their life in the short run, but the business that they started failed. Maybe they got rained out and damaged goods, or the police seized their wares or they were robbed. So within a year, they were no different than people who had received neither program. And 10 years later, they were no different.

Shankar Vedantam:

But the men who received the therapy and the cash gift?

Chris Blattman:

Persistently through the life of the project, after one month, after one year, after 10 years, the men who had received the therapy, followed by the cash, their behavior was totally transformed. They were half as likely to be selling drugs as the control group. They were half as likely to be engaged in thefts and muggings, to the point where it added up to an average of 26 fewer crimes per year per person that got both cash and therapy.

Shankar Vedantam:

I mean, that's an astonishing finding, Chris, 10 years after an eight-week intervention and $200. The net total cost of the intervention is probably $500 some dollars per person. Ten years later, you're seeing a 50% reduction in robberies? I mean, that is astonishing.

Chris Blattman:

It seems to come from the men who were the most far gone is where most of the crime and violence happens. Amongst those most far-gone guys, this intervention in the first month helped them climb out of that hole. The amazing thing is that once they climbed out of that hole, and even a month after the interventions, they were less likely to be engaged in crime, they stayed out.

Shankar Vedantam:

What do you think explains the fact that the cash transfer seems to have helped the cognitive behavioral therapy persist in terms of its effects? Especially when the cash transfers don't seem to have done anything on their own?

Chris Blattman:

I would say there is evidence that the therapy alone was impactful after 10 years. What the cash did, we think, is you finished therapy on week eight and then you had to go and eat the next day, or you still had to find a place to sleep. Many men had to turn back to mugging or crime or drug dealing just to make ends meet, or at least that was their best option. But if we gave you $200, one is you had a little bit of extra money in your pocket to survive a bad week without turning to crime. You could get some shelter, you could get some food, you could get some better clothing. You could also have enough money after that to maybe start a shoe shine stand or a little petty business. That gave you a chance to practice your skills.

If CBT is about practice, and if behavior shapes thinking, then it was this positive reinforcement and an ability to keep practicing the good behaviors and the new you and this new image and identity you'd put on for size. It really entrenched it such that some months later when the money was gone, you had an extra two, four, six, eight months of practice.

Shankar Vedantam:

There have been a number of programs in the United States as well that have asked, "Can we apply therapeutic interventions to basically address challenges of crime? Someone could look at your results in Liberia and say, "Great, Chris Blattman has found the magic pill. All we need to do is give people CBT and some money and we've solved the crime problem in many cities." What's the problem with that kind of thinking, Chris?

Chris Blattman:

Yeah, I mean, imagine you went to your doctor and you were halfway through describing what was wrong with you and she said, "Stop right there. Tylenol and radiation, we're going to cover all the bases. That worked for the last guy, it'll work for you." You would instinctively know that you need to get a new doctor. And yet we elect a mayor in Chicago or New York and we want them to tell us the secret is Tylenol and radiation. We want them to say, "Oh, New York did stop and frisk, or New York did hot spot policing, New York did CBT or something and just, let's take this off the shelf solution, must work for us." And we skip over that important step of diagnosis and say, "Wait a second, do we have the same kind of violence as that place? Do we really understand?" Because there are some different roots of violence and the treatment has to fit the diagnosis.

Shankar Vedantam:

Chris and Sara Heller, whom we heard from earlier in this program, are running an intervention along the lines of the Liberian study in Chicago. It's based partly on Johnson Boar's insight that you can change the course of people's lives with therapeutic interventions. And it's based on Sara's insight that thinking about people as potential victims rather than as would be perpetrators, has the capacity to both reduce their risk of becoming victims and their propensity to commit crime. The data out of the Chicago intervention is still preliminary.

Chris Blattman:

There's some incredibly promising results, a huge reduction in arrests for shooting and homicides. More than a 50% improvement. On the other hand, they're slightly more likely to be arrested for other types of violent offenses, not nearly as serious, but not running in the same direction like we expected. Then in the middle, they're actually less likely to get shot themselves. But that decrease is not so large, nor is it what we'd say is statistically significant. On average, it's ambiguous. On the other hand, if you weight these things by the cost to society and the cost to the young men, it's unambiguous that this is just a huge improvement.

Shankar Vedantam:

What works in Monrovia might not work exactly the same way in Chicago. The only way to figure out what works in a given area is to spend time diagnosing the problem and conducting careful evaluations of different interventions. But there are broad insights from the work that Chris and Sara and many other researchers are exploring. One of the most important is that the old model of crime show dramas -- with good guys chasing bad guys and bad guys never changing their ways -- that model is wrong.

Chris Blattman:

We assume these people are born into this and stuck and nothing will change. That they're hardened criminals or hardened killers. I think what we see is that's just not true. That most of these young men want a way out of this life. They will take it if you give it to them, and then if whatever you're offering them works, there's a good chance they'll get out of that life or at least be a lot less violent.

Shankar Vedantam:

Chris Blattman is an economist and political scientist at the University of Chicago. He's the author of Why We Fight: The Roots of War and The Paths to Peace.

Hidden Brain is produced by Hidden Brain Media. Our audio production team includes Brigid McCarthy, Annie Murphy Paul, Kristen Wong, Laura Kwerel, Ryan Katz, Autumn Barnes, and Andrew Chadwick. Tara Boyle is our executive producer. I'm Hidden Brain's executive editor.

Our unsung hero today is Kerri Hoffman. Kerri heads up the Public Radio Exchange, or PRX. She's one of the leading strategists across all of public radio and over the last few years, she has been a great friend to Hidden Brain. As I launched Hidden Brain Media, I've had many conversations with Kerri about podcasts and radio shows and the larger art of building stable, self-sustaining, high-quality journalism. She's always thoughtful, always generous, and wicked smart. Thank you, Kerri. I'm really grateful.

If you are a supporter of Hidden Brain, thank you. If you are not, please help us to produce more shows like this. When you support us, you are telling us you have our back. Go to support.hiddenbrain.org. Again, if you would like to help us keep building the show you love, go to support.hidden brain.org and pitch in.

I'm Shankar Vedantam, see you soon.


Podcast:

Subscribe to the Hidden Brain Podcast on your favorite podcast player so you never miss an episode.

Newsletter:

Go behind the scenes, see what Shankar is reading and find more useful resources and links.