Australopitenico wrote: »Australopitenico wrote: »
Hyeah. Microsatellites. So the special genetic markers that we use to recognize ancestry correlate well with perceived ancestry. Stop the fucking presses. That is very, very different from "race".
But microsatellites are not the only markers that correlate with ancestry, are they? This paper, as far as I understand it, shows differences in genes which, you know, actually do stuff: http://www.biomedcentral.com/1471-2148/11/16 Which again, you would expect. At the very least the phenotypic differences that define each "race" (skin color, hair type, prevalence of epicanthic folds, whatever) will have underlying genetic correlates. There may be other genotypic and phenotypic differences as well. Yes, the extent to which we group people in races is arbitrary, but it's no more arbitrary than any other formulation and maps largely onto broad regions of geographic origin in the distant past.Australopitenico wrote: »As for the article I told you:I believe it is inaccurate to refer to African Americans as a race or racial group (much as it is similarly inappropriate to refer to Latinos that way) -- unless you move away from the more classical definitions of race. We try to use the term race/ethnicity. There has been a lot of debate about whether genetic variation in the human population is continuous or discrete. From my view, it is both. This is what makes it challenging to create categories.I tend to think that race has been used more in terms of continental origins (Africa, East Asia, Europe, Americas). On that basis, one would not characterize African Americans as a racial group, but rather as an ethnic group. We sort of implied this in the Genome Biology paper. The reason is that African Americans typically have European as well as African ancestry (and possibly other ancestries as well) and are also culturally distinct from Africans. Sort of similar to Latinos - who from a genetic ancestry standpoint can be nearly anything. Hence our use of race/ethnicity.
And that's the thing, which is what you did with the microsatellites before. Science is not using race at all the way society understands it. Science is using genetic clusters and ancestral populations that define broad geographical areas. This is totally different to what we understand as "race". And you can't go around saying "guys race is totally a thing" without mentioning the caveat, or people will understand race as people define race, which is skin color, and since skin color is both polygenic and transmitted more or less independently from other genes, you can't look at a black or latino guy and say "his genetic makeup is X" because there's no way to know.
Just because Latinos are not a race, or African-Americans are not only one race does not mean race does not exist. I suppose you could argue there's no reason to privilege the racial level of classification over any other, but it is at least as valid a construction as any other one you want to look at, no? Risch himself uses race, as part of the descriptor race/ethnicity,Australopitenico wrote: »By the way that entire interview is the interviewer desperately trying to make Risch say that race is totally a thing while Risch tells him race is not a thing at all. It's pretty funny.
I read this completely differently from you so that's interesting.
This is from the article you yourself linked:Definitions can indeed be "clunky." I would use the phrase race/ethnicity rather than just race because in common parlance it is a better description. I tend to think that race has been used more in terms of continental origins (Africa, East Asia, Europe, Americas). On that basis, one would not characterize African Americans as a racial group, but rather as an ethnic group. We sort of implied this in the Genome Biology paper. The reason is that African Americans typically have European as well as African ancestry (and possibly other ancestries as well) and are also culturally distinct from Africans. Sort of similar to Latinos - who from a genetic ancestry standpoint can be nearly anything. Hence our use of race/ethnicity.
As you can see, despite both you and the interviewer trying and trying to make it seem like he approves and uses the concept of "race" as a biological term, he actually doesn't, he uses "race/ethnicity" which are NOT biological concepts.
Australopitenico wrote: »Also, you recognize that Latinos are not a race, and of course they are not, because you are trying to place a common label on people that only have in common some Native South American ancestry. Stupidly enough I, as an European Spaniard, have been classified as "Hispanic/Latino" when passing the US border. Which biologically is bonkers.
Ta-Nehisi has used an imagine of Walter White, the first African American head of the NAACP, to illustrate the pliability of the black identity. It certainly shows that there are no fixed definitions of race which are particularly useful. But that is a misconception of biological science, which is rife with exceptions and boundary conditions, and characterized by an instrumental perspective. The data above suggests that self-identified African Americans are characterized by some African ancestry, but over 90% are more than 50% African in ancestry. Walter White, who had five black great great great grandparents and 27 white ones, was almost certainly less than 20% African in ancestry. There are such people even today, but they are not typical, and do not disprove the reality that African Americans are predominantly of African ancestry.
Australopitenico wrote: »Well, the same thing applies to black people. There is a full continent with its own genetic history full of black people. As i said before, a black North-African is more similar genetically to a white North-African than to a black South African. Because skin color and other morphological characteristics are highly variable and not necessarily linked to the underlying genetic makeup. Also you have black people in Australia, in several islands and in India.
Asian is the same. What the fuck is "Asian". A guy with slanted eyes? Cool, we have another continent full of them with incredible genetic variability, so Asian is useless too.
White is the same thing. Useless.
So I hope now we have clarified that the usual concept of race is useless when we are talking about genetic studies.
The paper you mention did not study races. It has not made any data collecting across races. It has collected data from 4 very specific populations, each of these populations is from a different continent. And it has found that if you take 3 populations that are very separated geographically you can find genetic differences. Well no shit. It still does not show that those differences are in any way related to the phenotype.
Again, race is useless. Population is not, but population and race are totally different things. Right now black and white Americans are single population, as a group of people that are constantly exchanging genes, while the gene flow between black Americans and Africans is much, much smaller.The difference between race and race/ethnicity strikes me as a pretty pedantic one. Isn't race just another word for groups with similar ancestral geographic origin (on a continental scale)? How is that an invalid level of grouping? Risch uses it himself, he just isn't calling it that (he calls it european and african ancestry, instead of just saying white and black).
Grouping things at a continental level is meaningless biologically. And there is plenty of white people in North Africa.Furthermore IQ and personality traits have pretty high heritability, right?
Nope. No proof of that, we have no idea how much of intelligence is genetic and how much is due to culture/upbringing/social class.
Well, we DO know that IQ correlates amazingly well with social class, so there's that.
Posts
And yes. It's pretty safe to assume that a black American has African ancestry.
But from this study linked by your article: http://genomebiology.com/2009/10/12/R141
What this means is that African-Americans are actually very genetically diverse, having mixed markers from a lot of different African populations. They still show a good amount of African ancestry because until very recently races in the US didn't mix.
This study is actually meant as a "fuck you" to studies that only use Y cromosome or mitochondrial markers and overstate African ancestry, that's why they are doing a full-genome, comprehensive study.
So, I'm going to summarize my points:
-Classifying all Africans as the same race is stupid because Africans can belong to a lot of different populations.
-Black Americans are actually anything but genetically uniform, to the surprise of no one.
-Also unsurprisingly, black people tend to have some sort of African ancestry, because black skin is a very common trait in Africa.
-Yes, some genes are more common amongst specific POPULATIONS (not races).
-This says nothing about complex, polygenic, and not totally genetically-defined traits, such as intelligence.
-If me, as a biologist, classified all frogs on the same continent as the same population I would be kicked out of academy forever.
Edit: And lastly, "black", "latino", "white" and "asian" are not even ethnicities. Chinese and Japanese are not ethnically similar. Mexican and Brazilian are not, either.
Yes. All of that has nothing to do with my point, which was that race ~ ancestral geographic origin != ethnicity. wrt racial mixing just because you can mix red and blue to make purple doesn't mean red and blue don't exist.
"Classifying all frogs as the same order is stupid because frogs can belong to a lot of different species."
Except all frogs can be classified as the same order, anura.
Likewise, Africans all have something in common, they have ancestors of a common ancestral geographic origin. I am not saying Africans are frogs or comparing Africans to frogs. Of course, an order is way way more general than a race, which is more like a sub-sub-species in a taxonomic sense, and different races can and do (obviously) interbreed so the border between races is continuous and not discrete. Still (broadly) a valid concept.
Speaking of taxnonomy, here's Jerry Coyne on whether human races exist: http://whyevolutionistrue.wordpress.com/2012/02/28/are-there-human-races/
spoiler alert: his answer is yes.
I don't necessarily agree with everything he's written there, but I think the concept of race is a valid, if slippery one in modern biology. I quoted Richard Dawkins earlier, and I can go on quoting people, if only to assure you that this isn't just racist cranks saying this stuff. It's pretty mainstream biology, even if you don't subscribe to it.
Sure. But black skin isn't the only thing we use to define a race, and that was Edwards' whole point. I have an uncle who is an albino; my skin is darker than his, but he's more black-African than I am. These albino women are clearly still of the "black race" (so to speak) despite the color of their skin.
http://www.npr.org/2012/11/30/165643518/tanzanias-albinos-face-constant-threat-of-attack
Or see the negritos, who despite the name and their skin color, aren't black people either: http://www.nytimes.com/2005/03/14/opinion/14leroi.html?_r=0
In fact, I'm going to quote liberally from that op-ed since it demonstrates what my point has been all along:
I note here this was in The New York Times more than 8 years ago. The NYT is hardly a bastion of racial insensitivity, and the concept of a genetic race is hardly--to borrow a phrase from our progenitor thread--phrenology 2.0.
I mean we can get into the race and intelligence question if you like. Here's what I believe: I think it entirely possible there are genetic differences between different populations (read: races) in genes that, in normal, healthy individuals (i.e. not people with PKU or on the opposite end Gaucher disease), have an effect on the development of intelligence. If such differences exist, they're likely to be very, very small and dwarfed in effect by environmental factors. Crudely: I believe personality and intelligence have substantial genetic components, but I see no convincing evidence that the IQ differences between different races/nationalities/whoevers is genetically based. That said, I think the etiology of racial IQ differences is impossible to actually know at this moment. To quote Risch again:
:^:
Latino is an ethnicity; Mexican, Brazilian, Chinese, and Japansese are nationalities (that may be coeval with certain cultures); and black, white and asian are races, which are not the same as ethnicities, but are roughly similar to populations which share ancestral geographic origins. The concept of race doesn't make ethnicity, culture, nationality, tribe, and other kinds and levels of classification of humans into groups go away (and vice versa).
Also, IQ is basically just Phrenology 2.0. It's garbage. It measures nothing except how accustomed a particular person is to written tests in general and IQ tests in particular.
Race is a physical description of a group of people, ethnicity is a cultural description of a group of people. Both are never going to be 100% accurate as they attempt to compartmentalise the great human mixing pot continuum.
If everything goes well, we won't have discovered anything new, we won't have a substantial insight that radically changes the way we (don') perceive race and/or genetics.
But this can get ugly. Fast.
If someone is under the impression that X is a race, then it probably is, insofar as "race" is a totally arbitrary social construct to group large swathes of people together on an incredibly superficial level.
Also, change the thread title to:
Race and genetics: solved.
If people could easily separate their conceptions of abstract categories, useful for modeling but not actually accurately reflecting the world, from the actual concrete reality of the world, we'd have so many fewer arguments.
Certainly understanding the underlying genetics behind, say, skin color or hair texture is important. Genetic control of phenotypes is kind of my thing, so I am pretty down with that.
What I want to know is what sorts of information, what questions, are we really asking when we try and find genetic markers that segregate between perceived races?
What is the scientific issue we are addressing?
It usually comes back to "we want to find what portion of qualitative traits such as intelligence or violent behaviors can be explained due to genetic effects," with the underlying insinuation that if we find the genetics we can ignore the environmental component.
Like I said, this is where I got really upset with Razib's blog posts in the past. He boiled down to saying "If you control for environment, intelligence is heritable!" which is...
not really saying much, at the end of the day.
"If you remove all of the other factors that govern this trait, it is governed by genetics!" doesn't actually inform us of anything useful in regards to race, racial issues, or even biology and furthermore it gives too much of an easy out for racism to come back.
Ignoring the context to make your point is a constant theme in issues of racism and sexism, which is why I get so annoyed when that is what biologists do, in the name of "science".
A biologist can draw broad categories and those categories can be used appropriately within scientific literature.
But those biological categories align poorly with the social categories of race. The former can be disciplined, well-defined, rigorous, nuanced, with necessary exceptions and divisions. The latter rarely are; and as Coates points out, the criteria that we assign people to races (because it is an assignment) shifts over time - not in response to improved scientific knowledge, but in response to politics and economics.
the "no true scotch man" fallacy.
That is, I think, a very unfair and very inaccurate view of I.Q. I think that's (yet another) thread though.
--
wrt whether we need to have this thread--or talk in this thread about whether we need to have this thread: it was definitely becoming an offtopic subthread of its own somewhere else, so I split it off into its own thread in the interests of politeness and clarity. If you guys want to narrow the debate, I can make it clear in the OP that this thread is for the discussion of Lewontin's fallacy. Put another way: is use of the term race taxonomically valid in modern biology? It's a rather specific, academic question. The debate was perfectly polite, specific, and informative earlier, and I see no reason why that shouldn't continue to be the case going forwards.
--
@Arch
glad you found it, Arch! I agree that often the motivation for such studies is questionable at best. On the other hand I'm uncomfortable saying let's just not research this; especially for studies that, unlike say the Tuskegee experiments or the Japanese Unit 731 stuff, have no clear moral valence one way or the other. I don't think investigating racial differences in X or Y, or stumbling upon them, is an evil thing. If that's your balliwick, more power to you! I understand this is an area positively infested with what I would call murky motivations at best; but I also think there's a lot of potentially interesting research that can't be done (and discussions that can't be had) because of a fear of stepping on landmines (see the trepidation about this thread).
As I said in the last thread :
the "no true scotch man" fallacy.
Do we just mean geographically distinct populations that we can identify?
If so, how is that different than calling biological races "species" or "subspecies"?
It seems, from the data I've seen, that humans do cluster geographically. Does this map onto what we mean by "race"? And how are we defining race? What trait does race correlate with?
Skin color?
It just doesn't make sense, and most discussions about this tend to miss the forest for the trees. It is trying to conflate a social concept with a biological one, and being loose with terms.
A biologist talking about "human races" but meaning "genetically distinct geographic populations" is very different than a sociologist saying "there is a racial divide in IQ in America" which is very different than someone saying something like "I think Asians are all bad drivers".
I just keep coming at this thing with more and more questions about all the underlying assumptions, which makes me incredibly skeptical of the whole discussion, even in the scientific literature.
Registered just for the Mass Effect threads | Steam: click ^^^ | Origin: curlyhairedboy
okay, so what sorts of questions do you want to see "race and genetics" research address?
Simply saying "there are good questions here", isn't actually an answer to the claim that it doesn't actually propose any new or useful knowledge (which I admit, may be a spurious claim. I'm starting strong and working down).
I have a great paper for this, because gatdam I agree with it so hard.
The most important bits are Table 1 in box 3, and the quotes on the pages around it.
And the big one
I would extend this quote even further and say that given how little predictive power personal genomes currently possess, it makes me even more skeptical of any bit about "the genetics of race".
To pull this back to the initial discussion- that is, are there genetic differences in races that lead to intelligence differences- I have been pouring through the literature and the best answer I can come up with is "inconsistent results that fall around 50% heritability for intelligence, that is modulated by age, social status, nutrition level, and personal learning environment."
One of the more interesting studies showed an increase in heritability with age- a phenomenon they suggest is due to a bastardization of niche partitioning and assortative mating.
That is, people who inherit higher levels of intelligence will self-segregate into areas that further increase their intelligence and cause them to associate more with people who have similar levels of intelligence, thus increasing the overall heritability of the trait....which speaks something to the issue of cultural determination of intelligence more so than anything else, especially when combined with studies that show as socioeconomic levels fall, IQ is predicted more by environment, but as socioeconomic levels rise, IQ is predicted more so by genetic factors.
To tie this all back together- the biological idea of race when applied to the human categories of race (african-american, asian, caucasian, etc) are basically meaningless. When applied to geographic groups (North Americans, Japanese) it works better, but not perfect (see in the paper above, Craig Venter's genome being more similar to a Korean's genome than to Watson's genome). When applied to tricky concepts like intelligence, which are largely polygenic and pleoitropic and are generally modified by the environment, you see interesting patterns that only correlate with race due to contextual features in the region analyzed.
That is, blacks in america may have lower IQ (unconfirmed), and this has a heritable component, but that heritability is modulated by and depends on external environmental factors...which is why it is stupid to try and control for E in GxE interactions of traits, especially traits that have a lot of social and cultural baggage.
I've always been partial to the counter that it is still a useful concept because as a society at large it is treated that way. Yeah if you looking at narrow medical questions like "side effects of drug X" it's less useful or maybe not useful at all. But if I'm looking at something like "diabetes outcomes" then it becomes real useful real fast.
Huge amounts of discrimination cases hinge on disparate impact. Even simple stuff like 'must be clean shaven'
Which is, hey, essentially bigotry.
A lot of studies, thankfully, are moving away from IQ as a measure of intelligence, and finding less heritability or the same amount. The studies I talked about above (citations incoming) mostly ignored IQ.
Contrast that body of work to the work done on actual inherited mental disorders like Downs Syndrome; we know more or less exactly how Downs Syndrome happens, why it is passed along and what proteins are responsible for the condition. Nothing of that level of rigor exists within any biological study that purports to demonstrate inherited intelligence between previously isolated populations of humans, and i suspect it doesn't exist because no such mechanism exists, because any such links are imaginary.
* - general intelligence tests have fallen out of vogue, because "intelligence" hasn't been seen as a single variable since the 1940s. Rather, the dominant paradigm is that what we lazily call "intelligence" is an assortment of different cognitive abilities and characteristics, where some characteristics may be more or less relevant depending on context. Example: ability to remember seven numbers in short-term memory is more important in a classroom, ability to count a shifting group of animals quickly is more important on a ranch. Consequently, it's hard to find rigorous articles that refer to "intelligence" these days unless they're very broad lit reviews - they're more looking at performance on specific tasks, or perhaps a battery of related tasks, like remembering a sequence of numbers (digit span) or sorting cards by shifting criteria (set shifting/card sorting). The intelligence tests that remain, like Wechsler and Woodcock, tend to be expressed as multiple numbers. Eg, your "Wechsler score" matters less than your "Wechsler verbal comprehension score."
Anyboo, the ones that get published don't leap to any genetic conclusions - they might discuss genetics, but only in the context of other factors as well. Example: http://www.tandfonline.com/doi/abs/10.1076/clin.13.3.376.1735#.UnrJN_nBOSp
the "no true scotch man" fallacy.
because that leads to questions like, "Why did you include those tasks and not these other tasks?" and "how is task X weighted relative to task Y?"
the "no true scotch man" fallacy.
PSN:Furlion
The dominant paradigm in psych right now with pretty much all matters of human behavior (including intelligence) is that there are complex interactions between heredity and environment. It hasn't been nature vs. nurture for decades, it's now nature and nurture dancing in a lively cotillion.
So while, yes, I do see where you derive that from the quoted text, it is still the case that heredity matters very very much.
(Also, heredity =/= genes but that is just another big can of worms seriously worms everywhere)
the "no true scotch man" fallacy.
Depending on the region of their ancestry, a given person who self-identifies as Hispanic may genetically more closely resemble Caucasians, African-Americans, or Native Americans.
I mean, if you want to make a Razib-esque argument that races are totally a thing because you can cluster people genetically in categories that match up with racial categories
well
only if you acknowledge that much of latin america is a melting pot of admixture
the "no true scotch man" fallacy.
Okay, if genetics really is an important factor, then can you tell me which proteins are responsible for carrying 'intelligence' (if we accept a broader, Western-biased idea of 'intelligence', and yes of course there are more worms to be found here) from parents to offspring? Because I suspect that you can't.
On the flip side, to go back to Down's Syndrome, we know exactly what proteins are involved: an unfortunate redundant copy of chromosome 21. I suspect that anyone with specialized training would be able to go into further detail, but I'm hardly that person (and I'm sure I'm misusing the term 'protein', but you know what I mean).
All of the studies in the literature point to interesting-sounding correlative relationships between Population Group Being Studied and Particular Intelligence Trait, and while I don't necessarily doubt the existence of the correlation, the fact that no mechanism has been proposed suggests something that either is not genetically transmitted or has so little genetic impetus that it's not worth considering.
It is worth pointing out that Down's Syndrome is not in fact isolated to a complete duplication of 21. In fact the phenotypes can vary greatly depending on what part of the chromosome is duplicated. Also the gold standard for genetics and intelligence is probably Fragile X syndrome. Oddly enough we can tell you several genes which will negatively impact intelligence if they are interrupted, mutated, missing, or copied. Just none that actually boost it.
PSN:Furlion
Which was my point lol. We have no idea what makes you smart but we are getting better everyday at figuring out what can make you dumb.
I was not thinking translation but rather chromosomal breakage. I do not remember the statistics but I do remember the center where I worked on my phd had several patients without full triploidies. But yeah definitely a lot rarer.
PSN:Furlion
My point is that the genetic elements that give rise to severe mental retardation are most likely not the genetic elements responsible for intelligence. The loss of cognitive function stems from severe deficits in development that do not occur to the "normal" human and as such cannot be ascribed as regulators or rheostats for intelligence.
I don't accept a broader concept of general intelligence; I don't think that intelligence can be rigorously defined in a way that is not both contextual and domain-specific. (There is a measurable correlation across domains - the general intelligence factor - but the degree to which the general factor is amplified by cultural presumptions inherent to western intelligence testing is a huge unanswered question. And the general factor isn't a score, like IQ, that can be measured in an individual; it's a correlation observed across different tests.)
In any case, that's part of the conundrum. Twin studies; adoption studies; and population studies controlling for socioeconomic factors and education; have showed partial heritability for various cognitive performance tasks. However, nobody's been able to suggest direct etiologies for these observations - something is inherited but nobody seems to know what that something is (more likely plural - what those somethings are).
the "no true scotch man" fallacy.
Haha I did not even notice my phone putting translation in place of translocation.
PSN:Furlion
There are loads of genetic mutations disturbing normal neurologic development that only cut slightly into intelligence. That's why we have mild, moderate, severe, and profound levels of mental retardation.
Doc: That's right, twenty five years into the future. I've always dreamed on seeing the future, looking beyond my years, seeing the progress of mankind. I'll also be able to see who wins the next twenty-five world series.