If you spend too much time online (guilty!) you’re probably aware of the latest eugenics furore, sparked by Philippe Lemoine quote tweeting whether complex traits can be bred out of a population.
The original tweet from notable eugenicist Diana Fleischman, criticised the idea (from anthropologist Rebecca Sear) that complex polygenic traits can’t be bred out of a population. Things got heated when Sasha Gusev reasserted that highly polygenic conditions cannot be bred out. This provoked a backlash from people like Damien Morris, Richard Hanania, Geoffrey Miller, and Philippe himself, essentially criticising Sasha for attacking a straw-man. Of course no-one really thinks traits like criminality, alcoholism or low intellect can literally be bred out, but policies can be put in place to reduce their prevalence. These policies (eugenics) may be morally repugnant but that’s different than saying they just don’t work.
Sasha is accused of motte-and-bailey tactics. He argues that complex polygenic traits cannot practically be bred out of a population then retreats when the real argument is put forward - that policies can be used to improve the genetic health of a population.
The opposite side, encapsulated by Richard Dawkins in 2020, is that we should separate the morality of eugenics from the question of its practicality and, as animals, humans can be bred to amplify or suppress certain traits.
I enjoy the genetics discourse on Substack but I rarely offer an opinion given a) I’m not a geneticist and b) there are plenty of smart people who understand the field contributing to the discussion. However, in this instance the history of psychiatry offers valuable lessons. People seem to be unaware that we already had a movement to ‘breed out’ mental illness. So let’s revisit psychiatric eugenics.
Schizophrenia, an archetypal eugenic target
Eugenicists target schizophrenia with zeal - Diana Fleischman continually brought it up during a recent debate with Lyman Stone, arguing that sufferers should be incentivised to produce less offspring, for example by monetary compensation for long-acting contraception.
Schizophrenia is indeed substantially heritable. In most cases it has also a chronic, relapsing, lifelong course. It usually presents in adolescence or early adulthood, meaning that is has huge costs for the person affected, their family and wider society, despite having a relatively low prevalence (<1%) in the general population.
Unlike neurodiverse conditions like autism or ADHD, there are few advantages (perhaps none) associated with schizophrenia. Patients are tormented by voices. They experience delusions of being watched, persecuted, controlled by external agencies. They may believe they are being poisoned, that their mind is being read, that loved ones have been replaced by imposters. Over time, the so-called negative syndrome dominates: anhedonia, amotivation, social isolation. It is highly disabling, impairing the ability to fulfil roles in society or goals in life. If we magically had a cure for schizophrenia, it would be an enormous benefit to humanity.
Like other complex diseases (think of type II diabetes, inflammatory bowel disease, coronary artery disease), the genetic architecture of schizophrenia is highly polygenic, meaning the risk is spread over a number of genes. This includes common variations in the genome each having a tiny effect, as well as rarer variations in larger sections of DNA (copy number variants) with bigger effects. There are no single gene mutations that cause schizophrenia, it is the accumulation of genetic risk that makes an individual susceptible. Then perhaps in combination with environmental factors or by twist of fate, individuals cross a threshold and manifest with the clinical phenotype.
Given the genetic contribution to schizophrenia, it seems plausible (if unethical) that the disorder could be ‘bred out’ of a population. Would this work in practice? We don’t need to speculate, it has already been barbarically attempted, by the Nazis.
Psychiatric genocide
The more severely burdened should not propagate themselves... If we do nothing but make mental and physical cripples capable of propagating themselves, and the healthy stocks have to limit the number of their children because so much has to be done for the maintenance of others, if natural selection is generally suppressed, then unless we will get new measures our race must rapidly deteriorate.
The history of psychiatric eugenics precedes the rise of fascism and taints the reputation of our profession. This wasn’t a fringe movement, it was driven by leaders of the field who are still reverential figures. Eugene Bleuler who coined the name schizophrenia, favoured the sterilisation of people with the diagnosis. Emil Kraepelin, the father of modern psychiatry, promoted ideas of racial hygiene and ‘degeneration theory’ of the Aryan race. Kraepelin died before the Third Reich came to power but his followers played key roles in the Nazi eugenic programme.
Ernst Rüdin, mentored by Kraepelin, proposed that traits such as mental illness, criminality and alcoholism could be bred out of the German population. He help design the Nazi Law for the Prevention of Genetically Diseased Offspring, imposing compulsory sterilisation for genetic conditions such as ‘feeble-mindedness’, schizophrenia, manic-depressive insanity, genetic epilepsy, Huntington’s chorea, genetic blindness or deafness, and alcoholism. Approximately 400,000 Germans were forcibly sterilised between 1934-1939, many because of schizophrenia.
In later years, under Aktion T4, up to 300,000 were killed (sometimes euphemistically called ‘involuntary euthanasia’), for disabilities and mental illnesses. Distinguished psychiatrists of the day such as Carl Schneider, Max de Crinis, Werner Villinger, Friedrich Mauz, Friedrich Panse and Paul Nitsche were involved in the programme. Hans Asperger is accused of referring children to a clinic conducting euthanasia.
Fuller Torrey estimates between 220,000 and 269,500 people with schizophrenia were sterilised or killed, representing 73%-100% of all schizophrenia patients living in Germany between 1939 and 1945. The number of patients in psychiatric hospitals in 1945 was only 14% of what it had been in 1939. An unspeakable horror. Many were slowly starved to death in mental institutions. Others died by lethal injection or in gas chambers. What I find most shocking is that psychiatrists weren’t just the functionaries carrying out the programmes, they weren’t bystanders caught up in the wave of Nazism, or cogs in the machine. Instead, they were the intellect behind the movement and the architects of its implementation.
Thus almost all people with schizophrenia were systematically eliminated from the German population. The psychiatric genocide was ruthless and nearly complete. It was far more brutal, far more comprehensive than proposals from today’s eugenicists, which seem anodyne in comparison:
The proposals I endorsed in the debate were…
Incentives for violent criminals and individuals institutionalized for psychotic disorders to use long-term contraception or undergo sterilization.
Crucially, I was not arguing for coercion. No jailing. No force. No reproductive compulsion.
The eugenics programme instigated by the Nazis will, I hope, never be repeated. The existence of schizophrenia was more or less eradicated from a generation, using methods that even today’s most ardent eugenicist would find reprehensible.
What happened next to rates of schizophrenia in Germany may surprise the hereditarians.
Failure of Nazi eugenics
The prevalence rates of schizophrenia at the end of World War II were predictably low. This wasn’t due to the illness being bred out, of course, it was because virtually everyone with that diagnosis had been murdered. What happens when those with severe mental illness are removed from a population or forcibly sterilised? It turns out that new cases of mental disorders immediately bounce back and over time the prevalence returns to pre-war levels.
By 1965, rates of new cases of schizophrenia were higher in Germany than just about any other comparable countries. While horrifyingly effective in psychiatric genocide, the Nazi’s goal of eradicating schizophrenia from the Aryan bloodline was an abject failure. Schizophrenia, even when applying the most extreme eugenic measures, cannot be bred out of a population.
This should not surprise clinicians who see patients with schizophrenia, the vast majority (~90%) of whom do not have a family history of the illness. It won’t surprise evolutionary psychiatrists who are aware of the paradox that people with schizophrenia already have lower fertility, yet the disorder remains stable across time.
I heard the reason behind this more than 10 years ago at lecture in King’s College London by the Scottish psychiatrist and world-leading schizophrenia geneticist Professor Mick O’Donovan: most genetic risk alleles for carried not by people with schizophrenia but by people without mental disorders.
Complex disorders are complex
The reason why eugenics cannot selectively remove complex disorders like schizophrenia from a population is explained in this excellent Substack post by, you guessed it, Sasha Gusev. He discusses the threshold liability model in which there is an unobserved (latent) liability for developing an illness. This is modelled as a normally distributed bell curve, consisting of common genetic variants of tiny effect, rarer DNA copy number variants with larger effect, as well as environmental risk factors. It is only when the accumulation of risk factors crosses a certain threshold will an individual manifest with the phenotype of that illness, illustrated below.
For disorders with such imprecise boundaries as psychiatric illnesses, it is very plausible two individuals with similar liabilities for schizophrenia might not receive the same diagnosis. Furthermore, there may be individuals that have a relatively high liability to develop schizophrenia, which does not reach the threshold of a florid psychotic episode. Clinically, I think of people presenting with schizotypy who have odd beliefs that don’t quite become delusional or perceptual abnormalities that don’t quite have the quality of a hallucination.
If you think of the normal distribution curve, you would expect a greater number of individuals to have genetic risk for schizophrenia which doesn’t quite meet the threshold of illness than people diagnosed with the disorder. Said another way, most of the genetic risk for schizophrenia is harboured by unaffected individuals.
Sasha models what would happen if we removed all individuals with a 50% heritable polygenic condition with a normally distributed liability from a population, over successive generations. Even after ten successive generations, new cases of the disorder remain relatively common - highly polygenic disorders keep popping up, in contrast to monogenic (single gene) disorders which can be more easily bred out.
I’ve been using schizophrenia as an example favoured by eugenicists because I know it well. However other complex traits, like alcoholism or criminality, are likely to follow a similar threshold liability model. Even extreme measures like we saw in the 1930s would not be enough to substantially suppress new cases in a population.
Breeding out undesirable traits isn’t coming back, because it doesn’t work. However, a more socially acceptable way to reduce the genetic risk of mental illness has recently emerged, polygenic embryo screening. Could this be the new respectable face of psychiatric eugenics?
Maybe we’re all eugenicists
While I think Diana Fleischman is wide of the mark when it comes to breeding, she raises a more astute point in, You’re probably a eugenicist. In many ways modern society does accept what might be regarded as eugenic practices. For decades, antenatal screening has been employed to selectively terminate pregnancies in which certain abnormalities are detected, such as Down’s syndrome (Trisomy 21).
Despite rising maternal age (a strong risk factor for Down’s syndrome) the rates of babies born with this condition has fallen in recent years due to termination of such pregnancies. With some notable exceptions (such as the BBC documentary A World Without Down’s featuring the actor and campaigner Sally Phillips) there is very little opposition to this form of eugenics. It is almost universally accepted.
Pre-natal screening is typically used for malformations, chromosomal abnormalities and severe disabilities. However, now polygenic screening of embryos allows couples undergoing IVF to selectively choose offspring on the basis of traits of like intelligence and height, as well as risk for disorders like schizophrenia.
Typically, couples undergoing IVF will be able to select from a number of embryos to be implanted. Companies, such as Orchid and Herasight now offer embryo reports, calculating polygenic risk scores (based on adding together genetic risk markers across the genome) for a variety of traits and conditions.1
One claim is that this screening can produce a relative risk reduction of 50% for schizophrenia (though with a relatively low baseline prevalence this might not change the absolute risk by much). For most couples, shifting an absolute risk from 0.5% to 0.25% might not matter much, however, if they are concerned perhaps due to family history and already have a higher liability, shifting these risks might seem worthwhile.
This technology is not endorsed by the scientific establishment and I don’t know whether it is ready to be widely deployed but it seems everything is in place to select embryos based on their lifetime risk of developing a mental illness. The time to to discuss whether we, as a society, endorse this new form of psychiatric eugenics is now.
A recent Quillette article is a good primer on polygenic embryonic screening
I think the "you're probably a eugenicist" arguments obfuscate between selection on the phenotype and selection on the genotype. When someone selects a smart or charming partner, they are selecting on a phenotype that they think they are compatible with -- that's not eugenics! Actual eugenics in this case would be rejecting someone you find smart or charming because their sibling is a criminal and you think criminality is heritable and don't want to increase the criminal "genetic stock" of society. I think the latter, actual eugenics, would be considered pretty strange reasoning by most people. Genetic screening for monogenic conditions like trisomies falls more into a gray area but there again people are selecting on a genetic variant that has essentially a 1:1 mapping to a later expressed phenotype. Actual eugenics would be aborting a fetus because it is a heterozygous carrier of a recessive mutation that will never express itself phenotypically in that person -- and most people would not do this either.
As someone who would have been euthenaised by the Nazis, I'm for eugenics. I have a child with autism; if I had the choice, I wouldn't have knowingly had her. I have a disability, if I had the option of not being born, but being a non disabled different person, I'd choose being a different theoretical non disabled person. I understand that how a society cares for the vulnerable matters, but the reality is reducing preventable suffering from disability is my idea of a good thing. I'm not for a romanticised view of disability.
I know people think differently, but I value quality of life, and my disability has significantly decreased my quality of life. I don't think life per se is valuable when the quality of life is poor.