Mar 02 2010


The Diagnostics Statistical Manual for Mental Disorders, a much maligned document, is in the midst of its fourth major revision (the DSM-IV will be replaced by the DSM-V). This process has been going on for over a decade. The revisions are now being made public in order to have a two year period of public comment and debate about the details of the revisions.

This has led to a new round of criticism of the DSM, and through it psychiatry, from those who either do not sufficiently understand, in my opinion, the nature of psychiatric diagnosis, and from those who are anti-psychiatry because of ideology (Scientologists, for example).

At the extreme end of criticism are those who deny the very existence of anything that can be called mental illness. I have already dealt extensively with their arguments, and won’t repeat them here. But even those are not so extreme fall into some of the same logical fallacies when criticizing mental diagnoses. Recently George Will, for example, wrote an editorial which I think confuses medical diagnoses with taking moral positions (I will get to his commentary below).

Diagnosis in Psychiatry

I do acknowledge the extreme difficulty of establishing discrete diagnoses in psychiatry. The brain is the organ of mood, thought, and behavior, and it interacts in complex ways with the environment. So it is very difficult to tease apart those manifestations that should be considered a symptom of dysfunction from those that simply represent the range of normal variation or the effects of the environment.

Further, when there is demonstrable pathology we tend to think of those disorders as neurological, not psychiatric – even when they have overtly psychiatric manifestations (mood, thought, and behavior). If someone becomes psychotic (develops delusions and hallucinations) and is found to have a brain tumor, most people have no problem understanding that the tumor caused the psychosis – that this is a disease that needs to be treated. However, if another person becomes identically psychotic and no underlying pathology is found, suddenly this is a psychiatric disorder (not a neurological disease) and many people wring their hands about whether or not it is a real disorder, how to define it, and if it should be treated at all.

The challenge of psychiatric diagnosis also suffers from the fact that mood, thought, and behavior exist on a continuum – a Bell curve of variation. So there is often a demarcation problem – between healthy variation (the term “normal” is too problematic to use effectively), and mental disorder. Take any mental trait, such as anger. Some people are more angry than others. What if someone had no control over their anger and were perpetually on the verge of rage – triggered by the slightest perceived offense? It seems reasonable that at some extreme of anger we can meaningfully call this a mental disorder and explore ways to treat it. But there is no sharp line between healthy anger and a dysfunctional loss of self control.

This, of course, is the exact challenge of the DSM – coming up with reasonable criteria to designate when there is a disorder and some scheme of categorization. The definition of “disorder” that guides this process is the lack of a trait or ability usually possessed by people that leads to a demonstrable harm (like self-control).

Psychiatry also has a lot of historical baggage to contend with. In the dark ages of psychiatry, which we are barely out of in historical terms, diagnoses were not cleanly separated from the current fashionable moral judgments of the time. And so homosexuality was considered a mental disorder for a time. Treatments were likewise crude, even cruel. But it is a fallacy to blame the modern profession of psychiatry for the sins of the past – sins from which they have already redeemed themselves. It must also be noted that psychiatry was largely reformed from within.

Mental Illness and Morality

In his editorial, George Will argues that the DSM, including many of the new diagnoses being contemplated, represents medicalizing character traits, in order to exempt them from moral consideration. He commits many of the errors of mental illness denialists, however, in forming his arguments. He writes:

Today’s DSM defines “oppositional defiant disorder” as a pattern of “negativistic, defiant, disobedient and hostile behavior toward authority figures.” Symptoms include “often loses temper,” “often deliberately annoys people” or “is often touchy.” DSM omits this symptom: “is a teenager.”

Clearly Will thinks that a “normal” teenager might earn for themselves the label of “oppositional defiant disorder” – ignoring the crucial aspect of diagnosis that these traits must exist to an extreme degree, sufficient to cause demonstrable dysfunction. Therefore your average disobedient teenager would not (should not) be diagnosed. But a child who was unable to function at school because of frequent hostile disobedience might.

My wife, who is a counselor, actually worked with oppositional defiant children for a few months (those severe enough that they were enrolled in a special school) – they quickly burned her out (quite a feat, as she is a very tolerant person) and she decided to change jobs. There is no way to quantify severity of a mental disorder like this, and so some judgment is always required. But when you see the extreme cases you know what oppositional defiant disorder means, and you won’t confuse it for average teenage behavior.

He continues:

This DSM defines as “personality disorders” attributes that once were considered character flaws. “Antisocial personality disorder” is “a pervasive pattern of disregard for . . . the rights of others . . . callous, cynical . . . an inflated and arrogant self-appraisal.” “Histrionic personality disorder” is “excessive emotionality and attention-seeking.” “Narcissistic personality disorder” involves “grandiosity, need for admiration . . . boastful and pretentious.” And so on.

If every character blemish or emotional turbulence is a “disorder” akin to a physical disability, legal accommodations are mandatory. Under federal law, “disabilities” include any “mental impairment that substantially limits one or more major life activities”; “mental impairments” include “emotional or mental illness.” So there might be a legal entitlement to be a jerk. (See above, “antisocial personality disorder.”)

The DSM actually differentiates between a personality “trait” and a personality “disorder”. A disorder is no mere “character blemish” – it is a gross disfigurement. This is a crucial concept missing from Will’s editorial and highlights the danger of criticizing an entire profession without sufficient expertise to do so. Also, the personality disorders are considered the least “illnessy” of the mental illnesses. In fact they are described in the DSM and are part of psychiatric assessments so that their traits can be differentiated from more hard core mental illnesses, like schizophrenia or bipolar disorder.

Will continues with this fallacy a bit more, then turns to a new one:

Extremely irritable or aggressive children are frequently diagnosed as bipolar and treated with powerful antipsychotic drugs. This can be a damaging mistake if behavioral modification treatment can mitigate the problem.

Does Will really think he has a better grasp on the evidence for the relative safety and effectiveness of medication vs behavioral therapy for a variety of mental disorders than trained experts? But that aside, this hits upon one of the features of the proposed DSM-V revisions (albeit controversial even within psychiatry). New diagnoses were carved out in the DSM-V for the stated purpose of differentiating them from more severe disorders that are more likely to be treated medically.

Will links to this article in the Washington Post (apparently the spur for his editorial) which states:

Others expressed concern about the proposals to create new conditions such as “temper dysregulation with dysphoria,” or TDD. Supporters say it is intended to counter a huge increase in the number children being treated for bipolar disorder by creating a more specific diagnosis, though critics argued that it would only compound the problem of overtreatment.

I don’t know what the net effect will be of making a new category of TDD – it seems to be a matter of debate among those revising the DSM. But at least Will should understand and acknowledge that stated purpose by supporters – to create a less severe diagnosis and decrease the number of people who will quality for the more severe diagnosis and therefore likely get treated. Will assumes that labels lead to increased treatment, but the opposite may be true depending on how they are applied. At the very least this topic is more complex than Will’s simplistic formulation.

Finally we get to the core of Will’s point:

Furthermore, intellectual chaos can result from medicalizing the assessment of character. Today’s therapeutic ethos, which celebrates curing and disparages judging, expresses the liberal disposition to assume that crime and other problematic behaviors reflect social or biological causation. While this absolves the individual of responsibility, it also strips the individual of personhood and moral dignity.

This one paragraph could be the subject of a separate post, but I will try to summarize my points quickly. First, it must be realized that within medicine it is absolutely necessary to focus on curing and to “disparage judging”. Physicians need to be non-judgmental toward their patients, and should never impose their morality, religion, or politics onto their patients. That is a core principle of medical professionalism.

I will give Will the benefit of the doubt and assume he is not denying the necessity of this professional ethic. Rather, he seems to be lamenting the extension of this attitude into the realm of broader society – claiming that it is a “liberal disposition.” I don’t want to venture into a political argument on this blog, but I will just say that I see his point and will assume for the sake of argument that it is part of liberal political philosophy, in the name of social justice, to emphasize the situation and conditions that lead to individual behavior. Meanwhile, conservative philosophy emphasizes individual responsibility. This appears to be the lens through which Will is viewing the DSM.

But I think it is fallacious to criticize the DSM for trying to understand the human condition, simply because some may use that knowledge to bolster a political position with which Will disagrees. Rather, it would be more appropriate to simply argue that understanding the biological cause of behavior does not and should not absolve one of personal moral dignity or responsibility. Will instead has decided to attack the notion that behavior is biologically caused.

This relates to the deeper question of free will. Again, I will not delve into this complex issue, but briefly – if we take the premise that the mind is the brain, and the brain, as a materialistic physical entity, is deterministic, then we can conclude that all behavior is ultimately caused. Depending upon how one defines free will, it can therefore be argued that free will is ultimately an illusion as behavior is deterministic.

My take has been that, while this is true, we are still capable of making choices based upon internal reflection, and I call this free will (acknowledging that it is ultimately deterministic). But in any case, even those who deny the existence of free will acknowledge that we still need to have laws that hold individuals responsible for the choices they make. From a societal and legal point of view, we still need to act as if we possess free will – because those laws and moral boundaries are part of the environment that feed back on our behavior.

Will is in the curious position of making the same mistake that many of the liberals that he criticizes make – denying the underlying science rather than the inappropriate moral connection. It is better to assess the science for its own sake, and in this case not just the basic science but the medical utility of the DSM – and then separately argue that simply because a behavior can be meaningfully called a disorder that does not mean we do not hold people responsible for their behavior.

Admittedly this creates another demarcation problem – because we do draw a fuzzy line beyond which we mitigate guilt due to mental illness. This is now a legal question – the insanity defense. We recognize that someone’s mental function can be so compromised that they should not be held responsible for their behavior while in that state. There are established criteria for such a defense, and like many legal criteria they are complex and blurred at the edges, but can be reasonably applied in the real world. It would be a slippery slope argument to imply that any acknowledgment that extreme mental illness mitigates guilt leads to the medicalizing of all morality.


The issues raised by Will and other critics of the DSM and its current revision are complex. There may be problems with the current revisions – I would expect any such process to be messy and involve many controversies and trade offs. But I don’t think that Will and others with similar criticisms are hitting upon the true complexities of making mental diagnoses.

Most often criticisms of the DSM stem from looking at the list of symptoms that comprise the diagnostic criteria for a particular diagnosis and then not putting them into their proper context. This results in confusing healthy human variation with the extreme traits that constitute a disorder.

30 responses so far