At what level is racialism accepted in the scientific community?

Not a pertinent cite about genetics or the mechanics of the genes that they usually do not point or identify.

This is why you are really the one that has not paid attention. The makers of the surveys are an old darling of the pseudoscience racists.

As the cite you made points out: “So, when finding out about real opinions, anonymity is required. Rindermann, Coyle and Becker have replicated the last survey on experts done 30 years ago”

And your group before used a similar ruse:

http://boards.straightdope.com/sdmb/showpost.php?p=13693853&postcount=501

As I said that is just horse puckey, the survey did not include anthropologists or biologists, and few geneticists. Whoever did those surveys are playing loose and fast with what an expert in the matter is, as it was in reality a grab bag of diverse fields that they surveyed.

This is using a similar tactic as the survey the global warming deniers did showing that 32000 experts do not see that humans are causing the current warming that is being observed:

The tactic is to avoid surveying the experts that count so as to get the results the pseudo scientists want.

Same old same old.

Again, the issues remain, a very low reply rate, and the people consulted are indeed not looking at the whole of the evidence, as pointed by other examples like the other survey from meteorologists that pondered about climate science those levels of misunderstanding of what other more pertinent experts conclude is not extraordinary. Plain ignorance explains it, your claim about them looking at the totality of the science is silly since there is no science linked at in the survey you pointed out, what remains just the opinion of intelligence experts that should know better.

I’m also interested in this. When you say “self-identified African Americans with higher levels of European ancestry score higher on IQ tests,” I expect peer-reviewed published journal articles showing this.

I read one of your “cites” and it not only wasn’t it a published journal article, it would never be accepted as a journal article because of the glaring errors that were immediately evident on even a cursory examination.

There’s ongoing inquiry into this phenomenon:

http://coglab.wjh.harvard.edu/~cfc/Chabris2012a-FalsePositivesGenesIQ.pdf

The speculation being that the false positives were caused by environmental effects, particularly epigenetic changes.

AFAICT, the racist scum assessment comes from a few moves made by the “racialists”:

(1) Defense of the arbitrary separation of people into traditional racial categories instead of more coherent populations. There is no good defense for using Caucasians, Asians, and Africans as genetic populations.

(2) Using weak circumstantial inferences to reach traditional racist conclusions, when if anything the burden on people reaching those conclusions should be heightened rather than lowered.

(3) Denial of the history of white supremacy in the United States as way to minimize the likelihood that so-called “nurture” factors explain outcome disparities here.

I don’t think anyone is saying that it is inherently un-P.C. to scientifically conclude that a gene linked to intelligence is less prevalent in a given population. Instead, the objection is that this isn’t what the racialists are doing in any coherent way, and appear instead to be doing pseudoscience to serve racist agenda.

At most, (2) could be applied to this epigenetic hypothesis, but I think no one is arguing that we really know anything about how epigenetics works. The notion is more that the hypothetical possibility of this epigenetic mechanism casts even further doubt on notion that we can pretty well guess–based on current evidence–that intelligence is going be the subject of genes with different frequencies across large populations.

Epigenetics to explain variations between monozygotic twins, maybe, where the DNA gene pools are so similar.

It’s not a reasonable explanation at large scale for populations unless you have some credible driver for it.

Your third cite (bottom of P7) suggests the problem is that a given variant may have only a tiny effect, and therefore very large scale studies are needed for each candidate gene to get rid of false positives.

Interestingly, pre-Darwinian racial theories (e.g., those of Buffon) were often proto-epigenetic. Whatever the case, epigenetics can’t account for much cognitive ability variance within populations, so it’s very unlikely that it account for much between. Just do the math. A simplified model:

total variance = additive genetic + non additive genetic + shared environmental + unshared environment and error + epigenetic + other e…g, G x E

total variance = 100
additive genetic = lower bounds = 40 to 50 (age depending) as estimated by GCTA
additive variance not captured by common variants (and not picked up by GTCA) = ?
non additive genetic = at least 10 (age depending) (epigenetics wouldn’t, from what I have been told, mimic this)
shared environmental = 20 to 10 (some say zero by adulthood, but I can’t believe this) (age depending) based on Adopted sib correlations
unshared environmental + error = 20 based on twin/sibling discordance
Epigenetics = ?
Other = ?

There just isn’t much room for epigenetic influence.

In the pre-GCTA era this would have made a good argument. But GCTA establishes a substantial lower bounds narrow heritability, so there is little explanatory room for epigenetics – at least for IQ.

I don’t know anything about GCTA software, but I’m skeptical of your claims that it establishes hard boundaries for variance caused by different variables. My feeble googling suggests that it is good at estimating the variance associated with different sequence variations. How does it set the lower bounds for genetic heritability?

The thing is that a lot of the “racist scum” are not doing any of the things that you list (and there are other factors as well, FWIW), but are simply being lumped in and associated with the others who are. So the question still remains as to whether epigenetics proponents would be treated similarly.

[Personally I suspect that what might tip the scales is the epigenetics proponents emphasizing it being the result of persecution etc., which would cover them somewhat.]

You know, my first degree was Biochemical Engineering. That was over 20 years ago, of course, and I only worked in the area for a year afterwards. The field has changed a lot in the time since. However, I’ve tried to keep up with population genetics form an interest in the subject. So here is my semi-educated opinion on the matter:

This is the point where I would just put the book down and move on. To the best of my knowledge, this is not just wrong, but frothingly wrong.

Most genetic variety are found in Africa. The greatest genetic outlier among populations is the Khoe-San people. They diverge far more from the rest of humanity than for example Africans and Eskimos do. Or any other two population groups. In fact the genetic divergence between the Khoe and San subgroups is comparable to the difference between Europeans and American Indians.

Then there is Africa. The gentic diversity in Africa is bigger than the rest of the worlds total diversity.

It is like someone wanted to compare genetic variations in dog breeds and decided that a reasonable way to group dog genetic diversity was “Irish Setter” - “English Setter” - “Gordon Setter” - “Irish Red and White Setter” - and “Every other dog plus wolves

It is immediatly obvious that these are not fuctional groups of dog genetic diversity.

If you were to make honest groups of human population based on genetic variety to call “Race”, you’d end up with most of them being African groups. Wades “race” groups is not based on scientific knowledge. They are instead based on the “race” concept of 18th century supremacists and Aryan power groups.

Thats the point at which I’d leave the book behind as a waste of time.

On the raised subject of comparisons between African Americans and caucasians, remember that people exchange genes all the time, and most genes don’t cluster. Slavery ended 150 years ago, at which point african immigration to the US will have declined severely. American blacks are a group defined by the presence of a number of the genes that determine skin color. Whether any other genes have flollwed these genes down the generations is hard to say.

On eipgenetics, remember that the signal from epigenetics will not be the same in an individual as in a population. The Duch famine of 1944, as I remmeber, is where epigenetic effects were first noticed. The signal, in this case reduced height, is not neccessarily noticable in an individual, but will be exceptionally clear over a population.

The authors discuss the method here: Yang, J., Lee, S. H., Goddard, M. E., & Visscher, P. M. (2011). GCTA: a tool for genome-wide complex trait analysis. The American Journal of Human Genetics, 88(1), 76-82; Yang, J., Manolio, T. A., Pasquale, L. R., Boerwinkle, E., Caporaso, N., Cunningham, J. M., … & Visscher, P. M. (2011). Genome partitioning of genetic variation for complex traits using common SNPs. Nature genetics, 43(6), 519-525.

Here’s a pity discussion: “In GCTA, researchers use DNA data for hundreds of thousands of SNPs from unrelated individuals to estimate genetic influence on a particular trait. Unlike the hypothesis-testing approach of GWAS, GCTA does not specify which DNA variants are associated with a measured trait. Instead, it is a parameter-estimation approach that relates similarity in SNPs to phenotypic similarity between pairs of individuals. The use of a large sample, together with pair-by-pair comparisons, allows amplification of the weak signal derived from the low genetic similarity between unrelated subjects. Heritability is estimated as the extent to which genetic similarity can account for phenotypic similarity. (Plomin, R. (2012). Genetics: How intelligence changes with age.)”

The principle is the same as that with kinship studies, you are just indexing genetic similarity with SNPs. This allows you to look at the genetic/phenotypic correlation among random individuals, thus removing the confounds of epigenetics and shared environments.

Black African populations cluster together, meaning that a random Black African will be more genetically similar to another than to a non-Black African when a large number of alleles are looked at, because there were less barriers to interbreeding within Africa than between Africa and out-of-Africa.

http://www.sciencedirect.com/science/article/pii/S0002929710003642

On the other hand, there is higher than average genetic diversity because you began with more diverse populations to start.

Ridiculous. We have a pretty good idea about how West African African Americans are.

If epigenics counts for very little variance within groups, then it’s unlikely to count for much between groups. It could, of course, as could some magical x-factor that only works between populations. But it’s unlikely and would be very difficult if impossible to empirically evaluate.

Let’s see if I understand. So the software looks at a whole bunch of SNPs (not targeting any particular genes necessarily), and then comes up with some assessment of genetic similarity between two individuals based on similarity of SNPs. The more variation in the sampled SNPs, the more the two individuals can be assumed to be genetically distant.

It then looks to some kind of phenotypic similarity–say, thigh circumference–and estimates the extent to which genetic similarity accounts for the phenotypic similarity. I assume it does some fancy math here to compare how well the genetic variation tracks the phenotypic variation.

Is that it, or is there something more to it? Because if that’s it, I don’t think it does what you’re claiming.

When you take representative samples – i.e., don’t over-flavor with certain ethnic population – you get results like figure 7 here: Jombart, T., Devillard, S., & Balloux, F. (2010). Discriminant analysis of principal components: a new method for the analysis of genetically structured populations. BMC genetics, 11(1), 94.

Generally, black African groups cluster together because there was lots of boom-boom in Africa. Of course, if you load the sample with Black African, etc populations this will distort the results. The African American philosopher Q. Spencer explained this point:

Which is why 1st and 2nd generation Black immigrants do so well on the same measures, right?

That’s it. And the claim (repeated, not originated by me) is that it establishes the lower bounds narrow heritability.

That’s interesting because it’s usually said that epigenetics mimics additive genetics, not unshared environment i.e., makes twins similar, not different.

I’m not sure how that is responsive to my point. You can divide world populations up into two groups or twenty groups or two thousand groups. The more groups you have, the more internally genetically similar they are. The notion of “race” is that there are clear shelfs in the continuum of genetic diversity in three (or maybe four or maybe five, depending on what suits the racialist for any particular argument) groupings that are substantially greater in magnitude than any other divides in the continuum of genetic similarity. But, so far, the data does not prove that claim. I don’t see anything suggesting otherwise in figure 7 of that paper or your lengthy quote.

I have no idea whether Black immigrants do better on any given measure than African-Americans, nor why Black immigrants would be a meaningful category for analysis. Were you referring to some study in particular?

I think what you’re suggesting is that since Black people not from the United States are free from white supremacy (before they get here), we should expect them to perform better than African-Americans if white supremacy affects outcomes. But this logic is bad for at least three independent reasons. First, white supremacy is not a uniquely American phenomenon. Perhaps you’ve heard of South Africa? Second, immigrants are not a representative sample. They tend to be either more highly-motivated and successful than their countrymen, or more oppressed and vulnerable than their countrymen. And third, there is no scientific measure of whether someone is “Black” other than self-identification, which makes little or no sense in this context of world populations. Do most Aboriginal Australians identify as Black?

Where is that claim made? It isn’t made in the article you cited, as best I can tell. And it doesn’t make a lot of sense, in any event. To name one of a half-dozen ways it would be wrong, it would ignore any co-correlations between genotype and nurture. To offer a trivial example, I have no doubt that individuals with the genes for testicular development earn higher wages. But unless I also look to how much those genes co-correlate with social phenomena, then I can’t say much about the boundaries of heritability of earning a high wage. It doesn’t avoid this problem to look at overall genetic diversity.

On the contrary, what GCTA seems to be good for is helping to guess whether a trait has a genetic component or not when we haven’t identified a candidate gene. If genetic variation correlates to variation in the trait, it is a good candidate for heritability. But that’s very very different from claiming that GCTA sets a hard boundary for genetic/non-genetic/epigenetic explanations. AFAICT, it does not.

The thing about “race” is that it is an artificial decision about how to cluster. We could make a thousand races if we decided to do so. But as it turns out, because of the out-of-africa migration, what “race” does (assuming we are talking about the typical modern groupings, as for example, black/white/asian) is cluster gene pools into descendant lines from that migration anchor point, and that migration point–along with some others–does drive some differences in gene frequencies.

This idea that the total amount of diversity has anything to do with whether or not gene frequencies vary among races is a totally irrelevant and confused idea.

The hot-button questions around race, such as whether or not observed outcomes are the result of genes and not nurture, hinge on a single contention: gene frequencies driving those outcomes differ by race group.

Consider “greyhound” and “all other dogs” as two groupings. There are a group of genes whose frequency is markedly different in the greyhound group once you decide those are the two groupings you make. It may be that the groupings themselves are totally absurd. It may be that they are biologically reasonable. But if dogs were able to group themselves as people do, they could have a “greyhound” group and an “all other group.” Within those groups, the dogs who identify as “greyhound” would then clump themselves into a pool that has a different frequency of genes driving some outcomes–speed, perhaps; a typical shape, perhaps. The average speed of greyhounds would differ from the average speed of “all other dogs” and the reason would be the average difference in gene pools, even if “all other dogs” had a thousand times the amount of diversity and even if the division itself were felt to be arbitrary.

What drove that gene frequency? Well at some point the greyhound progenitors were isolated from “all other dogs” to an extent long enough to drive gene frequency differences, and the self-identified association with either “greyhound” or “all other dogs” would accurately enough reflect that history for the gene frequencies to drive average outcomes.

It is well-accepted current science that at approximately the L3-M/N split around the time of the (main) out of africa migration 60-ish kya, descendant M/N lines populated most of the rest of the world, and early on (say, 40-ish kya) probably absorbed some archaic lines via Neandertal and Denisovan gene introgression. Those genes are still well-represented in eurasian pools; not so much in sub-saharan pools because gene flow has not been so diffuse that every race is equally homogenized with every other race’s pool of genes.

If you do some reading in population genetics and paleoanthropology, you’ll begin to get a feel for how commonly accepted all of this is. You’ll see very common and very casual references to “african” (sub-saharan) and “non-african” frequencies for genes like archaic line genes; post L3/M-N genes like MCPH1 and so on. You will also see plenty of references to the relative frequency of other genes, such as the fact that 90+ % of sub-saharans carry a malaria resistance gene (DARC) while less than 5% of everyone else does.

Evolution and migration patterns drive gene frequencies; among the divisions for which it is trivially easy to show gene frequency differences is the continental grouping. In particular the continental grouping of sub-saharan africa and eurasian populations is very easy to show (north africa has much more modern gene flow in common with eurasia, but the sahara barrier has been mostly–though not perfectly–intact over long periods of time).

People do exchange genes all the time, but not to the point of homogeneity. In the US, a typical average given for self-identified blacks is close to 20% european ancestry (and blacks in the US over-represent certain west african lines in particular). For self-identified whites, it is probably less than a few percent recent african genes.

Many many genes do cluster by self-identified group; so many that some studies which use self-identification for race group, and microsatellite DNA/SNP clustering for continent of recent origin show almost perfect correlation. Those types of studies are done in an effort to decide whether or not self-identification is a reasonable way to approach race-based medicine (short answer: it’s a crude proxy, but individual genomics will shortly supplant it).

I hope this helps. It’s a common misperception that the argument about diversity has anything to do with an argument about whether or not “races” have average differences driven by genes. And ultimately the sensitivity about “scientific racialism” is more or less about that second question: do we see different race-grouped outcomes because of the effect of nurture alone, or because gene differences play a major role?

Diversity has absolutely nothing to do with that question. If I took a group of very very tall people drawn from all peoples, the outcome difference for height between them and everyone else would be genetic, on average, even if both groups otherwise had huge–or completely different–amounts of genetic diversity for everything else.

I hope this helps you think through this.

Self-identification into races does create a bifurcation of outcomes in two commonly observed areas: sports and quantified tests such as psychometric evaluations or academic tests.

Current data shows no effect on eliminating that gap by equalizing socioeconomic status, and the official position of most scientists and position papers (see, for example, the position paper on race and IQ from the American Psychological Association) is that while there is no evidence for genetics, the gap remains a “mystery.”

The argument for substantial race-based outcome differences because of genes has been discussed ad nauseum on this board, but I think it would be safe to say that the magnitude of those black/white/asian gaps is substantial, and given the failure of socioeconomic status or educational opportunity to eliminate the gap, is advanced by non-egalitarians as indirect evidence that the “shelves” for genetic continuum are there at a race level, even if any two populations are trivially clinal.

I can’t speak for anyone else, but it was very helpful to me after several posts where I was losing track and getting glassy eyed. Thanks.

mispost