Despite its sensationalist subtitle, this is a serious book. It takes the well-developed harmful dysfunction theory of mental illness and applies it to the anxiety disorders as defined in the “DSMs.” The harmful dysfunction theory, initially articulated by Wakefield 20 years ago (1, 2), argues that to be a psychiatric disorder, a syndrome must both cause harm to the individual and reflect an underlying dysfunction of an evolved mind/brain mechanism. The second part of this theory is the chief concern in this book. We can assume without controversy that the heart evolved to pump blood. In congestive heart failure, the heart is dysfunctional because it cannot do what it was designed to do: pump sufficient blood to meet the body’s needs.

This might seem simple, but it gets more complex when applied to psychiatric illness. Let us posit a fictional case to illustrate this point. Imagine 100,000 years ago, bites from a vicious feline species injured many early Homo sapiens. Genetic variants that caused an innate fear of the vicious feline emerged and were selected for as Homo sapiens evolved. Eventually, a proportion of the population became predisposed to developing a fear of the vicious feline. The resulting avoidance produced fewer bites and improved fitness. Fast-forward to the present. Without changing much in appearance, the vicious feline has evolved into the harmless house cat. However, lots of humans are still predisposed to fear them. A proposal is made to classify feline phobia as a type of specific phobia in DSM-5. The typical sufferer from feline phobia considers the fear to be completely irrational and wishes it would go away. “They are so nice and cuddly. They purr. I don’t understand why I have to panic every time I get close to one. Can’t you help me?”

Should we accept this proposal for DSM-5? Does feline phobia constitute a mental disorder? It clearly meets DSM-IV criteria for a “marked and persistent fear that is excessive or unreasonable” and all the other phobia criteria. But not so fast, say Horwitz and Wakefield. There is, they would argue, no dysfunction here. We humans evolved to be fearful of the vicious felines. The “feline-detection fear circuitry” is functioning normally in these individuals. You cannot call them disordered.

This book confronts the reader with two contradictory intuitions. In reflecting on what makes a person “disordered,” one approach is to evaluate the extent to which a people’s mind/brain system allows them to adapt functionally to the world around them or is somehow interfering with such adaptation. Some individuals with feline phobia are quite unhappy about their fears. These fears affect their lives negatively, and they want to get rid of them. From this perspective, calling them disordered seems sensible.

Horwitz and Wakefield posit a different intuition. You have to examine what the mind/brain was designed to do, they argue. If it is doing just that, even if that is currently causing harm, you cannot call it a disorder. Nothing is going wrong with the system.

This approach should not be lightly dismissed because it reflects something important about what we mean when we call something a disorder. Most of us want to assume that we treat disordered people when something has really gone wrong.

Thus, this book consists of a lengthy and at times forceful argument for the harmful dysfunction model in its application to anxiety disorders. The DSMs don’t come out very well in this exercise. By the Horwitz and Wakefield standard, the DSM definition of disorder is too broad, too focused on current dysfunction, and insufficiently attentive to the evolutionary past.

Horwitz and Wakefield have chosen their topic for this book wisely. Anxiety disorders provide a better terrain for their argument than does depression, the focus of their previous and closely related book (3). No one can doubt that severe anxiety occurs in individuals as a normal response to danger. Whether clinically significant depression can arise in a similar normal way is less clear than Horwitz and Wakefield contend.

In the meat of this book, the DSM anxiety disorders are discussed one by one. In each case, it is argued that the way in which DSM has drawn the boundaries of illness is flawed. Horwitz and Wakefield point out that DSM uses a hodge-podge of terms (e.g., “unreasonable,” “irrational,” or “excessive”) that all refer to whether the anxiety appears commensurate with the threat that the world actually presents to the patient. They are most vehement about social phobia and posttraumatic stress disorder, which they contend are being massively overdiagnosed. Space precludes me from here rehearsing their arguments. But in each case, they contend that DSM has ignored the fact that evolution has designed us to have anxieties that may be irrational by current standards (e.g., vicious felines, airplanes, tall buildings). Ignoring this fact, they argue, results in us being quite overinclusive in our diagnoses of anxiety disorders.

The greatest weakness of this overall excellent book is the excess confidence Horwitz and Wakefield display in the correctness of their theory. More self-criticism would have made for a better and a more balanced book. In the spirit of trying to even the playing field for those who go on to read this book, I would emphasize three limitations of their argument.

First, in most cases, we really have no good idea about what evolution has designed our mind/brain systems to do. Furthermore, very rarely can we do experiments to answer these questions. We are therefore at great risk of having competitive “just-so” stories about evolution that we judge on the grounds of plausibility, a slippery slope if there ever was one. If we did not know that long, long ago, felines were very vicious and prone to bite, what stories might have been constructed to explain our fear of them?

Second, Horwitz and Wakefield misrepresent evolution. Their argument depends critically on the assumption that evolution produces a single prototype of human functioning. But evolution works on populations not individuals. We know quite well that humans differ widely from one another in their vulnerability to anxiety disorders at least partly for genetic reasons. Some of us are going to have a lot of risk genes for anxiety disorders and others far fewer. Given this level of genetic variation, the harmful dysfunction theory would seem to lead us into the following rather untenable conclusions:

That is, a logical extension of the harmful dysfunction theory given genetic variation is that some individuals with the identical DSM syndrome should be considered ill and others not, depending on whether their “fear circuits” were “doing what they were designed to do.”

Third, it is not at all obvious what to do when the environment in which we evolved and current environments are seriously out of sync with one another. This was of course the point of my largely fictional story about vicious felines. But consider the current epidemic of type 2 diabetes, which could be a result of an evolutionarily selected fat storage system functioning as it should in the age of McDonald’s. Do we really wish to then argue, as dictated by the harmful dysfunction theory, that individuals with type 2 diabetes are not disordered and hence should not be eligible for insurance coverage because their metabolism is doing what it was evolved to do?

So the harmful dysfunction model for psychiatric illness has some significant problems. Despite this, however, it is a coherent, well-argued and thoughtful view about the boundaries we should set for mental disorder. Furthermore, I cannot suggest a much better approach, nor can the psychologist/philosopher Derek Bolton in his own excellent book on this topic (4). While we have our formal definition at the front of DSM, as a field we are actually in the uncomfortable position of not having a clear, philosophically coherent and easily implemented definition of a mental disorder. It is a devilishly hard problem.

For those interested in the fascinating problem of trying to define the boundaries of our disorders, reading this book will be time well spent. Indeed, in our mature moments, we should be glad that our field has attracted critics of such quality.