Hacker News Reader

Comment by armchairhacker | Hacker News Reader

Genetics has a huge influence on what someone is predisposed to and what they can achieve. Certain ethnic groups and families, on average, are more likely to have certain traits and less likely to have others.

But genetics doesn't matter, in that just because you have "bad" genetics doesn't mean you'll suck, whether it comes to weightlifting, intelligence, temperment, etc. Maybe you won't be in the top 0.0001%, but except for some very specific circumstances, you can defy the average. If you want to know if someone's bad at something, actually testing them is much more accurate than looking at their genetics.

If you look at the top 10 distance runners, they're all from Ethiopia and Kenya. But as you look further down the list into the top 100, you start to see Americans and others. Furthermore, the Americans' times are really close to the Ethiopians and Kenyans.

So, even if we know for a solid fact that some ethnic group is on average dumber than another, or more violent, it doesn't mean anything. Because it doesn't say anything about the individual people, who vary much more than the difference in averages.

View on HN

Sure, that's a good point generally but doesn't seem at all the problem the article is addressing. Skimming it, it's about progressives denying group differences at all and calling those who research them 'equivalent to Holocaust Deniers'.

From following genetics research and researchers, I've seen plenty of that, and I don't see the relevance of your point here.

The research isn't specifically about group differences, or differences between populations anyway. That issue inevitably gets all the attention, but it's just one dimension. The vast majority of genetic differences between any two individuals have nothing to do with whatever arbitrary groups society categorises them with, and most genetic differences that do correlate with anything like race are basically cosmetic.

Nevertheless genetic differences can make a huge difference to individuals. Many factors we uncontroversially categorise as disabilities have a genetic basis.

In UK schools children categorised as on the Autism spectrum, or dyslexic receive additional support and consideration with regard to exam rules. I don't want to sideline into an argument about to what extent these specifically have a genetic basis, but broadly if we are going to agree that such factors are taken into account, it's flat out incoherent to then ignore or deny research into the origins, characteristics and influence of these factors.

The backlash by progressives is usually specifically centered around the group differences part. And I don't buy that's the one that gets all the attention - all the negative one maybe, but most accepted and talked about genetic research isn't about that at all.

Different ethnic groups have meaningful difference in various allergies and disease risks, but that doesn’t get attention from the general public.

That stuff is basically all interesting but largely meaningless because it’s so binary. It doesn’t matter if your part of a group that’s 10x as likely to be lactose intolerant, you are either lactose intolerant or your not. Sure the dairy industry might care, but the only impact on individuals is what they can bring to very large family gatherings.

Even where it is actually meaningful to an ethnic group, people rarely from a single ethnic group. A large chunk of African Americans are whiter genetically than their black.

For lactose intolerance maybe it doesn't matter. There are a lot of diesesases for which screening is not worth it if your risk is too low due to a high false negative compared to your risk (which is part of why we screen young people for a lot less).

If you might remember a somewhat recent controversy when they tried to remove race in the eGFR (kidney effectiveness, sorta) calculations. This was not done for performance reasons, or to find a better measure. The impetus was the idea that it was racist, never mind the performance or benefit to black patients.

"most genetic differences that do correlate with anything like race are basically cosmetic."

But those that aren't cosmetic can matter a lot. For a noncontroversial example, sickle cell anemia. It absolutely makes sense to screen people of West African descent for sickle cell anemia.

To fine tune what you said, the article (to leave the actual scientific work in question aside for now) focuses on progressives denying genetic differences between groups, but modern behavioral genetics focusing on differences within groups, between individuals.

(Between groups being entirely too muddled by confounding factors)

And the stridently progressive crowd (read: Twitter) rage-cruising right past the distinction, and using arguments against the former to dismiss the latter as anathema.

And then, as these things typically go, coming to their senses and being somewhat embarrassed by where they end up, but unwilling to concede their obviously righteous point and so making increasingly intellectually ridiculous justifications.

Well, a major problem is that the groups being studied are often either very poorly defined - things like 'black people' or 'Ashkenazi jews', which are entirely social categories, with little to no effort in most IQ studies to actually establish genetic group markers.

Other times, studies look at groups with much more proeminent socio-economic differences than genetic ones (countries, neighborhoods). Here the genetic factor is impossible to isolate from the other factors, making genetic claims dubious.

> progressives denying group differences at all

In my experience, it depends on the group. It seems conservatives are more likely to defend the idea of IQ differences among racial groups, but also more likely to reject the studies showing that liberals tend to have higher IQ's than conservatives[1].

Fundamentally, though, I'm not sure what the use of highlighting group differences is. I tend to skeptical of such studies (in my own experience datasets can be very misleading if sliced the wrong way), but if the studies about political alignment are true, then what? Should I start separating people based on what group they belong to? As the above poster said, when you're measuring an individuals aptitude then the group aptitude is basically meaningless.

[1] https://www.cnn.com/2010/HEALTH/02/26/liberals.atheists.sex....

Comparing a single study to a whole subfield is just odd.

Also again, no. You generally don't make individual judgements based on group differences. You can make population-level decisions based on them or allocate funds more effectively or reach a deeper understanding on group differences with such research so socially outlawing it is counter-productive even if you personally won't use the results in your day to day life.

>In my experience, it depends on the group. It seems conservatives are more likely to defend the idea of IQ differences among racial groups, but also more likely to reject the studies showing that liberals tend to have higher IQ's than conservatives[1].

This may be true, and a perfectly fair point ... but it does not actually prove the argument wrong, it simply points out that groups are susceptible to motivated reasoning. ie, there are some measurable IQ differences among groups which may be related to genetics, and this point is not invalidated because different people supporting it or criticizing it are hypocrites.

The point is that progressives should stop being so afraid of accepting genetic differences, because they don't go against the type of policies most progressives are after. And because genetic differences are probably real, when progressives argue against racism etc. because "genetic differences don't exist" they actually hurt their position.

Policies that discriminate based on genetics are bad, not because genetic differences aren't real, but because they're a) inaccurate, especially compared to measuring the differences themselves, and b) usually immoral, because having good genetics is luck and doesn't make you deserve a better life. Chomsky (see above comment) probably does a better job explaining this.

The catch here would be thinking in distributions, and from that separating individuals from said distributions.

A current topic would be representation within a given domain. Particularly since in most representation cases (high paying jobs, celebrities, top universities, etc), we're already looking at the tails of the distribution. There, small shifts can heavily affect the quantity of users in the right tail, especially for domains with stronger requirements (see your case with the runners).

Of course I won't try to discount potential cultural impact, but anyone trying to sell you a story of how it's all nature or all nurture is typically spouting bullshit.

But it's a bad idea to compare the world's fastest runners with people getting high paying jobs - that's the main point of the GP's comment. Even for something as clearly and uncontroversially genetically influenced as running ability, the group differences only really matter for the top 100 people out of 7 billion. If you look at top 100k runners, the group differences are almost entirely gone. Even more, the top 100k runners are probably within a second or two of each other - entirely irrelevant if you were to hire based on a bar of performance, and not some elitist concept of hiring the very best possible.

Socially, who makes it to the top hundred matters a lot. We can know only so many people by name: Nobel awardees, composers like Bach and Dvořák.

Our own brain capacity for knowledge makes us concentrate on the top 100s. While the 100k top scientists do matter scientifically, they do not matter culturally.

> Even for something as clearly and uncontroversially genetically influenced as running ability, the group differences only really matter for the top 100 people out of 7 billion.

Sure, but there is no social contention arising from racial differences in the world's fastest runners. What gets us in trouble are the the racial differences in high-prestige jobs. To really understand the impact of differences in distributions, you need to look at some numbers. The excerpt below comes from Douglas Murray's new book. I am not going to attempt to convince you that racial differences in IQ exist. But he makes a strong case that if such racial differences did exist, the impact on the distribution of high-prestige jobs would be absolutely enormous:

"...Why are there so few minorities in these high-prestige jobs? It's a numbers game in which the odds are against a Latin (he uses geographic indicators in place of the colloquial 'black' and 'white', and 'latin' in place of 'hispanic') achieving one of those positions are high and the odds against an African are prohibitive, even if we assume that there is no racism whatsoever among the employers for high-prestige jobs.

To illustrate, I'll use the cohort of young Americans ages 25-29, the age at which the potential candidates for such jobs are coming out of law schools, medical schools, business schools, and graduate STEM departments. In 2019, there were 23.2 million Americans in that age group. About 228,000 people in that age group can be expected to have IQs of 135 or over.

The racial distribution of Americans ages 25-29 in 2019 was more multiracial than among the older population. Only 54% were European while 20 percent were Latin, 15 percent were African, and 6 percent were Asian. But that reduced dominance of Europeans in total population doesn't make a lot of difference in the 135+ pool. Employers seeking these exceptionally intelligent young adults were choosing from a pool that contained about 2,800 Africans [~1%] and 9,500 Latins [~3%] compared to 50,700 Asians [*~22%*] and 160,000 Europeans [~70%]." [~3% being 'other']

So, while the marginal IQ differences between individuals do not matter not one whit, when considered at the population level, these differences can hardly be ignored.

> But that reduced dominance of Europeans in total population doesn't make a lot of difference in the 135+ pool. Employers seeking these exceptionally intelligent young adults were choosing from a pool [...]

This is one mistake, and the danger of using IQ as anything more than a trend indicator. Employers don't need 'people with an IQ over ${ARBITRARY_CUTOFF}'. They need 'people who can do X and Y'. And IQs over 100 quickly stop having significant predictive power for mental abilities other than IQ tests. Even exams which focus on pass/fail of an aptitude rather than excellence show little 'racial' variance in general.

The numbers in that paragraph also seem entirely fantastical to me. It basically claims not only that there are IQ differences between the "races", but that they are huge.

Finally, and most importantly, the whole set of observations is meaningless in the way it is used. Even granted that there are such huge differences in IQ between various "racial" groups, the social justice response won't change - we need to focus on eliminating the social and environmental causes at the root of these differences. Murray could claim that the differences are genetic, but I am very curious what is the scientific methodology by which he defines these groups - how much genetic testing did he do determine whether one particular person is 'European', 'African', 'Latin', or 'Asian'? What genes in particular was he looking for that make someone European? (For that matter, are Italians and Spanish people "European" or "Latin"?? Is Eva Longoria, whose family came from Spain to America with a royal land grant 400 years ago "European" or "Latin"?).

> This is one mistake, and the danger of using IQ as anything more than a trend indicator. Employers don't need 'people with an IQ over ${ARBITRARY_CUTOFF}'. They need 'people who can do X and Y'. Even exams which focus on pass/fail of an aptitude rather than excellence show little 'racial' variance in general.

The scientific literature is very clear on this point--IQ is predictive of job performance. If you need people who can do X & Y, IQ is the best predictor[0] of whether a given individual will be able to do X & Y.

> And IQs over 100 quickly stop having significant predictive power for mental abilities other than IQ tests.

I would love to see a source for this claim.

> The numbers in that paragraph also seem entirely fantastical to me. It basically claims not only that there are IQ differences between the "races", but that they are huge.

The numbers in the paragraph are partly due to differences in IQ, and partly due to the nature of distributions. IIRC, Murray claims a mean IQ difference of ~15 points between 'Africans' and 'Asians'. Here's an image[0] of two distributions with an equivalent mean difference. Note how, the farther you get from the center, the greater the proportion of red:black. At 1 standard deviation, the ratio is 3:2, at 2sds, the ratio is 4:1!

> Even granted that there are such huge differences in IQ between various "racial" groups, the social justice response won't change - we need to focus on eliminating the social and environmental causes at the root of these differences.

The point of this conversation is that, if these differences are due to genetics, then no amount of 'eliminating causes' will address the root cause of these differences.

> Murray could claim that the differences are genetic, but I am very curious what is the scientific methodology by which he defines these groups - how much genetic testing did he do determine whether one particular person is 'European', 'African', 'Latin', or 'Asian'? What genes in particular was he looking for that make someone European? (For that matter, are Italians and Spanish people "European" or "Latin"?? Is Eva Longoria, whose family came from Spain to America with a royal land grant 400 years ago "European" or "Latin"?).

All the data he uses relies on self-reporting of race. Self-reporting is always how race is assigned. Self-reporting is how we know that certain minorities are 'underrepresented' in certain jobs--folks self-report their race when hired. It's the same way we know that there are differences in incomes and in educational achievement between races--self-reporting. You can doubt the veracity of self-reporting of race, but that doubt would have to apply to all data with a racial dimension. And, anyway, if race was some sort of fuzzy, meaningless construct (as you appear to be implying), then we would expect there to be little correlation between race and anything, let alone IQ. And yet the data show these differences (according to Murray).

Longoria would be 'Latin' (Hispanic). The whole race vs. ethnicity thing is weird, but he explains it well in the book.

[0]https://www.jstor.org/stable/20182140 [1]https://tamino.files.wordpress.com/2015/07/shiftmean.jpeg

> Self-reporting is always how race is assigned. Self-reporting is how we know that certain minorities are 'underrepresented' in certain jobs--folks self-report their race when hired.

Sure, because race is entriely social construct. It's impossible to find a genetic link between the vague notion of 'black' and IQ, because people who are in the social category 'black' or 'latin' or 'european' have a wide variety of genotypes.

As such, the only possible conclusion from IQ studies is that IQ is a measurement of social and perhaps early childhood environments. So, improving conditions for marginalized groups will improve IQ and implicitly improve other success metrics (to the extent those are correlated with IQ).

Note - IQs below 100 are a different matter entirely, and they mostly represent known developmental disorders such as Kleinefelter, or severe malnutrition in early life (a preventable condition, but irreversible).

> If you look at the top 10 distance runners, they're all from Ethiopia and Kenya. But as you look further down the list into the top 100, you start to see Americans and others. Furthermore, the Americans' times are really close to the Ethiopians and Kenyans

The Americans still almost certainly have very favourable genetics for distance running, they're just less favourable than the genetics of the Kenyan runners.

If you compared the American runners in the top 100 against the average American I would be very surprised if they had similar physical attributes (Lung capacity, muscle fibers, etc).

America in this case is a very bad country to measure with considering the amount of people that may actually be of very similar decent as your first examples..

East Africa and West Africa aren't the same places

Kenyans, Ethiopians and (white) Americans are not genetically homogenous. Even two siblings can be quite different genetically.

The great American distance runners have genetic talent - they didn’t just train harder than their middle school mates who were out of breath while the champions ran faster and longer than anyone else weeks after beginning training.

> have genetic talent - they didn’t just train harder than their middle school mates who were out of breath

Who knows if it was only due to genes? And more importantly, there isn't a gene that's about running faster. (sure, some genes are helpful and might get you in front - not discounting the work the athlete goes through)

Getting to the podium is a mix of several characteristics: Some people might be more prone to muscle recovery/injury resistance. Another one has a better muscle fiber mix. Another one has different muscle attachment positions. And another one feels less pain.

Though yes your genetics probably will push you towards a sport. Take the 100m finalists and put them to train marathon (and vice versa) and they'll probably go to the back of the pack even with good training.

Sure, there isn't a gene that makes you run faster, but no matter how hard you train, you'll have a hard time beating Michal Phelps unless you're also naturally built like Michael Phelps: https://www.biography.com/news/michael-phelp-perfect-body-sw...

The point isn't that there is a "best" set of genes. The point is that there can be genetic advantages or disadvantages. Physical reality has an impact on experiential reality.

Where it gets dangerous is when people start using this as an argument for eugenics or an "ought" rather than "is" (i.e. constructing hierarchies out of physical attributes).

Yeah, considering the actual day to day life of all people, Michael Phelps's genetic advantages are pointless. They confer advantage only in a specifically designed field that society has created and that is really a tremendous waste of time all things considered; it's just a performance.

But the aim of all this genetic stuff is not saying "well, Kenyans can make better marathon runners if we stick them into a sports contest." It's usually a lot more pernicious than that, and involves intelligence. And ordering society in some manner around genetic expressions in a way a lot more dangerous than recognizing some people are better at public performance than others.

The US military routinely tests recruits for IQ and assigns roles appropriate to that level. They even outright reject some people for being too stupid.

Nobody seems to be up in arms about it, why?

We also already do all sorts of soft eugenics in our own lives, from assortative mating to academic scores and schools one attends, social circles that one enters because of their achievements and so on. At some point it will become a lot more quantitive, to reduce the currently large factor of luck.

Societies that refuse to take advantage of optimizations this could bring will eventually lose out, for better or worse, this is entirely inevitable.

The progress of technology is not always progressive. There is a reason most of sci-fi reads downright dystopian.

Gene for running faster: https://www.frontiersin.org/articles/10.3389/fphys.2017.0108...

Amusingly, it is literally what you are writing about. Certain variants of that gene confer exactly a) better recovery b) sports injury resistance c) fast twitch fiber synthesis.

Simply because we have not pinpointed something with laser precision, propagated the knowledge to every human with internet access and papered it up with indisputable evidence strong enough to pacify the eternally-online exception finders and fact checkers - does not mean such things do not exist.

You are your genes. just a tiny 2% difference - and you would end up a dolphin.

> just a tiny 2% difference - and you would end up a dolphin

This is weirdly specific and I'm pretty sure my brain is going to remember this and bring it up on random occasions for the next ten years.

Sure, that's one gene that helps. Saying that you only need that to get to a podium would be a stretch.

And there might be other genes. Someone missing this beneficial change might have other helpful mutations. As even the article points

> In addition, whilst Kenyan and Ethiopian endurance runners are highly successful (Wilber and Pitsiladis, 2012), the frequency of the XX genotype within this group is very low at 8% (Ethiopian) and 1% (Kenyan) (Yang et al., 2007) (XX is the endurance variation)

Or are you saying that Track programs should genetically test for this gene and drop whoever doesn't have it?

> Or are you saying that Track programs should genetically test for this gene and drop whoever doesn't have it?

They don't have to. There is almost certainly a lot of self-selection going on. Anecdotally, I got one of those genetic kit tests done 10 years ago, and it included results on a "fast-twitch vs slow twitch muscle gene". I, who was a sprinter in high school who struggles immensely with long distance running, unsurprisingly have the gene for fast-twitch muscles. My wife, who has never participated in any kind of sport involving sprinting at any level but competes in long-distance races, unsurprisingly has the slow-twitch gene. Neither of us had any of these genetic tests available when we got involved in these activities. We just learned over time what we were naturally good at and what we struggled with. Does it mean that I can never run long-distance at the level my wife does, or that she could never have competed in sprinting? Maybe, but not necessarily. However if possible, it would take far, far more effort than sticking with the things we are good at.

Between people gravitating to what comes naturally, and other filters (like making or being cut from a track or XC team), a lot of the genetic filtering already occurs without actual testing.

> Who knows if it was only due to genes? And more importantly, there isn't a gene that's about running faster

Do you interpret “genetic talent” as “there is a single gene that explains everything about this skill”? I don’t.

Would you say it’s unreasonable to say cheetahs are faster than turtles due to genetics unless you can pinpoint the one gene that causes the running speed difference?

> Some people might be more prone to muscle recovery/injury resistance. Another one has a better muscle fiber mix. Another one has different muscle attachment positions. And another one feels less pain.

All genetic, so you’re agreeing with me then?

True, but the talent isn't as obvious. Similarly, you could have two very slow runners give birth to a fast runner, and most of their speed would ultimately come down to getting the right recessive genes.

Almost every trait, even willpower itself, is influenced by genetics. But for most traits it's very difficult if not impossible to reliably determine what genes influence what. Most things aren't determined by one gene, they're influenced by many genes, as well as randomness and your environment.

> Because it doesn't say anything about the individual people, who vary much more than the difference in averages.

It's not clear it says anything about the difference in averages. Say someone is born in Syria with great genetic potential to be a distance runner, you might expect they are unlikely to "live up to that potential" because of all the obvious things that get in their way.

Is it not possible that individuals who are biologically very well suited for distance running in Kenya have a better chance of ending up as distance runners than they do in America, because Kenya specializes in distance running and America does not, and because America actually specializes in other comparable things that are more remunerative? Or perhaps that in Kenya, amateur athleticism is a relatively lucrative career compared to outside options whereas in America it is not.

Let me model this mathematically; say countries are homogeneous such that all persons born in the country j are characterized by a two parameter model that for the sake of argument we'll take to be normal: mean_j and sd_j. A draw from this distribution assigns natural talent theta_i to each person i. If they choose to train and run, we see theta_i, so the observed outcome is: y_i = D_i theta + (1 - D_i) 0. Thus training to run is a treatment (D_i); if you are treated you reveal your talent theta, if you are control, you don't run and no one knows. Then the characteristics of the population as a whole are characterized by mean, standard deviation, and n (population) and the observed distribution is just the distribution of y_i.

In all countries D_i = f_j(theta_i, p_j); p_j refers to the fraction of the country that runs; in a country where everyone runs, the function is constant 1. In a country where runners are a random sample f_j has no theta_i term. You could imagine other functions that are positively or negatively correlated with talent. An easily implementation of this would be to draw a bernoulli variable with p_j and some correlation rho_j with theta_i; you can do this with a copula or the analytic formula for biserial normal-bernoulli correlation, but let's say the detals of the function are irrelevant.

Finally, competition reveals max(y_i) for a country, and thus for the planet.

It is easy to find a world where the identity of the winning country j is the country that has the highest mean. It's also possible that the identity reflects mean-sd-n (e.g. all countries have the same average but some have more variation; or more births = more outliers). But it's also easy to assign viable p_j or f_j(theta_i) p_j such that the "wrong" country wins because of cultural factors.

It's true the model is a toy model (countries aren't homogeneous, talent may well not be normal, theta is not perfectly revealed and training has varying degrees of quality, individual race times have stochasticity), but even accepting the toy limitations wherein every cultural factor goes through f_j(theta_i) and p_j, it's clear we don't have a bead on mean_j.

So, like, what now?

I don't follow your argument well. If a country trains more people to get into distance running, it could definitely skew their average even if their etnicity has bad running genetics.

But "doesn't say anything about the individual people" point is that, if you take any one person, no matter what their parents' genetics and environment say (barring extreme examples), they have a chance to become a great distance runner. A very slim chance, but a chance. Similarly, there are people born from poor families who show a suprising amount of intelligence.

I'm not really sure how to re-state any clearer -- just try to follow the math -- but my underlying claim was that "differences in distribution don't tell you anything about individuals" misses the point: yes, of course differences in distributions don't tell you about a given individua. I'm saying observed differences in winners don't provide very strong evidence for _differences in distributions_.

There are several issues: winning reveals the maximum of the distribution, not the mean or sd. An observed pattern of one country winning might reflect different distributional means. It might reflect same means and different standard deviations. It might reflect that the number of observations is larger and thus the probability of an outlier at a certain number of sd above the mean is higher. But beyond that, there are also obvious non-genetic processes that even in a toy distribution can produce a winning country that has lower genetic capability. The entire enterprise of saying "Kenyans win running, so Kenyans are more genetically talented at running" is unsupported.

> The entire enterprise of saying "Kenyans win running, so Kenyans are more genetically talented at running" is unsupported.

Or as my father once framed the argument: "It's because they've only recently come out of the jungle right" and when I yelled at him explained his reasoning with, "That's not racist, stands to reason they'd be better at physical things, we've evolved over millions of years to not need that sort of thing any more." at which point my mum told me not to call my dad a racist and to leave the dinner table, which I most certainly did.

It's amazing looking back how racist the 80s were when I was growing up. The framing has gone from outright racism that was "common knowledge" everyone accepted, and been dog-whistled into this seemingly academic debate about the effects of genetic differences between population groups, which is so far removed from the original that it's hard to tell who really is "just asking questions" and who wants to talk about "coming down from the jungle" without being labelled a racist.

On a more technical level, was your equation meant to be

y_i = D_i theta_i + (1 - D_i) 0

i.e. "theta" was supposed to be "theta_i"

Genes describe lot of things from obese tendency to height to IQ to your looks. If someone has an IQ of 70, there are very likely to live a difficult life. Yes IQ is not intelligence you will say but still there is a limit to everyday functioning if someone's IQ is low. Ask a conventional ugly person(specially female) about their life.

I agree with you, but an IQ of 70 is mentally disabled, so it's not really a good example of "bad" genetics, so much as it is an example of a disability.

That seems like a strange line to draw, surely plenty of disabilities are genetic, implying bad genetics no?

Sure, but it's unlikely that you would have an IQ that low, and not have a specific medical disorder that is causing some of your impairment. E.g, Down Syndrome is a chromosomal change, something that can be measured outside of a 'bad genetics'.

What? You somehow think that if there is a named thing then it's not genetics. It's genetics both if it's a named thing, and if it's yet unnamed thing too.

My point is that the distinction is important.

Of course, named disorder or not, genetics can be the cause/have a high impact.

There is a thin line between disability and just different genetics.

E.g. is it bad for a primate to have bad bite strength? What if that disability enhances their thinking ability (by both allowing their brain to expand more since the muscles holding the jaw are weaker and by forcing them to use something other than their bite strength)?

Isn't that a rather arbitrary distinction? Feels you are making it easy for yourself to just dismiss a certain amount of "not-normal" people as not relevant to the discussion? So, what do you define as "bad genetics", then?

Hard cases make bad law. In other words basing our precedents and rules on extreme cases is not a good idea.

While extreme cases are useful for illustration and thought experiments, they are quite distracting when the real issue is small variances in the top ranges of ability.

> from obese tendency

Which should make you rather skeptical of inferring just how innate IQ really is (assuming it's a valid concept).

The heritability correlations for obesity are only slightly weaker than IQ. Yet we know for a fact the logical complexity of losing weight is simple (doesn't mean it's easy). Anyone can and people do lose weight, even if statistically it may be rather unlikely.

It's completely plausible to me that intelligence can be simple but hard to increase too.

If it's not a statistical anomaly and the genes truly do contribute to obesity, the causation must be very indirect and situational.

Not to mention, obesity has increased by an order of magnitude in 50 years, with no signs of stopping. It's actually a nice example of how little genetic variance can actually matter at the population level, especially since it is a much more objective measure than IQ tests.

Every phenomenon has both sides. While many people grew obese, there are still significant outliers among us who chug beer, consume current obesogenic diet and stay slim.

Is it genetics? Is it gut microbiota? (E.g. genetics of other species?)

> Is it genetics? Is it gut microbiota? (E.g. genetics of other species?)

If you don't control for portion size comparing people who all have a burger and fries for dinner every night is basically useless, those meals could be hugely different in composition, portion size or how much is left over. Even if dinners are identical, snacking can be thousands of calories a day that you might never see happening. You've also missed what activity they do as well...

Two people who both "chug beer and eat greasy takeaways" whenever you see them out can be very different the rest of the time. If one stays slim and one doesn't it's far more likely to be that the former eats smaller portions overall, snacks less between foods and does more activity overall, before reaching for more exotic answers.

Portion size is notoriously hard to control for in freely moving people. If people self-report their foods, they are likely to make mistakes.

I know of one study only where the numbers can be trusted, and it was done on prisoners, where every single gram of food that passed their lips was accounted for. They were fed the same diets, but their weight gains varied a lot:

https://dm5migu4zj3pb.cloudfront.net/manuscripts/106000/1065...

The only place you can see weight gains is by looking at Table I and working it out yourself for each subject, and only the last subject PW had an outlying weight gain in absolute terms but his starting weight was also an outlier - his relative gain was the same as all the others as well.

The paper is about how the weight of fat cells changed at different subject weights, not how much subject weight changed for a fixed food size - the only mention of food portions is with respect to ensuring weight was maintained, gained and lost for the different stages. "After desired or maximum obtainable weight was reached" indicates they weren't using weight as the outcome here.

The only place you can see the weight gains is by looking at Table I and working it out yourself for each subject, and the relative gains are all very similar: 23.3%, 25.3%, 21.4%, 20.0% and 18.0%. I'm guessing they wanted them to put on 20-25% of their body weight to give consistent and measurable changes to the fat cells.

You can't discount environment either.

We have fat raccoons, that implies that genetics doesn't matter as much as we think we does.

Same with intelligence, how much is it genetics and how much is that the environment doesn't promote desired traits.

Obesity among US adults has gone from 10-15% in the 60s to 40% now. It's extremely unlikely that genetics changed in 60 years, but environmentals did.

I understand that some have genetics that make it harder for them to keep off weight, either by metabolism, neural hunger stimuli or otherwise, but then we're still back to the fact that environmental changes made the biggest effect, not genetics. Why can't this be the case in your other examples too?

TL;DR: Yes, genetics matter. What's more relevant is, how much?

>It's extremely unlikely that genetics changed in 60 years, but environmentals did.

Due to epigenetics, environmentals actually shape genetics to a degree.

Epigenetics does not change your DNA or what you pass down to your children, and is reversible, so I wouldn't call that "shaping genetics", at least not in the way I was referring to.

Heritable epigenetics is a (very cool) thing, though apparently somewhat overblown.

But epigenetic changes during foetal development could effect the likelihood of being successful at passing ones genes along at all, although I've no idea how you'd manage to isolate or replicate these effects in any scientifically useful fashion.

Someone in these discussions always brings up epigenetics at some point, its impact there far exceeding its impact on population genetics I would guess :) Nobody ever talks about genetic drift or gene flow however, probably because despite being actual things that affect population genetics talking about "nearly neutral drift in small populations" is really hard to wedge into an ideological shouting match dressed up as a scientific debate.

I've always viewed genetics and talent as essentially your maximum potential. Lots of ways to drag you down from being all that you can be (lead, malnutrition, a good knock on the head), but given a sufficiently good environment, genetics (or perhaps inherent ability might be better) then become the limiting factor on how far you can go.

Your random guy on the street ain't going to be Newton, no matter how much he studies or loves math. Similarly, the man born with muscular dystrophy isn't going to be setting marathon records.

This is a simplification of course, but I hope it conveys the point.

I couldn't agree more on an individual level.

These difference in distribution though start to appear when you do statistics. And often an extreme percentile is used as a measure of whether there is discrimination (top engineer at google, board member at an international company, olympians). And it only takes minor differences in distributions (mean, standard deviation) to have big differences on such extreme percentiles.

>Genetics has a huge influence on what someone is predisposed to and what they can achieve.

That is not true. We've known that since the 80s. See https://slatestarcodex.com/2017/05/30/hungarian-education-ii...

The is a great point. Id like to add, society doesn't require geniuses to operate.

Reply on news.ycombinator.com