Will Genetic Psychiatry Cause Another Holocaust?
My mildly optimistic response to Eric Turkheimer's "Gloomy Prospect" on whether genetics could someday help psychiatrists treat patients better.
The Conventional Wisdom on nature vs. nurture is implausibly contradictory: inherited genetic differences between individuals are thought to be very powerful in causing cancer and the like, but inherited genetic differences between races are believed to have been proven by The Science to be utterly superficial and insignificant. A DNA test, it is widely assumed, is less good at distinguishing an individual’s socially constructed race than even your lying eyes.
So, for some purposes, Nature is all-powerful, but for other closely-related purposes, Nurture is all-powerful.
It’s confusing when you think about it, so most people with three digit IQs don’t think hard about it. They don’t want to be one of those horrible people on the SPLC’s blacklist of “Extremists,” now do they?
In truth, most working left of center biological scientists who might help the public unravel this confusing ball of twine are prudently reticent in the crucial nature vs. nurture discourse.
One of the few working left of center geneticists to try to come up with a new general perspective on nature and nurture is U. of Texas Virginia [my apologies] behavioral geneticist Eric Turkheimer. His Gloomy Prospect Blog on Substack outlines his innovation: his response to hereditarians having all this fun new genome scan technology and data to play around with is that he doesn’t think we will ever be much able to untangle the influence of nature and nurture on human behavior. After all, anyone who thinks about it for long would realize that human behavior is hugely complicated. (I wholeheartedly agree that it is complicated.) Hence he calls his pessimism about the future effectualness of the behavioral sciences “The Gloomy Prospect.”
I’ll respond to his latest post
Essence Medicine and Outcome Medicine
A paradox
below, both above and below the paywall.
As I’ve been pointing out for years, one of the most formative events of 21st Century conventional wisdom on the central intellectual question of Nature vs. Nurture was the July 2000 White House ceremony celebrating the “completion” of the Human Genome Project.
It was actually completed later during the GWB Administration, much like Apollo 11 happened 6 months into the Nixon Administration. Still, Clinton, not surprisingly, wanted credit for overseeing it into the home stretch when success had become close to assured. Anyway, Clinton presiding over the Human Genome ceremony made the speeches much more impactful on upper midwit opinion than if GWB had been in charge of the shindig.
In particular, the late entrepreneur Craig Venter’s speech was a masterpiece of salesmanship at threading the politically fraught needle on Nature and Nurture: Yes, he implied, genetics would soon cure cancer and countless other ills the flesh is heir to. But, the Human Genome Project has also discovered more good news: no, race does not exist genetically!
“DR. VENTER: … The method used by Celera has determined the genetic code of five individuals. We have sequenced the genome of three females and two males, who have identified themselves has Hispanic, Asian, Caucasian or African American. We did this sampling not in an exclusionary way, but out of respect for the diversity that is America, and to help illustrate that the concept of race has no genetic or scientific basis.
“In the five Celera genomes, there is no way to tell one ethnicity from another.”
Venter, however, was pulling the global audience’s leg. Almost all of the genome decoded by Celera was … J. Craig Venter’s own personal very white genome.
Of course, what has happened since 2000 is basically the opposite of the message that so many people took away and still believe. In reality, we’ve made only slow (but hardly trivial) progress against cancer and the like (after all, your genes didn’t evolve in order to kill you). It appears that a lot of cancers might be due to fairly random developmental errors. (That doesn’t mean genetic advances will never be helpful against cancer, but it means that collecting enough data will be challenging than it had seemed when a popular prejudice was that cancer was caused by One Big Bad Gene.)
But … we now have astonishing DNA tests that can calculate your racial ancestry percentages with a high degree of accuracy. 23andMe, for instance, offers 45 different racial ancestry categories, as Emil Kirkegaard mentions in his new big paper doing surname analysis using 23andMe racial ancestry DNA data.
On the other hand, 23andMe’s stock is low now because it hasn’t made much progress on its ambitions to use its database of 14 million genomes to cure cancer and the like.
So, let me post a large chunk of Turkheimer’s most recent post because his pessimistic perspective on the future of the sciences of human behavior is worth pondering. It’s perhaps the most scientifically sophisticated left-of-center view of nature and nurture at present:
Essence Medicine and Outcome Medicine
A paradox
May 13, 2026
This is another post inspired by Kathryn Paige Harden’s Aston-Gottesman lecture, though I don’t really mean to put it on Paige— it’s just something she mentioned in passing, and probably thinks more about in her new book, which I haven’t read yet. So I don’t mean to attribute any of the ideas I am refuting to her specifically.
Turkheimer’s former grad student Kathryn Paige Harden (whose first book I reviewed here) has a new book out entitled Original Sin: On the Genetics of Vice, the Problem of Blame, and the Future of Forgiveness. Like Turkheimer, I haven’t read it yet.
Paige was talking about her ongoing empirical project on the GWAS of “externalizing behavior,” a latent variable or DSM diagnosis capturing the common variance in a broad domain of mental health problems, from drug and alcohol use to outright criminality. (Someone once described the difference between internalizing and externalising this way: Internalizing is about being unhappy; externalizing is about making other people unhappy.)
Anyway, in the course of talking about what might be accomplished by studying the molecular genetics of externalizing, she mentioned drug development, and the possibility that by understanding the genetics of conduct disorder we might be able to develop a pharmaceutical treatment for it. My stomach turned.
Of course, we already have a lot of pharmaceutical treatments for some types of conduct disorder, such as little boys who can’t sit still in the classroom being diagnosed with ADHD and being put on Ritalin to get them to stop squirming.
Maybe genetic science could help come up with better diagnoses and better pills in the future … and, in particular, better awareness of the times when pills would be worse than, “First, do no harm.”
It conjured a terrible image for me: troubled adolescents acting out in ways their parents don’t like, being given a diagnosis of a vague entity like conduct disorder, stigmatized as having defective genes or brains, and prescribed drugs to “cure” them. Yuck.
Uh, yeah, that’s what American society has been doing for several decades now, but often starting before adolescence. I am not particularly happy about this, but in my writings, I mostly offer a certain degree of skepticism but little in the way of specific advice for people facing these kind of problems, because I’m not an expert.
I certainly don’t feel yucky for hoping that advances in genetics will help Americans do what they are already doing, but in the future doing it more intelligently and thus with fewer bad side effects. I also don’t have the levels of optimism that genetic scans will prove to be cancer panaceas that will let us live to be 150 that Bill Clinton and Tony Blair expressed at the Human Genome Project ceremony in 2000. But, we have been in an era of a sort of Moore’s Law for gene scans getting much cheaper, so I figure we ought to ride it for awhile and see how far genetic science can take us toward solving various problems that it is reasonable to try to alleviate, such as clinical depression and so forth.
But then my opinion took a surprising turn. My daughter, an elementary school teacher, happened to be at the talk. I was expressing my misgivings afterward, and she pushed back. Was I opposed to prescribing stimulants for people with ADHD? (No I am not.) Am I opposed to prescribing SSRIs for depressed people? No again. In fact I generally endorse a libertarian view of psychiatric drugs— it it works for people and they consent to it, then it is up to them, not their doctor and certainly not to me.
What’s going on here? The distinction in my title is between what I call essences and outcomes. Outcomes, things like divorce (my standard example) or most forms of psychopathology, are recognizable classes of behavior that are the termination of a long interactive developmental process, to which genes and other biological structures may have originally been inputs, but which are etiologically independent of them by the time they occur. That is how something like divorce can be “heritable” (ie, correlated with genetic differences) without there being anything resembling divorce genes. Essences are behavioral entities that are direct representations of structures at a lower level of analysis. Aphasia is an essence, a behavioral manifestation of brain damage resulting from stroke. Huntington’s disease is an essence, a behavioral manifestation of a single dominant gene.
As far as anyone knows, all psychiatric disorders are outcomes, not essences. Psychiatric geneticists think I am nuts for saying so, but I genuinely believe there is no biological explanation of depression, any more than there is a biological explanation for divorce or bankruptcy.
Let’s talk about “divorce genes” using a real world situation.
Like Harden, I like to focus on lesbian eugenics (e.g., I first wrote about how Jodie Foster and Melissa Etheridge chose their respective sperm donors in VDARE in 2000), since nice liberals have a hard time calling lesbians Nazis. So liberals are more likely to engage their brains in thinking about the choices facing lesbians in choosing a sperm donor rather than just name-calling in sputtering rage.
Say that you are a lesbian in a committed gay marriage and now you and your lesbian partner want to have a baby. You both value monogamy highly because your respective childhoods were both made unhappy by both of your sets of parents’ divorces.
Your top two choices for a sperm donor are a happily married man vs. a man with the same qualities you value, except he has been divorced three times, at least twice largely due to his own fault as he admits. So, if all else is equal in the two willing donors, which one do you choose: the married man or the divorced man?
It strikes me as sensible for the lesbians to go with the married man in this case.
There are other situations, though, in which a gene scan would be all you have to go by. Obviously, GWAS analysis is, at present, not terribly strong at predicting much of any human behavior. However, these are early days.
In any case, the wisdom of an early 20th Century sportswriter named Hugh E. Keough, which was often cited by the more famous journalists Damon Runyon and Franklin P. Adams, seems relevant:
The race is not always to the swift, nor the battle to the strong, but that is the way to bet.
Turkheimer continues:
I won’t repeat that argument here, but here is the thing: just as genes can be correlated with something like depression that is not fundamentally genetic, there can be effective drugs for depression even though depression isn’t fundamentally biological or neurological.
The standard way to put this, one of those chestnuts that gets repeated a lot because it is irrefutable, is that headaches are not caused by an aspirin deficiency. Garden variety headaches, in fact, are outcomes, without specific causes (Meehl: specific genetic etiologies). Aspirin works by improving the phenomenology of headache, not by curing its cause.
Here is the paradox. In science and medicine, discoveries based on the elucidation of essences are the most prestigious and exciting, and drugs based on those discoveries are generally the most effective. Strep throat is an essence caused by the streptococcus bacillus, and penicillin cures that essence. Taking aspirin for fever and sore throat might be helpful, but is secondary to the essential medical treatment.
In psychopathology, the search for essential explanations is called the “medical model.” Biological psychiatry in general, and now GWAS in particular, are ongoing efforts, misguided if you ask me, to search for essential explanations of behavioral entities that are in fact outcomes.
I think many intellectuals like Turkheimer are still hung up denying Plato and Aristotle’s interest in “essences.” Therefore, they get over emotional about nature vs. nurture questions because genes are associated with “essences.”
The human race only recently developed even modestly sophisticated math for thinking about statistics. The fairly simple correlation coefficient was first unveiled by Francis Galton in 1888, two centuries after Newton had worked out the more complicated field of calculus. So, humans are still lagging at thinking realistically about statistical phenomenon. (For example, I'm constantly being denounced as an essentialist when I’m obviously somebody whose first turn of mind is statistical.)
But if you are thinking not like Plato and Aristotle, but like Keough and Runyon, then it would seem that genetic science may bring about better ways to bet. I’ll give an example involving psychiatric medicine below the paywall.
In general I think this is a waste of the time and effort of some very talented people and a waste of a great deal of money, but as a scientific effort it isn’t really up to me. If people are convinced that there is something to be found in the GWAS of conduct disorder, let them try. In this domain, I am libertarian about the science.
But the situation is reversed for the therapeutics. Let’s say someone found a drug that worked for conduct disorder like aspirin words for headaches, SSRIs for depression, or Adderall for ADHD. (Again, if you ask me, there is nothing wrong with the genes or brains of people with ADHD. It is a behavioral outcome, and stimulants help with the symptoms, not the causes.) Although an outcome-type drug for conduct disorder would almost certainly wind up being mis- and over-used (like stimulants and SSRIs) I could also imagine it being useful if used judiciously. Some kid endangering himself and others with out of control behavior, and a drug calms him down, helps him relate to others in meaningful ways. The key to using such a drug in a non-evil way would be specifically NOT believing that it is fixing a defect in the kid’s brain caused by deficiencies in his genes.
Well, it could also be caused by his “refrigerator mother,” or by an infection, or by a bump on the head (Roald Dahl attributed his creativity to a concussion suffered in a warplane crash), or by lead pollution, or by systemic racism, or by miasmas, or whatever. But what if in the future gene scans help you bet better on how to improve his conduct?
This is the fundamental paradox of behavior genetics, the topic of my Dobzhansky address at the meeting last year. The field thinks its task is to find essential explanations of behavioral syndromes, and is often willing to exaggerate its findings to convince people it is making progress, but those purported discoveries, and especially their exaggeration, again and again have led the field into biolgical determinism, eugenics, and racism. It is very difficult— I am inclined to think it is impossible— to have what seems like success according to the standards of natural science while maintaining a humanist attitude toward problematic behavior.
After all, anti-hereditarian nurturists like Stalin and Pol Pot had such an admirable track record when they got in power.
Likewise, practically everybody hired by the government to work with children in 2026 hates biological determinism, eugenics, and racism. Yet, together, they sure wind up putting a lot of boys on Ritalin.
In general, my suggestions for avoiding stupid extremes of policy like Stalinism and Hitlerism is to encourage the best minds to publicly discuss their views. Otherwise, the worst minds will be the ones to get attention.
Unfortunately, back in 2017 when Charles Murray was invited on Sam Harris’s podcast, Turkheimer, Richard E. Nisbett, and, to a lesser extent, Harden got mad that Murray was allowed to air his views, even though they admitted he was right about 4 of what they saw as the 5 big questions about race and IQ. I wrote at length in 2017 about this:
Vox: "Charles Murray Is Once Again Peddling Junk Science About Race and IQ"
After the paywall, I will give a pragmatic reason for why I’d like to see geneticists progress in understanding the genetic roots of mental health disorders:
Paywall here.
