Genetics and Environment or a Middle Path?
September 7, 2021 7:05 AM   Subscribe

 
I was in grad school (education) when Murray's book came out, and I read it carefully and recognized why, despite its relative plausibility, it was going to be the focus of some terrible debates because he was willing to go into recommendations for policy despite the narrow nature of the results he analyzed. Likewise with Hayden. Scientists are often naive about the role of denial and of societal agendas in the way research results are received. In conversations with the contractor who's doing my bathrooms, he often cites "research papers" to support his beliefs. He knows that much about science. But he doesn't know how much tunnel vision there is in science and how often it is a work in progress as well as a heated argument. People cherry-pick what they read and how they interpret it, no matter how advanced their research knowledge is. For instance, my advisor refused to let me use the term "classroom management" in my dissertation, in spite of the fact that I was literally looking at the ways teachers manage their classrooms. The phrase was just too fraught with inequity and power dynamics for my advisor.

Really good article.
posted by Peach at 7:58 AM on September 7, 2021 [10 favorites]


I was in grad school (education) when Murray's book came out, and I read it carefully and recognized why, despite its relative plausibility, it was going to be the focus of some terrible debates

Murray is still around and tweeting racistly.

At some point, Harden has to set aside her caveats and assert that sheer genetic luck plays a causal role in outcomes that matter: “If people are born with different genes, if the genetic Powerball lands on a different polygenic combination, then they differ not just in their height but also in their wealth.”

Right, we might just be measuring genes for physical beauty and good hair, which probably cause wealth and success on their own. If you have a genetically good immune system, you probably spend more time in school and less time home sick. Once you're measuring huge swathes of genes you could be picking up huge sets of things like that.

GWAS results are not “portable”: a study conducted on white Britons tells you little about people in Estonia or Nigeria.

Ah. So... what use are they, then? If they aren't valid in more than one population, what are they even measuring?
posted by BungaDunga at 8:10 AM on September 7, 2021 [11 favorites]


If we have to shield ourselves from science, because the right might use it badly, then we've lost the culture war and may as well turn ourselves in. The left--broadly construed--has had a real problem with accepting the difference between "this might be real" and "this might play badly in political discussions." I've been thinking about that while watching some of the folks I follow on Twitter react to this book. "Oh no, the right might do a racism if we talk about genetics"...as though the right were not going to plunge ahead with that no matter what, because that's one of their fundamental philosophical planks.

What came to mind reading this article--well, one of the things--was how political discussions around gayness went a few decades ago. That particular battle has been won on the cultural front, I believe; everyone assumes that gayness is a singular thing that will ultimately be shown to have a genetic cause. I remain as skeptical now, as I was back in the 90s when I first encountered the argument, because the terms of the debate were so focused on the politics. "We can't say homosexuality is a choice because the right will call it a sin and we'll lose rights!" So instead we went with "we're born this way" and the right called it a sin and aside from a supreme court decision and a few local laws, there are still no protections in the US, and we have lost the ability to talk about the role of taste, of personal aesthetic sense, of private history, of environment, as causative. (Which is not to say that there could not be a genetic component of homosexuality! But further discussion is foreclosed, and has been for decades.)
posted by mittens at 8:15 AM on September 7, 2021 [48 favorites]


In a recent paper, Harden asked, “You only have one life to live, but if you rewound the tape and started anew from the exact same genetic and environmental starting point, how differently could your life go?” She continued, “Overall, twin research suggests that, in your alternate life, you might not have gotten divorced, you might have made more money, you might be more extraverted or organized—but you are unlikely to be substantially different in your cognitive ability, education, or mental disease.” In the past few years, Harden noted, new molecular techniques have begun to shore up the basic finding that our personal trajectories owe a considerable debt to our genes.

Maybe I need to read the original paper, but I don't exactly understand the point here. If other-me could have an entirely different life (based on relationships or income), but I was still the same "me", then how can she say that "our personal trajectories owe a considerable debt to our genes"?

And in the end of the article she talks about "the heritability of divorce" anyway, so I sort of get the sense that Harden is saying that many things are hereditary, and yet environment can change things anyway? And vice versa? Which I think most people would agree with and would just quibble about the extent. So there's a lot of words here but it all seems very vague.

If genes play a significant role in educational attainment, then perhaps we ought to design our society such that you don’t need a college degree to secure health care.

Perhaps we should do that either way?


In 2018, she wrote an Op-Ed in the Times, arguing that progressives should embrace the potential of genetics to inform education policy. Dorothy Roberts, a professor of law, sociology, and Africana studies at the University of Pennsylvania, strongly disagreed: “There’s just no way that genetic testing is going to lead to a restructuring of society in a just way in the future—we have a hundred years of evidence for what happens when social outcomes are attributed to genetic differences, and it is always to stigmatize, control, and punish the people predicted to have socially devalued traits.”

I think that's very important to consider, and again I didn't really ever get a point from Harden. How would genetics inform educational policy? We already know education isn't equitable. The Op-Ed has 2 points by my read:
  • "First, these genetic results reveal the injustice of our so-called meritocracy.... But success in our educational system is partially a result of genetic luck."
  • Second, knowing which genes are associated with educational success will help scientists understand how different environments also affect that success...Understanding which environments cause improvements in children’s ability to think and learn is necessary if we want to invest wisely in interventions that can truly make a difference.
and then ends saying "acknowledging the reality of genetic differences between people is a necessary step for us to ensure a more just society."

But, just like the healthcare quote above, couldn't we say that society shouldn't be a meritocracy? She keeps saying we need make the world more just... can't we simply do that?
posted by Nonsteroidal Anti-Inflammatory Drug at 8:19 AM on September 7, 2021 [11 favorites]


Views associated with the “hereditarian left” have also been articulated by the psychiatrist and essayist Scott Alexander and the philosopher Peter Singer.

Okay, but Scott Alexander is not at all part of the mainstream left (he's deep into the "rationalist" world) and Singer has his own stuff. Not people I'd expect to be ideal in terms of convincing the left generally,
posted by BungaDunga at 8:30 AM on September 7, 2021 [9 favorites]


"The Left" struggles mightily with any expression of intersectionality, and the role of genetics not wholly versus but intertwining with environment (cultural and political are every bit as much of the human organism's environment as everything else) is definitely some of that.

A particularly cynical thought at:
An admirer of Darity’s work—especially on reparations for slavery—Harden was surprised that she’d elicited such rancor from someone with whom she was otherwise in near-total political agreement.

A man in a position of authority is angry at other men (Murray et al); can't hit at those other men, but can take his anger out on a woman and exert power over her future. The hell you say!
posted by Drastic at 8:50 AM on September 7, 2021 [3 favorites]


Deep appreciation for mittens's comment. I also find it pleasing to type mittens's
posted by elkevelvet at 8:57 AM on September 7, 2021 [8 favorites]


A man in a position of authority is angry at other men (Murray et al); can't hit at those other men, but can take his anger out on a woman and exert power over her future. The hell you say!

...Darity is a senior black economist who holds a chair at Duke, addressing what he plainly believes to be racist nonsense being circulated by a white colleague. I do not believe I would analyze this as you do.
posted by praemunire at 9:02 AM on September 7, 2021 [19 favorites]


This is a nurture-nature debate in 2021 without mention of epigenetics, which is itself a sort of middle path.
posted by grokus at 9:16 AM on September 7, 2021 [13 favorites]


Not at all an expert, or up on the literature, but hasn't there been work in recent years suggesting that genetics/genes are far less determining of an individual organism's physical and cognitive development? I.e., that genes are less important on their own and that a narrow focus just on genetics (which seems to be Harden's M.O.) obscures the way they interact with other bodily systems and external factors?
posted by Saxon Kane at 9:17 AM on September 7, 2021 [4 favorites]


Let me be blunt - Charles Murray is a bigot who has made a career of putting a pseudoscientfic veneer on that bigotry. Harden complains that people keep bringing up The Bell Curve, yet felt obligated to defend Murray - and seems unable to link the two points together. As long as the field has this issue with turning a blind eye to the bigots in their midst, they're going to continue to have people look askance at them.
posted by NoxAeternum at 9:20 AM on September 7, 2021 [21 favorites]


...Darity is a senior black economist who holds a chair at Duke, addressing what he plainly believes to be racist nonsense being circulated by a white colleague. I do not believe I would analyze this as you do.

Also worth noting that, far from not hitting back at Murray, Darity has been critiquing him and fellow white supremacists for the better part of forty years, including at least one collection of essays published right after The Bell Curve.
posted by Glegrinof the Pig-Man at 9:21 AM on September 7, 2021 [19 favorites]


I find it extremely difficult to have meaningful conversations on these topics because we so frequently need to fall back on claiming certain things are more or less impactful than we previously thought - but our previous thoughts were all personalized heuristics to begin with, so you telling me that now you believe genetics are more impactful than before tells very little about the starting or the ending state.
It all matters, and it is clear that we are refining the understanding of the degree and timing to which it matters, but we lack a consistent way to have discussion about it.
posted by meinvt at 9:21 AM on September 7, 2021 [6 favorites]


The whole topic (in the larger society, I don't mean here in this thread) would benefit from some basic visual diagrams. Showing two overlapping distribution graphs of Attribute X within two different populations (I started to write "bell curves" but then thought better of it), which overlap not entirely but by, say, 99%, might help people literally visualize the difference between large and small, er, differences - between groups as a whole, and between individuals in groups.
posted by PhineasGage at 9:26 AM on September 7, 2021 [1 favorite]


Harden complains that people keep bringing up The Bell Curve, yet felt obligated to defend Murray

But she didn't defend Murray. Direct quote:
“This is a very Christian thing I’m about to say, but it reminds me of the episode where Jesus is tempted by Satan in the desert,” she told me, in Bozeman. “There’s just enough truth in Murray that if you say, ‘This is all wrong,’ you paint yourself into a corner where you say intellectually dishonest things. Jesus has to say, ‘This part is true, and this part is false.’ ”
Furthermore:
The Vox piece, which Harden and Turkheimer wrote with the social psychologist Richard Nisbett, was headlined “Charles Murray is once again peddling junk science about race and IQ.”
posted by airmail at 9:43 AM on September 7, 2021 [9 favorites]


I read this and was surprised two things that were not mentioned.

1) The absence of any discussion of epigenetics. Heck we are learning more and more about how the gut microbiome may be the most important aspect of general health; and to not even discuss epigenetics seems weird. Especially if the GWAS studies are pointing to clusters of genes that are associated with the various things under consideration like educational attainment etc.

2) The fact that there is more genetic heterogeneity within any 'race' you want to specify; than between the 'races'. I put it in quotes simply because the term keeps changing in terms of who is included. The subtext in the article was how this will affect the discussion on race; without talking about this seems baffling.
posted by indianbadger1 at 9:49 AM on September 7, 2021 [27 favorites]


The whole topic (in the larger society, I don't mean here in this thread) would benefit from some basic visual diagrams. Showing two overlapping distribution graphs of Attribute X within two different populations (I started to write "bell curves" but then thought better of it), which overlap not entirely but by, say, 99%, might help people literally visualize the difference between large and small, er, differences - between groups as a whole, and between individuals in groups.

I am not sure that will help anything, Charles Murray's book has loads of charts. There's an IQ bell curve chart split by race on page 279.
posted by BungaDunga at 9:54 AM on September 7, 2021 [2 favorites]


Personally I think there's plenty of reason to be skeptical scientifically and morally about behavioral genetics (and evolutionary psychology, its bastard offspring). It's disingenuous to depict "progressive" rejection of aspects of this area of research as closing one's eyes and covering one's ears and ignoring "reality". Furthermore, the use of this kind of research by right wing (and "rationalist") groups is a real thing to be discussed and acted against. Science does not happen in a vacuum, sealed away from society, no matter how much we'd like to believe it.

Sorry for all the scare quotes.
posted by AlbertCalavicci at 10:02 AM on September 7, 2021 [5 favorites]


Without spelling out specific reasons for scientific skepticism, that's recapitulating the viewpoint the book and article are questioning.
posted by PhineasGage at 10:05 AM on September 7, 2021 [1 favorite]


I might not understand Darity but I understand the position of someone who has seen IQ testing used throughout history to subjugate and oppress people on a massive scale and be frustrated that this young white intellectual believes her nuanced take is the best, as if oblivious to the last 70 years of Black children being thrown into remediation, their schools underfunded, and their neighborhoods destroyed because conservative policy makers, financiers, et al conveniently adopted the most racist take on a nuanced conversation.

This conversation is not about being afraid of conservative takes, it's knowing and having experienced what will happen if you open up this can of worms. How many decades has it even been since No Child Left Behind was used to worsen educational outcomes? The Bell Curve and it's ilk aren't some kind of past dinosaur of US history, it's still an active, ongoing issue.

For example, beginning in 2002, the federal government spent almost a billion dollars on something called the Healthy Marriage Initiative, which sought to reduce marital conflict as a way of combatting poverty and juvenile crime. Harden was not surprised to hear that the policy had no discernible effect.

Right, because I'm sure this is completely a genetics-defined matter and not something that could be explained by bad design, worse implementation, and a completely reductive, pseudoscientific (and covertly Christian) assumptions about relationships that only someone like W Bush and his Focus on the Family friends could come up with with its embedded class/race/culture-agnostic definitions of "success."

she pointed to a study from last year as proof of concept

One single study, a profile in the New Yorker, and a pop-sci adjacent book that is supposed to help explain human nature. What do you call white privilege?

“They don’t want to hear that differences in I.Q. matter because they’re highly predictive of differential success in life

Could it be because IQ testing itself is rooted in white supremacist understanding of 'intelligence', is currently used to advance anti-Black policy agendas, and they'd rather do away with the whole thing? Could it be that even if things like IQ (or BMI) is useful at population levels, we've seen that using it at the individual interventional level is at best useless and at worst harmful, and we'd rather not go down that path of harm again?

There’s a politically correct left that’s still not open to these things.” Stuart Ritchie, an intelligence researcher, told me he thinks that Harden’s book might create its own audience: “There’s so much toxicity in this debate that it’ll take a long time to change people’s minds on it, if at all, but I think Paige’s book is just so clear in its explanation of the science.”

There's a progressive left that's cognizant of the historical discourse on this matter, the harm it's caused, who aren't interested in re-opening what was thought to be a closed discourse. This 'clear explanation' is of an unproven, hypothetical, and completely mechanistic understanding of human behavior and outcomes, and diplomatically what you could call 'naive' and frankly what you call ivory tower intellectualism at its worst. Removed and abstracted from society, things like this could be useful in a few decades when there are many more studies out and if you aren't promoting your reductive, pop-sci book on a touring circuit.

But fuck this white liberal bullshit that cares so little about the consequences of what is being platformed. How anybody has so much privilege that they think they can control how much harm their platforming and points bring into the world is something that I will never understand.
posted by paimapi at 10:12 AM on September 7, 2021 [40 favorites]


Because we're talking about Charles Murray, here's Shaun's obligatory 2.5 hour run down of scholarly failures so profound they can only have been deliberate. In general this article is notable for its failure to comment on the quality of any of this research or the extent to which any of these perspectives are supported. It's just straight-ahead IDW culture-war bullshit, and makes no effort to do anything more than "extremist bad, centrist good!" I'd like to see it re-written by some one with some interest in the science and without an axe to grind.

This "left" they talk about isn't stubbornly refusing to talk about genes out of a misguided wish to defend an "orthodoxy" or because it's too emotional to do science, what it's trying to do is keep the door closed on eugenics, phrenology, and the overriding desire to find biological reasons to recreate and reinforce social hierarchies, because these things are bad. And it doesn't seem like anyone can think of any use for this research other than the reinforcement of social hierarchies.

The essence of science isn't methodology or discovery, it's motivation and the motivation of anyone whose work girds the loins of eugenicists and fascists is suspect. I have no reason to think Harden is doing that deliberately but the article doesn't suggest that she has any good reasons to do it otherwise. She doesn't even seem to have thought about it carefully, or considered the pushback from people who's lives are already informed by the racial, political and economic reinforcement of social hierarchies. She seems (on the basis of this piece alone) to be extraordinarily naive, at best.

Harden was raised in a conservative environment, and though she later rejected much of her upbringing, she has maintained a convert’s distrust of orthodoxy.

Ah, well. At least her IDW speaking circuit credentials are shaping up nicely.
posted by klanawa at 10:30 AM on September 7, 2021 [24 favorites]


“Twin studies.” Call me when they randomly kids to environments rich with ACEs. I’m not going to be hold my breath. Of course parts of your genome are going to be correlate with aspects of your phenotype. It’s a fucking banal observation. As we speak researchers labor anonymously to uncover complicated developmental pathways leading to adverse outcomes like school failure and being diagnosed with ADHD. It’s almost always a fucking maze of environmental and generic factors. No fucking shit.

But if you couch it a certain way, you can ride the implication to fame and notoriety. So we cycle endlessly back to the 20th century phrenologists and their crap. It’s pointless; it tells us nothing about human development or how the brain works; and it furthers racist ideas in our society. I’d file it under the category of “not even wrong” except I know books like these exist to fuel a debate that’s more about our cultural and political conflicts than about understanding human cognition, behavior, and development.
posted by eagles123 at 10:36 AM on September 7, 2021 [3 favorites]


And it doesn't seem like anyone can think of any use for this research other than the reinforcement of social hierarchies.

I think Ezra Klein had an interesting point about this when discussing Murray with Sam Harris:
This is something you brought up earlier when you brought up that quote from Murray about luck, and I think it’s an important conversation. I think that if you follow Murrayism on this, if you were doing it without the political commitments he brings to it, it actually takes you to a very radical and interesting place.

If you say that our IQ is genetic and environmental, but at any rate, it’s not our fault, because we don’t choose either one of those, and there’s not much we can do about it. Not just our IQ, but something you’ve said is that, you know, a lot of traits come down like this — the big five personality traits, determination. Look, you can connect genetic inheritance to divorce. I think it’s a .2 or .4 correlation. So, if you begin to believe that, actually you begin to ask the question of, should, do we deserve what we have?

Should society be vastly more redistributed than it actually is? Should we be much less within this construct that what we’re getting, we’re getting because of hard work and determination and intelligence and the application of our talents? In fact, we need to move to something that is, I’m not literally advocating this, but more in the range of full socialism.

What I think is so interesting about the way he takes this debate — and I recognize this is not somewhere you took the debate, but I do think this is a useful thing to talk about — is that if you really did believe things immutable, if you really did believe that this was our inheritance both environmental and genetic and we can’t do much about it, then I think the implications of that are radical, and the implications aren’t that you take away help from people. It’s that you say pretty much what all of us has is primarily illegitimate. We didn’t do anything to earn it. I just happened to be born with the collection of talents that got me where I am. And as such, what we should spread around in society is much more vast.

Funnily enough, I don’t ever see people take that attitude on this. Again, the history of these ideas in America is they tend to be used to justify the status quo, not radically more generous versions of the status quo, but I do think that’s interesting, and I don’t understand why people don’t take that leap. I think that the implication of this is, it’s luck, and if you want to believe that — and, again, I don’t believe they’re immutable, I don’t think that’s what the evidence shows — but if you do believe they’re luck, I don’t think it takes you where he went in your conversation.
posted by Pyry at 10:46 AM on September 7, 2021 [18 favorites]


Yeah, I think Klein is noticing that the pro side on this seems to want a predictive way to categorize people rather than a responsive way to attend to their needs regardless of the conditions that put them in a state of need. But the pro side of this is pretty much universally opposed to the concept of "equality of outcomes" (a phrase they invented and falsely attributed to Marx, who opposed the idea), so Klein's idea conflicts with the core motivations of that side: they wouldn't even be interested in this topic if they thought it would lead to equality.
posted by klanawa at 10:55 AM on September 7, 2021 [11 favorites]


Without spelling out specific reasons for scientific skepticism, that's recapitulating the viewpoint the book and article are questioning.

It's worth noting that the article rather unquestioningly adopts the assumption that she is on the side of "the science." It's clear she believes that. I'm not sure it's so obvious.
posted by praemunire at 11:16 AM on September 7, 2021 [8 favorites]


Again, not taking sides, but that's the argument she's making. She's citing genetic research and psychological studies, i.e. what are commonly considered as "science." So to say things like "it's not so obvious" or it's "skeptical scientifically" - without being more specific - is just assertion rather than actually arguing against her views.
posted by PhineasGage at 11:24 AM on September 7, 2021 [4 favorites]


I learned about the existence of this article through Twitter, where a lot of the scientists in my own particular Twitter bubble are ripping into it as spectacularly irresponsible science journalism that contains a lot of misrepresentations - of both scientific evidence and of scientists who do not agree with Harden. A lot of scientists are real mad at this article.

Now, given the people I tend to follow, I'm really only seeing one side of it (the side that hates it). I'm also not enough of an expert in the field to form my own opinion about the scientific evidence. I mean, I've read extended Twitter threads that sound convincing to me, but I'm not a geneticist and I'm definitely not a behavioral geneticist.

I would not be surprised if there are rebuttal pieces coming out shortly, though.

I do doubt that if they do, they'll get half as much traction; these days. The "woke leftists won't let us pursue the truth" angle seems to be extremely popular in news and magazine articles these days. The New Yorker's cozy relationship with Steven Pinker, who I know for a fact deliberately misrepresents his opponents and criticism of him, makes me really skeptical that they did any due diligence here.
posted by Kutsuwamushi at 11:26 AM on September 7, 2021 [24 favorites]


Again, not taking sides, but that's the argument she's making. She's citing genetic research and psychological studies, i.e. what are commonly considered as "science." So to say things like "it's not so obvious" or it's "skeptical scientifically" - without being more specific - is just assertion rather than actually arguing against her views.

Is she? I know it's a long article, but if you read it with attention to detail and unwillingness to uncritically adopt the article's framing, you can see that she is, at best, advocating for a particular interpretation of a relatively new school of thought which is far from comfortably established as accurate. Which is not, in itself, a problem, it's what people making arguments in science do--but it is when it's presented as "leftists want to close their eyes to The Science!" You are not distinguishing (nor does the article wish you to) between citations to scientific data and citations to widely-accepted scientific conclusions.
posted by praemunire at 11:40 AM on September 7, 2021 [5 favorites]


No idea if this appeared in Kutsuwamuski's Twitter stream, but here's a thoughtful, detailed thread laying out objections to the article and book.
posted by PhineasGage at 11:41 AM on September 7, 2021 [11 favorites]


I wonder to what degree the relative standards of evidence in psychology (where Harden is a prof), medicine and biology play into this. Because my (very much casual) understanding of the current field of genetics is that it is not easy to make the kind of statements that Harden is making with any kind of certainty. We all know that psychology papers are very often not replicable. And then when writing a book, scientists are of course far more free to present weak findings as settled or certain, since there is no peer review process and fact checking is often minimal or non-existent. Is the difference in standards a factor here?
posted by ssg at 12:28 PM on September 7, 2021 [3 favorites]


I don't there's any question that some components of intelligence is heritable. But as pointed out in She Has Her Mother's Smile, something as simple as height can be affected by dozens of different genes. Something as complex as intelligence probably has hundreds of components. There's no single smartness gene.

That combined with the fact that good nutrition during developing years has markedly large effects, searching for genetic intelligence markers seems rather pointless other than to indentify minor effects.
posted by CheeseDigestsAll at 12:34 PM on September 7, 2021


what it's trying to do is keep the door closed on eugenics, phrenology, and the overriding desire to find biological reasons to recreate and reinforce social hierarchies, because these things are bad. And it doesn't seem like anyone can think of any use for this research other than the reinforcement of social hierarchies

That's where I land. What's the point of it, other than to cruelly dismiss people as beyond help? Her statement:
For her, accepting this is the necessary prelude to any conversation about what to do about it. “If you want to help people, you have to know what’s most effective, so you need the science,” she told me just doesn't hold any water. Pilot studies of different interventions are cheap and easy. If you want to help people, it's perfectly fine to throw 100 solutions at whatever problem you're concerned with and then do whatever generally works best.

Likewise: Harden can identify studies on a vast array of topics—Will coaching underresourced parents to speak more to their children reduce educational gaps? Does having dinner earlier improve familial relationships?—whose conclusions she considers dubious because the researchers controlled for everything except the fact that parents pass along to their children both a home environment and a genome.

If those results are spurious, it will quickly become obvious because the interventions based on them won't work. Then, we can do something else.
posted by GCU Sweet and Full of Grace at 12:48 PM on September 7, 2021 [1 favorite]


I teach biology to first generation college students.

1) Beyond a few well-understood alleles that clearly affect a single biochemical pathway, the vast majority of our conclusions about the connections between genotype and phenotype, especially behavioral conclusions, are drawn at the population scale and often fall apart when we look at individuals. Which is fine, because that's how population research works. But it means that conclusions cannot and should not be drawn about individuals from population data.

2) The idea that educational attainment of all things is inherited and one of those things about you that wouldn't be able to change if you suddenly relived your life over is so absurdly farfetched given the oodles of published research that show the benefits of certain interventions for first generation students. If educational attainment is heritable (and, the article implies, immutable), then why do small class size, a strong mentor relationship, peer support, and training in study skills all help our first gen students so much? Not to mention making sure that they have a safe place to sleep and enough food to eat?
posted by hydropsyche at 1:11 PM on September 7, 2021 [28 favorites]


I'm commenting as I read the article, making notes on my thoughts as I move through it. I've seen a certain amount of irritation about it on Twitter from other people in my field, but haven't gotten to it myself before today. I am also going to signpost, given the loving commentary on dog breed behavior genetics in the original piece, just how bad an analogy dog breeds are for human race (previously).

I do have to say that I am extraordinarily skeptical of many of behavior genetics' takeaways, less because I am a priori insulted by their evidence but more because I am wildly unimpressed by... well, a lot of their evidence. For example, one common method of study in the field is the Genome-Wide Association Survey, which takes vast numbers of people, measures them along some trait of interest, and samples vast numbers of single nucleotide polymorphisms across their genomes to find variants that might or might not indicate predisposition to that trait. These studies do generate significant relationships between certain alleles and certain traits, but these alleles almost always turn out to be either vanishingly rare or to have vanishingly small effects, and often both. I'm not convinced that these studies are a meaningful way to actually understand how and why genetic variation yields behavioral variation. In the fifteen years or so we've been conducting them, I will note that they have almost never resulted in clear insights that allow us to understand more about the way that genetic variation creates varied outcomes, and while the statistic p-values are often small... so are the R2 values that indicate how much of the variation they predict.

Another criticism I often have for behavior genetics comes from my background working on decision-making and plasticity. It is very hard to separate out the effects of the environment an individual exists in the context of and the effects of the specifics of that individual's genotype, especially because the expression of a genotype is often much more likely to vastly change phenotype than the genotype itself. Our brains are constantly changing the expression of neurotransmitters, receptors, and sensitivity to signaling in a thousand different places in response to the world we live on, and crucially we also change the environments around us just as our environments change us.

When we look at heritability for genetic traits, then--including many of the studies that behavior genetics rest on!--we measure a specific thing that people often do a bad job of explaining, which is the correlation between phenotype of related individuals--often the averaged phenotypes of, say, parents correlated with that of offspring. High heritability means most of the variation in the population that we see comes from genetic variability, and low heritability means most of that variation is coming from environmental variability. But: the amount of environmental variation you are incorporating into your tested sample matters. So does the amount of genetic variation! It's very easy to accidentally vary these things by failing to incorporate some axis of either type of variability into your design, and it's easy to accidentally measure a trait as highly heritable if you forget to incorporate enough (or the right kind) of environmental variability--or to measure that trait as negligibly heritable if you don't capture the true extent of genetic variability in your initial sample. These data sets are therefore incredibly reliant on the care initially taken to secure a sample that is as close as possible to your real-world dataset, and I think it is very reasonable to be very careful about overextending the results from those numbers to populations that differ in any way from the original sample population.

I note that my opinions on these studies are not remotely uncommon among scientists--indeed, the first time I ran across those criticisms of GWAS designs was during my undergraduate work in a population genetics lab, a solid ten years ago! So I am very much raising my eyebrows at the writers of the piece, because there really has been a lot of work in the slant of this piece to carefully avoid looking at the scientific criticisms of work like Harden's. That's especially true of the scientific criticisms that say less "we should never study this work" and more "we should be very careful about this work, because the policy implications are frightening." When there are many scientists who are critical about something, and it takes a long time to explain all the reasons why you are skeptical... well, maybe it behooves a popular writer searching for truth rather than a saleworthy narrative to ask those scientists why they hold these criticisms.

As I read--okay, we are going into some of these criticisms... but we are carefully going "ah, but Harden's work has taken them into account, so they don't hold water now!" Has it? We've discussed what a GWAS is in the article, but not why so many neuroscientists and geneticists roll our eyes at them and the shiny new Big Data findings they tend to pull out. When your results are so very dependent on your input data, and when the field is often.... lazy about that input data and measures of what the trait of interest actually is, why is your data and your approach automatically more trustworthy? More, so much of GWAS-related work relies on complex metrics of dimensional analysis that really bypass any deeper understanding of the underlying architecture of the data.

And when we talk about informing policy with data, how exactly do we do that when we barely understand what we're doing with respect to how even rare genetic variants of small effect produce various outcomes? It seems to me that the less we understand about the systematic way that genetics, gene expression, and environment interact to create a phenotypic outcome, the less impact that work should have on policy--that until we know how something is happening, we should be very careful about ascribing the label of science! to informing what we should do. This is a very different mindset to the one I see among GWAS fans and behavior genetics, and I find it very troubling indeed.

Behavior geneticists frequently quote an old disciplinary chestnut about how first-time parents are naïve behaviorists and that a second child turns them into convinced geneticists. In one chapter of her book, Harden mentions that her son struggles with a speech impairment. “Looking at how my children differ in their ability to articulate words, I can easily see the capricious hand of nature,” she writes. “When it comes to inheriting whatever combination of genetic variants allows one to pronounce a word like ‘squirrel’ by the age of three, my daughter was lucky. My son was not.”

This disciplinary chestnut, I must say, raises massive warning bells for me as someone who is interested in sex, gender, and biological determinism: almost exactly the same disciplinary chestnut is frequently repeated about gender in small children from parents who are committed to not passing on gender roles, and of course culturally we have extensive evidence that children are treated very differently from (or prior to!) birth based on their assigned genders. There is a definite tendency among behavior geneticists to act as if children in the same household share an identical early environmental set of experiences, especially if they are twins, but this isn't true--even children in the same household who are the same age may have very different social experiences based on the interactions they have with other members of that household and other non-household members of their social networks. When it comes to issues that shake among human cultural grounds, like race, ethnicity, and culture, you have to be incredibly careful about this before you go racing to declare that genetics is the root cause, because we are often bad at noticing how these different social systems work in practice.
posted by sciatrix at 1:36 PM on September 7, 2021 [30 favorites]


Catching up on the discussion here: oh yes abso-fucking-lutely where the fuck is epigenetics and gene regulation more generally in this lady's work?! did she not pay any attention to, say, the Meaney licking and grooming work in her meditations on why divorce maybe has a heritable component? where is gene regulation discussed and why is it totally missing from this article?!

I keep wanting to bang on my old chestnut in this discussion, which is: an organism without genetics does not exist, but an organism without environment is just a little test tube of nucleic acids and possibility. These things are not disentanglable! They interact with each other and in the case of gene regulation they are constantly changing in the brain all the time as a consequence of experience!

Why is this totally missing from this lady's work?!
posted by sciatrix at 1:48 PM on September 7, 2021 [11 favorites]


I do doubt that if they do, they'll get half as much traction; these days. The "woke leftists won't let us pursue the truth" angle seems to be extremely popular in news and magazine articles these days. The New Yorker's cozy relationship with Steven Pinker, who I know for a fact deliberately misrepresents his opponents and criticism of him, makes me really skeptical that they did any due diligence here.

I'm sure it won't come as a huge surprise that the "anti-woke" brigades seem to be positively giddy about this article entirely on the basis of it poking the left in the eye. The transphobes amongst that set are especially energized by it, which terrifies me, because twisting arguments about genetics is a major part of their strategy to codify trans people as abominations with mental health issues.
posted by Glegrinof the Pig-Man at 2:03 PM on September 7, 2021 [6 favorites]


It's worth remembering that labor historian Eric Loomis pointing out that scientific research is being done in a culture built on structural racism caused consternation among the usual suspects. There is a push to treat "science" as something "pure" and outside of society - often times by people who are looking for justification for their retrograde views.
posted by NoxAeternum at 2:23 PM on September 7, 2021 [7 favorites]


So I am very much raising my eyebrows at the writers of the piece, because there really has been a lot of work in the slant of this piece to carefully avoid looking at the scientific criticisms of work like Harden's.

Thanks, sciatrix, this is what I was trying to say, but you actually had more detailed scientific knowledge to make the point.
posted by praemunire at 2:34 PM on September 7, 2021 [2 favorites]


That's a crystalizing example. What Loomis is citing (inaccurate facial recognition) isn't "science" as that word is used in the book, article, and this thread, and his statement ("Science, statistics, and technology are all inherently racist because they are developed by racists who live in a racist society, whether they identify as racists or not") is explicitly what the article/book are disagreeing with. There are the two world-views, side-by-side.
posted by PhineasGage at 2:36 PM on September 7, 2021 [1 favorite]


Yes, I get that's what Harden is arguing. I just happen to, like Loomis, disagree with the position. And I disagree with it because a) science isn't some field of objective truth, but b) exists within a culture which has some significant issues with systemic racism which c) in turn impacts how science is both conducted and viewed, and yet d) people have a need for science to be free of racism:
I swear to God the need white people have for the things they respect or enjoy not to be racist is fucking mind boggling. If the effort put in to debating that androids don’t dream of electric klan rallies went into questioning why so many of our tools end up being used against our country’s ethnic minorities and how we can stop it the first debate wouldn’t need to take place.

Of course science is racist. The people who get to decide what to investigate and how and what conclusions to draw are overwhelmingly white people steeped in the racism of society. Fuck, it’s no shock that to this day poverty and poor choices is the answer of choice for diabetes in the Black American community, but that when the same trends were observed in post-war Holland it was traced back to the epigenetic influence of famine on the offspring of war survivors. Gee, I wonder if Black Americans have any history of famine or deprivation in their genetic past.

Of course statistics is racist. Look at any fucking debate on police brutality and the numbers crunched to prove that “Well, yes, police do beat up Black people more but they don’t shoot them more often as a percentage of their overall interaction.” Racist statistical analysis is accepted as legitimate all the time because, hey!, the math checks out!

Of course technology is racist. Ask anyone who ever had to photograph Black people how much harder it was to get the exposures and light balance right compared to shooting white subjects. Because technology is designed in this country with the white default in mind – consciously or not – it will disadvantage other groups.

Part of this is because people act like racism is a moral question when it’s not. It’s a systems question. Saying an object can’t be racist because objects don’t have moral agency is facile and tiresome. The object is racist because the system it inhabits and utilizes it is racist. There may be a way to get to systems that aren’t racist and therefore remove the racist component of these objects, but we’re not even close to that point and wasting time arguing that science is some sort of method that exists outside the boundaries of the society that uses it to interpret the world is insulting to everyone who was ever subject to degredation at the hands of people who were just doing what the science told them to do.
Harder complains that people keep bringing up The Bell Curve, but then calls repudiation of Charles Murray counterproductive - and doesn't see how the latter is what causes the former. As several posters have pointed out, the skepticism of the left is born out of how science can be twisted towards persecution, and that they want to see some acknowledgement of this. And yet we get the "argument" that science is somehow divorced from society (with no explanation of how), and that should be reassurance enough. And then scientists are surprised when it turns out that it isn't enough.
posted by NoxAeternum at 3:28 PM on September 7, 2021 [12 favorites]


I think Harden could be right, or not, either human flourishing capacity is 72% genetic (from the article), or it isn't. I wonder if the new book at least explains how this number was even obtained?

I think both Harden's politics and the criticisms can be right, it's not incompatible. Maybe in a context of 1000 years time humanity would be more ready to take on the genetic contributors to inequality--assuming that 72% factoid is even actually the truth.

Meanwhile the article is pretty light on the validity of this sort of science. Presumably these behavioral geneticists are all using machine learning by now? But is that good science, or not, is unclear.
posted by polymodus at 5:11 PM on September 7, 2021


Can I point out how weird it is to call a 10 month old “sly” for trying to grab things left at their level?

I haven’t read the actual papers, but my gut instinct as a data scientist is that a correlation pulled out of a big data set is a dangerous thing to make public policy on.

Polymodus - presumably 72% is an r-squared value of a fancy regression, and I would not call most machine learning algorithms “good science”.
posted by Valancy Rachel at 6:29 PM on September 7, 2021 [7 favorites]


I'm a brown STEM academic person that has a policy-relevant position, and I haven't been able to comment sensibly on this post all day because I just come off too screamy angry, so just a big yes to everything all the science people have said with a special thanks/shoutout to what I learned from sciatrix's informed detailed points. Also yes to everything Valancy Rachel just said - including that the analysis is very likely to be regression model based.
posted by BlueBlueElectricBlue at 7:29 PM on September 7, 2021 [15 favorites]


It's "seventy-two per cent of inequality at the primary-school level in the U.S. is within demographic groups rather than between them," which sounds like some sort of anova.

I can't think in anova very well -- that's not something my discipline uses very much at all -- but this is at least close to something like "Demographic differences explained 28% of variation in inequality, leaving 72% unexplained." The article doesn't explicitly say this, but the writing at this point is clearly trying to imply that this means lots of the remaining variation is genetic.

And, sure, some chunk of it is. But, not being able to find the World Bank data with trivial effort, some of that unexplained variation might just be due to interactive effects they didn't model between demographic variables. Or it might get sucked up by demographic variables that weren't included, or by different operationalizations of the same theoretical variables. And of course with any data on individual children we'd expect there to be a LOT of stuff that's pretty close to true random noise in whatever their DV is, whenever that was recorded. Which kids were coming down with a cold, or broke their pencil, or were given dry poffins for lunch?
posted by GCU Sweet and Full of Grace at 8:27 PM on September 7, 2021 [2 favorites]


I found the article very confusing. As others have noted it avoids certain issues (epigenetics I know knowthing about, but skirting mentions of race concepts is weird). But mostly it seems to me it contradicts itself, as several people have pointed above (your twin life would be different but the same!). The most surprising paragraph for me was:
Some of Harden’s most recent research has looked at curricular tracking for mathematics, an intuitive instance of how gene-environment interactions can create feedback loops. Poor schools, Harden has found, tend to let down all their students: those with innate math ability are rarely encouraged to pursue advanced classes, and those who struggle are allowed to drop the subject entirely—a situation that often forecloses the possibility of college. The most well-off schools are able to initiate virtuous cycles in the most gifted math students, and break vicious cycles in the less gifted, raising the ceiling and the floor for achievement.
So, the only thing that matters is the environment, since rich schools help both kinds of students and poor ones fail both...
posted by anzen-dai-ichi at 1:59 AM on September 8, 2021 [5 favorites]


Harden responds via Twitter to one element of criticism of the article. She points out the concerns that geneticists were "reinventing race as a biological reality rather than a social construction" and quotes Dorothy Roberts' "call for scientists to use terms like 'ancestry' and 'population' rather than 'race.'" She criticizes Murray's elision of the difference between ancestry and race: "Comments like these collapse race and genetic ancestry into a single idea, and in doing so, perpetuate the idea that racial inequalities in outcomes must be due to innate biological differences. It is, therefore, critical to understand why this collapse is a mistake."
posted by mittens at 7:35 AM on September 8, 2021 [1 favorite]


But Murray collapsing race and genetics isn't a mistake, and that's the heart of the problem here. His comment that Harden quotes isn't born out of misunderstanding - it's him saying "yeah, we all know why geneticists are using these terms now" with a wink, basically implying that geneticists are on his side. And you don't answer that with a longform answer that treats bigotry as good natured misunderstanding - you answer it by pointing out the bigotry and making it clear that you have no part of it.
posted by NoxAeternum at 8:03 AM on September 8, 2021 [4 favorites]


This Guardian article offers some clarification. It doesn't seem like she's pushing a race angle. She's also clear on the difference between doing well in school and intelligence. I'd be willing to read the book.
posted by CheeseDigestsAll at 12:54 AM on September 12, 2021


Even there, the only benefit that she can find for her research is that we could save a tiny bit of money by not even testing some interventions.
posted by GCU Sweet and Full of Grace at 5:11 AM on September 12, 2021


« Older “A cruise ship is not very good for people who...   |   When Billie Eilish met Stormzy Newer »


This thread has been archived and is closed to new comments