Skip to main content

The disease concept of addiction revisited

June 15, 2010

Despite the fact that The Disease Concept of Alcoholism by E.M. Jellinek was published 50 years ago, and although the idea that addiction is a disease is now widely accepted, it remains poorly understood. The latest contribution to the confusion is a book published last June called Addiction: A Disorder of Choice by Gene M. Heyman of Harvard University. It’s an ironic recapitulation of history—20 years earlier, Herbert Fingarette of the University of California advanced essentially the same argument in his book Heavy Drinking: The Myth of Alcoholism as a Disease. Both men based their conclusions on epidemiologic studies and surveys. Fingarette, a philosopher, argued solely from his reading of the scientific literature. Heyman, an academic psychologist, based his conclusions on epidemiologic surveys and laboratory research on the psychology of choice. It seems that neither writer spent much time in the clinical trenches, actually listening to alcoholics and drug addicts describe their lives.

It seems a fitting moment to re-examine the disease concept of alcoholism and other addictions.

To start, scientific data can no more “prove” that addiction is a disease than it can “prove” that the sky is blue. Either we all agree that the color of the sky is sufficiently like everything else we call “blue,” or we agree to call it something else. In the same way, asserting that addiction is a disease cannot be proven by scientific data. A disease concept is really a theory of addiction—a way of showing that addiction is like all the other things we generally accept as diseases.

Although it may sound strange, when we say that alcoholism or drug addiction is a disease, we are not talking about the behavior of drinking or using. Behavior might signify the presence of a disease, but behavior itself cannot be a disease. A disease isn’t something you do (voluntarily or otherwise); it’s something you have. The common sense inherent in our language reflects this same idea. We don’t speak of someone “high blood pressure-ing” or “pneumonia-ing.” We say a person has high blood pressure or has pneumonia. This is true for all diseases. The behavior we call a “seizure,” for example, might indicate an infection, a hemorrhage or a tumor in the brain. The seizure is the sign of a disease, not the disease itself.

If the behavior of drinking or using drugs is only the sign of addiction, then it is no surprise that measuring drinking or using behavior brings no uniform picture of the disorder. In virtually all illnesses, especially early in their course, signs and symptoms are remarkably variable. Just as fevers may be high or low, pain severe or mild, alcoholic drinking or addictive drug use may be heavy or light, intermittent or continuous, boisterous or quiet—all depending on biological, social and psychological factors influencing the individual with the disorder.

So if by calling addiction a disease we mean that sometimes drinking or using is a sign of something else, a result of something a person has, then we need to be clear about what that something is. Without a simple conception of what an addiction is (on par, for example, with what an infection is), we have no strong argument for the disease concept of addiction.

The experience of ‘powerlessness’
Part of the difficulty in establishing the disease concept of addiction is that the essence of the condition is known to us primarily through the reported experience of the person who has it. Although advances in brain imaging have begun to show us the disordered biochemistry underlying addiction, diagnosis is still based mostly on what patients tell us about their experience. As a result, the data are largely subjective and can be quantified “objectively” only indirectly. That’s why it is so important to listen carefully to the stories of alcoholics and addicts themselves—to hear what they say about what’s going on inside them. When we do that, we learn that they describe their experience as “powerlessness.”

But the idea of powerlessness is paradoxical. After all, many alcoholics and addicts do quit drinking and using for good. What happened to powerlessness, then? Choosing, “one day at a time,” not to drink or use sounds like having power, not like having lost it.

More than 100 years ago, in describing his own struggles with tobacco, Mark Twain gave us the solution to this puzzle when he said: “To cease smoking is the easiest thing I ever did. I ought to know because I’ve done it a thousand times.” Twain put his finger directly on the essential experience of addiction—when it is fully developed, it is an all-or-nothing experience. Although addictive behavior is remarkably varied, in the end virtually all addicts discover that abstinence is the only reliable foundation for recovery. “Quitting,” it turns out, is hard, but it isn’t the major problem. The bigger problem, brilliantly expressed in AA, is “staying quit—not starting again.

Like many other illnesses, addictions progress. The beginning is marked by the struggle for control (“never before 5 p.m.,” “only on weekends,” and so on). But as time goes on, control becomes increasingly difficult to achieve. Eventually, it is attained only by quitting. Indeed, episodes of quitting and relapsing are almost an unmistakable indication of the diagnosis of addiction (as opposed to mere abuse or misuse). At this stage, if people begin again, they spend increasing amounts of time and effort trying to maintain “control,” but in the end it is lost. This is what AA’s founders described as having become “powerless” over alcohol—not just for a particular episode of drinking, but repeatedly and inevitably for all drinking. In the end, all addicts discover that there is just no such thing as one.

Now comes another paradox. I call it the “control conundrum.” Non-alcoholics don’t have to work at controlling their drinking. Whether they are stronger or better people (as they sometimes like to think) is a matter of opinion. The fact remains that a normal drinker doesn’t have to struggle to control it. So here’s the puzzle: How can an illness be characterized by the loss of control when the healthy state of affairs is experienced as not needing control? How can an alcoholic lose what a non-alcoholic doesn’t have in the first place? The answer is that an addiction isn’t so much a loss of something as it is the development of something that has a life of its own, that takes control for itself. That something is an “automatism.”

Automaticity: toward a theory of addiction
The answer to all the confusion over the disease concept of addiction lies in conceiving of addictions as “automatisms,” or disorders of automaticity. Automatisms are permanent developments in the central nervous system. Some are instinctive (the control of breathing and heart rate, for example), and belong to what we call the autonomic nervous system. But others are acquired as we mature (standing, walking, talking) and become permanent parts of the voluntary nervous system. All such automatisms have two things in common. First, they are irreversible. Second, once they have been initiated, they function outside conscious control.

Swimming offers an excellent example of a common automatism. Once you “get” swimming, you can never go back to being a non-swimmer. Stay out of the water for 50 years, fall off the end of a boat dock, and watch what happens. Automatically, whether you intend to or not, you’ll swim. Of course, since you don’t want to drown, that’s a good thing. But suppose that for some reason it became extremely important for you never to swim again (dangerous currents, sharks, motor boats). What choice do you actually have? Since you cannot forget or “unlearn” being a swimmer, you literally cannot choose not to swim. Your only choice is to stay out of the water.

True, a swimmer might try to enjoy the water but avoid swimming by staying in the shallows. This would be analogous to the “setting limits” stage of an addiction. The problem is that it just isn’t going to work in the long run. Sooner or later, whether or not the swimmer intended to, if he’s back in the water, his feet will leave the bottom and he’ll be swimming again—automatically. As long as he doesn’t drift out into the deep water, there may well be no problems. But that isn’t the point. The point is that despite having resolved not to swim again, he is.

In order to succeed in controlling an automatism, one must become abstinent. For a swimmer, that means staying out of the water. For an alcoholic or drug addict, it means not drinking or using.

I am not saying that all automatisms are necessarily harmful. On the contrary, our lives are filled with, and indeed made possible by, automaticity. Complex automatisms—coordinated physical movements and speech, for example—free our attention for higher levels of consciousness and choice. But that freedom comes at a cost—the loss of choice at a lower level of consciousness—experienced as powerlessness.

Again, an example will be more useful than more description: DO NOT READ THIS!

In order to experience the meaning of the phrase (and the humor), you have sacrificed choice at the level of recognition—if you look at it, you cannot not read it. Had I written instead, “Ne olvasd ezt!”, chances are pretty good you wouldn’t have understood what a Hungarian would have found so amusing. You don’t have that automatism. In that case, you would be rather like a non-alcoholic drinker who doesn’t “get” why alcoholics simply don’t control themselves better.

Addictions, then, are complex automatisms, involving the progressive automatization of feelings (urges), thoughts (obsessions) and actions (behavior). In the end, an addiction becomes Mark Twain’s all-or-nothing experience of loss of control or powerlessness. Something inside has acquired a life of its own. When that something threatens the well-being of the whole (just like runaway blood pressure), then it is rightly considered a disease.

Is addiction like other diseases?
If this conception of addictive disease fits the class of events called “illness,” it shouldn't have to be forced into place. It should fit in the same the way other, well-accepted conditions do (the “Is the sky blue?” test). Does it?

In every disease, there is an agent of harm: a hostile germ, a defective protein, an abnormal growth that disrupts the harmonious balance of physiologic and psychological functions. Alcohol, cocaine, heroin, nicotine, etc., certainly fit that definition. They are neurotoxins—nerve poisons. But mere exposure to an agent of harm doesn’t always lead to illness. Most of us are exposed to potentially harmful germs all the time. We don’t become sick because our immune systems fight those germs off. Similarly, many people are exposed to alcohol and drugs, but many of them have resistance to becoming addicted. Many different factors combine to provide this protection, and, as with other illnesses, some people are more vulnerable to developing the disease when exposed to an agent of harm of this kind. Just as the failure of resistance to a germ is a consequence of biological (hereditary and acquired), social and psychological factors, so too is the development of an addiction shaped by heredity, biochemical effects of the toxin, social conditions (availability, cultural expectations), and co-existing psychopathology.

What about recovery and treatment? Is addiction like other diseases? The argument has been made that because many people recover from addictions without professional help, these problems should not be called diseases. The word “disease,” it is said, should be reserved for conditions that require medical treatment. But this is a narrow view. First, we recover from all kinds of illnesses (mild and serious) without professional help. Second, and equally important, an enormous proportion of modern healthcare services result at least in part from drug and alcohol use (cirrhosis, trauma, emphysema, AIDS, etc.). To treat only the consequences of addictions without attending to the underlying cause is very short-sighted.

Where addictions are different from other illnesses is in the degree of effort required for recovery. In this respect, all illnesses are on a spectrum. At one end (an infection, for example), the degree to which the patient has to work at recovery is relatively small. The body more or less does the work on its own. In the middle of the spectrum is something like having broken a leg. Unless there are complications, the body will heal the fracture, but if the patient will not work at rehabilitation, full function of the limb may not be restored. Addictions are at the other end of the spectrum, the end where almost no recovery takes place by itself. Here, long-term recovery requires the willingness to remain abstinent no matter what. That’s hard work.

Dangers of the disease concept
Are there dangers in the disease model of addiction? Yes, and in this, valid criticism of the medical perspective must be acknowledged. Addictions are chronic conditions in which the capacity for and exercise of choice play the major role in recovery. Like other patients who have lost control of a part of themselves (e.g., becoming paralyzed after a stroke), alcoholics and addicts must not only want to recover, they must be willing and able to work at recovery and rehabilitation. After 30 years of clinical experience with alcoholics and addicts, I believe that is best accomplished if they will become a dedicated, active member of a 12-Step group. Indeed, I conceive of my role and the role of treatment programs as helping people overcome obstacles to doing just that. There are many different kinds of obstacles: physiologic (withdrawal), psychological (denial, co-existing psychopathology), social (family dysfunction, unemployment), and treatment needs to address all of them. But the ultimate goal, to my way of thinking, is to help our patients find their way to a 12-Step group—not to substitute for it.

If the disease concept of addiction gives alcoholics and addicts the message that treatment providers are somehow going to do the work of rehabilitation for them, then it does them great harm. On the other hand, if the concept of addictions as disorders of automaticity clarifies the question of who is responsible for what, then it can be very helpful. The whole question can be summed up pretty simply, and the summary is consistent with illness in general: No one is responsible for having become sick, but everyone who has an illness is responsible for doing what they can to recover from it. It is an error, and a potentially harmful one, to call addiction a “chronic, relapsing brain disease.” Addictions do not relapse of themselves. People who have addictions relapse. That’s why they also can recover.

The terrible problem for the recovering addict is revealed in the first noble truth of Buddhism: “Life is suffering.” Put simply, none of us, recovering addict or “normie,” wants to suffer. Nevertheless, some suffering is inevitable. All of us strive to avoid it when we can and to escape it as quickly as possible when we cannot. But alcoholics and drug addicts have the extra burden of knowing precisely how to get rid of pain—drink or use. After a lifetime of changing their state of consciousness at the drop of a hat, the alcoholic or drug addict must become willing to experience “life on life’s terms—not because it’s morally better, but because it’s the only viable alternative to a path that leads to relapse.

This is why participation in one of the 12-Step groups, with their emphasis on spirituality, is so important. However one conceives of a Higher Power (the group, humanity, nature, God, whatever), without a sense of something greater than myself to which I am responsible, there is simply no reason to endure the pain inherent in living, let alone recovery. In this also, addiction is just like other illnesses. As the great French surgeon Ambroise Paré said nearly 500 years ago, “I merely dress the wound. God heals it.” There’s something in that lesson for all of us.

Richard S. Sandor, MD, practices psychiatry and addiction medicine in Santa Monica, Calif. He is a past president of the California Society of Addiction Medicine and is the author of Thinking Simply About Addiction (March 2009, Tarcher/Penguin). His e-mail address is

Back to Top