Notes from James Davies’ “Cracked, the Unhappy Truth About Psychiatry”


Found this book at a bookstore and was hooked after reading the first few pages, based on it’s topic and clear and well-written prose.

James Davies

James Davies

Chapter 1

James Davies, with a PhD in social and medical anthropology from Oxford, begins with a history of psychiatry starting in the 1970s and a crisis of confidence it faced. A series of experiments questioned the validity and reliability of psychiatric diagnosis.

The 1973 Rosenhan experiments on “Being Sane in Insane Places” questioned the validity of psychiatric diagnoses. Neurotypical confederates checked themselves into asylums claiming to have heard a voice once, but then once admitted acted normal and saw what they were diagnosed with and saw the great lengths it took to be deemed healthy again and get discharged from the institutions. After publishing the results, they were challenged by a hospital to send pseudopatient imposters back in. Rosenhan agreed to the challenge but did not send any patients. A month later the hospital reported they suspected 41 imposters.

Another experiment showed that diagnoses were not consistent between psychiatrists. It sent the same patients to different psychiatrists and showed that they got different diagnoses from psychiatrists around a third of the time. Additionally, the prevalence of different diagnoses seemed to be regional, with some diagnoses being more prevalent in certain countries.


These criticisms of psychiatry lead to a drastic rewrite of the 3rd edition of psychiatrists diagnostic manual, the DSM III, headed by Robert Spitzer, who attempted to increase diagnostic reliability by making the definitions of disorders more precise and adding an explicit checklist to each disorder. These checklists created thresholds between psychiatric disease and “normal” human experience that, while decided by expert consensus, were ultimately arbitrary.

Additionally, the DSM III removed many disorders, especially those that had been introduced by psychoanalysts, as that discipline was falling out of favor. He brings up an interesting story about the removal of homosexuality as a “sexual deviation,” which he says was largely due to pressure from the Gay Rights movement and came down to a vote at an American Psychiatric Association meeting in the 70s, with 5,854 psychiatrists voting to remove homosexuality as a disease and 3,810 voting to keep it in. Davies makes the point that many of these decisions were political and not directly based on any changes in scientific research.

Despite the reforms made in the DSM III and subsequent manuals, diagnostic reliability remains a difficulty. Interestingly, I looked up a citation Davies makes to Aboraya, 2006, which he says “showed that reliability actually has not improved in thirty years.”I’m uncertain about how Davies reached that conclusion from this paper, as it clearly states:

  • that while diagnostic reliability remains a problem, the third generation of psychiatric diagnoses “from 1980 to present… more reliability papers were published and the reliability of psychiatric diagnosis has improved,” and
  • “The development of the DSM-III and its subsequent versions has been a major accomplishment in the history of psychiatric nomenclature. Clinicians use the DSM criteria in clinical practice as an effective way to communicate the clinical picture, the course of illness, and efficacy of treatment.”

This citation seems academically sloppy and perhaps shows that Davies seeks to oversimplify a complex and murky issue into a one-sided story (though this also might reflect my innate bias against pop-science books).

Chapter 1 ends questioning the validity of psychiatric diagnoses even if we fix the reliability problem. Even if we could get every psychiatrist to agree on the diagnoses, does that mean it’s a real disease entity, or that we’ve just made a reliable but arbitrary construct? He argues that we need biomarkers to prove it’s a “discrete, identifiable biological disease.” While I agree, I think that psychiatric definitions do a good job of separating normal but different from disease, by often requiring that the disease is disruptive to the patients social relationships or occupational function. What makes psychiatric illnesses, diseases is that they are problematic for people’s lives, and people, whether the patient themselves or their friends and family, want something done about it. I’m unsure if we will be able to find or need to find biomarkers for every disease. While some diagnoses may ultimately be arbitrary, if they are clinically helpful and can show statistical and long-term improvements in patients quality of life, then they are valuable.

Let me know if you have any comments on this blog post. I hope to continue blogging my thoughts on this book as I read through it.

One Response to “Notes from James Davies’ “Cracked, the Unhappy Truth About Psychiatry””

  1. Definitely interested in reading more of your thoughts on this book. Davies seems to fall victim to the informal logical fallacy of “the Nirvana Fallacy” when he calls for the need of biomarkers to “prove” the existence of psychiatric disorders.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: