When this book first came out, in 2009, it was expected to raise controversy. In fact the reactions to its publication have been generally rather bland: Heyman's discussion of addiction as a disorder of choice may have been too narrow to have any power to shake or stir. The book may have some of the flaws and limitations that derive from its narrow focus, and its arguments may not be always perfectly logical, but if you are like me you will enjoy its critical look at often repeated assumptions about addiction, and you will find it full of useful references and useful for thinking about recovery. Even if it will not shake your view of the nature of addiction, it will force you to reassess the solidity of its foundations.

The book opens with a historical overview of the first drug ‘epidemic’ and the responses to addiction (chapters 1–2) and follows this with a series of first-hand accounts from individuals who developed and recovered from a substance use disorder (chapter 3). There is nothing new or distinctive about these introductory chapters: good historical accounts and first-hand stories of recovery can be found elsewhere. Where things hot up is in chapter 4, where Heyman begins his argument that addiction may not quite be as it is portrayed in mainstream literature. On the strength of, particularly, epidemiological evidence, he argues that the chronic disorder characterization of addiction that emerges from many textbooks and scientific papers is skewed, because all the evidence on which it is funded derives from samples recruited in treatment services. Only a minority of individuals with a substance use disorder, he points out, actually present to treatment services. Outside treatment, addiction is not a chronic disorder and remission is the norm rather than the exception. He also points to a clue as to why there may be such a discrepancy. The chronically addicted individuals of treatment services have been shown time and again to have high rates of psychiatric co-morbidity. Heyman suggests that what is necessary to recover from addictive disorders is a capacity to develop interests and responsibilities that would compete with addictive behaviours. This, he argues, is what distinguishes the addicted individuals of treatment services from their not-in-treatment counterparts. As a result of a pre-existing psychological disorder, he says, they may lack that capacity, and for this reason they remain addicted for longer. This is the strongest chapter of the book. Although the argument is not conclusively made (for example, co-morbidity data are lacking for not-in-treatment individuals with an addiction, and the discussion is based on extrapolation from the non-addicted population), chapter 4 reads like a useful antidote to the groupthink that all too often stifles mainstream scientific literature, and it offers a plausible account of why it is less difficult, even for well-resourced drug treatment services, to get people in treatment than to get them successfully through (if Heyman is right, the word ‘recovery’ here would be a misnomer).

If chapter 4 was the book's strongest, chapter 5 is the weakest. Here Heyman argues that not only is addiction not a chronic disorder, it is not a disease at all. To support this argument, he demonstrates the falsity of the three syllogisms: (a) addiction is characterized by genetic predisposition/changes in the brain/self-destructive behaviour; (b) all conditions with genetic predisposition/changes in the brain/self-destructive behaviour are involuntary; (c) addiction is involuntary (and therefore a disease). The problem with this approach is that these notions of disease are too narrow and could as easily be used to argue that, for example, depression should not be considered a disease. Equally narrow is Heyman's insistence on the juxtaposition of voluntary versus involuntary behaviour, rather than goal-directed versus habitual behaviour, which has been fruitfully applied to the study of addiction for at least the last 10 years (see Robbins et al. Reference Robbins, Everitt and Nutt2010). Furthermore, Heyman seems at times to get lost in his own arguments, for example when he uses a study on obsessive compulsive disorder (which he does not contend is a disease), to argue against the statement that addictive behaviours are involuntary and therefore aspects of a disease. Where Heyman is right is in his attempting to raise the reader's antennas against sloppy, worn, short-cut arguments that are used (and we all use from time to time) to explain the ‘diseaseness’ of addiction.

After attacking the ‘orthodox’ view of addiction, in chapters 6 and 7 Heyman offers his ‘heretic’ view, which is essentially couched in choice theory. In chapter 6 he describes general principles of choice and the two patterns of behaviour that can be generated from them: one depends on a piecemeal evaluation of the values of outcomes (‘local choice’); and the other depends on the evaluation of ‘packets’ of choices (‘global choice’). In chapter 7 he discusses the features of addictive drugs (as opposed to non-addictive rewards) that make it more likely that someone who has taken them will begin using local, rather than global choice more frequently than is good for them. The chapter closes with some observations on individual differences and on the effectiveness of treatment.

Heyman's description of addiction as resulting from an imbalance between local and global choices is neatly illustrated. The problem for me is that it sounds very much like countless previous accounts of addiction that emphasize the tendency of addicted individuals to neglect long-term benefits for the sake of immediate ones. Similarly, the concept that addiction can be understood in terms of normal mechanisms for motivation and choice having become imbalanced, as opposed to those mechanisms being ‘diseased’, has been put forward before (e.g. Orford, Reference Orford1992; Robbins & Everitt, Reference Robbins and Everitt1999). Heyman's model simply re-tells this concept from the perspective of a decision scientist, offering no advance in terms of predictive power or capacity to suggest improved treatment strategies (Heyman says that his model endorses contingency management, but this has already received the imprimatur of many scientific and governmental bodies worldwide).

In summary, I very much enjoyed reading this book, because it challenges concepts that many of us have become used to hearing and to which we have stopped applying our critical sense. One is the view that addiction is always a chronic disorder. Another is the notion that individuals with an addiction are better off in than outside treatment services, which in the UK has become a mantra of treatment services regulators (perhaps with some perverse effects, see Kimber et al. Reference Kimber, Copeland, Hickman, Macleod, McKenzie, De Angelis and Robertson2010). As the focus of treatment begins to move from engagement to recovery, Heyman's suggestion that chronically addicted individuals may have chronic problems with developing interests alternative to drug taking will be food for thought. For these reasons Heyman's book was well worth a read for me, even though his model of addiction, arrested as it is at the stage of description, may not merit elevation to the rank of ‘theory of addiction’.