On March 23, I called your attention to paleoanthropologist David Reich’s op-ed in the New York Times, “How genetics is changing our understanding of ‘race’.” I thought the article was quite good, one of the few articles that takes a pretty objective and open-minded stand on “race”. It noted that conventionally named races are social constructs (that is, there is no homogenous “black” or “caucasian” race that is diagnostically different from other races), but that even the social constructs reflect elements of history (different groups evolved in geographic isolation, and their genetic constitutions, which still reflect that isolation, can be used to reconstruct evolutionary history and give some help with medical diagnoses).
Reich notes that there has been ample time for different populations to have evolved genetic differences and that, although most variation in humans is found within groups rather than between them, we still do not expect all groups to be exactly equal in any trait that is genetically variable—an expectation that goes for both morphology and behavior. Finally, Reich makes the point—one that I’ve always emphasized—that even if there are group differences, that says nothing about how we should treat individuals, and we need to make it a moral principle that all individuals, regardless of gender, ethnicity, or other biological status, should be afforded equal opportunities, as well as personal treatment based on individual qualities rather than group membership. It is always unwise to predicate moral views on biological realities, for that makes your morality vulnerable to future empirical findings in ways we don’t want. (Of course some aspects of morality, like views on abortion, can vary depending on what science finds out. But I don’t see race this way, as I can’t imagine any genetic discovery that would alter the kind of equality I want in our species.)
This is a reasoned approach to the data, but there are many scholars who reject it, for they are aware of the history of eugenics in which “racial” differences were used to discriminate among (and even kill) people. That happened, and we must be aware of it. But the solution is not to simply deny science or reject whatever science finds about ethnic groups. Rather, we must ground our morality on a fundamental equality of humans based on their individuality.
We should never use ideology as a basis to accept or reject science. That way lies both madness and dissolution, as evidenced by the Lysenko affair in Soviet Russia, in which “Western genetics” was rejected in favor of a bogus form of heredity more congenial to the socialist view of human malleability. The result: millions starved to death. For both gender and ethnicity, much of the Left has an a priori assumption that all groups are equal—something Pinker questioned in his book The Blank Slate. The denial of science, or suppression of research, comes from the fear that any differences could be used to justify sexism, racism, bias, prejudice, and lack of opportunity. But Reich and I are concerned to show that we can have our genetic cake and eat it too: we can create a society of equal opportunity while still studying group and gender differences in our species.
If Reich’s essay had any weakness, it was the conflation of “race” with “population”, though he was concerned with that. As for me, I’m happy to abandon any traditional racial classification of humans, or even the word “race” itself, so long as we replace it with terms like “population” or “ethnic groups” that can help guide genetic and evolutionary research. These other terms have a biological reality not contained in the conventional (i.e., erroneous) use of “race”. Futher, recognizing populations and ethnic groups, fuzzy as they are, is essential in understanding the evolutionary history of humans—and has medical implications as well. What we always need to remember is that human evolution involved the differentiation of geographically isolated groups who evolved some differences, but now, with human movement, those differences are blurring, so that we have a fuzzy and overlapping set of populations. And yet those populations still show statistical differences that are useful. If they didn’t, you couldn’t spit in a test tube and have places like 23andMe give you a pretty accurate take on where your ancestors came from. Nor could we use genetic patterns in modern humans to reconstruct our migration throughout the world. The patterns remain, and have afforded immense understanding of our evolutionary history.
Reich’s article was excerpted from his new book, which just came out and will surely be worth reading (click on screenshot to order):
Reich’s piece, temperate as it was, was widely criticized by readers, who wrote letters to the New York Times. That paper then gave Reich the unprecedented opportunity to respond in a longish piece that was published two days ago. You can read it by clicking on the screenshot below.
Readers were still concerned that science could be used to justify prejudice and inequality, and Reich once again says that we don’t have to let that happen. Some readers repeated the conclusion (first suggested by my advisor Dick Lewontin) that the notion of group differences is meaningless since most variation in our species is within rather than between populations. That’s true, but people don’t understand that this doesn’t bar us from using constellations of genes to discern (not define!) groups and learn something about population structure and human evolution. Finally, in his response Reich lays out six principles which seem eminently reasonable. Number 6 is especially worth your attention.
From my point of view, it should be possible for everyone to hold in their heads the following six truths:
1. “Race” is fundamentally a social category — not a biological one — as anthropologists have shown.
2. There are clear genetic contributors to many traits, including behavior.
3. Present-day human populations, which often but not always are correlated to today’s “race” categories, have in a number of instances been largely isolated from one another for tens of thousands of years. These long separations have provided adequate opportunity for the frequencies of genetic variations to change.
4. Genetic variations are likely to affect behavior and cognition just as they affect other traits, even though we know that the average genetic influences on behavior and cognition are strongly affected by upbringing and are likely to be more modest than genetic influences on bodily traits or disease.
5. The genetic variations that influence behavior in one population will almost certainly have an effect on behavior in others populations, even if the ways those genetic variations manifest in each population may be very different. Given that all genetically determined traits differ somewhat among populations, we should expect that there will be differences in the average effects, including in traits like behavior.
6. To insist that no meaningful average differences among human populations are possible is harmful. It is perceived as misleading, even patronizing, by the general public. And it encourages people not to trust the honesty of scholars and instead to embrace theories that are not scientifically grounded and often racist.
In short, I think everyone can understand that very modest differences across human population in the genetic influences on behavior and cognition are to be expected. And I think everyone can understand that even if we do not yet have any idea about what the difference are, we do not need to be worried about what we will find because we can already be sure that any differences will be small (far smaller than those among individuals).
Reich’s original piece (and response) was not sufficient for 68 scholars, who wrote a joint piece, “How not to talk about race and genetics” (clearly named after Reich’s piece above), taking Reich to task for his “misunderstandings.” This piece was apparently submitted to the NYT but was rejected, so it was published in BuzzFeed.
The thing is, most of the things these scholars criticize were already taken into account by Reich, including the notion that conventional races are social constructs. But the 68 also object to the notion of “populations”, an objection that is unwise given that they admit later in the piece that there are differences between populations—they just don’t fall into the conventional categories of “race”—something that Reich already admitted.
Their main beef seems to be that those like Reich who unravel the genetic patterns of our species need to constantly consult with people like cultural anthropologists and social scientists. I’m not sure this is good advice, since those people have, by and large, tried to foist ideological views onto research on human groups. The bit below, for example, smacks of an unwarranted hubris:
Precisely because the problems of race are complex, scientists need to engage these issues with greater care and sophistication. Geneticists should work in collaboration with their social science and humanities colleagues to make certain that their biomedical discoveries make a positive difference in health care, including the care of those studied.
Of course discoveries should be used constructively, but is that the responsibility of people like Reich, who simply look at the frequencies of disease genes in different groups and the pattern of genetic differentiation across the globe? I don’t think so. How to use the findings of geneticists in medicine is the purview of bioethicists and physicians, not paleoanthropologists.
And have a look at this:
Even “male” and “female,” which Reich invokes as obviously biologically meaningful, has important limitations. While these categories help us to know and care for many human beings, they hinder our capacity to know and care for the millions of human beings born into this world not clearly “sexed.’ Further, overemphasizing the importance of the X and Y chromosomes in determining sex prevent us from seeing the other parts of the genome involved in sex.
Well, yes, there are people not clearly “sexed”, but the categories of “male and female” fit the overwhelming majority of humans, and have and can lead to useful research: both biological and medical. What we see here is more Pecksniffery that invokes rare exceptions to criticize a binary classification that, in the main, is correct and useful.
As for the signatories, there are some geneticists and biologists among them, but they’re outnumbered by anthropologists (I suspect mostly cultural anthropologists), sociologists, physicians, gender and ethnic studies professors, biomedical ethicists, historians, and professors of law. In short, just the mix of people you’d expect to object to Reich’s reasonable take. But you can read and judge for yourself.
h/t: Andrew
Racial-profiling is why blacks are overwhelmingly the ones diagnosed with sickle-cell anemia. [/sjw mode]
Glen Davidson
Also why the incidence of HIV infection is so high in some African countries.
Racial profiling is why the incidence of HIV infection is so high in some African countries? What on earth is that supposed to mean? In any event the incidence of HIV in some African countries is not due to some sort of ‘racial’ propensity to contract HIV, but to the neglect of preventive measures, on the governmental and individual level.
Yeah that was weird.
I think he was referencing ccr5 and hypotheses about Europe and the plague, smallpox, etc.
I was thinking ‘racial propensity’ in the sense that Africa is the home of wild chimpanzees (origin of HIV), Africans have a very low frequency of the CCR5 delta32 allele that is associated with resistance to HIV infection and Africa has a high incidence of TB which interacts with HIV co-infections.Plus of course the factors you mention. In other words nothing to do with racial profiling. I was trying to be sarcastic (obviously unsuccessfully. Apologies).
Your comment, meant I think to be provocative, is a non-sequitur. What’s actually true is that the sickle-cell phenotype was discovered and characterized in West African descendants in the West due to its higher prevalence in that population. There are, however, other populations quite far removed from Africa with a similarly high prevalence of the sickle-cell phenotype (West Bengal and Bangladesh for instance). There are also populations in Africa (South Africa actually) that have a comparatively low prevalence of the sickle-cell trait, despite being phenotypically members of a group that most Westerners (thinking US and Europe here) would identify as black. BTW, in apartheid South Africa both the native indigenous African, West African (which is primarily who African Americans are descended from) and West Bengalis would be “blacks”. Not surprising but go figure. This is where the cognitive dissonance seems to lie for those fixated on racial categorization, what the colloquial term means to those who use it and what it actually represents and why that came about.
Judge Starling on Tumblr who is the evolutionary biologist Dan Grauer has some pointed criticism of Reich.
I love Grauer, just his character is amazing, but he’s a neutralist. His criticism is expected.
*Graur
I was involved in a heated group discussion this weekend with a woman who insisted that there was really no biological difference between men and women. After I picked my jaw up from the floor, I pointed out that we not only had different organs and body shapes, but in fact different sets of chromosomes. Others pointed out things like testosterone levels and its effects.
She dismissed all that as irrelevant. She felt that these facts were the justification for patriarchical societies (which she insisted were recent constructs only a few hundred years old), and since such societies were unfair, then the facts must be wrong.
Of course, reasoning with her was futile. And that’s the roadblock one always encounters with idealogues of any political stripe.
You cannot use facts to debate a person whose views were not arrived at through reasonable analysis of facts.
Some of these folks do not even believe that there is such a thing as objective truth. So, not only will a demonstration of facts not influence their views, they dispute the very existence of facts.
Precisely. In the end, the woman shouted angrily that she was right and we were wrong, and stormed off in a huff.
I don’t doubt racists and eugenicist types will hate the idea that being white isn’t a sign of superiority from body to mind, and will therefore reject it outright. Like when it became apparent we likely all stemmed from the African continent.
Of course there are some genetic differences, for instance black males on average have higher levels of testosterone/ and free testosterone than white males. Not exactly a different species…
Racism is such a horrible blight, I hope findings like Reich’s can help in some way to dissipate that mentality.
Andrew Sullivan waded into this with a nice piece in NY Mag that will probably bring out the haters (I guess many folks will have seen it but in case)
http://nymag.com/daily/intelligencer/2018/03/denying-genetics-isnt-shutting-down-racism-its-fueling-it.html
Further to Glen Davidson’s comment above; I agree, nobody seems to get up in arms about racial health disparities research and the investment of funds to address these. As Reich notes African Americans suffer disproportionately from prostate cancer and there are a number of initiatives to address this (as there are for other diseases that disproportionately afflict minorities). A genetic and/or environmental basis for many of these conditions is slowly emerging.
I wonder how someone who maintains that race is purely a social construct with no biological basis could support affirmative action, or for that matter Black Lives Matter.
Since racism is a result of cultural conditioning, which is true, then those initiatives are necessary to counteract the resulting disadvantages. Just being a social construct doesn’t mean it’s not real.
That leads to a contradiction. The criterion for determining who should be advantaged by affirmative action is their declared race, as proven by the color of their skin and their overall appearance to whoever judges these things. You can’t logically be for affirmative action without being a “racist” in some sense (not the pejorative sense).
Is there any sense — other than the following senses — that I haven’t heard of?
a) a belief in the superiority of certain races vis-a-vis others.
b) antagonism towards other races based on the kind of belief defined in (a).
c) the theory that human abilities are determined by race.
These definitions seem perfectly adequate and genuinely descriptive without being pejorative. Why so you feel these definitions are pejorative?
“Racism” in everyday speech is a loaded term that means something clearly negative, except among some small number of deplorables out there who admit to it.
In neutral speech, a racist is someone who believes that races exist, and that’s the central question. It doesn’t necessarily imply that one race is superior to another.
In the case of affirmative action, a supporter has to be a racist in that neutral sense.
Some use the term racialist as the value neutral term, though many consider racist and racialist equivalent. It would be good if there could be two terms for two different things, but all too often in reality IMHO the former is a pretty strong indicator of the latter.
I’m happy to use “racialist” instead of “racist”, but it seems to be anodyne.
I suggest that you look up ‘racism’ and ‘racialism’ in a good dictionary.
According to Wikipedia: “Racialism is the belief that the human species is naturally divided into races, that are ostensibly distinct biological categories. Most dictionaries define the term racialism as synonymous with racism”.
https://en.wikipedia.org/wiki/Racialism
The Oxford Living Dictionary defines racialism as “another term for racism”.
https://en.oxforddictionaries.com/definition/racialism
‘In the case of affirmative action, a supporter has to be a racist in that neutral sense.’ That is pernicious nonsense.
So I looked up the dictionary definitions of “racism” and “racialism” that you suggested I do. The consensus is that they’re synonyms, with “racialist” more British and “racist” more American. Would you like to comment?
“Racist” in a value-neutral sense means someone who believes in races. That should be obvious. In a value-laden sense it means someone who promotes one race over others. No one likes to be called a racist (except for some value-laden racists who just don’t care), so the word is toxic. I get that.
“Since racism is a result of cultural conditioning, ”
You really think it is that simple?
The Trumpian Party is a close comparison to the racism of nearly 200 years ago and it comes out harshly in all areas of our society. From immigration, deportation, building of walls and even why more than 10% of the citizens in Puerto Rico continue without electrical power. No science required and none excepted.
sorry about that – accepted…
Here’s another perspective from Dan Graur.
[JAC: Sorry, you’ve just embedded his entire articles in the comment, which makes it too long. Just give the links so people can go over and see it. It might be salutary if you tell us exactly which of his criticisms are valid as well. ]
I think some of his criticisms are valid. There’s a lot of unknowns about the genetics of Educational Attainment and cognitive differences between races. But of course David Reich is certainly right in most of his article and certainly have no bad intention whatsoever, I just think he’s going to fast with some of his assertions.
Agreed.
Because of the long history of slavery and racism (which has been far from eliminated today) in the United States, for those who fear an even greater resurgence of the latter have developed a hypervigilance in seeking ideas or events that may fuel it. It is almost a form of PTSD. All perceived threats must be extirpated before they spread throughout the body politic. They have mentally determined (perhaps unconsciously) that if the truth must be suppressed to achieve this end, so be it. I sympathize with, although do not support, this viewpoint. I have little doubt that racists will grab upon Reich’s article, greatly distort it, and use it to promote their hateful views. This is a reality that cannot be ignored.
In the long run, attempts to suppress scientific findings will backfire; the truth will win out. To accomplish this as quickly as possible, the scientific community must speak out loudly and often, to show that Reich’s article is not racist. On the one hand, they must convince those who fear a racist resurgence that Reich’s views are not racist and that the scientific discoveries that he talks about can help minority communities. On the other hand, they must rebut the racist distortions whenever and wherever they are found. This is a mighty task for the scientific community (specifically geneticists), whose members I suspect are largely not very political. I hope they are up to it
I hope that they are up to it, too.
Wonder whether commenter GM will show up this fine Easter Sunday to downplay Uncle Joe’s Ukrainian Holodomor once again.
Maybe starvation is a social construct?
If you expect to rate as a gentleman, it’s “expectorate,” please.
Seriously, though, pound-for-pound, inch-for-inch, this is one of the best, most well-written apologias on the topic I’ve had the pleasure to read.
I did the linky thing over to Reich’s Times piece, and it deserves a similar accolade.
“Some readers repeated the conclusion (first suggested by my advisor Dick Lewontin) that the notion of group differences is meaningless since most variation in our species is within rather than between populations.”
As I always say when this comes up, this is simply a mathematical fallacy that Lewontin, Lande, and many population geneticists have committed. Even if the groups belonged to different species and shared no alleles, most of the variation (as calculated by Lewontin using entropy, or as calculated by Lande etc using “genetic diversity” (heterozygosity)) could still be within-group, when within-group diversity is high.
In other words, the fact that most of the “diversity” is within-group has no bearing on the degree of differentiation between groups, if diversity is measured by the formulas that population geneticists like Lewontin and Lande have typically used. Anyone can check this by simply calculating the results for a test system with high within-group diversity (try 20 equally-common alleles) but no shared alleles between groups.
These mathematical misconceptions may actually be distorting the picture of human evolution and similarity between groups. See my discussion with Greg in the comments under the original WEIT post about Reich’s article:
https://whyevolutionistrue.wordpress.com/2018/03/23/finally-a-sensible-discussion-of-race/#comment-1593756
Greg brought up a 2002 paper by Rosenberg et al supporting Lewontin. That paper shows no awareness of the mathematical fallacies being used. I dug around a bit and found a more recent paper by Rosenberg and others after they had noticed the fallacies empirically; their new conclusion was that purely mathematical effects were sometimes obscuring biological effects in the analysis of genetic differentiation in human populations.
My 2008 paper, and my earlier papers cited therein, explain this and show how to do the analysis properly:
https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1365-294X.2008.03887.x
It may well be that a correct analysis will confirm Lewontin’s conclusions, but we don’t know that.
I will explore Lou’s comment above when I have time, but meanwhile I have been wanting to ask two questions for some time, and might as well ask them here.
1.” Most variation in humans are within groups rather than between them”. What does this mean? I have never felt a firm grasp of this point. And,
2. ‘Races as we understand them are a social construct’. Maybe I already have a handle on this since it seems kind of obvious, but I want to make sure. Is this statement pointing out that the traditional races like ‘black’, ‘white’, etc. are really crudely defined groups, based on just a few distinctions like skin color and continental history? Meanwhile, those groups are really divided into several genetically and geographically defined groups?
Mark, I am not the best person to answer your first question, since I don’t think that the concept makes sense. But I can tell you what people often intend to mean by that statement. They take some measure of the diversity of a population, for example Shannon entropy (Lewontin)or heterozygosity (the probability that two randomly chosen individuals have different alleles at a given locus). They calculate this measure within each group, and take the average. That’s the within-group diversity or “variance”. Then they do the same for the pooled groups. They have the intuition that if the groups are very different genetically, then pooling them should greatly increase the heterozygosity. If they find that pooling the groups hardly increases the heterozygosity, then they cconclude that most of the diversity (or variance) is within-group, and so the groups don’t differ much.
The problem with that, in the case of heterozygosity, is that heterozygosity is a probability and can approach but never exceed unity. So when within-group diversity is high, i.e. close to unity, then the pooled heterozygosity must also be close to unity, so the difference between them (the “between-group variance”) must be close to zero, even if the groups are completely different (share no alleles at all).
Well, you humbly say that you may not be the best to address the first question, then you go ahead and address it in a way that I can pretty much understand! That was excellent, and so thank you!
How exactly would within group heterozygosity be high (that is a majority of bases in a given ROI have two or three alleles) and approach unity. And what do you mean by unity?
johnw, my argument is a mathematical one. I am just pointing out that Lewontin’s (and most other population geneticists’) inferences are not valid (i.e. the conclusion does not necessarily follow from the premises). For many loci the inferences may still be true, by accident, but this requires additional analysis; the ratio of within-group to total “variance” is not relevant.
There are loci, such as those of the Major Histocompatibility Complex, whose heterozygosities may be high.
I am not sure why you ask what I mean by “unity”. Unity is a word that means 1.000…
Understood, sorry wasn’t meaning to be a jerk. Just trying to wrap my head around heterozygosity approaching 1 and thinking in terms of genomic sequencing SNP data and that heterozygosity in humans is actually relatively low (for a primate), though obviously for a few loci could be 1.
Lewontin etc are dealing with whole genes, not SNPs, and so I am also addressing whole genes; now I think I understand why you asked your question about unity, since heterozygosity for SNPs can’t approach unity, since there are only four possible bases, but heterozygosity can approach unity for loci containing many alleles.
Ok here’s the links of the two Dan Graur’s articles I mentioned in my previous comment.
Article 1
Article 2
I hope this time it will work.
I have looked at the Dan Graur posts you have cited. He teaches biology at the University of Houston. I am not competent to judge his comments on Reich, but his conclusion is clear:
“Well, David Reich may be a big-shot geneticist at a very prestigious university, but his knowledge of population genetics and evolutionary biology seem to be meager at best, and his misrepresentation of the data and nomenclature can only be described as brazen.”
I hope that somebody knowledgeable of genetics will respond.
Dan Graur is the guy that wrote this infamous paper criticising ENCODE. He knows what he is talking about when it comes to population genetics.
http://gbe.oxfordjournals.org/content/5/3/578.abstract
Dan is always interesting. This part in the first link is especially worth considering:
“Let us now look at another paragraph in David Reich’s op-ed piece:
What a load of nonsense! For selection to operate and counteract the effects of random genetic drift, the effective population size should be large. Unfortunately, the long-term effective population size for all the humans in the world is barely 10,000—lower than that of chimpanzee. By necessity, the effective population size of each “race” separately is much smaller. So, the chances that 74 loci will experience significant changes in allele frequencies simultaneously in each of the four populations is zero. ”
The whole issue is very interesting but also complex. Right now I don’t know what to think but I will continue watching the debate.
From the little genetics I know, it seems to me that the “ability” of selection to counteract the effects of genetic drift depends not only on population size but also on the magnitude of selection coefficient. To me, the fact that separate mutations leading to lighter skin color were fixed in European and Asian populations suggests strong selection pressure towards lighter skin tones in areas with lower solar radiation. Selection was also important to increase the frequency of the sickle-cell mutation among some Africans and, I suppose, of thalassemias and G6PD deficiency among some Mediterraneans. Among livestock breeders, selection endorsed some mutations leading to lactase persistence. The latter change happened in much less time than 40,000 years.
Besides, I do not see why all changes due to genetic drift should be dismissed.
This comports with my understanding as well. Ceteris paribus, selection is more effective in larger populations but strength of the selection coefficient plays a role when ceteris is not paribus. I also fail to see why drift wouldn’t potentially play a role in genetic structure of human populations.
The traits you mention are influenced by just a few genes and a very small number of alleles. SNPs influencing the mode of expression (which is subjectively defined at best) of a behavioral phenotype is an entirely different thing.
Yes, and I am meanwhile thinking this through better, along those very lines. One cannot just state the primacy of drift in small populations when clearly there are many cases where selection still works.
One example I stumbled across is in an experiment on flies, who were selected to switch their preference from one species of host plant to another. It was found that they quickly did so in one generation, selecting upon well over 100 loci.
I’d love to see that paper. Can you please post the citation?
My impression of Dan Graur is that he prefers his population genetics to be idealized to theoretical populations and theoretical genes. It’s fine to say that David Reich is spouting nonsense because it’s theoretically unlikely for 74 loci to exhibit significant differences in allele frequencies between four geographically isolated populations but as 23andMe has demonstrated, they do. When observation conflicts with theoretical prediction isn’t it time to rethink the theory, not dismiss the observation?
What was that Sherlock Holmes quote? “It is a capital mistake to theorize before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts.”
The problem isn’t the existence of the correlation, it’s the attribution of causation to 74 loci (15 actually according to Graur) that are probably just SNPs. If you wanna say this genotype produces that phenotype, you still have a loooong way to go. Extraordinary claims require extraordinary evidence.
Greg Cochran wrote at his site West Hunter:
https://westhunt.wordpress.com/2018/04/02/the-usual-suspects/
“Next, Grauer talks about the long-term effective population size (EPS) in humans being around 10,000 – which he seems to think is too low for selection to work. He’s wrong, but that number is also the wrong number. He’s talking about the neutral-theory effective population size, which is a function [ harmonic mean] of long-term population size over the last million years or so – and it’s the wrong ‘effective population size’ for selection questions. There are different values of EPS for different questions. The right one is the “Variance EPS, which tells you the theoretical population size that yields the same noise in allele frequency change, right now. It basically measures how little a population is susceptible to drift.” The correct value of EPS, for different continental branches of humanity, has been hundreds of thousands to millions for a long time, well before agriculture was developed.”
@ MEH 0910
Whom would you trust if you have no background in population genetics?
Dan Graur:
An admittedly polemic professor in Genetics, author of tons of important papers on the topic, editor with some of the most important Journals in the field, author of highly respected book on molecular evolution.
or:
Gregory Cochran
Who is a Physicist by training, has almost no
papers on the topic (none of high profile) and believes that homosexuality is caused by infection… (I checked the references given in Wikipedia)
https://en.wikipedia.org/wiki/Gregory_Cochran
Best,
Michael
P.S.
I agree with our host & Reich that there are observable differences in human sub-populations (irrespective of how they were generated).
I am no geneticist but article 1 seems to depend on a fallacy. If 74 loci were selected a priori at random his claim of near impossibility would be sound, but that is not how those 74 were identified. This is a version of the blade of grass fallacy. Hit 74 golf balls. The chance they land on 74 preselected blades of grass is nearly zero, but the chance they land on the blades they actually landed on is 1.
You misunderstand. His point is that, in terms of allele frequency changes and subsequent changes in phenotype over thousands of years, all we have as examples are things like lactose tolerance and skin pigmentation which involve just a few genes at most. 74 loci altering in concert to produce a gradient of phenotypic expression of a complex (and ill-defined) phenotype is what he’s objecting to on the basis of prior credence. He has a good point. Also, the fact the authors making the claim did not apparently share the data or specify how they accounted for multiple comparisons is concerning. Mining for associations in GWAS or gene expression data are a tricky exercises and it is best to be modest in ones interpretations. My personal experience is in gene expression data mining and attempting to use microRNAs for diagnostics. In both cases, results in one cohort usually end up being useless in another.
No. The loci were identified ex post as it were, by examining the differing populations looking for variant loci.
If the goal is to not give racists any justification for their beliefs, how on earth would suppressing/hiding/ignoring/sweeping the truth under the carpet achieve that goal? The truth will be found by racists because the truth is discoverable and so what could give them more ammo than the fact that the left has been trying to hide the truth?
Suppressing the truth will aid racist ideologues not shut them down. Besides, who doesn’t think it’s a good idea to tell white supremacists that on average, Asians and Ashkenazi Jews are smarter than them.
“If the goal is to not give racists any justification for their beliefs, how on earth would suppressing/hiding/ignoring/sweeping the truth under the carpet achieve that goal? ”
I think you underestimate the power of this strategy.
“Besides, who doesn’t think it’s a good idea to tell white supremacists that on average, Asians and Ashkenazi Jews are smarter than them.”
… and you think this will convince the white supremacists to be more accepting of those clever Asian / Jewish smartasses? 😉
cr
At least some of 68 signatories are probably extremely concerned that some of the findings will contradict key pillars of their fields. Perhaps most of all among the gender study fields.
It’s really striking to me how much better a creationist/non-evolutionary model fits with gender theory rather evolution.
It’s as if human beings were sui generis among all creatures……
A problem I have with Lewontin’s analysis is that, while perhaps fine at the level of population genetic analysis, that principle of analysis (comparing within-group and among-group variance) is not what is used at the level of phylogenetic analysis, which relies on synapomorphies, ideally fixed ones. I wonder if Jerry and other knowledgeable individuals here agree with my perspective.
Appreciate the highly learned comments above, very instructive.
However, I have some trouble with point 4:
“Genetic variations are likely to affect behavior and cognition just as they affect other traits, even though we know that the average genetic influences on behavior and cognition are strongly affected by upbringing and are likely to be more modest than genetic influences on bodily traits or disease.”
Is that so? Is genetic influence on behaviour and cognition likely to be more modest than genetic influences on, say, disease? Many causes of diseases have a very strong environmental component. I suspect that ‘4’ is just a genuflection to PC-ness. I don’t buy it.
There is an unbelievable level of denial about the genetic component of behavior and cognition, not only between groups but also within groups. That is, if educated parents have a child who advances poorly in school, good luck to them trying to prove that this is not their fault!
(Upbringing is of course extremely important, but genes put a ceiling.)
The general result from behavioral genetics is that variation in the way a family raises a kid – over the range of possible ways that are fairly common in a typical western society – does not have much influence on IQ.
If you did something really unusual, perhaps it would make a big difference.
That’s not correct as far as I know. Adoption studies indicate that upbringing environment makes a huge difference and in terms of heritability of intelligense(whatever that is) changes things entirely. Genetics makes a bigger difference the further up the SES ladder you go, and lesser of a difference the further down you go. [Nisbet et al PMID: 22233090]
As long as Reich acknowledges that ‘race’ is fundamentally used as a social category , I have no quarrel with him.
Like Mark I would have to study Lou’s analysis more, even though I think I understand the mathematical corner case problem; say, is the corner case an actual problem in real human populations where genome wide heterozygosity apparently can be of the order 0.1 [ https://www.nature.com/articles/ejhg200848.pdf ]?
Meanwhile, I am baffled by this:
We know from the last decades that we can use ADMIXTURE to cluster groups in even smaller clusters, and from PCA analysis that the SNP variation has predictive components at the 0.1 % level that maps to European geography at roughly 100 km distances [ http://blogs.discovermagazine.com/gnxp/2008/08/genetic-map-of-europe-genes-vary-as-a-function-of-distance/#.WsFIzIi5vIU ]. But since those PC gradients are continent wide they seem to describe a fuzzification that has impressed on top of recent gene flows, presumably by recent generation marriages on those distances doing the smoothing. [Ha! at this point I googled up the reference link, and reading I see that Razib came to the same conclusion: “Note the rough correspondence to the geography of Europe in terms of spatial relations. This should be no surprise considering that for all practical purposes marriage networks move across two dimensions and drop off in likelihood as a function of distance.”]
Is this not confusing the two primary location gradients and their cause in old migrations and persistent pair marriage family customs, with the question of remaining population differences and their usefulness?
I fear that I have been ‘blurring’. All of my children are issue of a combination of different ‘traditional races’. So there. 🙂
Their main beef seems to be that those like Reich who unravel the genetic patterns of our species need to constantly consult with people like cultural anthropologists and social scientists.
in the chapter from the book reich says that geneticists need to break their alliance with anthropologists because of the latter’s confusions as to the science.
the book chapter is milder in style but more ominous for ppl like the above in substance.
that’s interesting
I would expect of him to promote better communication with anthropologists, if that was his criticism, not breaking alliances.
the way science is becoming more and more specialized, we need more interdisciplinary work
also, the book chapter is better cuz it doesn’t have to fit into the op-ed length constraint. some things were edited out which made things clearer….
This is all largely a moot point, of course, as Dr. Chopra and Pellitier assure me that all of us have the ability to “Change your genes, change your life.” I assume this is just an interesting musing until we all epigenetize? (I’m kidding, of course, I couldn’t help myself.)
I gotta say, I have a hard time wrapping my head around the statistics (Math! Ah!) of heritability. Until probably two days ago I shared what I now realize is a colloquial misconception of this term, that’s it’s an easy formula for “X percent of this trait is in your DNA. We know because we saw it under a microscope or something.” When looking at how heritability is actually determined, I now find I now have no idea what the logical conclusions of this construct actually are. Google tells me that heritability actually changes depending on the environment and the people included in the group you are studying in the first place – if that is the case, what it actually tells us about anything is not clear to me. I can kind of wrap my head around the idea that degree of variance in genes as compared to degree of variance in observed traits would presumably serve as a basis for sussing out something about the relationship between genes and traits. But then I read about implications in populations like the heritability of hair color in homogeneous populations actually being *zero*:
http://www.indiana.edu/~p1013447/dictionary/herit.htm
…and I think… “Yeah, ok, I don’t really get this.”
Maybe this is a side point and there’s some other measure that researchers have in mind (I notice the term ‘heritability’ is not used in this post, for example,) but it’s not clear to me what they would be. Given that the idea of heritability now makes little sense to me, it’s not clear to me what is even being discussed in these debates, as heritability sounds like a very fluid construct without clear contours. Maybe there are different ways of measuring it, but if not, I have a hard time understanding its logical conclusions when the heritability of eye color can be zero one minute and close to 100% as soon as you include other people in the sample. Given that, I have a hard time understanding how heritability can tell us much of *anything about a subset of people smaller than all of humankind. It seems to me that once you look at any smaller population you would see all those weird and fairly misleading framing effects. Obviously eye color *is heritable in the colloquial sense, and yet when you frame it by statistically isolating a given population, suddenly the statistics reflect the idea that it is not (which is not technically incorrect, I guess, but certainly misleading in terms of what we are actually looking to find out when studying heritability) – so wouldn’t the same be true of other traits? And if heritability varies wildly depending on framing, then how is it possible to talk about nature nurture at *all (again, only for populations that are smaller than ‘all of humanity.)
At a social, personal level, I really dislike the idea of talking traits and race at all – I don’t see what the point is. Is there really any compelling reason to conduct race-based science? If so I haven’t heard one. And at the level of scientific understanding, this is all kinda beyond me anyways, so I’m not sure what to make of all this.
“Is there really any compelling reason to conduct race-based science? If so I haven’t heard one.”
One argument I can think of;
lets say there are substantial scholastic differences between Australian aboriginal children vs people from European/Asian heritage, then surely the best/appropriate policies to help the children will be very different whether genetics is a major factor vs nuture, poverty etc.
Academic support should be available to any student who needs it, though, so I don’t see how race should be a factor there. Schools, at least in the US, already have systems in place for identifying children who are falling behind academically and getting services in place for them.
I can see an abstract case that we simply shouldn’t have taboos on *any area of potential inquiry, I guess – but I doubt there’s anyone who wouldn’t draw a line on that *somewhere.
Hi Roo,
Academic support should be available to any “student who needs it, though, so I don’t see how race should be a factor there”
I agree, but that is not how things work in the current political climate.
Imagine you are the Minister of Education and you are asked why the aboriginal children are so far behind in academic achievement.
It is not unreasonable for people to blame the teachers, institutionalized racism and a host of questionable social theories if group IQ differences are off the table.
The same applies to gender differences, if you have 70% women graduating as nursers and only 30% women as software engineers you might assume sexism, whereas the reason is innate preferences men and woman have in people vs things.
I suppose you are aware of the famous Google Memo which was about the lack of demographic representation in the IT sector.
‘It is not unreasonable…’, etc. So are you suggesting that institutionalised racism plays no part in these matters, and that the real problem is that aboriginal children lack intelligence?
“So are you suggesting that institutionalised racism plays no part in these matters”
No I do not, my bad for expressing myself poorly. I think both environmental (including current and historical discrimination) and genetic factors play a role.
I think the vitriol that has begun to accompany the conversation about equality of outcome is probably a reason *why we’re hearing this line of thinking right now (i.e., people want a perceived alternative to being blamed all the time, a retort to identity politics) but in practical terms, in my understanding at least, I don’t think science could *actually resolve those kinds of political fights either way. First, I think our understanding of genetic markers, their interactions with each other, the environment, and so on, is nowhere near that specific. Second, I think even if genetic causality could be clearly established, it wouldn’t speak to causality in an ultimate sense – one example I read was that, until a few decades ago, you hypothetically could have found clear genetic markers correlated to being a person who wears earrings (as they were worn mostly by women) and if you looked now, that apparently ‘genetic’ picture would have changed drastically. Similarly, a hundred years ago the genetic markers for hearing impairment would have been correlated with academic performance in a hugely significant way, an effect that would have changed a great deal as soon as hearing aids, schools utilizing sign language, and cochlear implants became common. Or, in the case of IQ, intensive ABA claims to show fairly dramatic increases, even though children in either a treatment or control group might have similar genes in terms of autism risk factors.
So on the educational front, I still don’t think that changes the basic dynamic – i.e., if one thing isn’t working, try something else. Research something else, develop something else, keep trying, until you find a method that does work. Maybe that’s at the societal level, like addressing racism, or maybe it’s coming up with teaching methods that are specific to different cultures or maybe it’s something different entirely like looking at vitamin intake or childhood stressors in different areas. But again, it’s not like in other cases of inequality we simply shrug and go “Well, genetics, too bad.” To use an extreme example, where genes (often) actually *do play a clear role – parents of special needs children are an incredibly formidable group, in my experience, and having a genetic explanation as concrete as Down Syndrome doesn’t change that at all. In fact they are often one of the *most vocal groups in pushing for more educational spending and reform. So on the political front, I don’t think it solves any of the problems of polarization and vitriol that we seen right now either – that’s more about interpersonal dynamics and being charitable to one another.
Thank you very much, Roo, for that good and fair-minded comment.
Thanks!
Hi Roo,
Your arguments are valid and well expressed.
Please remember my response was to your question
” Is there really any compelling reason to conduct race-based science?”
My point is just that if there are cognitive differences between groups then equality of outcome might not be achievable or realistic, thus a scientific understanding of group differences might be relevant.
Loat me at I agree, but….
Damore memo was gaslighting bullshit imho.
There is no logical rational reason to use so called racial differences to deprive people of fair treatment, or to fail to try to address past unfairness (for a debatable period) if it still has an impact.
“gaslighting bullshit imho.”
Really?
Please give me an example of “bullshit” in the memo.
“There is no logical rational reason to use so called racial differences to deprive people of fair treatment”.
He made no such argument. If find *your* comment pure bullshit.
i think the main problem with reich’s approach is his ignorance of how race is socially constructed. most people tend to use generic categorizations, rather than statistical ones.
reich seems to be making the really scientifically valid claim that “races” are somewhat proxies (even if crude) of genetic variation, but most people think of race as a generic category (scientific racists think like that too).
“scientific racists”
What is a scientific racist exactly?
Anyway, I hate the term “racist” which I take to mean people who are bigoted or hateful towards other ethnic groups.
But it seems people who postulate that there are genetic trait differences among population clusters (such as intelligence) are also considered racist.
I see no reason why there would not have evolved minor cognitive and character differences between human populations over the past 100,000 years.
There might be such silly differences such as
Italians being genetically more disposed towards extroversion that the Finns.
And I do not want to live it a world devoid of either introverts or extroverts, thus genetic diversity is good!
Just Google ‘scientific racism’ and you will find out all about it.
It was a rhetorical question.
Messy emotional terms like this makes it difficult to have a scientific discussion of race or population differences.
So why do you think it’s messy and emotional?
Here’s another rhetorical question that I’ve thought about a lot: What is the history of using the notion of race, a broad type of categorizing and naming others as most people use the concept, in terms of human prosperity or human suffering?
The term race is obviously messy and for valid historical reasons from slavery to eugenics it has emotional connotations.
I am not sure I follow your rhetorical point.
Do you mean racial prejudice correlates with economic/social disparities?
It is not messy and emotional. There actually was something called ‘scientific racism’ and historically it justified racism by means of (bad) science.
” There actually was something called ‘scientific racism’ ”
I am very aware of that, and because of the abuse of science historically it makes it very difficult to have an open and objective discussion of human group differences – does it not??
“Geneticists should work in collaboration with their social science and humanities colleagues to make certain that their biomedical discoveries make a positive difference in health care, including the care of those studied.”
Unauthorised discoveries will not be permitted.
Makes me wonder just how much of an oxymoron ‘social science’ is.
cr
This subject has resulted in a number of blog posts, starting from an unfortunate interview that Sam Harris,
https://samharris.org/podcasts/forbidden-knowledge/
had with Charles Murray, co-author of the infamous book, The Bell Curve.
I say unfortunate because Harris was far too kind to Murray, whose ultraconservative agenda colored his observations and conclusions. In particular, given that Murray had no evident expertise in the subject and evidently depended heavily on his co-author, Richard Hearnstein, who himself had a somewhat checkered history on the subject. Hearnstein had published a number of articles on the subject in the dubious “journal” Mankind Quarterly.
A few of the relevant links include:
https://pandasthumb.org/archives/2018/03/general-intelligence.html#more
https://freethoughtblogs.com/pharyngula/2018/03/30/science-and-reason-tell-us-that-everyone-outside-massachusetts-and-california-are-genetically-flawed/#comments
https://freethoughtblogs.com/pervertjustice/2018/03/29/destabilizing-the-genetics-of-g/
“To insist that no meaningful average differences among human populations are possible is harmful.”
My position is that modern geneticists define haplogroups based on non-coding segments of both the mtDNA and the Y-chromosome.
By definition, non-coding segments of these genomes are not biologically meaningful.
In most human populations there are multiple haplogroups because of sustained genetic mixing for millennia, a result of migration.
Luigi Luca Cavalli-Sforza, Paolo Menozzi, Alberto Piazza, The History and Geography of Human Genes.
The notion that human populations have been isolated for many thousands of years is not true for Eurasia and the islands of the Pacific. In Africa the great Bantu expansion began around 1000 BC and continued in at least two waves until modern times.
The timing of these and other migrations were proposed based on linguistic studies.
Isolation was true for the Americas in the sense that the continents were isolated, but within these continents migrations occurred. Both historical, archaeological and ethnological studies have documented the rise and fall of empires.
Canada is a good example. My mtDNA is Irish;. My Y-DNA is a rare form of Atlantic modal haplotype (AMH) now confined to Western Isles of Scotland. But my mother’s parents were Quebec Metis, what used to be called half-breeds. They moved to Ontario to escape discrimination and in Ontario passed as Europeans.
I do not doubt that “There are clear genetic contributors to many traits, including behavior.”
But in my opinion, “human populations” is a cop out for someone who is arguing that “race” exists in some biologically meaningful way.
I have not seen evidence to show a correlation between the haplogroups (proxy for race) and genetic traits, including behaviour, though I would be willing to change my opinion based on evidence.
To convince me the evidence would have to demonstrate more fundamental differences than cosmetic differences, such as the conformation of the eyelids or the shape of the nose.
I find one sticking point in the fact that the haplogroups are defined based on parts of the genomes that vary more or less randomly without being expressed the way genes are expressed. (Genes code for enzymes.)
What is needed to support the claim that races are biologically significant is to show that genes vary among human populations, not just the so-called “junk” DNA.
But we know that mtDNA hardly differs among human populations, except for a small percentage of specific base pairs that are not subject to selection. The Y-chromosome too does not differ by much, and again the base pairs that differ are not subject to selection.
When we do see differences that are biologically significant, almost all are pathological and hence subject to selection. The populations most affected seem to be inbred sub-populations among migrants that demonstrate the “founder effect”.
So I postulate (1) that race may be biological, but if so, we cannot know what the biological effects are because of the technology we use to define haplogroups as a proxy for race.
I postulate (2) that genes cannot vary much because most mutations are either lethal or pathological and thus subject to selection.
As for behavioural traits, I have lived and worked in 17 countries, 15 since 1970, and have noted differences that could be attributed to culture, but never a behavioural trait that I could plausibly attribute to a difference in haplotype (proxy for race).
What I find stunning is the opposite. Once you learn the language and customs of the people you associate with every day, you realize that humans have more in common with Pan paniscus than Pan troglodytes.
Well not all humans, but over 90% I would say, worldwide.