“…for he has no right to give names to objects which he cannot define.” –Charles Darwin
Do “races” exist as meaningful biological categories? Physical anthropologists and human biologists have been studying race (i.e., blacks vs. whites, or Europeans vs. Asians) for centuries. For most of that time, they subscribed to the perspective that race was a taxonomic category, and they sought to identify the biological characteristics (such as cranial shape or skin color) that characterized and defined these different groups. This perspective assumed that each individual was a member of a single racial category, that the differences between racial categories were biological, and that these categories were predictive of other traits (such as ancestry, temperament, intelligence, or health).
But it gradually became clear that this understanding was not scientifically sound. Groupings of people by skin color did not produce the same result as groupings of people by skull shape, nor of blood type. Furthermore, as scientists began to study human variation with the tools of genetics (in the process creating my fields, anthropological genetics and human population genetics), it became apparent that human genetic variation does not divide humans into a few discrete groups. There are virtually no sharp boundaries, either with physical features or with patterns of genetic diversity, that show where one population “ends” and the next “begins”.
These observations have led the majority of physical anthropologists, human biologists, and human geneticists in recent decades to conclude that the racial groups we recognize are social categories constructed in a specific cultural and historical setting, even if we consider physical features when categorizing people. These social categories can have biological consequences (for example, someone who experiences the stress of racism may be more likely to develop high blood pressure and hypertension than someone who does not).
Racial groupings differ from culture to culture. For example, although in the United States Chinese and Japanese peoples are usually viewed as one “race” (Asian), they are seen as members of different racial groups in South Africa. Racial groupings also vary over time within a single culture, as can be seen below in the United States census classifications of race over several decades.
However, according to former New York Times science writer Nicholas Wade, we should never have stopped thinking of race as a biological taxonomic category. In his new book, “A Troublesome Inheritance: Genes, Race and Human History”, Wade takes it upon himself to educate scientists about the errors of our interpretations of human genetic diversity.
Wade claims that the latest genomic findings actually support dividing humans into discrete races, and that the genetic makeup of different races contributes to behavioral and economic disparities. In a spectacular failure of logic, he asserts that those who disagree that races are meaningful biological categories in humans must ALSO think that human populations do not differ genetically, or have not been affected by evolution.
There is a lot to criticize in this book, particularly Wade’s imaginative storytelling in chapters 6-10 (“a much more speculative arena,” as he puts it). He explains that English populations have a “willingness to save and delay gratification”, which “seems considerably weaker in tribal societies” (pp. 184-185), and these differences must be genetically based, despite his admission that “the genetic underpinnings of human social behavior are for the most part still unknown” (p. 15), and numerous critiques of this hypothesis. In chapter 8, he asserts that Jews are adapted for capitalism in a manner analogous to the Eskimo’s adaptation to survival in an Arctic environment (p. 214) — an assertion unsupported by scientific evidence, to put it mildly. (Wade seems to be unaware of the consequences of laws prohibiting Jews from owning land and farming over much of Europe for centuries – and instead speculates that “their genes were adapted for success in capitalism”).
But others have already critiqued these aspects of his book. I’m far more interested in the central premise of Wade’s argument, which is passing unchallenged by all but a few reviews: “At least at the level of continental populations, races can be distinguished genetically, and this is sufficient to establish that they exist” (p. 122). If Wade is right and races are distinct biological categories, then we would reasonably expect that they would be unambiguously different from each other genetically and physically (as well as behaviorally, according to Wade). One should be able to define each race with a set of objective criteria, which could be used by any person to independently reach the same classifications (and number of classifications) as Wade. Furthermore, these categories should have predictive power; that is, features that define race should be in concordance with new discoveries of genetic diversity.
What is race?
To begin with, Wade can’t provide a clear definition of “race.” He tries to rely instead on loose associations rather than definitive characteristics, which forces him to conclude both that physical traits define race but that the traits can vary from person to person: “races are identified by clusters of traits, and to belong to a certain race, it’s not necessary to possess all of the identifying traits” (p. 121).
With such a shifty, casual footing, it’s no surprise that Wade’s conclusions are unsound. He can’t keep the number of races straight:
Wade can’t settle on a definite number of races because he can’t come up with a consistent, rigorous definition of what “race” means. He uses terms like “major race”, “race”, “subrace”, “group”, or “population,” but doesn’t provide any serious, objective ways to distinguish between these terms for arbitrary groupings of people arbitrary groups.
Rather than just announcing his subjective opinions about race, Wade wants to ground them in science. He tries to use genetics: “Such an arrangement, of portioning human variation into five continental races, is to some extent arbitrary. But it makes practical sense. The three major races are easy to recognize. The five-way division matches the known events of human population history. And, most significant of all, the division by continent is supported by genetics.” (p. 94)
To support his claim, Wade relies heavily on a 2002 paper (by Rosenberg et al.) that used a program called structure to group people based on similarities in markers distributed across the genome. He notes that the program identified five major clusters in this 2002 study, which corresponded to the major geographic regions (Africa, Eurasia, East Asia, Oceania, and America) of the world. Therefore, Wade argues, these results clearly show that humans are divided up into racial categories that match continents.
Charles Murray, author of The Bell Curve, who recently reviewed Wade’s book in the Wall Street Journal, agrees:
A computer given a random sampling of bits of DNA that are known to vary among humans—from among the millions of them—will cluster them into groups that correspond to the self-identified race or ethnicity of the subjects. This is not because the software assigns the computer that objective but because those are the clusters that provide the best statistical fit.
But Wade and Murray are both wrong. Structure didn’t simply identify five clusters. It also identified two, three, four, six, and seven clusters. (Rosenberg et al. 2002 actually identified up to 20 divisions, but 1-7 are the primary ones they discussed. They also divided their worldwide sample up into regions, and then ran structure within those regions, to look at more fine-scale population structure.)
Why? Researchers using structure have to define the number (K) of clusters in advance, because that’s what the program requires. The program was designed to partition individuals into whatever pre-specified number of clusters the researcher requests, regardless of whether that number of divisions really exists in nature. In other words, if the researcher tells structure to divide the sampled individuals into 4 clusters, structure will identify 4 groups no matter what–even if there is really only 1 group, or even if there are really 14 groups.
So, when Rosenberg et al. (2002) told structure to use K=6? They got six clusters, with the sixth corresponding to a northwestern Pakistani group, the Kalash. Does this make the Kalash a separate race? Wade doesn’t think so. When they told structure to use K=3? They got three clusters back, corresponding to Africa, Europe/Middle East/South Asia, and East Asia/Oceania/Americas. So are Native Americans and Australians not separate races? Rosenberg et al. never published any statistical evidence that justifies picking 5 races instead of 7, or 4, or 2 (although such methods do exist–see Bolnick et al. 2008). Wade seems to like K=5 simply because it matches his pre-conceived notions of what race should be:
“It might be reasonable to elevate the Indian and Middle Eastern groups to the level of major races, making seven in all. But then many more subpopulations could be declared races, so to keep things simple, the five-race, continent-based scheme seems the most practical for most purposes.” (p. 100)
Practical. Simple. Wade wants us to cut up human diversity into five races not because that’s what the statistical analyses show, but because thinking about it as a gradient is hard.
Wade isn’t even using the tools of genetics competently. The authors of the paper he relied on, as well as subsequent studies, showed that different runs of the program with the same data can even produce different results (Bolnick, 2008). Structure’s results are extremely sensitive to many different factors, including models, the type and number of genetic variants studied, and the number of populations included in the analysis (Rosenberg et al. 2005). When Rosenberg et al. (2005) expanded the 2002 dataset to include more genetic markers for the same population samples, they identified a somewhat different set of genetic clusters when K=6 (Native Americans were divided into two clusters and the Kalash of Central/South Asia did not form a separate cluster). In fact, Rosenberg et al. (2005) explicitly said:
“Our evidence for clustering should not be taken as evidence of our support of any particular concept of ‘biological race.’”
Finally, the creators of structure themselves caution that it will produce rather arbitrary clusters when sampled populations have been influenced by gene flow that is restricted by geographic distance (i.e. where more mating occurs between members of nearby populations than between populations that are located farther apart, a pattern we geneticists refer to as isolation by distance). As this pattern applies to the majority of human populations, it makes the results of structure problematic and difficult to interpret in many cases. These limitations are acknowledged by anthropological geneticists and population biologists, who interpret the results of structure cautiously. It’s very telling that Wade, a science reporter, chose to ignore the interpretations of the experts in favor of his own.
Human biological variation is real and important. I’ve studied it my entire professional career. We can see this variation most easily in physical traits and allele frequency differences between populations at extreme ends of a geographic continuum. Nobody is denying that. Let me repeat this: no one is denying that humans vary physically and genetically. All anthropologists and geneticists recognize that human differences exist. But Wade, and others who agree with him, have decided that certain patterns of variation—those which happen to support their predefined notions of what “races” must be—are more important than others.
Wade’s perspective fits with a larger pattern seen throughout history and around the world. Folk notions of what constitutes a race and how many races exist are extremely variable and culturally specific. For example, the Bible claims that all peoples of the world are descended from Noah’s three sons, mirroring the popular concept of three racial divisions (Caucasians, Africans, and Asians). On the other hand, the five-part division of races seems most “logical” to Wade. Anticipating confusion on this point he claims: “Those who assert that human races don’t exist like to point to the many, mutually inconsistent classification schemes that have recognized anywhere from 3 to 60 races. But the lack of agreement doesn’t mean that races don’t exist, only that it is a matter of judgment as to how to define them” (p. 92).
A matter of judgment. So, rather than being defined by empirical criteria, as Wade had asserted so confidently earlier in the book, it really is just a subjective judgment call. The differences between groups are so subtle and gradual that no objective lines can be drawn, so Wade draws his own on the basis of his own preconceptions.
How subtle is the gradient that Wade is chopping up? Humans are incredibly similar genetically. We only differ by about 0.1% of our genome. Compare that to chimpanzees, our closest relative. Individual chimps from the same population show more genetic differences than humans from different continents.
The genetic differences that exist in human populations are important, because they help us understand our evolutionary history. The most genetic diversity is found in populations in Africa, where our species originated. Subsequent migrations across the continents resulted in sampling a subset of the genetic diversity present in the ancestral populations; thousands of years of localized evolution and cultural practices have produced region-specific adaptations, such as the ability to thrive at high altitudes. These adaptations have influenced particular genes and traits, but the overall pattern of genetic variation is clinal, meaning that for the most part it varies gradually with geographic distance. Groups that live close together are more closely related to each other (and more genetically similar) than they are to groups farther away. (People marry and have children more frequently with people who live close to them than they do with people who live farther away). Other evolutionary forces (founder effects, selection, drift, and migration) have all contributed to patterns of genetic diversity that we see in populations today.
But these patterns of human diversity don’t give us a scientifically viable definition of race as a taxonomic unit. As Agustin Fuentes puts it, with emphasis added:
“when you compare people from Nigeria, Western Europe and Beijing you do get some patterned differences…but these specific groups do not reflect the entire continental areas of Africa, Europe, and Asia (the proposed “continental races” of African, Caucasian and Asian). There are no genetic patterns that link all populations in just Africa, just Asia or just Europe to one another to the exclusion of other populations in other places. If you compare geographically separated populations within the “continental” areas you get the same kind of variation as you would between them. Comparing Nigerians to Western Europeans to people from Beijing gives us the same kind of differences in variation patterns as does comparing people from Siberia, Tibet and Java, or from Finland, Wales and Yemen, or even Somalia, Liberia and South Africa— and none of these comparisons demonstrates “races.”
In fact if you use the common level of genetic differentiation between populations used by zoologists to classify biological races (which they called subspecies) in other mammals, all humans consistently show up as just one biological race.”
(Also see Templeton AR, 2013. Biological races in humans. Studies in History and Philosophy of Biological and Biomedical Sciences, http://dx.doi.org/10.1016/j.shpsc.2013.04.010)
Wade even seems to agree with population geneticists that there aren’t any races, just clinal distributions of genetic diversity: “Because there is no clear dividing line, there are no distinct races—that is the nature of variation within a species.” (p 92).
In other words, he can’t define distinct races. He just knows them when he sees them
I’ve focused a lot of this review on numerous technical details because I think that it’s very important that non-geneticists understand the degree to which Wade is distorting the results of recent research on genome-wide human variation. I won’t speculate whether this distortion is deliberate or a result of simple ignorance about genetics, but it is serious. There is a great deal more in this book that also needs to be critiqued, such as Wade’s assertion that the genetic differences between human groups determine behavioral differences, resurrecting the specter of “national character” and “racial temperaments”. But as I’ve shown here, Wade’s book is all pseudoscientific rubbish because he can’t justify his first and primary point: his claim that the human racial groups we recognize today culturally are scientifically meaningful, discrete biological divisions of humans. This claim provides a direct basis for the whole second half of the book where he makes those “speculative” arguments about national character. In other words, the entire book is a house of cards.
It’s also worth noting the extent to which Wade’s argument here is a variation on the Galileo fallacy: the fact that one bravely holds a minority view in science is considered to be sufficient evidence of the worth of one’s position. I’ve seen it used over and over again in responses to my criticisms of pseudoscience, and it’s no more persuasive for Wade than it is for creationists or homeopaths.
Further reading:
“If scientists were to make the arbitrary decision that biological race is real, can you think of a positive outcome?” –a nice piece by Holly Dunsworth: http://ecodevoevo.blogspot.com/2014/05/if-scientists-were-to-make-arbitrary.html
“The troublesome ignorance of Nicholas Wade”, also by Agustin Fuentes:
http://www.huffingtonpost.com/agustin-fuentes/the-troublesome-ignorance-of-nicholas-wade_b_5344248.html?utm_hp_ref=tw&utm_content=bufferfad4c&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer
“On the origin of white power” by Eric Michael Johnson:
http://blogs.scientificamerican.com/primate-diaries/2014/05/21/on-the-origin-of-white-power/
A critique of Structure:
Bolnick DA. Individual ancestry inference and the reification of race as a biological phenomenon. In: Koenig BA, Lee SS-J, Richardson SS, editors. Revisiting Race in a Genomic Age. New Brunswick, NJ: Rutgers University Press; 2008. pp. 77–85.
Jon Marks: “The genes made us do it: The new pseudoscience of race.”
http://inthesetimes.com/article/16674/the_genes_made_us_do_it
Barbujani and Colonna, 2010. Human genome diversity: frequently asked questions.
Click to access pr-1400-file_it-Barbujani-Colonna.pdf
******************************************
Many thanks to Deborah Bolnick, Colin McRoberts, Jay Kaufman, Jonathan Kahn, Troy Duster, and Rick Smith.
Please review my Site Policies before commenting. Disagreement with me is fine; bigotry is not.
Violent metaphors 
good review. a few comments
1) fwiw, i do think races are a useful biological category, if less useful than species. but species itself can be problematized in the same manner as race. the issue is instrumental utility. most biological classes in something like pop-gen are human reifications (e.g., folded site frequency spectrum?).
2) the issues with structure is why it’s important to also look at PCA, etc., which are less tied to a specific model (though they have assumptions).
3) i the idea that human populations are clinal is defensible, but as a matter of history i think it may be totally misleading in terms of how variation arose. in particular, it often assumes a sort of isolation by distance dynamic after out of africa diversification. as you know that’s not really the case; populations like south asians arose from a discrete admixture event ~4000 years ago. that event puts s asians between west eurasians and east eurasians (skewed a little toward west) in keeping with their geographic position, but that’s because two discrete clines are mixed up there. some distributions of genes (e.g.,convergent evolution of light skin in east and west eurasian) imply to me that in the pre-holocene period there was a lot of impediment to gene flow. the holocene ‘mushed’ a lot of this up so that most of the major populations are today compounds of admixtures. this is IMO a model that isn’t usually what people have in mind when you say clines.
4) minor dote, the number of K can be arrived at by things like like the ‘delta K’ method. obviously some K’s are ‘better fits’ than other K’s in terms of the model. for readers: if you take tuscans, han chinese, and yoruba, you can make K = 10, but K = 3 will be much more supported (e.g., show less variation in likelihood across runs, etc.)
5) finally, i think this paragraph needs to be updated, as it is true circa 2008:
The genetic differences that exist in human populations are important, because they help us understand our evolutionary history. The most genetic diversity is found in populations in Africa, where our species originated. Subsequent migrations across the continents resulted in sampling a subset of the genetic diversity present in the ancestral populations; thousands of years of localized evolution and cultural practices have produced region-specific adaptations, such as the ability to thrive at high altitudes. These adaptations have influenced particular genes and traits, but the overall pattern of genetic variation is clinal, meaning that for the most part it varies gradually with geographic distance. Groups that live close together are more closely related to each other (and more genetically similar) than they are to groups farther away. (People marry and have children more frequently with people who live close to them than they do with people who live farther away). Other evolutionary forces (founder effects, selection, drift, and migration) have all contributed to patterns of genetic diversity that we see in populations today.
http://biorxiv.org/content/early/2014/03/21/003517
(i would put up a blog post, but i’m kind of sick of writing about the genetics of race at this point 🙂
“good review”
Razib, this was a pretty vapid and somewhat ditzy critique of the biological race concept — but then maybe you don’t have a good handle on the concept yourself…..
Raff: “Groupings of people by skin color did not produce the same result as groupings of people by skull shape, nor of blood type.”
This is called the independent variation argument; it doesn’t stand up to multivariate analysis, a technique which, in practice, is hardly novel: Blumenbach (1795): “To this enumeration, however, I must prefix a double warning; first, that on account of the multifarious diversity of the characters, according to their degrees, one or two alone are not sufficient, but we must take several joined together; and then that this union of characters is not so constant but what it is liable to innumerable exceptions in all and singular of these varieties”
Raff: ” it became apparent that human genetic variation does not divide humans into a few discrete groups….”
Ok, but what did early racialists actually argue?
Blumenbach (1795): “Thus too there is with this that insensible transition by which as we saw the other varieties also run together”
Darwin (1871): “But the most weighty of all the arguments against treating the races of man as distinct species, is that they graduate into each other, independently in many cases, as far as we can judge, of their having intercrossed”
Also, there is a nonsequitor embedded here. One doesn’t need sharp boundaries to divide individuals into discrete populations. Rather, in absence of sharp boundaries more individuals will fall in an undefined region (e.g., zones of intergradation). Imagine carving out regions in space. One could define a region as a set of points in which each point is more similar in distance to others in the same region than to points in a different region. This would results in discrete categories (e.g., either region A or B or neither), with some points — the number dependent on the number of dimensions under consideration and the shape of space — in the undefined region. So, yes, you can more or less classify individual into discrete natural populations; hence Tal (2012): “The probability that a random pair of individuals from the same population is more genetically dissimilar than a random pair from distinct populations is primarily dependent on the number of informative polymorphic loci across genomes from the total population pool. This probability asymptotically approaches zero with a sufficiently large number of informative loci, even in the case of close or admixed population.”
Raff: “These observations have led the majority of physical anthropologists, human biologists, and human geneticists”
A while back, I looked through all of the published surveys which asked a variant of the question, “Do human biological races exist?” These surveys were mostly from the U.S, and Western Europe and didn’t include China and Russia, countries in which the reality of human races is overwhelmingly accepted. About 50% said “no”, 10% were undecided, and 40% said “yes”. The reasons given for rejecting races were silly, though: more variation within races, gene flow between populations, no pure races, etc. Clearly what was being rejected was not a typical concept of race, modern or historic (keep in mind that early pre-Darwinian racialists, being monogenists, believed that races (as opposed to species) were different inbred lineages of Adam, only several thousand years separated).
Raff: “Racial groupings differ from culture to culture. For example, although in the United States Chinese and Japanese peoples are usually viewed as one “race””
So, race is a polyseme and the different concepts lend themselves to different classification schemes. But Wade is dealing with a typical biological race concept in which races are understood to be subspecific natural populations (here “natural population” is an analogue of “natural classification” which is juxtaposed with “artificial classification”. So given this biological concept…
Raff: “One should be able to define each race with a set of objective criteria, which could be used by any person to independently reach the same classifications (and number of classifications) as Wade”
In biology there are a number of race concept, just as there are a number of species concepts (e.g, Autapomorphic species, Biospecies, Cladospecies, Cohesion species, etc.). Wade defensibly equates races with the populations geneticist’s races — now called “genetic populations”. So you either have to critique this high currency concept or launch a semiotic attack on the label. Good luck with ether — for example, compare Wade’s population genetic races to e.g., Boyd’s. As for an operational definition of “genetic population, some have offered: “Two individuals, I1 and I2, belong to the same genetic population if (a) their genetic relationship, measured with the coefficient of kinship, is greater than zero and (b) their kinship is much higher than kinship between them and some individual I3, which is said to belong to another genetic population.” (Aulchenko, 2010. Effects of population structure in genome-wide association studies).
Personally, I would operationalize them as: “sets of (subspecific) population whose members are more overall genetically dissimilar to members of other sets than they are to other members of their own set.” But then I like the idea of quasi (relational) essentialistic races. The point though is that since Wade identifies races with genetic populations, logically one would have to take on the population genetics concept or the race label. Unsurprisingly, nether was done.
Raff: “But these patterns of human diversity don’t give us a scientifically viable definition of race as a taxonomic unit. As Agustin Fuentes puts it, with emphasis added:”
So, Fuentes argues that there aren’t human races given (Ernst Mayr’s) — apparently scientifically viable — subspecies concept and some made up — by Templeton — genetic differentiation criteria and from this you get: there is no “scientifically viable definition of race as a taxonomic unit”?
(I will continue on request.)
Please continue, Chuck.
(Continued)
The last point touches upon a central problem with this critique. Either there are biologically valid concepts of race or there are not. But Raff wishes to have it both ways, so she can say that there are no races because there is no scientifically viable definition of them and because there are viable definitions of race and human populations don’t qualify as races by these. Well, which is it?
I can think of four prominent biological race concepts off of the top of my head: cladistic, evolutionary, population, and ecotypic. They are biological races seen from the perspective of different biological research programs. For a discussion of the former two, see the following defenses: Andreasen, R. O. (2004). The cladistic race concept: a defense; Hardimon, M. O. (2012). The Idea of a Scientific Concept of Race. To note, the population genetic concept does not purport to be a taxonomic unit, but rather a conceptual tool for understanding human biological variation. This concept refers to breeding populations (sometimes also confusingly called genetic populations) retrospectively understood.
The evolutionary race concept is Mayr’s evolutionary — as in evolutionary taxonomy –based one; when formally recognized, that is, when given a trinomen, these are called zoological subspecies or “geographic races” (Mayr and Ashlock, 1991) (not to be confused with Mayr’s “microgeographic racees” or with geographically delineated races in general). While only Mayr’s formally recognized races (zoological subspecies) are taxonomic units, his lesser races are nonetheless biological units — in the sense that “demes”, “morphs”, “clines” or other taxonomically unrecognized biological constructs are — they are subspecific natural populations which simply have not differentiated enough to warrant, for pragmatic reasons, being assigned a trinomen. I simply can’t see how one can grant the taxonomic validity of formally recognized races without also granting the biological validity of non-formally recognize ones — as if races that failed to meet the vague conventional subspecies qualifying criteria couldn’t be thought about. This consideration brings to mind Kant’s reply, when his race concept was challenged on the grounds that it didn’t describe a formal taxonomic unit: “The fact that this word does not occur in the description of nature [i.e., in Taxonomy] (but instead of it that of variety), cannot prevent the observer of nature from finding it necessary with respect to natural history.”
The final concept is the ecotypic one: King and Stansfield (1990): “A phenotypically and/or geographically distinctive subspecific group, composed of individuals inhabiting a defined geographical and/or ecological region, and possessing characteristic phenotypic and gene frequencies that distinguish it from other such groups. The number of racial groups that one wishes to recognize within a species is usually arbitrary but suitable for the purposes under investigation” I find this conception to be somewhat vague (e.g., “Can forms be ecotypes?”). but apparently ecologists find it useful.
The point here is that there are a bunch of viable well vetted biological race concepts — in addition to bio-anthropological ones They differ mainly in the way that species concepts differ (e.g., genetic species versus ecospecies), but they are related in that they generally — I don’t know about ecotypes — describe populations where members are arranged by genealogical and/or genotypic relationship, where what is of interest is overall similarity not just similarity in specific genetic characters such a specific chromosomes e.g., as in the male morph.
We need not worry about all these concepts though and their various operationalizations, since Wade focuses on the population genetic ones, which is very similar to Mayr’s evolutionary one in that the matter of concern is relative geneotypic similarity. Now, either “genetic populations” represent a viable biological construct or not. They obviously do — granted, the operationalizations are all over the place. And either these genetic populations can reasonably be labeled races or not. Again, they obviously can on a number of accounts. So we have our scientifically valid, viable race concept, which allows us to discuss human races. Now, with this as our basis, we can better examine your critique.
Raff: “To begin with, Wade can’t provide a clear definition of “race.” He tries to rely instead on loose associations rather than definitive characteristics”
Ok, but population genetic and evolutionary races are defined in terms of overall genetic similarity, not in terms of “definitive” characteristics. The situation is similar to that of twins. Twins are define by their coefficient of relatedness; this relatedness conditions character similarity; this character similarity can be used to diagnose twin status, but it doesn’t define it. Alternatively, ducks have wings; but they are not ducks because they have wings; rather, they have wings because they are of the duck lineage which has been subject to such and such evolutionary pressure. .
Raff: “With such a shifty, casual footing, it’s no surprise that Wade’s conclusions are unsound. He can’t keep the number of races straight:”
Ok, but everyone recognizes the nested nature of “genetic populations”. To quote Aulchenko (2010) again: “One can see that this definition is quantitative and rather flexible (if not to say arbitrary): what we call a “populations” depends on the choice of the threshold for the “much-higher” probability. Actually, what you define as “the same” genetic population depends in large part on the scope and aims of your study. In human genetics literature you may find references to a particularly genetically isolated populations, populations of some countries (e.g., “German population”, “population of United Kingdom”), European, Caucasoid or even general human population. Defining a population is about deciding on some probability threshold.
If you look at the old taxonomies, you will see that races will be prefixed with terms like “primary” or “major” or “continental” for this reason. This surely isn’t a novel point (see for example: Coon and Garn’s (1955): ” On the Number of Races of Mankind.” ) Generally, you can’t determine a true number of races or a true classification because there is no way to determine a “true” level of genetic relatedness; you can only determine a specific number and a specific classification given a pre-selected grain of genetic focus. This, of course, holds for all natural populations. The shifty biologists talk about x kingdoms, y classes, z species — they classify Sailer as an animal then a mammal then a primate — they can’t keep there natural populations straight! Madness.
Raff: “He uses terms like “major race”, “race”, “subrace”, “group”, or “population,” but doesn’t provide any serious, objective ways to distinguish between these terms for arbitrary groupings of people arbitrary groups”
Except, cluster analysis.
Because, in biology, arbitrary groupings are ones that don’t indicate evolutionary relationship, thus overall genotypic similarity. Mayr and Ashlock (1991): “classification based on convenient and conspicuous diagnostic characters without attention to characters indicating [evolutionary] relationship; often a classification based on a single arbitrarily chosen character instead of an evaluation of the totality of characters”.
What distinguishes (population genetic and evolutionary) races from other groupings is that they are not biologically arbitrary groups. Darwin (1859): “From the first dawn of life, all organic beings are found to resemble each other in descending degrees, so that they can be classed in groups under groups. This classification is evidently not arbitrary like the grouping of the stars in constellations’’ (or like forms or morphs). So there is a sense in which class and order are arbitrary groupings — as one could make an indefinite number of subdivisions or none– but there is a sense in which they are not, since members are arranged according to genetic similarity (genealogical for cadists and genotypic for evolutionary taxonomists).
I don’t see why this is difficult.
Raff: “The program was designed to partition individuals into whatever pre-specified number of clusters the researcher requests, regardless of whether that number of divisions really exists in nature.”
But seriously, how could the “number of divisions really exists in nature”? This is a genetic problem, of course. For example, how many particles does a carbon atom comprise? More importantly, no one ever claimed that some level of racial analysis was True — say, for example, in the way that it was once believed that species represented ontologically privileged level of analysis.
Raff: “The authors of the paper he relied on, as well as subsequent studies, showed that different runs of the program with the same data can even produce different results (Bolnick, 2008).”
The best way to do this would be to use whole genome analysis to sort people by continental level races. Alternatively, you can look at the results from multiple studies using different loci. At the continental level, the five major races keep popping out.
Raff: ” Human biological variation is real and important….But Wade, and others who agree with him, have decided that certain patterns of variation—those which happen to support their predefined notions of what “races” must be—are more important than others.”
Well, this is what it’s all about them: what “we” should see as important. It’s funny that you don’t notice the asymmetry here. Wade and others don’t argue that seeing genetic diversity in terms of race is the only way of seeing it, but that it is one valid way. No one argues, for example, that “Only race exists!” (Please show me a counter example, if you can.) Rather, the anti-race crowd — which you apparently fall into — argue that “Only non-race exists!” i.e., “Race does not exist!” This asymmetry is parallel to that between the environmentalists — “Only environment!” — and hereditarianism — “Genes too!” But of course, since you KNOW that you are right, since you know that “Only non-race exists!” you can’t help but seeing seeing pluralism i.e., “Race. too” as an imposition.
But we can go beyond that. It so happen that in biology, natural populations are deemed to be more important that artificial ones — they are given quasi privileged status. Hence, our taxonomic systems organize life by either genealogical (cladistics) or geneotypic (evolutionary taxonomy) similarity, and not by similarity in so called arbitrary character traits. There is a reason for this, of course, and it’s that such categories allow for more inductive inferences. If my “genetic populations” are based on similarity in two genetic characters, I have less inductive leverage than if they are based on evolutionary relationship and consequently similarity in terms of the whole genetic program. So, Wade and others have sound biological precedence for saying that race (qua subspecific natural population) is a generally more important grouping than others. It’s a well founded prejudice, no? And of course, we can make a similar argument for continental races relative to micogeographic ones. After all, these are the types of races that zoologists themselves privilege when it comes to other species. So there we go…
Raff: “Folk notions of what constitutes a race and how many races exist are extremely variable and culturally specific.”
Which is why the primary/major/continental level races/varieties of Berneir (1688), Linnaeus (1735), Buffon (1749), Kant (1775), Blumenbach (1795), Cuvier (1828), Coon and Garn (1955), Gill (1988), Risch et al. (2003), and Rosenberg et al. (2011) look remarkably the same: Why didn’t any manage to classify North Africans with other Africans or to classify South East Asians with South Asians? Of course, what you mean is that sociological races — ones defined by sociologists — are extremely variable and culturally specific. Which is why you should be criticizing those (e.g., “Asians Americans”). Or possibly what you mean is that biological and bio-anthropological race is a flexible tool that allows for nested classifications.
………..
I don’t know. I could keep going but I can’t find anything worthwhile here.
Thanks.
For an example of how better and better scientific tools keep coming up with basically the same racial map of the world, here’s the cover of population geneticist Cavalli-Sforza’s 1994 magnum opus The History and Geography of Human Genes:
http://isteve.blogspot.com/2014/05/nyt-review-of-troublesome-inheritance.html
There’s another way to think about race, although it’s not typically available to naturalists: genealogically. Naturalists use phenotypic clues to infer genealogy, but horse racing fans don’t bother worrying about things like color because they have complete genealogies available on each thoroughbred going back almost 20 generations:
http://takimag.com/article/the_great_white_horse_steve_sailer/print#axzz32Zl2yI5h
“although it’s not typically available to naturalists”
Here’s an excerpt from one of Darwin’s letters to Huxley:
“Grant that all structure of each race of man were perfectly known — grant that a perfect table of descent of each race was perfectly known. — grant all this, & then do you not think that most would prefer as the best classification, a genealogical one, even if it did occasionally put one race not quite so near to another, as it would have stood, if allocated by structure alone. Generally, we may safely presume, that the [phenotypic] resemblance of races & their pedigrees would go together” [comment: when he says “even if..” I imagine that he’s referring to situation like negritos and negros.]
Darwin largely established the modern view of natural classification, that based on descent. Though, he argued that “modification” should be taken into account (e.g., because of genetic divergence, birds shouldn’t be lumped with crocodiles). The evolutionary taxonomists — following Mayr — more or less reunderstand natural classifications to be based on genotype. Genealogy is back fitted:
“Once we accept the basic principle of biological classification, that organism are to be classified according to the information content of their genetic program, it is evident that [retrospective] monophyly must be required. Artificial taxa, containing descendants of different ancestors, would be unable to fill the demand one places on scientific theory, owing to the heterogeneity of the included genetic programs”
This genotypic understanding — which currently has less currency than the cladistic = genealogical only one — brings the classification principle somewhat back in line with Aristotle’s and forward in line with population genetic thinking. Of course, on the level of the individual organism, the genotypic and genealogical conceptions — both of which can be said to be “genetic” –produce the same, or at least very similar, results. And, on the population level, genotypic and Darwin’s “descent plus modification” also effectively act the same.
Regarding pre-Darwinian conceptions of race, my impression was that they were genealogical, too. During this time, some argued that human groups represented different species (polygenists), but most argued that they represented races in the sense of subspecific lineages (monogenist). The concept of race itself — largely developed by Buffon — was based on lineage (as in race de noble) and breed — so it should be a given that “race” was and is somehow associated with ancestry. There’s a lot of confusion on this, though, because anthropologists and sociologists hurriedly dig through the literature intent on debunking something and, in their rush, miss important disagreements such as the race/species one. So, they will say that race and species was used interchangeably, when, in reality, there was an ongoing race/species debate This confusion leads them to make all sorts of silly claims and to miss the whole point of the race concept.
Thanks. I’d never seen the Darwin quote before.
Dear Dr. Raff:
Here’s Mr. Wade’s response to your Huffington Post piece:
http://www.huffingtonpost.com/nicholas-wade/in-defense-of-a-troublesome-inheritance_b_5413333.html
And a feeble response it is. After carefully explaining that he’s not a racist, (never a good sign if you have to explain it), Wade derides two of his critics as, “anthropologists who, to judge by their webpages, do little primary research.” Which, of course, doesn’t compare with Wade’s credentials, since he has been writing about the subject “for years in a major newspaper.” It’s all a conspiracy by the “social scientists,” to silence Wade’s views. And the two anthropologists, (who actually are biological anthropologists), “have a strong personal interest … in attempting to trash my book.” Wade is clearly flailing about. Perhaps the fact that no credible biologist has come to the defense of his book is beginning to weigh on him.
When I prepare clients for a negotiation or confrontation, I often tell them, “If you have to tell someone how honest or smart you are, they shouldn’t believe you.”
Your advice works assuming that nobody implied that Wade was a racist. Since this isn’t so, your comment is not only irrelevant, it’s shows poor reasoning on your part, and I’d never hire you as an attorney.
The advice is actually particularly relevant when the question has been raised. if someone tells one of my clients, “we don’t trust you!”, then I don’t want them to answer by saying, “but we’re very honest people.” I want them to demonstrate the attribute: verify their statements to show honesty, go into detail to show that your proposal is clever, etc. It’s nothing more than the adage that actions speak louder than words.
Whether that applies to Wade, you can decide for yourself. In the meanwhile, I’ll strike “James the Commenter” off of my prospective client sheet.
Steve Sailor you write: “Race is ancestry”.
Wade and his sycophants cling to race like barnacles cling to shipwrecks.
They believe they are redefining what race is, but they are actually redefining what racism looks like, since The Bell Curve participating in the new sordid chapter of racial pseudoscience.
Wade’s biological determinism is racial folk ideas dressed in scientific jargon. Why don’t you point this out instead of encouraging a platform for racists.
If “racism” merely means:
(a) recognizing human biological races & (b) recognizing that some of these congenitally differ (implicit modifier: on average) in socially valued traits
What could be wrong with it as such?
Perhaps you could define “racism” as you mean it and we can discuss whether or not it’s something that should, in fact, be discouraged — and we can discuss in what way “racism” as defined above is related to your conception.
What is the precise scientific definition of racism? Is “racism” a continuous or discrete variable? If the former, how can “racists” and “non-racists” exist as discrete groups? If the latter, what specific traits differentiate a “racist” form a “non-racist”? Does racism come in degrees (a little racist, very racist)? If so, which degree really counts as “racist”…..
Yup. Racism, like race, is a social phenomenon.
yeah thanks for telling a biologist her business, steve. by the way is jim goad still writing for takimag? i liked his love letter to domestic violence after he fractured his wife’s skull.
with even a superficial understanding of anything about your slimy HBD crew it’s a wonder any of you have the audacity to poke your heads out from under your rocks, nevermind act like you are scientific authorities.
A typical response from Steve’s blog “It’s not a big secret that Jennifer Raff, Augustin Fuentes and Jon Marks are all far-left Cultural Marxists. If these lightweights weren’t given megaphones by MSM [mainstream media??], they would simply be ignored by anyonee with an IQ above room temp.”
and
“I don’t know why you and Chuck wasted so much time on that Colin fellow. I wouldn’t have. But that’s probably his goal (and what he was trained to do as a lawyer) — to wear people down by the sheer mass of his senseless sophistry.
Maybe most people just aren’t wired to get it, and there’s no sense in trying to explain it to them. I’ve come to suspect that’s just the way things are, and that’s why people need religion to convince them to think and behave in an adaptive manner.
Study the Bible and it’s clear that it’s largely an exercise in applied eugenics justified by a higher power.”
etc etc
When you search Steve Sailer’s name through google Pat Buchanan, John Derbyshire (who was unceremoniously catapulted from already ‘tacitly’ racist National Review for being too racist as well as being a military rape apologist) and scummy “pick up artist” Roosh V pop up as related people. Clearly quite the authority on anthropology and human genetics.
Steve Sailer’s scuzzy homeboy and takimag contributor Jim Goad celebrating domestic violence http://www.jimgoad.net/pdf/violence/lets.pdf
The future’s looking bright for HDB. Hey Steve why don’t you and the other racists pretending to know more than actual scientists crawl back under your shared rock you ridiculous charlatan.
Ironically, taking a look at Agustin Fuente’s picture, it’s obvious that most of his ancestors were white, especially whites from the Mediterranean:
http://news.nd.edu/for-the-media/nd-experts/faculty/agustin-fuentes/
And your point is…?
I’ll take a stab at what his point is: that most human beings do quite well using their eyes to tell them a great deal about the world and its people, and can’t stand it when people (in this case, a certain type of “academic”) insults their intelligence.
I don’t agree with Wade’s point that such people do what they do in denying race because they fear racism. (Maybe he doesn’t believe it either; perhaps it’s his bone to them to keep the peace.) They do what they do out of ego, self-righteousness, a need to perceive themselves as powerful. After all, no one could ACTUALLY be so dumb as to believe all the creatures of the world are the same, that they evolved the same, that there aren’t difference that go beyond the color of skin or fur. They know that it doesn’t matter when orange becomes red or red becomes orange, but they damn well know the two are enough distinct to call them by different names.
And they can’t stand paying the salary of these people who would obfuscate or downright lie to their face or to those of their kids.
Most human beings feel like they do quite well using their eyes to tell them a great deal about the world and its people. The usefulness of science comes largely from the fact that people aren’t nearly as good at it as we think we are. The shameful history of racial laws in America bears witness to the fact that it’s quite difficult to reliably sort people based on how they look; it feels easier than it is. Not only can looks be deceiving, there’s no objective standard to apply.
But I’ll agree that many racialists are just searching for a justification for sorting people based on whether they look black, white, asian or other.
Steve Sailer your obsession with Agustin Fuentes’ appearance is a bit creepy. Your blog: “…Professor Fuentes, who looks like he should be playing goalie for Argentina’s World Cup team.”
and this blog:
“Ironically, taking a look at Agustin Fuente’s picture, it’s obvious that most of his ancestors were white, especially whites from the Mediterranean.”
http://isteve.blogspot.com.au/2014/05/nicholas-wade-defends-troublesome.html
Steve Sailer, is your point that the merits of Agustin Fuentes’ critique can be discounted on the basis of what you think his ancestry might be? Please clarify what you mean.
Agustin Fuentes benefits from affirmative action due to having a Spanish surname, but he himself is as white as most of the Latin American upper class, certainly whiter than “white Hispanic” George Zimmerman.
Agustin Fuentes benefits from affirmative action due to having a Spanish surname
Is that a fact that you know, or a rhetorically convenient assumption you made for ideological reasons?
I agree with your two critiques of Raff’s argument that “human genetic variation does not divide humans into a few discrete groups.”
You have to be clearer about what you mean by “Wade defensibly equates races with… ‘genetic populations.'” Race *is* a polyseme and Wade dithers about the appropriate concept of race, never explicitly grounding any interpretation of the word in population genetics. And if you read the latter chapters of his book, it’s not at all clear that Wade is dealing with “a typical biological race concept.”
Hi Ben,
One can tell that Wade associates races with “genetic populations” (in the retrospective sense i.e., natural populations) based on his discussions. First, in the chapter, “Races as Clusters of Variation”, where Wade clarifies his conception, we are told:
“A necessary approach to studying racial variation is to look not for absolute differences but at how the genomes of individuals throughout the world cluster together in terms of their genetic similarity. The result is that everyone ends up in the cluster with which they share the most variation in common.”
Wade, then, goes onto cites genetic clustering research:
“In one of these more sophisticated studies, a team led by Noah Rosenberg of the University of Southern California and Marcus Feldman of Stanford University looked at the number of repeats at 377 sites on the genome of more than 1,000 people around the world. When this many sites are examined on a genome, it’s possible to assign segments of an individual’s genome to different races if he or she has mixed ancestry. This is because each race or ethnicity has a characteristic number of repeats at each genomic site.”
That is, he cites population geneticists who refer to their groupings as “genetic clusters” or “genetic populations”. Elsewhere Wade tells us, for example:
“The classification of humans into five continental based races is perfectly reasonable and is supported by genome clustering studies”
And:
“One might suppose that races differ in having different alleles of various genes. But, though a handful of such racially defining alleles do exist, the basis of race rests largely on something even slighter, a difference in the relative commonness, or frequency, of alleles, a situation discussed further in the next chapter”
I agree that he didn’t explicitly state that his races are (or, at least, are a type of) population geneticists’ genetic clusters/populations — instead, he just says that they are genetic clusters — “races as Clusters of Variation” — and then references population genetic research on genetic clusters — but he makes this point clear enough. (Perhaps not enough, though.)
Perhaps it helps to have read his Times articles. For example, in a February 2014 one, he states:
“The genetic atlas of human mixing events was published on Thursday in the journal Science by a team led by Simon Myers of Oxford University, Garrett Hellenthal of University College London and Daniel Falush of the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany. Having sampled genomes from around the world, they found they could detect about 95 distinguishable populations.
Though all humans have the same set of genes, their genomes are studded with mutations, which are differences in the sequence of DNA units in the genome. These mutations occur in patterns because whole sets of mutations are passed down from parent to child and hence will be common in a particular population.
Based on these patterns, geneticists can scan a person’s genome and assign the ancestry of each segment to a particular race or population.
(Tracing Ancestry, Researchers Produce a Genetic Atlas of Human Mixing Events )”
I suspect that his vagueness e.g., races AS clusters of variation versus races ARE clusters was tactical. Races, in the sense of sub specific natural populations, are generic biological phenomena. One can investigate them, and so conceptualize them, from different perspectives; the population genetic one, which deals with sets of individuals, is just one. Another is the evolutionary perspective, which deals with, as the unit of analysis, populations. Wade approaches races from both angles. For example, in a chapter he says:
“Races are a way station on the path through which evolution generates new species. The environment keeps changing, and organisms will perish unless they adapt. In the course of adaptation, different variations of a species will emerge in conditions where the species faces different challenges. These variations, or races, are fluid, not fixed. If the selective pressure that brought them into being should disappear, they will merge back into the general gene pool.”
This is an evolutionary perspective, one which is characteristically vague when it comes to the question of what makes an individual a member of one race and not another; it is since from this perspective the unit of analysis is the (sub)population (an entity, which unlike the individual organism, can evolve.)
If you’re going to approach the phenomena from multiple angles, then it’s best to not to over-specify the definition/ operationalization.
It’s only natural that a lot of people have problems thinking about complexity. All we can really ask is that they not be so smug about their obtuseness.
Hi Steve. You have a lot of questions outstanding on this thread. I know they’re difficult, but would you mind trying to answer them?
This was a particularly ditzy part:
Raff: (Also see Templeton AR, 2013. Biological races in humans. Studies in History and Philosophy of Biological and Biomedical Sciences, http://dx.doi.org/10.1016/j.shpsc.2013.04.010)…Wade even seems to agree with population geneticists that there aren’t any races, just clinal distributions of genetic diversity: “Because there is no clear dividing line, there are no distinct races—that is the nature of variation within a species.” (p 92).”
I commented on Templeton before. His argument is: (a) “biological race” should refer to zoological subspecies, (b) the criteria for being a zoological subspecies is having a Fst value greater that 0.25; (c) the Fst value between human populations is typically lower that 0.25. Of course, his Fst criteria, which makes little sense, was contrived – the paper he cites for this rule doesn’t discuss it — but rather discusses a multivariate version of the old 75% rule by which there would be human zoological subspecies. But Templeton keeps repeating it, so some people take it seriously More significantly, his argument rests on a word game. Races, we are told, are really “geographic races” — a term which Mayr used synonymously with zoological subspecies (i.e. formally recognized zoological races) and the subspecific natural populations that don’t make the conventional formal recognition cutoff are something else. Stop think. Of course, not accounted for are the evolutionary taxonomist’s microgeographic races and, more generally, non-formally recognized races i.e., the things that zoological subspecies are aside from being formally recognized. The whole argument is really silly — but at least — and this is the point — Templeton recognized the biological validity of at least one biological race concept. When one does, barring extreme disingenuity, one is forced to recognize that race is a fact of evolution; when you don’t you are left with conceptual schizophrenia e.g., ‘race isn’t a valid concept, therefore there are no human races like those of the Plain Zebra.’
And then we have the “just clinal distributions of genetic diversity” part. Cline once referred to character clines. Huxely (1939). “Some special term seems desirable to direct attention to variation within groups, and I propose the word cline, meaning a graduation in measurable characters…Clines may be of inter- or intra-group nature. Inter-group clines connect the mean values of the subspecies of a polytypic species (or of the species of a geographical subgenus or Artenkreis)”. The concept has since been extended — by whom? — to include population continua. But a character cline and a population continuum are very different. For example, an indigenous pigment character cline cuts across the southern hemisphere, yet a population continuum doesn’t. Generally, the presence of a population continuum isn’t inconsistent with that of biological races, it isn’t even inconsistent with the formal recognition of these: Quote: “The population continuum is that part of the species’ range where there is continuity of contact among local populations, some of which may be recognized as subspecies if sufficiently differentiated. (Gorilla biology: a multidisciplinary perspective”); see also: Mayr and Ashlock (1969;1991). The point is that populations carved out of a population continuum are still natural ones; ones carved out of a character cline (that doesn’t perfectly overlap with a population continuum) would constitute an artificial biological population and, thus, neither evolutionary, population genetic, nor cladistic races.
…
Whatever.
You’ve called this piece “ditzy” a couple of times now, but even your complaints make it clear that the post is engaging the evidence on a thoughtful, expert level. Do you think that if you throw another few gendered insults into the mix, you’ll succeed in getting under Dr. Raff’s skin? I think it’s unlikely; much nastier people failed to do so despite writing much more creative attacks in the vaccine threads.
The issue is that Dr. Raff is smart enough to know better. Statements like, “The differences between groups are so subtle and gradual that no objective lines can be drawn, so Wade draws his own on the basis of his own preconceptions” are silly. The Atlantic Ocean, for example, served as a highly objective line between races.
The Atlantic Ocean separates my family from our ancestors in Ireland and/or Scotland. Are we different races?
You’ll notice that zoos don’t define subspecies by present geographic location, either; you’re not the only animal dislocated form its ancestral geographic range. But what’s particularly odd about your comment is that the race concept was developed in part to explain why varieties of the same species retained their characteristic traits after relocation, that is, why some varieties are constant.
“Quite in line with this, the problem that Linnaeus saw in the well known case of “Ethiopians” was not to explain that and how their skin colour persisted, but rather that they possessed a distinctive character – black colour – that remained constant even under varying climatic and geographic conditions, while they nevertheless doubtlessly belonged to the same species as other humans: “Who would deny that the Ethiopian is of the same species as we humans”, as it says in a paragraph discussing difficulties in distinguishing varieties from species in the Critica botanica (1737), “and yet the Ethiopian brings forth black children in our countries.” (Cabbage, Tulips, Ethiopians–“Experiments” in early modern heredity.)
“How does Buffon reconcile such a statement with his theory of the production of races through the influence of climates and life conditions? This question is not very difficult to answer, but it is important concerning the very concept of “race”. Climate and life conditions act over time. They are transmitted through generations and inscribed in the
body through genealogy. For instance, «the germ of blackness is transmitted to children by their fathers and mothers so that in any country where a Negro may be born, he will be as black as if he were born in his own country.» (p. 523) Over time, relatively “constant races” are created this way. It means that, according to Buffon, history, kinships and transmission of characters over generations creates relatively constant varieties transmitted over time. (Race and Genealogy: Buffon and the Formation of the Concept of ‘Race’.)
Why do e.g., South Asians in Europe (Gypsies) look, in pigment, like South Asians in South Asia was a very old question. And one of the answers was “race” — more specially that they descend from a common lineage which somehow acquired characters that were somehow genealogically passed on.
Sailer claimed “the Atlantic Ocean … served as a highly objective line between races.” This doesn’t even fit his own idiosyncratic definition of race, but neither is it an objective discriminator under what appears to be yours. If a species is divided into subspecies on the basis of their geographic separation, then removing the geographic separation would also collapse the taxonomic separation.
“If a species is divided into subspecies on the basis of their geographic separation”
I imagine that you’re just being silly — but I don’t know. A species is divided into subspecies on the basis of genetic separation. In most species, this genetic separation results from decreased interbreeding due to geographic separation. This is why in zoos, the trinomen of an animal doesn’t change with relocation.
You may imagine whatever you wish. I’m no taxonomist, but my understanding is that subspecies (and even species) can be defined by their geographic isolation.
Zoos don’t change their animals’ designation because they aren’t considered to be native species of the zoo’s region.
” my understanding is that subspecies (and even species) can be defined by their geographic isolation.”
Here’s a quote from the paper:
“Once EVOLUTION was accepted, it became clear that variation at all levels in the taxonomic hierarchy was due to more or less SIMILAR causes; THE ONLY DIFFERENCE between variation above the level of genus or species and below was one of degree. Darwin realized that species could EVOLVE from intraspecific varieties. Darwin used the term ‘‘species’’ in a new and nonessentialist sense: ‘‘… the complete absence, in a well-investigated region, of varieties linking together two closely-allied forms, is probably the most important of all the criterions of their specific distinctness … Geographical distribution is often brought into play unconsciously and sometimes consciously; so that forms living in two widely separated areas, in which most of the other inhabitants are specifically distinct, are themselves usually ‘ looked at as distinct; but in truth this affords no aid in distinguishing geographical races from so-called good or true species’’ (Darwin, 1874). Darwin showed convincingly that there was NO ESSENTIAL DIFFERENCE between species and ‘‘varieties’’; species were simply varieties which had DIVERGED more, and which could coexist without intermediates being common.” (Emphasis added)
Terms like “evolve” and “diverge” signify genetic differences — no one uses “evolve” as a synonym for “migrate”.
“Diverge” in the sentence, “species were simply varieties which had DIVERGED more” refers to genetic, not geographical divergence. (Of, course, the genetic divergence (most often) results for geographical separation.)
Mallet doesn’t spell this out because it’s common biological sense.
“You may imagine whatever you wish. I’m no taxonomist, but my understanding is that subspecies (and even species) can be defined by their geographic isolation”
Find me a biologist who defines taxa in terms of geographic differences in absence of genetic ones.
Do you really believe that question makes sense sense in this context? Just curious.
It’s rhetorical. The objective line Sailer proposed doesn’t seem to actually be an objective line between races; it doesn’t divide me and my ancestors into two separate races. So what would? I keep looking for something other than whimsy that distinguishes one race from another, and I’m not seeing it.
Colin (2014) “So what would?”
Descent plus modification. Think “species”. Add “sub”. Work down from there.
That doesn’t answer the question. How does it apply? How do you objectively, in practice, determine a person’s race or distinguish between marginal cases?
“Racial clusters” is the language I will be using for go forward and encouraging others to use. If you want to focus on group traits from clusters along these very REAL these gradients, I see no reason why you should not. Others who simply disagree with you are going to use different approaches, definitions and conceptual frames. I know of no reason science must be approached the same way by all. Do you? Serious question. It’s not my intent to start a tense, unpleasant or sarcastic exchange with you, Colin. You seem like you have some substantial familiarity with the issues, and I’d like to learn some points from you I haven’t seen 1000x.
“How do you objectively, in practice, determine a person’s race or distinguish between marginal cases?”
Something like:
(a) start with a biological race concept e.g, population genetic
(b) operationalize it — that is, decide what exactly it is that makes an individual a member of one genetic division (i.e., one race) as opposed to another e.g., pairwise genetic similarity
(c) pick a grain of resolution i.e., determine how many genetic divisions you are interested in looking at
(d) decide if you want fuzzy or discrete sets
(e) obtain a reference sample e.g., hapmap.ncbi.nlm.nih.gov/
(f) categorize these individuals into your pre-selected number of natural population divisions + a nonassignable onw
(g) obtain a sample from your particular individual
(h) compare (f) to (g) — if you are working with discrete sets, the “marginal cases” may fall into the nonassignable group; from a zoological perspective, they fall into a zone of intergradation; however, if you are working with fuzzy sets you could give them values describing their relation to the defined divisions — basically, percent ancestry .
dsp – how do you define “racial clusters”? Do you mean clusters of different races? That doesn’t solve the problem of determining what a race is or who belongs to it. Do you mean clusteredness, as used in Structure studies? I don’t understand the concept all that well, but what I do know doesn’t explain how it could possibly be used to define “racial clusters.”
I think your response illustrates the exact kind of fuzziness and subjectivity we would expect to see in a discussion of a culturally, not empirically, defined phenomenon.
Finally, you probably shouldn’t seek to learn much from me. I’m a layperson, with opinions but no expertise in this field.
Chuck,
(a) assumes your conclusion–that you can define and apply such a concept. I haven’t seen it done.
(b) assumes your conclusion–that you can operationalize it. What I’m asking is how you accomplish this.
(c)-(h) are handwaving based on the assumptions in (a) and (b).
Let’s go back to my question: “How does it apply? How do you objectively, in practice, determine a person’s race or distinguish between marginal cases?”
“Start with a biological race concept” and “operationalize it” are vague descriptions of what an answer might conceivably look like on a conceptual scale, not actual answers to the question.
The fact that “race realists” can’t objectively define or apply “a biological race concept” is a strong indication that there isn’t a satisfactory definition to be found, much less operationalized. See, for example, Sailer’s frenetic insistence (at odds with his fellow travelers) that race is an extended family with inbreeding, and his inability to address, much less answer, the immediately obvious question of how one defines the extent of those families.
“Let’s go back to my question: “How does it apply? How do you objectively, in practice, determine a person’s race or distinguish between marginal cases?”
Must you be spoon fed? For example, here is a population genetic operationalization:
“sets of (subspecific) population whose members are more overall genetically dissimilar to members of other sets than they are to other members of their own set”
This is based on Hartl and Clark’s
http://www.amazon.com/Principles-Population-Genetics-Fourth-Edition/dp/0878933085
Here is another:
“Two individuals, I1 and I2, belong to the same genetic population if (a) their genetic relationship, measured with the coefficient of kinship, is greater than zero and (b) their kinship is much higher than kinship between them and some individual I3, which is said to belong to another genetic population.”
This is based on Aulchenko (2010)
books.google.com/books?hl=en&lr=&id=jD8ST0_kwakC&oi=fnd&pg=PA123&dq=%22Effects+of+population+structure+in+genome-wide+association%22&ots=h2bmNwRj-J&sig=Qo2wqG2I9DHVzvlkVF6hRhWSB9o#v=onepage&q=%22Effects%20of%20population%20structure%20in%20genome-wide%20association%22&f=false
As I mentioned a number of times, the specific operationalizations differ since there are different ways to define “genetic similarity”.
Now go through steps c-h.
Your position here assumes that population and race are equivalents; this rules out many of the tentative race definitions commenters here have offered. Are you taking the position that “race” is synonymous with “population”?
I know it’s not something that you want to hear, but please consider that your comments are not adequate support for your arrogant and condescending tone.
Hi Colin
re: “Your position here assumes that population and race are equivalents”
I originally said:
“(a) start with a biological race concept e.g, population genetic
(b) operationalize it — that is, decide what exactly it is that makes an individual a member of one genetic division (i.e., one race) as opposed to another e.g., pairwise genetic similarity”
In response, you said: [(a) and (b)] are vague descriptions of what an answer might conceivably look like”. And so, I offered a very precise operationalization. Yet, in reply, you wrote: “Your position here assumes that population and race are equivalents”. But my position assumes no such thing, since I originally said “start with a biological race concept e.g, population genetic”. To spell things out with big goofy crayons — since, apparently, I must — I didn’t say that a population genetic race concept was THE biological race concept, hence I used “e.g [exempli gratia] population genetic” not “i.e. [id est], population genetic”.
Recall that in one of my initial comments I said: “I can think of four prominent biological race concepts off of the top of my head: cladistic, evolutionary, population, and ecotypic. They are biological races seen from the perspective of different biological research programs.” I clearly recognize that there are multiple related biological races concepts. I noted also that most can be subsumed under a natural population conception of race. Compare this to the situation with species: de Queiroz, K. (1998). The general lineage concept of species, species criteria, and the process of speciation.
….
It sounds like we agree, then, that your specific operationalization rules out many of the possible definitions of race. It also sounds as if we agree that race is in practice an utterly subjective categorization: decide what definition you want to use, decide how granular you want your races to be, decide how many races you want and pick referents based on those assumptions. The result is a conceptualization of “race” that’s whatever you want it to be, producing whatever results best suit your ultimate goals.
“The result is a conceptualization of “race” that’s whatever you want it to be, producing whatever results best suit your ultimate goals”
When I specify a concept you accuse me of being too arbitrary (“Why not some other concept?”); when I generalize you accuse me of being too vague (“what does this mean specifically?”). Either you are being disingenuous or fuzzy-headed.
If the latter, to better understand, first distinguish between words and concepts. In and of themselves words means nothing; they gain meaning only through reference to concepts. Thus we need to focus on concepts. Second, recognize the nested nature of concepts: general to specific. For example:
a.Generic race (GR). Races as populations some how defined in terms of ancestry, real or imagined.
b.Biological race broadly constructed (BRBC). Races as populations defined in terms of biological ancestry.
c.Natural biological race (NBR). Races as populations defined in terms of overall biological ancestry.
d.Population genetic race (PGR). Races as populations defined in terms of overall biological ancestry understood from a population genetic framework..
f.Pairwise population genetic race (PPGR). Races as populations defined in terms of overall biological ancestry understood from a population genetic framework and operationalized in terms of pairwise similarity.
Concept (f) is nested within concept (a). Because it is more specific it is necessarily more exclusive and less general.
You argue that (f) is arbitrary because it’s a subtype of all possible concepts called “race”. (“But why choose this concept?”). This makes zero sense, You are now applying the goofy grain of resolution critique to the very concept of race; both applications fail for the same reason: biological racialists don’t claim that only one conception of race more generally understood — as in (a) — is true; they only claim that some biological concepts — b-f — are valid and that by these there are human biological races.
If someone here can show that by concepts b to f there are no human races or that some of the human races which are by these concepts don’t reasonably correspond to common sense once-said-to-be-biological races (e.g, continental races) or that these concepts themselves can not, given historical term usage, justifiably be called race concepts, I will try to eat my keyboard. If you can’t show this,what are you arguing against? If you are arguing that human races don’t exist because there is (a) no true number of them or (b) no true concept of them, please justify this requirement given the typical historic usage of the term “race” or given present biological race conceptions. If you are arguing that human races don’t exist because they are not discontinuous enough or differentiated enough or something else, please again justify these requirement in the same manner.
…
The pyrrhonism is simply overwhelming.
As an attorney, I am extremely comfortable within the realm of abstract hierarchies. My criticism of your position is, ultimately, no more than that you wish to ground culturally contrived concepts of “race” in the trappings of empirical biology. The arbitrariness of your racial categories, which depend entirely on your preconceptions and return results matching those preconceptions, gives me little confidence in the rigor of your methods. Your extraordinary arrogance also wears away at your credibility; I have often observed angry, marginalized amateurs pounding the keyboard to marginalize and demonize all who disagree with them, but I rarely see it in actual experts.
Having said that, is it possible that you’re right? It’s almost certain, because your central point appears to be that you can construct a definition of race that does whatever work you want it to, apart from generate new and significant insights into objective reality. I agree that you can create an argument supporting the proposition that N angels can fit on the head of a pin, especially since you can define N as you please.
As to your specific comments, you misunderstand me. Likely because my own comments have been unclear, so I apologize. I do not “argue that (f) is arbitrary because it’s a subtype of all possible concepts called ‘race’.” I argue that it is arbitrary because there is no objective reason to prefer it as a definition of “race” over a, b, c, or d.
You assert that “racialists don’t claim that only one conception of race … is true,” but that’s objectively false. I can think of at least two “racialists” in this very thread asserting one, and only one, definition of “race.” Surely they would agree, as we all would, that other definitions are possible, but they are promoting their own favored articulation in order to resolve the ambiguity that eats out the heart of the idea that human diversity is best understood as a function of race.
You also seem to believe that we are contending that there is no such thing as a “race concept.” That’s absurd, if it is your belief. Obviously there are lots of beliefs about race. My argument is that those beliefs are founded in little more than the proponents’ desire to carve humanity up into divisions that suit their preconceptions; its little more than “some people are black and some people are white and some people are yellow.” Even as a layman I’m aware that human variation is vastly more complex than that, but the “racialists” are intent on creating buckets based on convenience rather than objective reality. I think Steve Sailer is a good example of this, having adopted a transparently subjective definition of “race” with no hope of pegging its divisions to anything other than his own whimsy. Your efforts seem more tied to actual variations in human populations, albeit no more effective in practice (from what I can tell).
In other words, I do contend there is “no true number of [races].” (I can’t say there’s no “true concept” of them, because the truth of a concept is such an ambiguous quality; I’m perfectly comfortable saying an utterly subjective division is “true,” depending on the truth criteria set by the believer.) There are any number of races, depending on how you want to define the term. The argument that there are three, or five, or seven, doesn’t reflect a significant biological distinction nearly as much as it reflects culturally-determined preconceptions. Those preconceptions are real, and “racialists” seem to be attempting to import that reality into a more substantive biological context.
@Colin, as a lawyer, you should be able to distinguish between categories that are fuzzy in their application sometimes and pure subjectivity. Even hard-edge rules have border cases sometimes. IF they are important enough to result in litigation, we get an authoritative ruling, but it could be completely arbitrary which side of the line we draw.
Science, at least life science, requires concepts that have fuzzy borders too. How many ecosystems are there on earth? Phyla? Taxonomy resolves some of these disputes through consensus (from the loser’s point-of-view, arbitrarily).
Folk conceptions of race matter for medical diagnosis and treatment. That really should be enough.
The important point is that racial equality does not depend on, nor is it threatened by, any developments in taxonomy or population genetics. It is a moral commitment, so it doesn’t really matter what empirical facts are discovered. The Canadian Constitution also requires equality without discrimination based on disability, and I would argue that is just as essential to a democratic society as racial equality. There is a sense in which disability absolutely is socially constructed. But of course disabilities can be rooted in genetics. Some disabilities (cystic fibrosis or sickle-cell anemia) are almost exclusively found in particular races. Those are important empirical facts, but they have no implication for legal or political equality at all. (At least, they don’t imply lack of a right to equality. It may be true that disadvantaged minorities are less likely to get appropriate medical treatment.)
Pithlord, I’m not sure why you think I’m opposed to subjectivity as a rule. I’m not so uneducated in science that I’m not aware that taxonomy often relies on subjective distinctions. I’m dubious that humans can be categorized in the same way as other species, given our rather extreme proclivities to interbreed despite geographic barriers. Insofar as there is a group of committed “racialists” who are deeply committed to drawing racial lines and calling them the result of science, I’m skeptical of their motives. I see too much consistency with plain old fashioned racism and race obsession; see Steve Sailer’s recent comments complaining that the California shootings were the result of anti-blonde racism. (I don’t mean to imply that all “racialists,” or Steve Sailer in particular, are racists; only that the tone and tenor of their crusade makes me skeptical of their objectivity.) So when they draw arbitrary lines call it science, I think it’s important to point out that it’s merely a new framework imposed on the same old folk concepts of race.
Of course I could be wrong–perhaps geneticists will one day come to a consensus that there is a way to divide humans into races that makes biological sense independent of our cultural preconceptions. But it doesn’t appear that day has come, nor do I believe that “racialist” bloggers are hastening it.
As for your comment that “folk conceptions of race matter for medical diagnosis and treatment,” does that have any significance other than that folk conceptions of race are useful for making guesses about actual ancestry? (See my more substantive comment on that point in this thread as well.) That’s not the same thing as saying that those folk concepts are well grounded in biological realities. For example, wealth is a cultural concept. A doctor might look at the quality of my shoes and decide that I’m rich enough to be eating a high-cholesterol diet, and order tests accordingly. The cultural construct of material wealth was useful to that doctor, but doesn’t support a division of humans into “rich” and “poor” genetic races.
Colin,
I am not complaining that you are opposed to subjectivity. Rather, you seem to confuse the “subjective” with any fuzzy concept that might require judgment. I would think a lawyer would see the difference. If I like strawberry ice cream, that’s just subjectivity. If I think the rule “No vehicles in the park” applies to cars and not to poodles, but might apply to bicycles, that’s not subjectivity in any interesting sense.
You make an empirical claim when you refer to “our extreme proclivities to interbreed despite geographic barriers”, and it is an obviously overblown one. It’s true that some humans will mate with pretty much anyone in the right circumstances. But those right circumstances may be hard to come by, which is why humans are homogamous (tend to mate with those like them). Before Colombus, it just wasn’t possible for a European peasant to fall in love with an Inuit hunter gatherer. Since Colombus, we are in a slow-motion admixture event, but it has a way to go yet.
You are correct to suspect the motives of the “HBDers.” But motives don’t resolve scientific issues. Better to insist that the norm of human equality is independent of any findings in population genetics.
My own unprofessional view is that while there are clearly racially-based genetic differences in appearance and disease resistance/susceptbility, I don’t think there is empirical evidence for genetically-based differences in intelligence, since IQ rises by a standard deviation a generation. It is more likely that disadvantage leads to lower IQs than that lower IQs lead to disadvantage. I also don’t really see any reason to think there would be geographically different selection pressures for any kind of general cognitive ability: as Diamond points out, hunter gatherers probably have more cognitive demands on them than farmers. That doesn’t rule out other evolutionary mechanisms of founder’s effects, drift and so on, but in the absence of evidence, I would think we would go with the null hypothesis.
However, even if it were true that there is a genetic basis for “g”, and this is differently distributed among racially-defined populations, it shouldn’t have any legal or political implications, unless you think more intelligent people should have more rights than less intelligent people. I actually think that kind of bias is just as invidious as racism, and undoubtedly is more socially acceptable.
If I gave you the impression that I don’t distinguish between “fuzzy concepts” and subjectivity then I apologize. I do, or at least I think I do. Subjectivity comes into play as one method (and not the only one) used to impose clear boundaries on fuzzy concepts. Your example is itself a bit fuzzy, though; whether a bicycle is a vehicle is itself a fuzzy concept, resolved (in the absence of an objective standard, such as a regulatory clarification or precedent) through subjective reasoning. A court might answer the question by asking, for instance, whether a bicyclist is more like a pedestrian than like a motorist; the answer to that question is necessarily subjective, isn’t it? Even if the decider builds his opinion on objective factors, it’s ultimately still an opinion. But that’s neither here nor there, as I suspect our discussion here is more based on mutual misunderstanding than a fundamental disagreement about what subjectivity is.
I’ll take your word for the slow-motion nature of the current admixture event. My impression is that taxonomists, anthropologists, biologists and other scientists generally don’t consider humans to be more than one species, or to be divisible into subspecies; I’m also under the impression that there is a vocal minority consisting largely but not entirely of “racialists” who disagree. Do you think that’s an accurate summary?
I absolutely agree that motives don’t resolve scientific issues. As I have to look to experts to interpret data for me, I make credibility determinations about which expert to listen to. Motives are relevant to that process, and not dispositive of the underlying issues. And I further agree, enthusiastically, “that the norm of human equality is independent of any findings in population genetics.” I like the norm as well as the way you’ve phrased it. I don’t believe my opinions about race are predicated on my desire to defend that norm, but of course my own opinions about my own opinions are fallible (as well as subjective and recursive).
I’m curious, do you have an example in mind of a right preferentially provided to people with higher intelligence? I’m not sure if you’re describing a hypothetical problem or a current one.
“Having said that, is it possible that you’re right? It’s almost certain, because your central point appears to be that you can construct a definition of race that does whatever work you want it to, apart from generate new and significant insights into objective reality….”
No, my point is that given concepts and operationalizations that biologists have already constructed to understand genetic diversity in non-human species, there are humans “biological races” by which I mean biologists’ races. Let’s take a definition from a dictionary of zoology:
“An interbreeding group of individuals, all of whom are genetically distinct from the members of other such groups of the same species. Usually these groups are geographically isolated from one another, so there are barriers to gene flow”. (Allaby, 2010. A Dictionary of Zoology)
http://www.oxfordreference.com/view/10.1093/acref/9780199233410.001.0001/acref-9780199233410-e-7338).
This is nearly identical to the population genetic race definition which I offered above. Clearly, I’m not doing to definition construction; I’m simply applying biologists’ definitions (and doing a little exegesis on the side when these are conceptually vague). Now, I agree that race concepts are constructs; epistemologically, I’m a conventionalist, as such, I see most concepts as constructs. My point is that given biologists’ definitions there are human races. Do you agree or not? If so I will address some of your other points.
That definition raises some obvious questions that you seem to have overlooked:
– How is that definition different from the definition of “population,” as applied to humans?
– How distinct does a population have to be among animals to be “genetically distinct from the members of other such groups”? Are the races you’re trying to define so distinct? (Don’t humans only differ by about 1%?)
– Are the current geographical barriers to gene flow, if any, significant under this standard given human breeding habits?
All you’ve done is identify a measure by which taxonomists can distinguish groups in animals, and slapped the word “race” on it. If you’re trying to connect this to the folk definitions of race, you’ve also skipped the step of showing that this measure maps to those groups (if you can even define them). If you’re not connecting it to folk definitions of race, then you’ve skipped the step of showing that these measures are applicable to human beings.
On a more conventional level, obviously most biologists don’t agree that “given biologists’ definitions there are human races.” Why not?
“All you’ve done is identify a measure by which taxonomists can distinguish groups in animals, and slapped the word “race” on it.”
I cited a dictionary of zoology’s definition of “race”. Thus I am offering a referential account: accordingly, a biological race is what such and such biologists say it is. Do you agree that there are human biological races by this particular definition? As for correspondence, I have already shown that some “folk race classifications” such as classic continental ones qualify as races by this definition. Of course, your statement ” If you’re not connecting” is a non sequitur, since the said concept could allow for (e.g, bio-medically) useful classifications which are orthogonal to “folk” classifications.
[By the way, where is your evidence for this: “obviously most biologists don’t agree that “given biologists’ definitions there are human races.” I’m only aware of three surveys concerning biologists’ opinions, one from 1983, n=147 (73% yes, 30% no), one from 2001, n=22 (45% yes, 55% no), and one from 2003 n= 56 (53% yes, 30% no) (Lieberman et al. 1992; Morning, 2011; Strkalj, 2008); has some new data been released?]
Yes, you cited a zoological definition of race. I don’t believe that it’s intended to apply to humans, which appears to be an assumption you’re making. Even if we assume arguendo that it is applicable to humans, I can’t agree that there are races by that definition; see the questions I set out in response to your earlier citation.
Thank you for the data regarding biologists’ opinions on race; I wasn’t aware of it. I’ll revise my assumption that most biologists believe there isn’t. Why do you think those biologists who disagree with you, disagree with you?
“Yes, you cited a zoological definition of race. I don’t believe that it’s intended to apply to humans.”
At what point in human evolution do you imagine that humans ceased to be animals such that this pan species definition could not, in principle, apply to them? If you think that it doesn’t in practice explain why.
….
– How is that definition different from the definition of “population,” as applied to humans?
In biology, the term “population” is generic. It can, for example, refer to the concept of breeding population(which is defined in terms of probability of sharing descendents). Generally, “race” specifies a particular type of biological population. Some, of course, use the term “population” as a euphemism for the term “race” to refer to the type of populations that “race” frequently refers to i.e., natural populations. But this all gets confusing.
– How distinct does a population have to be among animals to be “genetically distinct from the members of other such groups”?
Not much. I cited tal 2012 above: “The probability that a random pair of individuals from the same population is more genetically dissimilar than a random pair from distinct populations is primarily dependent on the number of informative polymorphic loci across genomes from the total population pool. This probability asymptotically approaches zero with a sufficiently large number of informative loci, even in the case of close or admixed population.” I’m 90% sure that “distinct” in this definition means genetically differentiable — in the sense that members of populations don’t overlap in N-dimensional genetic space.
– Are the current geographical barriers to gene flow, if any, significant under this standard given human breeding habits?
I don’t see why this would matter. Imagine perfect geographic genetic homogeneity tomorrow; we would still need to describe past (and future) variation.
“At what point in human evolution do you imagine that humans ceased to be animals such that this pan species definition could not, in principle, apply to them? If you think that it doesn’t in practice explain why.”
I’d be curious to hear from an actual expert, but that definition seems to make assumptions that don’t apply to humans—the most obvious is the assumption that geographic barriers are a significant barrier to reproduction, which isn’t true (or is true in a vastly different way) in humans. The fact that humans are animals doesn’t mean that terminology describing non-humans can be cleanly applied to humans. Whether it can in this case seems to be something neither of us can speak to with certainty; you’re assuming so, but frankly (and obviously) I’m deeply skeptical of the rationales developed by an amateur race scientist and self-described “racialist.”
With regard to your citation to tal 2012, aren’t you equivocating between races and populations? You claim that population is a “generic” term in biology, but I don’t think that’s true. I think that biologists can define the term quite precisely as they apply it. The piece you cited explains that populations are detectibly distinct from one another. You’re taking that position and dropping it into a zoological definition of “race.” I’m not satisfied that’s sound reasoning, although I don’t for a moment imagine that you’ll agree.
“Imagine perfect geographic genetic homogeneity tomorrow; we would still need to describe past (and future) variation.”
It seems to me that anthropologists and geneticists are perfectly able to describe past and future variation using rigorous, defined terms and sound reasoning—they don’t fall back on vague notions of race to do their work. What need do you see your racialist theories filling?
“I’d be curious to hear from an actual expert, but that definition seems to make assumptions that don’t apply to humans—the most obvious is the assumption that geographic barriers are a significant barrier to reproduction, which isn’t true (or is true in a vastly different way) in humans.”
You’re either arguing:
(1) “I doubt that the members of such and such human populations are in fact more overall related to each other. How could they be so, given the history of human gene flow? As such, I doubt that such and such human populations meet the explicit requirements of this definition.”
This type of argument might have worked in the 80s when the genetic evidence wasn’t plentiful enough; but it’s untenable now.
Or:
(2) “While human populations clearly meet the explicit requirements of this definition, I suspect that the authors had in mind others — for example, that between populations there were “significant” barriers to reproduction, ones large enough not just to ensure that members of populations met the stated requirement, but large enough to ensure that these populations differed by quite a bit. I doubt that human populations would meet this intended requirement.”
The author offers some examples such as the island races of the St Kilda wren … but I was unable to find data on genetic differentiation. Before I spend more time looking, verify that you mean (2). In the meantime, we can look at related definitions. For example:
“In population genetics, a race is a group of organisms in a species that are genetically more similar to each other than they are to the members of other such groups. Populations that have undergone some degree of genetic differentiation as measured by, for example, Fst, therefore qualify as races.”
Removing all ambiguity, the authors continue:
“Using this definition, the human population contains many races. Each Yanomama village represents, in a certain sense, a separate “races”, and the Yanomama as a whole also form a distinct “race”. Such fine distinctions are rarely useful, however. It is usually more convenient to group populations into larger units that still qualify as races in the definition given. These larger units often coincide with races based on physical characteristics such as skin color, hair texture, and body conformation. Contemporary anthropologists tend to avoid “race” as a descriptive term for human groups because culture and linguistic differences, which are important, are often discordant with genetic differences and sometimes discordant with each other.” (Hartl and Clark. Principles of Population Genetics, Third Edition. 1997. Page 121-122.
I probably should have just cited this one from the beginning — since it’s so close to the one I originally offered and since it’s so very explicit.
So what is it that you’re objecting to?
These definitions, that is, the ones by which there are human biological races, can’t be too marginal since otherwise few biologists would accept the existence of human races — yet, as the survey data shows, many do.
You might argue that there are just too many definitions, rendering the issue somewhat subjective. That’s not untrue — by my argument was merely that there are human biological races given some common biological definitions.
Speaking of which, I came across the following discussion:
” But some, like Jody Hey, think that species do not exist except in the minds of biologists and their public. So for them, zero.
Final score: 26-27, 7, 2, 1 or 0.
What to think? My solution is this:
There is one species concept (and it refers to real species).
There are two explanations of why real species are species (see my microbial paper, 2007): ecological adaptation and reproductive reach.
There are seven distinct definitions of “species”, and 27 variations and mixtures.
And there are n+1 definitions of “species” in a room of n biologists.””
(Wilkins, J. (2010). How many species concepts are there. The Guardian, London.)
As with species, so with subspecies and race. You have one major biological concept i.e., intra species natural populations, four major research programs (ecological, evolutionary, cladistic, population genetic), at least four major race concepts that fall in line with these programs, and scores of definitions or sub-concepts. Because there are so many definitions, and because human populations are so race-like, it’s easy for me to find ones by which there are human biological races. This might prove your point that there is much subjectivity involved; but it also proves mine that the definitions are biological and that I — an amateur race scientist –am not the one dreaming them up.
Once you agree with this, I will address your semiotic attack —
“aren’t you equivocating between races and populations? …What need do you see your racialist theories filling”
This is just more hair splitting, heading down what I suspect is an infinite and circular path. I’m going to draw a line under it, since I suspect we’ve both read (and written) enough arguments to know that “You agree with X and then I’ll address Y” is a symptom of a conversation that is well and truly bogged down.
Let’s try to break out of it. My point isn’t that there is “much subjectivity involved,” although that’s very close. My point is rather that “race” is a subjective concept without nearly as much objective grounding as “species,” to the extent that there is virtually no practical purpose to defining or focusing on race as a concept. That is, lots of useful and valid theories have “much subjectivity involved.” But they also have something else that makes them more than just angels-on-the-head-of-a-pin sophistry. Further, may of the various race theories smack of cargo cult science to me: racialists want to put their folk notions of race on a firmer footing, so they snatch up the tools of anthropology, pick the K value that most closely matches their desired outcome, and start crying persecution when scientists balk at the process.
It sounds as if your point is that although racialists can’t find an objective definition of race, the various definitions they do have are grounded in biology. Am I right? I don’t intend to put words in your mouth.
Saying “the definitions are biological” isn’t saying much. If I divide humanity into groups based on whether they have an even or odd number of whorls in their left thumbprint, I’m using a definition that’s grounded firmly in biology. But the division is pointless and doesn’t reflect a significant division in the species. That definition would be a social construct, not reflective of anything more than superficial differences between individuals. Rather like race.
So let me ask again, not to split hairs but to get at the fundamental disconnect between us: what’s the need your racialist theories fulfill? Since scientists seem perfectly able to describe and analyze human diversity using defined categories such as population, what job do the various racialist theories do?
Colin: “You agree with X and then I’ll address Y” is a symptom of a conversation that is well and truly bogged down.”
Said Cratylus to Socrates when he recognized where the dialogue was headed.
I’ll take it that you concede the argument.
Colin: ” My point is rather that “race” is a subjective concept without nearly as much objective grounding as “species,” to the extent that there is virtually no practical purpose to defining or focusing on race as a concept.”
One doesn’t focus on race as a concept, rather one applies the conceptual tool of biological race to focus on an aspects of biological diversity. If this conceptual tool has little practically utility then why is it so well established in the biological sciences. Or do you want to argue that this conceptual tool has virtually no practical utility only in regards to humans? (Yet how could this constitute an argument against the existence of human biological races — as opposed to merely an argument against research on differences between them? ) But I just cited a well known population genetics book which maintains otherwise. And a google scholar search turns up papers such as: Risch, et al. (2002). Categorization of humans in biomedical research: genes, race and disease. Genome Biol, Do you agree, at very least, that some geneticists apply this conceptual tool to explore practically useful concerns – and that some geneticists consider the application of this tool to human genetic diversity to be practically useful?
Colin: “If I divide humanity into groups based on whether they have an even or odd number of whorls in their left thumbprint, I’m using a definition that’s grounded firmly in biology.”
Yet I already anticipated this. I wrote:
“It so happen that in biology, natural populations are deemed to be more important that artificial ones — they are given quasi privileged status. Hence, our taxonomic systems organize life by either genealogical (cladistics) or geneotypic (evolutionary taxonomy) similarity, and not by similarity in so called arbitrary character traits. There is a reason for this, of course, and it’s that such categories allow for more inductive inferences. If my “genetic populations” are based on similarity in two genetic characters, I have less inductive leverage than if they are based on evolutionary relationship and consequently similarity in terms of the whole genetic program. So, Wade and others have sound biological precedence for saying that race (qua subspecific natural population) is a generally more important grouping than others. It’s a well founded prejudice, no?”
Do you agree that biologists, at least from Darwin on, tend to privilege biological natural populations over biological artificial ones? Do you agree that this privilege is epistemically justified on the grounds that the former allows for more inductive leverage?
…
Three more points for Chuck — what are we playing up to?
Colin: The race concept has virtually no practical utility in biology.
Chuck: Why?
Colin: Because the concept is too subjective.
Chuck: What do you mean?
Colin: I mean that there are a bunch of different operationalizations of the concept floating around.
Chuck: Why might that be?
Colin: Uh…..you know, this type of dialogue is symptomatic of a conversation that is well and truly bogged down.
I’ll take it that you concede the argument…. Three more points for Chuck
Carve as many notches in your keyboard as you like; they’re worth what you paid for them.
One doesn’t focus on race as a concept, rather one applies the conceptual tool of biological race to focus on an aspects of biological diversity.
This is empirically false. See, for example, Steve Sailer pontificating about race as a measure of some nebulous, undefinable extended family (but not too extended, so as to preserve the white-black divisions he finds so compelling).
If this conceptual tool has little practically utility then why is it so well established in the biological sciences.
It isn’t. See, for example, the response experts have when an amateur like Wade treats it as if it were. Your response is primarily, “Look at this scientist who uses the word race!” Yes, legitimate scientists use the word. But they don’t seem to be using it to mean the same thing–in other words, they’re using the word, not some coherent concept.
Instead, they seem to be using it as a culturally-defined concept and fitting it to the data. Maybe I’m wrong about that–I’m not familiar with all, or even a decent chunk of, the literature. But they way you and other racialists flounder about trying to find a consistent framework for the concept of race paints a pretty dire picture. It seems to be more about preserving the utility of race as a social tool than actually accomplishing science with it.
Colin: The race concept has virtually no practical utility in biology.
Chuck: Why?
Colin: Because the concept is too subjective.
Chuck: What do you mean?
Colin: I mean that there are a bunch of different operationalizations of the concept floating around.
Chuck: Why might that be?
Colin: Uh…..you know, this type of dialogue is symptomatic of a conversation that is well and truly bogged down.
Yes, the kind of dialog where you award yourself points and celebrate your own cleverness with hypothetical exchanges is symptomatic of a conversation that is well and truly bogged down.
I’ll make another attempt. Race isn’t useless because it’s subjective. I’d say rather that race is useless because the only function it seems to have is as a proxy for other, better-defined concepts like population or specific ancestry. The subjectivity of race may be a consequence of that; there’s no application of it that accomplishes very much, so we wind up with various flavors of racialist promoting various definitions and insisting they’re doing science.
The rest of us are left wondering why racialists care so passionately about race. What does it do? What’s the point of abandoning useful, well-defined metrics in favor of a loose collection of concepts aimed generally at supporting the cultural construct? One commenter pointed out that doctors may use race as a rough proxy for determining how to treat patients. Assuming the results are empirically supportable, the utility of race in that case would be that it’s a very easy proxy to apply in circumstances where the fact that it’s very inaccurate might not matter so much; if the patient is unconscious, calling race by looks is obviously easier than asking them for a genealogy, even though the visual assessment isn’t likely to tell you all that much about their specific genes.
Colin: “It is empirically false [that ” One doesn’t focus on race as a concept, rather one applies”].. See, for example, Steve Sailer pontificating about race as a measure of some nebulous, undefinable …”
The population genetic definition which I offered can readily be translated into Saileresque: an extended family that has inbred enough for members to be more overall genotypically similar to each other than to members of other extended families. Thus the Sailer race concept encompasses this. See below.
Colin: “I’d say rather that race is useless because the only function it seems to have is as a proxy for other, better-defined concepts like population or specific ancestry.”
So, in a parallel manner, the concept “vehicle” is useless because it only functions as a proxy for more specific concepts such as “automobile” or “boat”? The epistemic utility of a more general concept is its ability to tie together the specific forms of it.
Look, what would you like to call the general pan research program (population genetic, ecology, zoology, phylogeny, etc.) concept of “sub-specific biological groups where members are arranged according to overall genetic relatedness”? Population? This is overly confusing because population can refer to e.g., breeding populations. Typical definition: “a group of interbreeding individuals that exist together at the same time.” Also, population has a statistical meaning — set of individuals. Genetic population? Better — but ‘genetic population’ often is used to refer to breeding populations, too. How about a pure neologism: “Geme”? This works for me. But how do we handle past usage. Just re-translate everything going back to e.g. Darwin? Hmmm. Or do we say that “geme” (a play off of “deme”) is a refined version of the vague race concept? I can work with this. So you agree that there are human biological gemes*, right — for example, the continental gemes* such as mongoloids, caucasoids, etc.
*also still called “races”
….
I’m not opposed to changing the term — if we do so consistently going way back — or with refining the concept — if we note that this is the new and improved version of what was once called “races” — just with this verbal hocus pocus.
Colin: “The rest of us are left wondering why racialists care so passionately about race. What does it do?”
It does what such concepts are meant to do: it allows us to organize our knowledge and to see relationships, especially evolutionary ones. What’s the utility of derecognizing the biological geme* concept — or of proscribing the use of the term “race” as a synonym for geme*. Yes, we know that “it’s a loaded term” — loaded with information.
What Sailer could say doesn’t affect what he is saying, and what he is saying is inconsistent with your assertion that “one doesn’t focus on race as a concept.” And the fact that your concept of race flows smoothly into his illustrates how nebulous his concept is; his concept of race could put everyone into one race or divide the species into millions of races. There’s nothing to prefer one result over another other than his desire to see a number of races matching his cultural preconceptions.
Your preferred definition seems equally fuzzy. The variables you select (such as K or the number of people being sorted) determine the size and number of the resulting races. You can just juggle those inputs until you get the result you want, like twisting an aperture ring until the subject you’re most interested in comes into focus. The fact that you’re using genes and alleles to make those distinctions doesn’t make the ultimate result more grounded in biology than in your cultural definition of the targeted result.
I very much appreciate your use of an analogy; I love analogies. I think this one is poor, though. “Vehicle” is an easily-defined concept encompassing a number of equally well-defined sub-concepts. There are some fuzzy boundaries (Is a bicycle a vehicle? Are roller skates?) but they exist on the edges of a discrete core concept.
“Race,” on the other hand, is a mess. In this thread alone, you want it to be groups that are more self-similar than not, Sailer wants it to be extended families, and MW says it’s whatever groups people can visually distinguish 99% of the time. There are probably others in the subthreads I haven’t been following. What all these definitions have in common is that they only return similar results under one condition: when the person applying them starts with preconceptions consistent with those results.
Calling “racialism” just the study of “sub-specific biological groups where members are arranged according to overall genetic relatedness” is acknowledging its arbitrary lines. Anthropologists can and do study people and their genetic differences without adopting the freighted, useless language of racism.
You seem to enjoy abusing little words like “fuzzy”. I provided you with a very precise concept. This concept left undetermined a number of values as does e.g., the concept ‘population’, but it was not “fuzzy” as the concept allows for a clear determination of which groupings are biological races and which are not. Presidential elections? No. Speeded contests? No. Lineages with no pedigree collapse? No. Throw out some possibilities and I will give you an answer — biological races or no — if this isn’t the test of conceptual fuzziness, what is? Now, as Sailer leaves the degree of pedigree collapse open ended, his definition is technically fuzzy in the statistical sense. Is it unclear? You can try the classification test with him.
It is true that both concepts allow for lumping and splitting. This, though, is a feature not a bug as the race concept is a tool for analyzing biodiversity. Specific racial classifications are merely outputs, while the race concept is the conceptual program into which genetic data is fed. The fact that the race concepts themselves don’t specify classifications — and never have — should be taken as proof that they aren’t simply contrived to justify “cultural preconceptions” about human groupings. Your reasoning makes little sense: “Human continental race classifications are not determined by biological race concepts, therefore these concepts are useless and just represent a means of justifying human continental classifications.” Ok, there — and I guess that algebraic principles are just back-fitted rationalizations for giving students with lots of red marks on their papers low grades.
But you say: “What all these definitions have in common is that they only return similar results under one condition: when the person applying them starts with preconceptions consistent with those results.” Yet the classifications can be generated using unsupervised genetic analysis. Just take a look at the pretty pictures here: http://www.biomedcentral.com/1471-2156/11/94/figure/F7 Whatever the case, I don’t fixate on specific classifications. Just as I know how to distinguish words from concepts, I know how to distinguish race concepts from race classifications.
Your particular definition of race is better specified than most, but that’s not saying much, is it? I hardly think I’m abusing the word “fuzzy” when the concepts you’re throwing around are only vaguely defined. What’s race? Your self-similarity concept? Sailer’s infinite families? The other guy’s visual recognition test? “Fuzzy” is a generous term for the way racialists define race; the only common themes are that you know it when you see it, what you see happens to map very neatly to your preconceptions, and you spend an awful lot of time thinking feverishly about it.
Yet the classifications can be generated using unsupervised genetic analysis.
I think it would be fairer to say that classifications can be generated using many different analyses, which return many different results; racialists pick and choose those they find most convenient, which are those corresponding with their preconceptions. The article you linked to (which, to be fair to the authors, doesn’t talk about race at all), for example, finds four clusters instead of the “three major races” Wade says are easy to recognize or the five he says are supported by genetics and continental geography. Why aren’t the results consistent? It appears to me that the definitions of race are idiosyncratic and designed to return classifications that fit cultural notions of race. The racialist theories aren’t coming organically from the study of human diversity, they’re the result of constructing colorable scientific arguments to support cultural notions of race. It somewhat resembles the cargo cult practices of creationism, picking and choosing scientific experts to create a “sciency” if not scientific framework around a non-scientific ideology.
One common thread through my responses is that I’m no expert; I’m unfamiliar with most of the research in the field, and poorly equipped to pick up new research and read it for comprehension. That’s nothing unusual in this world, where we’re commonly confronted with problems outside our various wheelhouses. Like most people in that situation, I resolve these dilemmas (at least to my own satisfaction) through a combination of my own knowledge and assessing the credibility of the voices on each side of the question. For what it’s worth, you come across relatively poorly in that analysis. Your condescending tone and occasional nastiness make it difficult to take you seriously, particularly when your own expertise seems fairly shallow here. (I realize I’m being a bit hypocritical here, but I don’t expect anyone to take my opinions on this subject seriously.) I would find the HBD crowd generally much more persuasive if there were anything approaching a consensus on what “race” is or how to determine it. Many HBD commenters insist that this is coming, and that in just a few years it will be the new scientific orthodoxy. Well, maybe so! I’ll have lots of crow to eat. But until then, with the subject-matter experts lined up on one side and a crowd of suspiciously febrile bloggers on the other, it doesn’t seem terribly likely.
I think it’s pretty clear that we disagree, and where we disagree. Obviously we haven’t persuaded each other of much, although I appreciate that you got me to do some reading I wouldn’t have otherwise. With the start of a busy week and a fresh Wade article I’m probably not going to continue this conversation. But I do promise to read your inevitably, if undeservedly, triumphant last word.
Try thinking relativistically. A hundred miles of ocean between Scotland and Ireland meant some restriction on gene flow, but nowhere near as much as the 1600 miles between Africa and South American meant before 1492.
What African populations? Thinking relativistically, even the pre-1492 division between South American and any arbitrary African population is less than the division between African populations, isn’t it? You’ve said elsewhere that race is ancestry, but the South Americans and Spaniards shared most of their ancestry even before 1492. I don’t see any evidence that the resulting differences are more than skin deep, or are determinible in any way other than calling your gut feelings.
“Thinking relativistically, even the pre-1492 division between South American and any arbitrary African population is less than the division between African populations, isn’t it?”
No. You fell for one of the most common myths.
You need to read my 2000 article Seven Dumb Ideas about Race:
https://www.vdare.com/articles/053100-cavalli-sforza-ii-and-seven-dumb-ideas-about-race
I’m sorry, but I don’t see what in that article is supposed to be responsive to my point. I may be misunderstanding your position, but please elaborate.
” isn’t it? ”
No. Take a look at figure 1, here:
McEvoy, et. al. (2010). Whole-genome genetic diversity in a sample of Australians with deep Aboriginal ancestry. The American Journal of Human Genetics, 87(2), 297-305.
You’ll notice that the genetic distance between two S.S. African populations is less than that between one S.S, African and one out of African population.
The connecting lines are scaled on genetic distance.
Sorry, I’m not familiar with the literature. Where does it explain the scaling of the connecting lines? I see the explanation that the groupings are determined by the K value selected, but not a description of the actual distance between those groupings.
The measure of inter-population genetic distance is fixation index. en.wikipedia.org/wiki/Fixation_index The branches are scaled on this.
See the “Scale” and the line with 0.01,
“The matrix of pairwise interpopulation genetic distances (FST values) was used to construct a neighbor-joining (NJ) phylogenetic tree that summarizes the relationship of the 52 populations to each other ( Figure 1A).”
The point is that S.S, African populations cluster together (at K=5) because there is less genetic distance between them relative to out-of-African populations. .
Dtizy: “silly or scatterbrained.”
I used “ditzy” descriptively. This was a silly piece in which a number of sophomoric criticisms, fallacies, and nonsequiturs were made and in which major internal inconsistencies can be found. I detailed some above. My characterization was generous compared to her characterization of Wade’s book as “pseudoscientific rubbish”. If you want, I will detail, paragraph by paragraph, the problems.
To take another example:
“Rosenberg et al. never published any statistical evidence that justifies picking 5 races instead of 7, or 4, or 2 (although such methods do exist–see Bolnick et al. 2008).”
But in Rosenberg, et al. (2005), we are told:
“When the number of loci, sample size, and correlation model were held constant, K = 2 (that is, two clusters) generally produced smaller clusteredness than did the larger values of K (Figures 3 and 4; Table 1). For the correlated allele frequencies model, K = 5 and K = 6 tended to have higher clusteredness than did K = 3 and K = 4, whereas the reverse was true for the uncorrelated model (Figure 4).”
As Dienekes notes, “some numbers of K fit the data better than others”
dienekes.blogspot.com/2005/12/clusters-strike-back-ii.html
So, if Jennifer meant Rosenberg et al. (2002), the statement while correct is deceptive as we see in Rosenberg et al. (2005) that there is “statistical evidence that justifies picking 5 races instead of 7, or 4, or 2” — as least as far as there can be. That is, there is no “real” — to use her term (quote: “number of divisions really exists in nature”) — level of genetic analysis, but preferring some divisions over others can be justified on empirical grounds GIVEN some idealized notion of races. And since Jennifer contrasts races with clines (quote: “there aren’t any races, just clinal distributions of genetic diversity”), presumably she would think that clustered populations make for better races; thus we have an idealized notion. More clustered, less clinal — note: for reasons discussed above, I consider the race/cline dichotomy to be conceptually confused.– populations make for better races. Thus based on Rosenberg et al. and given this prejudice we can say that K=5 is preferable to 7, or 4, or 2.
….
Look, she really did not think this through.
Having read your criticism and the original piece, it seems pretty clear that you’re using “ditzy” for the same reason you use Dr. Raff’s first name and keep casually insisting she doesn’t understand or hasn’t considered the question–as a lazy, personal dig at someone making an argument you don’t like.
Your comment here reads almost as if you were trying to create something that looks like a response to the original article, but serves only as a container for those snide remarks. For example, you hold up Rosenberg (2005) as if it rebuts Dr. Raff’s point, but she explicitly cites that paper: “Structure’s results are extremely sensitive to many different factors, including models, the type and number of genetic variants studied, and the number of populations included in the analysis (Rosenberg et al. 2005).” (You must have read that sentence, it comes right after the one you quoted.)
Nothing you wrote seems to address the point that Rosenberg “never published any statistical evidence that justifies picking 5 races instead of 7, or 4, or 2.” You seem to be claiming that clusteredness justifies a K of 5 (or 6?), but why? It’s another arbitrary criteria added to the stack–the assumptions you make going into the Structure analysis determine the number of “races.” Why is that assumption better than any other?
Your last paragraph seems to answer that question by assuming that Dr. Raff must believe “that clustered populations make for better races.” But did she write that? If clusteredness is the assortment of individuals into discrete groups, then wouldn’t it be even clearer if we used a much higher K to get a more granular assortment? That wouldn’t map cleanly to “race,” though…
Obviously I’m not a scientist, and can’t review the literature in substance. So it’s possible that you’ve made some deep, insightful point here. But based on your use of casual, snide personal comments to attack Dr. Raff’s credibility, I am confident that neither one of us believes that.
You said: “Dr. Raff’s first name”
I still call my medical doctors “Dr,” That’s about it.
You said: “You seem to be claiming that clusteredness justifies a K of 5 (or 6?)”
I was clear on this point; reread what I said again.
There can not be a true level of genetic analysis; no proponent of biological race claims that there can. So Jennifer’s critique regarding STRUCTURE not establishing a “real” number is vapid. That said, you could prefer some level of genetic analysis GIVEN some pragmatic or other concerns. For example, taxonomists only formally recognize — they prefer — races that “differ enough”? Why? Because to do otherwise would lead to taxonomic chaos since it’s races all the way down. This pragmatic concern arises regarding human races, too; thus Leroi (2005) noted:
“[t]here is nothing very fundamental about the concept of the major continental races; they’re just the easiest way to divide things up. Study enough genes in enough people and one could sort the world’s population into 10, 100, perhaps 1000 groups, each located somewhere on the map. (Leroi, 2005)”
Now, I noted that Jennifer seems to see “clines”, meaning something like population continua, as antithetical to races. She says: “there aren’t any races, just clinal distributions”. Am I misreading this? If not, then it follows that less “clinal”, meaning less continuous, populations would make for better races (given her conception), no? Well, Rosenberg (2005) happens to present a clusteredness analysis. And some of the populations were found to be less “clinal”, meaning less continuous. So, given Jennifer’s apparent cline versus race conception and given Rosenberg (2005)’s results, some populations make for better races than others — and that would be k=5 (or 6) versus K=7, or 4, or 2. Thus, you have empirical justification to pick a number, GIVEN some conception about preferable races. Of course, if you have no preferences, then you couldn’t possibly empirically objectively prefer any level of analysis. Well, duh.
But why is this preference better than others? I’m not the one who entertains it. I see no population continua problem. But Jennifer does. So ask her,
….
Does anyone else understand this point? ….
What is a “true level of genetic analysis”? At some point you may wish to consider that your failure to communicate effectively is not actually due to the vapidity of the learned professionals you are insulting. Your writing is not as clear as you think it is.
It seems that you’re taking the position that (a) the inputs one selects in a STRUCTURE analysis determine the results; (b) preconceptions about where the lines between races should lie provide a justification for selecting such inputs; (c) wanting greater clusteredness supports selecting a K of 5 (or thereabouts); and (d) Dr. Raff should want to choose K=5 because it would resolve her concerns about the clinal nature of supposedly racial divisions.
I’d agree with (a) and (b); I’m not conversant enough with the subject to fully understand (c), but I think I agree with it. (D) is nonsensical to me. Does the Rosenberg (2005) paper indicate that populations are actually clustered in such a way that is only revealed if one selects K=5, or that selecting K=5 imposes clustering on those populations? The question may be nonsensical; I’m obviously a layperson.
Incidentally, your arrogant and insulting style made me curious about the depth of your own expertise. Elsewhere in this thread you used a Gravatar associated with John Fuerst. What’s your background? Are you published anywhere other than the Human Varieties blog and the Stormfront forum?
“What is a “true level of genetic analysis”?”
The philosopher-biologist Quayshawn Spencer discussed this issue recently. Quote:
“As for (b), Hochman (2013, p. 348) claims that there is a ‘‘grain-of- resolution problem’’ insofar as there is no ‘‘principled reason’’ for calling one collection of genetic clusters of human populations “‘races’’ and not the rest. For example, we might be tempted to call K = 5 ‘races’ in Rosenberg et al.’s studies because it matches current U.S. Census racial groups. However, according to Hochman (2013, p. 348), the ‘‘chosen number’’ should not simply reflect a ‘‘folk assumption’’ since such a move would be an arbitrary reason for designating each cluster a subspecies. (Spencer, 2014. The unnatural racial naturalism)”
Hochman’s ‘‘grain-of- resolution’ is my “level of genetic analysis” and Jennifer Raff’s “number of clusters”; as Hochman notes, there is no ‘‘principled reason’’, that is, there is no reason in principle, for calling the genetic divisions based on one level of analysis but not others races. Of course, you can stipulate certain levels e.g., primary/major/continental races but there are no purely genetic grounds for saying that these e.g., k=5 really are the true or real races. There is no true, with a capital T, level of analysis. Ok, but a typical solution is simply to grant that the natural population divisions at different levels of subspecific genetic analysis are races (to note, following
Dobzhansky and others I take an extremist Horton Hears a Who, a race is a race no matter how undifferentiated, approach; others are more exclusive); hence, Wade notes that there are primary races, continental races, and subraces.
Now, as said, one can nonetheless prefer some grain-of- resolution over others, just as taxonomists do. So, for example, Risch et al. 2002 tells us: “genetic differentiation is greatest when defined on a continental basis’’ If this is true, we can use this fact to construct an empirically grounded justification for preferring (that is, focusing on) continental level races, granted that we consider “genetic differentiation” to be an important consideration . This isn’t to say that the other divisions cease to be races, rather they are less prioritized. Now, Wade more or less makes this point. Which is why Jennifer’s criticism in this regards, like others, is misplaced. But maybe she didn’t well understand him. Let’s look at some quotes:
Wade: “True, races are not discrete entities and have no absolute boundaries, as already discussed, but that doesn’t mean they don’t exist. The classification of humans into five continental based races is perfectly reasonable and is supported by genome clustering studies. In addition, classification into the three major races of African, East Asian and European is supported by the physical anthropology of human skull types and dentition..”
Here Wade doesn’t say that the “continental based races” are THE races but are race; they are continental level ones. He notes that one can also have a three MAJOR race classification. There is no inconsistency here because the five continental races consist of the three MAJOR races plus two less differentiated — due to time of separation — continental ones. Thus Wade says:
“A practical way of classifying human variation is therefore to recognize five races based on continent of origin. These are the three principal races—Africans, East Asians and Caucasians—and the two other continent-based groups of Native Americans and Australian aborigines…”
Moving on.
Wade: ” Within each continental race are smaller groupings which, to avoid terms like subrace or subpopulation, that might be assumed to imply inferiority, may be called ethnicities. Thus Finns, Icelanders, Jews and other groups with recognizable genetics are ethnicities within the Caucasian race.”
Here Wade recognizes the nested nature of race. He just calls the divisions produced by finer partitions, the sub- races,”ethnicities” so as not to imply inferiority.
Wade: “Such an arrangement, of portioning human variation into five continental races, is to some extent arbitrary. But it makes practical sense. The three major races are easy to recognize. The five-way division matches the known events of human population history. And most significant of all, the division by continent is supported by genetics.”
Here Wade recognizes the arbitrariness of selecting a grain of focus.
Wade: ” Lewontin’s answer came out to 6.3%, meaning that of all the variations in the 17 kinds of protein he had looked at, only 6.3% lay between races, while a further 8.3% lay between ethnic groups within races. These two sources of variation add up to around 15%, leaving the rest as common to the population as a whole…..The 15% genetic difference between races, in other words, is not random noise but contains information about how individuals are more closely related to members of the same race than those of other races. ”
The final sentence of this excerpt makes it clear that Wade considers ethnic groups (or sub-races) to actually be races — as the “15% genetic difference between races” is partitioned between [continental] races and between ethnic groups. Knowing this, the first sentence can be read: “Lewontin’s answer came out to 6.3%, meaning that of all the variations in the 17 kinds of protein he had looked at, only 6.3% lay between continental races, while a further 8.3% lay between sub races within continental races.”
Wade: “A necessary approach to studying racial variation is to look not for absolute differences but at how the genomes of individuals throughout the world cluster together in terms of their genetic similarity. The result is that everyone ends up in the cluster with which they share the most variation in common.”
Here Wade clarifies what races are. They are populations defined in terms of overall genetic similarity i.e., natural populations and not in terms of specific characters e.g., artificial populations.
Wade: “A variation on the no distinct boundary argument is the objection that the features deemed distinctive of a particular race, like dark skin or hair type, are often inherited independently and appear in various combinations…..But as already noted, races are identified by clusters of traits..,,Even when it is not immediately obvious what race a person belongs to from bodily appearance, as may often be the case with people of mixed-race ancestry, race can nonetheless be distinguished at the genomic level.”
Here Wade reiterates that races are defined multivariately — that is, in terms of overall similarity, Unfortunately, he fails to make it clear in this passage that the defining feature is (overall) genetic similarity and that phenotypic similarity is an index of this. But the last sentence, and others — such as that quoted above — indicate that this is what he’s thinking of — and thus that his race concept doesn’t substantially differ from typical biological natural population ones.
“It seems that you’re taking the position that”
Regarding the level of genetic analysis issue, below is Kitcher’s 2007 discussion: :
“Faced with the statistical analysis, and especially with the illuminating figures that present the data, it is tempting to say that here we have a completely objective division of the human species into infraspecific groups. We have put the question, and nature has spoken: there are races, or something akin to them. That conclusion, however, has to be hedged with qualifications. First, it is important to understand the question that has actually been put. Given rich data about individuals and bits of their DNA sequences, computer programs have sought divisions, being told in advance how many clusters they are to find. So, for example, we might ask, “If our species were to be divided into just two groups on the basis of genetic similarity, how would geographical populations be assigned to those groups?” and we would discover that the two clusters are “anchored by Africa and America” (Eurasian populations would be lumped with the African ones). Ask for three groups, and Eurasia is split off; ask for four, and East Asian populations form a distinct fourth group; ask for five, and Oceania is separated from the other East Asian populations. 21 So there is a genuine issue about level or fineness of grain, one that can only be settled on pragmatic grounds: the clusters, or races, will be picked out by fixing the number so that the resulting division best accords with the inquiries we find valuable. Picking out new clusters preserves, in an important sense, the boundaries that have already been drawn. You may find new subdivisions within a previously identified unit, but you do not generate new clusters that straddle earlier ones. If two populations are assigned to different clusters at one value of the parameter, they remain separated at all higher values. On this basis, one might conclude that the pragmatic component in dividing the species is relatively insignificant, just a matter of finding the appropriate level in an objective tree-structure”. (Does ‘Race’ Have a Future?)
Races are nested within races. As a result, you can split major geographical races into minor geographical ones and you can lump the latter into the former. Thus one must select — when doing a genetic analysis — a grain of resolution. That is (a). (b) is that you can empirically justify selecting a grain of resolution GIVEN some views about what makes for better races. For example, you could add a subroutine to STRUCTURE and tell it to select the grain of resolution, the K=, that shows the greatest level off differentiation or the most clusteredness or the smallest dissimilarity fraction. If you did this, you would (typically) get K=5. (c) is that Jennifer does consider some natural populations to make for better races than others e.g.,ones that are not located on a population continuum — or, in her terms, are not clinal.. Given this and given the empirical results, one can justify selecting a particular grain of resolution and so K=5
I’m not saying that one should select k=5; I’m saying that one can empirically justifiably do this, given Jennifer’s (apparent) better-race standards.
…
You said: “Incidentally, your arrogant and insulting style made me curious about the depth of your own expertise”
Now and then, I post at humanvarieties.org/ and openpsych.net/ODP/ Since topics often involve race, the SF crowd now and then links to HV. As for research, supposedly I’m a second author of a paper in a third rate chemistry journal — though, I lost contact with the advising professor, so I don’t know which. Also, I have half of a phd in experimental psychology from a second rate university; as for biology, I once had a secondary teaching license — that was when I had the crazy idea of teaching in the U.S., after having a blast teaching abroad, I have a bartending certificate — from my early 20s — not sure if that counts. Ya, so when a schmuck like myself can figure things out better than learned professionals that tells you something about either those professionals or their profession, no?
@chuck
It tells us more about your opinion of yourself.
In other words, there are as many races as there are angels that can dance on the head of a pin. It is a subjective categorization, with “race realists” using their folk ideas about race to select criteria that permit them to carve humanity up into groups fitting their preconceptions about racial identity.
“Ya, so when a schmuck like myself can figure things out better than learned professionals that tells you something about either those professionals or their profession, no?”
Ask the creationists asking the same question about evolutionary biologists, or the naturopaths asking the same question about oncologists, or the tax protesters asking the same questions about tax lawyers, or the Planet X fans asking the same question about astronomers.
Colin,
Are you ever going to realize that Chuck has thought vastly longer, harder, and more productively have about this subject? Or will you continue to exult in your ignorance?
Steve,
Would you please answer the question that several people (including Colin) have asked you in various forms downthread? (“how far back do you trace your family groups in order to meet your definition of race”?)
Or, even better, would you put your definition of race (“partially inbred extended families”) into population genetics terms? What values of R (the coefficient of relationship) are you using to define these extended families?
Steve,
Lots of people have “thought vastly longer, harder, and more productively” about race than I have, from frothing Stormfront bloggers to actual geneticists. (Although “productively” is as subjective a term as “race,” obviously.) How hard he’s thought about the subject and how much he’s written on the subject matter much less than the substance of his conclusions. From what I can tell from his writing here, he’s no more able to define race objectively than you are, and no less willing to restrain his resentment at the scientific establishment that disagrees with his preferred conclusions.
Chuck,
Writing about population genetics programs in laypersons’ terms is always a tricky balance between simplifying the math (and concepts) enough to be intelligible to an average person, but not oversimplifying to the point of distortion. I’ll do my best here. I agree with Colin that you haven’t expressed your ideas very clearly, but I’ll try to address what I understand to be your concerns. (I’ve drawn from Schwartz and McKelvey, 2009, Jorde and Wooding 2004, and a bit from the papers at the end of this comment).
There are two fundamental assumptions underlying clustering analysis : 1) there are biologically meaningful clusters to be discovered, 2) the effects of structural elements (like gradients) are small relative to those elements that contribute to the clustering.
STRUCTURE starts with the assumption that there are K populations (where K is assigned by the user). It assigns individual genotypes to those populations using a Markov Chain Monte Carlo method that minimizes deviations from Hardy Weinberg and also minimizes linkage disequilibrium. STRUCTURE will provide posterior probability of the data for a given K: Pr (X|D). Using this estimate to choose between values of K is not without some controversy—the authors of STRUCTURE caution that it “merely provides an approximation” and “biological interpretation of K may not be straightforward.” Evanno et al. (2005) observed that since the log probability of data wasn’t maximized at the correct value for K, they recommend the measure of ∆K, a second order rate of change of the likelihood function with respect to K, as a better estimate of which K most accurately described the data. A lot of authors follow this practice, although it has been argued against (Waples and Gaggiotti, 2006).
Now, as to your points about Rosenberg et al. Neither his 2002 nor 2005 papers report the LnP(D) (or ∆K) for any of the values for K, so we have no way of distinguishing which is the “best” value of K. (I use “best” in quotes because LnP(D) is a somewhat disputed metric, as I discussed above, but at least it’s some way of evaluating between different Ks). Bolnick 2008 reports information about his unpublished LnP(D) values (provided by personal communication from N. Rosenberg): “…no single value of K clearly maximized the probability of the observed data. Probabilities increased sharply from K=1 to K=4 but were fairly similar for values of K ranging from 4 to 20. The probability of the observed data was higher for K=6 than for smaller values of K, but not as high for some replicates of larger values of K. The highest Pr (X|K) was associated with a particular replicate of K=16, but that value of K was also associated with very low probabilities when the individuals were grouped into 16 clusters in other ways. Consequently it is uncertain which number of genetic clusters best fits this data set, but there is no clear evidence that K=6 is the best estimate. “ (p77)
If you take a look at their 2005 figure 2, you’ll see some discussion of likelihood values and how they differ from run to run (they don’t report them explicitly). They note that different runs resulted in different population subdivisions (“The highest-likelihood run of the ten runs with K = 6, shown in the figure, had a different pattern from the other nine runs (not shown). These other runs, instead of subdividing native Americans into two clusters, subdivided a cluster roughly similar to the Kalash cluster seen in [3], except with a less pronounced separation of the Kalash population.”)
The point of their work was not to provide statistical support for different values for K, but rather a refutation to the claim made by Serre and Paabo (2004) that their derived clusters were sampling artifacts.
You say “Thus based on Rosenberg et al. and given this prejudice we can say that K=5 is preferable to 7, or 4, or 2.” K=5 does indeed seem preferable to K=4 or 2, but Rosenberg et al. 2005 NEVER TESTED ABOVE K=6, so you have no justification for saying that K=5 is preferable to any higher value of K.
In fact (and I think I point this out in my post above), Rosenberg and co-authors explicitly say, with regard to the clusters they identify that they are 1) very small fraction of human genetic variation and 2) “should not be taken as evidence of our support of any particular concept of ‘biological race’”
Either you haven’t actually read Rosenberg et al. 2005 carefully, or you have simply misunderstood the point of their study.
As I discussed in my post above, STRUCTURE is not designed to be applied to populations that experience isolation by distance (IBD), as is true of most human populations. The authors of the program explicitly warn against this (but everyone does it anyway). Guillot et al. (2009) discuss this further:
“Another confounding factor of the clustering algorithms is IBD. All the models make sense fully only at a scale that is small enough to ignore its effect. At larger spatial scales, any species is affected by IBD and assuming within-cluster panmixia becomes inappropriate.” and “The general effect reported is that the presence of clinal variations tends to be interpreted as the presence of clusters and a number of clusters larger than one is generally inferred, even though no barrier to gene flow was present. “
STRUCTURE can absolutely be a useful tool for inferring individual ancestry, but only with an understanding of assumptions inherent in the clustering algorithms, and cautious interpretation of the results. Because of these caveats, careful and rigorous scientists generally view the “best” clustering scheme as a starting point: for generating testable hypotheses about ancestry and population history, NOT (as Wade and you and a number of other non-geneticists on this forum) as the ending point and definitive proof of discrete races.
Based on your evidently superficial understanding of how STRUCTURE works, you and Wade (and several others) don’t seem to be up to date on the ongoing discussion in the biological literature about it. May I recommend that you read the following sources, to get a better understanding of the field? They’re by no means an exhaustive survey, but they’ll get you started.
An excellent explanation of the math behind cluster analysis can be found in Chapter 8 of the book “Analyzing multivariate data” by James Lattin
Then, I’d recommend reading the following articles on STRUCTURE:
Evanno G, Regnaut S, Goudet J (2005) Detecting the number of
clusters of individuals using the software STRUCTURE: a
simulation study. Mol Ecol 14:2611–2620
Latch E, Dharmarajan G, Glaubitz JC, Rhodes OE Jr (2006) Relative
performance of Bayesian clustering software for inferring
population substructure and individual assignment at low levels
of population differentiation. Conserv Genet 7:1566–1572
Latch E, Rhodes OE Jr (2006) Evidence for bias in estimates of local
genetic structure due to sampling scheme. Anim Conserv 9:
308–315
Pritchard JK, Wen X, Falush D (2007) Documentation for structure
software: version 2.2. University of Chicago, Chicago, pp 1–36
Also, Long JC, Li J, Healy ME (2009) Human DNA sequences; More variation and less race. American Journal of Physical Anthropology 139: 23-34 would be an excellent article for you to read to get an overview of the problems with your argument.
Hi, If you have time and you’re open to it, can you comment [non-technically if possible] on this aside from r khan in comment # 1 above?
“species itself can be problematized in the same manner as race”
I’m a strict layperson here, but as a purely rhetorical matter that statement is very slippery. Whether A “can be problematized in the same manner as” B is not the same thing as saying that A and B have the same problems.
In this case, it seems to me (again, as a layman) that species are generally distinguishable from one another in an objective sense: individuals from different species can’t produce viable offspring. No such objective distinguishment is possible for race. Species get fuzzy around the edges and in special cases, which is why it’s fair to say you can “problematize” the concept, but that fuzziness is the exception rather than the rule as it is in race.
In other words, compare a cash transaction with a promise of a future favor. It’s virtually impossible to objectively value a particular good or service in terms of some future favor. It can also be hard to value a good precisely in dollar terms, but that’s a much less common and much less serious problem with a currency transaction.
(I’m sorry if that’s a terrible analogy; it fits my way of thinking but might not work for other people.)
What are you talking about? People value the promises of future favours all the time. It’s called options trading.
The precision of a price depends on the liquidity of the market — not on anything intrinsic to the item being valued (which can be a promise, or a thing, or currency). Of course, we estimate value all the time in non-liquid markets (by making comparisons to prices in more liquid ones.) This creates room for judgment. But room for judgment does not equal “subjective.” When people talk about the (general) price of crude oil, they are actually doing just this, but no one would say that the price of crude oil is subjective. I wouldn’t like to pay anything for gasoline, but somehow I have to.
Actual science (as opposed to the mythological version you seem to have in your head) relies on judgment all the time. Its concepts are ALL social constructs by the standards you are setting here.
The metaphysical rubber hits the pragmatic road when you get to medicine. Races is useful for diagnosis and treatment. That’s all the scientific validity it needs.
You’re overthinking an underspecified analogy, but for the sake of argumentation let’s step it up: the value of an option is not a promise of “a future favor,” but a precise and contractually defined exchange. If the contract weren’t specific–“I promise to do a favor for Pithlord,” as opposed to “I promise to deliver fifty bushels of hog ears to Pithlord on September 1”–it would be much less valuable (and likely unenforceable).
On a more relevant point, you are equivocating regarding the nature of subjectivity. When I speculate as to the future value of oil, my belief is subjective. The price that emerges as a consequence of my subjective belief (as well as all the other subjective beliefs of market participants) is not subjective.
I have no objection to “room for judgment” in science. I’m extremely comfortable with subjectivity, and with social constructs. The difficulty comes when “racialists” take a subjective construct and ascribe it actual, objective biological significance.
In other words, “I’m going to draw an arbitrary line between these two groups and call them races” is fine with me, as long as it’s well understood that the division is arbitrary and subjective. (Or, if the line is based on defined criteria, understood that those criteria are arbitrary.) Claiming that the same divisions happen to arise naturally and inescapably from biology is more problematic. I’m not persuaded that the divisions are more than post hoc justifications draped over culturally-constructed racial divides.
Your last point is the most interesting to me. As a layperson, I often turn to practicality when assessing scientific questions that I’m not equipped to analyze myself. If, arguendo, race is useful for diagnosis and treatment, what does it really mean? Is it anything more than a quick and dirty proxy for making a best guess as to the person’s specific ancestry?
For example, I’m usually considered to be ethnically Irish or Scottish. Let’s say there’s a test that should particularly be run on people who share a mutation that uniquely arose in Scotland ten generations or so back. A doctor using race to determine treatment would look at my name and my face and probably recommend that I undergo that test. In my particular case, though, it would be an error; my name is more or less accidental and my ancestry is mostly Danish. (Something I discovered relatively late in life; for quite a while, I would have agreed with the doctor that I was “racially” Scottish.)
The doctor’s guess would have been a decent one–people with my name and complexion are more likely to have ancestors from Scotland within the given time frame than people named Obama with dark skin. But that guess would also have been based on his and my assumptions about race, which are culturally-derived. And it would have been wrong, about me and possibly about the hypothetical Obama.
Having said all that, I’m not opposed to using race as a medical diagnostic tool, assuming as we are that the experts agree it’s useful. But all that’s doing is using a cultural construct as a quick and dirty proxy for actual ancestry, isn’t it?
A mistake about whether you are Danish or Scottish is unlikely to cause premature death or unnecessary disability because Danes and Scots are relatively closely related. We just don’t know enoguh about any racial differences for it to be medically relevant.
That is manifestly not the case for African Americans. Inappropriately diagnosing them or treating them based on validation that used European subjects can kill people. Diagnosis and treatment are (or ought to be) Bayesian. You should use the knowledge you have about background probabilities in the population that a person is a member of to decide on what tests you order, what drugs you prescribe and so on. Medicine is always dealing with probabilities, so the possibility of counter-example shouldn’t bother anyone.
I agree that there isn’t much point in pursuing the options/favour example. Philosophically, though, you still seem to think that something is either “subjective” or “actual, objective” and of “biological significance.” The distinction is really one of dgree. Sorry, but nature did not develop for the convenience of those with tidy minds.
That’s what we lawyers call “fighting the hypo.”
Assuming that racial assumptions actually are useful in making treatment decisions, I’m not opposed to using them. I think we probably agree generally on this point. You’re saying doctors should use the cultural construction of race to make best guesses about their patients’ genes, acknowledging that those guesses are often going to be wrong because those racial groups are imperfect proxies. Is that right?
Again, I’m not separating “subjective” from “significant.” I apologize for my lack of clarity, and I’m glad for this conversation in that it’s forcing me to sharpen my rhetoric. I think my primary question is whether common conceptions about race reflect anything more than the cultural construct, or at most very flawed proxies for very broad assumptions about genealogy. I don’t believe they do.
Among others, I made the following points:
Position (1): Biological race proponents generally do not claim that there is a privileged grain of genetic resolution or a true number of races.
We might take some examples:
Leroi, A. M. (2005). A family tree in every gene. Journal of genetics, 84(1), 3-6.
Sesardic, N. (2013). Confusions about race: A new installment.
Shiao, J. L., Bode, T., Beyer, A., & Selvig, D. (2012). The genomic challenge to the social construction of race. Sociological Theory, 30(2), 67-88.
This position is hardly a new; Mayr, Dobzhansky, Coon, Darwin, Kant, etc. all acknowledged the validity of different levels of racial analysis. The typical position was and is similar to that expressed by Leroi:
[T]here is nothing very fundamental about the concept of the major continental races; they’re just the easiest way to divide things up. Study enough genes in enough people and one could sort the world’s population into 10, 100, perhaps 1000 groups, each located somewhere on the map.”
Wade maintained a similar view; hence, he (a) recognized different grains of focus (e.g., a three primary race systems and a five continental race one), (b) noted that the choice was arbitrary in the sense that one could choose otherwise, and (c) pragmatically grounded his choice of focus. You don’t dispute any of this, presumably because you can not. Instead, you trot out the number or races strawposition.
Perhaps you might explain why you feel that for race to be a valid biological construct there must be a biological or genetic way of establishing a preferable or correct or true or privileged grain of resolution? Do you feel that this requirement holds for other biological entities such as demes or taxon in general, if not why?
Position (2): One can justify a preference for a grain of resolution on the basis of genetic and biological data, such as that presented by Rosenberg et al. (2002;2005), given some prior conceptions of what better races look like, conceptions such as those you yourself seem to hold.
I noted for example, that you consider genetic discontinuities — in your parlance non clinality — to be one important quality; you also seem to consider magnitude of genetic differentiation to be another. I then pointed out that Rosenberg et al. (2002;2005) presented evidence that support the position that (a) continental level races exhibit higher levels of clusterability than some other grains of resolution [K5].
My point was not novel — For example, here was Quayshawn Spencer’s (2014) discussion:
” First, some levels of genetic clustering corresponded nicely with certain folk racial classifications. For example, at K = 3, the clusters picked out Mongloids, Caucasoids, and Negroids. Furthermore, at K = 5, the clusters picked out current U.S. Census races: Sub-Saharan Africans (hereafter ‘‘Black Africans’’), Caucasians, East Asians, Amerindians, and Oceanians. K = 6 was similar to K = 5, except it had a single Pakistani population (the Kalash) in a part all by itself. A second interesting result was that the level of populations with the highest among-part genetic variance (r2a) was K = 5, which was 4.3 ± 0.40 at 95% confidence (Rosenberg et al., 2002, p. 2382). This last result seemed to support Risch et al.’s famous claim that ‘‘genetic differentiation is greatest when defined on a continental basis’’ (Risch, Burchard, Ziv, & Tang, 2002, p. 3).”
Now, it needs to be stated at this point that Wade only obliquely made the above argument. He mentioned the two points, (a) and (b) above:
“Many larger and more sophisticated surveys have been done since, and all have come to the same conclusion, that “genetic differentiation is greatest when defined on a continental basis…The Rosenberg-Feldman team then reanalyzed their data and gave their survey finer resolution by looking at 993 sites, not just 377, on each of the genomes in their study. They found that the clusters are real. Although there are gradients of genetic diversity, there is also a clustering into the continental groups described in their first article.”
Yet he did not clearly construct an argument and he characterizes the choice of continental races (K=5) as both somewhat arbitrary and practical.
Regardless, I do maintain that one can justify K=5 using Rosenberg et. al’s data.. You dispute this — or at least a straw version of it e.g., that the results provide “definitive proof of discrete races”, by arguing that ∆K is a best measure of the “best” K and that the results for ∆K are not give. Yet I gave a clear and unambiguous justification for using clusterability and degree of genetic differentiation as a measure of preferability: your own prior conception of what ideal races should look like. (Recall that I myself have no problem with microraces or with races locate in a population continuum; to my mind, the acceptance of these (as historically understood races) is very defensible; I, instead, justify prioritizing continental races on the basis of my limited memory span; 5-7 works well for me.) Now, you might argue that we should take into account ∆K, but in absence of this value we only have ones for clusterability and degree of genetic differentiation. Given this situation, you can only argue that clusterability and/or degree of genetic differentiation fails as a justifiable ways for justifying a K. I’m waiting for that argument. Make it or concede the point.
You also argue that the authors did not attempt “to provide statistical support for different values for K”, but I never asserted that they did — but only that, contrary to what you wrote, one can construct a justification based on the results which they presented. Also, you argue that the authors NEVER TESTED ABOVE K=6; yet Rosenberg et al. 2002 did — see table 1. In that table, you will see that the between region variance is lower at K=7 than at K=5. Am I misreading this? Regarding your other comments concerning STRUCTURE, at best they weaken the two justifications offered. However, weak justification is still justification, and you have yet to provide countervailing evidence.
More generally, STRUCTURE can only provide evidence for or against the coherence of certain racial partitions e.g., the lumping South and East Asians into a pan Asian race. It does this by evidencing that predefined populations represent natural populations, which is what races have frequently been understood to be. Rosenberg et. al. (2002) note that this — showing that predefined populations represent natural ones i.e., ones defined in terms of genetic relationship — was the point of their study:
“Most studies of human variation begin by sampling from predefined “populations.” These populations are usually defined on the basis of culture or geography and might not reflect underlying genetic relationships”
Note that the clusters generated by structure are not races themselves, rather they are the statistical outputs which one would expect where predefined “populations” natural populations — at least, discontinuous ones. This being the case, challenging the validity of STRUCTURE’s results does not challenge the validity of the race concept per se — just the evidence that some groups often said to be races qua natural populations are in fact these. But there are multiple lines of evidence indicating that the classically defined continental level races are in fact natural populations — for example, evidence from the PCA method, classic and newer. Throwing out STRUCTURE’s results would, then, still leave intact the other evidence.
…
Which points above do you disagree with? And do you agree that your original post was somewhat silly?
Surely you can do better.
In other words, “Given this long, detailed explanation of why you are correct that race is a completely subjective measure determined by one’s preconceptions rather than useful objective standards, wouldn’t you agree that you are a silly woman and also foolish?”
No one would agree with that, and no reasonable person would write it expecting agreement. Your comment isn’t a serious argument, but a vehicle for delivering more petty asides at someone who has offended you by not taking your frothing race obsession seriously.
If Jennifer can show that my claims are “pseudoscientific rubbish” — by showing that they’re internally or externally inconsistent — I’ll happily confess to a bout of foolery and recant. I expect her to do the same, If she can’t show that her own claims are not.
Waiting….
While you’re waiting, you might wish to go back and re-read Rosenberg et al. 2002 and 2005 and that list of articles I included with my last comment to you, because your understanding of cluster analysis is very superficial.
I’ll reiterate a few points that you’re clearly ignoring and/or misunderstanding:
1) As the authors of STRUCTURE note in the program’s documentation, it is not designed to be applicable to populations that experience IBD. (This is true of human populations). You can read it for yourself here: http://pritchardlab.stanford.edu/structure_software/release_versions/v2.3.4/structure_doc.pdf
2) Rosenberg et al. 2002 tested K=1-20. They did not publish any statistical justification [Pr (X|K)] to justify any particular value of K over another. Bolnick 2008 actually saw their unpublished Pr (X|K) values and described them (you can read the quoted description in my comment above). I realize that you don’t have experience in this kind of statistical analysis (that’s why I provided all those links for you to read), but Pr (X|K) is not the same thing as ∆K, nor of Rosenberg et al’s 2005 “clusterdness” statistic. You really need to learn the definitions of each statistic, as well as the pros and cons of using each, before we can proceed further in any discussion of these studies.
3) Rosenberg et al. 2005 (You appear to mix these two studies up) does not test values of K over 6. Nor do they provide ∆K or Pr(X|K) for any measure of K=1-5. They are not trying to find the “best” measure of K in this study anyway–they are attempting to rebut the criticism of their sampling method made by Serre and Pääbo (another study which you really need to read).
You’ll spend less time waiting for my replies to your points if you spend less time being rude. I’ll get to your comments when I get to them–after I’ve responded to the more serious and civil correspondents (not to mention Wade himself).
Hi Jennifer,
Your argument now is:
(a) “the authors of STRUCTURE note in the program’s documentation, it is not designed to be applicable to populations that experience IBD”
I will note, firstly, that I find your excessive focus on STRUCTURE to be odd — there are a number of other methods (e.g., DAPC) in use which produce convergent results. As such, attacking an inferred population structures by attacking the program STRUCTURE is akin to attacking heritability estimates by attacking MZ twin research. Whatever the case, this is what we’re told by Pritchard et al.:
“The underlying structure model is not well suited to data … [which exhibits IBD]. When this occurs, the inferred value of K, and the corresponding allele frequencies in each group can be rather arbitrary. Depending on the sampling scheme, most individuals may have mixed membership in multiple groups. That is, the algorithm will attempt to model the allele frequencies across the region using weighted averages of K distinct components. In such situations, interpreting the results may be challenging”
This means that in absence discontinuous jumps, one can not place high trust in the coherence of the clusters identified by STRUCTURE — which, of course, is all the more reason to prefer, at least when limiting consideration to this method, those sets of clusters which correspond to those populations between which there are jumps, such as K=5. But how do we know that there are such jumps at K=5. Apart from looking at google map or from experience traveling, we do because Rosenberg et al. (2005) tested for clusteredness, which effectively tests for the degree of genetic discontinuity. That is, Rosenberg et al.’s (2005) clusteredness results provide evidence that at k=5 there is less IBD relative to other Ks, which was the whole point of that 2005 cline-cluster paper, no?
So I can reformulate the argument:
P1. more clusteredness = larger jumps in genetic distance (Rosenberg, 2005)
P2. larger jumps in genetic distance = less IBD (Prichard et al. 2010 section 5.4)
P3. less IBD = more trustable, less arbitrary, more coherent (easy to interpret) partitions (Prichard et al. 2010 section 5.4)
P4. more coherent clusters more likely to better identify races as natural populations
P5. partitions that better identify races are preferable (inferred from Jennifer, 2014)
P6. K=5 shows more clusteredness (Rosenberg, 2005)
C K=5 shows more clusteredness (6), thus more likely to identify races as natural populations (1-4), thus, more preferable (5).
But maybe you intended this — argument (a) — as one to be directed against K \= 5?
Recall that my position with regards to programs such as STRUCTURE is very circumspect. I said: “Note that the clusters generated by structure…. And, in fact, my evidence for races — adopting Hartl and Clark’s definition — comes from analyses such as:
Tal, O. (2013). -individual genetic distance & Gao & Martin (2009). Using allele sharing distance for detecting human population stratification
Regarding STRUCTURE, my criticism was very narrow: Contrary to what you keep claiming, Rosenberg et al. 2002/2005 did provide evidence that could be used to justify K=5. Next argument:
(b) “They did not publish any statistical justification [Pr (X|K)] to justify any particular value of K over another…but Pr (X|K) is not the same thing as ∆K, nor of Rosenberg et al’s 2005 “clusterdness” statistic.”
Here you are just being disingenuous. My argument was very clear. I even explained the way to construct a counter. I will merely restate what I said:
“You dispute this … by arguing that [Pr (X|K) or] ∆K is a best measure of the “best” K and that the results for [Pr (X|K) or] ∆K are not give. Yet I gave a clear and unambiguous justification for using clusterability and degree of genetic differentiation as a measure of preferability: your own prior conception of what ideal races should look like///Now, you might argue that we should take into account [Pr (X|K) or ∆K, but in absence of this value we only have ones for clusterability and degree of genetic differentiation. Given this situation, you can only argue that clusterability and/or degree of genetic differentiation fails as a justifiable ways for justifying a K.
In response, you only pointed out that I neglected to include Pr (X|K) along with ∆K. But even so, nothing I said appears to be inaccurate. You still haven’t explained by clusteredness and highest among-part genetic variance are voiceless.
Next argument:
(c) “Rosenberg et al. 2005 (You appear to mix these two studies up) does not test values of K over 6.”
The confusion arose because I initially wrote:
“So, if Jennifer meant Rosenberg et al. (2002), the statement while correct is deceptive as we see in Rosenberg et al. (2005) that there is “statistical evidence that justifies picking 5 races instead of 7, or 4, or 2″ — as least as far as there can be.”
In response to: “So, when Rosenberg et al. (2002 …Rosenberg et al. never published any statistical evidence that justifies picking 5 races instead of 7, or 4, or 2….”
I was wrong because both Rosenberg et al. (2002) and Rosenberg et al. (2005) provide evidence that can be used to justify 5 versus “7, or 4, or 2”. Rosenberg et al. (2002) provides evidence for 5 versus 7 (lower among-part genetic variance than K=5) and Rosenberg et al. (2005) provides evidence for 5 versus 2 and 4 (less clusteredness than k=5). So, in fact, your statement was incorrect, not deceptive.
See — I admit when I make mistakes! (Unlike some others.)
“1) As the authors of STRUCTURE note in the program’s documentation, it is not designed to be applicable to populations that experience IBD. (This is true of human populations).”
You should probably let the creator of structure know that it is “not designed to be applicable . . . to human populations”. He seems to be confused on this point, considering the paper that introduces structure includes an example application to human data (and indeed a primary impetus for creating structure appears to have been the desire to correct for population structure in human association studies).
Pritchard is also a co-author on the 2002 and 2005 Rosenberg papers.
The issue of isolation by distance would be a reason to be cautious about over-interpreting clusters within Europe at a high levels of K, for example. Not, as Chuck points out, a reason to ignore clusters that comprise major continental populations with clear discontinuities between them.
“Which is why the primary/major/continental level races/varieties of Berneir (1688), Linnaeus (1735), Buffon (1749), Kant (1775), Blumenbach (1795), Cuvier (1828), Coon and Garn (1955), Gill (1988), Risch et al. (2003), and Rosenberg et al. (2011) look remarkably the same.”
All those you listed are still not the same. Garn attempted to solve the problem by arguing different classifications are a product of geographical hierarchy (continental > local > micro races) however classifiers cannot even agree on the largest divisions:
1. Russian anthropologists (Nesturkh, Cheboksarov) during the Soviet Union lumped Africans, Australians and most other people below or (what they regarded as to be) near the equator – as a single cross-continental “equatorial race”.
2. Christoph Meiners (who coined “caucasian race” in 1785) only recognized the existence of two races. Africans he clustered with the “Mongolians”, cross-continental.
3. Calvin Ira Kephart’s “Races of Mankind: Their Origin and Migration” (1960) proposes that Indians and Africans are part of the same cross-continental racial division.
“So there is a sense in which class and order are arbitrary groupings — as one could make an indefinite number of subdivisions or none– but there is a sense in which they are not, since members are arranged according to genetic similarity (genealogical for cadists and genotypic for evolutionary taxonomists).”
No this is false. Even what is (falsely) called “whole genome sequencing” virtually never covers 100% of the genome, but instead 95 – 99.9%. You cannot objectively determine genetic similarity between individuals without 100% of their genome sequenced.
The goal of taxonomy is to arrange organisms this way.
My point was that this arrangement, in principle, is not arbitrary in the sense meant by Darwin when he said: “this classification is evidently not arbitrary like the grouping of the stars in constellations.” That is, evolutionary-genetic relationship is a fact of the world. It is, however, arbitrary in another sense, since there is no ontologically privileged level of genetic analysis. As for the possibility of taxonomic accuracy, since genetic character is correlated (in natural populations), the probability of correct classification increases with the number of loci. Thus one doesn’t need to look at every genetic marker. The number one needs to look at depends on the amount of genetic differentiation between populations. Among humans this isn’t super large, so one has to look at a bunch of loci.
“look remarkably the same”
By “remarkably the same” I meant “remarkably similar”. I apologize for the confusion. A quick google search shows that other people similarly (mis)use the phrase. Regardless, the major/primary/continental taxonomies look “remarkably similar” (to me):
Linnaeus: Europaeus, Afer, Asiaticus, Americanus
Buffon: European, Negros, East Asian, Other SE. Asians & Polynesians, Tartars
Kant: White, Hindustani, Negro, Mongol
Blumenbach: Caucasian, Ethiopian, Mongolian, Malay, Americanus
Cuvier: Caucasoid, Negroid, Mongoloid
Coon and Garn: Caucasian, Indian, SS. African. East Asian, Polynesian, American, Melanesian
Gill: Caucasoid, Negroid, Mongoloid, Polynesian, American, Australoid
Risch et al.: Caucasian, African, Asian, American, Pacific Islander
Rosenberg et al. Europe, S.S. Africa, East Asian, Americans, Oceania
I keep seeing the big three. And the other divisions mostly make biological (how you lump and split populations) or methodological (anthropomorphic versus genetic measures) sense. I am hardly the only one who has made this point. .
“however classifiers cannot even agree on the largest divisions”
But, taking into account the method variance, data limitations, and conceptual differences (e.g., pre-Darwin versus post-Darwin), they reasonably agree across time. That is, the taxonomic consistency regarding major human varieties/races compares favorably with that regarding the divisions of many other species — which are under constant revisions. That should be our standard, no?
The point you are making is covered (or rather debunked) by Ridley (1993) in regards to species and race classification. He shows that even if there is 100% classification agreement by “independent observers” – this does not make species/races objective:
“The fact that independent observing humans see much the same species [or races] in nature does not show that species [or races] are real… The most it shows is that all human brains are wired up with a similar perceptual cluster statistic.”
“this does not make species/races objective”
But I was responding to:
“Folk notions of what constitutes a race and how many races exist are extremely variable and culturally specific.”
If all human brains were “wired up with a similar perceptual cluster statistic”, they would be less variable and less culturally specific, no?
Dear Mythos:
Please, stop for a moment to consider this: there is an objective genealogical reality underlying this. Genealogy is the least subjective structure of knowledge possible. For some reason, contemporary people have a hard time grasping genealogical structure, but if you are interested, here’s a good 2006 introductory article by Steven Pinker in The New Republic:
https://groups.google.com/forum/?hl=en#!msg/rec.arts.movies.past-films/bxmkikmoI2M/tkQwKVf6B0wJ
But the genealogy ties us together above the race level, doesn’t it? We all share certain matrilinear and patrilinear common ancestors that transcend the presumptive race boundaries, as far as I’m aware. (Please correct me if I’m wrong!) So why are we several races instead of one race stemming from mitochondrial Eve?
Our genealogies might be objective. Where you carve those genealogies to create “race” isn’t.
Hi Colin,
Fair question. In biology, races are generally understood to be divisions of a species. Thus, technically, there can only be zero or more than one races. (Since a partition results in more than 1 population, there can’t be 1 race in a species.) If you count archaic humans e.g., Neanderthals as different human races — instead of as different species — one could sensibly speak of the “modern human race” — though this wouldn’t exclude the existence of multiple modern human races at a finer grain of focus.
ok, not to spam ,but old post of mine http://blogs.discovermagazine.com/gnxp/2013/05/why-race-as-a-biological-construct-matters/
Having read your piece, I’m not sure I understand why you think race matters. For example, you say, “within groups like African Americans which are admixed there is variation in disease risk by genomic fraction.” Doesn’t that imply that “race” is nothing more than a subjective, flawed proxy for ancestry?
Race is ancestry.
Any two individuals share the vast majority of their ancestry; we’re the very tips on the evolutionary tree. HBD places a lot of meaning in the tiny gaps between the leaves, without much evidence to support it.
“Any two individuals share the vast majority of their ancestry”
Yes, but a lot what is shared is also shared with non-human animals, so your point is quite meaningless.
The clusters that are produced by cluster analysis contain individuals that are always more similar to one another than the similarity between a random member of cluster A and a random member of cluster B. A.W.F. Edwards made this point. He said that the possibility of the above not being correct decreases to 0% as the number of loci in the analysis increases.
I will now address the K=X argument:
If we were to do cluster analysis on a sample of humans and chimps, and we set K=2, then all the humans would cluster with one another, and all the chimps would cluster with one another, but ,according to you, the two clusters are (a) meaningless; and (2) not indicative of reality because the K is arbitrary.
But since the above conclusion is clearly wrong, the idea that the arbitrariness of K renders cluster analysis meaningless and wrong is false.
Your first argument is a reduction to absurdity. Every human shares the vast majority of their human ancestry with every other human; saying “race is ancestry” avoids the obvious question of where to draw the line. It’s not a small criticism. Several people have asked Steve Sailer that in this forum, and he doesn’t seem to have an answer.
Your second argument shows how flawed the Structure analysis is when it comes to defining races: the inputs determine the outputs. As for your specific example, I would say that the clustering results reflect the population divide between humans and chimps. The “race” result is nothing but a very flawed proxy for determining those populations and/or (as Sailer says) ancestry. But it’s such a flawed tool that I don’t see any advantage to relying on it.
“The “race” result is nothing but a very flawed proxy for determining those populations and/or (as Sailer says) ancestry. But it’s such a flawed tool that I don’t see any advantage to relying on it.”
Sorry, but are you referring to cluster analysis or race when you say “a flawed tool?”
—
“As for your specific example, I would say that the clustering results reflect the population divide between humans and chimps”
If it the arbitrariness of K is a non-issue for chimps and humans, why is it an issue for humans? After all, humans and chimps are just different combinations of alleles (and in some case different genes).
Race. I’m not expert enough to have an informed opinion on cluster analysis.
I may have misused the terminology when I referred to the “clustering results” w/r/t an analysis of humans and chimps. It’s probably clearer to point out that we can identify objective differences between humans and chimps, not least that (I assume) no matter what we make K chimps will always stand out from human populations in a Structure analysis. Is that correct?
“I may have misused the terminology when I referred to the “clustering results” w/r/t an analysis of humans and chimps. It’s probably clearer to point out that we can identify objective differences between humans and chimps, not least that (I assume) no matter what we make K chimps will always stand out from human populations in a Structure analysis. Is that correct?”
Don’t know. i am a layman.
” It’s probably clearer to point out that we can identify objective differences between humans and chimps”
Same is true of human races. Is the difference in allele distributions between “populations,races,x) not an objective difference?
Is the difference in allele distributions between “populations,races,x) not an objective difference?
The difference in allele distribution isn’t in question–what is is whether groups break down in a way that’s detectible and maps to anyone’s conception of race groups. The answer seems to be “no,” given how many different definitions of “race” are floating around.
“saying “race is ancestry” avoids the obvious question of where to draw the line.”
I would say that where you draw the line is arbitrary, but all lines are correct in the sense that everyone encircled by the line is more similar to one another than are a random member encircled by that line and someone outside of the line. This was proved by A.W.F. Edwards.
Is this how you’re defining race?
“Is this how you are defining race?”
yes, hsu did the math.
http://infoproc.blogspot.com.au/2007/01/metric-on-space-of-genomes-and.html
Asking where to draw the line is the same as asking what value for K, so this follows:
“K is arbitrary but all values of K are correct I would say that where you draw the line is arbitrary, but all lines are correct.
Here is why:
Everybody inside of cluster A has more similarity than a random member of A has with a random member of cluster B.
BUT– and please grasp this — it’s possible for members inside of cluster A , call these individuals A.1, to have more similarity than a random member of A.1 has with a random member of A*(A* is A – A.1).
If one wanted A.1 to form its own cluster, then one would simply increase K.
Let me give you an example:
1. Two random Amerindians have more similarity than a random Amerindian has with a random Mongolian.
2. A random Amerindian and a random Mongolian have more similarity than a random Amerindian has with a a random African.
Both statements are correct, but they arise under different values of K.”
*CORRECTION* ““K is arbitrary but all values of K are correct I would say that where you draw the line is arbitrary, but all lines are correct.
“The difference in allele distribution isn’t in question–what is is whether groups break down in a way that’s detectible and maps to anyone’s conception of race groups.
Yes, they do. See
http://infoproc.blogspot.com.au/2007/01/metric-on-space-of-genomes-and.html
The answer seems to be “no,” given how many different definitions of “race” are floating around.”
Nope. Here is the definition of race.
http://infoproc.blogspot.com.au/2007/01/metric-on-space-of-genomes-and.html
Yeah, that’s not “the” definition of race. That’s a definition of race. There are others, from prominent race theorists, in this thread alone. (See Steve Sailor’s genealogy comments, for example.)
This is a masterpiece of blogging. Written by a scientist who reads the papers, and understands them.
Thank you very much indeed
Roundup of all reviews of Wade’s book at:
Link removed for spam. –JR
I think Jennifer is missing the point. It doesn’t matter how many races there are. As someone else noted, just because someone can’t list how many shapes there are or how many colors there are on the color spectrum, it doesn’t mean that shape or color don’t exist.
Seriously, the “race is a social construct” idea is one of the most ridiculous ideas ever put forward. Chinese geneticists 100 years from now will be laughing that silly Americans actually believed this.
Shapes and colors can be unambiguously, objectively defined–geometrically in the case of shapes, and by wavelengths of light for colors. How do you objectively define races?
If there isn’t an objective way to discern one race from another, then doesn’t your statement devolve back to, “I can divide people however I choose into arbitrary groups”? I agree, you can do that–but I don’t see why it matters.
You lump or split them based on inferences about genealogy:
http://isteve.blogspot.com/2014/05/the-race-faq.html
Why inferences about genealogy? If there is a fundamental, underlying significance to race, wouldn’t it leave objective genetic evidence?
These aren’t rhetorical questions–I’m genuinely having trouble understanding the HBD position. Everyone seems to agree that race is impossible to define and determine objectively, and it seems effectively useless as a concept unless loaded down with preconceptions.
People did not have the concept of wavelength centuries ago, therefore colors
could not be objectively defined by them. Did color matter to those people?
Could color be a meaningful thing to them?
This is a really good response. My first response is that the fact that people weren’t able to put their finger on the objective reality of color doesn’t mean there wasn’t one–their categorization was still founded on a real distinction, albeit one they couldn’t articulate. That’s different from “race,” which isn’t founded on anything other than our social constructs (or, alternately, is just a synonym for “population”).
Moreover, without objective criteria, all it means to say something is “yellow” or “blue” is to say, “that thing looks yellow or blue to me.” If you can’t access or define the objective reality of the category, isn’t it just subjective?
If you can’t access or define the objective reality of the category, isn’t it just subjective?
Kant answered this in the Critique of Pure Reason. Unfortunately, I never quite got through it.
Then can we be sure he ever wrote it?
To say that something is a social construct is different than to say that it doesn’t exist. Indeed, whether or not something exists (and how) and whether or not something is socially constructed (and how) are separate issues. Saying that color, for example, is socially constructed doesn’t mean that color is just made up or non-existent, but rather that our understandings and perceptions of colors are mediated by social interaction. Like it or not, humans create categories — categories don’t create themselves or exist as independent properties of the objective world. It is therefore completely legitimate to ask whether particular social categories correspond to the objective world or correspond to some other factor (like culture). Race, especially as it is used in everyday life, does not appear to be a social categorization scheme that accurately corresponds to the objective world — and this is the conclusion of geneticists, I might add.
All categories, including the most scientific ones, are mediated through social interaction. So you get the unhelpful result that all categories are “social constructs”.
I think the conclusions of geneticists, including Dr. Raff, are that genetic variation is continuous, not discrete. The assumption is that folk concepts of race don’t realize this, but I am not sure whether that is actually true. It might have been true of race in the Jim Crow south, but not in Brazil, and maybe not today in America.
If you can’t define what you mean by race in clear, unambiguous, terms, then how can I know what you are talking about? Justice Stewart’s (U.S. Supreme court) ” I know it when I see it” (with reference to pornography) makes for very difficult arguments, for it makes every person’s interpretation equally valid-or in other words, there is no definition. That seems to be the argument with race-“I know it when I see it”. If you can’t define what you mean, then I can’t test what you mean. If I can’t test it because it is inherently untestable (because you can’t define the concept)-then it is outside the purview of science. Science relies on fixed definitions-so that we all know what we are talking about. If you can’t define race as a set of observable traits, with fixed definitions-science is out the door.
@ Scott,
They manage to do so, indeed need to do so, for the purposes of medical and pharmaceutical research.* It is a nuisance for GWAS — you have to correct for population structure as a possible confound for the real effect you are looking for.
* http://genomebiology.com/2002/3/7/comment/2007
Using “black” is less accurate than specific ethnicity which is less accurate than using individual traits in assessing risk factor. There might be more “black” people on average with a specific trait that has a specific risk factor for a specific medical issue but that does not mean all black people have a risk factor for it. Millions and millions of “black” wont have that trait(s)/or another trait(s) and thus very different risk factor which is most variable by individual. What that paper promotes is using something more risky and less accurate. Those averages are constantly changing too, very rapidly. Its simply bad science, you people need to accept it.
Humans are not spread out into a tree its a trellis, a rapidly changing one at that.
It seems that the differences we see are really just micro-variations, ripples on a deep ocean of shared genetic material. To address the colour analogy mentioned above – even though the number of potential colours is infinite, we can measure and explain exactly how each colour differs from every other colour – namely through the wavelength of the light. It sounds like no such measurement exists for differentiating races. We can measure the genetic difference between two people, but it sounds like there is as much diversity within ‘races’ as between ‘races’. To be more precise, despite all of our phenotypic diversity, there are no sub-species of homo sapiens.
My question (mostly to Jennifer, but to anybody who wants to answer) is, how can we talk about these micro-variations in a meaningful way, without bringing Social Darwinism and other cultural-political baggage into the discussion? Certainly some of these micro-variations are worth discussing, no?
It would be useful and helpful to approach “these micro-variations” that are commonly known as racial differences, as well as all other recognizable variations, from a perspective of appreciation for the richness that diversity brings to our lives.
By this, I do not mean that we should hold one or a few groups above other groups, because each group has “gifts” to offer humanity as a whole. Some of these “gifts” might be more highly valued at some times or in some situations, while others might be more highly valued at other times or in other situations. A broad and wise perspective would never put down what any group has to contribute, because it would include the recognition that in the long term, all contributions to the richness of life will be valued, each in their optimum time and place.
It is useful, though, to recognize that some groups do seem to have special needs that we can and should all help them meet. So, for example, certain illnesses and diseases seem to be associated with particular groups who therefor need appropriate health services to help them deal with these issues. The same applies to gender related health issues, and although we tend to feel we recognize males and females when we see them, gender is not always clear cut either. We continue to deal with many inequities based on gender that get in the way of our ability to truly appreciate the richness that gender related diversity brings us in life.
Much of what we find to be contentious in life could, by simply changing our perspectives and attitudes, instead become enjoyed and appreciated as enriching our life experience. Taking this approach would greatly raise our quality of life in countless ways. I think we ought to try it.
This is a crucial question, I think. To often the issue is (inaccurately) characterized as being a choice between two strawmen: a) humans are divided into bounded races based on objective biological properties or b) humans cannot be thought of as more or less biologically similar in any way. (More bluntly, either races are real biological types or they are totally made up with no reference to objective reality). I think the vast majority of physical and social scientists are no where near these two extremes, though, and that most agree that while there are patterns of greater and lesser relatedness, these patterns don’t reduce down to bounded groups (as race as historically been used and understood).
One way to address the micro variations in genetic similarity without dragging along the baggage of racial essentialism, social Darwinism, etc. is to stop using language that suggests that these micro variations in genetic similarity can be understood as constituting “groups”. We would be clearer and better served by a vocabulary for genetic (dis)similarity that did not imply that people who are placed in the same social category or have a particular degree of relatedness are part of the same “group”. Calling a set of people a “group” semantically evokes clear boundaries, ingroup and outgroup social dynamics, shared identity, mutual orientation, and all sorts of other things that aren’t directly relevant to variations in genetic similarity. We can and should ask questions about genetic relatedness among humans, but when we talk about that genetic relatedness in groupist terms we risk assuming that genetic relatedness corresponds to social groups or simply confusing the two.
“how can we talk about these micro-variations in a meaningful way?” Take a look at the anthropological genetics literature–this is exactly what our field does.
I’ll try, but I don’t think I’ll jump right into the journals. What book(s) would you recommend for a basic introduction to the field?
Comparing my field (linguistics) to yours, classifying languages at the macro level is relatively simple and straightforward. Even a language like English, which has a heavy Romance influence from both Norman French and Latin, is uncontroversially classified as a Germanic language. Defining ‘race’ is more akin to the micro perspective in linguistics – defining or counting dialects (as someone mentioned above). How many dialects of English are spoken in North America? It’s impossible to answer, and is essentially a useless question – but that doesn’t mean that dialects don’t exist. The better question is, how does speech/language differ in this area from that area, or in this social group from that social group?
With this understanding, linguists use the word dialect in a relative way. We can talk about the American dialect of English, the New York dialect, the Brooklyn dialect, an ‘upper class’ dialect, etc. It’s relative to whatever we’re comparing it with. American can be compared to Canadian or British. New York can be compared to Boston, Brooklyn can be compared to the Bronx. In everyday usage, the word ‘race’ doesn’t usually have this flexibility, however, and can only be used to describe the extreme macro level (ie. roughly equivalent to language families, or even language stocks, in linguistics). Maybe ‘clan’ would be a better choice – we could talk about the Asian clan, the European clan, but then also the South Asian clan, the East Asian clan, the Gujarati clan, the Han clan, etc, right down to the level of actual traditional clans, and even individual family trees. What terms are used in anthropological genetics?
While Wade clearly reaches too far with his speculations and conclusions, his basic premise – that there are statistically significant genetic and phenotypic differences between certain groups of homo sapiens – is clearly true, right? Even if we can’t be divided into subspecies, we clearly differ based on our regions of origin, and presumably genetic markers have been found that correspond to at least some of the physical traits that forensic anthropologists and others use to approximately determine someone’s ancestry. To me, the biggest issue is the term ‘race’, which is simply not refined enough, and too politically loaded to be useful in this conversation.
Right, race is relative, which isn’t surprising because, fundamentally, it is about who your relatives are.
Oops, also just saw your further reading list at the end of the blog post, so ignore that question…
Addendum to my previous comment: the word ‘clan’ would be defined as it usually is, meaning a group of people thought to be descended from a common ancestor (or possibly small group of ancestors).
OK, I see now that human ancestry is, for the most part, a trellis, not a tree, so by the definition of ‘clan’ above we are all part of the same clan, until very recently in human history. That is, we all share a huge number of common ancestors due to admixture.
Templeton’s 2013 paper in Science is what helped me understand this:
http://www.sciencedirect.com/science/article/pii/S1369848613000460
Science relies on fixed definitions
then we shouldn’t talk about species, should we? BSC is not universally accepted, and even when it works well (e.g., placental mammals) it isn’t definitive and crisp.
Yes but human’s are not like other animals. They don’t have a say in where the line is drawn around them but humans do and that can have very bad affects on both sides of said line…. and the poor sods in the middle of it.
Its not as simple as lions and tigers. You know that, Wade knows that, but you don’t want to realize it.
Classifying people into groups based on anything, especially shared average differences in the exact same things IS EXTREMELY STUPID. Human beings are not like dog breeds, or any other animal.
Humans are way too complicated in psychological traits and too much of it shared for these sort categories to be useful. It will only cause a massive fight.
Separating people based on some temporary average difference in the number of jocks vs nerds is ridiculous, does not matter how genetic or not.
Human beings are not like dog breeds, or any other animal.
well, in a genetic sense i think they are.
Eh, only in the very loosest sense (to the point that I don’t really think it’s a useful comparisson). There’s meaning in the fact that TreeMix diagrams for dogs are so spiky, but very spindly for humans.
In other words no they are not like dog breeds in a genetic sense. Plain and simple. They are not like monkeys or chimps(who are way more diverse), which “some” people think they aren’t.
http://www.ox.ac.uk/media/news_stories/2012/120302.html
I don’t know how we can compare the reality of “species” as a category to that of “race” if we can’t define what “race” is. I can’t do it any better than Wade can. Can you? What are we talking about when you refer to “race”?
And how many different definitions of species are there? How often have there been major legal and regulatory battles over whether some local variant qualifies under the Endangered Species Act?
The definition of “species” can become fuzzy around the edges, but that doesn’t make “race” any more solid. In fact, it suggests just how absurd it is to pretend that “race” has an objective, biological reality when even species aren’t always clear.
The only thing that is certain in human affairs is that everybody has a family tree of ancestors with one father, one mother, two grandfathers, two grandmothers, and so forth. Thus, it makes sense to think in terms of extended families.
How does that support categorizing humans into races? What differentiates one extended family grouping from another?
Colin, if you read about “clades”, you’ll find “species” is a much fuzzier concept that we used to think. And Steve, one father and one mother, yes, but positing two separate grandfathers and two separate grandmothers is just the most common case, hardly universal.
Colin, if you read about “clades”, you’ll find “species” is a much fuzzier concept that we used to think.
I’m aware that the concept is fuzzy around the edges, as I said. But again, that doesn’t make “race” any more solid a concept.
“And Steve, one father and one mother, yes, but positing two separate grandfathers and two separate grandmothers is just the most common case, hardly universal.”
Everybody has four separate slots for grandparents on their family tree. Whether four separate individuals fill them is another question. If you go back 40 generations, you have over a trillion slots to fill, but there weren’t a trillion people alive back then, so somebody had to do double duty. That’s what makes racial groups: partial inbreeding.
While the notion of species has been challenged of late, due to the fact that in some relatively rare cases, certain species have been known to interbreed with other species, all that means is that the definition needs to be slightly modified: e.g.: “A species is a group of living organisms consisting of similar individuals capable of exchanging genes or interbreeding, almost, but not necessarily always, exclusively with one another.” Does the necessity of clarifying in that manner necessarily call the whole notion of “species” into question? And if it does, then how does that somehow validate race?
Razib, if you think you can produce a scientifically acceptable definition of race, of any kind, either precise OR vague, then by all means post one. Otherwise, in the immortal words of Wittgenstein, “remain silent.”
“how can we talk about these micro-variations in a meaningful way, without bringing Social Darwinism and other cultural-political baggage into the discussion?”
This sums it up. The motivation to deny biological reality of race is politically motivated by “Cultural Marxism” (as it was for Gould and Lewontin), not scientifically motivated. Don’t get me wrong. I do not like racism. (I’m actually Chinese-American.) But I doubt you need to worry about a resurgence of racism in this day and age. And using these Marxist notions in science is very dangerous!
Razib is right. If the same analysis used against race is used against other concepts, they also could all fall apart. Many definitions in biology would fall apart under this irresponsible method.
As I recently told Jerry Coyne, the “race is a social construct” notion is not only a lie, it’s a reprehensible and pernicious lie. And one that needs to be destroyed.
So if race isn’t a social construct, what is it? How do you define it in an objective way?
A racial group is a partly inbred extended family.
Would you define this in mathematical terms, please? How much variation?
Splitting and lumping questions are highly relativistic — they depend upon what your needs are at the moment. For example, are the Amish a racial group? They are descended from about 200 founders around ten generations ago, with little inflow of genes from converts. For Census and affirmative action purposes, the U.S. government considers the Amish part of the white race. On the other hand, they have developed their own profile of congenital diseases, so if you are a doctor attempting to diagnose an ill Amish child, you should look up their racial tendencies in terms of susceptibilities to genetic defects.
Try thinking about the general problem of racial categorization from the bottom up rather than the top down — it provides a valuable new perspective. Start with families. How much genetic variation is their among siblings? Well, compared to identical twins, quite a bit, but less than among first cousins. And how much variation is their among first cousins? More than among siblings, but less than among second cousins. And so forth and so on.
What makes a racial group a special case of extended family, however, is that their family trees show some degree of inbreeding.
In other words, there’s no objective definition?
How much variation is there for races/sub-species etc in other species?
http://infoproc.blogspot.com.au/2007/01/metric-on-space-of-genomes-and.html
Hsu did the math.Trust an Asian to come though with the math 😀
Layman– please review my comment policy. I encourage discussions of race on this thread, but your “joke” is based on a racist stereotype, and that’s unacceptable here.
I have no idea what that means. It’s far too vague. And it fails to take either genetics OR morphology into account.
I read and teach Marx in an academic setting on a regular basis. That said, I find one of the strongest indicators that someone doesn’t understand the first thing about Marx’s works or academic Marxism is use of the phrase “Cultural Marxism”. First of all, that phrase already means something within (post/neo-)Marxist scholarship — and that meaning is not ‘trying to impose leftist political goals through vague cultural/intellectual means’. But more crucially, the use of the phrase in the latter sense is actually antithetical to Marxist historical analysis, so it makes it kind of obvious that the speaker is criticizing something without understanding its most basic premises.
Once again Jenn, you bring Scientific Intelligence forward. The common reader is NOT as informed with the clear understanding of genetics that you possess. Debunking information and naming it correctly (pseudoscience) is a hallmark of your capacity to be a voice of integrity within Science Educators.
Thank you so much, Lyn 🙂
Well argued. I think Jared Diamond’s Guns, Germs, and Steel also made excellent arguments as to how different races developed differently in the context of environmental determinism. While this is not likely all the answer, it is a better answer than the idea that one particular race is genetically superior to another. As you correctly point out, the genetic differences among races are not clear or consistent, so it seems likely that other factors are relevant in charting the success or failure of a group of people.
If I showed you 100 pictures — 50 Koreans & 50 Australian Aboriginals — and asked you to label their race correctly, how many times do you suppose you’d get it wrong? Guessing that # is close to 0, but sure, it’s not a “thing.”
As Dr. Raff said, no scientist denies that there are genetic variations between people. The variations you select, however, are arbitrary and overly simplistic. Skin color varies within a group of people within an area or tribe or group or nation so much that its usage as a “racial” definition is pretty much useless. You can find people with the same skin color from India to Australia to South America to Africa. Do you know what they all have in common, aside from melanin? Their ancestors all grew up in a part of the world that had high UV input. That’s it. Does that mean to you that they are all the same “race”? You picked people from quite different latitudes (and solar UV input), so of course they have different skin colors, making your example cherry-picked. You picked people with and without epicanthic folds, as well. How convenient for you. How about looking at the variation across skin colors and eye folds within Korea? How about doing the same in Australia. NOW, how about instead of cherry picking in your favor, we select native Koreans and Australians who actually look closer to each other than to the stereotypical Korean and aborigine? Those wide variations are what prove Dr. Raff’s point, not yours. The variations within populations are so large that they overlap with other populations more than they draw distinctions, making race an arbitrary choice (or “thing” in your “jargon”), not a scientific “thing”.
If something can be sorted with the human eye with close to 100% accuracy, it’s empirically non-trivial. And what is with constructivists and skin color. The races vary in many other ways. An aboriginal could be an albino and still be easily distinguishable from a Korean. As for the visually ambiguous cases, I’d ask you to produce examples of those if you could, but even if 1 out of every 100 case was ambiguous, that doesn’t negate the “thing.”
… and the point is, this one can’t be in general.
It absolutely can. If we sort humanity into 5 groups, how many ambiguous cases will result? Surely someone can produce a Dane who looks like a Maori or a Han who looks like a Cape Verdean.
Generally speaking, a sophisticated, worldly individual can sort people they meet into continental-scale races with an accuracy rate of at least 90%, and probably 95%. Richard Dawkins offered a thought experiment:
“Well, suppose we took full-face photographs of 20 randomly chosen natives of each of the following countries: Japan, Uganda, Iceland, Sri Lanka, Papua New Guinea and Egypt. If we presented 120 people with all 120 photographs, my guess is that every single one of them would achieve 100 per cent success in sorting them into six different categories… I haven`t done the experiment, but I am confident that you will agree with me on what the result would be.”
Personally, I think a lot of people would get some of the Papuans and Ugandans confused at first. After some experience, though, you’d make fewer and fewer mistakes.
It really astonishes me how we have these wonderful instruments to aid the human perceptions- senses which can be fooled easily, but people insist upon letting themselves be fooled. When arguing the science of genetics, the proof must use genetics, not sorting with the human eye. The human eye cannot see DNA, MW. Phenotype is not genotype.
First, define race – use genetics, not just superficial traits like skin color. (See my comment above why skin color and eye shape are not the same as race. You did not refute my argument, you ignored it.) Then second, prove your assertion that what you defined as races can be separated, not just by the unmodified human eye but using tools that cannot be fooled. These are not minor details. These cannot be assumed as true or accepted as axiomatic. No scientist would let you make these claims without proof. So prove.
The 21st Century genome studies have very largely validated the accuracy of mid-20th Century physical anthropology studies of race based on traits that could be measured with calipers and the like. For example, Carleton Coon’s 1965 bestseller The Living Races of Man, the last major work of the pre-genetic era, stands up quite well today. It contains a few surmises that have been falsified — e.g., the Ainu of Northern Japan don’t actually have much Caucasian ancestry, as Coon theorized — but those are closer to the exceptions that show how much Coon got right.
Steve- which genome studies? Citations, please.
Google Razib or Dienekes for blogs that have been keeping the public up to date on the science of population genomics for most of this century.
Go take some less visible genetic characteristics. Go take people from 50 places around Africa and 50 places around Europe and Asia, and sort them into, say, 10 groups. If you look at human evolution, there’s generally been a lot more variability among the people whose ancestors stayed in Africa than the ones who left (especially if you only count the ones who left later, not the ones who became the Neandertals.)
And while you’re at it, go compare African-Americans to some of those people.
First, races can’t be “sorted with the human eye with close to 100% accuracy.” Anyone with experience with eyewitness testimony will tell you that your expectations heavily affect your perceptions; if you know someone identifies as black, for example, it will seem obvious that they’re black to you. But if you don’t know someone’s racial identification, it can often be difficult to sort them mentally based on their appearance. In other words, it feels easy. That doesn’t mean it’s actually something people can do with great accuracy.
For example, you claim that it’s easy to distinguish races without skin color, but Walter White shows how slippery that identification can be. Both of his parents were born into slavery, but I wouldn’t have guessed that based on his picture.
Second, obviously people can easily pick out significant morphological differences. We can say this guy’s black and that guy’s white because they have very different skin tones–we can even objectively show it with a pantone chart. But that’s not race! All you’re identifying there is skin tone. Two different people with black skin can be–and likely are–more different from each other than either is from an average white person.
“For example, you claim that it’s easy to distinguish races without skin color, but Walter White shows how slippery that identification can be. Both of his parents were born into slavery, but I wouldn’t have guessed that based on his picture.”
I’m not sure a person with mixed ancestry is the best example to use (although he sounds like an interesting fellow). I picked Korean and Australian Aboriginals to highlight situations where there is literally zero ambiguity (JerryA claims he can find a person from one of those populations who looks like the other, but I’m not holding my breath). If this was all so arbitrary we would not find ourselves with cases where there is literally no way to mistake one group for another.
It sounds like all you’re claiming is that you can tell two people apart–I think that’s obviously true. And some features are inheritable–again, that’s obviously true. But how do you extend that out to a concept of “race” that isn’t just drawing arbitrary lines based on a few arbitrary visible characteristics?
If set A & set B are mutually exclusive taxonomically and our set of sets is operating on a gradient, it stands to reason that somewhere very specific on that gradient, is the point at which one can draw a line between the two sets (which will catch other sets along the way). Given that we have multiple sets that are mutually exclusive (Danes and Nigerians, as another example), we can keep drawing lines until we have a maximal # of hard edges, and that will give you a rough estimate on the # of meta-sets (or “races” for our purposes) we are dealing with. Probably turn out to be about 5 or 6, I’m speculating.
So how do you determine the boundaries of the sets? Please be specific–this is the sort of thing that people assume is possible, but I think is probably not.
Well, since people are holding definitions to such high standards, I’ll just say 99%: if you can’t disambiguate individuals from two groups with less than near-perfect accuracy you don’t have a formal set boundary. I’m looking to maximize the # of hard edges since people seem to be quite enamored with fuzziness at the moment. I’d say a 99% heuristic would let you do things like separate all of Asia from most of Oceania (& most of SSA, now that I’m thinking about it).
So is race based entirely on physical appearance? And it really is an “I know it when I see it” phenomenon?
That’s been my operating assumption from the start. If you have set A and set B, the clinalists would have you believe that the differences between groups are so subtle that total disambiguity begins right at the edges of the sets in question and nowhere in between. That is obviously false as you can easily disambiguate A’s neighbors from B’s neighbors with the same reliability that you can disambiguate A and B. If that is not valid taxonomy, then knowledge does not exist.
I’m glad you’re trying to establish some definitions (I’m a lawyer, so I hate having conversations without them!), but I’m not sure this one works. If we take two African populations that people can’t visually distinguish with 99% accuracy–call them A1 and A2–then they’ll likely be more genetically distinct from one another than either is from a Korean or Finnish population. That is, the visual signs of “race” would lead most people to say that A1 and A2 are the same, but their underlying genetics don’t support that judgment.
“If we take two African populations that people can’t visually distinguish with 99% accuracy–call them A1 and A2–then they’ll likely be more genetically distinct from one another than either is from a Korean or Finnish population.”
I’m sorry, I hear that kind of thing all the time, but it’s wrong.
Can you elaborate?
Impossible .
The definition of a cluster is that everyone inside of it is always more similar to one another than are a random member of the cluster and someone not in the cluster.
So if you concede that A1 and A2 exist, then it’s mathematically impossible for a member of A1 to be closer to a Finn, who is not in A1, than with a random member of A1.
http://www.ncbi.nlm.nih.gov/pubmed/12879450
Isn’t that assuming your conclusion?
Read what Dr. Raff says: “Groups that live close together are more closely related to each other (and more genetically similar) than they are to groups farther away.”
So is that the definition of race? Does it require that members of a race live close to one another? How much difference in their genes is permitted before they stop being members of their parents’ race (or races)?
No one–no one, no one, no one–is denying that genetic variation exists between individuals and populations. The question is whether there is anything such as “race” that exists as more than an “I know it when I see it” subjective judgment call. If not, it puts a large hole in the idea that behavioral variation tracks our folk ideas about race.
As I’ve said before, the definition I’ve been using is whether you can tell populations or groupings of populations apart with 99% accuracy. If this were any other species on the planet, that would be sufficient to evidence to establish the existence of subspecies. What is it about humans that makes them so special?
Like every racialist, you’re free to invent your own definitions of “race”; you pretty much have to, in fact, since it’s such a fluid and poorly specified concept. Yours happens to be explicitly an “I know it when I see it” definition, or more precisely, “I know it when 99 people out of 100 see it.” I don’t think that’s a particularly useful or reliable concept, but who knows? Maybe your research will revolutionize the field of biological anthropology.
As an aside, while we’ve established I’m no biologist, I’m pretty sure that “99% of people can tell them apart” is not how we determine subspecies. 100 out of 100 people could distinguish Great Danes from Chihuahuas, but they’re both Canis lupus familiaris.
“NOW, how about instead of cherry picking in your favor, we select native Koreans and Australians who actually look closer to each other than to the stereotypical Korean and aborigine?”
The reason why there’s a stereotypical Korean and aboriginal look is because they’re members of different races. Think about it….
That’s a circular argument. There are differences between people in different parts of the world, adapting to local conditions. When scientists tried to define those characteristics genetically, they found that the differences within a group were wider than the differences between groups. Hence, not subspecies, i.e. not races. You try to define a race scientifically and get it published in a peer-reviewed genetics journal, I’ll wait. (And wait.)
Dude, you’re wrong.
…with even as few as 20 randomly chosen genetic markers. Two Caucasians are more similar to each other genetically than a Caucasian and an Asian.””
Risch et al
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC139378/
woops, you said native Australian. Anyway, the same argument is true regardless.
Are there any full-blooded Koreans who look like full blooded Australian Aborigines? I’d love to see pictures.
“NOW, how about instead of cherry picking in your favor, we select native Koreans and Australians who actually look closer to each other than to the stereotypical Korean and aborigine?”
Google Image Search should help you find some, if they exist. I look forward to seeing them.
As a sociologist who specializes in racial and ethnic categorization and its consequences, I thank you for this clear expression of the social constructivist stance on race. While it seems at a least a few of your readers could use some remedial sociology/socio-cultural anthrolopogy, I’m always pleased to be reminded of the complementarity of our respective fields.
Fantastic post. To address the “color” example brought up in the comments, anthropologists showed ages ago that color definitions are also socially constructed. That doesn’t make them less real, it just means that the categories humans put light into are not clearly defined by our biology or the nature of the light. Same goes for races: socially constructed ideas of race are definitely real things that affect how people live and identify themselves. Our biology doesn’t determine how we define races, our cultures do, which is why a race definition is so hard to pin down.
-Why couldn’t one reinforce the other?
Well, for one point, there have never been any genes identified as “adapted for success in capitalism”. They have never been discovered, just asserted without evidence by Wade. If I had not read this today about a modern book, I would have thought it came from the early 1900’s before DNA was found. That’s not science at all, that’s just bad reporting.
I don’t see how your comment is helpful at all. Are you denying that time preference is heritable to any degree?
Is there any evidence that it’s a heritable trait at all?
Well, there is a result about dopamine genes : those with certain variants of the genes are (statistically speaking) more impulsive and more seeking immediate rewards. The variants are distributed in different frequencies in different human populations, in addition.
Wade would first have to show that genes “adapted for success in capitalism” even exist. Otherwise, it’s just a baseless assertion (a clueless statement made without evidence) that can be dismissed as nonsense. Honestly, if I did not know this book was written recently, I would have thought it came from the early 1900s before DNA was discovered. It is that backward.
Did you actually read Wade ? By “adapted for capitalism” he clearly explains that the mechanism is via unusual degree of intelligence amongst Ashkenazi Jews. Intelligence correlates with things like time preference, cooperativeness, trust, etc. which would be characteristics conducive to capitalism.
Yes, and it’s all speculation and hand waving. What alleles determine ‘time preference’? Please be specific.
In that post, I was replying to JerryA, not to you. But to your question : are all inferences of evolution by natural selection that are generally agreed upon, also accompanied by allele identification ? Or is this just a special requirement for inferences of evolution by natural selection when it comes to human behaviour, especially group differences therein ?
Why do you want to draw a line in the sand over a technical question that will undoubtedly be resolved with a few decades?
How do you find alleles without first speculating that they exist?
Asking for you to provide evidence to back your ideas is not “drawing a line in the sand”. It is the basic minimum level of reality demanded by any scientific field. Otherwise, you’re… what is the proper phrase? Oh yes, you’re making shit up. (That’s the less polite version of ‘speculation and hand waving’, but I’m not as nice as Dr. Raff.)
You could until a few years ago have asked what alleles determine height? Nonetheless, it’s pretty well accepted that height is a heritable trait. It also appears there are differences in frequency of those alleles between North & South Europeans.
Similarly, behavioural traits are heritable too – even if the exact genes haven’t been identified. As the cost of genome sequencing falls they will have a better idea.
Mendel established solid evidence for the heritability of height (at least in pea plants). Where’s the evidence for the heritability of “time preference”?
It seems like the Sailer/Wade position is to speculate that some group that we can’t define has a predisposition to some genetic features we can’t identify that determine “time preference,” suitability to capitalism, and other behavioral traits. Is that accurate? It’s not a rhetorical question, I’m honestly trying to understand the argument—since everyone seems to agree that it’s impossible to objectively define race, I want to know what it is that Sailer and Wade claim exists.
Colin
Here’s a blog post reporting on a study that shows that impulse control –obviously related to time preference — is significantly heritable (0.5, with .38 additive, .12 non-additive genetic effects).
http://blogs.discovermagazine.com/gnxp/2012/08/the-heritability-of-impulse-control/
One thing one comes to realize by studying these areas is that almost any cognitive or emotional trait you might think of is in fact heritable.
It may seem absurd at first blush to think that might be so, because it may seem that a very specific trait — such as greater inclination to be divorced — can’t possibly have a set of genes devoted to such a function. But that just misunderstands how genes work. Genes work on basic processes in the body and brain. For the most part, we don’t understand these basic processes. But these processes have effects on all kinds of traits, a great many of the traits being only partially determined by the processes–the other factors deriving from something in what is loosely described as the environment.
In the case of cognitive and emotional traits, it’s probably the case that there is a relatively small number of more fundamental cognitive/emotional traits affected by genes that in turn affect more complex traits, such as an inclination to be divorced. One may imagine that, say, impulse control is a more basic trait that underlies that inclination, as well as many others.
So while it may again seem absurd to say that an inclination to divorce has an important genetic base, it’s clear enough how it may come about.
And one can see more generally how all kinds of cognitive/emotional traits can have a genetic base, and how indeed virtually any such trait will demonstrate SOME degree of heritability.
Right.
There is an enormous gulf between one writeup of one paper discussing the partially heritable nature of a trait “related to time preference” and the assertion that “almost any cognitive or emotional trait you might think of is in fact heritable.” It certainly sounds as if you have a conclusion in search of the evidence to support it.
What I gather from your response is that no, there is no significant evidence that “time preference” is heritable; there is some evidence that a related trait is partially heritable. That analysis found the effect was strongest in children, not adults, and no more significant than environmental factors.
The HBD position is speculation piled atop speculation.
Additionally, assuming that “time preference” (or impulse control) is inherited in any pattern that maps to racial groups?
Sorry, flubbed that one – I meant to ask whether, assuming “time preference” or a related trait is inheritable, it is actually inherited in a pattern that maps to a racial group?
It pretty much has to, although the differences could be small enough to be undetectable. The odds that the frequencies are identical across racial groups are essentially zero, and the odds that different genes created completely identical results are even lower.
“It pretty much has to” sounds a lot like, “I don’t know but I really want the answer to be yes.” Why is it yes? How do you know? Your logic would mean that every inheritable trait would post hoc map to the racial groups. But, if as other “race realists” have said in this forum, races are defined by how they look, why would behavioral traits correlate with things like skin color?
There’s simply no mathematical way that a large number of genes spread across millions of individuals could exactly mirror another group without the hand of god. It would be proof the intelligent design people are right if you could prove that they were the same.
Imagine taking two groups of a million people and trying to find an exact match for height in each of the groups, down to, say, a micron. It’s not theoretically impossible, but it is practically impossible.
You don’t need the same allele to do the same thing.
I read and teach Marx in an academic setting on a regular basis. That said, I find one of the strongest indicators that someone doesn’t understand the first thing about Marx’s works or academic Marxism is use of the phrase “Cultural Marxism”. First of all, that phrase already means something within (post/neo-)Marxist scholarship — and that meaning is not ‘trying to impose leftist political goals through vague cultural/intellectual means’. But more crucially, the use of the phrase in the latter sense is actually antithetical to Marxist historical analysis, so it makes it kind of obvious that the speaker is criticizing something without understanding its most basic premises.
“Cultural Marxism” within Marxist circles tends to be associated with Raymond Williams. But I think that right-wing critics mean Gramsci and the Frankfurt School. They are aware that this represented a movement away from Engels & Kautsky -style economic determinism. They generally don’t care who is the real Marxist, any more than you care who whether Hayek or Friedman represented real libertarianism, or Sunnis or Shi’ites are the real Muslims.
Gramsci’s prison writings are obviously deliberately obscure, but I think it is fair to say he thought (a) class rule depended more on cultural institutions than Second International Marxists had thought and (b) the left should focus on developing its organic intellectuals for the purposes of developing cultural institutions which would shift hegemony to the left. There is a class analysis on the right that claims that after World War II, a “new class” that depends for its income on the state and is hostile to private enterprise (except as a source of revenue for the state) succeeded in taking over the traditionally proletarian left and became dominant in cultural (and to a lesser extent scientific) institutions. The premise is that this represents the unfolding of Gramsci’s strategy.
The idea that reality is socially constructed is both flattering to word-spinning intellectuals and is the shibboleth of their movement. It inevitably faces some resistance from proletarian common sense and from the natural sciences. The denial of the existence of race, as opposed to the moral principle that races are equal, strikes a blow against proletarian common sense, much to the delight of the New Class. Natural science would suggest that all categories are fuzzy, which is not the same as saying they aren’t useful and real. Scientific eliminationism (“there are no chairs — only atoms”) confronts common sense, but it doesn’t really support the new class view that all that matters are words.
@Coco :
Would you say the same to those who oppose “Social Darwinism”, or would you rather go after those who bastardised Darwin’s ideas in the first place by introducing them to political discourse ?
As a layman, I have an uneducated question:
How does the idea of “race” differ from the one that is described by “subspecies” ?
Is it not the case that a taxonimist freely gets to decide whether he recognises subspecies as such when he thinks there exist at least 2 groups which have inherited sufficiently distinct genetic traits through seperated breeding over prolonged periods of time while still being able to interbreed and produce fertile offspring ?
Is that not exactly what “race” is about ?
Genetic differentiation between populations correlates with the geographic distance separating them, which is clinal. There are no genetic barriers or discontinuities.
Genetic clusters only appear if you exclude the “intermediate” populations between/or on the edge of continents:
“Suspicion is also warranted by the fact that as geographically intermediate regions are added to the data, the genetic markers used to identify continental clusters become less powerful.” – Glasgow, Joshua. (2009). A Theory of Race. Routledge. p. 104
See also Handley et al. (2007):
“These clines might seem to contrast with work that has described human genetic variation as ‘clustered’” and: >75% of the total variance of pairwise FST can be captured by geographic distance alone.” – Handley, Lori J. Lawson, Manica, Andrea, Goudet, Jérôme, & Balloux, François. 2007. Going the distance: Human population genetics in a clinal world. Trends in Genetics, 23, 432–439
As Frank Livingstone wrote in 1962: “There are no races, there are only clines”, modern genetic data has confirmed this. So called “race realists” have no or little understanding of science and Jonathan Marks is correct when he considers them to be like creationists.
The clinal model is reasonable as far as it goes, but how far does it go? Take a look at a globe and you’ll noticed that gene flow was historically limited by oceans, deserts, and mountain ranges. It’s only 1600 miles from West Africa and Brazil, but how much gene flow was there across it in 1491?
There was gene flow from West Africa to East Africa, from East Africa to the Middle East, from the Middle East to South Asia, from South Asia to East Asia, from East Asia to Siberia, from Siberia to the North American Arctic, and from the North American Arctic to the sub-Arctic Americas, including Brazil.
For all we know, there was enough gene flow such that if there were no regionally-specific selection effects, beneficial mutations could spread through the whole world. If, as you seem to claim, innate abstract cognitive capacity (whatever the genetic substrate of IQ may be) is always beneficial, why wouldn’t it spread through the pre-industrial world? Hunter gatherers needed all the problem-solving ability they could get.
The Flynn effect proves there can be massive increases in raw IQ scores without significant change in population genetics. Why would I assume that differences in standardized test scores between the descendents of slaves and the descendents of European immigrants would be explained by pre-industrial differences in selection pressures, rather than more obvious and recent historical causes?
We have a pretty good idea these days of the amount of gene flow between continental racial groups during prehistory and it was surprisingly limited. Think about what it would take for a favorable mutation to get from West Africa to Brazil or vice-versa. And, would it be favorable in all environments in-between, such as the Bering Strait?
But you don’t need much gene flow if the selection coefficient is high enough. And we know that IQ scores can vary massively over a few generations based on cultural change.
Everyone in this debate is stuck with their priors.
“We have a pretty good idea these days of the amount of gene flow between continental racial groups during prehistory and it was surprisingly limited.” Your source for this? And what do you mean by ‘limited’? How much is ‘limited’?
What is the evidence about pre-Colombian gene flow among human populations?
If you buy Jensen’s idea of “g”, it would seem to have immense selective advantages for any pre-industrial population. Diamond’s point that a hunter-gatherer has to solve more cognitive problems to survive than a farmer does seems sensible.
Regional variations in selection pressures seems more plausible if you have a massively modular view of the mind. “g” would have to be massively polygenic and the selection pressure for improvement would seem to be strong everywhere. On that basis, my intuitive model would be that “g” would go up everywhere over the last 50,000 years.
Endemic disease resistance would be different.
“As I recently told Jerry Coyne, the “race is a social construct” notion is not only a lie, it’s a reprehensible and pernicious lie. And one that needs to be destroyed.”
Coyne follows Dobzhansky’s populationist (re)definition of race as a group of individuals that differ in some of their genes. This means Italians, Londoners, Amish villagers, a rain-forest tribe or a house of 5 incestuous individuals are all races (yes, the latter is horrible but it still passes). Dobzhansky’s idea of “races” just considers them synonymous with any breeding population.
The obvious problem with Dobzhansky’s definition is: what does it have to do with race in the first place? To quote Montagu:
“Why use the word “race” here when what is being done is precisely what should be done, namely, to describe populations in terms of their gene frequency differences? What, in point of fact, has the antiquated, mystical conception of “race” to do with this? The answer is: Nothing. Indeed, the very notion of “race” is antithetical to the study of population genetics, for the former traditionally deals with fixed clearcut differences, and the latter with fluid or fluctuating differences. It seems to me an unrealistic procedure to maintain that this late in the day we can readapt the term “race” to mean something utterly different from what it has always most obfuscatingly and ambiguously meant.”
The term “race” has been used in English for about 500 years. In Shakespeare it has a number of different but interestingly overlapping uses: lineage (“noble race of kings”), breed or herd (especially of horse), contest of speed, and root. Much of English-language use of “race” was originally concerned with breeding racehorses, developing lineages that are more likely to win races.
The more you talk about race, the more it sounds like you agree that it’s a nebulous, nearly impossible-to-define, utterly subjective categorization. In other words, a social construct.
Difference can be continuous without being subjective. The term “social construct” is not really up for the distinctions that need to be made.
And differences can be somewhat continuous and somewhat lumpy.
How strange. All the commenters dismissing the idea of race as a social construct seem to have failed to read the first three paragraphs of this article, which very clearly explain the concept. It is not a denial of biological differences between groups of individuals, but a recognition that the signifiers people use to denote what they call “race” are poorly defined and do not establish coherent groups that correspond to relationships by descent.
Dear P.Z.,
You do know you are wrong, don’t you?
Steve
“Race is incoherent — biological differences exist, just please oh please don’t ask us to give examples.”
“Race is scientific–just please oh please don’t ask us to define it in empirical terms.”
What’s a “cline?”
I’m here to help!
http://bit.ly/1m94PSv
It’s all relative:
http://www.vdare.com/articles/its-all-relative-putting-race-in-its-proper-perspective
Social conceptions of race impose discrete categories on underlying continuous difference. That isn’t at all unusual, and that isn’t why race is problematic. Social conceptions of age do the same thing. Race is problematic because of the history of slavery, colonialism and apartheid.
Interesting observation by Steve Hsu about the definition debate:
“I’m really not interested in arguing about the meaning of a term like “race” — it is used quite differently by cultural anthropologists, biologists, the man in the street, etc. Consequently it has become an impediment to understanding.
But we can still make well-defined statements such as “Allele frequencies tend to be more similar between two Nigerians than between a Nigerian and a Korean” or “The number of base pair differences between two Nigerians is less than between a Nigerian and a Korean”. These are both now known to be true, but that was not the case 30 or 50 years ago. If you want to argue with me about what I mean by Korean or Nigerian then you are really just diverting the conversation, because those terms are about as well defined as “cat” or “car”.
I don’t want to make any strong claims about the *consequences* of these observations until the science is on firmer footing. It’s an important first step to note that some claims about *genetic* level variation in humans (the Montagu/Lewontin-type claims still taught in university classes) turned out to be incorrect.”
http://infoproc.blogspot.co.nz/2014/05/whats-new-since-montagu.html
Thank you, pzmyers! You’ve put that very succinctly and it ought to be the last word on the matter. Unfortunately there are way too many who will insist that just because they can tell (or think they can tell) a “negro” from an east Asian “just by looking at them” that “race” has to be recognized as a scientific concept. Of course, it’s equally “obvious” that the world is flat, isn’t it?
What I was responding to was the following statement by pzmeyers: “the signifiers people use to denote what they call “race” are poorly defined and do not establish coherent groups that correspond to relationships by descent.”
I can’t think of a more succinct and effective way of putting it. It doesn’t matter how often you can spot a “negroid” or “mongoloid” or “caucasoid,” and it doesn’t matter what sort of clusters someone can find (if he looks hard enough) in some STRUCTURE display. The concept of race will remain poorly defined and inconsistent with genetically determined lines of descent.
That really ought to be the end of it. But of course, as with the endless debates over the reality of “intelligent design,” there’s no end in sight.
Wade made it quite clear that by “adapted to capitalism” he meant Jews had an unusually high IQ, which allowed them to succeed disproportionately under capitalism. Wade uses the combination of Cochran-Harpending (natural selection) and Eckstein-Botticini (attrition by out-conversion) to explain that unusual intelligence. There’s no specific confirmation of those theories relating to Ashkenazi Jewish intelligence, but surely no one denies that intelligence is at least partly heritable ?
“and numerous critiques of this hypothesis [of Gregory Clark, as borrowed by Wade]”
Those two critiques are by the same person — Deirdre McCloskey — and she doesn’t even seem to understand what “regression to the mean” means. And Clark himself addressed those arguments in the European Review of Economic History from which I excerpt the “genetically relevant” part here
Every field of classification runs into the splitting v. lumping tradeoff. How many languages are there? Religions? Mountain ranges? Species?
http://en.wikipedia.org/wiki/Lumpers_and_splitters
Not that’s not true. The classification of chemical elements for example by their atomic number is objective and there is no “splitter” or “lumper” debate in chemistry [elements are natural kinds]. Each element has a unique atomic number.
Races unlike chemical elements in contrast are not real – they are mentally constructed and arbitrary. If you look at the literature you will see that species are constantly debated as whether they qualify as natural kinds. Philosophers and scientists seem to be split 50/50 in regards to species, that is why there is a “species problem”. There is however no “race problem”. The overwhelming consensus is that races don’t exist in a natural sense. Of course, this doesn’t mean they aren’t artificial kinds (e.g. like “tall people”).
Plato advised “carving nature at the joints” and there are obvious joints (i.e., partial disjunctions). The Atlantic Ocean in 1491 was a sizable impediment to gene flow. Same for some other oceans, some deserts (e.g., the Sahara), and some mountain ranges (e.g., the Himalayas).
I wasn’t aware that consensus was the appropriate criteria by which we determine the ontological status of abstract concepts.
The remarkable thing about this review and just about every other negative review of Wade is how little attention is paid to the deepest message of his book, namely that the genomic studies now show that human evolution has been “recent, copious, and regional”, and that among the regions of the genome which have been subjected to evolution in recent times are areas connected with brain function.
If that point is true, what does it matter if we talk of races, or if we declare there are 5 races or 20, or if we insist in describing the entire human species as be captured by clines?
The point is, subgroups (or different regions of clinal variation) have been subject to distinct evolutionary paths, and the genome has changed in a way specific to that group or clinal region.
What that means is that it’s perfectly reasonable to expect that on cognitive and emotional traits each group or clinal region may present different distributions — for all we know, very different distributions.
And what that may readily mean is that, say, the average IQ of certain groups from certain areas may be much higher than the average IQ of other groups from other areas.
Does taking away the language of race make it any easier to swallow that possibility?
“recent, copious, and regional” is a very facile phrase to throw around, but I don’t think you actually understand population genetics. You seem to be missing the point that genetic variation is measurable and quantifiable. What Wade does with his chapter purporting that race is genetically based is MISREPORT the scientific literature characterizing human genetic variation.
The reason that you can’t settle on the number of races is that you cannot define race empirically. So you have no justification to claim the mantle of science for your position. “I know it when I see it” doesn’t cut it here.
I have no idea what you are even arguing.
The point I made couldn’t be simpler. Even if we refuse, for whatever reason, to acknowledge the existence of races, what cannot be denied is genetic variability across subpopulations in different geographical regions. The science shows that human evolution is “recent, copious, and regional”, so that populations in different regions have been subjected to different selection pressures, changing the genomes of those subpopulations in distinct ways. Some of the regions of the genome thus differentially altered relate to brain functioning.
Thus it is entirely consistent with this picture that, say, different subpopulations in different continents will differ significantly on IQ based on genes. There is certainly no reason to believe that these geographically separated subpopulations sufficient gene flow between them, or were so little different in their selection pressures, to prevent that from happening. So it is entirely possible that, say, based on genes, subpopulations in subSaharan Africa have average IQs a standard deviation (or more) below the average of subpopulations in Europe.
Again, this is true whether we want to talk about race or not.
My overarching point is that the sorts of conclusions you most want to avoid, namely, such things as that the average IQ of subpopulations in different continents may differ substantially, aren’t in any way obviated by niggling over the applicability of “race” as a concept.
You’re arguing against a strawman with this. Who’s denying genetic variability in different geographic regions? Not me.
“Human biological variation is real and important. I’ve studied it my entire professional career. We can see this variation most easily in physical traits and allele frequency differences between populations at extreme ends of a geographic continuum. Nobody is denying that. Let me repeat this: no one is denying that humans vary physically and genetically. All anthropologists and geneticists recognize that human differences exist. But Wade, and others who agree with him, have decided that certain patterns of variation—those which happen to support their predefined notions of what “races” must be—are more important than others.”
Selection is NOT the only evolutionary force. Can you think of other evolutionary forces which may have contributed to regional genetic differences? And it’s hardly ‘niggling’ to demonstrate that the ‘factual basis’ of Wade’s claims for the genetic distinctiveness of ‘races’ is anything but.
Then, since you are so much accepting of the genetic differences between human groups, that you are fully on board with the idea that groups might readily differ on cognitive and emotional traits as well? (It’s just awfully curious that you mention physical traits, but not cognitive or emotional ones.)
Once one grants that evolution is “recent, copious, and regional”, on what possible scientific ground can one argue that cognitive and emotional traits should be excepted?
The problem is, it is very well established that many cognitive and emotional traits have a genetic base, as shown by twin studies, adoption studies, sibling studies, etc. If identical twins show that they are much more alike on a trait such as IQ than fraternal twins, it’s a little hard to believe that genes don’t play an important role, you know?
So why should such a trait be excepted from natural selection if all manner of physical traits clearly evolve?
I’m shocked that she broke off contact here. Just shocked.
Don’t flatter yourself–it’s a holiday weekend. I’ll get to it when I get to it.
Myself? I had nothing to do with it. Where would I begin?
Anyway, you’ve ducked the hard questions all over this thread, but poke your head in the easy ones enough to show that it’s deliberate.
The problem for you is pretty obvious: when it comes to genes here, you’re clinging to the rapidly disappearing chunks of ignorance we still have about the genome. When it comes to race, you’re not doing much but playing word games, as your later comment about your work shows.
Easy questions don’t take as much time to answer as the more complex ones, and I want to do my best to provide thoughtful responses to some of the interesting points that people raise here. But that takes time, and I prioritize my work and personal life over blogging.
Don’t worry – unless she’s a Cartesian dualist she already conceded the argument.
“We can see this variation most easily in physical traits and allele frequency differences between populations at extreme ends of a geographic continuum. Nobody is denying that.”
The brain is a physical organ.
Which neither implies nor requires that all behavioral variation is genetic in nature.
The difficulty is that cognitive and emotional traits obviously have a cultural history. The Flynn effect shows that raw IQ scores go up by a standard deviation a generation! That obviously can’t be accounted for by any change in allele frequency. Any middle aged person can compare his or her children’s generation to his or her grandparents in terms of willingness to express emotion, tolerance of sexual expression and diversity, attitudes towards gender roles and so on. By “recent”, I take it Wade means the last 10,000 years or so, but everything about your everyday life would make no sense to a pre-industrial person.
Different population groups (“races” if you insist) also have different cultures and different histories. It is difficult to disprove the possibility that some of the current statistical differences between them are due to genetic variation, but a twin study or an adoption study is just not a solid basis to resurrect the race science of a century ago.
Dont’ races (however you want to define them) have huge internal variations when it comes to cognitive and behavioral attributes, like IQ, aggression, creativity, etc? All of these are statistical aggregations, and you describe them to be so. Isn’t a statistical measure fundamentally different from a “trait”, like lactose intolerance, light skin, epicanthic folds, or adaptation to high altitude? Trait is something that every individual of that “race” possesses, whereas high IQ (for example) is something that a small percent of that “race” possesses. When comparing such statistical aggregates across races, how can you discount all non-genetic variables (culture, environment, history, etc.)? Am I missing something?
“The reason that you can’t settle on the number of races …”
How many extended families do you belong to?
“The point is, subgroups (or different regions of clinal variation) have been subject to distinct evolutionary paths, and the genome has changed in a way specific to that group or clinal region.”
Great. If that’s the whole point, just tell us which mutations in which populations have had what effect? Just one trait specifically associated with intelligence that has undergone a selective sweep would be enough.
Since we’re not sure how to measure intelligence, what the neural correlates of intelligence are, what genes correlate with neural development, what variations in those genes cause differences in neural development, or what populations have those specific genetic variations, I’d say you’ve got your work cut out for you. But if these differences are as obvious as you say they are, you must have figured all this out already. Otherwise, you’d just be dressing up centuries old bigotry with new pseudoscience.
Your argument sounds just like creationist arguments against evolution. “Oh, you say that the eye ‘somehow evolved’ by natural selection. But you can’t tell me how exactly it evolved that way, can you? So I have no reason to believe in this crazy theory of yours until you do!”
Creationists use a God-of-the-gaps argument. You use a Equality-of-the-gaps argument: until I can specify every detail in the mechanism whereby, say, IQ is determined genetically, you’re not going to believe it.
Look, no one would impose these sorts of constraints on other features. We don’t know in any real detail which genes make Pygmies short or which genes make Tutsi relatively tall. But no one seriously disputes that that difference hangs on genetic differences.
So it should be with such things as IQ — if we think like scientists instead of ideological hacks.
And the point of Wade’s book is, explicitly, not to argue that any specific cognitive or emotional trait has already been established by science to differ between population groups based on genes, but merely to open us to the very strong possibility that such traits and configurations of genes exist.
“Creationists use a God-of-the-gaps argument. You use a Equality-of-the-gaps argument: until I can specify every detail in the mechanism whereby, say, IQ is determined genetically, you’re not going to believe it.”
Nope. You have to show the differences is genetic rather than cultural. And show that IQ is actually measuring something phenotypic. And show that that phenotype actually varies between populations. Show that the thing you’re claiming is actually true instead of just assuming that it is.
Otherwise, we’re free to just assume the reason Western children enjoy Harry Potter more than Asian children is genetic and adaptive.
“merely to open us to the very strong possibility that such traits and configurations of genes exist.”
People have been all too open for all too long. We’ve got the 1,000 Genomes Project and the HapMap project. If there were any significant signal here, we’d be swimming in conclusive data by now. If you want to convince the me, stop making excuses and prove your case.
It also might help me take you seriously if you somehow differentiated your claims from centuries of pseudoscientific speculation on this topic.
Do you have anything supporting your point of view besides abuse?
Look, in a post below I link to a study demonstrating the link between genetic variation and IQ based on genomics. This simply confirms what twin, adoption, sibling and other studies demonstrated regarding the high heritability of IQ in particular. Only to the most credulous believers in the dominantly environmental hypothesis would this result come as a surprise.
Here, again, is the link:
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3182557/
“And show that IQ is actually measuring something phenotypic. And show that that phenotype actually varies between populations.”
Do you even know what you are talking about here? Look, if measured IQs differ between, say, those of subSaharan descent and those of European descent, then, by definition, it is a phenotypic difference. The only question is the source of that phenotypic difference: genes, environment, or some combination of both (obviously it’s some combination, but in what proportion?)
No one is arguing against there being a genetic component to intelligence. That is your strawman. What you need to show is that variations in intelligence have any connection whatsoever to your definition of race. Not IQ as measured by culturally specific and provably biased tests, but intelligence. By the way, define race. You’re making logical leaps over huge chasms, then getting offended when people point out your or Wade’s total lack of connections.
“Provably biased” ?!? Damn, i guess I had to write to all psychometricians that all their papers written on how IQ test are not biased are wrong.
Since we already have enough semantic confusion about “race” I woildnt want to add to it on “bias.” Thebfundamentalmpoint is we know raw IQ scores go up a standard deviation a generation. So they can’t possibly be a direct measure of underlying genetic differences. (No evolutionary change could possibly take place at that speed.)
An objective definition of race is (everyone with trait X) vs (everyone without trait X).
For instance, I don’t like walnuts. They make my tongue itch. Given that I am a superior human being (no proof needed!), and that I am Walnut-Intolerant, we can therefore conclude that I am a superior human being *because* I am Walnut-Intolerant.
[And one day, we shall rise up and overthrow you filthy Walnut-Eaters…]
There is a vast body of inferences in the biological and evolutionary-psychological literature that evolution by natural selection has occurred, most of which are not accompanied by evidences from the genome (especially for the higher animals). Am I wrong ? Those inferences have most often been assessed by data from phenotype using classical quantitative genetic techniques. Am I wrong ? The idea is to further test these “classical” inferences with evidence from the genome, yes ? So why hold genetic arguments about group differences in the social world to the higher standard of genomic proof ? Those all can be ultimately corroborated or falsified by investigations into the genotype. But in the mean time there’s enough social science evidence to suggest that genetic differences must be at least partly responsible for some of the group differences in social outcomes.
“But in the mean time there’s enough social science evidence to suggest that genetic differences must be at least partly responsible for some of the group differences in social outcomes.”
Show your work.
“What can be asserted without evidence can be dismissed without evidence.” -Hitchens
http://www.tylervigen.com
I almost forgot to include this helpful website you can use to find more correlations to blindly assume we should take seriously. Are you going to preach we should start investigating why Nicolas Cage keeps drowning people, or does that not confirm your prejudices sufficiently?
Behavioural genetics uses twin & adoption studies.
As I understand it — and I’m willing to be corrected — the actual number of twin/adoption studies is very very low (like, <100), because the time interval between "hey, these (twins who were deliberately separated and then adopted and raised by different families) provide an interesting test for environment-vs-genetics" and "hey, we should stop deliberately separating orphaned twins, it's cruel" was quite small.
Coke and Pepsi use blind taste tests.
I can use non sequiturs, too.
To Danny in Canada : No, it’s in the hundreds. The limitations of these studies are several, foremost in my opinion that they are almost all confined to middle-class households in developed countries. But there’s a virtual consensus amongst psychologists that heritability is much higher in developed countries than in developing ones, so that particular limitation is perhaps not that troublesome. We know developing countries have a lot of room for improving their environments. The question then becomes how room is there in developed countries. The two issues act as constraints on each other, by the way.
To Danny in Canada : plus the twin & adoption studies results are now supported by at least 1 GWAS study ( http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3182557/ ) and a totally separate line of evidence ( http://press.princeton.edu/titles/10181.html )
The high heritability of IQ in particular is now beyond serious dispute, based on genomic studies (which simply confirm what twin, adoption, and sibling studies have already shown).
The breakthrough study on this is by Visscher et alia (plenty of et alias!)
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3182557/
It is frankly a disgrace that this result and its implications are not better known.
Come on!
http://pss.sagepub.com/content/23/11/1314.short
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3592970/
GWAS are not even close to conclusive. They fail to replicate, and they find contradictory results. At best, they’re a starting point for further investigation. This isn’t settled science. Pretending it is tells me more about you than it does about the state of the research.
If the difference is real, we’ll find out soon. But it’s going to take more than GWAS and a handful of twin studies. It’ll take replication, a body of literature, and deeper investigation.
And that still won’t tell us anything conclusive about the relative effects of culture and genetics. We’ve still got a LOT to learn before we start making sweeping generalizations.
Of course it’s a starting point. But do you believe intelligence is not inherited at all based on the totality of the current evidence ? Complete agnosticism is the correct stance ?
No. I think being highly skeptical of these claims is more warranted, because people have a long history of making claims like this and being wrong. And people have a long history of doing terrible things based on flimsy claims like this. For all meaningful decisions any of us are going to make in our lives, a genetic component of intelligence might as well be nonexistent. At best, this is a very, very weak signal.
Intelligence is a very complex, multifaceted phenomenon. It’s not captured well by a single measurement like IQ. If the phenomena we call “intelligence” are inherited, they’re probably not associated with a single gene or even a handful of genes. Factors influencing such a complex phenomenon are probably spread throughout the genome, and are probably highly subject to environmental conditions.
We’re going to have to understand genetics and intelligence much better before we can tease out a genetic component that has anything like a practical relevance (if one even exists.) For now, though, it’s not really worth considering. Yes, the moon has a measurable gravitational influence on me. Does it factor in to my decision making? Not really.
If there is something resembling a refutation of the particular GWAS study I offered up, I haven’t seen it, and you haven’t presented it.
That’s what makes it a breakthrough study: the techniques and data are far more powerful than in previous studies. Even a famous advocate of the importance of the environment, Eric Turkheimer, concedes that it establishes what twin studies had independently supported:
“Thanks to the Visscher program of research, it should now be impossible to argue that the whole body of quantitative genetic research showing the universal importance of genes for human development was somehow based on a sanguine view of the equal environments assumption in twin studies, putting an end to an entire misguided school of thought among traditional opponents of classical quantitative (and by association behavioral) genetics (e.g.,
Joseph, 2010; Kamin & Goldberger, 2002).”
Click to access StillMissingFinal.pdf
Turkheimer complains that the study doesn’t give a picture of the etiology whereby genes may figure into bringing about various traits> But that is another matter, and mostly irrelevant to the issue at hand, namely, whether genetics actually plays a critical role in IQ, however it may do so, and is therefore something that natural selection can work on.
I should note that the breakthrough technique in the study in question is actually called GCTA (genome-wide complex trait analysis), to differentiate it from previous GWAS techniques.
While there’s still some controversy about the statistical techniques involved, they have stood their ground well so far.
It doesn’t matter how many races there are for race to be real. This Cultural Marxist obsession to “prove” that “race is a social construct” reminds of me Creationists using whatever tools they can find to prove evolution is false.
And even if race is clinal, it doesn’t matter. Color is clinal and exists. Race deniers are the new Creationists.
As another commenter notes above, the funny thing is that we know evolution is recent, copious and regional, so we know we’re soon going to discover genetic cognitive and behavioral differences between groups whether we use ‘race’ or ‘regional clinal population groups’, etc.
The truth will eventually prevail and such a day will be a sad day for all the Creationists in this comment thread.
It’s funny (or maybe just sad) that DJ is calling the scientists in this argument “creationists”. Creationism has much more in common with Wade’s assertion than with the scientific view that race is an arbitrary concept. Wade does not define race, just says “it is there, see? I can see ‘it’, so disregard genetics and geneticists.”. Race is not skin color or epicanthic folds, folks, that’s melanin and fat, as I said above. Creationists do the same as Wade- they assume that a God exists, then argue without proof that God created the universe. When scientists ask for proof, we’re not trying to stifle the discussion, we are asking creationists and the people here who agree with Wade to *show your work*. You are making an assertion, with no clear definition of what is race, no statistical proof that Ashkenazi are more intelligent, no clear connection between intelligence and capitalism, no clear genetic or epigenetic markers, nothing but “but I can see it by eye” hand-waving wild claims that you demand we accept. Sorry, no. Scientists have to prove our hypotheses by showing direct connections between A and B. Show your work.
By the way, the reason why MW and candid_observer don’t understand some of Dr. Raff’s arguments is because she is using precise terminology in the fields of genetics and anthropology. (I had to look them up.) Given that we’re arguing genetics and anthropology, that should give MW and c_o a clue about how far over their heads they are… sadly it does not.
And pseudoerasmus, this is Dr. Raff’s blog. You are her guest. What kind of bloated ego does it take to tell your host that she should not take part in a discussion on her own blog?
“And pseudoerasmus, this is Dr. Raff’s blog. You are her guest. What kind of bloated ego does it take to tell your host that she should not take part in a discussion on her own blog?
Where did I do say that ????
pseudoerasmus stated to Dr. Raff: “In that post, I was replying to JerryA, not to you.”
Granted, you then went on to reply to her comment, but your first sentence was rather dismissive. In commenting here, we are all discussing her blog post.
Thanks for the vote of confidence, JerryA. I’m asking for specific details because some of them I believe have been lacking (such as the ambiguous individuals who would be within but indeterminate between the category “Korean” and “Aboriginal”). Over the past week or so, I’ve see many people say that they _never_ denied biological differences between groups, but I have yet to see anyone actually name a difference (probably afraid of getting yelled at).
As for “cline,” I know what it is, but I’d like to see the Doctor’s own working definition to see if it’s as precise, objective, mathematical, and totally non-ambiguous as people seem to think all classifying terms need to be.
Bingo!
A cline is a fine concept, even though it involves much hand-waving. So is culture, environment, nurture, education, and so forth. Just because they are hazy in definition doesn’t mean they aren’t real and important.
If you want to test something scientifically you need to know what it is you’re testing. So, Mr. Sailer, please tell us what it is that is being tested when testing for “race”? And how does one go about doing that?
A racial group is an extended family whose ancestry is partly inbred.
But that’s all of us. If you trace humans back far enough in time, we’re all part of a single “extended family whose ancestry is partly inbred.” So how far back do you go? Just far enough to create the racial groups you want to find, like Wade with his misuse of Structure?
docgee: You should ask Neil Risch, he’s really excellent at it.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1196372/
Wow. 99.86% of the time he could match a genetic cluster to the person’s self identified race.
A racial group is a partly inbred extended family.
There’s nothing more non-fuzzy than biological family trees. And there’s nothing more inevitable than some degree of inbreeding the further back we go because the number of slots on our family tree goes up by three powers of ten for every 250 years or so we go into the past. So, individuals tend to be more closely genealogically related to some people than to other people.
In practice, closeness of genealogical relationship turns out to be very closely related to genomic closeness which turns out to be very closely related to what an worldly person can see by eyeballing individuals.
How far up the ladder does the inbreeding have to be while still serving to define a “race”? And are you testing for it, or assuming that it exists?
It is, I suppose, an assumption that there’s inbreeding 40 generations back on anyone’s tree, as it can’t be completely, absolutely ruled out that everyone had a trillion simultaneouslyiving ancestors in a single generation 1000 years ago
Probably not the way things really were, though.
It’s more than an assumption. You don’t have to go very many generations back before you get billions of unique ancestors for every current human, and that number would get bigger as you looked further back in time. Obviously that’s not possible. We all share ancestors, which means–depending on your boundary criteria–we’re all inbreeding all the time.
The question of where we draw the line might seem trivial to you, but I don’t think you can answer it in an objective or consistent way. I know Sailer can’t; it’s the obvious question about his idiosyncratic definition of “race,” but he changes the subject every time it comes up. He doesn’t have a solid criteria to use. Do you?
How many grains of sand are required for there to be a “heap”?
It’s a subjective answer, just like race.
If you want to know who the true creationists are, see this image:
Get with the times, docgee. We’re all clinialists these days. Now, since it’s already been stipulated that groups differ in terms of biology and genetics, we’ll get down to determining, say, the differences between a cline picked totally at random from East Asia and a cline picked totally at random from the Outback in order to determine psycho-physical characteristics, and continue to do so until things take whatever rough shape they might.
Thanks so much, Jennifer. This is exactly the sort of review I’ve been hoping for, and only rarely finding. Far too many reviews of this book, even the negative ones, too readily accept Wade’s simplistic and misleading defense of “race” as scientifically established by all that “new” genetic research he so loves to cite.
Imo his misinterpretation of this research goes beyond even the very real problems you’ve raised. Looking specifically at the STRUCTURE graph from Rosenberg et al., we find, under k=5, a clustering of the world into what Wade sees as five distinct “races.” As you correctly point out, this division is NOT produced automatically from the computer, as there is nothing special about that particular k value. But even if we give Wade the benefit of the doubt on that, there are serious problems with regarding this as a racial clustering.
For one thing, the African Bushmen have a highly distinctive morphology and genome, and were generally regarded as a separate race by an older generation of anthropologists – and yet they are grouped with all the rest of Africa in the k = 5 clustering. Secondly, it’s important to remember that the most commonly accepted “racial” breakdown among anthropologists and laymen alike was threefold: caucasoid, negroid and mongoloid. Since Native Americans were typically grouped in the third category, their inclusion here as a distinct population cannot be seen as evidence supporting the traditional view of race, but on the contrary, evidence that genetic clusterings of this sort have nothing to do with race, being based on migration patterns over vast geographical areas and long periods of time.
Finally, the category “Oceania” is extremely misleading, as only “Melanesians” and “New Guineans” are represented. NB: Australian aborigines are not represented in this particular study. By anyone’s standard, it is ludicrous to claim, on the basis of this or any breakdown, that Oceanians represent a distinct race, or even a distinctly morpho-genetic population.
Wade is clearly seeing what he wants to see. Rosenberg et al. make no such claim, as you note, and for very good reasons. Again, thanks so much for an extremely lucid and sensible review.
Great image that people are tweeting on twitter. It reads:
“The Cultural Marxist War against Darwinism
Creationists: evolution is a social construct, not biologically real.
Liberal Creationists: race is a social construct, not biologically real.”
Image at: http://oi62.tinypic.com/21ovkpj.jpg
This perfectly sums up the idiocy of the “social construct” positions — both of religious and liberal creationists.
That… makes no sense whatsoever. “Cultural Marxist War”? “Darwinism”? Only morons call geneticists “Darwinists”. It’s a science, not a belief system or opinion. “Liberal Creationist” is made up gibberish, as is the first line. It’s all mixed up garbage. Creationism is religion, not science. Similarly, what used to be called “race” by our culture has been proved artificial, a way to justify bigotry and oppression. That may be why the genetic research showing it is not inherent in our genes is having such backlash. Not scientific refutation using facts, just loud arguments and bluster (like “100% detection” using the unaided eyeball). When you and others here were asked to define your terms or show your work, you can’t do it. Yes, you have the right to say whatever you want, but I am also free to mock it. This tweet is twaddle.
Fwd:
Updated Image people are Tweeting:
It reads:
“The Cultural Marxist War against Darwinism
Creationists: evolution is a social construct, not biologically real.
Liberal Creationists: race is a social construct, not biologically real.
Charles Darwin: I’m not a creationist; I’ll use the word ‘race’ in title of my Origin of Species”
Image at: http://oi61.tinypic.com/9kyn2w.jpg
This pretty well sums up how silly the “social construct” position is – whether by religious creationists or liberal creationists.
David. You’ve published your image several times to this thread, and we’ve all had a good laugh at it. (Are you honestly citing a nonsensical meme as ‘scientific’ evidence for your argument? You’ll find similar tactics used by the anti-vaccine crowd over in the other threads on this site). Anyway, you’ve posted it enough times here, and more instances will be considered spamming.
“You’re arguing against a strawman with this. Who’s denying genetic variability in different geographic regions? Not me.”
Jennifer,
Are you willing to acknowledge Wade’s contention that human evolution has been “recent, copious, and regional”?
You might be interested to know that a 1 standard deviation height difference has arisen between Northern and Southern Europeans (“regional”) over the last 10,000 years or less (“recent”). The evidence is overwhelming that this difference is partially genetic in cause, and due to selection (see Nature article below). Add to this examples like EDAR, lactose tolerance, altitude adaptation, disease resistance, etc. and you easily get to “copious”.
http://www.nature.com/ng/journal/v44/n9/full/ng.2368.html
I would like to hear your argument against why selection pressures of this sort could not lead to differences in average temperaments or cognitive abilities.
If you don’t have such an argument, then you are acknowledging that different geographical groups might differ in the ways that Wade suggests. Whatever faults you want to ascribe to Wade, you should admit that he has the courage to discuss these possibilities in the face of withering (and often unfounded) criticism.
In the face of this more meaningful analysis, why get hung up on the definition of “race” or the number of “races”?
Galton,
“Are you willing to acknowledge Wade’s contention that human evolution has been “recent, copious, and regional”?” I’d go further than “recent”. I’d say that evolution is ongoing. As for “copious”, how precisely are you quantifying that? Remember that we geneticists and evolutionary biologists actually measure changes in allele frequencies in order to model evolution. As for regional, please be more specific as well. Which regions? What are their boundaries? That phrase is very easy to repeat, as Wade does quite often, but it’s incredibly vague. I’m interested in having a conversation about population genetics and human evolution, but it’s going to require you actually grappling with the data, as I did in my critique of Wade’s interpretation of the Structure results. It will also require you to have an understanding of evolutionary forces beyond just selection. Drift and gene flow, for example, are also major contributors to current patterns of human variation, but you guys don’t seem to acknowledge their effects.
You bring up height differences between Northern and Southern European populations. This is a subject that has been studied quite intensively, and it’s not as straightforward as you seem to think it is. Here is an open-access preprint by Jeremy Berg and Graham Coop that you might find illuminating: http://arxiv.org/pdf/1307.7759v2.pdf
In particular, I’d like to direct your attention to pages 19-21.
@Jennifer: Thanks for your reply. Berg and Coop discuss the Turchin et al. paper I cited earlier and their analysis *supports* its conclusions. The possible confounds discussed on pages 19-21 can *reduce* the statistical power to detect selection, but are not likely to cause drift or founder effects to appear as selection. Basically, it is not easy to cause a coordinated allele frequency difference (over hundreds of alleles!) in two groups. By the way, you might be interested to know that there are very strong correlations between effect sizes for height (and other GWAS) alleles in different population groups, supporting the interpretation of largely additive genetic architectures. (See, e.g., http://www.plosgenetics.org/article/info%3Adoi%2F10.1371%2Fjournal.pgen.1003566 or http://www.cell.com/AJHG/abstract/S0002-9297(13)00328-5 )
Back to the topic at hand: do you or do you not accept the findings of Turchin et al. (and Berg and Coop!) that the N/S Euro height divergence is likely due to selection? If so, then you accept that human evolution has been *regional* (N vs S Europe) and *recent* (post African exodus). We can leave “copious” off the table for now.
This updating of priors, thanks to the heroic work of the GIANT collaboration and modern genotyping technology, does not depend at all on whether we want to define N and S Euros as different “races” or subspecies or clines or whatever. The point is that geographical groups can differ due to local selection effects, including on complex traits (i.e., due to soft sweeps or small allele frequency differences).
Do you have an argument against similar divergent selection pressures on cognitive or behavioral traits? Or is it just a prior that perhaps deserves reconsideration?
Sometime in the near future a similar analysis of SNPs correlated to cognitive traits will be possible, and we may find that allele frequencies differ systematically between groups. I would hope that you, a population geneticist, are excited and open-minded about new insights into recent human evolution.
http://www.ibc7.org/article/journal_v.php?sid=312
http://www.ibc7.org/article/journal_v.php?sid=317
@Galton. Our and Turchin et al.’s research does indeed show that loci involved in height, as mapped to date, show greater systematic differences between northern and southern European populations than would be expected under a model of genetic drift alone (Raff was not arguing this point). First, note that this signal is really quite subtle, and some of this change is very likely due to drift and gene flow in addition to selection. Also all of these variants segregate in all of these populations, so they say little about an individual; i.e. most the variance is still between individuals within populations rather than among populations.
It is plausible that the observed differences in height between populations are partially genetic in nature and have been subject to selection. However, we do not know this for sure, that’s because of the large environmental contribution to between population differences in phenotypes like height (like the majority of human phenotypes). Also, as outlined in that section of the paper only a small fraction of the variance in height within Europeans populations has been mapped, and there is a potential for GxE and GxE. It is possible that as yet unmapped loci could completely change our understanding of the genetics of height in Europe. Equally, we do not know how systematic environmental differences could skew our view. While for height many loci do seem to act in an additive manner, we do not know that this plays out the same for those loci driving our selection signal, because our signal comes in aggregate across many loci. All of these are fascinating questions, and worthy of study, but also mean that interpretation of between population differences is very tricky.
There’s still a huge amount we do not understand about how selection and drift have shaped height, or any phenotype, within Europe. We also do not know that the differences reflect direct selection on height, the agent of selection, or whether our observations are due to selection on some other phenotype (that may not even be on our radar in current environmental conditions, nor may this correlated phenotype differ among populations today). We do not know the timing of this selection, we do not know whether it represents a long term trend or if this is just a snap shot of some long term fluctuation, or even if this selection took place in Europe or is the result of differential gene flow from populations who themselves diverged in height.
As a human geneticist and a population geneticist, I find it incredibly humbling that we are only just beginning to understand the long-term role of selection and drift in shaping a phenotype as well studied as height. Undoubtedly the we will learn a lot over the coming decade about how drift, selection, and migration have shaped the genetic basis of phenotypes across populations, but this will only come about by the careful study of these phenotypes and their genetic and environmental components. Simply claiming that differences in phenotypes between populations could be genetic and could be due to selection (especially when they conform to our personal biases), rather than working to understanding the complex interplay of environmental and genetic differences, falls into one of the basic traps that all evolutionary biologists strive to avoid.
Graham Coop
Thank you for your reply, Dr. Coop. Galton, does that clarify things for you?
@Graham: Thanks for your comment. I don’t disagree with the caveats you list above. This is a complex subject and in science no hypothesis is ever *proven* to be correct.
However, I am curious as to what your actual confidence level is in the hypothesis: selection is at least *partially* responsible for the N-S height gradient in Europe.
Q1: At what confidence level do you believe this hypothesis? (e.g., at 80 percent confidence level?)
Q2: Do you acknowledge that it is very plausible that our confidence level in the hypothesis will rise significantly in the net decade? (Hint: rumor has it that GCTA on whole genomes accounts for nearly all of the additive heritability (~ 0.8) for height suggested by twin studies…)
Unless you assert that you are highly confident that the hypothesis is *wrong* (which would seem to contradict the tone of your paper), then you are comfortable with the *possibility* that local selection effects can be strong enough to produce *part* of a 1 population SD difference between geographical groups, over tens of thousands of years (post African exodus).
I’ll emphasize again that such group differences do not depend on any specific definition of “race” or subspecies or cline. Variation exists, and can be investigated, regardless of specific terminology. That such variation exists in human groups is Wade’s main assertion, not the specific definition of “race”.
@Jennifer: your readers deserve to know your answers to Q1 and Q2 as well!
@Galton as I have explained in my comment above it is quite plausible that part of the difference in height is genetic in nature, and again as I said above our understanding of height differences will improve dramatically over the next decade (we are actively working on this area). However, what is not helpful is simply claiming that differences between populations in partially heritable phenotypes are genetic and are adaptive, nor is it meaningful to point to a genetic basis of differences in one phenotype and extrapolate to other phenotypes. Even in simpler systems this line of argument has repeatedly proved to be a weak assumption. I have no interest in engaging in debate about this type of speculation, and it is not productive.
Graham Coop
@Graham: Thanks again for your reply. I certainly agree that many population differences will turn out *not* to be due to genetic causes, let alone local adaptation. Wade speculates excessively in the second part of his book about such things.
But our host Jennifer is attacking the *first* part of Wade’s book (which mainly summarizes known results on population structure, certain alleles of large effect whose frequencies vary by population, etc.), claiming a knockout blow because, for example, the definition of “race” is fuzzy, or the amount of variation is “small” in some sense.
Our discussion shows that her points are *irrelevant* to the main question: could genetic variation among humans be partly responsible for group differences (e.g., different averages) on complex traits?
My contention is that among those who have a good understanding of genomics and population genetics, the answer is Yes.
Cosmic rays can theoretically flip bits in computer mamory, although the odds are infinit3simal.
Could cosmic rays be partly responsible for the typos in the preceding sentence? Among those who have a good understanding of physics and computer hardware, the answer is yes.
But those experts wouldn’t stop there, and would be the first to tell you that answer is far too simplistic.
And of course, everyone else recognizes it as an excuse for my having made typos. In other words, it’s not brave wisdom: it’s a simplistic excuse for reaching a convenient conclusion.
***In fact if you use the common level of genetic differentiation between populations used by zoologists to classify biological races (which they called subspecies) in other mammals, all humans consistently show up as just one biological race.”***
Not true. Many other sub-species have lower or similar levels of differentiation. Woodley (2010) notes:
“Chimpanzees for example exhibit very similar degrees of observed heterozygosity to humans (0.63–0.73 vs. 0.588–0.807) yet have been divided into four subspecies.
Some species such as the grey wolf actually exhibit lower levels of observed heterozygosity than humans (0.528 vs. 0.588– 0.807) yet have been divided into as many as 37 subspecies.”
Also, the claim that you need a Fst Fst >0.25 is based on a misrepresentation of what Sewell Wright wrote in 1978. For example Graves (2010) says Wright suggested was equated with geographical races. This is INCORRECT. Wright, in fact wrote:
“It is, however, customary to use the term race rather than subspecies for the major subdivisions of the human species as well as for minor ones. The occurrence of a few conspicuous differences, probably due to selection for adaptation to widely different environmental conditions, does not necessarily imply much difference in general. Nei and Roychoudhury (1974) have shown that the differences among negroids, caucasoids, and mongoloids in the protein and blood group loci are slight compared with those between individuals within any one of them. There is disagreement on the number of major races that should be recognized. At a minimum, the Australoids are added to the three referred to above.
….
Diversification is much greater on the average among than within the major races (F (DS) = 0.0715, F (ST) = 0.1248) but there is more uniformity within ).0.042 to 0.141) than among (0.026 to 0.402). The great differences among loci indicates that more than pure sampling drift is involved.”
(Wright, 1978. Evolution and the Genetics of Populations, Vol. 4: Variability Within and Among Natural Populations. Univ. Chicago Press, Chicago, Illinois. p 439-440)”
Thanks. It’s worth comparing the degree of genetic difference among racial groups to genetic difference among close relatives. The latter is surprisingly high. For example, genetic anthropologist Henry Harpending pointed out that he would have a hard time picking an unknown biological grandchild of his out from a crowd of white children, but he would have less trouble picking a white child out from a crowd of black children.
What if the “white” child’s parents were “black”?
The flood of comments here have served only to increase my admiration of Jennifer Raff’s calm and very well-informed opinions.
I checked on one commenter in particular. Steve Sailer’s blog says “I’m a reporter, movie critic for The American Conservative”, Why am I not surprised? It seems that his grasp of science and mathematics is such that he has difficulty in grasping the subtlety of the problem. But his political views are very obvious indeed.
Care to elaborate on the errors of his position? Steve Sailer tackles race fairly regularly on his blog, including the genetic variations of population. He’s extremely well read on the subject, and I enjoy his dispassionate but opinionated takes on this subject and many others. If you believe readers of this blog will find Sailer’s credentials and employer disqualifying, you’re living in a bubble.
He may be well read, but Jennifer actually does it. It’s always possible to pick papers that confirm your own views, in almost any area. That’s not science, it’s propaganda. Do you really believe that Sailer (or most of the other commenters here) understands the mathematics of population genetics? Do you? I very much doubt it.
It isn’t a question that can be settled by armwaving, or by claiming that you can recognise faces. It’s a highly technical question. I thought that Dr Raff explained it very well, yet commenters keep accusing her of saying things that she didn’t say. In particular she did not say “race is a social construct”. That’s a rather silly and ill-defined idea that is used by sociologists and post-modernists, not by geneticists. It’s a total red-herring/
I cannot actually see any disagreement between Sailer and Dr. Raff on any descriptive or empirical point. Their disagreements are political and moral.
Actually, I’ve come up with a better idea for conceptualizing race than the traditional Linnaean one:
http://www.vdare.com/articles/its-all-relative-putting-race-in-its-proper-perspective
I don’t expect people to grasp it when first exposed to it because it turns thinking about race upside down from race as subspecies to race as extended family. The only thinker who seems to have come up with it on his own is philosopher Philip Kitcher. But, it turns out to be extremely helpful.
So the descendants of Queen Victoria constitute a race, as do the descendants of mitochondrial Eve. Once you go back a small number of generations, you have as many races as ancestors. Billions of races.
Whatever the objections to that definition, you apparently agree with Dr. Raff’s criticism of Mr. Wade.
“So the descendants of Queen Victoria constitute a race …”
This inbred lineage’s congenital tendency toward hemophilia played a historic role in the downfall of czarism in Russia. Indeed, a lot of Europeans besides Russians resented that their royals were turning into an inbreeding racial group of primarily German origin. You’ll notice that House of Windsor has rebuilt its popularity over the last 33 years by marrying attractive nonroyal girls of British stock, with Prince William and especially baby Prince George being popular with the British people.
People keep asking you how you determine how far back to trace your family groups, given that it can result equally in one race or millions. Do you have an answer?
Not Sailer, but it obviously doesn’t matter. You can slice it up at any level to get the level of detail you want. And that’s half the point of what Sailer keeps writing. It’s a definition that relies on another definition we’re all familiar with the fuzziness of.
An extended family can be defined by a common ancestor. For course, some extended families could be defined by more than one common ancestors. (Presumably, the descendents of Queen Victoria are the same as the descendents of Prince Albert). More than a few generations back, pretty much all extended families would be “partially inbred.” So really, you have billions of partially-inbred extended families to choose from, and you pick the ones that are useful for your purposes.
Yes to both answers–it’s just a framework for creating the desired results
@ David Colquhoun,
Jennifer cites a statement that “In fact if you use the common level of genetic differentiation between populations used by zoologists to classify biological races (which they called subspecies) in other mammals, all humans consistently show up as just one biological race.”
What is the common level of genetic differentiation used by zoologists? The claim seems to go back to a misrepresentation of what Sewell Wright wrote in the 1970’s.
Also, there are numerous other species with _less_ genetic differentiation between populations but nonetheless are identified as having sub-species.
Yes. There is less genetic variation between the Asiatic Lion and the African Lion than there is between the ‘races’ of humans.
Great post, David. I was particularly impressed by your contribution to the understanding of cluster analysis. You really teased out the subtleties here. In contrast, Steve has contributed nothing of value to the discussion. Keep up the good work!
Jennifer, I have an idea that to your way of thinking, anyone who disagrees with you is a bigot – and highly peculiar into the bargain.
Now. I need to check. Do you think that lack of a work ethic is a bad thing, or something that needs to be changed? Or does it even matter?
Is your REAL beef with genetic genealogy, or even with genealogy? If you don’t think having a work ethic matters, you may also not believe family history matters, or that much of anything matters. Those of us who deeply research our family history, and their genetics, want to know in great depth who our ancestors were, what they were, how they lived, how they thought. We put great effort into identifying their ethnicity, even from long ago. We get attacked alot, by violently upset people who insist that it doesn’t matter. Their own family trees extend no further than their grandparents and a great grandparent or two. Their idea of family history is a scrapbook of family photos. One thing that puzzles me is what can so upset people to whom so little matters. Like, duh, man, whatever.
As to caring about ancestry, everyone on my genealogical genetic forums takes great pride in whatever their ancestors turn out to be, even when for instance they had no clue some of their ancestors were African. Genetics also provides tools to test suspicions and family stories about actual ethnic ancestry, as well as knock down my mother’s insistence that OUR Dewolf ancestors aren’t the wealthy and prominent 19th century slave traders. (Many of my most prominent Family Finder genetic matches are of that family.)
Now. Nicholas Wade is shallow and ditzy, and many of his ideas could use further development, but that does not make them untrue. People of British descent DO have more of a work ethic than the rest of the world. The reason for that is the Protestant ethic. This feature of Capitalist society developed in England and spread from England. It was carried by cultural, economic, British and American imperialism.
The actual progenitor of this cultural development were people of Saxon descent. They remained in a unique position relative to both the social structure of England, and the means of production, through medieval times and became Protestants, and developed industry. Their entire morality and outlook on life were as different as night from day from people of “Celtic” ancestry in Great Britain. Sexual morality, work ethic, ambition, the notion that people control their lives, are all key ways people of Saxon descent and people of Celtic descent vary, even within Britain. The rest of the world lacks our stress on work ethic because they do not fully participate in the Capitalist way of life – for better and worse.
The Puritan migration is an example of something geneticists assert; all mass migrations are genetically selective. Deep historical factors drive them, and they almost always involve some group that long ago had a distinct history, that makes them genetically different from those around them. Most of your article is a quite vituperative attack on the very idea that people with distinct histories are genetically distinct as well; that our genes can for instance tell me anything at all about my own ancestors’ history, or tell the English important information on how the Protestant ethic developed. I guess you must think it’s wrong to care about these things. Most likely, you are not inclined by temperament to care about much at all, except when you get upset by those of us who do, who you really don’t understand at all.
As to race, we have the people of Southeast Asia, the people of Africa, and the people of Europe and western Asia. These groups have been isolated from each other for tens of thousands of years. Recently important major structural factors distinguish their histories, and their fates relative to the world system and to the socioeconomic systems wehre they migrate. Anyone who, like Nicholas Wade, describes them in terms of obvious physical differences may be shallow, but they’re also human, and I think the same way. In fact most of us do – unless we’re too sanctimonious or incapble of caring enough to observe the world around us. Clearly, as of late, it also means that most people should be fired from our jobs, divested of whatever we own, and never make a living again. That’s, most people, now, not a few low-life dregs of whatever. Jesus taught us first and foremost not to be hypocrites.
It is interesting that nineteenth century ideas never die. Whatever the differences between the Protestant and Catholic peoples of the British Isles may be, they are surely not the product of long-term reproductive isolation.
“People of British descent DO have more of a work ethic than the rest of the world.”
What nonsense! Is that supposed to be a scientific argument?
When I first started following the HBD scene about a year ago, I was a little shocked that there are now hundreds of HBD blogs and hundreds of HBD tweeters, some of whom have thousands of followers.
But then you see the American Anthropology Association promulgating lies like “race is a social construct” and it’s little wonder.
If the AAA had any concern for the truth (which it doesn’t) it would put in bold font at the top of its letterhead: “Race is a biological reality but racism is morally wrong”. But this would be logical and true, so don’t expect so much from people completely indoctrinated by Cultural Marxism and at war with human nature.
And so, HBD will continue to grow in popularity and organizations like the AAA will continue to become a bigger joke…
One thing I have noticed about these “HBD”ers is that they (a) start with the proposition that they are just pointing out some genetic diversity in the global population, with no value judgments, (b) then they say IQ is the most important thing on earth and is essentially immutable (despite the Flynn effect, which they don’t like) and (c) try to convince you that a more Mexican America in the future would be a disaster. This makes the observer suspect that (a) is just a gateway drug.
Of course, that does not mean (a), properly qualified, isn’t also true. But it isn’t properly qualified, and from the reviews (I have not read the book), it looks like Wade doesn’t properly qualify it. He admits in the fine print that half his book is pure speculation, but that is precisely the dangerous half.
“He admits in the fine print that half his book is pure speculation, but that is precisely the dangerous half.”
It’s hardly “in the fine print:” it’s last four paragraphs of the first chapter, which you can read here:
http://isteve.blogspot.com/2014/05/wades-speculation-disclaimer.html
There are multiple ambiguities in the concept of “social construct”.
In one sense, any category is a social construct. Astronomers decide Pluto is a dwarf planet. They have good scientific reasons for doing so, but these are reasons of cost-benefit analysis. It is more useful for astronomy to classify Pluto with Ceres than with Neptune. There is no external-to-science criterion of dwarf planet vs. planet.
Of course, Pluto, Neptune and Ceres would exist if there were no astronomers. But they wouldn’t be classified. Even really difficult to argue with classifications (like between the body of a planet and empty space) wouldn’t be made if there were no classifiers. And the classifiers work within a historically-developed body of knowledge, which could have developed in different ways.
Folk concepts like tables and chairs are like these scientific concepts in that they are ultimately determined by human convenience and could be fuzzy in some cases. At the same time, we can’t just will a chair into existence or even socially agree that it is there.
On the other hand, the Denver Broncos and the House of Representatives are socially constructed in the further sense that there is no history-independent reason that teams and legislatures are particularly sensible ways of dividing up the world. It’s semiotics all the way down.
This distinction is orthogonal to the discrete/continuous distinction. Castes and classes are both socially constructed, but the first is discrete, while the second is continuous. Development of a multi-cellular organism from zygote to death is continuous, although for voting and legality of abortion purposes we (socially) divide it into discrete bits.
I assume (without any professional knowledge) that Professor Raff is correct that human genetic diversity is clinal. I spent some time in Ethiopia, and it was pretty clear to me from naive observation that all Ethiopians are African but some are more African than others. Ethiopians will bore you on the subject if you let them. If you travel north up the Nile basin, people would get more and more Caucasian. Crowds of Egyptian protesters include people who could be from Greece. A group of Greeks will include someone who could be from Bavaria. Und so weiter.
But an Anglo slave trader in the eighteenth century had no trouble distinguishing his deck hands from his human cargo. That is presumably because the gene flow from Nigeria to England and back was pretty slow. The two groups were discrete. There was no history-independent reason that he categorized the one group as slaves and the others as free labourers. But a society that decided it was illegal to own white Christians, but legal to own black heathens, could reliably rely on phenotypical markers to distinguish who was in which category.
One reaction to that history is to say “no one should own anybody.” A different, less sensible, reaction is to say, “there is no way to distinguish people whose ancestry is from West Africa from people whose ancestry is from the British Isles.” It is perfectly true that a system of slavery and then apartheid had difficulty dealing with people of mixed ancestry, and had to create arbitrary rules. If you ignore the moral issues, though, those rules are just like a voting age: legal systems sometimes find it convenient to make what is continuous discrete. The conceptual categories were still imposed on a pre-social reality of natural diversity.
It would be crazy to make the normative proposition of human equality contingent on an empirical claim about the nature of human genetic diversity. We want GPs and barbers to be able to make the distinction between people whose ancestry is mostly from West Africa and those whose ancestry is mostly from Northern Europe. It is when they don’t make that distinction, that equality is actually threatened.
There’s another sense in which racial groups are socially constructed: biologically. Presumably the first North American Indians came to a social decision to leave Beringia behind and commit themselves and their descendants to the unknown North America ahead of them. That group decision meant that their descendants would evolve under different selection pressures and with little genetic exchange with their relatives in Beringia.
Right now, the Amish in America are creating a sizable group (now numbering over 200,000) that has had relatively little gene flow in from the surrounding population (although there is substantial gene flow out into the general white American population) for about 10 generations in America. That’s very much a social construction, yet the Amish are developing their own suite of typical congenital problems. As Cochran and Harpending have argued, the substantial outflow of people in each generation who stop being Amish because it doesn’t suit them is likely increasing the average “plainness quotient” of those who choose to stay Amish:
http://isteve.blogspot.com/2014/03/cochran-harpending-paper-on-amish.html
There’s another sense in which racial groups are socially constructed: biologically.
That was actually Foucault’s interest. He was interested in how early modern “scientific” discourses tried to, and did, change the body. He didn’t have even a superficial knowledge of genetics, but he was very aware of the anxiety about the demographics, fighting strength and so on of the French people, and the motivation of much of the early social science to improve such things so that a repetition of 1871 could be avoided (partial success in 1914, not so much in 1940, better luck since then).
My point is just that scientists tend to have a cartoon version of post-modernism in their heads.
Scientific breeding of livestock was central to British economic and intellectual progress (e.g., Darwin and Galton). Thoroughbred racing became the pastime of Anglophiles all over Europe and Argentina.
Anglo-Americans had a rough “one drop rule” of racial categorization for part-Africans that leads to, say, Barack Obama declaring himself black and only black on the 2010 Census. Other groups, such as the Spanish and Portuguese, had a “clinal” system of racial categorization with distinctions based on fractions of ancestry. George Zimmerman, for example, would be a triracial “pardo” in his mother’s native Peru.
It’s not clear that, say, slaves were treated better in Brazil than in Virginia because of this different system. Both systems of categorization had various consequences.
The U.S., it’s worth noting, did not have a one drop system of segregation for people of part-American Indian background. The fact that Herbert Hoover’s Vice President Charles Curtis had grown up speaking Kaw on an Indian reservation added glamor to his image.
In other words, race is a social construct, in the same sense that adolescence or retirement are.
Different cultures can construct somewhat different retirement systems: e.g., the Japanese tend to retire at 55 from corporate jobs and get less responsible jobs, while Americans try to work longer at their highest jobs. But every culture deals one way or another with the fact that people run out of gas eventually.
Human genetic diversity tends to be fairly clinal along flat, well-watered landscapes. But 69% of the earth’s surface is water, other parts are deserts or mountain ranges. So, for instance, there was no cline to speak of between West Africa and Brazil in 1400. The earliest mutual ancestor of two typical people in West Africa and Brazil in 1400 might be 500 to 1,000 generations ago. So, while diversity is somewhat clinal in some places, that’s only if you ignore giant boundaries like oceans, the Sahara, and the Himalayas.
(ps—I’m not a professor yet 🙂 )
Reading this I’m reminded of that joke of the biologist, physicist and mathematician traveling on train in Scotland and they see a black sheep. Thats the joke.
On the money as usual… There is a good article at http://rationalwiki.org/wiki/Racial_realism
“Human genetics doesn’t work like race realists think it does.
Race realists spend a great deal of time and effort pointing out genetic differences between geographically separated populations in gene clustering research and insisting this is evidence for “races”.
In gene clustering research a set of populations is typically determined via subjective descriptors in ethnicity, language and geographics and people can be reliably identified as members of these groups. However, this way of categorizing people depends fundamentally on the quantity and method used to create the aforementioned framework of ancestral populations. Depending on what you subjectively chose to be aforementioned populations people may or may not end up in the same group. This is completely different from the problem of “races”, which presupposes that there is just one objectively and biologically demarcatable set of populations among all humans.”
This issue really ticks people off. I wonder why?
Because people rely on Rational Wiki to inform their opinions about disputes surrounding quantitative genomics and the philosophy of language.
@ Tom Goodfellow,
Well, if they are merely social constructs with no biological meaning you wouldn’t get clusters? BGI Cognitive Genomics Member Steve Hsu comments:
“A common argument is that 99.9 percent genetic similarity cannot leave room for “consequential” differences. But modern humans and Neanderthals are almost as similar (~ 99.8 percent; we have high accuracy sequences now for Neanderthals), and there are significant differences between us and them: both physical and cognitive. However, because humans and Neanderthals are known to have interbred, we are still part of the same species. (Would it be fair to refer to them as a separate “race”? Is the modern-Neanderthal difference merely a social construct?) Furthermore, this 0.1 percent genetic variation accounts for human diversity encompassing Confucious, Einstein, Shaq and Shakespeare.
Genetic variation is patterned — two individuals who trace their ancestry to the same geographical region (e.g., two Japanese) will have about 15 percent fewer total differences between them than if we were to compare individuals from widely separated ancestries (e.g., a Nigerian and a Japanese). This means hundreds of thousands of fewer differences between individuals from the same group than for two randomly selected people from different groups.
Gene variants (alleles) which are common in one population (e.g., 90 percent of Japanese have version A) can be rare in another (e.g., only 20 percent of Nigerians have version A). Differences in allele frequencies are correlated across populations.
From these correlations one can easily identify a genome (or even a small chunk of DNA as long as it includes many alleles) as belonging to a particular ancestral group. To oversimplify: one simply asks whether the DNA chunk in question has mostly the variants that are common in one group as opposed to another. Even if the differences in allele frequency are small — e.g., allele X is 62 percent likely in Japanese, versus 57 percent likely in Nigerians — once we consider thousands of such alleles the statistical signal becomes apparent. Each individual (or chunk of DNA) can be associated with a particular ancestral group.
Is this genetic difference consequential? Does it make two Nigerians more similar, on average, to each other than to a random European? Obviously, on some superficial phenotypes such as skin color or nose shape, the answer is yes.”
http://infoproc.blogspot.co.nz/2014/05/whats-new-since-montagu.html
I’ll agree with you I know nothing about this but I’m trying to correct my deficiency and understand this important issue because “race theory” has had consequences in the past ie justifying slavery, Ku Klux Klan, exterminations of Jews, creating arbitrary walls between groups etc
The way I understand it essentially the way one classes the relationship between populations genetic information is linguistically similar to separating sounds of a wave or light on a spectrum, where one starts and one finishes depends on the protocol/method one uses. The result of this is that difference/similarity between populations/individuals can be more complex and counterintuitive than popular opinion. Is that controversial?
Can you please tell me what the difference is between “consequential” differences and old fashioned plain “differences”?
Do you disagree with Dr Raff’s conclusions?
Re: XochtitlJimenez. It’s a wiki. I invite you and any other interested parites to add to it and make it better, if you are a professional in quantitative genomics or the philosophy of language consider it a civic duty. And If you have substantial arguments to what I wrote then I’m all ears.
Can you define race for me?
And if by “they” you mean race then don’t you think it’s a big leap to say that “clustering” can be used to argue that the differences in genes can explain the “rise of the West,” as Wade does. This determinism is confusing correlation with causation. Wade is shoehorning evidence, he’s a closet racist hiding behind fancy words, logic twisted by wishful thinking and confirmation bias.
Wade:
“Their social descent had the far-reaching genetic consequence that they carried with them inheritance for the same behaviors that had made their parents rich. The values of the upper middle class—nonviolence, literacy, thrift and patience—were thus infused into lower economic classes and throughout society.”
Does that sound fishy to you?
And Hsu:
“Hypothesis 1: (the PC mantra) The only group differences that exist between the clusters (races) are innocuous and superficial, for example related to skin color, hair color, body type, etc.
Hypothesis 2: (the dangerous one) Group differences exist which might affect important (let us say, deep rather than superficial) and measurable characteristics, such as cognitive abilities, personality, athletic prowess, etc.”
A straw man and a massive jump from the data. The word race has so many definitions that Hsu is inviting racists to commit equivocation, his “clusters” give a veneer of science to racial categories that are arbitrary.
Geography can go a lot further than genetics it explaining complexity of human social organization and power, see: Guns Germs and Steel.
@XochtitlJimenez If you think that this debate is only within the realms of “quantitative genomics and the philosophy of language,” then you are being willfully naive.
http://blogs.scientificamerican.com/primate-diaries/2014/05/21/on-the-origin-of-white-power/
@villandra24
“whatever their ancestors turn out to be, even when for instance they had no clue some of their ancestors were African.”
Oh, the horror. Here’s a news flash for you, everyone’s ancestors were African.
Here’s my question for Ms. Raff and anyone else who says race is a social construct: If so, then isn’t any system based on this non-existent thing called “race” also unscientific and without foundation?
Doesn’t it follow that all laws and regulations relating to affirmative action, racial preferences, etc. must be as invalid as Wade’s book? Heck, they’re inherently less valid, because laws sort people into a handful of socially-constructed “races” based on little more than self-identification. No genetic information required. And they’re not only classifying by race, they’re dispensing valuable things like university slots, public housing, jobs, and promotions, all based on something unscientific.
Even statistical studies regarding race must be similarly contradictory: e.g., no “race” can ever be said to be over- or under-represented, because it’s impossible to define “race.”
I think there’s an interesting paradox here: overall, the same people most committed to the “social construct” view are also likely to be supporters of racial preferences and the desire to racially classify people in the census, studies, etc. On the other side, the people who believe in the reality of racial differences (including all racists) seem to be the ones most likely to advocate abolishing racial preferences. Funny, that.
No, it doesn’t follow because laws can sensibly be based on social constructs. Money is a social construct.
The thing is Pothlord, that when people say its a social construct they imply there are no biological differences. For instance, they presumably think that genetically a Korean and a Finn would be as similar to each other as they are to somone from within their ethnic group. But that’s not the case. Also, in the context of medical research race, whether you want to call it a social construct or not, has biological significance.
edit: sorry, I misstyped your handle – Pithlord.
Even more absurd, does arguing that age is (in some plausible sense) a social construct imply being opposed to senior’s discounts?
Even more absurd, does arguing that age is (in some plausible sense) a social construct imply being opposed to senior’s discounts?
It does if there is no objective definition of “senior,” and instead the discounts are handed out according to anyone who self-identifies as a “senior,” and based on how old someone appears to someone else.
Is it really your contention that the people who say race is a social construct are arguing that there are no biological differences between a Korean and a Finn? If that’s your belief, then I would suggest that you really, truly don’t understand their arguments.
I see people disagreeing profoundly on political and moral issues, and then inappropriately transferring this to what is basically a semantic dispute in science. Is “age” a social construct? In a sense, yes. In a sense, no. Whatever. No one cares. But the issue about the relaionship between scientific knowledge concerning genetic differences among human populations and social concepts of race map directly onto age.
Referencing your age analogy, Americans cared a remarkable amount about age in the 1960s. I suspect that was because other identity politics concerns were muted, especially after Kennedy broke the barrier for Catholics becoming President. With the old Protestant – Catholic rivalry diminishing and immigration low, that opened the way for other identity disagreements, including the Generation Gap. Today, we almost never hear about the Generation Gap in part because we have so much racial and ethnic diversity. Today, young blacks and Hispanics aren’t supposed to rebel against their elders because they are supposed to be loyal to their group.
The association of race with biological difference in medicine is real ONLY in the sense that the perception of someone’s “race” can be a clue to his/her genetic makeup. So despite the fact that many individuals perceived as “black” are of primarily European heritage, it still makes sense to identify all “black people” as possible carriers of a disease such as sickle cell anemia, because there is no easy way to determine someone’s actual biological ancestry. The fact that racial perceptions are useful as clues does NOT make race a science. It’s a useful tool for medicine, that’s it.
Affirmative Action works a bit differently. It is completely based on race as a social construct, because the need for Affirmative Action arose as a result of racial stereotyping. As far as discrimination, past and present, is concerned, it matters not if you are mostly of European ancestry — so long as you are identified as “black” you are subject to discrimination. The discrimination is NOT based on your genes, so the genetic aspect is irrelevant. It’s based on how you are perceived. And so is AA.
To treat race as a social construct is not to dismiss it as unimportant. In many ways it is far more important in our society than genetic science. The difference is that one is not science and the other is.
Your view that AA is required to remedy past discrimination is not the view of the US Supreme Court. Its view is that it is justified to provide for a more diverse group of students, cops or whatever than would occur without affirmative action. Differences in representation in elite institutions or under market conditions are not largely caused by present-day discrimination.
Racial identification is a useful tool in medical diagnosis (sometimes). That is because it is an imperfect proxy for genetic realities. Whether an imperfect but socially useful proxy for underlying continuously distributed biological differences is a “social construct” would depend on what you mean by that rather ambiguous phrase. Anyway, social constructs are not necessarily scientifically useless.
“So despite the fact that many individuals perceived as “black” are of primarily European heritage,”
In the United States, there really weren’t many people who self-identify as black who are less than about half black, at least not until after the civil rights era. You hear about it a lot because African-American elites in the NAACP and West Indian elites such as the Attorney General tend to be quite white, but that takes a lot of careful breeding.
An early genomic study about a dozen years ago of self-identifying black adults in the U.S. found only 10% were less than half sub-Saharan African by ancestry. The technology has improved a lot since then, so the number might have changed but if you look at people passing down the street, you’ll see that is roughly true in people over 40 or so.
That’s not true in places that didn’t have a One Drop Rule like Brazil and that’s not true for races to which the One Drop Rule didn’t apply such as American Indians. I’ll leave it to you to work out the genealogical logic.
Is “retirement” a social construct? Does a “yes” to that question imply that there is no biological significance to age?
@Pithlord: So, it’s impossible to define races scientifically, because races are a social construct without clear edges, but there’s nothing wrong with laws that treat people differently according to race, because races are just a social construct? Yeah, that makes sense.
papayasf,
Sarcasm sounds better when you understand what you are responding to. In three lines, you attributed to me three propositions I don’t agree with, while not responding to my point. I didn’t say races are a “social construct” (since I think that is a fundamentally ambiguous concept, and races are a social construct in one sense and not in another sense). I don’t think that whether they have clear edges is relevant to whether they are a social construct (courtesy is a social construct, and it doesn’t have hard edges; citizenship is a social construct and it usually does have hard edges). I didn’t say that laws are justified because they apply to social constructs (I can’t even think of an example of anyone who would think that).
I responded to your argument that there is an inconsistency between believing that race is a social construct and thinking it should be used in law. I pointed out that law uses social constructs all the time (money, jurisdictional and property boundaries, etc., etc.)
In some cases, law uses discrete categories that are socially imposed on continuous biological difference. Social Security is an example. There is a biological process of senescence. It does not happen overnight on anyone’s 65th birthday. No one even thinks it happens overnight on someone’s 65th birthday. But a system of legal entitlements usually finds it convenient to use clear cut discrete categories, even if they are imperfect proxies for the underlying difference. This is the rules vs. standards debate in law.
For the purposes of the debate our host provoked, race is basically like “old age”, “middle age”, “adolescence” and “childhood.” There are continuous underlying biological differences. Different societies carve things up different ways. If you want to call that a social construct, fill your boots. Planets are also social constructs.
Race is different from age because of the history of colonialism, slavery and segregation. And because racial/ethnic conflict is a major potential threat to civil order everywhere, while the “generation gap” is just a premise for sit coms. Basically, everyone on this thread could probably agree with my last paragraph if they weren’t emotionally and politically invested in this conflict. No one really worries whether old age is a social construct or not.
Whether affirmative action is justified depends on whether you think there are good reasons for the law to take race into account sometimes. I happen to think that there are. You do not want a city in which half the population are African Americans and all the police are Irish. That’s not even good for the Irish. So that’s an easy case for justifying affirmative action. The fact that there was in 1491 a genetic gradient along via Central Africa, East Africa, the Nile Basin, and the Middle East between the Irish and the non-European ancestors of African Americans is neither here nor there.
Many commenters here seem to confuse “scientific” with the kind of hard-edged divisions that are more common in law than in nature. Legal systems try to develop “bright line” categorization rules — e.g., in the U.S. you can’t vote if you are under 18-years-old. (It used to be age 21.) That keeps things simple.
The social science concept related to this strict division is “maturity.” We can use the tools of social science to come up with some very hazy data on when people tend to be mature enough to vote, then argue over what the hard and fast law should be.
Plato spoke of trying to carve nature at the joints. That well recognizes the inherent messiness of biological science. It’s not a reason to declare messy subject matters “unscientific.”
No matter what you call the basis for the idea of race, social construct, or genetic, or whatever, laws are needed that treat races differently to rectify the wrongs that have been done to people in the name of race.
Also to ensure reasonable representativeness in elite institutions (and non-elite, but important institutions like the police) in the future.
At the risk of going off-topic, the problem with the “rectifying past wrongs” argument is the problem of all forms of “group justice”: it conflicts with individual justice. You end up punishing and rewarding individuals, who may well have suffered or committed no wrongs, based merely on their membership in different groups (which are supposedly impossible to scientifically define).
In the past, various classes of people have been discriminated against, for their religion, the color of ther skin, etc. This goes on today in, for example, housing, employment, and other areas. Laws are needed to protect entire classes of people from discrimination. Are you punishing individuals because they are unable to discriminate? You are inventing problems where none exist.
Yes, law often uses discrete categories because it reduces the decision and compliance costs of getting to the underlying messy variable. It isn’t really possible to treat people as individuals. Well, maybe your mother can, but a judge can’t. Law always treats people as instantiations of categories. Equity comes along and tries to deal with the injustice caused by doing this by taking into account more context. But that in turn creates inconsistency of treatment, and delay, and excessive discretion, which requires equity to turn into a kind of law. And so on.
Life isn’t fair. White Americans are basically the group in history that life has been least unfair to, so enough with the whining.
It’s “Dr” Raff, actually, if you’re using titles. But no need to be formal around here. Jennifer is fine.
***Because there is no clear dividing line, there are no distinct races—that is the nature of variation within a species.” (p 92).***
This objection was met decades ago by Dobzhansky:
Professor Fried has correctly pointed out that there is no careful and objective
definition of race that would permit delimitation of races as exact, nonoverlapping,
discrete entities. Indeed, such criteria do not exist because if they did,
we would not have races, we would have distinct species. (Dobzhansky in Mead
1968, 165)
Click to access SesardicRasse.pdf
Reblogged this on Shane O'Mara's Blog and commented:
This is a superb and important piece. I would make a few comments from brain science regarding this debate. There has been a very serious and large-scale effort in neuropsychiatric genetics and neurological genetics to pin down genes of relevance to conditions such as depression, schizophrenia and degenerative disorders such as Alzheimer’s. To put it mildly, these ‘in your face brain’ disorders (there’s little subtlety about these conditions) have yielded little by way of really meaningful effect sizes or gene relationships. The literature is replete with examples of disappearing effect sizes, where initial large effects disappear on replication or increased sample size (here are a few: BDNF; 5’-HTTLPR; genes for ‘alcoholism’; have a look at http://deevybee.blogspot.ie/2012/04/getting-genetic-effect-sizes-in.html). Instead, the picture is very complicated with multiply-related small effects – and this is in well-controlled studies with registers and a disabling psychopathological condition.
The transgenic literature where gene candidates have been modulated (I use the term generously to mean ko/ki or conditional) has been similarly bedevilled. The behavioural impairments in the AD mouse are small and subtle and really require a combination of insult or senescence and triple transgenes to show really reasonable effects.
And in the case where pure genetic determinism would seem to unavoidable (Huntington’s), the data in the animal models show that the condition is moderated dramatically by environmental enrichment and substantial aerobic exercise. Even in autistic spectrum disorders, ‘Family studies have shown that the risk of having ASD is 10-20 per cent if you have an affected first-degree relative’ (Gallagher and Mitchell), which is a long way from a sense of determinism. To jump from this pretty mechanist literature to genetic determinism and IQ (or intellectual function) is a stretch too far, when we know so little.
What’s my overall point? It is this: IQ data are observational data, and causal inferences about underlying neurogenetics from such observational data should only be made with extreme care. And even more so when we know from experimental data that ‘Poverty impedes cognitive function’ by as much as a standard deviation (Science. 2013 Aug 30;341(6149):976-80; Science. 2012 Nov 2;338(6107):682-5; http://shaneomara.wordpress.com/2013/12/01/contra-boris-johnson-iq-doesnt-ensure-life-success-non-cognitive-skills-and-poverty-matter-greatly-too/). There’s a lot more: there have been many candidate genes to enhance cognitive function (and indeed drugs too). When you actually look at the data, there are many profound problems – such as hypoalgesia or excess neuronal excitability – or even simply inconsistent or non-replicable effects. A little less myside bias in reasoning would be good in this domain.
I’ve added some more details and citations to my comment above at: http://shaneomara.wordpress.com/2014/05/23/nicholas-wade-and-race-building-a-scientific-facade/
We know very little about the genes for height and exactly how they function. Those genes appear to be of great multitude and of very small effect size–a famous recent GWAS now demonstrates this, as another GWAS, modeled on the the one for heights, also demonstrates that at minimum 51% of fluid g (a key component of IQ) is due to genetic variation. The links can, I believe, be found elsewhere in these comments.
No, we don’t yet understand the etiology related to the many genes in either the case of height or of IQ. But the crucial thing going forward is that we know that they exist, and, however they manage to do it, play a major role in height and IQ respectively.
People in brain science and other disciplines need to discard their ideology and come to a reckoning with these facts, and the sooner the better.
This article has so many logical fallacies in it, I at first thought it was some sort of parody. But these fallacies have all been addressed by others so I won’t dwell on them.
Anyway, as was noted at a bio-evo workshop I attended, the reason the whole “race is a social construct” crowd (with is either lying or stupid) has sacrificed scientific integrity for politics is because at this point in time racial characteristics are permanent.
However, in the near future, as someone at the workshop noted, we probably will be able to genetically modify people so that blacks and amerindians then could both behaviorally and cognitively be more like whites / Europeans and NE Asians. At this point, all people will affirm the biological reality of race because the problems will be able to be fixed.
Nonetheless, honesty is important regardless of the consequences, and it’s better now to be honest about the biological reality of race. Please, stop the lies. It only makes you look ridiculous.
What looks ridiculous is the idea that blacks need to be “modified” so they can be more like whites.
May I be genetically-modified so that I bear no resemblance to DJ?
I’m curious, DJ, which workshop did you attend?
Holy mother of God. You don’t even try to hide your bigotry in a cloud of “high-minded” words.
It’s very illuminating, isn’t it?