People who disagree about whether certain drugs such as cocaine and heroin should be illegal tend to argue from strong ideological positions. Those who defend a legal ban emphasize the need for society to reinforce weak personal self-control and thereby help preserve the character and dignity of human beings; those who wish the drugs to be legal argue that human freedom must be protected from a government that, despite lacking any power to decide what people read, tries to decide what they shall eat or snort. To reinforce their views, those who support a legal ban say that the costs are worth the reduction in the number of users, while those who object to a ban claim that the costs exceed any possible benefits.

Aware of the two sides in this endless argument, readers may approach Gene Heyman’s new book with suspicion. Heyman, a lecturer in psychology at Harvard Medical School, surely will be endorsing one side or another in this debate. But no: Heyman does not disclose his views on the legal issue; instead, he offers a remarkable book about the extent to which people can choose to use or not to use narcotic drugs. He argues that addiction is not a disease, it is a choice.

This is not the conventional view among many, probably most, researchers, managers of drug treatment programs, and the mass media. The National Institute on Drug Abuse has distributed statements saying that addiction is a disease, and the Diagnostic and Statistical Manual (DSM) of the American Psychiatric Association defines “substance dependence” as “compulsive” drug use that leads a user to continue drug use despite significant drug-related problems. Clinicians usually echo these sentiments. Recent findings that drug addiction is influenced by the genetic makeup of some people strengthen the argument; after all, if genes determine our behavior, then there is nothing we can do about it.

Heyman raises some serious questions about this view. For one thing, the percentage of Americans who become dependent on narcotic drugs has changed dramatically over time. Among people born between 1917 and 1936, less than 1% abused drugs, but among those born between 1952 and 1963, nearly 14% did. If baby boomers are 14 times more likely than Depression-era Americans to become addicts during their lives, then something else must be going on. (By comparison, the two generations are quite alike in the chances of becoming schizophrenic or depressed.)

Another analysis supports this view. The percentage of women who test positive for cocaine use is four times greater in poor inner-city neighborhoods than in smaller urban areas, and that percentage in turn is four times greater than the percentage in poor rural areas. To most people there is nothing surprising about this fact; many would explain it by saying that “of course” impoverished inner-city women are more at risk for drug use, just as they are more at risk for being the victims of a crime. But when we think this, we are admitting that people have a choice: in some places they create a market for drug use, and in other places they do not. Drug abuse, in short, is not an equal opportunity disorder.

From these facts Heyman draws a conclusion: “drug availability and changes in attitudes, values, and perhaps sanctions or perceived sanctions explain the large differences.” But if attitudes and sanctions affect drug use, how can we explain the familiar claims that people in drug treatment programs are rarely if ever cured and that “once an addict, always an addict”? The explanation is easy: these claims are not true.

Heyman draws on three major national surveys to show the falsity of the argument that addiction is a disease. The Epidemiological Catchment Area Study (ECA), done in the early 1980s, surveyed 19,000 people. Among those who had become dependent on drugs by age 24, more than half later reported not a single drug-related symptom. By age 37, roughly 75% reported no drug symptom.

The National Comorbidity Survey (NCS), done in the early 1990s and again in the early 2000s, came to the same conclusion: 74% of the people who had been addicts were now in remission. As with the ECA, the recovery rate was much higher than in the case of psychiatric disorders. The National Epidemiologic Survey on Alcohol and Related Conditions (NESARC), done in the early 2000s with more than 43,000 subjects, came to pretty much the same conclusion.

* * *

Why, then, do so many clinicians say that addiction is a disease? Perhaps, Heyman suggests, it is because they only work with addicts in treatment and are not aware of the life history of drug dependence. And some may think that if addiction is shaped by inherited traits (as it is), then nothing can be done about it.

But virtually every aspect of human behavior has a significant genetic component. Heyman notes that although alcoholism and religious beliefs are importantly shaped by genes, people also retain control over the extent to which genes determine behavior. People at risk for alcoholism will usually not become alcoholic if they live in a dry county or have a spouse who insists on sobriety; people who are genetically predisposed to have fundamentalist beliefs may not join a church or go to meetings where those beliefs are expressed. As I have written elsewhere, genes shape about half of our personality dimensions. This means, for example, that people who are more likely to be neurotic have a steeper hill to climb to avoid that ailment than people who are not at risk. Steeper, but not unconquerable.

Though almost all voluntary activities are shaped by genetics, they are also shaped by the consequences of activities themselves. These consequences, Heyman writes, include benefits, costs, and values. A truly involuntary activity, like a true disease, is elicited by a stimulus, e.g., a bacterial invasion or a body blow: benefits and costs make no difference.

If you believe that addiction is a disease and is not influenced by human choice, then surely it would be cruel and unjust to punish addicts. This would be like scolding, fining, or imprisoning persons for always getting lost because they had Alzheimer’s. Instead, we should presumably require health insurance companies to provide the same coverage for heroin or cocaine addiction as they now do for cancer or heart disease.

Some people take this view under the mistaken impression that people can no more control their addiction than they can prevent the onset of Alzheimer’s. But the data Heyman has gathered shows that most people do overcome addiction. Addiction is not Alzheimer’s.

This, of course, leaves open the question of whether drug dealers should be punished. If we punish a person who sells cocaine, would that not be akin to punishing one who sells alcohol or cigarettes? Answering that question depends on how you compare liquor or nicotine addiction with cocaine addiction, and what you think are the benefits and costs of rendering the trade legal or illegal. Heyman does not address these questions. In my opinion, it would be a grave mistake to add to the number of people who have a destructive addiction to alcohol an additional large number of people addicted to coke, heroin, or methamphetamine.

To be clear on this matter we must decide what proportion of the population we should let become self-destructive addicts, how bad that addiction is for the people who suffer from it, and what the costs and benefits are of enforcing a law against drug abuse. I believe that the abuse of narcotics is destructive of human character, that legalizing certain narcotics would significantly increase the number of addicts, and that the costs of their behavior (on themselves, their families, their job prospects, and their education) outweigh the costs of enforcing laws that keep addiction levels at lower rates than they would be if the drugs were legal.

* * *

Heyman and the reader may or may not agree with my view, but everyone should acknowledge that drug abusers do in fact respond to incentives. One of the most dramatic incentives is being married. Married people are about as likely as single people to have schizophrenia and are more likely to suffer from depression, anxiety, and various phobias. But married people are less than half as likely as single people to be drug addicts. The data do not permit us to say whether marriage discourages addiction or whether addicts have a hard time getting married, but clearly it makes a difference.

The drug treatment programs that work involve managing consequences. A Vermont program reduced addiction by paying people in vouchers if they stayed clean, and Alcoholics Anonymous (AA) helps those who stay in the program acquire a mentor.

People with strong religious beliefs are less likely than atheists to become drug dependent, and more generally addiction is less common when the values of the culture are hostile to it and more common when those values erode. One of the ways society makes its values clear is by making actions against those values illegal and reserving praise for people who act in accordance with them. Though Heyman does not mention it, judicial programs such as Project HOPE in Honolulu that quickly punish arrested drug users with small penalties sharply reduce drug use and the attendant criminality.

Whatever your views on drug use, Heyman’s book is very helpful for understanding the nature of drug addiction.