“I Don’t Know What You’ve Done With My Husband But He’s a Changed Man”

Johnson BORH: All of a sudden we heard of the war.

The war was a civil war, the first of two in Liberia, West Africa. It began in 1989. Johnson Borh was 18 years old. This audio is from an ABC News report.

REPORTER: As rebels fire blindly, some duck for cover. There is little place to hide.

Borh was in high school at the time, getting ready to move up to a vocational school. But, he says, he was conscripted into the NPFL, the National Patriotic Front of Liberia. That was a rebel group led by Charles Taylor, who would come to be known as an infamously brutal warlord. The NPFL was fighting against the government of Samuel Doe. When the war started, Johnson Borh was thrown right into the action.

BORH: It was chaotic. When you are on the front line, the very first thing you think about is that your opponent should not be able to overrun you.

He became a company commander.

BORH: I was around when people were tortured. I was around when people were killed because I was part of the rank.

The population of Liberia at the time was barely 2 million. During the war, roughly a quarter of a million people died, many of them civilians. Borh told our producer, Christopher Werth, about one distinguishing aspect of the Liberian civil war: the use of child soldiers.

BORH: There were a lot of child soldiers. Understand that we even had a unit called SBU, Small Boy Unit, that was also headed by child soldiers. They are not afraid to go anywhere that they are requested to go. They have not experienced life so that fear of death would not be there, especially if some of them are taking drugs and all that kind of stuff or they have arms in their hand. Child soldiers were always used in the instances of torture. Even if there was a case wherein such a person needed to be executed, you always would want to use child soldiers to do that.

Christopher WERTH: Did you have to be involved in something like that? Either ordering an execution or torture or anything like that?

BORH: Not really. But I could say most likely, because if you are leading the troop and you have a troop under you, we have a saying that, “What the unit does the commander is responsible.” You, the commander, will always be responsible for those kinds of stuff.

Borh was eventually wounded in combat.

BORH: I feel like I could have died because I went off. I was out of blood, so as soon as I got to the hospital they had to put some blood in me and drips in me.

But in the scheme of things, being wounded was not the worst thing that could have happened.

BORH: Because that was one way you could get escape from the war. Either you run away, or if you are a wounded soldier, then they let you go because they know they can’t actually make use of you. It was at that time that I left the war.

Charles Taylor’s rebels prevailed and he took over the presidency of Liberia for six years. He was eventually convicted of war crimes and crimes against humanity; he’s now serving 50 years in a British prison. After two civil wars and more than a dozen years of on-and-off fighting, Liberia finally began to settle down. Elections were held, and the country tried to find a new normal. But it had to deal with the fact that a generation of its young men didn’t really know what normal meant.

BORH: That’s the problem that we face in Liberia. Because of the number of child soldiers that were involved with the war, their reinsertion into normal community is a very serious issue.

Chris BLATTMAN: Liberia is a very small country. It’s only about 4 million people.

That’s Chris Blattman. He is an economist and political scientist at Columbia University who’s spent time in Monrovia, Liberia’s capital.

BLATTMAN: It’s tropical. It’s one of the rainiest countries in the world. Picture a city in the middle of a swamp in a jungle, like Washington, D.C., except run down. Run down not just by the tropics, but really run down by the war and years of decline.

Blattman is essentially a development scholar and, therefore, pays particular attention to poverty and the associated violence. He works primarily in Africa and is an adviser to the Liberian government. Liberia has made gains since the wars, but it’s hardly flourishing. Gross national income per capita is $400; on the U.N.’s Human Development Index, Liberia ranks 175th out of 187 countries. Adult literacy is below 50 percent; life expectancy is only about 60. The homicide rate is about 10 per 100,000 people – which, just to put that in context, is about double New York City’s, but only a little more than half as bad as Chicago. Even so, the U.S. State Department ranks the threat of robbery and political violence in Liberia as high. Chris Blattman has tried to forecast when and where violence may break out. Part of the problem, it seems, is an artifact of war.

BLATTMAN: There was nobody whose family wasn’t touched by the war, almost. Every family, if they didn’t flee, had somebody involved in the war in some way. Most people reintegrate very well, but years after the war there’s who knows what percent, but some chunk of these young men were still doing really poorly. You have guys who have lived lives of violence who haven’t reintegrated, some of whom are engaged in crime.

Johnson Borh was seeing this firsthand. He was by now in his mid-20s. He had happened upon a couple of humanitarian organizations and he’d been exposed to trauma- healing and peace-building. And he’d gotten some training on how to work with troubled ex-combatants who’d experienced a lot of the things he had.

BORH: I was using those skills to get into their base. With the new skills that I have had to discourage them from going back to war. Because of our situation in Liberia, the war was a form of virus because people were leaving from one faction to another in a different part of the country. That was how it was and people kept recruiting child soldiers and what have you. In fact, those that were still in the war, we try to discourage them. Spread a message to all the youth around the community that war is not good.

In 2000, Borh and another ex-combatant started an organization called the Network for Empowerment and Progressive Initiative, or NEPI. The idea was to help former soldiers and other young men who were in trouble or heading for trouble. NEPI partnered with larger, international organizations — and, at a conference in Washington, D.C., Johnson Borh met Chris Blattman. It sounded to Blattman as if Borh’s NEPI was having a dramatic impact on the Liberians it worked with. Blattman wound up working closely with Borh and his colleagues in Liberia so he could see for himself exactly what they were doing.

BLATTMAN: I think they stumbled somewhat accidentally onto C.B.T.

C.B.T., or cognitive behavioral therapy, is a form of psychotherapy that tries to teach behavior change without necessarily attacking the underlying psychological roots of that behavior.

BLATTMAN: They cobbled together curricula — from the U.S. and who knows where — and training manuals, and this and that and their own experiences. They experimented and experimented and experimented all the time. When I met them, almost 15, 20 years after they’d left the war, they had this program that seemed to bring in these really outcast young men in one end and then, with some probability, spit out changed people.

Today on Freakonomics Radio: what’s the evidence that people truly can change with something as simple and cheap as C.B.T.? Also, what happens if, besides the therapy, you throw in a cash incentive? And could the same kind of behavioral insights be applied to domestic violence?

Simon RUDA: I think one woman in particular said, “I don’t know what you’ve done with my husband, but he’s a changed man.”

*      *      *

In our previous episode, we looked at a couple of programs in Chicago that use cognitive behavioral therapy, or C.B.T., to try to keep young men from committing crimes and dropping out of school. One of them was called Becoming a Man, or BAM, run by Tony DiVittorio.

Tony DiVITTORIO: I was working with these youth who were referred to me for anger issues, cutting class, yelling at their teachers, what have you.

A team of researchers from the University of Chicago’s Crime Lab helped expand the programs and, importantly, they randomized the treatment, so they could evaluate if it was effective.

Sara HELLER: The first time that we saw these effects, we thought, “Wow, can this be right?”

The criminologist Sara Heller helped evaluate BAM.

HELLER: The first year of BAM, we saw a 44 percent decline in violent crime arrests, and a 36 percent decline in other nonviolent, non-property, non-drug arrests. We see an increase in school engagement that we think might end up eventually, once the kids are old enough, improving graduation rates by maybe 7 to 22 percent.

The reason C.B.T. was effective, Heller and her colleagues argue, is that it teaches people to stop and think about their behaviors in a way that they rarely do.

HELLER: The basic theory rests on something that we’re calling automaticity or automatic behavior. Automatic behavior is something we all engage in all the time, right? It’s behaving without consciously thinking about it.

And then, if done right, C.B.T. teaches people to disrupt those behaviors. It gets them to reframe their perspective, to think differently about their negative emotions, and how to respond to them. But it seems as if the benefits of C.B.T. are relatively short-lived, that they fade out after a year or two. Now, as Sara Heller points out, this doesn’t mean the therapy isn’t worthwhile.

HELLER: If there’s a way to reduce violent crime, even for just a year at a pretty low cost, we should do it.

And the cost is relatively very low, especially compared with some of the more traditional anti-crime programs that governments tend to favor.

HELLER: Although we would love the effects to last forever, so we’re working to figure out what combination of interventions for how long can help the effects last.

One combination of interventions the Chicago programs didn’t try was C.B.T. and cash. Which brings us back to Liberia — to Johnson Borh, the former company commander in the rebel army and, NEPI, the organization he helped set up for other former fighters. And to Chris Blattman, the Columbia researcher who was so impressed with NEPI that he wanted to see it firsthand. Over several years, Blattman went back and forth between the U.S. and Liberia, spending several weeks at a time in Monrovia. He and his collaborators — Julian Jamison, another economist, and Margaret Sheridan, a psychologist — helped Borh and his team refine their curriculum and raise funding. They began recruiting young men who were known troublemakers. And they’d try to teach these men anger management and self-control using the same kind of therapeutic and role-playing exercises used by other C.B.T. programs like Becoming a Man in Chicago.

BLATTMAN: You have to control your anger in the case where someone’s trying to say provoke you, or something nasty happens, or somebody insults you. They might start in the classroom, the very basic, “What’s the baby step you take? Now, I’m going to have you two stand up here, and now you’ve got to insult him.” Right? They know it’s a joke. He comes up with something and everybody laughs. It’s not a real situation of anger, but the guy who’s insulted now has to come up with something he has to say, or explain what he’s going to do, and role-play how he’s not going to react.

Then the group would brainstorm ways to avoid confrontation, diffuse the situation. They’d talk about what works, what doesn’t work. The leaders would also teach the former soldiers seemingly simple things they never learned to do: go to a bank, buy food in a supermarket. Blattman says it was important for these guys to change their self-image — to be able to see themselves as a part of normal society, rather than outcasts.

BLATTMAN: The average guy that we recruited was dressed a little bit rattily.

BORH: They hadn’t cut their hair. They hadn’t cut their nails. They hadn’t take their bath. The only thing they think about is food and drugs.

BLATTMAN: These guys know that stealing is bad. They know that drug dealing is bad, or at least that society considers it a bad thing. They know that drug using is considered bad. They don’t disagree on what the moral principles are in society. They just don’t necessarily believe that that’s their group. “That’s bad for you, but that’s not bad for guys like me.” That was their image. That was their self-image. The goal was to get them to try thinking like mainstream society members.

So as part of the program, the men were encouraged to clean themselves up, wear clean clothes. The program even offered free haircuts.

BLATTMAN: A guy comes in with a razor blade and a chair and will shave off your dreads so that you basically have a shaved head, which is the respectable look in Liberia.

What Blattman wanted to learn was just how much a person’s self-image could change at this point and how effective it would be to teach life skills to adult men. Self-control, for instance. There’s been a lot of research on how self-control develops in childhood – like the famous marshmallow experiments by Walter Mischel. Young children were sat in a room and offered one marshmallow, which they could eat right away. Or, if they chose to wait 15 minutes — alone, in the room, without eating the marshmallow — they’d get to eat two marshmallows, as a reward for their patience. This audio is from a CBS News report.

NEWS REPORT: If you had to choose, would you like to have one marshmallow? Or would you like to have two marshmallows?

BLATTMAN: They look, after 20 years, at these children’s life outcomes. They find that the ones who display more self-control as a preschooler or even younger, are more self-control associated with better jobs, better pay, less crime. When they’ve had programs to try to build these skills, foster these character skills in very young children, like in preschool, those seem to have a big payoff in terms of less crime and better work and happier lives later in life. This is terrific. We know this is really valuable. But there’s been an assumption that adults are no longer malleable, and we need to throw adults under the bus and put all our social spending into preschoolers to have a better future generation rather than say these guys can actually change.

The C.B.T. programs in Chicago proved remarkably effective with criminally-inclined teenagers. How would a similar program work with criminally inclined adult men in Liberia? To find out, Blattman, Borh, and their teams set up a recruiting effort. They approached 1,500 men in Monrovia, aged roughly 15 to 35, who more or less admitted to being involved in criminal activity, or what’s known in Liberia as “the hustle.” Blattman says this could include anything from pickpocketing and robbery to some other kind of scam, often in combination with some legal ways to make money — small jobs, a little farming. Out of the 1,500 men they approached, they got about 1,000 takers. For the sake of the research, Blattman randomized them.

BLATTMAN: They knew that there was going to be a lottery. We explained that some of you are going to get it, some of you are not. We’re going to follow you for over a year because we want to see how this works out. You’re going to play what they call ‘lucky ticket’ in Liberia.

Blattman also wanted to try another element that the Chicago researchers didn’t use, a cash incentive. Some of the participants in the study would be given $200. Now, in Liberia, $200 goes a long way.

BLATTMAN: Two hundred bucks is about three or four months of what they could earn from their so-called hustle. You could think of it like four months of wages. We did it for two reasons. First, cash was actually a way to measure a change in behavior. If you’re more self-disciplined, if you’re more perseverant, if you’re more self-controlled as a result of this therapy, maybe we’ll see it show up in your spending. Maybe you’re more likely to spend it on a small business and less likely to spend it on booze or a meal in the market.

That money — most of which, by the way, was provided by American or other Western aid agencies and foundations — it could also be used as investment capital.

BLATTMAN: Now you can buy a bunch of scratch cards and sell them. Now you can buy things in bulk and sell it at a higher profit. Now you can have a little shoeshine kit.

So here’s how the experiment was set up. Of the roughly 1,000 men who were recruited, one-quarter of them — 250 — wouldn’t get any help at all; they’d become the control group. Another quarter of the men would get C.B.T. One quarter would get C.B.T. and the $200 cash. The last quarter of the men would get just the $200 cash. That way, Blattman could measure which of the treatments, if any, were effective. Johnson Borh was among the first to see the results.

BORH: The results were remarkable. It was very successful.

Criminal behavior fell substantially among all three treatment groups.

BLATTMAN: If you got the cash or if you got the therapy, or if you got both, within a month, you had shaped up. You were committing a lot less crime. You’re going from about one crime a week to half a crime a week, if that’s such a thing. You were less likely to be aggressive, you were less likely to be selling drugs by about half. 20 percent of the control group, I believe, was selling drugs. That was about 10 percent if you received one of the treatments.

Now, you might argue that if you’ve just been given $200, you have less need to sell drugs — but still. These were the results after one month. But what did the results look like after one year? It turns out that the effects had diminished after a year for the men who’d received just therapy or just cash. But, the men who received both therapy and cash were still doing much better.

BLATTMAN: What we found is that these changes in their crime and all of these things we call antisocial behaviors — like aggression, and cheating, and things of this nature, and everyday violence — those were still down a lot, by about 40 or 50 percent in the groups that received both cash and therapy.

As Blattman suspected, receiving a big cash infusion let some of the men start up small businesses.

BLATTMAN: The guy who made the most money went out of town to a distillery, bought a giant barrel of some kind of alcohol, came back, and then sold it by the cupful and made a killing.

And yet, Blattman points out, a lot of the men who received cash, whether with therapy or without, had lost all of it. One reason is that robbery and burglary are pretty common in Liberia. Blattman’s own office, in Monrovia, was broken into, and his computers were stolen. But as Blattman interprets the data from his study, even if men lost their money, they were still better off because they got to adjust their self-image.

BLATTMAN: C.B.T. is about practice. The guys who got the cash got to practice being a businessman, had to put self-control into action by making self-controlled decisions with that money, reinvesting profits. They got to appear to themselves and to the community like normal, peaceful members of society. Even after that was stolen away from them after a few months, it’s like maybe an extra eight or 12 weeks of practice. Maybe it’s like doubling the amount of therapy they had.

Blattman believes this is groundbreaking territory.

BLATTMAN: This is the first time anyone has tried this in a developing country, least of all a fragile state with this kind of population.

One major caveat is that the study relies on self-reported data, which we’ve repeatedly warned you on this show to be leery of. In this case, the participants simply told researchers whether they’d committed a crime or used drugs; and what they spent their money on. Indeed, Blattman was worried the subjects were just telling the researchers what they wanted to hear.

BLATTMAN: When we first saw these results we thought, “No way.” This is just too big. We don’t believe it. No one else is going to believe it.

So as a precaution, Blattman set up a validation group. Researchers followed around some of the subjects, roughly 1 in 12 of the men who were enrolled. They’d hang out for several days, talk with the friends and family. Blattman says they could see no difference between what the men had told the researchers and what they actually did — which makes it easier to believe the data on the men who weren’t followed. Blattman also points out that what we generally consider a more concrete form of data — arrest records from Chicago, for instance — often exists in a sort of gray zone.

BLATTMAN: This is true of any crime study. In the U.S., we say, “We feel better. We have arrest data.” But the one thing maybe events of the last year in the U.S. have really brought to mind is actually the guys who are arrested, that doesn’t necessarily mean you committed a crime. It might mean you looked like the wrong kind of person. You can’t believe that data. You can’t believe our data. They might be telling us what we want to hear. Always take all of these things with a grain of salt. We’re very cautious. But we asked these guys multiple times, building a relationship with trust. Then we went and we like hung out with them as long as we conceivably could without it feeling really weird. [We] found no sense that anything was amiss with what we were learning.

And, again, even if the Liberian study was half as effective as it was, at twice the price, it would still be pretty cheap. It cost about $500 per person.

BLATTMAN: Five hundred dollars per person is for both programs, both therapy and cash, is a bargain compared to most development programs. An average skills-training program usually costs $1,000 or $2,000. The other way to think of it is we saw crimes drop by about, go from about 50 crimes per person to 25 crimes per person. This is basically $20 a crime. If what they were stealing in each crime was worth $20 or more on average, then society is actually better off before we even think about the changes in these guys, other aspects of their lives, like less aggression, less violence.

So even though these C.B.T. studies — in Liberia and Chicago — are relatively tiny, Chris Blattman believes it’s time to consider C.B.T. as a viable tool in preventing crime, especially if — as in both the Liberia and Chicago cases — you can target the biggest troublemakers you can find.

BLATTMAN: We have evidence from Chicago and we have evidence from Liberia, and we have evidence from a lot of incarceration and other types of institutions in the U.S. That’s not enough to say, “Let’s do this everywhere.” But it actually looks better than a lot of things we do. It fits with the way psychologists think about human beings who are pretty similar everywhere. We think we should try this in a couple more places before going gangbusters and scaling it up, which is our plan. But I’m optimistic this has a lot of potential for those 10 or 20 or 30,000 highest-risk men in any given urban center.

Coming up after the break: C.B.T. is not the only new idea being used to try to prevent crime. What might happen if, for instance, you write a simple message on a wall for people who end up in a police holding cell?

Simon RUDA: You have literally a captive audience. Could we write a message which might help an offender think differently about what they’ve done and about who they are?

And, if you’re enjoying this podcast, you should probably just go ahead and subscribe to it, on iTunes or wherever you get your podcasts. You can also help us out by writing a review or giving it a rating. My personal favorite is the five-star rating and the rave review, but, you know, your results may vary.

*      *      *

If you’ve listened to even a handful of our previous episodes, you know we’re fond of discussing clever, simple solutions that are built on a fairly scientific understanding of human behavior. In most cases, that means we patrol the peripheries of academia, maybe private industry. But occasionally you’ll run across a government, or at least a government unit, that also believes in such behavioral insights. What might you call such a unit? How about the Behavioral Insights Team? That’s what they go by in the U.K., where a group of scholars and other clever people have been working for several years now on injecting behavioral science into public policy. You may have heard us mention the Behavioral Insights Team in the past — they’re also known as the Nudge Unit, after the book Nudge by Richard Thaler and Cass Sunstein. It has become a bit of a bible in how behavioral tricks or nudges, can help a government increase the rate of pension take-up, showing up for court …

RUDA: Organ donation, recruitment of soldiers to the army.

That’s Simon Ruda. He is head of Home Affairs, Security and International Development for the Behavioral Insights Team.

RUDA: The remit of the team was really to investigate how some of the nascent findings in the fields of behavioral economics and experimental psychology could be applied to public policymaking in the broadest perspective, to see how we could deliver better results for the country.

Even though the Nudge Unit was personally cheered on by Prime Minister David Cameron, and was set up principally within the walls of Number 10, there was a lot of skepticism from British lawmakers, the media, the general public.

RUDA: But what became apparent quite quickly is that we were able to demonstrate robust impacts of the work that we did — surprising everybody — because we saw quite big positive impacts, positive gains from reasonably low-cost interventions.

One of their early big wins had to do with tax compliance.

RUDA: We demonstrated that making really small and simple changes to the way tax letters were written — tax letters that reminded people that they owed tax to the government — significantly increased the speed at which people re-paid their taxes.

For instance, based on the knowledge that the herd mentality is very prominent among us humans, the Nudge Unit added some very simple language to these letters. It read:

RUDA: “Nine out of ten people in the U.K. have already paid their taxes. You’re one of the few people who haven’t.” That leads to a significant increase in people paying their taxes. When we get into it a little bit more, we see that different messages have different effects for people who owe different amounts of money. For example, if you tell an average population that actually these taxes are really important ‘cause they pay for roads, schools, etc., it doesn’t make that much difference. In fact, I don’t think we found a significant impact at all. But for people with debts of over 30,000 pounds, it makes a big impact.

The Behavioral Insights Team has since become a private company with, as it says, “a social purpose,” though it’s still partly owned by the British government. It has even established satellite units — in Sydney, and, just this summer, in New York. When we were chatting with Simon Ruda, we asked him if he knew of any good work being done using behavioral insights to reduce crime, as we’ve been exploring in Liberia and Chicago. He mentioned a British anti-domestic-violence effort called Project CARA, or Cautioning Against Relationship Abuse. It’s built on the recognition that not all acts of domestic violence are the same:

Heather STRANG: The kind of domestic abuse — domestic violence that attracts a lot of public attention, of course, gives rise to the greatest concern by both police and the public at large — is actually a relatively rare phenomenon.

That’s Heather Strang. She’s research director at Cambridge University’s Institute of Criminology.

STRANG: Only around perhaps two percent of all cases of domestic abuse, domestic violence that comes to police attention is at that high-harm, high-frequency level.

Meanwhile, some 80 percent of domestic-abuse arrests are for what Strang calls a low level of abuse.

STRANG: In about half of those cases, there’s no physical violence at all and nothing in fact to charge an offender about. So we know that there is still a lot of misery and a lot of harm associated with those lower-level incidents, and of course a huge amount of police resources expended on those cases.

But Strang’s point is that the criminal-justice system has tended to not make that distinction when responding to domestic-abuse calls.

STRANG: It’s important to understand when victims call the police in relation to a domestic incident, what is it that they want to happen? The assumption is that they call the police because they want their partner arrested. But we find that only around 28 percent of our victims actually wanted their offender arrested. An even smaller proportion wanted them prosecuted, even though 100 percent of these offenders were arrested because that’s what happens when you call the police. But two-thirds of them simply wanted the police to calm the situation. Almost two-thirds said that their primary reason for calling was actually to get help for their partner rather than punishment for them because all of those two-thirds actually wanted the relationship to continue. That is something that is often overlooked in this field.

In terms of getting help for the partner, judges sometimes do require an offender to get counseling, but that’s just for that two percent of high-level abuse cases that actually make it to court.

STRANG: You have a very small number of cases, usually quite serious cases and usually some very resentful offenders who believe that they are being very hard done by the process that they’ve experienced. They’re not in a very positive frame of mind to undertake this program. They are often quite onerous programs, some of them lasting 26 weeks. Almost nobody ever finishes them and so they’re back in court for failing to complete the program.

Every evaluation of these programs, according to Strang, shows that they haven’t worked; that they haven’t reduced the likelihood that those offenders will commit domestic abuse again. So, a few years back, the police constabulary in Hampshire — that’s a county in southern England — they tried something different. They partnered with a local charity to create some workshops aimed at that much larger group of low-level domestic-abuse offenders.

STRANG: This is an unusual program in many ways because it is not punitive in its orientation. The objectives are to get the participants to understand something about why they behave as they have done, the way in which they use violence in their relationships, and to understand really what emotional abuse is all about. It’s also to get them to recognize their own physical symptoms of anger and the feelings leading to the anger and then onto violence. What we were interested in investigating was the effectiveness of this program in tackling that huge bulk of calls that come to police attention.

So Heather Strang set up a randomized control trial to measure the effectiveness. Here’s how it worked: when British police are called on a domestic-abuse complaint, the offender is generally arrested. Let’s assume, for the sake of pronoun use he is a male, as most offenders are, though not all, of course. Then, as part of this new approach, he’s diverted into what’s called conditional cautioning. That is essentially a promise by offenders that they’ll be on their best behavior.

STRANG: At that point they are randomly assigned either to simply be required to be of good behavior for four months, or they are randomly assigned to be of good behavior for four months, but also undergo this intervention.

This intervention consists of two workshops, about four weeks apart. Each workshop is about five hours, with a break for lunch.

STRANG: It’s on Saturday mornings. They’re held in a hotel because the workshop providers believe that it’s very important to show a respectful attitude to these offenders. It’s restricted only to adult male offenders and to those who admit the offense. They must have no prior record of violence towards victims in the preceding two years.

It’s also restricted to men who’ve abused their intimate partners, not other family members. And the victims have to agree that the treatment won’t put them in further danger. In the beginning, Strang says, the men are reluctant to engage.

STRANG: The facilitators expect this. This is normal. There’s often initially a lot of victim-blaming, a lot of excusing of behavior. [The facilitators] don’t confront it directly. They allow the participants enough space to talk themselves out, to explain why they’ve behaved as they have. Of course, they hear the stories of all the other people in the room as well. By careful questioning by the facilitators about different ways in which they perceive their relationship with their victim operating, rather than any direct confrontation, there is a slow unpacking of what appears to be going on. As an evaluator, we are getting astonishingly good results.

How good is “astonishingly” good? Strang has been looking at 400 domestic-abuse cases. In half of them, at least two years have already passed since the workshops began. Compared to the men who didn’t get the workshops, there was a 40 percent reduction in committing further domestic abuse and there was a 20 percent reduction in other crimes.

STRANG: We have interviewed many of the victims whose cases have been dealt with this way, and they themselves indicate that there has been a marked improvement in the behavior of their partners following the program.

Simon Ruda from the Behavioral Insights Team read through the findings generated by Strang and her colleagues.

RUDA: One woman in particular said, “I don’t know what you’ve done with my husband, but he’s a changed man.”

Heather Strang didn’t necessarily think of the domestic-abuse intervention as cognitive behavioral therapy, or C.B.T., but she does see a lot of commonality in getting people to think about and understand their behavior.

STRANG: Bringing people around to a greater degree of self-knowledge about the way that they’re behaving and the impact of their behavior seems to be key to this program.

RUDA: Helping people to realize what leads to the offense, and the ramifications of the offense, the effect on the offender, can be a far more powerful way of reducing reoffending than just the threat of punishment.

The point, says Ruda, is that people can change.

RUDA: It’s possible that the way we think about ourselves, no matter what stage in life, has a big impact on our behavior.

And on that front, the Behavioral Insights Team is currently planning a new crime-fighting trial in another part of England.

RUDA: Someone that we work with in the police in West Midlands — Chief Superintendent Alex Murray — he came to us. We’ve been working with him on a number of areas. He spotted the opportunity to consider, “How are we best making use of the time that we have with offenders who are in police custody?”

On average, Ruda says, offenders spend about 15 hours in police custody.

RUDA: How are we using that opportunity? You have literally a captive audience. Let’s think about what we can do.

The West Midlands police superintendent had been thinking …

RUDA: He observed that when we hold offenders in police custody they tend to stare at blank walls inside a cell. But what could we write there?

What could they write there? We all know that advertising and marketing, if done right, can profoundly affect our future behavior. So, what kind of message, written on a holding cell wall, might help change an offender’s future behavior? What if that message seemed like something written by one of the previous occupants of the cell?

RUDA: The message might read, “Five years ago on this date I was here, like you are now. This is what I’d done, and this is me looking sad, moody, tough. But this is the impact that my actions were having on those who were close to me. My children, my parents, etc. Then I discovered that if I try to do this thing, whatever it is, and I work at it, it’s gonna make it better.” The story goes on, and the guy does what he does, and all of a sudden he can improve. He can change and life’s much better for him and his family. I thought that might be quite a nice thing to see if we can prompt offenders who have found themselves in that situation to reflect on their actions. To reflect on whether or not they can actually do something a bit differently. Try [to] shift the offender from the mindset of, “I’ve done this action because this is who I am. I am a criminal, therefore this is what I do,” into a mindset of, “This is something that I’ve done but it doesn’t necessarily need to define my life.” This building on some of the work of Carol Dweck, who’s done some excellent work on something she refers to as “growth mindsets.” The hypothesis is that if people really understand how they can work on something and improve themselves, they’re more likely to achieve that than if they believe that they’re made a certain way and that way defines how they behave.

Among the many potential virtues of posting messages like these is the cost, which is very low.

RUDA: We’re already gathering the data. They’re already in the cells. We’re not really investing that much to test the effectiveness of this. But I do believe in the approach: trying to show offenders that actually this isn’t you. This is a behavior. If you want to change it you can.

They’ll be looking to see if messages on cell walls lead to lower rates of re-arrest. If police officers are spending less time pacifying angry prisoners. To that point, Ruda says that projects like this one – which, granted, is a little weirder than the typical government intervention – have not always captured the imagination of senior government officials.

RUDA: A few years ago, we’d have a great idea and we’d go to the relevant ministry and they’d say, “That’s really interesting, but I’m busy.”

But, Ruda says, the reception is generally warmer among police officers and others on the front lines of delivering public-sector services.

RUDA: The appetite was much greater among them because they can see this is going to help them do their jobs. Once you run a trial, all of a sudden, you get some good results. The ministers are then interested.

Ruda believes that behavioral nudges are proving so effective that they might even work in one of the thorniest realms of law enforcement: fighting extremists.

RUDA: There’s this hypothesis around “psychological inoculation.” The hypothesis is that you give younger people — particularly those whose opinions of the world and everything haven’t been formed rigidly yet — ideas of prejudice that are so ridiculous that it’s easy for them to dismiss them. “All women are stupid because they have long hair,” or something ridiculous like that. The hypothesis is that once the cognitive process starts to rebut this and think more logically about it, it becomes more resistant to slightly more sophisticated arguments of prejudice that it may have the displeasure to hear later. This is an interesting angle which is worthy of further pursuit.

In fact, Chris Blattman — the Columbia researcher who works in Liberia — he says he’s heard from people in the U.S. State Department who want to know whether cognitive behavioral therapy might be helpful in countering violent extremism.

BLATTMAN: It’s a slight exaggeration, but if you talk about development project X to the State Department or U.S.A.I.D. or a member of the U.S. government, there’s a 50 percent chance that they say, “Will that work in Afghanistan? Or will that work against terrorism?” Because on some level, right or wrong, this is where a lot of the attention is focused. Anything that says we can change the person, we can essentially re-socialize people who are about violence into having a different identity, that sounds tempting because some of these extremist organizations are essentially offering people a particular self-image. They’re offering them something immaterial that seems to be very valuable. Programs like this, might do that in reverse.

This week and last week, we’ve been looking at new ideas for crime prevention. We all know an ounce of prevention is worth a pound of cure — although in the case of crime, it may be more like a ton of cure. Just think of all the costs associated with crime — for the victim, for society, for the criminal. The costs are financial but also psychic; and they are huge. What’s encouraging about the crime-prevention stories we’ve been hearing — what’s different about these stories — is that these treatments are working on people who’ve already started getting into trouble: juvenile delinquents in Chicago, former child soldiers in Liberia, domestic-abuse offenders in Britain. And the treatment helps turn them around. Most people who do bad things, violent things, presumably aren’t doing them for pleasure. When someone gets in a fight or beats up a spouse or kills someone, it’s generally because they’ve lost control. They’ve made a rash and poor decision. So, can the modern boom in decision science, in behavioral science, help change the way we think about crime itself? I would argue that the evidence we’ve put forth – from Chicago and Liberia and Britain — suggests the answer is yes.

In fact, it suggests that the way we think about crime and punishment generally is perhaps wildly, hopelessly outdated. Like performing a lobotomy on anybody who has a really bad case of the blues. Because of our inability to prevent more crime, especially among people who’ve already committed some, do we routinely turn small-time offenders into lifelong criminals? Have we turned prisons into housing projects? Have we turned our police forces into a de-facto occupying army whose primary remit is to wait for bad things to happen then try to round up the offenders? That’s not what police were originally intended to do in a civilized society. In Britain, in 1829, Sir Richard Mayne wrote Instructions to the New Police of the Metropolis, which begins, “The primary object of an efficient police is the prevention of crime: the next that of detection and punishment of offenders if crime is committed. To these ends, all the efforts of police must be directed.”

As the former London cop Stevyn Colgan writes in a forthcoming memoir, called Why Did the Policeman Cross the Road?: “So why then, I wondered, did my basic training not include any crime prevention? Not one minute of my whole 16-week initial training programme was allocated to the subject. Instead, the syllabus focused exclusively on catching the bad guys or, more precisely, catching the bad guys after the crime had been committed.” All this has made me want to make a future episode – or maybe several, actually – about policing and crime. Working title: “What Are Cops For?” I’d love to hear your ideas on this topic from any angle at all, or any suggestions for what kind of stories we should tell, what kind of data we should pursue. Write to us at radio@freakonomics.com, or give a shout on Twitter. As always, many thanks.