The Book of Woe, and why NIMH said goodbye to DSM-5

IMG_1078This book review and comment are a little off my beat, for they are about DSM-5 in general, not just its diagnosis of trauma, which I addressed in my post of January 20, 2015,

The full title of Gary Greenberg’s book is The Book of Woe: The DSM and the Unmaking of Psychiatry. Greenberg is a psychologist and journalist. Some of the juiciest pieces of this book appeared in The New Yorker between 2010 and 2013 (Google Greenberg + DSM + New Yorker; they are all there).

The book begins with a story about Sandy, one of Greenberg’s patients. When Greenberg first saw him Sandy could barely go out of his house. By the end of therapy, Sandy had a job, a girlfriend, and what most of us would call a life. Sandy and Greenberg exchanged emails after Sandy moved halfway across the country from Connecticut. Eventually, the emails stopped. A couple of years later Greenberg received a call from Sandy in the middle of the night. It ran something like this.

You’ve got to help me. They’ve sucked all the bones out of my body. I’m here in this hotel room and my bones are gone. My mother and my father and James. They’ve done this to me. And I don’t want to die. Please don’t let them kill me. You’re the only one who can help. Good-bye. Good-bye. (p. 9)

Greenberg tried to call Sandy back at the hotel he had called from, but there was no answer. He never heard from Sandy again.

The book turns on what to make of Sandy’s phone call.

(1) I think that we best understand it when we understand what Sandy is using this desperate statement to really talk about, for his is a metaphor with meaning. But what’s the meaning?

(2) I don’t think we really understand what Sandy says when we label it as psychotic or schizophrenic. But we may be able to say that a group of people like this were helped by a drug called Clozapine or Risperidone.

(3) For many, including the leadership of the National Institute of Mental Health (NIMH), the goal is locate the bundle of neurons in Sandy’s brain that is causing his delusion, characterize their atypical firing, and we have diagnosed Sandy’s disease.

Still, it’s complicated. Drugs have side effects. Sandy might be helped by understanding the metaphorical language his delusion is speaking, even if this didn’t make it go away. It seems like in the end we should want all three, with meaning taking the lead but not always. It depends where Sandy is at.

DSM: Not about the world

The psychiatrists who wrote the DSM III, which started it all, understood that they were not saying anything about the world. They were trying to find a reliable way to communicate among themselves.  Psychiatrists know that certain symptoms tend to cluster together, and psychiatrists could reliably identify these patients. But, even (or especially) Robert Spitzer, who directed the team of psychiatrists who drew up DSM III, acknowledged that the book did not solve the validity problem. It wasn’t supposed to. The APA originally hired him, says Spitzer, to

achieve only the smallest of bureaucratic goals—to bring the DSM into harmony with the World Health Organization’s International Classification of Diseases, known as the ICD. (p. 41)

The validity problem is the problem of whether the DSM is characterizing any real entity in the world. Reliability, getting a bunch of psychiatrists to agree that certain symptoms tend to accompany each other in ways that can be called disorders, is much simpler. Even so, the reliability of most DSM-5 diagnoses is low, between .40 and .60 on Cohen’s kappa (I don’t really understand kappa, but I think Greenberg does, pp. 225-228).

From this small beginning the DSM III took off to rule the psychiatric world, at least in the United States, selling over a half million copies. DSM-5 is likely to be similarly successful (at $199 a pop), only there is an embarrassing story (at least for the people who wrote DSM-5) behind it. Not a single disorder in the DSM is diagnosed by its physical causes, but only by its organization of symptoms. It would be equivalent to diagnosing a man with a heart attack as suffering from chest pain disorder. Physical medicine can point to entities such as narrowed arteries, germs, or abdominal adhesions as the cause of heart attack, streptococcal pneumonia, or abdominal pain. The DSM points only to symptoms.

DSM-5 had a chance to do it differently

In an article published in The American Journal of Psychiatry, a group of eminent experts called for the DSM-5 to include a disorder they called melancholia.

Melancholia is unique in that it can be distinguished from other depressions by clear physical and chemical markers: a particular pattern of disturbed sleep (more REM, less deep sleep), a test that reveals persistently high cortisol levels, and a heightened responsiveness to tricyclic antidepressants. “The proposal included plenty of standard scientific evidence—clinical and lab studies, case histories, literature reviews—and, with its tie to cortisol, melancholia seemed to fit in with emerging theories about depression and stress.” (pp. 336-337)

Why wasn’t melancholia included? The promoters of melancholia were told that the mood disorder people wouldn’t like it because it would be the only diagnosis with a physical basis. In other words, it would show the DSM-5 up for what it was: a collection of symptoms that was organized more in accord with the political influence of those who proposed the disorder than with nature.

About the new diagnosis of melancholia being in accord with nature we should be a little skeptical. A proponent of melancholia complained that

carving out a well-defined type of mood disorder . . . is a small step in the development of the classification, but it is one that has been extracted from Nature [in capitals] grudgingly, and deserves greater attention and consideration within . . . DSM-5. (p. 338)

Melancholia isn’t extracted from nature. It is as much a social construction as any other DSM diagnosis. The difference is that it has identifiable physical and clinical markers, so that it more closely resembles the social construction of pneumonia, once thought to be caused by miasma, now believed (on the basis of sound, socially constructed physical evidence) to be caused by bacteria or virus, depending on the type of pneumonia.

NIMH sends its regrets

At almost the same time as DSM-5 was being released, the National Institutes of Mental Health, was refusing to fund any more research based on the DSM. I quote from the press release.

National Institute of Mental Health (NIMH) announcement
By Thomas Insel on April 29, 2013

Patients with mental disorders deserve better. . . . That is why NIMH will be re-orienting its research away from DSM categories  . . The weakness is its lack of validity. Unlike our definitions of ischemic heart disease, lymphoma, or AIDS, the DSM diagnoses are based on a consensus about clusters of clinical symptoms, not any objective laboratory measure.

Henceforth, the NIMH research goal is to fill in the “Draft Research Domain Criteria Matrix,” which links 5 basic natural formations, such as “systems for social processes,” including attachment and fear, with eight columns of units of analysis, such as genes, molecules, and cells.

The goal is to move from mind to brain, so that there will no longer be any need to talk about mind at all. It’s all about electricity and meat, as Greenberg puts it. And electricity and meat can be measured.


As Greenberg recognizes, there is something going on in our culture today that is conducive to seeing ourselves as the people of the brain. (p. 345) I’d say what is going on is the loss of the mytho-poetic imagination, so that we no longer have interesting stories to tell about the reality of our inner lives, for these stories come from a cultural reservoir that is quickly drying up. In this regard, there is not much to choose between DSM and the NIMH.

Perhaps the best thing to do is remember that people’s psychological symptoms generally have meaning, and that talking about our suffering is often the best thing we can do. Suffering comes in many guises, and there is no need to label every one. “Tell me what ails you, and let’s see if we can’t make some sense out of it together. And if your suffering gets too bad, certain drugs or treatments might help, at least for a little while.” Is this an impossibly naïve way to begin?


Gary Greenberg, The Book of Woe: The DSM and the Unmaking of Psychiatry. NY: Penguin, 2013.


Comments (5)

  1. James

    Despite his being labeled an Anti-psychiatrist, a term he vehemently opposed (see for example his book ‘Antipsychiatry: Quackery Squared’), Thomas Szasz repeatedly called for psychotherapy to move away from the medical model and for people to recognize that the mind is not synonymous with the brain (cf. ‘The Meaning of Mind: Language, Morality, and Neuroscience’).

    “Perhaps the best thing to do is remember that people’s psychological symptoms generally have meaning, and that talking about our suffering is often the best thing we can do,” sounds like something Szasz would have said, and have the backing of Erich Fromm (see his ‘The Forgotten Language’) and C.G. Jung (cf. ‘Man and His Symbols’).

    Thanks for posting this review. It stands as a beacon of hope for those of us who believe making meaning is more important than labeling and categorizing people into clusters of symptoms.


      James, I suppose making meaning is the trick, and while I wasn’t writing about trauma in particular in this post, I think what trauma does is make it impossible to tell a meaningful story about oneself to oneself or others.

      Trouble is, many traumatized people can eventually tell meaningful stories, and they remain traumatized. Meaning isn’t enough, or perhaps it is too much: unbearable meaning. Including the meaning that my suffering has no larger meaning: it didn’t make me better or the world better. How people come to terms with that I’m not sure. Nietzsche liked Greek tragedy. I think living in a community with others and the practice of rituals helps. Both are an endangered species these days, at least in the modern world.

      • James

        Yes, ritual does help a great deal; in fact, the “symptoms” of mental illness have been likened to rituals, as I am sure you are aware. Only difference is, they are private and have meaning only for the single individual practicing them.

        Perhaps there’s something we can do as educators to recommend recovery of the forgotten language of ritual? The Freemasons have managed to keep age-old rituals alive and well for centuries. I wonder if there’s a way we can somehow borrow from them and make their rituals more widely accessible.


          Yes, James, I think you are right on all counts, at least in principle.

          While PTSD is not a ritual, there are cultural idioms of distress that tell people suffering from trauma how it is expressed in our culture. Too bad these memes aren’t a little more gentle on the self, but I suppose they can’t be and still express trauma.

          And of course there are more private rituals that people with OCD practice.

          About ritual and trauma, I think that AA (Alcoholics Anonymous) practices ritual: belief in a higher power, testimony with no cross-talk (that is, others can’t comment on the testimony), a ritual prayer (the serenity prayer), etc. I wonder if local groups like TA, Trauma Anonymous, wouldn’t help (I just made up TA).

          There is, I suppose, a minimum of formal ritual in these groups, but many people go to AA 4 – 5 times a week, and it takes on the quality of a ritual. Why not for TA?

          I think a leader would be necessary at first, but someone who has suffered trauma and come to terms with it and knows the general idea of the group would probably be enough.

          Adaptions would be necessary, maybe some limited crosstalk, but the idea seems sound to me. Of course, versions of this are sponsored by the VA, but I think a freestanding TA would have advantages.

          So, I guess what I am saying is that there are lots of rituals, and while PTSD itself has the quality of a ritual, so could its amelioration.

  2. You bring up a good point. I wonder how hinavg a child diagnosed in infancy would affect parental experiences. On the one hand, the diagnosis might be harder to accept because the baby’s development still seems typical to the naked eye. On the other hand, parents could be spared the agony of suspecting a problem and then waiting for a diagnosis, and could take comfort in knowing that their child will receive help as early as possible (assuming that infantile interventions are developed and available, which is a big assumption ).

Leave a Reply

Your email address will not be published. Required fields are marked *