David and Goliath: The Triumph of the Underdog

by Malcolm Gladwell

  Reynolds and his group collected thousands of signatures to qualify for a statewide referendum. There are countless referendum ideas in every California election season, and most never see the light of day. But Three Strikes struck a nerve. It passed with the support of an astonishing 72 percent of the state’s voters, and in the spring of 1994, Three Strikes was signed into law, almost word for word the way it was written up in Mike Reynolds’s backyard. The criminologist Franklin Zimring called it “the largest penal experiment in American history.” There were eighty thousand people behind bars in California’s prisons in 1989. Within ten years, that number would double—and along the way, the crime rate in California came tumbling down. Between 1994 and 1998, the homicide rate in California dropped 41.4 percent, rape dropped 10.9 percent, robbery dropped by 38.7 percent, assault dropped by 22.1 percent, burglary dropped by 29.9 percent, and auto theft dropped by 36.6 percent. Mike Reynolds pledged, on his daughter’s deathbed, to ensure that what happened to Kimber would never happen to anyone else—and out of his grief came a revolution.

  “Back then, we were seeing twelve murders a day in the state of California. Today it’s about six,” Reynolds said. “So every day that goes by, I like to think that there’s six people alive that wouldn’t have been prior to this.” He was sitting in the office of his house in Fresno, surrounded by pictures of himself with dignitaries of one sort or another and plaques and signed certificates and framed letters—all testifying to the extraordinary role he has played in the politics of America’s largest state. “Every once in a while during the course of your life, you might have an opportunity to save somebody else’s life,” he went on. “You know, pull ’em out of a burning building, rescue ’em from drowning or some other crazy thing. But how many people get a chance to save six people’s lives each and every day? I mean, I think, I’m so lucky.”

  He paused, as if he were going back over all that had happened in the nearly twenty years since he made that promise to Kimber. He was remarkably articulate and persuasive. It was obvious how, even in the midst of overwhelming grief, he would have been so compelling all those years ago on the Ray Appleton show. He started up again: “Think about the guy that invented safety belts. Do you know his name? I don’t. I’ve got no clue. But think about how many guys that are safe, or people that are safe, as a result of safety belts or air bags or tamper-proof medicine containers. I could sit here and go right through it. Simple devices that are made by Joe Average, just like me, that have gone on to save numerous lives. Yet we’re not looking for any kudos, we’re not looking for any pats on the back. All we’re looking for is results, and the results are my greatest reward.”

  The British came to Northern Ireland with the best of intentions and ended up in the middle of thirty years of bloodshed and mayhem. They did not get what they wanted, because they did not understand that power has an important limitation. It has to be seen as legitimate, or else its use has the opposite of its intended effect. Mike Reynolds came to wield extraordinary influence in his home state. There are few other Californians of his generation whose actions and ideas have touched as many people as his have. But in his case, power seemed to have achieved its purpose. Just look at the California crime statistics. He got what he wanted, didn’t he?

  Nothing could be further from the truth.

  3.

  Let us go back to the theory of the inverted-U curve that we discussed in the chapter on class size. Inverted-U curves are all about limits. They illustrate the fact that “more” is not always better; there comes a point, in fact, when the extra resources that the powerful think of as their greatest advantage only serve to make things worse. The inverted-U shape clearly describes the effects of class size, and it clearly applies as well to the connection between parenting and wealth. But a few years ago, a number of scholars began to make a more ambitious argument, an argument that would end up pulling Mike Reynolds and his claims for Three Strikes into the center of two decades of controversy. What if the relationship between punishment and crime was also an inverted U? In other words, what if—past a certain point—cracking down on crime stopped having any effect on criminals and maybe even started to make crime worse?

  At the time Three Strikes was passed, no one considered this possibility. Mike Reynolds and his supporters assumed that every extra criminal they locked up, and every extra year they added to the average sentence, would bring about a corresponding decrease in crime.

  “Back then, even first-degree murder was just sixteen years, and you’d do eight,” Mike Reynolds explained. He was describing California before his Three Strikes revolution. “It became a very viable option to go into the crime business. The human psyche follows the course of least resistance. The course of least resistance is what’s easy, and it’s a hell of a lot easier to go out and rob and steal and suck drugs than it is to go out and bust your ass forty hours a week and punch in on a job and take a lot of shit off customers. Who needs that? I can go out there and wave a gun around and make as much as I want as fast as I want, and if I get caught, ninety-five percent of all cases get plea-bargained down. They charge me with this, I’ll admit to that, and so let’s make a deal. And then third, I’m going to only serve half the time. Weigh all three, the odds are you’re going to do one hell of a lot of crime before you ever in fact get caught and prosecuted.”

  Reynolds was making a version of the argument that Leites and Wolf made in their classic work on deterrence: Fundamental to our analysis is the assumption that the population, as individuals or groups, behaves “rationally,” that it calculates costs and benefits to the extent that they can be related to different courses of action, and makes choices accordingly. In Reynolds’s view, criminals found the benefits of committing a crime in California much greater than the risks. The answer, he felt, was to raise the costs of committing a crime so high that it was no longer easier to rob and steal than to work an honest job. And for those who continued to break the law—even in the face of those altered odds—Three Strikes said, Lock them up for the rest of their lives, so they never have a chance to commit another crime again. When it came to law and order, Reynolds and the voters of California believed, “more” was always better.

  But is it? Here’s where the inverted-U theorist steps in. Let’s start with the first assumption—that criminals respond to increases in the cost of crime by committing fewer crimes. This is clearly true when the penalties for breaking the law are really low. One of the best known case studies in criminology is about what happened in the fall of 1969 when the Montreal police went on strike for sixteen hours. Montreal was—and still is—a world-class city in a country that is considered one of the most law-abiding and stable in the world. So, what happened? Chaos. There were so many bank robberies that day—in broad daylight—that virtually every bank in the city had to close. Looters descended on downtown Montreal, smashing windows. Most shocking of all, a long-standing dispute between the city’s taxi drivers and a local car service called Murray Hill Limousine Service over the right to pick up passengers from the airport exploded into violence, as if the two sides were warring principalities in medieval Europe. The taxi drivers descended on Murray Hill with gasoline bombs. Murray Hill’s security guards opened fire. The taxi drivers then set a bus on fire and sent it crashing through the locked doors of the Murray Hill garage. This is Canada we’re talking about. As soon as the police returned to work, however, order was restored. The threat of arrest and punishment worked.

  Clearly, then, there’s a big difference between having no penalties for breaking the law and having some penalties—just as there’s a big difference between a class of forty students and a class of twenty-five. On the left side of the inverted-U curve, interventions make a difference.

  But remember, the logic of the inverted-U curve is that the same strategies that work really well at first stop working past a certain point, and that’s exactly what many criminologists argue happens with punishment.

  Some years ago,
for example, the criminologists Richard Wright and Scott Decker interviewed eighty-six convicted armed robbers. Most of what they heard were comments like this:

  I put forth an effort to try not to think about [getting caught.…It’s] too much of a distraction. You can’t concentrate on doing anything if you are thinking, “What’s gonna happen if it doesn’t go right?” As time went on, if I had made up my mind to do a robbery, [I decided] to be totally focused on that and nothing else.

  Or this:

  That’s why [my partners and I] get high so much. [We] get high and get stupid, then we don’t trip off of [the threat of getting caught]. Whatever happens, happens.…You just don’t care at the time.

  Even when pressed, the criminals interviewed by Decker and Wright “remained indifferent to threatened sanctions.” They just weren’t thinking that far ahead.

  The murder of his daughter made Reynolds want to put the fear of God into California’s would-be criminals—to make them think twice before crossing the line. But that strategy doesn’t work if criminals think like this. Joe Davis and Douglas Walker—the two thugs who cornered Kimber Reynolds outside the Daily Planet—were crystal-meth addicts. Earlier that day, they had attempted a carjacking in broad daylight. And remember what Walker said? I wasn’t really thinking much a nothing, you know. When it happens, it happens, you know. It just happened suddenly. We were just out doing what we do. I mean, that’s all I can tell you. Is this the sort of person to think twice?

  “I’ve talked to family friends who knew Joe and his brother, and they asked him why he shot Kimber,” Reynolds once said, looking back on that tragic evening. “And he said that he already had the purse, so that wasn’t an issue. But that he’d shot her, instead, because of the way she was looking at him. He shot her because he didn’t think she was taking him seriously, and wasn’t giving him any respect.” Reynolds’s own words contradict the logic of Three Strikes. Joe Davis killed Kimber Reynolds because she would not give him the respect he thought he deserved as he held a gun to her head and grabbed at her purse. How on earth does changing the severity of punishment deter someone whose brain works like that? You and I are sensitive to increased punishment, because you and I are people with a stake in society. But criminals aren’t. As the criminologist David Kennedy writes: “It may simply be that those who stand ready today to take a chance, often on impulse, often while impaired, on what they view as a very small likelihood of an already very serious sanction will stand ready tomorrow to take the same chance on what they still view as a very small likelihood of a somewhat more serious sanction.”2

  The second argument for Three Strikes—that every extra year a criminal is behind bars is another year he can’t commit a crime—is just as problematic. The math doesn’t add up. The average age of a California criminal in 2011 at the moment he was convicted of his Third Strike offense, for example, was forty-three. Before Three Strikes came along, that man might have served something like five years for a typical felony and been released at the age of forty-eight. With Three Strikes, he would serve, at minimum, twenty-five years—and get out at sixty-eight. Logically, the question to ask is: How many crimes do criminals commit between the ages of forty-eight and sixty-eight? Not that many. Take a look at the following graphs, which show the relationship between age and crime both for aggravated assault and murder and for robbery and burglary.

  Longer sentences work on young men. But once someone passes that crucial midtwenties mark, all longer sentences do is protect us from dangerous criminals at the point that they become less dangerous. Once again, what starts out as a promising strategy stops working.

  Now for the crucial question: Is there a right side to the crime-and-punishment curve—a point where cracking down starts to actually make things worse? The criminologist who has made this argument most persuasively is Todd Clear, and his reasoning goes something like this:

  Prison has a direct effect on crime: it puts a bad person behind bars, where he can’t victimize anyone else. But it also has an indirect effect on crime, in that it affects all the people with whom that criminal comes into contact. A very high number of the men who get sent to prison, for example, are fathers. (One-fourth of juveniles convicted of crimes have children.) And the effect on a child of having a father sent away to prison is devastating. Some criminals are lousy fathers: abusive, volatile, absent. But many are not. Their earnings—both from crime and legal jobs—help support their families. For a child, losing a father to prison is an undesirable difficulty. Having a parent incarcerated increases a child’s chances of juvenile delinquency between 300 and 400 percent; it increases the odds of a serious psychiatric disorder by 250 percent.

  Once the criminal has served his time, he returns to his old neighborhood. There’s a good chance he’s been psychologically damaged by his time behind bars. His employment prospects have plummeted. While in prison, he’s lost many of his noncriminal friends and replaced them with fellow-criminal friends. And now he’s back, placing even more strain emotionally and financially on the home that he shattered by leaving in the first place. Incarceration creates collateral damage. In most cases, the harm done by imprisonment is smaller than the benefits; we’re still better off for putting people behind bars. But Clear’s point is that if you lock up too many people for too long, the collateral damage starts to outweigh the benefit.3

  Clear and a colleague—Dina Rose—tested his hypothesis in Tallahassee, Florida.4 They went across the city and compared the number of people sent to prison in a given neighborhood in one year with the crime rate in that same neighborhood the following year—and tried to estimate, mathematically, if there was a point where the inverted-U curve starts to turn. They found it. “If more than two percent of the neighborhood goes to prison,” Clear concluded, “the effect on crime starts to reverse.”

  This is what Jaffe was talking about in Brownsville. The damage she was trying to repair with her hugs and turkeys wasn’t caused by an absence of law and order. It was caused by too much law and order: so many fathers and brothers and cousins in prison that people in the neighborhood had come to see the law as their enemy. Brownsville was on the right side of the inverted U. In California in 1989, there were seventy-six thousand people behind bars. Ten years later, largely because of Three Strikes, that number had more than doubled. On a per capita basis, by the turn of the twenty-first century, California had between five and eight times as many people in prison as did Canada or Western Europe. Don’t you think it’s possible that Three Strikes turned some neighborhoods in California into the equivalent of Brownsville?

  Reynolds is convinced that his crusade saved six lives a day, because crime rates came tumbling down in California after Three Strikes was passed. But upon closer examination, it turns out that those reductions started before Three Strikes went into effect. And while crime rates came tumbling down in California in the 1990s, they also came tumbling down in many other parts of the United States in the same period, even in places that didn’t crack down on crime at all. The more Three Strikes was studied, the more elusive its effects were seen to be. Some criminologists concluded that it did lower crime. Others said that it worked but that the money spent on locking criminals up would have been better spent elsewhere. One recent study says that Three Strikes brought down the overall level of crime but, paradoxically, increased the number of violent crimes. Perhaps the largest group of studies can find no effect at all, and there is even a set of studies that argue that Three Strikes raised crime rates.5 The state of California conducted the greatest penal experiment in American history, and after twenty years and tens of billions of dollars, nobody could ascertain whether that experiment did any good.6 In November of 2012, California finally gave up. In a state referendum, the law was radically scaled back.7

  4.

  Wilma Derksen was at home, trying to clean up the family room in the basement, when her daughter Candace called. It was a Friday afternoon in November, a decade before Kimber Reynolds walked out of he
r parents’ home for the last time. The Derksens lived in Winnipeg, Manitoba, on the prairies of central Canada, and at that time of year, the temperature outside was well below freezing. Candace was thirteen. She was giggling, flirting with a young boy from her school. She wanted her mother to come and pick her up. Wilma did a series of calculations in her head. The Derksens had one car. Wilma had to pick up her husband, Cliff, from work. But he wouldn’t be finished for another hour. She had two other children—a two-year-old and a nine-year-old. She could hear them quarreling in the other room. She would have to bundle them up first, pick up Candace, then go and pick up her husband. It would be an hour in the car with three hungry children. There was a bus. Candace was thirteen, no longer a child. The house was a mess.

  “Candace, do you have money for the bus?”

  “Yup.”

  “I can’t pick you up,” her mother said.

  Derksen returned to her vacuuming. She folded laundry. She bustled about. Then she stopped. Something seemed wrong. She looked at the clock. Candace should have been home by now. The weather outside had suddenly turned colder. It was snowing. She remembered that Candace hadn’t dressed warmly. She began to pace between the window in the front of the house and the kitchen window in the back overlooking the alleyway. Candace might come in from either direction. The minutes passed. It was time to pick up her husband. She packed up her other two children, got in the car, and drove slowly along Talbot Avenue, the road that connected the Derksens’ neighborhood to Candace’s school. She peered inside the windows of the 7-Eleven, where her daughter sometimes lingered. She drove to the school. The doors were locked. “Mom, where is she?” her nine-year-old daughter asked. They drove to Cliff’s office.