There are several aspects of this controversy that interest me greatly. One of them has to do with the reputation of science writers. Nicholas Wade was a science writer for Nature (1967-1971), Science (1972-1982), and the New York Times (1982-2012). I've often heard heard him being referred to as one of the best science writers, particularly by other science writers. That's an opinion that I've never shared and I'm glad to see him get his comeuppance.
Science writers aren't doing so well these days.
The chief reasons why so many people are loath to admit the genetic variability of socially and culturally significant traits are two. First, human equality is stubbornly confused with identity, and diversity with inequality, as though to be entitled to an equality of opportunity, people would have to be identical twins. Human diversity is not incompatible with equality. Secondly, it is futile to look for one-to-one correspondence between cultural forms and genetic traits. Cultural forms are not determined by genes, but their emergence and maintenance are made possible by the genetically conditioned human diversityThe issue of human races came up in my local newspaper in a context that's unrelated to the debate over Wade's book. A biologist at Ryerson University published a letter in the Toronto Star today. Here's a link to the letter: The only race is the human race. It was written by Mario C. Estable, Associate Professor, Department of Biology, Ryerson University. (Ryerson University is in downtown Toronto). Mario Estable is a biochemist/molecular biologist. He wrote ...
The word "race" was invented by Linnaeus where he categorized people in much the same manner he categorized other lifeforms. The assumption was that people from different geographical regions were of differences so great that they were of different "races." By race, he meant a separate group of interbreeding individuals of the same species that had been separated from all others for millions of years, diverging from the other groups.I understand that there can be legitimate debate over the existence of biological races in the species Homo sapies. But you don't contribute to that debate by saying things that aren't true. If you are a biologist, and a professor, you have a responsibility to make sure that what you are saying is correct.
Now, fast forward to modern times. It’s well known that all of the Homo sapiens currently alive are recent descendants of a group of individuals, fewer than 200, that literally walked out of Africa into the Middle East and Europe and then the rest of the world, a mere 60,000 years ago.
This means that human diversity on earth can’t have resulted from millions of years of separation of some groups from other humans. In other words, races do not exist for Homo sapiens. We are all very genetically related. Period!
So statements like "biracial" or "race" are simply "racist" because they perpetuate the ignorant division of humans by superficial meaningless traits. Races don’t exist, so stop referring to people as belonging to one race or another — or two races (biracial). There is only one race, the human race.
In this case, it's not correct. It is simply not true that fewer than 200 individuals are the ancestors of every human living outside of Africa and it's certainly not true that they are the ancestors of every human who remained in Africa. Mario Estable seems to be forgetting about Africans as though they don't count as "all of the Homo sapiens currently alive."
We know for a fact that the genetic makeup of non-Africans includes contributions from people who left Africa long before the latest Out-of-Africa wave and any biologist writing a letter to a newspaper should know this. We don't know if the latest wave left 60,000 years ago or much earlier. We know for a fact that more than 200 people were involved.
Linnaeus did not invent the word "race." It was in common use long before Linnaeus published his classification scheme in 1735. This is easily confirmed by checking the Wikipedia entry on human races. That article would have also alerted any biologist to the legitimate controversy and prevented him/her from propagating untruths.
What do biologists know about human races? Quite a lot, actually. It's just not true that scientists all agree that human races don't exist.
Here's a repeat of the rules I imposed in some previous postings.
Let me sound a note of caution to those who wish to comment. The fact that humans races might be genetically different says absolutely nothing at all about equality and racism. For this thread only, I will delete any comments where the author is confused about this distinction. This is a discussion about science and whether biological races actually exist.My position on this topic is very close to that of Bruce T. Lahn and Lanny Ebenstein as expressed in an opinion piece in Nature (Lahn and Ebenstein, 2009).
The current moral position is a sort of 'biological egalitarianism'. This dominant position emerged in recent decades largely to correct grave historical injustices, including genocide, that were committed with the support of pseudoscientific understandings of group diversity. The racial-hygiene theory promoted by German geneticists Fritz Lenz, Eugene Fischer and others during the Nazi era is one notorious example of such pseudoscience. Biological egalitarianism is the view that no or almost no meaningful genetically based biological differences exist among human groups, with the exception of a few superficial traits such as skin colour. Proponents of this view seem to hope that, by promoting biological sameness, discrimination against groups or individuals will become groundless.
We believe that this position, although well-intentioned, is illogical and even dangerous, as it implies that if significant group diversity were established, discrimination might thereby be justified. We reject this position. Equality of opportunity and respect for human dignity should be humankind's common aspirations, notwithstanding human differences no matter how big or small. We also think that biological egalitarianism may not remain viable in light of the growing body of empirical data (see box).
[Image Credit: The second image is obviously the cover of Scientific American from December 2003. This is one of the most blatant examples of political correctness ever published in a prestigious journal and it's one more example of the decline of Scientific American. It doesn't take much to recognize that the faces on the cover are identical except for skin color. As if that's all there is to human populations.]
Lahn, B.T. and Ebenstein, L. (2009) Let's celebrate human genetic diversity. Nature 461:726-728 [Nature]
Thanks for the Dobzhansky quote. Right on the money.
ReplyDeleteDid Dr Estable really write these things, or were his actual words creatively re-edited by some journalist? As it stands, the quotation suggests that the "stay-at-home" African populations do not belong to Homo sapiens. I'm also surprised to learn that Linnaeus regarded "a race" as "a separate group of interbreeding individuals of the same species that had been separated from all others for millions of years". Millions of years? Linnaeus??
ReplyDeleteI was surprised by the described view of Linnaeus as well, although he did live at a time when geologists were beginning to understand that the earth was older than accounts suggested by Genesis. But millions of years old in Linnaeus' time? I do not know. Linnaeus did dabble in evolutionary thoughts later in life.
DeleteDarwin and his contemporaries generally described the races as being something very similar to separate species -- and so races were then viewed as profoundly different groups separated for a long time. So it seems more likely that Linnaeus would have seen races as profoundly different groups as well, rather than as barely different groups.
Marcoli: No. Darwin was a monogenist and in "Descent of Man" emphasized the similarities between races and that all humans were one species. Darwin said (rightly or wrongly) that all evolutionists would argue that all human races were one species. (Which turned out to be wrong, since later on the evolutionist Haeckel argued for separate species.) Darwin knew that in arguing for one species, he was going up against anti-evolutionists who were more racist and believed in separate species.
DeleteIt was the ANTI-Darwinists who believed human races were separate species (polygenism): in particular, the racist creationist hero Louis Agassiz, his co-author Josiah Knott, George Gliddon, and the Anthropology Society of London. The ASL was founded as a haven for anti-Darwinists and racists.
Both sides compared black people to apes, but the comparisons were much more hysterical and demeaning among the anti-Darwinists.
Agassiz coined the phrase, that the gap between whites and blacks is larger than the gap between blacks and apes-- a phrase copied and pasted for some 90 years after its appearance in the 1854 racist creationist magnum opus, "Types of Mankind." That book, very influential and published 5 years before Darwin's "Origin", was co-authored by creationist hero Agassiz, crackpot Josiah Nott who used to give what he called "lectures on niggerology" to Southern slaveowners, Bible expert George Gliddon, and skull measurement guy Dr. Samuel Morton. Nott's contribution includes crude cartoons, not well drawn, comparing each kind of negro to various species of ape. He also "proved" mulattoes were less fertile, die early and would go extinct (a point later raised by creationist hero Agassiz when he wrote a letter to Pres. Lincoln strongly recommending that the US impose racial segregation after the Civil War, until the happy day that blacks go extinct in the North-- can't tolerate the climate you know.) The alleged infertility of cross-breeds was an argument used by anti-Darwinists against both evolutionary theory and racial intermarriage.
In the 1854 book Agassiz writes a long, detailed introduction in which he employs an "irreducible complexity" argument to prove God created the races separately: ecosystems (he calls them "kingdoms") have many interdependent species and are thus irreducibly complex; therefore God had to create all the species in each ecosystem simultaneously; human races are part of each continents' ecosystem; therefore God created blacks in Africa, Asians in Asia, etc.
In the preface an author looks forward to the extermination of native tribes. Dr. Samuel Morton posthumously contributes a chapter on the smaller brain sizes of non-whites. George Gliddon presents a genealogical table drawn from the Bible.
Racist polygenism was originally a creationist thing, and they perfected it. Anyway in Hitler's Table Talk, ~1941, Hitler gives a paraphrase of Agassiz's "distance to apes" quote. Creationist Richard Weikart, of "From Darwin to Hitler" infamy, presents Hitler's paraphrase of a creationist hero as proof that Darwin caused the Holocaust implemented by Christians.
Nowadays you can see what creationist Ken Ham says about race right here https://answersingenesis.org/racism/are-there-really-different-races/
DeleteI especially love this bit describing the changes after the Tower of Babel incident:
"Because of the new language and geographic barriers, the groups no longer freely mixed with other groups, and the result was a splitting of the gene pool. Different cultures formed, with certain features becoming predominant within each group. The characteristics of each became more and more prominent as new generations of children were born. If we were to travel back in time to Babel, and mix up the people into completely different family groups, then people groups with completely different characteristics might result. For instance, we might find a fair-skinned group with tight, curly dark hair that has blue, almond-shaped eyes. Or a group with very dark skin, blue eyes, and straight brown hair."
I love his idea of mixed-up characteristics in the last two lines. He has no knowledge of genetics!
But I could have sworn that the Creation "Museum" has a display describing how Africans are descendants of Noah's son, Ham, who was cursed. Ken Ham tries to debunk that in this article.
The characteristics of each became more and more prominent as new generations of children were born.
DeleteSounds like evolution to me.
"Cultural forms are not determined by genes, but their emergence and maintenance are made possible by the genetically conditioned human diversity." I know of no evidence whatsoever that cultural forms are maintained by the hereditary diversity of the group. The notion that cultures don't change for certain ethnic groups is near as I can tell always a myth. Dobzhansky explicitly says that there is no one-to-one correspondence between cultural forms and genetics, which is true. But it not at all clear what it could possibly mean to say that the "genetic variability of socially and culturally significant traits" of groups leads to the emergence of different cultural forms other than the implication that some groups have different capacities for some categories of social and cultural traits. I do not think there is any evidence that young of any group has failed to learn the culture of a different group if they were raised in it.
ReplyDeleteLahn and Ebenstein write "Equality of opportunity and respect for human dignity should be humankind's common aspirations, notwithstanding human differences no matter how big or small." Unfortunately equality of opportunity has no widely accepted meaning. It can mean a formal political equality as we in the US had during segregation. I believe there isn't real equality of opportunity today either but denying this is controversial, just as it was under Jim Crow. Historically it is precisely the logical (and desirable for those who benefited!) implication that a racially diverse group should be treated appropriately to its differences that led to the promotion of racial thinking. Contra Lahn and Ebenstein, it was the triumph of experience over ideology that led to the claim that there were no significant differences between African Americans and the rest of the US population. Separate but equal was never equality of opportunity. Lahn and Ebenstein's claim that biological egalitarianism is responsible for justifying discrimination is wildly wrong.
Or to put it another way, human diversity does indeed display differences, not identity. Children for example are wildly different from adult. And there are adults who have wildly different capacities. Not only is logical to treat these categories of people differently, it is widely accepted that we as a society do so. We do not expect children to avail themselves of their equal opportunity to work, nor do we give them the same political rights. It's true that mentally handicapped adults are simply permitted to enjoy their equal opportunity to fit into society but it is arguable that this is not really a good thing. It is simply not sufficient to talk about equality of opportunity. If one protests that the group differences are nothing of this sort, I can only say that is biological egalitarianism!
So, you think there are races. How many are there? What are they? How are they delimited? The fact that nobody can answer these questions (or, rather, that everyone answers them differently) suggests that they do not, objectively, exist. Nobody doubts that there is genetic diversity in humans and that some of that diversity is geographically structured. But that isn't enough. The geographically structured diversity has to come in discrete, diagnosable segments rather than smooth clines. What do you have?
ReplyDeleteJohn, why does the "diversity [have] to come in discrete, diagnosable segments rather than smooth clines"? We don't ask the same of species designations for asexually reproducing life forms, for example; and there are lively discussions about species classifications even among sexually reproducing life forms. For sub-species classifications, as human races would be, "discrete, diagnosable segments" seem (to me, anyway) even less necessary or logical.
DeleteWhat I tend to wonder about race is how useful it is as a way of classification. What is gained in our understanding of people thereby? As with many sub-species classifications, e.g., gender, variation within the classification can be as great or greater than average variation between/among classifications. So what is the use of identifying such variations?
Harshman: "The geographically structured diversity has to come in discrete, diagnosable segments rather than smooth clines."
DeleteExactly. If that's part of the definition of "race" when applied to animals and plants, then it must be part of the definition of "race" when applied to humans.
If "race" when applied to plants and animals generally signifies LARGE, geographically discrete, correlated inherited differences, but "race" when applied to humans refers to SMALL, geographically continuous, correlated inherited differences, then our use of the term "human races" is either inaccurate, tendentious, or adopted for political advantage-- or all of the above.
Let me rephrase that for you:
DeleteSo, you think there are species. How many are there? What are they? How are they delimited? The fact that nobody can answer these questions (or, rather, that everyone answers them differently) suggests that they do not, objectively, exist. Nobody doubts that there is genetic diversity in animals and that some of that diversity is geographically structured. But that isn't enough. The geographically structured diversity has to come in discrete, diagnosable segments rather than smooth clines. What do you have?
Ah, but we can answer that for species, most of the time. Species are of course abstractions of diversity, as are subspecies, but they tend to be clear enough most of the time, and we tend to agree on what they are, most of the time. Human subspecies, or races, not so much.
DeleteWe don't ask the same of species designations for asexually reproducing life forms, for example; and there are lively discussions about species classifications even among sexually reproducing life forms.
DeleteMost asexually reproducing metazoans I know of fall into nice, discrete groups, presumably because of ecological selection around different phenotypes. Other taxa, I don't know so much, and perhaps it makes little sense to talk about "species" in bacteria. Obligate outcrossing, sexually reproducing species are a little simpler, and we have that good old biological species concept to fall back on. It works pretty well in most cases. The major exception regards allopatric populations, in which it can be difficult to tell if we should call them species or subspecies. And that's what much of any controversy is about. If you like the phylogenetic species concept, you generally promote subspecies to species and don't use the subspecies category at all. In birds, at least, continuous, clinal variation between morphotypes is generally grounds for failing to recognize subspecies. Your mileage may vary.
The problem with saying that human races exist is that it is difficult to find a consistent non-arbitrary biological justification for the groupings. Cluster analyses, PCA plots, etc., often give contradictory results. Even the study by Rosenberg, which said there were 5 human groups was subject to an alternative but equally probable interpretation finding 16 groups. I realize much has been done since the 2002 Rosenberg study but I'm just using this one as an example.
ReplyDeleteBut of course the process of denoting "biological differences among groups = races" doesn't actually get at the question of "do races exist" as we've long known that there are differences among human groups (just take a stroll down any large city); the question is more of which differences and how many differences do we use to assign someone to this group versus that group. Hence, we can find all sorts of biological differences--that's trivial--but how we use these differences to erect categorical groupings of humans--well, that's not trivial, and it's also a highly cultural and arbitrary exercise. For example, Lahn/Ebenstein note that there is a pigmentation-associated SNP that "is present in almost all Europeans but is nearly absent in east Asians and Africans." This sounds like a good biological justification for some sort of racial divisions, but Lahn/Ebenstein have a priori reinforced the traditional culturally-defined notions of race in the USA: white, black, asian. Otherwise, why bring up these regions unless one assumes they are meaningful in the first place. (Tendentiously, I note that they could have just used latitude/longitude coordinates to present this result).
What is interesting to me is that people find it far more problematic if scientists say there are 3 or 4 or 5 "races" (a low number); whereas there is far less controversy if they say there are 10, 12, or 15 "races" (a high number). The history of racial studies, it seems in part, is the history of trying to find the lowest number of human groups. A study saying there is 3 versus 32 genetically delineated groups of humans is the difference between publishing in Nature and publishing in Am. J. Hum. Biol.
Some attribution: the finding that Rosenberg's study had an equally likely probability of delineating 5 versus 16 groups was from a chapter by Deborah Bolnick in a book called "race in the genomic age." My second paragraph is largely informed by Jon Marks, who was one of my teachers in grad school and I pretty much read everything that he writes; so I'm largely parroting Jon.
DeleteYour point about the insistence that races are objectively real, vs. disagreement among experts about how to even COUNT the races, was orginally made by Darwin in a marvelous passage in "Descent of Man."
DeleteLast I heard there were 5 major races, where 3 of them would check the box marked 'black' on the US Census.
ReplyDeleteI think all this hand-wringing over whether races exists occurs over concern that some closet-bigot in the US Congress will take some remark by a geneticist, misunderstand and misinterpret it and use it to justify spending less on inner-city schools.
The notion of a 'species' may be imprecise but I dont think 'race' is. The fact that there can be many definitions doesnt invalidate it as a category, especially when you consider the historical and culteral aspect. Its unfortunate that we need to have these conversations on race but the vast majority of the population recognize that race, in some form, exists. When biologists claim that race cant be defined or that it doesnt exist they do nothing more than remove themsevles from the conversation.
Lantog: "the vast majority of the population recognize that race, in some form, exists. "
DeletePeople perceive correlated inherited differences. Yes, dark skin is correlated with kinky hair. But the existence of correlated inherited differences is not enough to prove that human races are real.
Diogenes. Different racial and ethnic groups share a history, language, culture, cuisine etc along with a set of superficial characteristcs that have a genetic basis. It seems to me thats all thats needed to justify the category of race. It doesnt matter if there are varying degrees of overlap with other groups or mixing.
DeleteI'm going to step back and reread some posts. Maybe we're talking past each other because I'm misunderstanding what everyone is saying.
Nick can explain it better than I can. There was a good post on this at Panda's Thumb some while back.
Deletehttp://www.pandasthumb.org/archives/2012/02/continuous-geog.html
DeleteYep, the discreteness of the concept of "race" is the huge block to applying it successfully to humans. To a first approximation, all humans are extremely similar (~99.9% identical) and what genetic differences there are are mostly within-population (85% of the 0.1%) rather than between-population (15% of the 0.1%). To a second approximation, there is some geographic structure in the remaining bits of genetic variability (the 15% of the 0.1%), but it is dominated by a continuous, isolation-by-distance pattern and cannot be portrayed as discrete except by being dumb about (a) spatially discrete genetic sampling and/or (b) inference programs that assume ahead of time that there are a certain number of known or unknown clusters into which the data will be categorized. To a third approximation, it looks like there is a bit of a discrete signal beyond simple linear distance (something like 10% of the 15% of the 0.1%), representing the relative difficulty of crossing water between continents, crossing the Himalayas, etc., but now we are talking about something like 0.0015% of the genetic data. With modern sequencing techniques we can still detect these patterns but this does not thereby confirm traditional notions of race / subspecies, which typically amount to at least a few hundred thousand years of genetic isolation.
ReplyDelete(All IIRC)
Diogenes seconded your remarks so I’ll follow up my discussion with him here. I’ve heard those numbers on within and between-population comparisons and I think they’re irrelevant. If I was more adept at Genebank and BLAST searches I’d do the following: I’d find a locus involved in skin pigmentation and download it and surrounding sequence from a few individuals of Icelandic descent and a few individuals of Nigerian descent. I would bet there would be more differences among the Icelandic individuals and among the Nigerian individuals than between the consensus sequences for Icelandic and Nigerian. But of course phenotypically the 2 populations are at the opposite extremes with no overlap (with a minor exception) This tells us that genetics is completely uninformative on the issue. I would further assert that skin color is a perfectly reasonable criterion to define race since it correlates within recognized groups and differs widely among groups. (height would not be a reasonable criterion)
ReplyDeleteOne could shut the whole conversation down by simply asserting that race is a cultural construct based on superficial traits that have a genetic basis. If biologists claim the category itself has no biological basis (which I don’t think is quite true) they haven’t in any way disputed the concept, they’ve just said they have no qualifications to speak on the subject with authority.
So, redheads are a different race than blondes, then. Glad it's so simple!
DeleteWe all agree you can distinguish Icelanders from Nigerians, just as we all agree you can distinguish red from green. You can in fact distinguish any two widely separated points on a continuum. That's all you have here.
DeleteIf skin color is a marker of race, what do you do with all those intermediates? And are South Indians, Melanesians, and Australians the same race as Nigerians? I don't think this is going to work for you.
Why do you want there to be races?
Lantog: "skin color is a perfectly reasonable criterion to define race since it correlates within recognized groups and differs widely among groups."
DeleteWell that's circular logic then, because you start out with society's definitions of "groups." Who defines a group?
Anyway, "race" for animals and plants is not defined this way, by first asking what the "groups" are, then working backwards to prove your starting premise.
John brings up Melanesians. Historically, the classification of Melanesian vs. Polynesian was a mess. Then there are Negritos, South Indians, Khoisan etc.-- how to you even COUNT the races?
Wait - Estable actually thinks that a group of individuals "literally walked out of Africa into the Middle East and Europe and then the rest of the world"?
ReplyDelete(I'd give him a break but he did go out of his way to put the word "literally" in there.)
I enjoy reading this blog. I'm admittedly a spectator just watching and learning. What I've leaned is Diogenes has somehow turned the comment section into an episode of Hoarders. Would I be out of line by suggesting he start his own blog?
ReplyDeleteDiogenes comments are one of the primary reasons that I read this blog (apart from Larry's own contributions/thoughts).
Deleteyou can skip over the Diogenes comments easily enough. I suspect most regular readers do not.
DeleteThis is a discussion that needs to be better publicized, because the real scientific position seems to be held by neither set of activists ("race egalitarians and race realists", I suppose you could call them). Clearly, humans are highly diverse and not equally related to the same ancestors in recent history. Africans have a higher net contribution from African ancestors... but here's the trouble! "Africans" is not a useful biological label, it's too heterotypic. We can certainly make any random grouping of things that share some commonality (Africans live in Africa) but the Victorian continental race definitions are nearly useless as a homotypic label.
ReplyDeleteIf the scientific use of the term race differentiates "Bantu-Zaire" from "Khoi-San", two groups with vastly different genetic markers, both living in Africa, I'm on board. Any race definition that puts Indians, Han Chinese and Japanese into a single category "Asian" is a worthless definition in terms of genetic studies. You can either have 1 human group or 100+ if we're going to try to use non-arbitrary divisions.
Even here, though, I agree with several other comments pointing out that human population diversity is dominated by clines except at a few natural barriers. Our ancestors were the ones who managed to slip over to the neighboring village or overseas port and prevent genetic stagnation. If my memory of pop gen serves, it doesn't take many such romantic optimists to keep populations from diverging.
If my memory of pop gen serves, it doesn't take many such romantic optimists to keep populations from diverging.
DeleteOne migrant per generation keeps populations from diverging through drift. They might still diverge through selection, though.
"They might still diverge through selection, though."
DeleteAgreed! What human phenotypes have fitness values that would outpace drift in the last 10,000 years (500 generations)? This is an honest question, and one that I fear is often overlooked.
Human evolution continued from the brain up... but what (strongly) genetic behavioral traits are under (strongly) positive or negative selection?
My position would be that the majority of human inter-group differences are drift-dominated or related to nutrition or disease conditions with high selection. More contingency than destiny.
John wrote "One migrant per generation keeps populations from diverging through drift. They might still diverge through selection, though."
DeleteThat's wrong. It is based on the misinterpretation of Gst or Fst as a measure of allelic differentiation. See my article "Gst and its relatives do not measure differentiation", Molecular Ecology (2008) 17: 4015-4026. For further discussion and simulations see my interchange with Nolan Kane at
http://www.molecularecologist.com/2011/03/should-i-use-fst-gst-or-d-2/
Lou Jost
The paper is paywalled, of course (Wiley). The discussion is not. But for convenience, could you provide an expression for the number of migrants per generation necessary to prevent divergence by drift? Does it have 1 as an estimated value under some parameter values?
DeleteJohn, allelic differentiation (at neutral loci under the finite island model or stepping stone model) is controlled by (to a good approximation) nu/m, where n is the number of demes, u is the mutation rate per generation, and m is the migration rate per generation.
DeleteThe classic argument in favor of Nm as the controlling factor goes like this:
1. Choose Gst as a measure of genetic differentiation, and apply it to the finite island model at equilibrium so that Gst is written in terms of the model parameters.
2. Check which values of the parameters make Gst small. This happens when Nm >>1.
3. Conclude that since Nm >>1 makes Gst small, the demes must show little differentiation in this regime.
The mistake is that Gst can be close to zero even when the demes are completely differentiated, so the conclusion does not follow. In order to make this argument correctly, we need to use a measure of genetic differentiation that takes its maximum value iff differentiation is complete (no shared alleles) and its minimum value iff the demes are genetically identical. I derived such a measure from first principles in the paper I mentioned, and then followed the steps above to discover that nu/m is the real controlling factor for differentiation.
While we're on this subject, Nick above makes a related misinterpretation when he implies that between-group differences are small when 85% of genetic diversity is within-group. Those figures usually use heterozygosity or (as in Lewontin) entropy in the ratio of within-group to total diversity, and those measures don't behave as most pop geneticists expect. The population could have 85% of its "diversity" within-group, even if the groups share no alleles with each other (one could even get that result comparing two different species, if a high-diversity locus is chosen). The measures can be fixed by transforming them as described by my papers.
Lou Jost
Thanks, but it would be really nice if you could answer my question. Can you present a formula for the number of migrants per generation necessary to prevent two populations from diverging by drift? Let us say for convenience that the number of demes exchanging migrants is exactly two.
DeleteI did. When m/u >>1 there will be little or no differentiation. When there are two demes, the actual formula for the expected value of differentiation at equilibrium is approx 1/[1+ m/u].
DeleteLou
So m <<1 should actually be sufficient to prevent significant differentiation at most loci other than microsatellites?
DeleteThe u is the mutation rate per generation per allele. So if mutation rate per base is 10^-8, u might be 10^-5 per locus per generation. At such loci, m of 10^-5 would show significant differentiation. For populations in the tens of millions, that would amount to an absolute number of migrants Nm of around 100 rather than 1. But all this, like your Nm=1 figure, is based on spatially homogeneous panmictic demes. This assumption is grossly violated by geographically widely dispersed human distributions with important geographic bottlenecks. A 2-d stepping-stone model would be more apt, and this would support more migration without erasing genetic differences between the population centers.
DeleteIt seems to me that pygmies are the best example of a well-defined human race. Not as identifiable are populations adapted to milk-drinking as adults and populations adapted to high altitudes. But those populations do not match up with any popular notion of race.
ReplyDeleteIf you discard phenotypically distinguishable populations, you can sequence genomes and find genotypically distinguishable populations. But what "race" could mean when you have to use twenty-first century technology to find it is beyond me. Also, I don't think the evolution is gene selection and therefore genes are everything mode of thinking very convincing.
I sent a link to this post to Mario Estable and we've exchanged a view email messages. He does not agree with my views on the existence of human races and he does not agree with the information I posted.
ReplyDeleteHowever, his main criticism is that I am attacking him personally instead of dealing with the scientific facts. He warns me that this could lead to potential legal action for defamation.
Distinguishing between ad hominem attacks and the substance of an argument is a constant problem. Professor Estable published a letter in the Toronto Star and he signed it "Mario C. Estable, Associate Professor, Department of Biology, Ryerson University." Any reasonable person would assume that the scientific facts in such a letter are accurate since they have to do with biology and since Professor Estable lists his credentials. Presumably he did this in order to make it look like he was speaking from authority.
I took the opportunity to question whether all biologists would agree with Professor Estable and I pointed out factual errors in his letter. I concluded by saying that some biologists don't know very much about human races and they probably shouldn't pretend to be authorities.
I presented evidence that what Professor Estable wrote in the newspaper was factually incorrect and that his claims to be speaking for all biologists were untrue. It's too bad that Professor Estable doesn't want to debate the scientific evidence because that could be an interesting discussion.
You presented evidence (or at least alluded to evidence) that he was wrong about some of his points, but you presented no evidence that he was wrong about his main point, that there are no races.
DeleteWhat evidence do you have that there are races? (And what do you mean by that word?) Are you one of the biologists who know a lot about human races and should pretend to be authorities?
I presented an argument that the issue of human races was controversial. It is simply not true, as Professor Estable implies, that biologists are all agreed that human races do not exist.
DeleteI have an opinion on this issue, as do you. Both of us have acquired quite a bit of knowledge about the pros and cons of races. I think it's fair to say that we know a lot simply because we know that it's complicated and unsettled.
I don't agree with you but you are not wrong (yet). However, if you were to say that all biologists think like you, then you would be wrong.
What do we disagree on and why?
DeleteWe've been over this many times in the past fifteen years. We disagree on whether the species Homo sapiens is genetically homogeneous or whether it is subdivided into subgroups that have been genetically isolated from each other for long enough to be distinct and idenitifiable.
DeleteNo we don't. We all agree that the species is not genetically homogeneous and that there is geographic structure to a small proportion of the non-homogeneity. When it's necessary to distort an opponent's position in order to make an argument, you should be suspicious of that argument.
DeleteSo, what subgroups have been genetically isolated from each other for long enough to be distinct and identifiable? I don't see a lot of genetic isolation; I see almost entirely clinal variation. And please don't tell me that Mandinka look different from Malays. Taking distant points out of a continuum shows nothing.
I want to think about the bigger problem of whether human races actually exist or not. I think they do, but the debate tends to get very confusing because it's tied into racism.
ReplyDeleteThe problem with saying something like this is that it would have to be accompanied by a definition of the term "race". I can easily believe in luminiferous aether as long as I define it to be the same as interstellar vacuum. And in the present case people may have very different ideas of what "race" is, with very different implications of what that means for public policy.
Personally, my position is this: Supraspecific taxa are natural units by virtue of being monophyletic groups of species. Species themselves are more controversial but there seems to be a common understanding that they are the smallest clusters of individuals that are clearly separated from each other, with none or very few exceptional intermediates between them.
It follows then that all infraspecific taxa, such as races, varieties and subspecies, are not that clearly separated from each other because if they were we would call them species. And that means that races are pretty much impossible to delimit objectively, making them a matter of subjective choice as opposed to empirical science.
I wouldn't go that far. If we adopt the "biological" species concept, there could be clearly separable units within a biological species, as there are different criteria for species (actual or potential interbreeding) and subspecies (diagnosable fixed differences); and there clearly are subspecies within some species. Anas crecca is the first example I think of, as European and American populations can be distinguished at a glance, but they will cheerfully hybridize whenever they come into contact. The question is whether such groups exist within H. sapiens. Don't see any, myself. Larry says he does, but so far we don't know what.
DeleteAnas crecca is the first example I think of, as European and American populations can be distinguished at a glance, but they will cheerfully hybridize whenever they come into contact. The question is whether such groups exist within H. sapiens. Don't see any, myself.
DeleteSo if you pick 50 random africans, 50 random japanese and 50 random Norwegians you can't "distinguish them at a glance"?
John Harshman,
DeleteThat is why I wrote that species are a bit more controversial. In my area people rarely use the BSC, instead I'd argue most botanists intuitively use the Genotypic Cluster Concept of Mallet (1995) even if many of them wouldn't even know that name. They recognise the smallest clusters that are really distinct, and the stuff below it is by definition subjective.
The more important point is, and we are agree on it, even if you can show it for Anas crecca you cannot show it for humans. As several other people in this thread have pointed out, if you ask two scientists how many human races there are you are likely to get three different answers. That shows that although humans clearly differ from each other individually the circumscription of races in our species is highly subjective, not a matter of scientific fact.
Pedro Pereira,
The problem would still be that across all humans there is a complete gradient between those three "groups".
Alex,
Deletethere is a complete gradient in ring species, and we consider species A to be different from species E on the other side.
Notice that I'm not defending any particular position here. All I'm saying is that you CAN recognize differences between some major human groups, and it seems like special pleading to argue otherwise. If they should be considered as races, and where do we draw the line, and how many groups actually qualify, is of course another question. Much of it will be arbitrary, as so much is in science, and maybe there is no real benefit that can be derived from trying to force an answer for this particular case.
For me the point is simple, even if the answer is not: race in humans should be seen excately through the same criteria as in any other metazoans. Personaly, I don't have an answer, but I want consistency in arguments.
Just to clarify my position:
DeleteI personaly think that trying to fit human groups into "races" wont probably work because any choice of groups will be too arbitrary and no bioscientist will ever agree if one should be bin people into 4, 5, 20, or 33 groups. There simply is no good criteria to do so that feels somewhat "natural".
But I think it doesn't make sense to state that one "can't see any group differences" in human groups. If one couldn't it should be obvious that racism would have never have arisen in the first place. We CAN see. Anyone can think of particular phenotypic characteristics that are Indian, or Pigmy, or Somali, or San, or whatever. But where do you draw the line? If we can't agree where the lines are drawn, and if any clustering depends ultimately on an arbitrary choice of criteria, then there's no point in arguing for the recognition of races.
Well, but that is pretty much my position!
Delete1. Yes, there are genetic differences between humans, even adaptive ones, e.g. the capability to get along with less oxygen in Aymara or Sherpas or greater tolerance of cold in Inuit.
2. No, races are not a scientific or even just useful category because there is no objective way to delimit them. Also, we tend to mix up anyway the moment people get together in the same area.
As for ring species, I was recently surprised to read this: http://whyevolutionistrue.wordpress.com/2014/07/16/there-are-no-ring-species/
What Alex said. But to restate: nobody disagrees that there is genetic variation in humans, some of which is geographically structured. (And I don't know why people keep thinking that this is an issue, unless it's just easier to argue against than the real issue.) You can indeed tell Norwegians from Japanese. But there is a continuous cline in characteristics across Europe and Asia from Norway to Japan. Where does the Norwegian "race" end and the Japanese one begin?
DeleteRing species, if there were any, wouldn't be relevant. It involves reproductive isolation between two populations but not between others. Since there's no reproductive isolation among any human populations, there's nothing to see there.
In one of Spencer Wells' television shows he visited some people in central Asia whose appearance was a fascinating midpoint between what we would think of as "European," "Chinese," "Middle Eastern," and "Indian." So interesting to see people who you'd have a hard time slotting into categories of appearance that we are accustomed to thinking of as reasonably distinct.
DeleteAs for ring species, I was recently surprised to read this: http://whyevolutionistrue.wordpress.com/2014/07/16/there-are-no-ring-species/
DeleteThat's a nice article, Alex. I don't frequent Coyne's blog becuse I dislike his general attitude, so I missed it. Maybe I should go there more and just make an effort to ignore the less savory side.
So interesting to see people who you'd have a hard time slotting into categories of appearance that we are accustomed to thinking of as reasonably distinct.
DeleteWell, that's because they ARE reasonably distinct. Otherwise we wouldn't be having this conversation nor would we find that group in Central Asia fascinating.
Anyway, I think we all generally agree. It would be more interesting if Moran chimed in, though.
Some of the commentators continue the huge error of picking a few points out of a continuum and concluding that they have shown there are "groups" and this is "obvious". Discrete sampling from a continuum DOES NOT DEMONSTRATE that there are discrete categories in real life. All it means is that your sampling was discrete. There's really no way for a discussion to progress until this is accepted by everyone.
DeleteNick,
DeleteI'm told that forensic experts can tell from looking at a skeleton whether someone was Asian, European, or African. How do you think they do this?
I'm no expert but I also have a pretty good track record identifying my students as Asian, African, or European (ancestry). Have I just been incredibly lucky?
We've met. I'm guessing that most of your ancestors come from Europe? Would I be able to do this if there weren't some approximation of "discrete categories"?
That does not mean that every single person fits neatly into one of the discrete categories. I don't think there are any examples or races or subspecies where this is the case. Do you?
Answering the questions in order:
DeleteBy picking a handful of skeletal traits and defining that as Asian or
whatever. As near as I can tell, for example, shovel incisors are pretty much the sole trait deemed to identify "Asians." The size of the zygomatic arch appears to be the distinguishing trait between "African" and "Caucasians."
If anyone has identified an African-American as being of African ancestry because of the hair texture and skin color, this is a good example of how forensic anthropologists do this. The scientific fact is that African-Americans are a mixed population, yet they are arbitrarily placed in the African category despite a large fraction (the majority?) being in the scientifically distinguishable clinal variation. No, it's not a matter of luck, it's a matter of using nonscientific criteria to identify social groups instead of biological ones.
Guessing "most" of Nick's ancestors came from Europe isn't good enough. If these major races really existed it would be possible to estimate the likelihood of crossbreeding. In the US, those who are convinced they can tell the mulattos or whatever are wrong.
Yes, because the categories are social, artificially discreet conventions.
I think this is rhetoric. I think the correct standard should be that you can fit, say 86% of humanity, into racial groups.
Amen to the author here questioning this Wade guy.
ReplyDeletehe truly, and others, want to bring back formerly obscure ideas about racial/sexual identity being a genetic influence on smarts and morality. its truly the presumptions behind these ideas in the early 1900's. Including in germany but not just there.
its too late and unneeded to use this to explain differences in people or people groups in history regarding intelligence and morality.
They can't say they are being brave. it requires bravery to fight these people as they always seem to have acceptance in high places.
i'm not saying they are evil but give mankind a chance before such unlikely conclusions.
However evolutionary thought is behind all this in concepts. Brain size rules still in humanoid fossil judgements and animals.
YEC teaches that there are no races and so no racial differences between mankind.
We all come from Noahs family.
after that man separated into bigger clans and then groups. Including language suddenly changing at babel.
there are no races but upon migration these segregated groups changed in body looks.
it does mean they changed as segregated groups. so no general evolution affecting a single tribe but all groups affected the same.
therefore the only reason i look like Russians/slavic peoples is because of like climate influences. otherwise i am as unrelated to them as to northern indians in india.
only related from a early indo european origin as language shows.
evolutionism struggles with likeness in people groups and so must say there was a original white tribe that under evolution attained its traits and later segregated into different languages.
its a odd thing that mankind is evidence that evolution does not create changes but other mechanisms in biology does.
anyways race has no influence on man but identity does as we frow up in our group.
in fact Wades mistake, like others, is testing people groups aFTER they moved to other peoples civilizations.
its not a controlled experiment.
Every human being who ever existed UPON BIRTH IN CANADA would be as moral and smart as anyone else.
However they must be born here otherwise they will be like thier own group in history.
Creationists have a advantage on these issues and should seize them but are uncomfortable to WADE into to race/sex things even if our position is perfect for equality of smarts in all men.
We are missing a opportunity but its a sad history and still not over about clowns saying this or that race is superior beyond free will.
yEC is ahead of the curve here and indeed evolutionists are more divided then they let on.
This is the one blog I regularly read where I learn as much from the comments as from the original blog posts. And I learn a LOT from Larry's posts.
ReplyDelete(Of course, what I learn from Robert Byers comments tends to be orthogonal to what I learn from the rest of the blog and comments.)
"tends to be orthogonal" -
DeleteNicely put. :)
"Linnaeus did not invent the word "race." It was in common use long before Linnaeus published his classification scheme in 1735. This is easily confirmed by checking the Wikipedia entry on human races. That article would have also alerted any biologist to the legitimate controversy and preventing him/her from propagating untruths."
ReplyDeleteLinnaeus certainly did not invent the term but to suggest the term was in common use long before the early modern period and easily confirmed by a glance at wiki is historically naive.
The question surrounding race and its historical usage before the modern period is not straightforward and requires more than a brief glance at wiki.
The question surrounding race and its historical usage before the modern period is not straightforward and requires more than a brief glance at wiki.
DeleteI don't really care to investigate the history of the word "race." However, if I were going to write a letter to a newspaper where I intended to claim that Linnaeus invented the word, I'd check Wikipedia to make sure I was on the right track.
What part of my statement do you think is wrong? Is it the phrase "long before" that bothers you?
Race as a word before modern times would mean nothing.
ReplyDeleteEvolution must believe in race because it must see genetic differences, looks I mean, in mankind.
so they must see segregated population groups being selected on.
So these groups or gene races are real to evolutionism.
They are not real to creationism.
No selection took place in creating man.
instead everyone got zapped who lived in those areas however segregated in groups.
there was never a original white tribe. let evolution must believe there was.
unless they say convergent evolution was affecting segregated population groups and with like results.
Then some evolutionists always made a next step that smarts in these segregated populations also was selected on and with results.
then they cheat on how they score results. no controlled experiments. just scoring your local school with tests in english!!
once again its peoples looks which confuses evolutionary theory.
Our looks must come from other biological mechanisms and not from selection on winners in population groups.
Evolutionists always had a problem with this stuff about people.
Creationists have a chance to make gains but got no nerve.
When I said "long before" I meant decades, as in the refererence mentioned in the wikipedia article.
ReplyDeleteBut that's irrelevant. The question before us is not semantic or historical. It concerns the biological meaning of "race"—a concept that's used in other species. Is there a scientific reason for not using that word (or a synonym) to describe genetically isolated subpopulations of humans?
Well we already have the word "population" or if you prefer the term "Mendelian Population".
ReplyDeleteCan you give an idea of which genetically isolated subpopulations you're thinking of? There really doesn't seem to be much of that going on.
ReplyDeleteI don't know that the word "race" is much used in biology these days. "Subspecies" is more common. The only case I can think of are "host races" in Rhagoletis. But I'm glad to equate the terms "race" and "subspecies", if that would help.
The Tasmanians had been isolated from the rest of the species for some 14 thousand years at the time of European arrival. The various Andamanese peoples (some of them still uncontacted) may have been isolated for 60 thousand years, though I'm not sure how complete their isolation was in the past millennia. Other than those, are there any other such populations?
ReplyDeleteSome recently discovered tribes in the Amazon, maybe.
ReplyDeleteOK, so to summarize, these are the races of H. sapiens: 1) Tasmanians (extinct), 2) various Andamanese peoples (more than one race?), and 3) some recently discovered Amazon tribes (more than one race?), and 4) presumably, everyone else.
ReplyDeleteYes, that's about it, though I have doubts about those Amazon tribes:
ReplyDelete(1) The peopling of South America was relatively recent in comparison with that of the Indo-Pacific region.
(2) Recent archeological findings show that the Amazon rainforest was quite densely settled in Pre-Columbian times, and there was nothing "primitive" about the local populations. They lived in large and well-organised communities, practised agriculture and made pottery as early as the sixth millennium BC. Their isolation may well be secondary and caused by depopulation and civilisational collapse as a result of the European conquest.
So it's two extant races, really, plus one that has recently gone extinct (unless the Andamanese can be split into more than one).
"I'd check Wikipedia to make sure I was on the right track."
ReplyDeleteThe articles somewhat fuzzy and misleading. Race comes to mean descended from common stock in English in the mid 16th century (at least according to the sources I have read: put in a disclaimer as the subject is prone to error as its so political and social).
One academic writing on medieval race concepts noted C.S Lewis's statement that people like to express approval or disapproval of things much more than they like describing them.
Certainly seems applicable to this subject and not just in regard to the middle ages. Accurate and clear description does seem rather absent with this subject.
Since almost nothing clear can be said about human "races" is it at least fair to say that they are less distinct than animal "breeds"?
ReplyDeletehttp://en.wikipedia.org/wiki/Breed
because clusters of breed characteristics are clearly delineated.
Differences in people are more then many animal breeds traits.
DeleteA absolute fact is people look very different from each other and its in the genes.
I understand breeds of some animal revert back to the original easily after being let alone. not us.
Therefore evolutionism mIST say man is segregated in original populations under a influence that changed them from a original single group.
No way around it.
the creationist can say its from our bodies merely reponding to a influence with a innate response.
The evolutionist must have selection .
So for all intents and purposes race is a real thing to evolution.
A population whose genes have come to be different from a original population and is stuck tight.
A creationist can say its not tight but could change with a triggering mechanism after a threshold is crossed.
So segregated populations did come to look like each other without genetic reproductive association.
In other words Celts look like Germans but this is unrelated to biological origins as a original common people.
Both just adapted traits from like influences. They were segregated populations but don't have thier mutual looks from a original tribe assuming selection results not from selection after segregation.
They are all innately responding in like fashion from already innate mechanisms.
Race is not the origin of traits for man. Just location, location, location without selection.
Race must be divorced from traits. This is the rub.
Evolution believers must say traits are from selection on a population and so race is not just a segregated breeding pop but their biology is defined by it.
So then a few cats every now and then say intellectual and moral traits are based on race.
A line of reasoning. Evolutionism has a tar baby problem. If outer loiks are from race selection then why not a option for inner workings above the neck.
Creationism has no problem and sees all men as just from Noah and his wife and no outer or inner selection ever happened. Its a myth.
Humans have geographically patterned variation that I think we can call races, probably because as a plant taxonomist I think that while intraspecific groupings mean something, they often don't mean much.
ReplyDeleteIdeally, subspecies would be geographically distinct population sets, characterized by fixed genetic differences. Sometimes this is the case, e.g. with populations on different islands, not longer interbreeding because of geography. Naming examples like this as different species is also reasonable.
Naming the extremes as gradual clines as subspecies seems unreasonable.
However, there are many intermediate conditions. If population sets A and B are separated by a mountain range, they may meet in the gaps and interbreed there. Interbreeding is small compared to the whole range, and calling these things subspecies can be useful. Even calling them species may be justified.
How far do we go in naming differences? I know species where subspecies (named, geographically based population sets) are probably defined by alternate sets of alleles at three or four loci. In fact, I know of a couple taxa where intraspecific taxa (formerly species!) are named by the phenotypic results of two loci, each with two alleles. (As the word about that example has gotten out, those taxa are no longer recognized.) I expect subspecies to be distinguishable much of the time, but I also expect intermediate individuals.
So in fact (though not in theory), subspecies are geography-based but the degree of genetic differentiation involved and the degree of interbreeding that occurs vary greatly and depend on somewhat arbitrary decisions about where to draw lines.
"Races" are even less well defined than subspecies.
Humans have geographically based (or formerly geographically based) differences in skin color, hair, eye color, facial features and such medically important features as blood types, vitamin metabolism, sunburn rates, skin cancers, and metabolism of salt, alcohol, and sugar. We can classify these differences as racial.
If we cared as little about classification of humans as we care about classification of plants, we would surely classify humans into races -- and have heated but unimportant arguments about how much to split and how to treat population sets that clearly originated as interracial hybrids. The problem is that once we name things, we think they are things that are more distinct than they really are. We over-interpret races. We confuse cultural and genetic causes for behavior. We assign values to groups rather than individuals. (I think immediately of evidence for black/white differences in intelligence. Students seemed incapable of seeing that if studies show an equal range but slightly different average [which could be, probably is, due to cultural issues including consequences of poverty due to discrimination], saying blacks are less intelligent than white isn't really justified by the data, and using that generalization to say anything at all about individual X of either race is just stupid.)
Personally, I come from a large, biracial or quadriracial (depending on your classification system) extended family that includes many fine biracial nieces and nephews. Racial differences are interesting, but traits I value most in people, such as niceness, responsibility, and religious and political views, don't segregate along racial lines.
So are there human races? Yes, but that means a lot less than you might think.
Barbara. I don't agree with your evidence you present that there are races.
ReplyDeleteYou are basing races existence based on biological traits. you presume these traits came after populations segregated unless they have the same traits thought segregated.
So you are saying the population had selection working upon it and the origin for the different traits.
The race exists because the traits are from biological segregation and so selection acting with reproduction.
yet if the traits differences or likeness did not come from selection on populations but came from innate mechanisms dealing with like needs in some geography place then the races are not based on selection on reproduction .
They would not be races. They would not look the way they do because of reproductive segregation with selection.
Therefore your race idea is entirely based on WHY people look different from each other. !!
So evolutionists are forced to say race or reproductive pop's with selection is real.
Creationists can say its not real.
No evolution equals no race. Its a small matter that people are segregated from each other including reproduction results. We looik the way we do for general reractions in some geographic are and it happened quick and not long ago. After the complete segregation of language groups.
Therefore there are no races because biological looks differences are not from evolution acting on segregated populations.
if all men looked the same completely then there would no claims of different races.
The equation here is that its nOT about biology results but about WHY the results.
The why determines the conclusion of the existence of races in mankind.
"Is there a scientific reason for not using that word (or a synonym) to describe genetically isolated subpopulations of humans?"
ReplyDeleteIs there a genetically isolated subpopulation of humans? That's a very dubious proposition today. Since that's gone by the wayside, it follows that the term race should go by the wayside.
There are populations that have been geographically isolated to some degree. But if you compare the phenotypic variation between humans to the phenotypic variation in horses, I think you should conclude that human "races" are similar to so-called "color" breeds in horses. These are distinguished from "true" breeds for good reason. It concerns the biological meaning of "race"—a concept that's used in other species. Is there a scientific reason for not using that word (or a synonym) to describe genetically isolated subpopulations of humans?
As for synonyms such deme or ecotype, I think they are rejected in reference to humanity because when someone can identify a population that might count as a deme or ecotype, it doesn't match up with any popular usage of "race" in ordinary language, especially none of the so-called "major" races. "Breed" is not used because it refers to artificial selection and is therefore inappropriate for humans.
The real question is when it is scientfically appropriate to use the term "race" in reference to humans? When referring to the local gene pool.
But is there any scientific need for referring to "race" in human when meaning something like Asian, African or European? No.
Let's assume that we kill all dogs except for Chihuahuas and Grand Danois. Now there can't be any interbreeding. Would we now have two different species?
ReplyDelete