A few years ago, I read a fascinating little book entitled Heavy Drinking: The Myth of Alcoholism as a Disease by philosopher Herbert Fingarette. Drawing on a slew of psychological studies, Fingarette presented a compelling case against the disease model of addiction, including the common claim that the alcoholic cannot control his/her drinking.

Back in January, when I answered a question on the nature of addiction, I wanted to re-acquaint myself with Fingarette’s basic arguments. Happily, I found a fabulous article by him — Alcoholism: the mythical disease — that offers many of the same arguments as the book.

As a philosopher, the issue of most interest to me here concerns free will and responsibility — namely, do “alcoholics” lack control over their drinking? In this article, as well as in the book, Fingarette presents some fascinating empirical evidence on that score. (Since the article is available freely as a PDF, I’ll quote the whole section.) Fingarette writes:

In fact, alcoholics do have substantial control over their drinking, and they do respond to circumstances. Contrary to what the public has been led to believe, this is not disputed by experts. Many studies have described conditions under which diagnosed alcoholics will drink moderately or excessively, or will choose not to drink at all. Far from being driven by an overwhelming “craving,” they turn out to be responsive to common incentives and disincentives, to appeals and arguments, to rules and regulations. Alcohol does not automatically trigger uncontrolled drinking. Resisting our usual appeals and ignoring reasons we consider forceful are not results of alcohol’s chemical effect but of the fact that the heavy drinker has different values, fears, and strategies. Thus, in their usual settings alcoholics behave without concern for what others regard as rational considerations.

But when alcoholics in treatment in a hospital setting, for example, are told that they are not to drink, they typically follow the rule. In some studies they have been informed that alcoholic beverages are available, but that they should abstain. Having decided to cooperate, they voluntarily refrain from drinking. More significantly, it has been reported that the occasional few who cheated nevertheless did not drink to excess but voluntarily limited themselves to a drink or two in order to keep their rule violation from being detected. In short, when what they value is at stake, alcoholics control their drinking accordingly.

Alcoholics have been tested in situations in which they can perform light but boring work to “earn” liquor; their preference is to avoid the boring activity and forgo the additional drinking. When promised money if they drink only moderately, they drink moderately enough to earn the money. When threatened with denial of social privileges if they drink more than a certain amount, they drink moderately, as directed. The list of such experiments is extensive. The conclusions are easily confirmed by carefully observing one’s own heavy-drinking acquaintances, provided one ignores the stereotype of “the alcoholic.”

Some people object that these experiments take place in “protected” settings and are therefore invalid. This gets things backwards. The point is that it is precisely settings, circumstances, and motivations that are the crucial influences on how alcoholics choose to drink. The alcohol per se — either its availability or its actual presence in the person’s system — is not decisive.

Indeed, the alcohol per se or its ready availability seems to be irrelevant to how the alcoholic drinks. Among the most persuasive experiments demonstrating the irrelevance of alcohol to the alcoholic’s drinking are several studies in which alcoholic subjects were deceived about whether they were drinking an alcoholic or nonalcoholic beverage. Alan Marlatt and his colleagues, for example, asked a group of alcoholics to help them “taste-rate” three different brands of the same beverage. Each individual subject was installed in a private room with three large pitchers of beverage, each pitcher supposedly containing a different brand of the same beverage. Their task, of course, was phony. Unknown to them, the subjects had been assigned to one of four groups. One group was told that the beverage in the three pitchers was tonic water — which was true. But a second group was told that the beverage was a tonic-and-vodka mix — though in fact it, too, was pure tonic water. Those in the third group were told that the beverage was tonic-and-vodka — which in fact it was. Those in the fourth group were told that it was simply tonic water — whereas in fact it too was tonic-and-vodka. The subjects were left alone (actually observed through a one-way window) and allowed to “taste” the drinks at will, which they did. The total amount drunk and the rapidity of sips were secretly recorded.

The results of this study (and several similar ones) were illuminating. First, none of the alcoholic subjects drank all the beverage — even though, according to the disease theory, those who were actually drinking vodka ought to have proceeded to drink uncontrollably. Second, all of those who believed they were drinking vodka — whether they really were or had been deceived — drank more and faster. Conversely, all of those who believed they were drinking pure tonic — though some were actually drinking vodka — drank less and more slowly. The inference is unambiguous: the actual presence or absence of alcohol in the system made no difference in the drinking pattern; what the alcoholics believed was in the beverage did make a difference — in fact, all the difference.

These results fit into a more general pattern revealed by similar experiments on other aspects of alcohol-related behavior in both alcoholics and non-alcoholics: change the beliefs about the presence of alcohol (or the effect it is supposed to have), and the behavior changes. But the alcohol itself plays no measurable role.

Mark Keller, one of the early leaders of the alcoholism movement, has responded to such evidence by redefining (or as he would say, “reexplaining”) the key concept of “loss of control.” We are now told that this concept never connoted an automatically induced inability to stop drinking. Like other sophisticated advocates of the disease concept, Keller now means that one “can’t be sure.” The alcoholic who has resolved to stop drinking may or may not stand by his resolution. We are told that “loss of control” is compatible, though unpredictably, with temporary, long-term, or indefinite remission. Here medical terms such as “remission” provide a facade of scientific expertise, but the substance of what we are told is that “loss of control” is consistent with just about anything. This precludes prediction, and of course explains nothing. If it retains any empirical content at all, it amounts to a platitude: someone who for years has relied on a certain way of handling life’s stresses may resolve to change, but he or she “can’t be sure” whether that promise will be fully kept. This is reasonable. But it is not a scientific explanation of an inner process that causes drinking. Similarly, the idea that “craving” causes the alcoholic to drink uncontrollably has been tacitly modified. It was plausible in its original sense, which is still the popular understanding: an inordinately powerful, “overwhelming,” and “irresistible” desire. But the current experimental work regards “mild craving” as a form of “‘craving.” Of course the whole point of “craving” as an explanation of a supposed irresistible compulsion to drink is abandoned here. But the word is retained — and the public is misled.

There have been other adjustments in response to new evidence, designed to retain the “disease” terminology at whatever cost. We now read that “of course alcoholism is an illness that consists of not just one but many diseases, having different forms and causes.” We also hear — in pronouncements addressed to more knowledgeable audiences — that alcoholism is a disease with biological, psychological, social, cultural, economic, and even spiritual dimensions, all of them important. This is a startling amplification of the meaning of “disease,” to the point where it can refer to any human problem. It is an important step toward expanding the medicalization of human problems — a trend that has been deservedly criticized in recent years.

Fascinating, no? If you’re interested in the phenomenon of addiction, check out the whole article! Its other findings may surprise you.

Also, if you’ve not yet heard that 27 January 2013 discussion of the nature of addiction on Philosophy in Action Radio, you can listen to or download the podcast here:

• Duration: 30:48
• Download: MP3 Segment

For more details, check out the question’s archive page. The full episode – where I answered questions on the nature of addiction, unions for government employees, materialism in marriage, mandatory child support, and more – is available as a podcast too.

Note: I published a version of the above commentary in Philosophy in Action’s Newsletter a while back. Subscribe today!