“Speaking for myself, I have never once suggested that the heritability of anything was going to turn out to be zero”
Look, I know you think that there is some smaller heritability for behavioral traits and that this is a sign of being reasonable and that anyone who suggests otherwise is an ideologue, but I have to say that your field has made some assumptions about heritability existing (whether large or small), but not zero for decades. This is a bad assumption, and frankly, bad science to start with and, in my view, has caused a lot of wasted time. All of your studies start with the assumption that traits are heritable (because of twin studies) and it’s a matter of to what extent. Studies should start with the null hypothesis and whether your results are better than the null (and statistically significant). However, the studies to date have never once tested against actual null traits. Instead, there is simply an implied null of zero. I’m sure you know that even a null trait (chop sticks, social security number, zip code, etc.) will not give a zero result. Or, if you think they will, show me the money. I’ve been making this point for 25 years. So if IQ and zip code and major depression all get similar results for “heritability,” what can you say about the results? Your field has an ideological bent and I would argue that, consciously or unconsciously, testing against the null is taboo. This has led to endless attempts to explain why results haven’t stood up to replication. This is why it took 15 years to see that candidate studies were giving spurious results and this is why GWAS won’t die, even though they have produced almost nothing of substance. There is no need to argue the point, really. If you want to show that Jay Joseph (and me) are wrong in suggesting the null, it is within your power to prove it. Otherwise, you are just spouting opinions.
Behavioral genetics doesn't assume psychological and behavioral traits are heritable because of twin studies. Those are observations that modify a prior belief of existing heritability and/or genetic effects on behavior. These traits are assumed to be heritable because of evolutionary theory. If cognitive abilities or behaviors ever mattered to survival and reproductive success (which of course they did and still do), then the genome shapes them in meaningful ways. The complexity arises because these things could be understood as "extended phenotypes." The genome has reciprocal interactions with species-level conditions like required parental care/provisioning and sociality.
There is also an abundance of indisputable empirical evidence showing specific genetic changes predictably causing specific neuro/psych/behavioral outcomes, e.g. de novo/inherited pathogenic variants in MECP2.
Everything but the actual study to prove me wrong is subjective opinion. “Genes do things” and genetic variants determine our behavioral differences are two different points. If you are making assumptions based on theory, regardless of whether you think the theory is correct, you still should be testing against the null. That is really basic.
There's another sleight of hand going on near the end, but I don't regard Turkheimer as ever having been some kind of 'anti-hereditarian' champion--he has always been closer to an unwitting centrist. The only thing he consistently gets correct is that heritability doesn't necessarily tell you much about historical malleability, but even here he fails to account for the fuller ramifications (because he is a psychologist first and foremost in his training). I also find it curious that people like to claim that Sasha Gusev is the forefront of genetic representation--no, he's just a very public figure with a kind of niche celebrity status. This is not the same thing. Not a bad guy, though, but certainly not the end-all-be-all authority for any side.
The aforementioned sleight of hand, mind you, consists in saying something like this: "if I had different genes, would I have a different IQ? The answer is probably, yes."
Setting aside the slippery caveat afforded thanks to the word 'probably', the broader sleight of hand arises with the assumed insinuation that there is something fundamental to the genes which primarily causes the dictum of one's genetic *potential* to produce a different IQ score. This conflation of discrepancies is itself a logically extrapolated leap and a conflation of the genetic set with a linear disparity in potential between other genetic sets, because the question of potential as it concerns genetics is its own issue--there's nothing necessarily true about the idea that if, say for example, you took someone's exact life history, with every last circumstance engendered therein, and somehow magically cloned this life history as a counterfactual and grafted it onto a person with a different set of genes, their IQ would be *incommensurably different* with regards to genetic *potential*. All you'd be doing in this magical, hypothetical, counterfactual circumstance is identifying a difference in the reaction norm specific to that person. Which is to say, yes, let's suppose that the outcome shifts in accordance with the genetic difference even if the life-environmental history transpires identically from one hypothetical iteration of a person to the next... this in no way precludes the possibility that another different IQ score could manifest in that very same person, even with their different set of genes and their initially differentiated IQ score.
Why?
Because the potential of that person still technically hasn't been addressed. All you're doing in such a scenario is conflating the outcome with the notion of potential, when in actuality it is entirely conceivable that the difference is merely an issue of the norm(s) of reaction.
Thus, if we continue along this exampled path, let's reimagine things in an initially similar conceptual vein--but we'll then reach a point where the logic deviates from Turkheimer's: Let's suppose that someone experiences their own life history and its environmental circumstances and arrives at having an IQ of 100. Let's then magically hypothesize an exact circumstantial recreation of that same story, except with a different set of genes for this 'someone', and let's see a result which causes an outcome of an IQ which is now 120.
Counter-intuitively, as far as the current science actually demonstrates (due to the fact that such empiricism is an entirely agnostic and conjectural issue as far as the capacity of any evidential resolution is currently going) it *does not follow* that this 20 point difference is an absolute value of capacity particular to the possibility space of one genetic set compared to another. All that this can actually be said to show (in a definitive sense) is that one of the reaction norms was more positively amenable to the environmental circumstances with that arrangement of genes; it does not suggest that no other set of environmental circumstances could produce a substantially different outcome--importantly, this means that in theory, the first set of genes in the example which resulted in an IQ of 100 could still theoretically ascertain an IQ of 120 via different environmental *requirements* compared to the exampled second set of genes which resulted in an immediate IQ of 120... Thereby allowing for equivalency despite the different genetic sets, but only via *a different set of environments*. Likewise, the second set of genes which resulted in an immediate IQ of 120 could potentially have a depressed score of 100 but in the context of a different array of environmental requirements. The shaping of the genome needn't necessarily be significantly divergent in terms of its range of commensurable possibilities, especially insofar as there is great universal benefit in trending towards specific utilities. Of course, sometimes specific mutations can cause specific mechanical outcomes, but this is not always going to hold as meaningfully analogous across the spectrum of the usually expected range apparent to the nature of a given trait. So, all that is actually *empirically* describable is that the environmental sensitivity requirements differ from one set of genes to another. The issue of which set of genes inherently contains 'greater potential' relative to the hypothetical range of environmental minima and maxima is still altogether inconclusive.
It is entirely conceivable, contrary to the popular hereditarian narrative, as well as the current academic consensus which is that of the soft-hereditarian sleight of hand found in the works of people like Turkheimer (who want to have their metaphorical cake and eat it too... it is somehow at once the case that we'll never be able to precisely mechanically understand behavioral causation yet somehow we can safely infer genetic influence based merely on assumed inferences mediated through correlational formulas of calculus which are replete with their own epistemic assumptions), that the actual degree of true 'upper difference in potential' between people is negligible or highly canalized (assuming the satisfaction of unique optimizations particular to the requirements of *individuals*--which likewise necessitates that there cannot exist any kind of univocal or linearly singular 'x' factor at the heart of environmental conditions and the variance of their outcomes), with phenotypic variance generally only constituting the temperaments which manifest relative to different norms of reaction.
The recent 'convergence towards the 30% figure' is just another ephemerally passing theory in the wind, as 80% once was, with the finality of the debate altogether unsettled. Turkheimer, being a classical scientific realist, would like to assume that the process of veracious discovery follows a trajectory in which we are readily excavating ourselves ever closer to the refinement of truth, naively reposed away from the issue of paradigm shifts. But paradigm shifts can disrupt the ways in which we even conceive of the question of the truth to begin with, retroactively reframing it. As such, the real truth will probably not see the light of day for hundreds of years still, as the real truth would require testing norms of reaction across a practically infinite variety of ranges. This can be 'simulated' through a massive ocean of assumptive guesswork baked into the calculus posited by the geneticists, but this is not necessarily correspondent to correct values due to the required assumptions imputed into the formulation, and thus, the resulting data. A determinately accurate calculation might be possible with advances in quantum computing and AI working in tandem to emulate such a thing, but it's a far-off matter of the future.
All of this also assumes that it is even conceptually coherent to attempt the latticing of heritability onto a trait which is only theoretically assumed to exist and which may or may not even be measuring what it purports to be measuring to an ambiguous degree (IQ).
For now, the only thing which will truly differ compared to the past is the empirical matter of what public perception looks like, and in that respect, Turkheimer is indeed correct--many people will accept the prevailing narrative (or lean into full hereditarianism), and few people, even academically, will be able or willing to dispute it, even despite the inner contentiousness of its claim to 'truth'.
Well, I wouldn't call Sasha Gusev a hereditarian at all. He's definitely more on the opposite side, he's just not particularly radical about it despite what the Twitter crowd might insist.
Perhaps you meant 'anti-hereditarian'? It would depend on the field. The issue can't be accurately conceptualized as exclusive to the question of population (or molecular) genetics, given the overt intersection with psychometrics, so it's already fundamentally interdisciplinary (and I'd argue that it entails many other facets which are oft unspoken or unrecognized in popular discourse, i.e. anthropology, philosophy of science, neuroscience, AI, and so on). If you want an example on par for quality with Sasha from the academic-psychometric avenue, you could check out Peter Hans Schönemann... although I'd argue he's better than Sasha, not merely equal to. But you have to recognize that scholarly achievement isn't exclusively predicated around the cachet of the university someone occupies--this would be a bit of a just world fallacy. Plenty of great researchers never went to Harvard, but of course credentials matter in some capacity, so I'll preface this with the clarification that Peter is not a crank by any means and is still conventionally credible.
If you want someone more immediately biologically centered (who is also 'academically prestigious'), you could check out David Scott Moore or Richard C. Francis, for a start. I have other authors in mind as well but they're much more 'controversial', and I'm going to assume that you are not looking for heterodox material in your preferences, so I'll leave it at that for now.
Oh, I misinterpreted your initial comment to suggest an equivalency between Sasha and the association of Hereditarianism, but you meant that you wanted an equivalently accredited figure from the opposite side of the aisle, I see. if you want 'relatively' academically prestigious hereditarians who are contemporary (and not total charlatans like Jordan Lasker), you could check out someone like Razib Khan, or Meng Hu, or the ole' Cochran and Harpending folks. There are older names out there who are still kicking; I've neglected to mention them here because they're not in the public eye much nowadays. Needless to say, I'm not in agreement with any of these people.
I'm no expert, but let me repeat a point I've made before. Imagine a mom and dad who have their first child at 30 and their 5th child at 40. Won't their egg and sperm cells at age 40 be different in ways that are likely to affect the embryos? And won't they be different parents with their fifth child? And won't their parenting methods be different? And won't the fifth child be affected by the four siblings? (Perhaps an older sibling will become a surrogate parent.)So, isn't w/i family heritability hugely impacted by environment?
the set of all factors we don't know about. "Luck" can be viewed as a feature that incorporates quite a lot of cascading interactions that elude identification, much less measurement.
idk, sounds dubious. what difference-making existents are there that explain outcomes aside from biology-chemistry-physics? all of it is luck btw, because nothing controls reality. reality doesn't need to be made to happen, it just happens. what's he suggesting by 'that something else'? magic, aliens, astrology, higher-dimensiomal beings... or what? (btw all of these, if real, would also just be mechanisms like genes lol, so still all luck).
idk what you mean by mystery here, nor do i think that there's any strong reason to infer that there are difference-making existents that 'are mysterious'. if you just mean ignorance, that's not dubious, just trivially the default state, as everyone is extremely more ignorant about most things than not.
Faustroll contends that some phenomena are destined to elude any possibility of definitive resolution. A maxim implicit to the proper study of 'Pataphysics. The laws governing exceptions.
“Speaking for myself, I have never once suggested that the heritability of anything was going to turn out to be zero”
Look, I know you think that there is some smaller heritability for behavioral traits and that this is a sign of being reasonable and that anyone who suggests otherwise is an ideologue, but I have to say that your field has made some assumptions about heritability existing (whether large or small), but not zero for decades. This is a bad assumption, and frankly, bad science to start with and, in my view, has caused a lot of wasted time. All of your studies start with the assumption that traits are heritable (because of twin studies) and it’s a matter of to what extent. Studies should start with the null hypothesis and whether your results are better than the null (and statistically significant). However, the studies to date have never once tested against actual null traits. Instead, there is simply an implied null of zero. I’m sure you know that even a null trait (chop sticks, social security number, zip code, etc.) will not give a zero result. Or, if you think they will, show me the money. I’ve been making this point for 25 years. So if IQ and zip code and major depression all get similar results for “heritability,” what can you say about the results? Your field has an ideological bent and I would argue that, consciously or unconsciously, testing against the null is taboo. This has led to endless attempts to explain why results haven’t stood up to replication. This is why it took 15 years to see that candidate studies were giving spurious results and this is why GWAS won’t die, even though they have produced almost nothing of substance. There is no need to argue the point, really. If you want to show that Jay Joseph (and me) are wrong in suggesting the null, it is within your power to prove it. Otherwise, you are just spouting opinions.
Behavioral genetics doesn't assume psychological and behavioral traits are heritable because of twin studies. Those are observations that modify a prior belief of existing heritability and/or genetic effects on behavior. These traits are assumed to be heritable because of evolutionary theory. If cognitive abilities or behaviors ever mattered to survival and reproductive success (which of course they did and still do), then the genome shapes them in meaningful ways. The complexity arises because these things could be understood as "extended phenotypes." The genome has reciprocal interactions with species-level conditions like required parental care/provisioning and sociality.
There is also an abundance of indisputable empirical evidence showing specific genetic changes predictably causing specific neuro/psych/behavioral outcomes, e.g. de novo/inherited pathogenic variants in MECP2.
Everything but the actual study to prove me wrong is subjective opinion. “Genes do things” and genetic variants determine our behavioral differences are two different points. If you are making assumptions based on theory, regardless of whether you think the theory is correct, you still should be testing against the null. That is really basic.
There's another sleight of hand going on near the end, but I don't regard Turkheimer as ever having been some kind of 'anti-hereditarian' champion--he has always been closer to an unwitting centrist. The only thing he consistently gets correct is that heritability doesn't necessarily tell you much about historical malleability, but even here he fails to account for the fuller ramifications (because he is a psychologist first and foremost in his training). I also find it curious that people like to claim that Sasha Gusev is the forefront of genetic representation--no, he's just a very public figure with a kind of niche celebrity status. This is not the same thing. Not a bad guy, though, but certainly not the end-all-be-all authority for any side.
The aforementioned sleight of hand, mind you, consists in saying something like this: "if I had different genes, would I have a different IQ? The answer is probably, yes."
Setting aside the slippery caveat afforded thanks to the word 'probably', the broader sleight of hand arises with the assumed insinuation that there is something fundamental to the genes which primarily causes the dictum of one's genetic *potential* to produce a different IQ score. This conflation of discrepancies is itself a logically extrapolated leap and a conflation of the genetic set with a linear disparity in potential between other genetic sets, because the question of potential as it concerns genetics is its own issue--there's nothing necessarily true about the idea that if, say for example, you took someone's exact life history, with every last circumstance engendered therein, and somehow magically cloned this life history as a counterfactual and grafted it onto a person with a different set of genes, their IQ would be *incommensurably different* with regards to genetic *potential*. All you'd be doing in this magical, hypothetical, counterfactual circumstance is identifying a difference in the reaction norm specific to that person. Which is to say, yes, let's suppose that the outcome shifts in accordance with the genetic difference even if the life-environmental history transpires identically from one hypothetical iteration of a person to the next... this in no way precludes the possibility that another different IQ score could manifest in that very same person, even with their different set of genes and their initially differentiated IQ score.
Why?
Because the potential of that person still technically hasn't been addressed. All you're doing in such a scenario is conflating the outcome with the notion of potential, when in actuality it is entirely conceivable that the difference is merely an issue of the norm(s) of reaction.
Thus, if we continue along this exampled path, let's reimagine things in an initially similar conceptual vein--but we'll then reach a point where the logic deviates from Turkheimer's: Let's suppose that someone experiences their own life history and its environmental circumstances and arrives at having an IQ of 100. Let's then magically hypothesize an exact circumstantial recreation of that same story, except with a different set of genes for this 'someone', and let's see a result which causes an outcome of an IQ which is now 120.
Counter-intuitively, as far as the current science actually demonstrates (due to the fact that such empiricism is an entirely agnostic and conjectural issue as far as the capacity of any evidential resolution is currently going) it *does not follow* that this 20 point difference is an absolute value of capacity particular to the possibility space of one genetic set compared to another. All that this can actually be said to show (in a definitive sense) is that one of the reaction norms was more positively amenable to the environmental circumstances with that arrangement of genes; it does not suggest that no other set of environmental circumstances could produce a substantially different outcome--importantly, this means that in theory, the first set of genes in the example which resulted in an IQ of 100 could still theoretically ascertain an IQ of 120 via different environmental *requirements* compared to the exampled second set of genes which resulted in an immediate IQ of 120... Thereby allowing for equivalency despite the different genetic sets, but only via *a different set of environments*. Likewise, the second set of genes which resulted in an immediate IQ of 120 could potentially have a depressed score of 100 but in the context of a different array of environmental requirements. The shaping of the genome needn't necessarily be significantly divergent in terms of its range of commensurable possibilities, especially insofar as there is great universal benefit in trending towards specific utilities. Of course, sometimes specific mutations can cause specific mechanical outcomes, but this is not always going to hold as meaningfully analogous across the spectrum of the usually expected range apparent to the nature of a given trait. So, all that is actually *empirically* describable is that the environmental sensitivity requirements differ from one set of genes to another. The issue of which set of genes inherently contains 'greater potential' relative to the hypothetical range of environmental minima and maxima is still altogether inconclusive.
It is entirely conceivable, contrary to the popular hereditarian narrative, as well as the current academic consensus which is that of the soft-hereditarian sleight of hand found in the works of people like Turkheimer (who want to have their metaphorical cake and eat it too... it is somehow at once the case that we'll never be able to precisely mechanically understand behavioral causation yet somehow we can safely infer genetic influence based merely on assumed inferences mediated through correlational formulas of calculus which are replete with their own epistemic assumptions), that the actual degree of true 'upper difference in potential' between people is negligible or highly canalized (assuming the satisfaction of unique optimizations particular to the requirements of *individuals*--which likewise necessitates that there cannot exist any kind of univocal or linearly singular 'x' factor at the heart of environmental conditions and the variance of their outcomes), with phenotypic variance generally only constituting the temperaments which manifest relative to different norms of reaction.
The recent 'convergence towards the 30% figure' is just another ephemerally passing theory in the wind, as 80% once was, with the finality of the debate altogether unsettled. Turkheimer, being a classical scientific realist, would like to assume that the process of veracious discovery follows a trajectory in which we are readily excavating ourselves ever closer to the refinement of truth, naively reposed away from the issue of paradigm shifts. But paradigm shifts can disrupt the ways in which we even conceive of the question of the truth to begin with, retroactively reframing it. As such, the real truth will probably not see the light of day for hundreds of years still, as the real truth would require testing norms of reaction across a practically infinite variety of ranges. This can be 'simulated' through a massive ocean of assumptive guesswork baked into the calculus posited by the geneticists, but this is not necessarily correspondent to correct values due to the required assumptions imputed into the formulation, and thus, the resulting data. A determinately accurate calculation might be possible with advances in quantum computing and AI working in tandem to emulate such a thing, but it's a far-off matter of the future.
All of this also assumes that it is even conceptually coherent to attempt the latticing of heritability onto a trait which is only theoretically assumed to exist and which may or may not even be measuring what it purports to be measuring to an ambiguous degree (IQ).
For now, the only thing which will truly differ compared to the past is the empirical matter of what public perception looks like, and in that respect, Turkheimer is indeed correct--many people will accept the prevailing narrative (or lean into full hereditarianism), and few people, even academically, will be able or willing to dispute it, even despite the inner contentiousness of its claim to 'truth'.
Well, I wouldn't call Sasha Gusev a hereditarian at all. He's definitely more on the opposite side, he's just not particularly radical about it despite what the Twitter crowd might insist.
Perhaps you meant 'anti-hereditarian'? It would depend on the field. The issue can't be accurately conceptualized as exclusive to the question of population (or molecular) genetics, given the overt intersection with psychometrics, so it's already fundamentally interdisciplinary (and I'd argue that it entails many other facets which are oft unspoken or unrecognized in popular discourse, i.e. anthropology, philosophy of science, neuroscience, AI, and so on). If you want an example on par for quality with Sasha from the academic-psychometric avenue, you could check out Peter Hans Schönemann... although I'd argue he's better than Sasha, not merely equal to. But you have to recognize that scholarly achievement isn't exclusively predicated around the cachet of the university someone occupies--this would be a bit of a just world fallacy. Plenty of great researchers never went to Harvard, but of course credentials matter in some capacity, so I'll preface this with the clarification that Peter is not a crank by any means and is still conventionally credible.
If you want someone more immediately biologically centered (who is also 'academically prestigious'), you could check out David Scott Moore or Richard C. Francis, for a start. I have other authors in mind as well but they're much more 'controversial', and I'm going to assume that you are not looking for heterodox material in your preferences, so I'll leave it at that for now.
Oh, I misinterpreted your initial comment to suggest an equivalency between Sasha and the association of Hereditarianism, but you meant that you wanted an equivalently accredited figure from the opposite side of the aisle, I see. if you want 'relatively' academically prestigious hereditarians who are contemporary (and not total charlatans like Jordan Lasker), you could check out someone like Razib Khan, or Meng Hu, or the ole' Cochran and Harpending folks. There are older names out there who are still kicking; I've neglected to mention them here because they're not in the public eye much nowadays. Needless to say, I'm not in agreement with any of these people.
I'm no expert, but let me repeat a point I've made before. Imagine a mom and dad who have their first child at 30 and their 5th child at 40. Won't their egg and sperm cells at age 40 be different in ways that are likely to affect the embryos? And won't they be different parents with their fifth child? And won't their parenting methods be different? And won't the fifth child be affected by the four siblings? (Perhaps an older sibling will become a surrogate parent.)So, isn't w/i family heritability hugely impacted by environment?
Robert Hunter, from a long-ago interview (paraphrased): "there's that something else that isn't nature, it isn't nurture- what is it?"
You mean THE Robert Hunter? Do you have a source? I'd be interested if you can find it.
yes. It's a great, really deep interview. I think I can find the source with some work.
Randomness, development noise, luck many more similar terms, that get compounded. Is there any other thing ?
the set of all factors we don't know about. "Luck" can be viewed as a feature that incorporates quite a lot of cascading interactions that elude identification, much less measurement.
Goodness, that is quite a quote.
???
yes, indeed. It is a mystery.
idk, sounds dubious. what difference-making existents are there that explain outcomes aside from biology-chemistry-physics? all of it is luck btw, because nothing controls reality. reality doesn't need to be made to happen, it just happens. what's he suggesting by 'that something else'? magic, aliens, astrology, higher-dimensiomal beings... or what? (btw all of these, if real, would also just be mechanisms like genes lol, so still all luck).
You think mysteries are dubious?
Positivism always starts from a default position of Enclosure. smh
idk what you mean by mystery here, nor do i think that there's any strong reason to infer that there are difference-making existents that 'are mysterious'. if you just mean ignorance, that's not dubious, just trivially the default state, as everyone is extremely more ignorant about most things than not.
Faustroll contends that some phenomena are destined to elude any possibility of definitive resolution. A maxim implicit to the proper study of 'Pataphysics. The laws governing exceptions.