I generally avoid writing book reviews because most publications offer very little space and I seem to get assigned books that almost no one would ever want to read. The following certainly fell into that category, but it seemed just pretentious enough, and foolish enough, to warrant the ridiculous amount of work it took. And this journal gave me more space. It appeared in

Addiction (formerly the British Journal of Addiction)
Vol. 95 No. 3; p. 449-452
March 2000

This is a strange book, and a hard one to get a handle on. The one-paragraph Preface explains that the book came out "of a Working Group on Addiction that met annually from 1992 to 1997", and that it was funded by the Norwegian Institute for Alcohol and Drug Research, the Norwegian Directorate for the Prevention of Alcohol and Drug Problems and the Russell Sage Foundation. That's it. From the contributors page, one learns that five of the 10 authors are American and five are Norwegian, including both editors. There is no other information about how the group came together or what they were trying to do.

The first page of the Introduction by Jon Elster & Ole-Jorgen Skog explains that addiction raises the question of "why people engage in behaviors that they know will harm them." It also says that this book seeks the answers in what it calls "theoretical" writings. The contributors come from a variety of fields, although rational choice economics, analytical philosophy and some schools of psychiatry appear to be their primary professional and conceptual vocabularies.

After reading the book all the way through it becomes clear that the volume is above all an extended response to the argument of economist Gary Becker and his co-author Kevin Murphy, that addiction is rational behavior; but Elster & Skog never say this. Nor do they say that Becker is a leading proponent of Chicago School neo-classical economics. From the references one learns that Becker & Murphy made their case in an economics journal article in 1988. Becker's argument is one of a number of theories that are reviewed briefly in the introduction. It then reappears repeatedly in the chapters by different authors, coming from very different perspectives. By the end it becomes clear that responding to Becker's argument that addiction is rational behavior has been the dominant intellectual motif in the book -- and the reason for the subtitle.

The book touches on some other themes, and one of them is about the importance of brain chemistry for understanding addiction. Elster is a professor of Political Science, and Skog is a professor of Sociology and Human Geography. The last paragraph of their introductory chapter says:

"We conclude by stating our view that the future of addiction studies is likely to be heavily shaped by developments in brain chemistry. The most promising line of research might involve the integration of neurobiological and decision-theoretic approaches. Although this general idea is now steadily gaining importance, it has not been extensively applied to the analysis of addiction ... The neurobiological study of addiction, like similar studies of emotion, promises to be a key to progress in understanding of the mind-body problem."

I found this to be a peculiar way for two social scientists to end the introduction to their book about addiction. Elster & Skog have no apparent expertise in brain chemistry or neurobiology. Nor do they explain why we should accept their opinion that brain chemistry may offer "the most promising line of research" in addiction, even when coupled with their preferred type of economic theory. They might be right of course, but why should anyone think so?

The only chapter that discusses brain chemistry at length is "The Neurobiology of Chemical Addiction" by Eliot L. Gardner & James David, professors of Psychiatry in New York City. This is the longest chapter in the book and, paradoxically, it suggests a less optimistic view of the capacity of neurobiology to explain relapses and episodic bingeing, and therefore to explain "why people engage in behaviors that they know will harm them." Gardner & David write an up-to-date summary of findings from neurobiological research, mainly laboratory studies of rats with electrodes wired to their brains and rats given drugs. Their brief conclusion compares "addicted humans with addicted laboratory animals."

Gardner & David say that at the moments when addicted human beings are actually using drugs they "remind us of our experimental animals." However, the authors suggest that men and women least resemble laboratory animals when they desire or crave a drug after not having used it for a while. "The role of episodic craving in relapse is increasingly familiar to both clinicians and researchers" the authors write, "yet the neurobiological correlates of this craving remain obscure and unexplored." In other words, neurobiological research on addiction can not explain the behavior of men and women when a drug has not been in their brain for a while. Therefore, neurobiology helps little in accounting for abstinence, relapse or periodic binges of heavy drug use -- three fairly important phenomena among human addicts.

Brain chemistry does not figure prominently in any of the other chapters, but micro-economic theories do. Two of the chapters, in fact, rely heavily on econometric formulas. They also discuss Becker's article a great deal, which itself relies heavily on such formulas. The chapter by Karl Ove Moene, a Professor of Economics in Oslo, entitled "Addiction and Social Interaction", has equations on nearly every page. The chapter ends with one paragraph entitled "Conclusions" that says:

"The main message of this paper is that history matters for total drug use in a society. The simple model that we have explored identifies spillovers and ratchets in drug use that cannot so easily be captured in the theory of rational addiction by Becker and others."

In short, Moene's equations lead him to conclude that Becker's theory is inadequate for comprehending addiction.

Ole-Jorgen Skog's chapter, "Rationality, Irrationality, and Addiction--notes on Becker and Murphy's theory of addiction," also employs many formulas, graphs with curves and other econometric and statistical tools. His two-page conclusion includes the following comments:

"Becker and Murphy have obviously succeeded in proving one quite basic point: that addiction is not necessarily incompatible with rationality ... Nonetheless, it is not at all obvious that the state described by Becker and Murphy's theory is what addiction is all about. Nor is it clear that real addicts are in fact rational in the strict sense ... I can see no good reason to doubt that people -- including people suffering from an addiction -- by and large try to make the best of it given the circumstances...."

Skog then ends with a poignant personal story.

"My own experience as a former heavy cigarette smoker ... is best described by the term 'unstable preferences.' Over and over again I relapsed, giving in to short-term motives, over-shadowing the long-term motives. As Ainslie implies, one has to struggle with oneself to become dynamically consistent -- it is not a gift given to us at birth."

Skog appears to be using his own experience with the quite common problem of relapse to further show the limitations of Becker's idea of rational addiction.

The only chapter focusing on policy is by Helge Waal, a Professor of Psychiatry in Oslo, entitled "To Legalize or Not to Legalize: is that the question?" Waal is against legalizing the currently illegal drugs, and apparently even against much "liberalization." He does indicate support for "a harm reduction perspective", but, as he puts it, "without a one-sided focus on the harms caused by restrictive policies." Waal thinks that "drug war" is a bad metaphor for drug policy and that we need a new metaphor. I am not sure about Norway, but in the United States "drug war" is not a metaphor, it is a fairly literal description of current US drug policy -- including tanks, guns, airplanes, helicopters and the rest, just like in the movies. As I write this, the US Drug Czar is holding a conference with all the "Latin American Drug Czars" to discuss how to use that equipment more effectively. The question is not the name, but the policy -- and Waal evidently approves of much of it.

Waal also seems to believe that some drug policy is not really rational -- is not based on a rational measuring of costs and benefits. If we can accept that sometimes drug policy makers are not choosing rationally, then why does anybody think that drug addicts would be? Indeed, the editors never explain why so much intellectual firepower has been devoted to critiquing a narrow, highly technical argument made by two economists in an economics journal.

The other chapters in the book look at addiction from a variety of perspectives, focusing to a greater or lesser extent on the limitations of viewing addiction as rational behavior. The essay by Olav Gjelsvik, a Professor of Philosophy in Oslo, is entitled "Addiction, Weakness of the Will, and Relapse", and concludes that "weakness of the will" is, in fact, important in understanding addiction. The chapter by George Ainslie, the Chief of Psychiatry at a Pennsylvanian hospital, is entitled "The Dangers of Willpower." It offers a surprising blend of psychiatric and econometric vocabularies, as its last two paragraphs illustrate:

"Neither addiction nor dissociation represent the straightforward function of an elementary principle; rather hyperbolic discount curves can be expected to create them semistably in the course of intertemporal bargaining. Significant clinical addictions will usually occupy lapse districts (areas that the will has stabilized by its failure) and these addictions may ally with other addiction-range interests and even long-range interests. Likewise, dissociation has other motives besides the management of traumatic memories."

"As theories of this complex subject proliferate, a unifying discipline will be indispensable: that of specifying how each tenet explains robust contradictory behaviors within the constraint of maximizing expected utility. This discipline seems attainable if and only if we understand the discount curves from expected rewards to be hyperbolic."

Unlike Elster & Skog, this psychiatrist sees the most promising line of research not in brain chemistry, but rather in a distinctive combination of psychiatry and micro-economics. He even thinks that various perspectives can come together in a unifying discipline "if and only if" it uses the concept of "hyperbolic discount curves" and related nations.

Elster's essay on "Gambling and Addiction" begins by asking whether compulsive gambling is "usefully seen as an addiction." In the course of discussing this he writes: "I take it for proven that compulsive gamblers typically -- not only occasionally -- behave irrationally." He then asks: "Does this invalidate the Becker-Murphy model of rational addiction?" He concludes that some compulsive gamblers do "violate ... assumptions of the Becker-Murphy model." In particular, says Elster, "they do not stick to their plans, and they do not form rational beliefs." Therefore, "the behavior of these gamblers cannot be predicted by the model."

The last two chapters appear to reject Becker & Murphy's thesis from the outset. George Loewenstein, a professor of Social Science in Pittsburgh, offers "A Visceral Account of Addiction." The last sentence of his chapter reads: "I view addiction as one of many types of human behaviors that are not usefully viewed as rational." The Epilogue by Thomas Schelling, an economist in Maryland, briefly reviews a range of states -- depression, euphoria, drunkenness, fear, panic, rage, temper, phobias, and more -- not well captured by the concept "rational." I found both essays to be among the most interesting in the book.

To summarize: five American and five Norwegian professors from a variety of disciplines have approached the question of whether addiction is rational. They have taken the question seriously, thought hard about it from their various perspectives, and said that in important ways addiction is not rational.

I hope I am not being overly critical when I suggest that this is not earth-shaking news. In a page bound with the book, Cambridge University Press points out that the contributors also "demonstrate that addicts are capable of making choices and responding to incentives." As the readers of Addiction know, this is not a new finding either.

The editors and some of the authors also seem to believe that drug policy is made taking into consideration the conclusions of people like themselves. In the United States at least, it does not seem to me that the views of theoretically-oriented professors help explain almost anything about the character of drug policy -- which must be found instead in history, culture, society, politics and macro-economics.

In the 1850s, 20 years after the American total abstinence movement began (and just as the Norwegian one was getting under way), famed showman P. T. Barnum -- who made a fortune displaying midgets, giants, and other "freaks" -- wrote a temperance pamphlet. This is how he began:

"If as a judge of rarities, I were asked what I should consider the greatest literary curiosity, I am inclined to think that I should respond: The man who can furnish a new idea on the subject of Temperance."

In the study of intemperance or addiction, as in other fields, there are often new facts and data that sometimes lead to new understandings; but, as this book shows, new ideas are still quite rare.

HARRY G. LEVINE
Queens College,
City University of New York, USA