Many people I know who have multiple children know this because some aspects of personality show up almost the moment a baby is born. My parents and my brother marveled at how different their two children (my brother and I and then my two nieces) were from infancy. However, sociologists ignore genetics for a very good reason that has nothing to do with ignorance: you can’t do anything about it and the path to trying to is full of dragons we do not want to wake. Environment, on the other hand does potentially lend some opportunity for change, and there are enough examples of people who have broken the poverty cycle to keep optimism alive.
If humans could be prevented from lazily concluding that genetics is destiny, maybe we could include genetics in any analysis, but they can’t, and that’s how eugenics, among other awful practices were promulgated. Not just the left, but any human, would censor such practices. There was a time not so long ago that poverty and stupidity were treated as inevitably linked and poor people were forcibly or deceitfully sterilized.
Ah, another case of willful mischaracterization. Promulgation itself may fall under free speech, but (as I’m quite sure you know) that’s not what I’m talking about. I was talking about the practices that all fall under the label eugenics. Talk about eugenics all you want, but most decent people would abhor putting it into practice.
Ah, another case of imprecise wording. Promulgation does fall under free speech.
As for eugenics (m-w.com) - that's "the practice or advocacy of controlled selective breeding of human populations "
1-Note: advocacy is free speech;
2-I don't like the idea of this practice;
3-The critical questions is: by whom? If a woman decides to pay to use a genius sperm bank, should that be outlawed? No, imo, even though I don't like it. Should a national government have laws mandating eugenics for all citizens? No, imo.
Claiming to speak for all, or even for all decent people, is a risky business.
I'm not an expert in genetics but everything I've read suggests that genes play a significant role in determining intelligence. Difficult, if at all possible, to assign percentages. Very possible that that is true of criminality as well.
I think poverty is correlated - not perfectly certainly - with physical/violent crime because of the path of least resistance for the poor towards acquiring wealth.
There is no doubt however, that any mention of genetics relating to any such behaviour/personality characteristics results in immediate false smears by the left of racism, ...
A correlation between crime and poverty would suggest that crime should have been astronomical during the Great Depression. I'd be curious about any data on the question.
Coming from a family that went through divorce and remarriage with a second round of children, I can see major factors that appear to be genetic as well as environmental in terms of the outcomes in the children. I think the reason that the genetic component of the explanation is rarely discussed goes back to a few major societal taboos working together. First, the early work in eugenics and attempts genetical "engineer" away social problems in the 20's and 30's led to some very, bad policies and outcomes which have made the entire topic taboo for research and discussion. It's much easier to dismiss the genetic role as unscientific racist crap than have to consider that it may not be so. This taboo goes deeper however because it runs counter to a very important theme in Judeo-Christian belief systems which place major focus on free will and the individuals responsibility, and ability, to alter their behavior to reject sin and embrace higher social goodness in their attitudes and actions. If much of these tendencies is genetic, that undermines our belief in the triumph of the soul over our base "genetic" nature as animals. No one wants to admit even to themselves that they are "doomed" by their genetics...just as no one wants to deal with those insufferable people who think they are "blessed" in similar fashion by the circumstances of their birth. We all know that a gifted athlete from a family of such gifted athletes got good genes...but its much easier and more comforting to try to pretend it was all a matter of opportunity and hard work by the athlete that to say....that Olympian is just a better physical specimen than I am!
"If much of these tendencies is genetic, that undermines our belief in the triumph of the soul over our base "genetic" nature as animals. "
Much (and no one knows what much precisely means) doesn't mean we are predestined zombies with no free will. That is certainly not Christian theology, which does not deny genetics.
I so wish this didn't haunt every day of my life with a profoundly unwell adopted child from a profoundly unwell mother. What we do know is that although IQ appears mostly stable, we can fiddle around the edges and try to give kids born to those very psychologically unwell parents an 'enriched environment' as early as possible- loads of resources- parenting support, killer good daycare, extracurriculars, and vouchers to live in middle class neighbourhoods. 5 extra IQ points would go a very long way for those kinds of kids and save us all, esp. them a lot of agony as adults. You don't have to look hard for these kids. The foster care system makes it easy to find them.
The twin studies that the author is basing his beliefs on use simple variance decompositions that *assume away* a bunch of covariances to make the problem tractable. By construction they shunt a lot of heritability into the additive (i.e. “genetic”) bucket, which is why you always get something like 50% genetic, 50% random environment, 0% shared environment whether it’s criminality or IQ or doomscrolling. The frequent reappearance of this split is not some profound fact about nature/nurture; it’s just the result of (fairly radical) mathematical approximations that were common in early studies.
In the 21st century we can do better. Studies that use large genomic datasets coupled with methods that use exogenous variation (i.e. are plausibly measuring causation) find much lower heritabilities (<10%) for most social outcomes.
Hi Shine, first you're right that, in classical twin studies, gene-environment interactions can add into the heritability. To explain that for everyone: take two kids sent to same school, one with genes for being good at maths, the other with genes for being bad at maths. The first child ends up studying maths more and does more homework, gets more encouragment from teachers; this reinforces his maths ability. After school that child might get a job involving continued use of math. The second child gets less of that, drops maths as sonn as he can, and after his final maths class avoids maths entirely. In a classical twin study, comparing outcomes as age 20, the difference looks like a genetic effect.
This is known and understood, and is why heritability estimates change with age. E.g. From a Plomin 2016 review: "the heritability of intelligence increases significantly from 41% in childhood (age 9) to 55% in adolescence (age 12) and to 66% in young adulthood (age 17)" -- for exactly such reasons, the older the child get the more they have "created" the environment they are in.
Is it then wrong to consider this to be a genetic effect? Well, it's in line with how the world works. The "heritability" tells you about actual outcomes for people with different genes born into the world we are in. It does not tell you about a hypothetical world where kids who are good at maths get punished for it, and kids who are bad at maths get intensive one-to-one instruction.
Twin studies have been extensively critiqued, and most people in the field regard them as robust. That is, they may be somewhat out but are not wildly out. One does, though, need to be clear on what "heritability" means and the fact that it does indeed tell you about how genes interact with the range of environments in the study.
Second major reply: The only studies giving very low heritabilities for such traits (below 10%) are GWAS studies that attempt to identify each gene that is contributing to a trait (what you call "plausibly measuring causation"). The problem is that, any complex behavioural trait such a "intelligence" or "anti-social behaviour" is affected by many thousands of genes. If there are thousands of genes affecting a trait, then each gene only has a tiny effect on that trait. That makes it almost impossible to detect, since to find an effect size that small you need sample sizes that are way above current studies. This is why such studies only report genetic contributions of ~ 10%, they are only finding the few-hundred genes with the largest effects and cannot find the many thousands of genes that are also contributing. Note that even Sasha Gusev has given up arguing for these very low values.
Third major reply: in recent years, newer methods have come along -- methods such as "sibling regression" and "relatedness disequilibrium regression" -- that do not suffer from the above problem with GWAS studies, and which also use a very different methods from twin studies, so are not vulnerable to the confounding discussed above. The problem with these is that, being new, we don't (yet) have corroboration from multiple studies from different teams reporting the same answer (indeed, some of the results are not consistent with each other). These give heritabilities that are much high higher than the GWAS values, but somewhat lower than twin-studies values. Still, overall they do corroborate the twin studies in showing that genes are a major influence.
For example, here is Alex Young (one of the experts in these newer methods) in a Tweet a few months back: "IQ is probably more than 40% heritable. The only sibling based estimate we have, which should be more robust than other estimates, gave 75% (with a big SE of 20%). Plus this estimate is downward biased by assortative mating. So I would say 40% is a lower bound on the heritability of well measured IQ". This is lower than the values of 60% to 70% from twin studies, but then he explicitly says that it's a lower bound.
To sum up, most likely, genes are indeed the single biggest factor in the variation of behavioural traits, and the heritabilities are almost certainly high enough to justify the arguments made in my post.
Sasha Gusev (of Harvard Medical) has been at this for a while and written many explainers at his substack: theinfinitesimal.substack.com
I don’t know of anyone that can do it better. It takes some effort to go through these topics and some math is unavoidable, but that’s the nature of this beast. People want quick takeaways like X is 40% heritable, but those numbers don’t fall out of the data! They are the product of models and necessarily involve assumptions.
It's possible that 'right-wing commentators' are unwilling to be associated with the bizarre positions of prominent gene theorists of recent decades: those who have popularised the idea of 'the selfish gene' (while, at the same time, insisting that humans are a kind of computer); those who are convinced that altruism is an illusion, declaring that the only reason people act virtuously is because, and in the precise degree that, they are genetically related to the beneficiary; or (a recent example) that kind parenting is misguided because siblings are engaged in an all-against-all fight (https://iai.tv/articles/gentle-parenting-is-doomed-to-fail-auid-3493). (Whether it's even coherent to assert that kindness is both an illusion, i.e. doesn't actually exist, and misguided - which would presume existence - is another question.) Thanks, but no thanks.
I think that you're strawmanning people you disagree with (and also misunderstanding concepts such as the "selfish gene"). For starters, the theory of reciprocal altruism does not depend on kinship.
'We are survival machines - robot vehicles blindly programmed to preserve the selfish molecules known as genes' (Richard Dawkins, The Selfish Gene, 1976, p. x, Preface). This early use of 'programmed,' a term from computing, must have been intended - as it was taken - to liken humans to those then exciting machines. It is repeated often throughout the work.
I was thinking of so-called kin altruism: 'consanguinity allows the actor to trade some fraction of its own reproduction for that of the recipients' (Krupp and Maciejewski, The evolution of extraordinary self-sacrifice. Sci Rep 12, 90 (2022). https://doi.org/10.1038/s41598-021-04192-2022, introduction).
Thank you for asking. Everything. Genes can't be selfish, any more than they can be envious, hardworking, or liberal. Of course, when pressed, Dawkins and his fans have tended to acknowledge it was 'just a metaphor,' probably in order to catch eyes and sell books. But that means it was an unserious project in the first place, notwithstanding its massive subsequent impact.
Humans are not 'robot machines,' and they are not 'programmed.' They are organisms which have many inbuilt drives and tendencies, largely towards survival and reproduction. I would say that altruism does not require any consideration of algorithms: it's an evident outgrowth of family - largely in origin maternal - love, and social cooperation, in a highly intelligent being.
And the 'programmed' thing? As observers picked up at the time, Dawkins, with his 'robot' image, had at a stroke brought teleology back into biology - a robot requires a designer. Ever since, the problem has been to try to remove implications of teleology while retaining terms such as 'programmed,' 'built,' 'structured,' 'designed,' etc - which hasn't been done as far as I can see, and probably can't be.
In other words, you're misunderstanding all of this.
"Selfish" was always intended in the technical sense (which had been adopted in the field prior to Dawkins) of "acting to maximise its fraction of the gene pool". This was all explained in the book. It never was intended to suggest that genes had feelings.
Yes, humans are "robot machines" in that they are built by and programmed by genes. And you're right, altruism doesn't come from "consideration of algorithms" it comes from our nature, our feelings. But that nature has been programmed into us by evolution, and we can describe that evolution by algorithms.
Yes, the "robot" image invokes teleology, in that the genes build the robot in order to survive and propagate themselves into the next generation. But that is "blind" in that the genes are not thinking, deliberating entities, but they act that way because the genes that act that way have survived and propagated best. So, no, there's no problem over teleology.
Again, all of this is actually explained very well by Dawkins's books.
This interpretation might be why so many of my professional biologist friends hated Dawkins' book and hypothesis. However, as an outsider who knows nothing, I liked his book.
I understand there are other criticisms of the book and Dawkins. I even looked into it a bit at one point, but since it is not my field, the reasons did not stick with me.
I think that one of the major issues is the reductionism that eliminates one or more factors in order to privilege to the preferred factor(s).
What this article does not discuss is multigenerational emotional process, as described in natural systems theory. This is distinct from genes and functions in ways similar to how genes do. This concept does not eliminate or ignore the role of genes, but rather provides another process that may interact with genes and also is influential in shaping behavior.
Just as attributing everything to genes (as this post appears not to do) is a reductionistic error, so is denying the role of genes (which this post challenges). This denial falls short of he ideal of determining knowable facts as well as possible and developing responses that are grounded in what the evidence indicates.
However, there's not really any evidence for "multigenerational emotional processes" of this nature. Indeed, one of the central messages from twin studies and adoption studies is that "shared environment", and thus parenting, has little effect on children's personalities. Psychology has a fairly bad track record of making up theories that have little empirical support.
First, the basic issue involved has to do with variations in whether genes turn on or off, so that those with identical genes may have significantly different physical and behavioral characteristics. What accounts for these variations? This brings us to the field of epigenetics.
Second, as I understand Murray Bowen’s Natural Systems Theory, it differs in many significant ways from much of conventional psychology. So, this theory is best understood on its own terms, with its own definitions. I see Bowen Theory as providing a striking contrast to “tabula rasa.”
Relevant for this discussion, Bowen Theory attends to variations in family stress/anxiety, which relates to the biochemistry that provides environmental signals to genes regarding “turning on or turning off.” When family is understood in multigenerational terms (as Bowen Theory does), this is in parallel to what has been called “an intergenerational epigenetic inheritance.”
To oversimplify, many biological processes are related to the degree and scope of chronic inflammation. Chronic inflammation is related to chronic anxiety, which is related to the emotional processes in multigenerational families. The pathways of action are understood systemically and reciprocally, which is much broader and multidirectional than frameworks of cause and effect or of the internal processes of individuals.
Family emotional process is considered to be an aspect of epigenetic context. This process functions in ways that are similar to genes, are influenced genetically in some ways, and can have epigenetic effects.
Thus, this perspective does not deny or minimize the role of genes. Rather, it describes some family systems factors in epigenetics. Michael Kerr’s Bowen Theory’s Secrets is one of many works that describes multigenerational family emotional process.
If you’ve studied Bowen’s Natural Systems Theory and are not persuaded, you are not alone. Much work is in process to develop and test these concepts, and more needs to be done.
I am one who finds a natural systems approach to be plausible and practical. It also fits with my experience with my brothers (who are identical twins) and with my aversion to reductionism.
If you’re not familiar with Bowen Theory, you may choose to look at much more though presentations than my very brief summery and application. I thought your post was somewhat congenial with my understanding of systems theory. You may find it so as well, or you may see more contrasts than I do.
Evidence for multi-generational epigenetic effects in humans is actually pretty weak. This topic has been hyped by the media, but there's little evidence that psychological traits are affected by happenings before the child is conceived (though a mother's ill-health or malnourishment during pregnancy can matter).
While epigenetics is hugely important for the development of a body (it is what causes different genes to be expressed in different parts of the embryo, creating the different body parts), the problem for multi-generational theories is that epigenetic markers get wiped clean each generation.
But, if you think that I'm wrong, feel free to write a piece pointing to hard evidence for multi-generational epigenetics and send it to Dorian. I'd be interested to read it!
Although I agree with much of the original post, we've looking through different lenses.
I suggest consideration of a systems model and its findings, alongside of the research that is cited. This may generate some further ideas and perspectives.
There are always environmental and behavioral influences that can minimize or exacerbate predispositions.
Genetics are like boxes of crayons. One can still create decent enough pictures with a set of 24 if you didn’t get the 288, it just takes a lot more management of the process.
Here's something that 'right-wing commentators' might be more inclined to get behind - partly because we can't choose our genes, but we can make a lot of other, better, choices (it's not as if 'nurture-only,' and the genetic lottery, are the only two pathways involved):
This is clearly a fascinating topic that I know very little about. My own subjective impression based on personal observations of my family would suggest that there is a strong genetic component to human capability. I was a fairly long-term donor to two fertility clinics, partly based on my own intuition that eugenics is not a completely discredited and pointless endeavor, in spite of its ugly history.
Because of the litigiousness of American society, I have not availed myself of the websites that exist to track down any offspring that I might have from my contributions. Of course, I am curious about what has occurred. There are legal means of protection that could be employed, but I have not done so yet. Perhaps I will consider this avenue in the future.
Call this view that all people are the same blank slate "Sociological Lysenkoism". But the genetics is more complicated than that. "Behind every great fortune lies a great crime."
Having studied proxy war, terrorism and its leadership I have more than most to say. Let's take this quote, "the same kind of dispositions that lead someone to think about knocking over a liquor store at gunpoint..." Those dispositions are exactly what motivates private equity that does things like buy a controlling interest in a corporation doing well, get a massive loan using the pension fund as collateral, wait 2 years and a day, then distribute the loan funds to themselves, bankrupt the company, sell it off for parts, and leave the taxpayer holding the bag to pay the pensions. Mitt Romney's fortune is almost exactly what the pension funds he essentially stole were worth.
Psychopaths are highly represented in CEOs, dictators, and ordinary social climbers---it's a spectrum. These people see opportunities to steal, de facto and de jure, and take those opportunities without conscience. Carpe Diem! Psychopaths make it big regularly and tend to have kids.
I knew, and am still tenuously connected to the man who ran the Gamestop gambit and made a fortune, bankrupting the big boys who tried to claim that only they were allowed to rip people off like that. This man is manic depressive, and on a manic swing he's unstoppable. Now he's got mansions on multiple continents and can do anything money can buy.
Ludwig Boltzman was bipolar and eventually killed himself, but he accomplished feats of math and physics in his on phase. Isaac Newton was probably bipolar as well, and accomplished feats of at least Botlzman's scale when he was in an up phase.
In other words you have to look at the whole social group or tribe and how genes benefit or harm them. Psychopaths and sociopaths see the world in terms of power relations---its' their talent. Autism, schizophrenia, manic-depression, sociopathy, psychopathy, all have negative effects on individuals and can have terrible effects on others. But they also have huge benefits enough of the time that the group that has these traits does better, much better. Think of the rare autistic genes expressed in the stone age. Their hyper-focus on "things" and ability to see patterns may well be what allowed humans to start technological progress. On the other end of that list, psychopaths organize societies for war and conquering which makes that subgroup rich and organizes humanity into nation-states that live or die by organization and technological development.
Scholars and well-meaning people with tunnel vision have the goal of eradicating everything from autism and manic-depression to sociopathy. The same goes for gene therapy. But doing so is the most dangerous thing we could do, because the true organism of humanity is the entire group. This is true for the same reason that programmed death is perfectly conserved in all animals and in humans. Without it, evolution stops, and that guarantees death of the species.
The other missing factor is the power of ideology. We see that starkly in islam which has an ideology that explicitly is for war, and blesses assassination, mass murder, rape out-groups for recreation, torture, and slavery. The other major ideologies (we call religions) mostly have opposite values and this is reflected in their societies. (FWIW - I think that the social algorithm of islam that kills masses of rebellious young women every year, together with inbreeding, has lowered their collective IQ. We see the results of little technological accomplishment which is a fair intelligence test. Jews do the opposite, and their average IQ is higher, which means their outliers are also smarter.)
The social mass of a society depends on the interaction of genes with ideologies. Marxism is an ideology with values firmly rooted in Christian thought. Progressives also grew out of Christian reform thought.
There is a book well worth reading by Holloway, "Heavens on Earth: Utopian Communities in America, 1680-1880". Unless you understand what Holloway presents, you do not understand Marxism, because it was religious communism in North America, that was where Marx got his communist idea. But in Marx' time secular communism had shown unequivocally that it always failed. Without the over-arching ideology that transcends this world there was no basis for shared altruism to stand for lifetimes. So Marx was a bad scholar in that way. Today, the Hutterites, who became so successful in Canada that they were legally barred from buying more land for fear they would own the country, these Hutterites are the most successful example of a religious communist society that originated in the Pietist period of what became the USA and Canada.
Many people I know who have multiple children know this because some aspects of personality show up almost the moment a baby is born. My parents and my brother marveled at how different their two children (my brother and I and then my two nieces) were from infancy. However, sociologists ignore genetics for a very good reason that has nothing to do with ignorance: you can’t do anything about it and the path to trying to is full of dragons we do not want to wake. Environment, on the other hand does potentially lend some opportunity for change, and there are enough examples of people who have broken the poverty cycle to keep optimism alive.
Many sociologists - and others - ignore genetics out of fear of the left's cancelings and censorings.
If humans could be prevented from lazily concluding that genetics is destiny, maybe we could include genetics in any analysis, but they can’t, and that’s how eugenics, among other awful practices were promulgated. Not just the left, but any human, would censor such practices. There was a time not so long ago that poverty and stupidity were treated as inevitably linked and poor people were forcibly or deceitfully sterilized.
Practices such as promulgation fall under the umbrella of free speech. No, all humans would not ban that. In fact, most would allow it.
Ah, another case of willful mischaracterization. Promulgation itself may fall under free speech, but (as I’m quite sure you know) that’s not what I’m talking about. I was talking about the practices that all fall under the label eugenics. Talk about eugenics all you want, but most decent people would abhor putting it into practice.
Ah, another case of imprecise wording. Promulgation does fall under free speech.
As for eugenics (m-w.com) - that's "the practice or advocacy of controlled selective breeding of human populations "
1-Note: advocacy is free speech;
2-I don't like the idea of this practice;
3-The critical questions is: by whom? If a woman decides to pay to use a genius sperm bank, should that be outlawed? No, imo, even though I don't like it. Should a national government have laws mandating eugenics for all citizens? No, imo.
Claiming to speak for all, or even for all decent people, is a risky business.
I'm not an expert in genetics but everything I've read suggests that genes play a significant role in determining intelligence. Difficult, if at all possible, to assign percentages. Very possible that that is true of criminality as well.
I think poverty is correlated - not perfectly certainly - with physical/violent crime because of the path of least resistance for the poor towards acquiring wealth.
There is no doubt however, that any mention of genetics relating to any such behaviour/personality characteristics results in immediate false smears by the left of racism, ...
A correlation between crime and poverty would suggest that crime should have been astronomical during the Great Depression. I'd be curious about any data on the question.
Interesting point. All I can say is that this subject is hugely complex and multifactorial.
Coming from a family that went through divorce and remarriage with a second round of children, I can see major factors that appear to be genetic as well as environmental in terms of the outcomes in the children. I think the reason that the genetic component of the explanation is rarely discussed goes back to a few major societal taboos working together. First, the early work in eugenics and attempts genetical "engineer" away social problems in the 20's and 30's led to some very, bad policies and outcomes which have made the entire topic taboo for research and discussion. It's much easier to dismiss the genetic role as unscientific racist crap than have to consider that it may not be so. This taboo goes deeper however because it runs counter to a very important theme in Judeo-Christian belief systems which place major focus on free will and the individuals responsibility, and ability, to alter their behavior to reject sin and embrace higher social goodness in their attitudes and actions. If much of these tendencies is genetic, that undermines our belief in the triumph of the soul over our base "genetic" nature as animals. No one wants to admit even to themselves that they are "doomed" by their genetics...just as no one wants to deal with those insufferable people who think they are "blessed" in similar fashion by the circumstances of their birth. We all know that a gifted athlete from a family of such gifted athletes got good genes...but its much easier and more comforting to try to pretend it was all a matter of opportunity and hard work by the athlete that to say....that Olympian is just a better physical specimen than I am!
"If much of these tendencies is genetic, that undermines our belief in the triumph of the soul over our base "genetic" nature as animals. "
Much (and no one knows what much precisely means) doesn't mean we are predestined zombies with no free will. That is certainly not Christian theology, which does not deny genetics.
I agree...I was attempting to explain why people avoid the genetic component.
I so wish this didn't haunt every day of my life with a profoundly unwell adopted child from a profoundly unwell mother. What we do know is that although IQ appears mostly stable, we can fiddle around the edges and try to give kids born to those very psychologically unwell parents an 'enriched environment' as early as possible- loads of resources- parenting support, killer good daycare, extracurriculars, and vouchers to live in middle class neighbourhoods. 5 extra IQ points would go a very long way for those kinds of kids and save us all, esp. them a lot of agony as adults. You don't have to look hard for these kids. The foster care system makes it easy to find them.
I think many people are terrified of implications. Even me.
The twin studies that the author is basing his beliefs on use simple variance decompositions that *assume away* a bunch of covariances to make the problem tractable. By construction they shunt a lot of heritability into the additive (i.e. “genetic”) bucket, which is why you always get something like 50% genetic, 50% random environment, 0% shared environment whether it’s criminality or IQ or doomscrolling. The frequent reappearance of this split is not some profound fact about nature/nurture; it’s just the result of (fairly radical) mathematical approximations that were common in early studies.
In the 21st century we can do better. Studies that use large genomic datasets coupled with methods that use exogenous variation (i.e. are plausibly measuring causation) find much lower heritabilities (<10%) for most social outcomes.
Hi Shine, first you're right that, in classical twin studies, gene-environment interactions can add into the heritability. To explain that for everyone: take two kids sent to same school, one with genes for being good at maths, the other with genes for being bad at maths. The first child ends up studying maths more and does more homework, gets more encouragment from teachers; this reinforces his maths ability. After school that child might get a job involving continued use of math. The second child gets less of that, drops maths as sonn as he can, and after his final maths class avoids maths entirely. In a classical twin study, comparing outcomes as age 20, the difference looks like a genetic effect.
This is known and understood, and is why heritability estimates change with age. E.g. From a Plomin 2016 review: "the heritability of intelligence increases significantly from 41% in childhood (age 9) to 55% in adolescence (age 12) and to 66% in young adulthood (age 17)" -- for exactly such reasons, the older the child get the more they have "created" the environment they are in.
Is it then wrong to consider this to be a genetic effect? Well, it's in line with how the world works. The "heritability" tells you about actual outcomes for people with different genes born into the world we are in. It does not tell you about a hypothetical world where kids who are good at maths get punished for it, and kids who are bad at maths get intensive one-to-one instruction.
Twin studies have been extensively critiqued, and most people in the field regard them as robust. That is, they may be somewhat out but are not wildly out. One does, though, need to be clear on what "heritability" means and the fact that it does indeed tell you about how genes interact with the range of environments in the study.
Second major reply: The only studies giving very low heritabilities for such traits (below 10%) are GWAS studies that attempt to identify each gene that is contributing to a trait (what you call "plausibly measuring causation"). The problem is that, any complex behavioural trait such a "intelligence" or "anti-social behaviour" is affected by many thousands of genes. If there are thousands of genes affecting a trait, then each gene only has a tiny effect on that trait. That makes it almost impossible to detect, since to find an effect size that small you need sample sizes that are way above current studies. This is why such studies only report genetic contributions of ~ 10%, they are only finding the few-hundred genes with the largest effects and cannot find the many thousands of genes that are also contributing. Note that even Sasha Gusev has given up arguing for these very low values.
Third major reply: in recent years, newer methods have come along -- methods such as "sibling regression" and "relatedness disequilibrium regression" -- that do not suffer from the above problem with GWAS studies, and which also use a very different methods from twin studies, so are not vulnerable to the confounding discussed above. The problem with these is that, being new, we don't (yet) have corroboration from multiple studies from different teams reporting the same answer (indeed, some of the results are not consistent with each other). These give heritabilities that are much high higher than the GWAS values, but somewhat lower than twin-studies values. Still, overall they do corroborate the twin studies in showing that genes are a major influence.
For example, here is Alex Young (one of the experts in these newer methods) in a Tweet a few months back: "IQ is probably more than 40% heritable. The only sibling based estimate we have, which should be more robust than other estimates, gave 75% (with a big SE of 20%). Plus this estimate is downward biased by assortative mating. So I would say 40% is a lower bound on the heritability of well measured IQ". This is lower than the values of 60% to 70% from twin studies, but then he explicitly says that it's a lower bound.
To sum up, most likely, genes are indeed the single biggest factor in the variation of behavioural traits, and the heritabilities are almost certainly high enough to justify the arguments made in my post.
This is very interesting. Would you write a post explaining this in more detail?
Sasha Gusev (of Harvard Medical) has been at this for a while and written many explainers at his substack: theinfinitesimal.substack.com
I don’t know of anyone that can do it better. It takes some effort to go through these topics and some math is unavoidable, but that’s the nature of this beast. People want quick takeaways like X is 40% heritable, but those numbers don’t fall out of the data! They are the product of models and necessarily involve assumptions.
It's possible that 'right-wing commentators' are unwilling to be associated with the bizarre positions of prominent gene theorists of recent decades: those who have popularised the idea of 'the selfish gene' (while, at the same time, insisting that humans are a kind of computer); those who are convinced that altruism is an illusion, declaring that the only reason people act virtuously is because, and in the precise degree that, they are genetically related to the beneficiary; or (a recent example) that kind parenting is misguided because siblings are engaged in an all-against-all fight (https://iai.tv/articles/gentle-parenting-is-doomed-to-fail-auid-3493). (Whether it's even coherent to assert that kindness is both an illusion, i.e. doesn't actually exist, and misguided - which would presume existence - is another question.) Thanks, but no thanks.
I think that you're strawmanning people you disagree with (and also misunderstanding concepts such as the "selfish gene"). For starters, the theory of reciprocal altruism does not depend on kinship.
'We are survival machines - robot vehicles blindly programmed to preserve the selfish molecules known as genes' (Richard Dawkins, The Selfish Gene, 1976, p. x, Preface). This early use of 'programmed,' a term from computing, must have been intended - as it was taken - to liken humans to those then exciting machines. It is repeated often throughout the work.
I was thinking of so-called kin altruism: 'consanguinity allows the actor to trade some fraction of its own reproduction for that of the recipients' (Krupp and Maciejewski, The evolution of extraordinary self-sacrifice. Sci Rep 12, 90 (2022). https://doi.org/10.1038/s41598-021-04192-2022, introduction).
What do you think is wrong with Dawkins’s stance, as quoted?
Thank you for asking. Everything. Genes can't be selfish, any more than they can be envious, hardworking, or liberal. Of course, when pressed, Dawkins and his fans have tended to acknowledge it was 'just a metaphor,' probably in order to catch eyes and sell books. But that means it was an unserious project in the first place, notwithstanding its massive subsequent impact.
Humans are not 'robot machines,' and they are not 'programmed.' They are organisms which have many inbuilt drives and tendencies, largely towards survival and reproduction. I would say that altruism does not require any consideration of algorithms: it's an evident outgrowth of family - largely in origin maternal - love, and social cooperation, in a highly intelligent being.
And the 'programmed' thing? As observers picked up at the time, Dawkins, with his 'robot' image, had at a stroke brought teleology back into biology - a robot requires a designer. Ever since, the problem has been to try to remove implications of teleology while retaining terms such as 'programmed,' 'built,' 'structured,' 'designed,' etc - which hasn't been done as far as I can see, and probably can't be.
In other words, you're misunderstanding all of this.
"Selfish" was always intended in the technical sense (which had been adopted in the field prior to Dawkins) of "acting to maximise its fraction of the gene pool". This was all explained in the book. It never was intended to suggest that genes had feelings.
Yes, humans are "robot machines" in that they are built by and programmed by genes. And you're right, altruism doesn't come from "consideration of algorithms" it comes from our nature, our feelings. But that nature has been programmed into us by evolution, and we can describe that evolution by algorithms.
Yes, the "robot" image invokes teleology, in that the genes build the robot in order to survive and propagate themselves into the next generation. But that is "blind" in that the genes are not thinking, deliberating entities, but they act that way because the genes that act that way have survived and propagated best. So, no, there's no problem over teleology.
Again, all of this is actually explained very well by Dawkins's books.
People a lot smarter than me have thought there is 'a problem over teleology': people such as Thomas Nagel:
https://www.sciencedirect.com/science/article/abs/pii/S1369848615001740)
A lot of teachers struggle with presenting evolution without any teleology:
https://link.springer.com/content/pdf/10.1186/s12052-020-00130-y.pdf
No doubt I am foolish, but It's probably in my genes, and certainly in my upbringing, that I care about all this Darwinian and Dawkinsian stuff:
https://en.wikipedia.org/wiki/Darwinian_Fairytales
Thank you for replying.
This interpretation might be why so many of my professional biologist friends hated Dawkins' book and hypothesis. However, as an outsider who knows nothing, I liked his book.
I understand there are other criticisms of the book and Dawkins. I even looked into it a bit at one point, but since it is not my field, the reasons did not stick with me.
I think that one of the major issues is the reductionism that eliminates one or more factors in order to privilege to the preferred factor(s).
What this article does not discuss is multigenerational emotional process, as described in natural systems theory. This is distinct from genes and functions in ways similar to how genes do. This concept does not eliminate or ignore the role of genes, but rather provides another process that may interact with genes and also is influential in shaping behavior.
Just as attributing everything to genes (as this post appears not to do) is a reductionistic error, so is denying the role of genes (which this post challenges). This denial falls short of he ideal of determining knowable facts as well as possible and developing responses that are grounded in what the evidence indicates.
However, there's not really any evidence for "multigenerational emotional processes" of this nature. Indeed, one of the central messages from twin studies and adoption studies is that "shared environment", and thus parenting, has little effect on children's personalities. Psychology has a fairly bad track record of making up theories that have little empirical support.
Thanks for the opportunity to develop my comment.
First, the basic issue involved has to do with variations in whether genes turn on or off, so that those with identical genes may have significantly different physical and behavioral characteristics. What accounts for these variations? This brings us to the field of epigenetics.
Second, as I understand Murray Bowen’s Natural Systems Theory, it differs in many significant ways from much of conventional psychology. So, this theory is best understood on its own terms, with its own definitions. I see Bowen Theory as providing a striking contrast to “tabula rasa.”
Relevant for this discussion, Bowen Theory attends to variations in family stress/anxiety, which relates to the biochemistry that provides environmental signals to genes regarding “turning on or turning off.” When family is understood in multigenerational terms (as Bowen Theory does), this is in parallel to what has been called “an intergenerational epigenetic inheritance.”
To oversimplify, many biological processes are related to the degree and scope of chronic inflammation. Chronic inflammation is related to chronic anxiety, which is related to the emotional processes in multigenerational families. The pathways of action are understood systemically and reciprocally, which is much broader and multidirectional than frameworks of cause and effect or of the internal processes of individuals.
Family emotional process is considered to be an aspect of epigenetic context. This process functions in ways that are similar to genes, are influenced genetically in some ways, and can have epigenetic effects.
Thus, this perspective does not deny or minimize the role of genes. Rather, it describes some family systems factors in epigenetics. Michael Kerr’s Bowen Theory’s Secrets is one of many works that describes multigenerational family emotional process.
If you’ve studied Bowen’s Natural Systems Theory and are not persuaded, you are not alone. Much work is in process to develop and test these concepts, and more needs to be done.
I am one who finds a natural systems approach to be plausible and practical. It also fits with my experience with my brothers (who are identical twins) and with my aversion to reductionism.
If you’re not familiar with Bowen Theory, you may choose to look at much more though presentations than my very brief summery and application. I thought your post was somewhat congenial with my understanding of systems theory. You may find it so as well, or you may see more contrasts than I do.
Thanks for stimulating my thinking.
Evidence for multi-generational epigenetic effects in humans is actually pretty weak. This topic has been hyped by the media, but there's little evidence that psychological traits are affected by happenings before the child is conceived (though a mother's ill-health or malnourishment during pregnancy can matter).
While epigenetics is hugely important for the development of a body (it is what causes different genes to be expressed in different parts of the embryo, creating the different body parts), the problem for multi-generational theories is that epigenetic markers get wiped clean each generation.
Here is a blog post from 2018 by Kevin Mitchell, reviewing the evidence for multi-generational epigenetics in humans (I find this rebuttal quite persuasive): http://www.wiringthebrain.com/2018/05/grandmas-trauma-critical-appraisal-of.html
Jerry Coyne has written about this multiple times on his website, and is also sceptical. See the posts under this link: https://whyevolutionistrue.com/category/epigenetics/
But, if you think that I'm wrong, feel free to write a piece pointing to hard evidence for multi-generational epigenetics and send it to Dorian. I'd be interested to read it!
Although I agree with much of the original post, we've looking through different lenses.
I suggest consideration of a systems model and its findings, alongside of the research that is cited. This may generate some further ideas and perspectives.
There are always environmental and behavioral influences that can minimize or exacerbate predispositions.
Genetics are like boxes of crayons. One can still create decent enough pictures with a set of 24 if you didn’t get the 288, it just takes a lot more management of the process.
My genes tell me not to listen to you all
So why do you? You certainly post enough, which tells me that you aren’t listening to your genes.
its my genes ....don't blame me
Here's something that 'right-wing commentators' might be more inclined to get behind - partly because we can't choose our genes, but we can make a lot of other, better, choices (it's not as if 'nurture-only,' and the genetic lottery, are the only two pathways involved):
https://andrewperlot.substack.com/p/our-iqs-are-stagnant-but-this-still
The path to wisdom is paved with reality.
This is clearly a fascinating topic that I know very little about. My own subjective impression based on personal observations of my family would suggest that there is a strong genetic component to human capability. I was a fairly long-term donor to two fertility clinics, partly based on my own intuition that eugenics is not a completely discredited and pointless endeavor, in spite of its ugly history.
Because of the litigiousness of American society, I have not availed myself of the websites that exist to track down any offspring that I might have from my contributions. Of course, I am curious about what has occurred. There are legal means of protection that could be employed, but I have not done so yet. Perhaps I will consider this avenue in the future.
Call this view that all people are the same blank slate "Sociological Lysenkoism". But the genetics is more complicated than that. "Behind every great fortune lies a great crime."
Having studied proxy war, terrorism and its leadership I have more than most to say. Let's take this quote, "the same kind of dispositions that lead someone to think about knocking over a liquor store at gunpoint..." Those dispositions are exactly what motivates private equity that does things like buy a controlling interest in a corporation doing well, get a massive loan using the pension fund as collateral, wait 2 years and a day, then distribute the loan funds to themselves, bankrupt the company, sell it off for parts, and leave the taxpayer holding the bag to pay the pensions. Mitt Romney's fortune is almost exactly what the pension funds he essentially stole were worth.
Psychopaths are highly represented in CEOs, dictators, and ordinary social climbers---it's a spectrum. These people see opportunities to steal, de facto and de jure, and take those opportunities without conscience. Carpe Diem! Psychopaths make it big regularly and tend to have kids.
I knew, and am still tenuously connected to the man who ran the Gamestop gambit and made a fortune, bankrupting the big boys who tried to claim that only they were allowed to rip people off like that. This man is manic depressive, and on a manic swing he's unstoppable. Now he's got mansions on multiple continents and can do anything money can buy.
Ludwig Boltzman was bipolar and eventually killed himself, but he accomplished feats of math and physics in his on phase. Isaac Newton was probably bipolar as well, and accomplished feats of at least Botlzman's scale when he was in an up phase.
In other words you have to look at the whole social group or tribe and how genes benefit or harm them. Psychopaths and sociopaths see the world in terms of power relations---its' their talent. Autism, schizophrenia, manic-depression, sociopathy, psychopathy, all have negative effects on individuals and can have terrible effects on others. But they also have huge benefits enough of the time that the group that has these traits does better, much better. Think of the rare autistic genes expressed in the stone age. Their hyper-focus on "things" and ability to see patterns may well be what allowed humans to start technological progress. On the other end of that list, psychopaths organize societies for war and conquering which makes that subgroup rich and organizes humanity into nation-states that live or die by organization and technological development.
Scholars and well-meaning people with tunnel vision have the goal of eradicating everything from autism and manic-depression to sociopathy. The same goes for gene therapy. But doing so is the most dangerous thing we could do, because the true organism of humanity is the entire group. This is true for the same reason that programmed death is perfectly conserved in all animals and in humans. Without it, evolution stops, and that guarantees death of the species.
The other missing factor is the power of ideology. We see that starkly in islam which has an ideology that explicitly is for war, and blesses assassination, mass murder, rape out-groups for recreation, torture, and slavery. The other major ideologies (we call religions) mostly have opposite values and this is reflected in their societies. (FWIW - I think that the social algorithm of islam that kills masses of rebellious young women every year, together with inbreeding, has lowered their collective IQ. We see the results of little technological accomplishment which is a fair intelligence test. Jews do the opposite, and their average IQ is higher, which means their outliers are also smarter.)
The social mass of a society depends on the interaction of genes with ideologies. Marxism is an ideology with values firmly rooted in Christian thought. Progressives also grew out of Christian reform thought.
There is a book well worth reading by Holloway, "Heavens on Earth: Utopian Communities in America, 1680-1880". Unless you understand what Holloway presents, you do not understand Marxism, because it was religious communism in North America, that was where Marx got his communist idea. But in Marx' time secular communism had shown unequivocally that it always failed. Without the over-arching ideology that transcends this world there was no basis for shared altruism to stand for lifetimes. So Marx was a bad scholar in that way. Today, the Hutterites, who became so successful in Canada that they were legally barred from buying more land for fear they would own the country, these Hutterites are the most successful example of a religious communist society that originated in the Pietist period of what became the USA and Canada.