98 Comments

Tell you what, Bryan: next time I'm psychotic I'll come to your house, and we'll see how you feel about my preferences.

Remember, if you report this comment, you're not thinking like Szasz....

Expand full comment

Is the idea here that all preferences are good? Because that's crazy. Obviously lots of people prefer to commit crimes, the fact crimes are bad doesn't mean those people are unable to chose otherwise or they have a disease. It just means they're choosing to do something bad.

Expand full comment

People who have had their consciousness commandeered by a flaw in their neurology can't choose anything. Without conscious control there is no choice.

Expand full comment

That's point. Somehow these people who supposedly have had "their consciousness commandeered by a flaw in their neurology" suddenly make different choices given the right incentives. There is no choice without conscious control, true. But that's the point. They do show control, just way after you or I would begin to behave because they prefer their behavior to the bad consequences.

What is hard for us to generally accept is why someone would behavior self-destructively. Do people actually prefer to be depressed, anxious, alcoholic, etc.? As onlookers, we can think through others situations without involving ourselves at all in the motivating feelings that are so crucial to action. It might be obvious to a everyone that a skinny guy needs to go to the gym if he wants to build muscle, but the skinny guy might prefer staying home more than doing what it takes to build muscle.

To analogize to depression, let's say that it is reasonable for a person to become depressed after being dumped. If they decide then that they find meeting new people intimidating and so cannot find another partner, how are they supposed to resolve their depression? Without change, they cannot. So functionally, they prefer depression. But if that person were offered 1 million dollars to find a partner in the next week, maybe they'd take their friends' advice and get their shit together.

Expand full comment

Is there a list somewhere of those mental illnesses you both agree are not preferences?

Expand full comment

Bryan, Scott - I think assembling this list is a really good suggestion!

I might add one more column of "and how does my assessment of preference vs. illness change how it should be treated?" (as in "treatment" not social treatment)

Expand full comment

This is an interesting perspective, but it really is wrong. Several psychiatric conditions do support common pathway models and have other validity evidence that makes them concerning. For several conditions, they are adequate representations of commonly caused conditions and we often have treatments for them, as is the case for schizophrenia, where the meds work.

I have never seen any reason - from Tversky on - to think of the better-defined mental illnesses as individual differences in preferences. Preferences do not behave like traits, but mental illnesses do, and that affects treatment options. Take a page from Kahneman & Tversky (1979, pp's 289-290).

Expand full comment

I don't disagree with you, but it does seem like you're using "do the meds work" as the key criteria of whether its a trait or an illness. But I guess I don't understand why that matters, apart from "how should we socially judge this person?" There are a large number of mental and physical illnesses that can be improved (by the judgement of the patient) by a wide range/combination of meds, actions, choices, etc...

If I have schizophrenia and I treat it without meds and I get better, that doesn't prove you wrong. If I treat it exclusively with meds and get better, that doesn't prove Caplan wrong. Because it doesn't matter to me! What matters is: "by my standards and accounting for the cost of treatment, did I get better?" If you want to say "schizophrenia that you can treat without meds doesn't count" then I am 100% fine with that - but I think you (and Caplan) are fighting over definitions instead of reality.

Expand full comment

"Do the meds work" isn't a good criterion for whether something is a disease because 1) many illnesses are untreatable, and 2) you can use drugs to stop non-disease behaviors and preferences easily (chemical castration of homosexuals).

Expand full comment

Agreed, with the addition that may also be an indication that determining if something is a disease or not isn't that helpful (where I define helpful as "helps make the patient better of).

Expand full comment

In my example, the key criteria was "is there an associated common pathway model?"

Whether the meds work or not cannot tell us about whether a trait is real, but it can lend evidence to the idea if we see that there is a treatment effect and it is not heterogeneous (ex: https://jamanetwork.com/journals/jamapsychiatry/article-abstract/2776610). Homogeneous treatment effects are unrealistic for many traits because treatment is generally not targeted at the etiology of a disease comprehensively, so this can only be used to provide positive, not negative evidence wrt whether a psychiatric diagnosis tags an actual trait.

The concerns raised by PercyPrior have to do with measurement invariance, which is often tenuous, rather than etiology, which I was concerned with.

Expand full comment
Jun 15, 2023·edited Jun 15, 2023

The DSM counts you as being schizophrenic if you have 2 or more of the following for more than 2 months:

1) Delusional beliefs (Oxford Dictionary: "a false belief or judgment about external reality, held despite incontrovertible evidence to the contrary, occurring especially in mental conditions") Bryan's objection: what counts as "incontrovertible evidence to the contrary?" If the standard is "certainty," there isn't incontrovertible evidence that we aren't brains in vats, so it must be something more like "really really good evidence to the contrary." But what about religion? Many (most?) religious folks are happy to acknowledge that their beliefs are taken on faith, in full knowledge of the lack of evidence. Maybe Bryan's objection fails because religious beliefs are not subject to incontrovertible evidence *to the contrary,* just lacking in positive support.

Bryan also objects that incentive programs at psych wards have caused people to stop attesting to their delusions. I don't find this convincing because it's too easy for a patient to simply lie about their beliefs for rewards.

2) Hallucinations (usually auditory). Problem: most people have an internal monologue. You could say it has to be a voice heard from the outside, but the DSM considers a voice which can only be heard "on the inside" to count, assuming one does not identify with or control the voice. I'm not sure what to make of that either, though, because many of my thoughts are spontaneous and uncontrolled (just try to control what your next thought will be, and notice how often it comes to you unbidden), and I don't necessarily identify with my thoughts either. (I'm not entirely sure what it even means to "identify" with a thought. To think the thought is you? I am more than any thought I have. Does it mean "to think the thought is a part of me?" Why wouldn't hallucinations be "a part of" a schizophrenic?)

3) Disorganized speech (incoherence, frequent derailment). I don't know if Bryan has an objection to this, and Scott says in his experience this doesn't respond to incentives. That is, he will offer a patient who wants to be released the opportunity to be discharged if they can just string together an even slightly coherent sentence without their medication, and even still they cannot do it.

4) Grossly disorganized or catatonic behavior: IDK what Bryan says about this criterion either, maybe there's a study somewhere where this responds to extreme incentives too.

The takeaway: you could plausibly meet the criteria for schizophrenia despite being mentally healthy if you fit items 1 and 2 for more than two months (by being religious and having an internal monologue). But maybe the solution isn't to jettison the concept of schizophrenia as a disease, but to refine the diagnostic criteria?

Expand full comment

Yeh thats why diagnosticians are taught to use clinical judgement and not use the diagnostic criteria as a checkbox.

Hence the years of training on interview technique (assuming the symtpom , open ended etc) and thousands of patient encounters.

You dont fill out a dsm form and send it to the dsm police when you diagnose someone , you do an in depth interview and take into account the entire case and the patient as a whole. You get collateral. You examine if subjective and objective line up and if its causing dysfunction or damage to the persons well being.

No ones out their trying to "gotcha" people into a diagnosis if the treatment would be of no utility.

A broken medical system with poor incentives isnt psychiatrys fault , they use the same criteria in countries that sont have for profit healthcare systems, the criteria werent designed to buttress US inpatient hospital numbers , insurance companies and governments dont want to pay for that and have a lot more money and power than all thr psychiatrists combined.

Expand full comment

> Grossly disorganized ...behavior

Anti-ideological Pragmatism, eg, US domestic politics and foreign policy

Expand full comment

About 30 years ago, I dated a girl who had an autistic sister. She wasn't the "nerdy" kind of autistic, she was the "can't speak at all (as far as anyone knows) and huddles in a corner grunting" kind of autistic. She was 20 years old but obviously never going to be capable of taking care of herself. Sometimes her parents made us take her with us on dates, both to give themselves some rest and (presumably) to limit our friskiness.

There are levels of psychiatric disease that can reasonably by interpreted as preferences. There are also people who are truly, fundamentally broken. I suspect that the utterly mad form a large part of Dr. Alexander's reference pool and much less of Dr. Caplan's, and I suspect that to be the basic cause of your disagreement.

Expand full comment

I presume that sister would be incapable of taking care of herself even with a gun to her head. Thus, by Caplan's standards she would have a disability rather than just a preference.

Expand full comment

What if she could take care of herself if a gun was pointed to her head?

But that it took that level of extreme force and it had to be applied constantly.

Would she no longer be disabled because under those extreme circumstances she could summon the willpower overcome her problem?

I mean people are capable of all sorts of things if the threat of corporal punishment and death are ever present.

But nobody would seriously suggest that someone follow this girl around with a gun to her head. Nor I would guess to cut off whatever support she gets because she could “in theory” overcome her issues.

Expand full comment

As I said elsewhere, I prefer the term behavior instead of preferences for various reasons. This is one more reason. She has a maladaptive behavior How do you know that with the right therapy or treatment (maybe one nobody yet knows how to apply) she couldn't be helped and become able to function and take care of herself?

Expand full comment

Emil Kirkegaard's attempt to link homosexuality to pedophilia is an old homophobic trope.

Go to his blog and you find he is depraved himself. He is into BDSM:

"This subject is of some personal interest to me becus ive slept with a girl who likes this stuff. It is kinda fun, even if one has to be careful. I probably wudnt do it again, without some a prerecorded video or something that i can show in the case that she dies. In that way, i wont be sentenced for murder."

https://rationalwiki.org/wiki/Emil_O._W._Kirkegaard#Sadomasochism_and_erotic_asphyxiation

Projection much?

Expand full comment

He didn't claim to be "into it" himself. He said he slept with a girl who was.

Expand full comment

While it'd true that it is internally consistent, I feel like the "disease = constrain" framework is not a very useful way to think about mental illness/health. Consider a person with OCD who feels a compulsive urge to wash his hands every 15 minutes. Probably this person would fail the gun-to-the-head test, so, according to Brian's definition, he does not suffer from mental illness, but simply has a very strong preference for washing his hands.

It's true that this approach gives a consistent model of mental illness, but are we really prepared to say that the compulsive hand-washed does not suffer from any mental problems? Even if this person admits that his behavior is irrational (i.e., he understands that the benefits of the incessant hand washing are small compared to the cost)? Even if he finds his compulsion extremely disruptive?

If you define preferences to be whatever-people-actually-do-when-they-don't-have-a-gun-to-the-head, of course this is a preference and not disease. Personally, though, any definition of disease that does not count a life altering compulsion to wash your hands as a disease seems contrary to common sense.

To my mind, an important feature of mental illness that Brian's view is missing is the existence of intrusive thoughts of the kind and intensity that lead OCD patients to engage in their compulsions. It's like having a little voice in your head incessantly telling you to do something. You can write off these thoughts as preferences, but, at the risk of repeating myself, this seems to me contrary to common sense.

Expand full comment

Really, I think the problem is that the market-based theory Bryan is trying to fit mental illness into is based on the idea of markets being efficient at fulfilling preferences for rationally self-interested actors. Notice that this implies that rationality, or at least average mental functioning approaching it, is a prerequisite the entire system itself is founded upon. So mental illness, where there is a disturbance to mental functioning severe enough to impede everyday rational action, is by definition outside the domain of analysis of markets and economistic preferences. You can't use a system of analysis on a domain where it's undefined, and I think that's what's happening with any attempts to explain mental illness in this manner, and why it falls so short in terms of meshing with basic common sense.

Expand full comment

You don't seem to acknowledge that different order preferences contradict one and other. I would define a psychological illness not in terms of social approval but subjective higher order preference: a psychological illness is a behavioral pattern or psychological trait that people who possess it would rather not possess. Would you acknowledge that someone can both be addicted to heroin, and thus have the impulse to do heroin, and also have the desire to not have said impulse?

By my definition, then, for a homosexual who wishes he could be cured of his homosexuality, his sexuality is a disease. Whether we classify a trait as a disease usually depends on whether those who have it would rather not have it (except in cases where the behavior harms others, but I think that's a minority of psychological illnesses). Sometimes, as in the case of homosexuality, that's culturally determined, but for some traits, like extreme addictions, or OCD or attention deficit disorder, these are behavioral patterns that few people in any society want to keep if they can help it. You can redefine them as preferences if you please, but this seems like a purely a semantic victory. Fine, then they're preferences that people would, at a higher level, prefer not to have. The only important question is whether psychiatry is effective at curing people of these adverse 'lower order preferences.' Unless Bryan is positing that most people don't actually want to be cured of them.

Expand full comment

What about schizophrenia? Do schizophrenics just have a "preference" for hallucinating and having delusions? Do those with bipolar disorder have a "preference" for wild mood swings? The idea of mental illness just as abnormal preference breaks down when you stop just looking at like, alcoholism.

Expand full comment

Forgive me if this has already been addressed, but how do you account for situations where there's a first-order preference towards doing something, but a second-order preference towards not indulging that preference, and the illness comes about from the first order preference overriding the second order one? Take, for example, an addict. They have a preference or desire to do drugs at some moment, and they indulge in it, but at a higher level they have a genuine, internally-motivated desire to not indulge in that preference, not even because of external social sanction, but because they recognize it's destroying their body or their ability to feel pleasure outside of the drug. Or for an example more squarely within traditional mental illness, someone with OCD who in the moment has an overwhelming compulsion to perform some ritual action, where in the moment doing so is satisfying a desire, but at a higher level they wish they could get themselves to stop, because the compulsion is taking up all their time, or harming their bodies, or interfering with any other number of preferences they are unable to satisfy in the long run because of the OCD preferences being overwhelming in the moment. In this case, it seems like there is genuinely a higher level preference against expressing the addictive or mentally ill traits, and the problem that defines it as an illness is that these lower-level, more immediate preferences prevent that. I think the problem is refusing to acknowledge that preferences can exist at multiple levels of abstraction on top of each other, and recognizing that lower level preferences can interfere and prevent the attainment of more deeply held, but harder to act on, higher level preferences. Seen like this, I think you can return to the common sense model of genuinely respecting individual preferences, while recognizing that mental illnesses are an exception, because they represent a disordered conflict between lower and higher level preferences that needs to be remedied in order to allow an individual to genuinely express and be able to act on their higher level preferences (i.e. not shooting up heroin all day or washing their hands all day, which I think even the worst heroin addicts or sufferers of OCD would agree are genuine preferences they hold, despite their actions).

Expand full comment

Would you agree, Bryan, that incentives mattering is only important if the incentives are reliable over time? An alcoholic can refrain from drinking while you hold a gun to his head, but I can refrain from eating while a gun is held to my head too, and that doesn't imply that I don't need calories, just that I can postpone taking them in for some period. Incentives are a test of necessity, but not a perfect one.

Expand full comment

You'll eventually die if you don't eat or drink, the alcoholic will survive without alcohol. Mark Kleiman noted that doctors with access to opiates sometimes get addicted, and they way they came up with to treat such doctors was... monitoring with swift consequences for slipping up.

Expand full comment

Bad example. In some extreme cases alcoholics genuinely *will* die without alcohol, and the only medically responsible treatment is *less* alcohol than they’ll voluntarily consume on their own rather than *no* alcohol.

Expand full comment

Well, that's kinda illustrating the point. Is the actual crux "if going without doesn't, in itself kill you" or is there some level of delayed, non-fatal, or indirect harm that's still sufficient to constitute "need"? Or is literally anything that can be endured/foregone without fatal consequences a preference?

Expand full comment

Precisely. Simple hunger and thirst are excellent examples. Not all necessity is immediate or omnipresent. It can be put off, delayed, or temporarily suppressed. That does not make it any less of a necessity.

Expand full comment

Mental illness is neither a disease nor a preference nor a myth. It is a causal system with multiple mutually interacting parts, genetic, situational, environmental, random, that lead to poor adaptation and personal misery. You appear to have little understanding of recent empirical and conceptual work on psychopathology (hierarchical taxonomy, network analysis, harmful dysfunction a la Wakefield).

Expand full comment

For most mental illnesses, the preference changes you're talking about are actually a consequence of distorted beliefs. For example, depressed or socially anxious people have systematically inaccurate perceptions of how their social interactions go, and this helps explain why they prefer to stay home. Beliefs aren't under direct voluntary control.

Also, mental illness preferences usually go along with other involuntary symptoms. Memory problems, hallucinations, loss of IQ points.

Someone as smart as you should have thought of these points. Your persistent adherence to the Szasz view is a strong sign of biased thinking on your part.

Expand full comment

I always thought the functional definition of mental illness was any mental state and/or possibly resultant behaviors that

1. Present a danger to yourself

2. Present a danger to others

Or,

3. Impede one's ability to live a normal life

The third is the thorniest and most vague, but in my mind the definition should have nothing to do with "essence" but with presentations that justify medical intervention. The essential categorizations come later and are pragmatic. In my mind, homosexuality, however non-normative one may believe it to be, fits none of the three unless by "normal life" you mean heterosexual life.

As a side note, I don't buy that the number of gay people has skyrocketed. I think more people identify with the IDEA of fluidity of preference in their self-conception and the number of genuinely same-sex attracted individuals is roughly the same. I cannot, however, prove it.

Expand full comment

Probably lots of people will ask this, but how do you feel about things commonly categorized as mental illnesses that pass the gun-to-the-head test?

- An insomniac who cannot fall asleep despite threats of gruesome violence

- Someone with anxiety who cannot reduce physical, measurable, symptoms of anxiety (heart rate, sweat production...)

Do you believe in a wide range of psychosomatic illnesses then?

Expand full comment

It seems to me that you are engaging in a debate over vocabulary. The terms "mental illness" and "preferences" mean different things depending on the context. "Preferences" implies desirability, while "mental illness" implies deviancy or disease. "Neurodiversity" appears to be a better term - humans are neurologically diverse: some of that diversity is desirable, some is not; some is genetic, some is environmental (which doesn't mean it is voluntary or can be "unlearned"); some allows some voluntary control, some do not. Sometimes there are ways to enhance our ability to function for those with certain neurological arrangements. For instance, many forms of ADHD can be significantly improved temporarily with medication - that doesn't mean ADHD is "deviant," just that it inhibits learning and concentration. Untreated, it has a higher than normal propensity for drug addiction (i.e. self medication) and/or petty crime, as well as poverty due to low performance. Some forms of high functioning autism provide advantages with learning but interferes with social functioning. Counseling can help. Dyslexia inhibits learning, but can be overcome with modified instructional routines. These are just conditions I have some knowledge of. More serious conditions such as schizophrenia or autism require more significant care. I wouldn't call those "preferences", no matter how technically you want to define the term.

Expand full comment

> Neurodiversity: the Nazi claim of different brains of different races. There is a human nature. We are not shrimp or crows.

Expand full comment

There's a diversity of brains within any single race.

Expand full comment

Push it one step further. There are no races. Each individual has a human brain w/free will. Cannibals have the same human brain as Hollywood directors, Asian rice paddy farmers and waiters in Paris. Each individual chooses to focus his mind or evade focusing. Or some combination. Thats life.

Atlas Shrugged-Ayn Rand

Expand full comment

Race is indeed a social construct, not a biological one. Neurodiversity refers to relatively random variation in the architecture of the brain, and has nothing to do with race. Human brains are not blank hard drives. The brain's architecture is unrelated to "race," as the term "race" is based on outward physical appearance (in the US) and/or cultural practice, ancestral heritage.

Neurodiversity has the opposite meaning of the Nazi conception of race, as the Nazi's brutally enforced their ideal uniform construct. The terms neurodiversity and neurotypical are unrelated to the Nazi vision of labeling dissidents as mentally ill, as well as executing and experimenting on children who didn’t fit their vision of normal.

"Neurodiversity" is a term that labels diverse brains as normal, fully human, and intrinsically valuable. While certain neurological conditions can impose hardships on individuals who are therefore deserving of treatment, it is inappropriate to use derogatory labels to describe those individuals.

This comment conjures up a debate over the origins of autism and Asperger’s Syndrome, and whether Hans Asperger was a Nazi collaborator who sent children to their death, or whether he saved some kids. Our understanding of autism and what is no longer called Asperger’s Syndrome don’t really come from Hans Asperger. While first identifying the condition, he didn’t really publish much of value.

Expand full comment

Race and sociais a false idea of the free wills of some people. There are no social constructs.

Expand full comment

Race and social construct are false ideas. There is only the free will choice to focus and reason or evade and rationalize. Ideas are products of the minds processing of the evidence of the senses. Innate ideas are the absurdity of knowledge prior to knowledge. Neurodiversity has been invalidly used by the new racial/sexual communists to rationalize irrationality. Value is a product of a focused mind. There are no mystical intrinsic values. Suffering is not a mystical revelation obligating sacrifice.

Both Nazis and the new racial/sexual communists say there are basically biologically different types of humans, each w/their own interests and ideas. Hitler thought that Judiasm was good-for Jews. He rejected objectivity.

Expand full comment

According to your own criteria , homosexual attraction is actually not a preference . Getting an erection when seeing an attractive nude man , and failing to get an erection when seeing an attractive nude woman are involuntary reactions .

Expand full comment