# How many races are there? **URL:** \ **Category:** Great Debates\ **Created:** [September 5, 2002, 8:34pm UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179 "2002-09-05T20:34:33Z")\ **Posts on this page:** 20\ **Page:** 3
**Author:** ![tomndebb](https://avatars.discourse-cdn.com/v4/letter/t/b9e5f3/32.png) [@tomndebb](https://boards.straightdope.com/u/tomndebb)\ **Post date:** [September 15, 2002, 10:29pm UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/41 "2002-09-15T22:29:33Z")
> [@](#): > > Then read the Race FAQ. It explains how the “strong” position is incorrect. From the FAQ: > [@](#): > > First, the “strong” interpretation: A person’s race has absolutely no predictive value about that person’s genetic makeup, with the minor exception of genes that influence skin color, hair color, etc. Then: > [@](#): > > However, there is a problem with the “strong” interpretation: it’s clearly false. In other words, a person’s race does have predictive value about his or her genetic makeup. The classic example of this point - that race has predictive value about alleles - is the allele for Sickle Cell Anemia which is more common among blacks than non-blacks. But a person’s “race” has no predictive value regarding their possession of the Sickle Cell trait. Millions of “blacks” do not have the Sickle Cell trait and millions of “blacks” _are not susceptible to_ the Sickle Cell trait. It is not predictive about a person’s genetic makeup (which was the statement you assigned to the “strong” argument). ---
**Author:** ![Tars\_Tarkas](https://avatars.discourse-cdn.com/v4/letter/t/6a8cbe/32.png) [@Tars\_Tarkas](https://boards.straightdope.com/u/Tars_Tarkas)\ **Post date:** [September 15, 2002, 11:19pm UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/42 "2002-09-15T23:19:35Z")
Just when you think it is safe to tread in GD… > [@](#): > > However, there is a problem with the “strong” interpretation: it’s clearly false. In other words, a person’s race does have predictive value about his or her genetic makeup. The classic example of this point - that race has predictive value about alleles - is the allele for Sickle Cell Anemia which is more common among blacks than non-blacks. Sounds like an argument for POPULATIONS to me. (since it is not universal, only SOME blacks are more likely to have it, and SOME medditerranians [spe;;ed horribly wrong] are more likely to have it.) > [@](#): > > it does not rule out the possibility of significant genetic differences among the races I’m not sure we’ve found anything major. I could be wrong, writing this off the cuff as i have to go soon. > [@](#): > > Let’s face reality: If you claim that “race doesn’t exist,” and what you actually mean is that “race should not be used in science because of fuzziness and historical baggage,” you’re misleading your audience. We are saying race doesn’t exist as a scientific construct. It IS a sociological construct (as has been said in several if not all of these threads). ---
**Author:** ![lucwarm](https://avatars.discourse-cdn.com/v4/letter/l/e19adc/32.png) [@lucwarm](https://boards.straightdope.com/u/lucwarm)\ **Post date:** [September 16, 2002, 12:43am UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/43 "2002-09-16T00:43:14Z")
> [@](#): > > \*Originally posted by tomndebb \* > \*\*Millions of “blacks” do not have the Sickle Cell trait and millions of “blacks” _are not susceptible to_ the Sickle Cell trait. It is not predictive about a person’s genetic makeup (which was the statement you assigned to the “strong” argument). \*\* If you know that a person is black, it follows that there is a higher probability that the person carries the SCA allele than otherwise. Thus, a person’s race has predictive value as to his or her genetic makeup. It sounds as though you don’t understand what “predictive value” means. (It’s also possible that I’m using the term in a non-standard way.) I suggest you google on the term “predictive value” and think about it a bit. ---
**Author:** ![lucwarm](https://avatars.discourse-cdn.com/v4/letter/l/e19adc/32.png) [@lucwarm](https://boards.straightdope.com/u/lucwarm)\ **Post date:** [September 16, 2002, 12:54am UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/44 "2002-09-16T00:54:33Z")
\*\* > [@](#): > > Sounds like an argument for POPULATIONS to me. (since it is not universal, only SOME blacks are more likely to have it, and SOME medditerranians [spe;;ed horribly wrong] are more likely to have it.)\*\* I’m not sure what you mean by an “argument for POPULATIONS.” In any event, the fact that the SCA allele is distributed as it is lends support to the “medium” interpretation described in the Race FAQ. If you want to argue all day in favor of the “medium” interpretation, I can’t stop you. \*\* > [@](#): > > We are saying race doesn’t exist as a scientific construct. \*\* Ok, let me rephrase what is in the FAQ: > [@](#): > > Let’s face reality: If you claim that “race doesn’t exist **as a scientific construct** ,” and what you actually mean is that “race should not be used in science because of fuzziness and historical baggage,” you’re misleading your audience. By the way, do you mind if I quote your previous post on that website? ---
**Author:** ![tomndebb](https://avatars.discourse-cdn.com/v4/letter/t/b9e5f3/32.png) [@tomndebb](https://boards.straightdope.com/u/tomndebb)\ **Post date:** [September 16, 2002, 4:05am UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/45 "2002-09-16T04:05:13Z")
> [@](#): > > If you know that a person is black, it follows that there is a higher probability that the person carries the SCA allele than otherwise. And if I go to Capetown and pull 50 “blacks” and 50 “whites” off the street and test them for Sickle Cell and they all test negative, what has your Predictive Value shown us? (True Positive / (True Positive + False Positive)) × 100 = Predictive Value (0 / (0 + 0)) × 100 = 0 (or undefined with the division by 0) ---
**Author:** ![Tars\_Tarkas](https://avatars.discourse-cdn.com/v4/letter/t/6a8cbe/32.png) [@Tars\_Tarkas](https://boards.straightdope.com/u/Tars_Tarkas)\ **Post date:** [September 16, 2002, 4:20am UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/46 "2002-09-16T04:20:56Z")
Which part? if you do the “I’m not sure we’ve found anything major.” make sure to add the "I could be wrong, writing this off the cuff as i have to go soon. " otherwise i could end up looking like a fool. You should probably include a link to this thread at the bottom. I used populations (which was capitalized for emphasis, not yelling, hindisight is 20/20) because that is the term scientists usually use. Populations fit here because the North Africans are the ones that have most of the SCA trait while the sub-saharans have less. Here are some links for you to read: [http://www.aaanet.org/stmts/racepp.htm](http://www.aaanet.org/stmts/racepp.htm) which i got from [http://www-sul.stanford.edu/depts/ssrg/misc/race.html](http://www-sul.stanford.edu/depts/ssrg/misc/race.html) (some of the links require membership) [http://www.annieappleseedproject.org/racclingoods.html](http://www.annieappleseedproject.org/racclingoods.html) ---
**Author:** ![tomndebb](https://avatars.discourse-cdn.com/v4/letter/t/b9e5f3/32.png) [@tomndebb](https://boards.straightdope.com/u/tomndebb)\ **Post date:** [September 20, 2002, 2:02am UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/47 "2002-09-20T02:02:05Z")
Hmmm. Some posts have been lost to the crash. OK. From my memory, **lucwarm** is actually using “predictive value” not as a measure of the validity of testing (which is the only way I have seen it used in a technical setting), but as a “general view of the proportion of occurrences.” This would mean that my interpretation that the perceived “race” of an individual had to be “predictive” of values in _every individual_ was a misinterpretation of his intent. Since the whole “strong” version is **lucwarm** ’s invention, I would say that he gets to set the rules of what it means. ---
**Author:** ![lucwarm](https://avatars.discourse-cdn.com/v4/letter/l/e19adc/32.png) [@lucwarm](https://boards.straightdope.com/u/lucwarm)\ **Post date:** [September 20, 2002, 9:30am UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/48 "2002-09-20T09:30:06Z")
> [@](#): > > \*Originally posted by tomndebb \* > \*\*Hmmm. Some posts have been lost to the crash. > > OK. From my memory, **lucwarm** is actually using “predictive value” not as a measure of the validity of testing (which is the only way I have seen it used in a technical setting), but as a “general view of the proportion of occurrences.” \*\* Just to make things clear, I am using the term as follows: A condition “A” has predictive value as to trait “X” if the satisfaction of condition A changes the probability that an individual bears trait X. i.e. if P(X) != P(X|A) or, more specifically, if “S” represents the trait of carrying the SCA allele, and “B” represents the condition of being black, then P(S) != P(S|B). In essence, your post above attempts to argue that both sides of the inequality are equal by adding an extra condition - being from Capetown. So you are saying that: P(S|C) = P(S|B&C). Possibly true, but irrelevant. \*\* > [@](#): > > This would mean that my interpretation that the perceived “race” of an individual had to be “predictive” of values in _every individual_ was a misinterpretation of his intent.\*\* Agree. \*\* > [@](#): > > Since the whole “strong” version is **lucwarm** ’s invention, I would say that he gets to set the rules of what it means. \*\* Sadly the “strong” viewpoint is based on actual posts I have seen on these boards. ---
**Author:** ![Telemark](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/telemark/32/372_2.png) [@Telemark](https://boards.straightdope.com/u/Telemark)\ **Post date:** [September 20, 2002, 3:13pm UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/49 "2002-09-20T15:13:05Z")
If I am understanding your agrument correctly, you have used circular reasoning. How do you define “black” in this case? Is it people of African ancestory? North African, Southern African (I understand only one of the two groups have a true statistical correlation to the SCA allele)? People with dark skin and curly hair? If the latter, you must include Australian Aborigines, correct? After all, it gives you the exact same statistical correlation so there is no reason not to include them. And if I found another group with dark skin and curly hair you’d have to include them in your race, right? How about Standard Poodles? They seem to qualify, and amazingly, they produce the exact same statistical correlation. It seems that any arbitrary collection of populations will produce the statistical correlation you are describing. Assume it’s the north African population that actually has the SCA allele (I’m taking this from notes lost in the past few days, sorry no cite), the fact that you can add in the southern population and still get a result says very little. The fact that you can use Lithuanian farmers instead of southern Africans and get the same result shows that this method of defining race has little or no use. ---
**Author:** ![lucwarm](https://avatars.discourse-cdn.com/v4/letter/l/e19adc/32.png) [@lucwarm](https://boards.straightdope.com/u/lucwarm)\ **Post date:** [September 20, 2002, 10:56pm UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/50 "2002-09-20T22:56:25Z")
> [@](#): > > \*Originally posted by Telemark \* > \*\*If I am understanding your agrument correctly, \*\* I’m not sure if this comment is addressed to me, but assuming it is, you don’t understand the argument at all. See below. \*\* > [@](#): > > The fact that you can use Lithuanian farmers instead of southern Africans and get the same result shows that this method of defining race has little or no use. \*\* And this is a decent argument in favor of the “medium” viewpoint discussed in the Race FAQ. Irrelevant to the “strong” viewpoint, however. You do understand that there’s a difference, don’t you? Anyway, it’s interesting how people seem so irresistably drawn to making points of the sort you just made. ---
**Author:** ![system](https://global.discourse-cdn.com/straightdope/original/2X/e/e489c3b7d8fce19c4b355dd4fc3f88cc39c34b87.png) [@system](https://boards.straightdope.com/u/system)\ **Post date:** [September 24, 2002, 3:46pm UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/51 "2002-09-24T15:46:09Z")
There is a simple answer here. Data. And what the primary investigators say. Now, lucwarm et al continue a fine little game of “but what if” sophistry, or more clearly, argument to ignorance (argumentum ad ignorantiam). A fine little logical fallacy our ever so-well-informed contrarians keep falling back on, resting their essential claim that race is true proposition of some utility (\*\* in the context of the biological argument \*\*) only because it hasn’t been proved false to some “principled” standard, although our dear lucwarm has failed to define in a rigorous _scientific_ manner what principled actually indicates in the context of applying this knowledge or unknowledge as the case may be. Of course, he also seems unable and unwilling to substantiate much of his argumentation and accusations, preferring straw men. However, it has readily been apparent that there is a general illiteracy in re the substance throughout this argument in re the contrarian position. I draw attention to the illiteracy of the argument in the prior thread that somehow Whale and Human could be conceived of as a “ring species” – an illiterate abuse of the concept worthy of the creationists. Or here in this thread that somehow the lumping of disparate groups together to create statistical artefacts is a useful exercise. To undertake a counter argument in a technical field, one has the burden of first informing oneself as to the meaning and usage of terminology and standards in that field, rather than to blunder about blindly flailing about without the slightest grounding. Further that, the hand-waiving claim that if somehow “race” is not a category of examination (in the context of human genetics) that genetic differences will go unfound, as we will see. A clear argument to ignorance insofar as the wealth of **primary** research to date identifying such differences, research which has consistently led such researchers to view race as a category that obstructs understanding actual structuring of human populations by imposing unsupported I draw attention, \*\*\* once more \*\*\* to the material which was evidently not understood the first time around – although this is hardly a novelty at this time. Indeed let me repeat myself from . > [@](#): > > Data continued > The issues are quite clear. > > (a) For the better part of the last ten years the genetic data has consistently pointed away from the old classic races having real utility for scientific inquiry into populations, on a biological basis. Ten years of accumulated science keeps running in the same direction. The old race concept leads to mis-categorization and gets in the way of forming new, better understandings of human population structure. Now some folks seem to be unable to wrap their minds around the nuance of differentiating between populations – defined on objective, biological bases—and the hoary old races. Population differences, however trivial on a macro-basis clearly exist, but they’re not well structured along the old race concept lines. > (b) That leaves race as a sociological concept – which on one level is fine if the researcher or practitioner is aware of the nature of the concept and does not reify it as a coherent biological one. It strikes me that time and time again we’ve seen studies which have pointed again and again to inappropriate treatment, inappropriate categorization of an issue as biological rather than socio-economic/societal (e.g. differing treatments received by black Americans even controlling for economic status.) – in short, generalized bad medicine and bad science caused by the very kind of sloppy thinking that our ‘race is useful’ crowd engages in. Let me draw attention to (a). Perhaps it is difficult for our contrarians to grasp this, but it remains the fact that race has been an obstacle even in medical inquiry. The recent NEJM article and editorial inviting primary researchers and medical practitioners to abandon race as their primary means of thinking about population differences _on a biological level_ was motivated by the accumulation of “revisions” of medical “facts” finding issues previosuly assumed (due to improper conceptualization) to be rooted in race-biology were in fact rooted in socio-environmental differences. (“Racial disparities in Medical Care” Editorial. NEJM May 2001, p. 1471-3, also see for example Schwartz LM, et al. “Misunderstandings about the effects of race and sex on physicians’referrals for cardiac catheterization.” NEJM 341 1 Jul 1999, p. 279-83; 286-7) Differential treatment patterns (e.g. heart disease, for unknown reasons black Americans receive differential treatment (sub-standard) resulting in a large statistical difference in results) – the hand-waving counter argument that the contrarians simply don’t think that socio-economic/environmental differences can produce large group differences is utterly unsupported by data. That is there is clear data from a variety of different ethnic groups/races/populations which clearly indicate such differences can and do produce large, observable differentiations in population health outcomes – which is to say the expression of their underlying genetics. I further note that the contrarians continue to labor under the largely inexplicable attachment to the idea they are “more closely” related to someone of their own race – the question asked in the other thread I believe, and nonsensically several times, was whether it was not more likely to be more closely related genetically to someone of one’s own race than not. That question is in fact quite simple. Some probability calculations should resolve it, if the askers would simply reflect on the general observation that around 85% of **variation** does not map unto groups, only around 10-15% does map unto groups, that variation being tied to, of course, regional environmental selection factors such as climate etc. I believe that some simple probability calculations can relieve them of their misapprehensions. Now this brings us the much ignored paper, I have cited several times Chiara Romualdi et al “Patterns of Human Diversity, within and among Continents, Inferred from Biallelic DNA Polymorphisms” Genome Vol. 12, Issue 4, 602-612, April 2002 which noted: > [@](#): > > [emphasis added: once again] > Previous studies have reported that about 85% of human diversity at Short Tandem Repeat (STR) and Restriction Fragment Length Polymorphism (RFLP) autosomal loci is due to differences between individuals of the same population, whereas differences among continental groups account for only 10% of the overall genetic variance. [The general observation] **These findings conflict with popular notions of distinct and relatively homogeneous human races, and may also call into question the apparent usefulness of ethnic classification in, for example, medical diagnostics.** Here, we present new data on 21 Alu insertions in 32 populations. We analyze these data along with three other large, globally dispersed data sets consisting of apparently neutral biallelic nuclear markers, as well as with a -globin data set possibly subject to selection. We confirm the previous results for the autosomal data, and find a higher diversity among continents for Y-chromosome loci. We also extend the analyses to address two questions: (1) whether differences between continental groups, although small, are nevertheless large enough to confidently assign individuals to their continent on the basis of their genotypes; \* (2) whether the observed genotypes naturally cluster into continental or population groups when the sample source location is ignored.\* Using a range of statistical methods, we show that classification errors are at best around 30% for autosomal biallelic polymorphisms and 27% for the Y chromosome. Two data sets suggest the existence of three and four major groups of genotypes worldwide, respectively, **and the two groupings are _inconsistent._** These results suggest that, at random biallelic loci, there is little evidence, if any, of a clear subdivision of humans into biologically defined groups. This is but the most recent paper, and this one clearly responding to the question of whether one can tease meaning out of macro-structures in populations. And, **once again,** the answer is no, the data does not support the tired old macro-populations known as the classical races. Returning to the argument then: Even those traits derive from multiple alleles such that the same, more or less, skin color can not be said to mean that the persons ipso facto even share the same alleles for that phenotype. Different paths to the same surface expression. Further, as cited in the article above, a mere 10-15% of the overall package of our genetic diversity varies by region. 90% varies on an individual basis: around 90% of your variation is individual and the remainder is regional such that 90% of the difference in between any two humans does not map unto whatever regional grouping one may want. Now, let me dredge up past questions from prior threads that I should have responded to, but had not the time. > [@](#): > > \*Originally posted by IzzyR \* > How significant is this lack of evidence? …. It would seem to me that since race has at best an extremely limited utility in science - if any at all – \*\* it is likely that the matter has never and will never be fully studied. \*\* emphasis added, focus on the issue of evidence and not the underlying question of sport. **Absolute and complete misunderstanding.** Ongoing research on populations is what gives us the data. If there are ‘racial,’ or better some population differences, they _ **will be** _ uncovered. They will be uncovered and better understood by structuring that research in a manner which best fits the data. In a real sense, the question you ask suffers from incomprehension of where the science is going. There is no lack of study of population differences – that is the essential science on this – other than of course lacks driven by limited resources. Structuring and determining the research on ‘racial’ grounds simply does not help with uncovering, rather it introduces _inefficient_ hypotheses. Your worry then is almost entirely misplaced. If the data emerges which indicates say upper Sahelian derived populations have some package of traits which give an advantage at the margin – well then it **will emerge** , and then when the alleles are identified, research will look for them. Abandoning the race concept as the driver and looking for specific traits allows one to come up with better, more accurate and ultimately more efficient hypotheses and thus better research. Simple as that. In other words, if you’re worried that somehow population factors will be missed the **best** way not to miss them, given our current state of knowledge is to ditch the classical race concept and build up rigorous population models – and not allow the sociological issues to confuse the primary research. Further returning to a past question on differences and performance: > [@](#): > > Which would leave it in the realm of speculative and possibly unimportant but possible. In the area of athletic accomplishment, the evidence in favor of a Black advantage is only at the upper end. There is no evidence (that I know of) to suggest an average advantage. In light of this, the significance of any possible genetic differences being very small is lessened. This because even a very small difference in the overall distribution can produce a disproportionately large difference at the extreme tails. So while it would not support “race in the classical sense” it would possibly be a cause of the observed disparity. Wonderful, now this is what I was getting at. Now you’re advancing something which makes sense on the evidence. Indeed, if you look back to the linked discussion you’ll note that this was among the hypotheses I advanced as areas where someone looked for biological explanations to complement the socio-cultural factors would want to look. ---
**Author:** ![system](https://global.discourse-cdn.com/straightdope/original/2X/e/e489c3b7d8fce19c4b355dd4fc3f88cc39c34b87.png) [@system](https://boards.straightdope.com/u/system)\ **Post date:** [September 24, 2002, 4:05pm UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/52 "2002-09-24T16:05:12Z")
Now, I am going to break my promise to myself and introduce yet more materials. It is perhaps an exercise in almost delusional optimism that this primary literature, from peer reviewed journals, will make the slightest dent in our sophists’ argumentum ad ignorantiam, however there is enough of interest here that it may be worth while. First, the brand new review and comment on the literature by David B. Goldstein and Lounès Chikhi “Human Migrations And Population Structure: What We Know and Why it Matters” Annu. Rev. Genomics Hum. Genet. 2002. 3:129-152; and if I do not lose steam L.B. Jorde, W.S. Watkins and M.J. Bamshad1 “Population genomics: a bridge from evolutionary history to genetic medicine” Human Molecular Genetics, 2001, Vol. 10, No. 20 2199-2207. I also would like to draw attention to the following: Luísa Pereira, Isabelle Dupanloup, Zoë H. Rosser, Mark A. Jobling and Guido Barbujani “Y-Chromosome Mismatch Distributions in Europe” Molecular Biology and Evolution 18:1259-1271 (2001) which is less directly relevant but interesting article illustrating direction of research and structure of populations. Michael F. Hammer and Stephen L. Zegura1 “The Human Y Chromosome Haplogroup Tree: Nomenclature and Phylogeography of Its Major Divisions” Annu. Rev. Anthropol. 2002. 31:303-321. Again an interesting overview which I looked at very rapidly, notes further support for OoA in a complex form, uses multidimensional anlaysis to get at groupings, and also notes, that although Africa exhibits greater divergence among haplogroups, Asia contains the largest number of major haplogroups, with a multidimensional scaling analysis plot indicates that the Americas, Africa, and East Asia are outliers, whereas the rest of the world forms a large central cluster. Interestingly, they advance a high percentage for the examined NRY haplogroups, 43% of the total variance, as attributable to differences among populations. Now, these citations aside let us look at what Goldstein and Chikhi have to say first. The article presents an engaged overview of the current literature and issues, and argues for a more sophisticated and engaged use of population genetics, historical observations on populations based on the same, and medical genetics. It also directly addresses the issues often raised here. I think the article is interesting enough to quote extensively from it. Much of the commentary and analysis revolves around the importance of linkage disequilibrium analysis (the non-random association between alleles at different sites/loci) as a means of getting at complex traits (as opposed to monogenetic) – which would include almost all non-infectious disease of concern to most modern medicine. However, as they note, linkage disequilibrium in any given population is driven by numerous factors of which admixture, genetic drift, bottlenecks, rate of mutation are important. These bear directly on the historically focused genetic research. There is, then, a direct tie between better genetic medicine and better, more sophisticated understanding of genetic populations and their histories. (As I have frequently argued) As the authors state “the optimal design of association studies depends not only on the relationship between genetic variation and disease status but it must also be guided by what is known about the pattern of LD [nb: linkage disequilibrium] in the population and genomic region under study.” They note for example that it “ is also clear … that demographic events such as bottlenecks and admixture have elevated LD in particular populations. For example, non-African populations generally have much higher levels of LD than African ones (Tishkoff et al. 1996) (Reich et al. 2001) and admixture was shown to have dramatically increased LD in a Bantu-Semitic hybrid populations (Wilson & Goldstein 2000).” Now they also further discuss the complexity of untangling disease-allelic relationships, which bears quoting en extenso: > [@](#): > > Equally important to efforts to map the genes underlying common diseases is the genetic control of common disease. Chakravarti (Chakravarti 1999) has outlined two extreme possibilities concerning the nature of the mutations that influence common disease. At one extreme the mutations are rare and often population specific, and at the other extreme they are common and generally found in multiple human populations. The second possibility is usually referred to as the common disease common variant, or the CD/CV hypothesis. One argument that has been marshalled in favour of CD/CV hypothesis is that common diseases might result from a mismatch between alleles and their environments and/or their genetic backgrounds. Indeed a flagship example for the CD/CV hypothesis is the APOE_4 allele, which appears to be an example of such a mismatch. APOE_4 is a risk factor for coronary artery and Alzheimer’s disease and ranges in frequency from less than 10% to over 40% in different human populations. Strikingly, APOE_4 frequencies appear to be associated with the historical diets of populations. In particular, APOE_4 frequencies are lowest in populations that have long practised agriculture (Carbo and Scacchi 1999). Thus APOE\*4 may be an example of thrifty allele that is deleterious in high nutrient environments (Carbo and Scacchi 1999). The truth is, however, that we have extremely few examples of alleles that underlie the common diseases, so there is no way to know which view is closer to reality, in general. > > The issue is critical, however, because population-based gene mapping will only succeed for mutations that conform to the CD/CV hypothesis. Unfortunately, there is one discouraging piece of evidence. While we know very little about the genetic control of the common diseases, we know a great deal about the genetic control of simple disease caused by a single locus. And as emphasized by Chakravarti (Chakravarti 1999), the message here is unambiguous. Most loci underlying Mendelian diseases have a diverse allelic spectrum. That is, many different mutations, often of low frequency, cause disease, and these mutations often differ among populations. Now, perhaps of most interest to us is there discussion of “Geographic structure of medically-relevant mutations” Again, their discussion merits quoting en extenso, from the beginning: First they situate the issue as follows, readers will note that their argumentation on this issue is precisely what I have presented to date, of course providing further citational support: > [@](#): > > One definite and obvious consequence of the complexity of human demographic history is that races in any meaningful sense of the term do not exist in the human species. The term race as popularly imagined implies groups that can be cleanly separated from one another, and within our species, there simply aren’t any such groups. Rather, differences among groups of humans are always graded, and decisions about whom to cluster with whom on genetic grounds always must include arbitrary criteria. > > Not only are differences among groups graded, they are also modest. No matter how such groups are defined, it is well known that the majority of the genetic variation in the human species is due to differences between individuals within rather than between groups. Starting with electrophoretic analyses by Lewontin (Lewontin 1972) and moving through the entire range of genetic markers, authors have always found that differences between groups count for less than 15% of the total genetic variation in our species (Nei & Roychoudhury 1974) (Nei & Roychoudhury 1993)), (Barbujani et al. 1997b) (Jorde et al. 2000). > > The explanation for this pattern is surely that we are a demographically young species, with most of our genetic variation tracing to a relatively small precursor population, or populations, in Africa, that recently colonized the rest of the world. Questions about the pattern of genetic variation in human populations, however, have more than just academic significance. For example, Exner et al (2001) recently reported that enalapril, an angiotensin-converting enzyme inhibitor, has lower efficacy in Black patients compared with Whites, and the FDA (Food and Drug Administration) has recently approved clinical trials of a combination of drugs for congestive heart failure specifically for use in African Americans (Anon, 2001). The use of racial/ethnic labels in this context, however, has been sharply criticized as a dangerous example of reification (e.g. Schwartz 2001). We are sympathetic to this criticism for the reasons outlined in the paragraph above, but we also note that average genetic differences in individuals from different parts of the world may, in some cases, be relevant to drug evaluation. _It would be obvious to any serious human population geneticist, however, that racial labels would not be an appropriate approach for dealing with geographic structuring of variable drug response._ Emphasis added. Why? Well, as argued and as my citations supra indicated, variation does not map onto races, it is incoherent on that level. Or to further follow the authors: > [@](#): > > An alternative to the use of racial labels in the context of drug evaluation was recently outlined by Wilson et al. (Wilson et al. 2001a) who used a panel of 39 microsatellite markers to assess relatedness among individuals in a set of populations samples from throughout the world. The geographic labels of the individuals were suppressed, and a likelihood based model (see Pritchard et al. (Pritchard et al. 2000) for a description of the method) was used to estimate the amount of ancestry each individual has from each of the inferred sub groups. In the Wilson et al. (Wilson et al. 2001b) study, including groups from Europe (Norway, Ashkenazi Jews), Central and East Asia (Armenia, South China, PNG), East and West Africa, and Afro-Caribbeans, the clustering procedure identified four genetic clusters. Wilson et al then compared the frequencies of mutations at genes encoding drug metabolizing enzymes among these genetic clusters as a surrogate for whether the clusters might respond differently, on average, to certain classes of drugs. Wilson et al found that 4 of 6 mutations considered have highly significant differences among the clusters, suggesting that there is considerable scope for differences among the clusters in drug response. Wilson et al also emphasized that these clusters had a generally poor correspondence with ethnic labels, suggesting that the latter should not be used a guide to genetic structure in the context of drug evaluation. For example, Bantu individuals from Southern Africa were largely included in one cluster, but individuals from Ethiopia were largely included in a separate cluster that included most Europeans. Additionally, the two Asian populations (South West China and Papua New Guinea) were largely in separate clusters. > > It would be entirely incorrect, however, to interpret these results as providing a new definition of races. Even these clusters, which in a sense are the most genetically distinct groupings that can be found, always show graded differences (even ignoring model assumptions). And in the context of drug response, the distribution of mutations makes quite clear that any given response present in one group is likely to be present in the others, although possibly in different frequency. Thus the most significant unit of analysis will always be the individual, though knowledge of cluster membership may be of clinical relevance when the causal genes are not known, and therefore cannot be typed directly. I do note the emphasized sections of text, which further confirm similar observations by Roumaldi et al, as cited above. Again the evidence speaks strongly against the utility of the classic race concept in a bio-medical context for confusing rather than clarifying the source of medical problems. Now an interesting side point, but of utility for illustrating the importance understanding population histories is their discussion of “Population differentiation and gene function” connected with a side issue on Neanderthal. They propose that looking SNPs (Single Nucleotide Polymorphisms: most frequently found DNA sequence variations in the genome) across multiple populations as a way to get at a variety of important insights as loci under selection may be identified through differences in the average degree of differentiation between discrete populations. They argue that the “dramatic increase in genomic data” should allow for assessment of large sets of variants and identification of empirical outliers. > [@](#): > > The key point, however, is that soon it will be possible to assemble large sets of SNPs that are far from known genes. The distributions for these, presumably neutral patterns would be compared to SNPs of interest. A particularly encouraging aspect of this program is that it can be carried out for haplotypes, as opposed to SNPs. The genomic pattern of recombination apparently results in extremely simple haplotype diversity over genomic stretches ranging in size from about 5 to about 100 kb (e.g. (Daly et al. 2001) (Goldstein 2001). Hence it should be possible to build up a picture of the expected neutral distribution of haplotype differentiation among populations, against which haplotype differentiation in specific genes, or classes of genes may be compared. For example, Wilson et al (Wilson et al. 2001a) showed that there is a relatively high level of differentiation of functional mutations at genes for drug metabolizing enzymes. Because of the role of these loci in detoxifying compounds in the diet, it is possible that they show greater differentiation among populations than other classes of genes, perhaps in correlation with historical diets. This is now the sort of hypothesis that can be directly evaluated through the study of human genetic diversity. An interesting suggestion of where and how differences btw populations, in terms of distributional frequencies will be located. Well, this is already far too long, I think I shall place the Neandertal materials in the Cecil column thread on “why white people look different,” as it is more directly relevant. The essence in the end is the same. Or let me put it in another way, let’s face reality, my dear contrarians have yet to present any kind of scientific argument. Or to illustrate another way, in response to Telemark pointedly drawing lucwarm’s uneducated attention to the incoherence of his argument, he responds: > [@](#): > > The fact that you can use Lithuanian farmers instead of southern Africans and get the same result shows that this method of defining race has little or no use. > > > [@](#): > > > > And this is a decent argument in favor of the “medium” viewpoint discussed in the Race FAQ. Irrelevant to the “strong” viewpoint, however. You do understand that there’s a difference, don’t you? I believe he understood the arguments defined on that unlearned collection maundering to be straw men. Rather, he was pointing out that your “method” is nothing but a useless statistical artefact without underlying explanatory value. > [@](#): > > Anyway, it’s interesting how people seem so irresistably drawn to making points of the sort you just made. Yes, interesting. It probably would be ever more interesting to observe the high degree of correlation between an education in genetics and biology and the advancing of pointed commentary in regards to the incoherence of arguments made by those not so educated. Or the dodges and lack of substance in the replies by the same. ---
**Author:** ![sweetsensation](https://avatars.discourse-cdn.com/v4/letter/s/da6949/32.png) [@sweetsensation](https://boards.straightdope.com/u/sweetsensation)\ **Post date:** [September 24, 2002, 4:46pm UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/53 "2002-09-24T16:46:07Z")
I believe there is one race. i dont see why every breaks them down. there are many diffent skin clor many diffent backround belief places but ine race. bruna ---
**Author:** ![The\_Flying\_Dutchman](https://avatars.discourse-cdn.com/v4/letter/t/9fc348/32.png) [@The\_Flying\_Dutchman](https://boards.straightdope.com/u/The_Flying_Dutchman)\ **Post date:** [September 24, 2002, 4:47pm UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/54 "2002-09-24T16:47:58Z")
> [@](#): > > _Collounsbury_ > Further returning to a past question on differences and performance: > > ## quote: > > ## Which would leave it in the realm of speculative and possibly unimportant but possible. In the area of athletic accomplishment, the evidence in favor of a Black advantage is only at the upper end. There is no evidence (that I know of) to suggest an average advantage. In light of this, the significance of any possible genetic differences being very small is lessened. This because even a very small difference in the overall distribution can produce a disproportionately large difference at the extreme tails. So while it would not support “race in the classical sense” it would possibly be a cause of the observed disparity. > > Wonderful, now this is what I was getting at. Now you’re advancing something which makes sense on the evidence. Indeed, if you look back to the linked discussion you’ll note that this was among the hypotheses I advanced as areas where someone looked for biological explanations to complement the socio-cultural factors would want to look. Well **Jon Entine** would agree with that. So would I. Now if only you volunteered this little tidbit to answer the innumerable queries regarding why blacks seem to dominate athletically two years ago rather than leave us with what appeared to be the _only_ alternative answer namely “cultural influences” I’m afraid I missed your post where you first acknowledged this. What linked discussion? > [@](#): > > This because even a very small difference in the overall distribution can produce a disproportionately large difference at the extreme tails. As far as I know, I was the first poster to present **Jon Entine** conclusions particularly the qualification that evidence of the “black advantage” exists only at the elite level. Never once did you acknowlege the rationality of my position in this regard but instead ridiculed me and **Jon Entine** each and every time you responded to me. What made you change your mind? ---
**Author:** ![system](https://global.discourse-cdn.com/straightdope/original/2X/e/e489c3b7d8fce19c4b355dd4fc3f88cc39c34b87.png) [@system](https://boards.straightdope.com/u/system)\ **Post date:** [September 24, 2002, 5:03pm UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/55 "2002-09-24T17:03:33Z")
> [@](#): > > \*Originally posted by grienspace \* > Well **Jon Entine** would agree with that. Jon Entire advanced an illiterate mish-mash of misunderstanding, his “black excellence” does not respond to the data. > [@](#): > > So would I. Now if only you volunteered this little tidbit to answer the innumerable queries regarding why blacks seem to dominate athletically two years ago rather than leave us with what appeared to be the _only_ alternative answer namely “cultural influences” The hypotheses I advanced were done so long before, in regards to restricted specified populations and a warm-climate phenotypical advantage hypothesis _in conjuction_ with socio-cultural constraints and influences. All on the same thread. Your lack of understanding of the same, notwithstanding. > [@](#): > > As far as I know, I was the first poster to present **Jon Entine** conclusions particularly the qualification that evidence of the “black advantage” exists only at the elite level. Never once did you acknowlege the rationality of my position in this regard but instead ridiculed me and **Jon Entine** each and every time you responded to me. What made you change your mind? No change of mind, not one whit. Your position, and Entine’s nonsensical ignorant argument, are still fully deserving of ridicule for their over-reaching, unfounded conclusions. I was encouraging Izzy’s further specification of where he was thinking and attempting to encourage him to think further-- evidently I became too generous for it encouraged the unformed to scurry forth. ---
**Author:** ![The\_Flying\_Dutchman](https://avatars.discourse-cdn.com/v4/letter/t/9fc348/32.png) [@The\_Flying\_Dutchman](https://boards.straightdope.com/u/The_Flying_Dutchman)\ **Post date:** [September 24, 2002, 5:16pm UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/56 "2002-09-24T17:16:04Z")
Again. What linked discussion in your Part I post are you referring to ? ---
**Author:** ![system](https://global.discourse-cdn.com/straightdope/original/2X/e/e489c3b7d8fce19c4b355dd4fc3f88cc39c34b87.png) [@system](https://boards.straightdope.com/u/system)\ **Post date:** [September 24, 2002, 5:31pm UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/57 "2002-09-24T17:31:02Z")
I don’t recall - this was a partial piece I decided to rescue from my harddrive’s obscurity. In retrospect I should have eliminated portions to focus on the immediate issue. ---
**Author:** ![lucwarm](https://avatars.discourse-cdn.com/v4/letter/l/e19adc/32.png) [@lucwarm](https://boards.straightdope.com/u/lucwarm)\ **Post date:** [September 25, 2002, 1:37am UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/58 "2002-09-25T01:37:08Z")
\*\* > [@](#): > > Or here in this thread that somehow the lumping of disparate groups together to create statistical artefacts is a useful exercise.\*\* Fascinating how people continue to confuse the “strong” and “medium” viewpoints. It’s like moths drawn to a flame. Anyway, I’m not going to bother responding to the rest of your nonsense. I don’t really see much point in your conclusory meanderings, but perhaps they are therapeutic to you. ---
**Author:** ![system](https://global.discourse-cdn.com/straightdope/original/2X/e/e489c3b7d8fce19c4b355dd4fc3f88cc39c34b87.png) [@system](https://boards.straightdope.com/u/system)\ **Post date:** [September 25, 2002, 1:28pm UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/59 "2002-09-25T13:28:57Z")
> [@](#): > > \*Originally posted by lucwarm \* > Fascinating how people continue to confuse the “strong” and “medium” viewpoints. It’s like moths drawn to a flame. No, not really. Rather strawmen are strawmen, and the data is the data. > [@](#): > > Anyway, I’m not going to bother responding to the rest of your nonsense. I don’t really see much point in your conclusory meanderings, but perhaps they are therapeutic to you. Yes, nonsense. Of course _you_ don’t see a point, lacking a grounding in what you pretend to speak to. Uncomfortable that, this primary research thing, actual data as opposed to empty sophistry, unsupported by any glimmering of science, eh my dear fellow? ---
**Author:** ![lucwarm](https://avatars.discourse-cdn.com/v4/letter/l/e19adc/32.png) [@lucwarm](https://boards.straightdope.com/u/lucwarm)\ **Post date:** [September 25, 2002, 3:53pm UTC](https://boards.straightdope.com/t/how-many-races-are-there/127179/60 "2002-09-25T15:53:46Z")
> [@](#): > > \*Originally posted by Collounsbury \* > \*\* > > strawmen are strawmen, and the data is the data.\*\* Feel free to point out (in a non-conclusory manner) where and how I have misrepresented anyone’s position. \*\* > [@](#): > > Of course _you_ don’t see a point, lacking a grounding in what you pretend to speak to.\*\* Umm, I don’t see a point because your posts are so nonsensical. \*\* > [@](#): > > Uncomfortable that, this primary research thing, actual data as opposed to empty sophistry, unsupported by any glimmering of science, eh my dear fellow? \*\* None of the data you have posted contradicts anything I’ve said. I imagine you are unaware of this, since your posts display such minimal comprehension of the relevant issues. [Previous page](https://boards.straightdope.com/t/how-many-races-are-there/127179.md?page=2) [Next page](https://boards.straightdope.com/t/how-many-races-are-there/127179.md?page=4)