Jump to content

Talk:Human Genetic Diversity: Lewontin's Fallacy

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia


Sources for response to and impact of Edwards 2003

[edit]

This article has been through a lot of controversy over the years. It was nominated for deletion, but kept, its name has been changed, and it has undergone considerable edit-warring, with a current posting of a talk page notice about Arbitration Committee discretionary sanctions. I've been watching this page for a few years, and in recent weeks there has been a flurry of talk page discussion here, alas with remarkably little updating of the article text. I've been reading university textbooks on human genetics "for fun" since the 1980s, and for even longer I've been visiting my state flagship university's vast BioMedical Library to look up topics on human medicine and health care policy. On the hypothesis that better sources build better articles as all of us here collaborate to build an encyclopedia, I thought I would suggest some sources for improving this article and related articles. The Wikipedia guidelines on reliable sources in medicine provide a helpful framework for evaluating sources.

The guidelines on reliable sources for medicine remind editors that "it is vital that the biomedical information in all types of articles be based on reliable, third-party, published sources and accurately reflect current medical knowledge."

Ideal sources for such content includes literature reviews or systematic reviews published in reputable medical journals, academic and professional books written by experts in the relevant field and from a respected publisher, and medical guidelines or position statements from nationally or internationally recognised expert bodies.

The guidlines, consistent with the general Wikipedia guidelines on reliable sources, remind us that all "Wikipedia articles should be based on reliable, published secondary sources" (emphasis in original). They helpfully define a primary source in medicine as one in which the authors directly participated in the research or documented their personal experiences. By contrast, a secondary source summarizes one or more primary or secondary sources, usually to provide an overview of the current understanding of a medical topic. The general Wikipedia guidelines let us know that "Articles should rely on secondary sources whenever possible. For example, a review article, monograph, or textbook is better than a primary research paper. When relying on primary sources, extreme caution is advised: Wikipedians should never interpret the content of primary sources for themselves."

On the topic of human population genetics and variation within and among human populations, a widely cited primary research article is the 1972 article by Richard Lewontin, which I have seen cited in many of the review articles, monographs, and textbooks I have read over the years, and to which the Edwards paper that this Wikipedia article is about responds.

  • Lewontin, Richard (1972). "The Apportionment of Human Diversity" (PDF). Evolutionary Biology. 6. Springer: 381–398. Archived from the original on 25 November 2010. Retrieved 23 November 2013. {{cite journal}}: Unknown parameter |deadurl= ignored (|url-status= suggested) (help)

As Wikipedians, we can evaluate where the findings in Lewontin's article fit in the current understanding of the topic of human genetic variation by reading current reliable secondary sources in medicine.

Some years after Lewontin published his primary research article on human diversity, when his primary research results had been replicated in many other studies and his bottom line conclusion that "about 85% of the total genetical variation is due to individual differences within populations and only 15% to differences between populations or ethnic groups" had been taken up by many textbooks on genetics and medicine, A. W. F. Edwards wrote a commentary essay in the journal BioEssays

which is the topic of this Wikipedia article and in which Edwards proposes a statistical model for classifying individuals into groupings based on haplotype data. Edwards wrote, "There is nothing wrong with Lewontin’s statistical analysis of variation, only with the belief that it is relevant to classification," pointing to his own work with Luigi Luca Cavalli-Sforza, the author of the book

which I read soon after it was published in 1994. In his 2003 article, Edwards cites a lot of publications from his collaboration with Cavalli-Sforza. Edwards also mentions that collaboration prominently in his subsequent review article

in which he describes their method for tracing ancestry with genes. Edwards even shows a photograph of Cavalli-Sforza with him in 1963 in his 2009 article, emphasizing their scholarly friendship.

So I wanted to look up Cavalli-Sforza's current views as well while I traced citations of the Lewontin 1972 article and the Edwards 2003 article in subsequent secondary sources. Wikipedia has a suggested template for articles about nonfiction books, and notability guidelines for nonfiction books, but nothing comparable for research journal articles, very few of which have their own Wikipedia articles. It looks to me from those analogous guidelines that discussing the scholarly reception of the Edwards 2003 essay in more detail would be helpful for updating this Wikipedia article.

Through searches with Google, Google Scholar, and Google Books, both from my home office computer and from a university library computer, I found a number of books and articles that cite both the Lewontin paper and the Edwards paper. Through a specialized set of wide-reaching keyword searches (for example, "Lewontin Edwards") on the university library's vast database subscriptions, I was able to obtain the full text of many of those articles and of whole books that discuss what current science says about grouping individuals of species Homo sapiens into race groups. I also found more up to date discussions by Luigi Luca Cavalli-Sforza of the Human Genome Diversity Project.

Listed here are sources that have the following characteristics: (1) they cite both previous articles by Lewontin and the 2003 article by Edwards, discussing the underlying factual disagreement between those authors, (2) they are Wikipedia reliable sources for medicine (in particular, they are secondary sources such as review articles or textbooks rather than primary research articles), and (3) they are currently available to me in full text through book-buying, library lending, author sharing of full text on the Internet, or a university library database. They are arranged in approximate chronological order, so that you can see how the newer sources cite and evaluate the previous sources as genetics research continues. The sources listed here are not exhaustive, but they are varied and authoritative, and they cite most of the dozens of primary research articles on the topic, analyzing and summarizing the current scientific consensus.

  • Whitmarsh, Ian; Jones, David S., eds. (2010). What's the Use of Race?: Modern Governance and the Biology of Difference. Cambridge (MA): MIT Press. ISBN 978-0-262-51424-8. {{cite book}}: Unknown parameter |laydate= ignored (help); Unknown parameter |laysummary= ignored (help)
  • Ramachandran, Sohini; Tang, Hua; Gutenkunst, Ryan N.; Bustamante, Carlos D. (2010). "Chapter 20: Genetics and Genomics of Human Population Structure". In Speicher, Michael R.; Antonarakis, Stylianos E.; Motulsky, Arno G. (eds.). Vogel and Motulsky's Human Genetics: Problems and Approaches (PDF). Heidelberg: Springer Scientific. pp. 589–615. doi:10.1007/978-3-540-37654-5. ISBN 978-3-540-37653-8. Retrieved 29 October 2013. Most studies of human population genetics begin by citing a seminal 1972 paper by Richard Lewontin bearing the title of this subsection [29]. Given the central role this work has played in our field, we will begin by discussing it briefly and return to its conclusions throughout the chapter. In this paper, Lewontin summarized patterns of variation across 17 polymorphic human loci (including classical blood groups such as ABO and M/N as well as enzymes which exhibit electrophoretic variation) genotyped in individuals across classically defined 'races' (Caucasian, African, Mongoloid, South Asian Aborigines, Amerinds, Oceanians, Australian Aborigines [29] ). A key conclusion of the paper is that 85.4% of the total genetic variation observed occurred within each group. That is, he reported that the vast majority of genetic differences are found within populations rather than between them. In this paper and his book The Genetic Basis of Evolutionary Change [30], Lewontin concluded that genetic variation, therefore, provided no basis for human racial classifications. ... His finding has been reproduced in study after study up through the present: two random individuals from any one group (which could be a continent or even a local population) are almost as different as any two random individuals from the entire world (see proportion of variation within populations in Table 20.1 and [20]). {{cite book}}: Unknown parameter |laydate= ignored (help); Unknown parameter |laysummary= ignored (help)
  • Tattersall, Ian; DeSalle, Rob (1 September 2011). Race?: Debunking a Scientific Myth. Texas A&M University Anthropology series number fifteen. Texas A&M University Press. ISBN 978-1-60344-425-5. Retrieved 17 November 2013. Actually, the plant geneticist Jeffry Mitton had made the same observation in 1970, without finding that Lewontin's conclusion was fallacious. And Lewontin himself not long ago pointed out that the 85 percent within-group genetic variability figure has remained remarkably stable as studies and genetic markers have multiplied, whether you define populations on linguistic or physical grounds. What's more, with a hugely larger and more refined database to deal with, D. J. Witherspoon and colleagues concluded in 2007 that although, armed with enough genetic information, you could assign most individuals to 'their' population quite reliably, 'individuals are frequently more similar to members of other populations than to members of their own.' {{cite book}}: Unknown parameter |laydate= ignored (help); Unknown parameter |laysummary= ignored (help)
  • Barbujani, Guido; Colonna, Vincenza (15 September 2011). "Chapter 6: Genetic Basis of Human Biodiversity: An Update". In Zachos, Frank E.; Habel, Jan Christian (eds.). Biodiversity Hotspots: Distribution and Protection of Conservation Priority Areas. Springer. pp. 97–119. doi:10.1007/978-3-642-20992-5_6. ISBN 978-3-642-20992-5. Retrieved 23 November 2013. The massive efforts to study the human genome in detail have produced extraordinary amounts of genetic data. Although we still fail to understand the molecular bases of most complex traits, including many common diseases, we now have a clearer idea of the degree of genetic resemblance between humans and other primate species. We also know that humans are genetically very close to each other, indeed more than any other primates, that most of our genetic diversity is accounted for by individual differences within populations, and that only a small fraction of the species' genetic variance falls between populations and geographic groups thereof.
  • Bliss, Catherine (23 May 2012). Race Decoded: The Genomic Fight for Social Justice. Stanford University Press. ISBN 978-0-8047-7408-6. {{cite book}}: Unknown parameter |laydate= ignored (help); Unknown parameter |laysummary= ignored (help)
  • Barbujani, Guido; Ghirotto, S.; Tassi, F. (2013). "Nine things to remember about human genome diversity". Tissue Antigens. 82 (3): 155–164. doi:10.1111/tan.12165. ISSN 0001-2815. The small genomic differences between populations and the extensive allele sharing across continents explain why historical attempts to identify, once and for good, major biological groups in humans have always failed. ... We argue that racial labels may not only obscure important differences between patients but also that they have become positively useless now that cheap and reliable methods for genotyping are making it possible to pursue the development of truly personalized medicine.

By the way, the Barbujani, Ghirotto, and Tassi (2013) article has a very interesting discussion of SNP typing overlaps across the entire individual genome among some of the first human beings to have their entire individual genomes sequenced, with an especially interesting Venn diagram that would be a good graphic to add to this article.

An author who is intimately familiar with Edwards's statistical approach, because he has been a collaborator in fieldwork and co-author on primary research articles with Edwards, is Luigi Luca Cavalli-Sforza. Cavalli-Sforza is a medical doctor who was a student of Ronald Fisher in statistics, who has devoted most of his career to genetic research. In an invited review article for the 2007 Annual Review of Genomics and Human Genetics, Cavalli-Sforza joins issue directly with the underlying factual disagreement among previous authors, but cites different previous publications.

GENETIC VARIATION BETWEEN AND WITHIN POPULATIONS, AND THE RACE PROBLEM

In the early 1980s, Lewontin (11) showed that when genetic variation for protein markers is estimated by comparing two or more random individuals from the same populations, or two or more individuals from the whole world, the former is 85% as large as the latter. This means that the variation between populations is the residual 15%, and hence relatively trivial. Later research carried out on a limited number of populations and mostly, though not only, on protein markers has confirmed this analysis. The Rosenberg et al. data actually bring down Lewontin’s estimate to 5%, or even less. Therefore, the variation between populations is even smaller than the original 15%, and we also know that the exact value depends on the choice of populations and markers. But the between-population variation, even if it is very small is certainly enough to reconstruct the genetic history of populations—that is their evolution—but is it enough for distinguishing races in some useful way? The comparison with other mammals shows that humans are almost at the lower extreme of the scale of between-population variation. Even so, subtle statistical methods let us assign individuals to the populations of origin, even distinguishing populations from the same continent, if we use enough genetic markers. But is this enough for distinguishing races? Darwin already had an answer. He gave two reasons for doubting the usefulness of races: (1) most characters show a clear geographic continuity, and (2) taxonomists generated a great variety of race classifications. Darwin lists the numbers of races estimated by his contemporaries, which varied from 2 to 63 races.

Rosenberg et al. (16 and later work) analyzed the relative statistical power of the most efficient subdivisions of the data with a number of clusters varying from 2 to 6, and showed that five clusters have a reasonable statistical power. Note that this result is certainly influenced by the populations chosen for the analysis. The five clusters are not very different from those of a few partitions that had already existed in the literature for some time, and the clusters are: (a) a sub-Saharan African cluster, (b) North Africa–Europe plus a part of western Asia that is approximately bounded eastward by the central Asian desert and mountains, (c) the eastern rest of Asia, (d ) Oceania, and (e) the Americas. But what good is this partition? The Ramachandran et al. (15) analysis of the same data provides a very close prediction of the genetic differences between the same populations by the simplest geographic tool: the geographic distance between the two populations, and two populations from the same continent are on average geographically closer than two from different ones. However, the Rosenberg et al. analysis (16) adds the important conclusion that the standard classification into classical continents must be modified to replace continental boundaries with the real geographic barriers: major oceans, or deserts like the Sahara, or other deserts and major mountains like those of central Asia. These barriers have certainly decreased, but they have not entirely suppressed genetic exchanges across them. Thus, the Rosenberg et al. analysis confirms a pattern of variation based on pseudocontinents that does not eliminate the basic geographic continuity of genetic variation. In fact, the extension by Ramachandran et al. of the original Rosenberg et al. analysis showed that populations that are geographically close have an overwhelming genetic similarity, well beyond that suggested by continental or pseudocontinental partitions.

A year later Cavalli-Sforza joined seventeen other genetics researchers as co-authors of a review article, published as an "open letter" to other scholars, on using racial categories in human genetics.

  • Lee, Sandra; Mountain, Joanna; Koenig, Barbara; Altman, Russ; Brown, Melissa; Camarillo, Albert; Cavalli-Sforza, Luca; Cho, Mildred; Eberhardt, Jennifer; Feldman, Marcus; Ford, Richard; Greely, Henry; King, Roy; Markus, Hazel; Satz, Debra; Snipp, Matthew; Steele, Claude; Underhill, Peter (2008). "The ethics of characterizing difference: guiding principles on using racial categories in human genetics" (PDF). Genome Biology. 9 (7): 404. doi:10.1186/gb-2008-9-7-404. ISSN 1465-6906. Retrieved 3 December 2013. We recognize that racial and ethnic categories are created and maintained within sociopolitical contexts and have shifted in meaning over time Human genetic variation within continents is, for the most part, geographically continuous and clinal, particularly in regions of the world that have not received many immigrants in recent centuries [18]. Genetic data cannot reveal an individual's full geographic ancestry precisely, although emerging research has been used to identify geographic ancestry at the continental and subcontinental levels [3,19]. Genetic clusters, however, are far from being equivalent to sociopolitical racial or ethnic categories. {{cite journal}}: Unknown parameter |displayauthors= ignored (|display-authors= suggested) (help)CS1 maint: unflagged free DOI (link)

I look forward to updating this article extensively in collaboration with other Wikipedians on the basis of these and any other reliable secondary sources you find. P.S. A special thank you to Wikipedians BlackHades and ArtifexMayhem, both of whom suggested sources on my user talk page as I was looking up sources for editing this and related articles. -- WeijiBaikeBianji (talk, how I edit) 19:45, 3 December 2013 (UTC)[reply]

Browsing through these sources, there appears to be common theme that is repeatedly mentioned. The sources consistently state that there is no consensus or agreement among scientists on the race concept and that many scientists accept the existence of biological races and find the classification useful, while many others don't. Jonathan Marks might have summed this up perfectly:

"Race, we must rather conclude, is underdetermined by genetics (Morning 2011). That is to say, it is genetically real when geneticists who believe it is real brandish their particular genetic data and statistical analysis (Risch et al. 2002, Edwards 2003), and it is unreal when geneticists who do not believe it is real brandish their genetic data and statistical analysis...the number of races of Europeans, say, may range from one (on the latest US Census, in 2010) to three (Ripley 1899) to five (Boyd 1963) to twelve (Coon 1939). None of these conclusions is more right than any of the others, for they are all coproduced by the natural facts of difference and by the cultural facts of classifying."

Marks, Jonathan (October 2013). "The Nature/Culture of Genetic Facts"

All the scientists are looking at and agree on the same objective data, but it is the interpretation of the data using subjective definitions and classifications that creates the disagreement. WeijiBaikeBianji's list of sources seem to either be from scientists that either are opponents of the race concept or neutral on the topic. Given the fact that all the sources, even the ones that oppose the race concept, consistently state there is disagreement and a lack of consensus in the scientific fields on the race concept (while citing plenty of sources from both sides of the arguments), a well balanced collection of sources for the article is needed. It cannot consist of only neutral or opposing perspectives. It must include proponent perspectives as well or else it creates WP:NPOV and WP:DUE issues. BlackHades (talk) 22:19, 3 December 2013 (UTC)[reply]
I think if you have read other writings by Jonathan Marks (as I have, many times), you would better be able to grasp his meaning there. (I give you credit for trying to grapple with the meaning of a sentence that I think is inartfully written.) It is widely agreed by a lot of scientists that "race" has social meaning. In other words, to be treated as belonging to one race category or another, in one society or another, makes a big difference in an individual's life, up to and including changing health outcomes and educational opportunity. A classic nonfiction book on this reality is Black Like Me by Howard Griffin.
Griffin, John Howard (1961). Black Like Me. Robert Bonazzi (afterword). New York: Signet. ISBN 0-451-19203-6. {{cite book}}: Unknown parameter |laydate= ignored (help); Unknown parameter |laysummary= ignored (help)
I wonder how many people would dare to try Griffin's experiment, even today? Meanwhile, there is a huge increase in the actual data we have now about genomic variance among human individuals, and the following two statements are well established facts: (1) "about 85% of the total genetical variation is due to individual differences within populations and only 15% to differences between populations or ethnic groups" and (2) "individuals are frequently more similar to members of other populations than to members of their own." These statements occur repeatedly in the current secondary sources because they are both based on numerous replicated primary research studies and accord with the theoretical framework of human genetics that has developed based on data gathered in the last decade. The writings of Guido Barbujani (like Edwards, a collaborator with Luigi Luca Cavalli-Sforza) are quite helpful in summing up what the data actually show, as Barbujani is a biologist. He notes in several of his recent writings that the first few full human genome sequences of individuals included the sequence of Seong-Jin Kim, a Korean scientist, who shares more SNP (single-nucleotide polymorphism commonalities with Craig Venter than James Watson does, and similarly shares more SNP commonalities with Watson than Venter does. Look up his articles or the primary research article by Ahn et al that Barbujani cites, and wrap your mind around what the data show. -- WeijiBaikeBianji (talk, how I edit) 17:27, 4 December 2013 (UTC)[reply]
You're correct on point #1 and very misleading on point #2. To the extent that #2 is correct, it is only when just a few loci is sampled. Even still, individuals are much more likely to be more similar to members of their own population than other populations. This is the case even when just a few loci is sampled. But if enough loci is sampled, the probability that an individual will be more similar to members outside their population group than inside it, is zero.

"How often is a pair of individuals from one population genetically more dissimilar than two individuals chosen from two different populations?” depends on the number of polymorphisms used to define that dissimilarity and the populations being compared. The answer, Formula can be read from Figure 2. Given 10 loci, three distinct populations, and the full spectrum of polymorphisms (Figure 2E), the answer is Formula ≅ 0.3, or nearly one-third of the time. With 100 loci, the answer is ∼20% of the time and even using 1000 loci, Formula ≅ 10%. However, if genetic similarity is measured over many thousands of loci, the answer becomes “never” when individuals are sampled from geographically separated populations."

Witherspoon, D. J., Wooding, S., Rogers, A. R., Marchani, E. E., Watkins, W. S., Batzer, M. A., & Jorde, L. B. (2007). Genetic similarities within and between human populations. Genetics, 176(1), 351-359.

The prevailing/majority position in the field of biology is that race is both biological and social. This is what the sources from "Nature" journals show which for unknown reasons you declined to use. There are many more high quality sources I can provide that will argue this same point. There are of course certainly biologists that argue that race is entirely social and I never said there wasn't. But I don't understand the constant need to pretend the other side doesn't exist. Particularly when the other side happens to be the majority. BlackHades (talk) 19:49, 4 December 2013 (UTC)[reply]
Come on, BlackHades, at least read the article closely enough to understand the structure of the argument in the Witherspoon paper. (The authors of the secondary sources I have cited all managed to do that.) From farther along in the paper than the cherry-picked quotation you found, which gives an intermediate step in the argument if certain real-world considerations are ignored, comes the authors' bottom line: "The fact that, given enough genetic data, individuals can be correctly assigned to their populations of origin is compatible with the observation that most human genetic variation is found within populations, not between them. It is also compatible with our finding that, even when the most distinct populations are considered and hundreds of loci are used, individuals are frequently more similar to members of other populations than to members of their own population. Thus, caution should be used when using geographic or genetic ancestry to make inferences about individual phenotypes." And that's what all the current sources say, because this has been empirically demonstrated on repeated occasions by now. To edit Wikipedia articles to an acceptable standard of verifiability, we have to read sources more closely and make sure to identify good sources in the first place. -- WeijiBaikeBianji (talk, how I edit) 21:05, 4 December 2013 (UTC)[reply]
I never said or implied that the fact that individuals can be correctly assigned if enough genetic data is used, is in any way, not compatible with human genetic variation being mostly within population. (I clearly stated point #1 is correct) Nor did I ever argue or imply that samples of hundreds of loci equates to never being more dissimilar to a member within population than outside population. (100 loci sample equates to being more dissimilar to a member within population than outside population 20% of the time) The line "Thus, caution should be used when using geographic or genetic ancestry to make inferences about individual phenotypes" I absolutely fully agree with this line.
Nothing you wrote acknowledges the point I was making. When several thousands of loci is sampled, the probability that an individual will be more dissimilar to a member within population than a member outside of it is 0%. Meaning if one were to compare the entire genome of an individual and compare it to the entire genome of someone within population and someone outside the population, the individual will 100% of the time be more similar to the individual within population. This is why your point #2 is very misleading and incomplete. When small amounts of loci is sampled, your point #2 can be true in these scenarios, however it is never true if you were to compare entire genomes of individuals. This is what I was trying to explain to you and you haven't provided any reliable source that would refute this specific point. BlackHades (talk) 22:35, 4 December 2013 (UTC)[reply]
Figure 2. The intersection of single-nucleotide polymorphism alleles in the genomes of Craig Venter, James Watson, and Seong-Jin Kim.[1]
BlackHades, you are making broad generalizations that are not supported by the sources. The claim of 100% classification accuracy is only possible, as stated in Witherspoon, "when individuals are sampled from geographically separated populations." Witherspoon also disposes of the "theoretical illustrations" used by Risch et al. (2002) and Edwards (2003) that suggested "if enough loci are considered, two individuals from the same population may be genetically more similar (i.e., more closely related) to each other than to any individual from another population" and lead to Risch's claim that "two Caucasians are more similar to each other genetically than a Caucasian and an Asian." A claim shown to be incorrect by Bamshad et al. (2004 ).
The sources provide above by WeijiBaikeBianji do refute your claim that "if one were to compare the entire genome of an individual and compare it to the entire genome of someone within population and someone outside the population, the individual will 100% of the time be more similar to the individual within population." For example;

Today, developments in DNA sequencing technology allow us to compare completely sequenced genomes. Ahn et al. (54) observed that two US scientists of European origin, namely James Watson (11) and Craig Venter (2), share fewer SNPs (461,000) than either of them shares with a Korean scientist, Seong-Jin Kim (569,000 and 481,000, respectively)(Figure 2). Of course, this does not mean that, on average, people of European origin are genetically closer to Asians than to other Europeans. However, it does show that patterns of genetic resemblance are far more complicated than any scheme of racial classification can account for. On the basis of the subjects' physical aspect, a physician would consider Venter's DNA, and not Kim's, a better approximation to Watson's DNA. Despite ideological statements to the contrary (55, 56) racial labels are positively misleading in medicine, and wherever one is to infer individual genome characteristics.
Barbujani, G.; Ghirotto, S.; Tassi, F. (2013). "Nine things to remember about human genome diversity". Tissue Antigens. 82 (3): 155–164. doi:10.1111/tan.12165. PMID 24032721.
(cited "ideological statements to the contrary" can be found in (55) Burchard et al. (2003) and (56) Risch et al. (2002))

See also: Barbujani, G.; Colonna, V. (2010). "Human genome diversity: Frequently asked questions" (PDF). Trends in Genetics. 26 (7): 285–295. doi:10.1016/j.tig.2010.04.002. PMID 20471132.
Very few sources, if any (that satisfy WP:RS), refute the fact that "individuals are frequently more similar to members of other populations than to members of their own", find it "very misleading". In fact the vast majority of high quality reliable sources don't find any genetic basis for racial classifications at all. — ArtifexMayhem (talk) 10:59, 5 December 2013 (UTC)[reply]
The Venter genome and especially the Watson genome were sequenced using early technology that was extremely error-prone. I've heard that parts of the published Watson genome appear positively non-human due to these errors. Tal[2] wrote:
The study of Ahn et al. (2009) suggests that the pairwise distances among three individuals, a Korean (“SJK”), Craig Venter and James Watson, measured by multilocus ASD, are roughly similar despite the distinct geographical origin of SJK in relation to Venter and Watson (see also their Fig. 2E). These results are surprising in light of our model for �n, which predicts that for worldwide distant populations (FST > 0.13) the probability for such an occurrence is virtually zero given as little as 200 independent and informative SNPs (Appendix F, Fig. F.1). In fact, with roughly 3.5 million SNPs sequenced in each individual genome, the pairwise distances Venter–Watson and Venter–SJK (or Watson–SJK) must show substantial discrepancy, since the ratio of average pairwise distances RAD is above 1.3 already at FST = 0.10 (see Fig. 5A). The paradoxical result is most likely an artifact of the high error rate and low coverage in Watson’s SNP calling (Yngvadottir et al., 2009). (emphasis added)
--Victor Chmara (talk) 11:36, 5 December 2013 (UTC)[reply]

Through a variety of search tools available through library databases, I found a number of books and articles that cite both the Lewontin paper and the Edwards paper, bringing up to date the scholarly response to Edwards's commentary article on Lewontin's much-cited paper. I invite all other Wikipedians here to mention, for completeness, any reliable secondary sources on human genetics that you know about that specifically discuss the claims of Lewontin's original 1972 paper and contrast those with the claims of Edwards's 2003 commentary article, in light of current human genetic research. Those would be on-point references for the much overdo top-to-bottom revision of this article. Your kind suggestions of sources (or your thoughtful examination of the sources already mentioned here in this talk page section) would be quite helpful for collaborative editing of this Wikipedia article. Please let us know. -- WeijiBaikeBianji (talk, how I edit) 14:26, 19 September 2014 (UTC)[reply]

I found a number of books and articles that cite both the Lewontin paper and the Edwards paper and mentioned those with complete citations in this talk page section. I see, reviewing the article today, that by contrast many of the references cited in this Wikipedia article do not respond specifically to the Lewontin 1972 paper or to the Edwards 2003 commentary on the Lewontin paper, but are about broader issues for which there are other Wikipedia articles. In the spirit of a kind editor suggestion in another talk page section here, I propose rewriting this Wikipedia article from top to bottom with careful attention to the reliable, secondary sources that actually discuss the Lewontin paper in the context of also discussing the Edwards 2003 paper (because the Edwards 2003 paper, as such, is the topic of this article). That will mean chopping out a lot of what is now in the article, and also adding in references and the statements those references contain that have never been in this article. With this project in mind, I again ask editors watching this page to recommend reliable, secondary sources that specifically address the Edwards response to Lewontin and the responses of other scholars, in turn, to Edwards. -- WeijiBaikeBianji (talk, how I edit) 17:28, 23 September 2014 (UTC)[reply]

New review articles on human population genetics based on studies of ancient DNA

[edit]

Edwards made some predictions, based on a mathematical model rather than empirical data, in his commentary on Lewontin's article. A lot of science has happened meanwhile.

Two review articles in prominent journals about human population genetics are bringing together analysis of the many recent studies of human DNA, including DNA from ancient individuals.

  • Pickrell, Joseph K.; Reich, David (September 2014). "Toward a new history and geography of human genes informed by ancient DNA". Trends in Genetics. 30 (9): 377–389, 378. doi:10.1016/j.tig.2014.07.007. PMC 4163019. PMID 25168683. Retrieved 16 September 2014. However, the data also often contradict models of population replacement: when two distinct population groups come together during demographic expansions the result is often genetic admixture rather than complete replacement. This suggests that new types of models – with admixture at their center – are necessary for describing key aspects of human history ([14–16] for early examples of admixture models). {{cite journal}}: Invalid |ref=harv (help)

Earlier studies of this issue were based on more limited samples (fewer genes, and fewer human individuals from fewer regions and only recent times). As more samples of more genes from more individuals from more places and times are gathered, the molecular evidence is making it increasingly clear that human beings have been moving back and forth across the Earth's surface and mixing genes over long distances ever since their earliest ancestors moved out of the human homeland in Africa. The picture that emerges backs up the raw data that Edwards found in his field studies, but is also supportive of the overall conclusion reached by Lewontin. -- WeijiBaikeBianji (talk, how I edit) 16:46, 18 September 2014 (UTC)[reply]

Do you have a source which explicitly links these studies with the Lewontin/Edwards controversy? AndyTheGrump (talk) 16:55, 18 September 2014 (UTC)[reply]
I think some of the sources cited above in an earlier section on this talk page (found by bibliographic work looking for sources that specifically mentioned both Edwards and Lewontin) are more directly on point for editing this particular article, and I urge any interested person to consult them for that purpose. The Hawks review article is interesting to me because many people think that Hawks has a different point of view from the view he expresses in that article. The Pickrell Reich review article (it is a secondary source itself, as the Hawks review article is, ready to use for editing Wikipedia articles) is very newly published (just this week). It may be helpful to editors to show how further research has put both Edwards's ideas and Lewontin's to the test of more deeply looking at the data. -- WeijiBaikeBianji (talk, how I edit) 18:41, 18 September 2014 (UTC)[reply]
That would constitute original research. AndyTheGrump (talk) 18:45, 18 September 2014 (UTC)[reply]
I actually agree with you, as a matter of directly inserting references from these two new articles into the Wikipedia article text of the article here, but that kind of close reading of the WP:OR content guideline suggests just what kind of edits I would like to see less of on the article SAT (for example), which has a lot of edits not drawn from on-point reliable, secondary sources. That's the main reason these two articles are mentioned on the talk page here (and not in the article text): I mention these articles here precisely because some editors of this Wikipedia article have wondered what the scientific consensus is about the latest molecular genetic evidence, and I think the two review articles show, by what they jointly agree about, what that consensus is, and reduce arguments about direct edits to this Wikipedia article from the most on-point sources. Those directly on-point reliable secondary sources, cited above in a previous talk page section, are not ideological, but just aware of the newest developments in the research on the molecular genetics of humankind. -- WeijiBaikeBianji (talk, how I edit) 19:19, 18 September 2014 (UTC)[reply]
Article talk pages are intended solely for discussions of direct relevance to article content - they are not a forum for general discussions. AndyTheGrump (talk) 19:34, 18 September 2014 (UTC)[reply]

The studies WBB cites have no implications for Lewontin's fallacy. Edwards's argument has been empirically confirmed in many studies, e.g., [3]. See also this article.--Victor Chmara (talk) 09:13, 19 September 2014 (UTC)[reply]

Victor, the first study you kindly cite actually mentions a data-based counterexample to Edwards's predicted model: "RISCH et al. (2002) and EDWARDS (2003) used theoretical illustrations to show why accurate classification is possible despite the slight differences in allele frequencies between populations. These illustrations suggest that, if enough loci are considered, two individuals from the same population may be genetically more similar (i.e., more closely related) to each other than to any individual from another population (as foreshadowed by POWELL and TAYLOR 1978). Accordingly, RISCH et al. (2002, p. 2007.5) state that 'two Caucasians are more similar to each other genetically than a Caucasian and an Asian.' However, in a reanalysis of data from 377 microsatellite loci typed in 1056 individuals, Europeans proved to be more similar to Asians than to other Europeans 38% of the time (BAMSHAD et al. 2004; population definitions and data from ROSENBERG et al. 2002)." Tal's paper is about a mathematical model rather than about empirical data (as is the commentary by Edwards that this article is about, for that matter). I'll depart from this section of the talk page to go back to the section of the talk page that is specifically about follow-up secondary sources to call for more of those. -- WeijiBaikeBianji (talk, how I edit) 14:18, 19 September 2014 (UTC)[reply]
Please read the Witherspoon study. It shows that studies finding that members of different races are sometimes more similar to each other than to members of their own groups suffer from poor resolution. If thousands (rather than dozens or hundreds) of genetic markers are used, there is no such overlap between geographically separated populations. In fact, Tal shows that if enough markers are used, there is no overlap between geographically close populations, either.--Victor Chmara (talk) 08:35, 22 September 2014 (UTC)[reply]

Mistake in text

[edit]

The text says "Philosophers Jonathan Kaplan and Rasmus Winther have argued that while Edwards's argument is correct it does not invalidate Lewontin's original argument, because racial groups being genetically distinct on average does not mean that racial groups are the most basic biological divisions of the world's population." Lewontin's original argument was that lower variation between groups meant race was of no taxonomic significance. This argument is indeed completely invalidated, and any other point is a different argument. Zhang500 (talk) 02:55, 9 July 2015 (UTC)[reply]

Do you have Lewontin's paper at hand as you bring this up? What was Lewontin's statement then, in his own words? Have you checked the more recent paper by Lewontin and Marcus Feldman (I've seen it as a book chapter, cited on this talk page, as I recall) in which Lewontin updates his comments on the issue? What did he say in that year? -- WeijiBaikeBianji (talk, how I edit) 11:58, 9 July 2015 (UTC)[reply]
Here it is.[4] He relies entirely on a single broad measure of diversity across the whole human subspecies, and since the between race component is "small", he claims race is of no taxonomic significance. That's his argument - there is nothing about any more complex structuring such as "subsets" or the like. Can you link to the Feldman paper? Also do you have access to Kaplan and Winther? How do they define "basic biological division"? Zhang500 (talk) 03:57, 10 July 2015 (UTC)[reply]

Copying discussion from user talk page

[edit]

From User:ArtifexMayhem's talk page.[5]

{start copy}

Talk:Human Genetic Diversity: Lewontin's Fallacy Could you explain why you re-inserted demonstrably false material? Reverts must be supported with reasoning. Zhang500 (talk) 11:14, 18 July 2015 (UTC)[reply]

The burden is really on you, Zhang500, to explain why you think the reliably sourced material you are deleting from the stable version of the article is "demonstrably false". In other words, I think the edit by ArtifexMayhem reflects the best current, published, reliable sources. -- WeijiBaikeBianji (talk, how I edit) 11:53, 18 July 2015 (UTC)[reply]
What, exactly, is demonstrably false in said material? — ArtifexMayhem (talk) 15:05, 18 July 2015 (UTC)[reply]

{end copy}

Lewontin offers no argument for the taxonomic insignificance of races other than low Fst. Since low Fst is shown to not in itself invalidate a taxonomy, Lewontin is incorrect in his argument. Therefore it is impossible that "while Edwards's argument is correct it does not invalidate Lewontin's original argument" or that "Edwards and Lewontin are therefore both correct". This was explained in the section OP. Zhang500 (talk) 05:38, 19 July 2015 (UTC)[reply]
Your line of reasoning is not supported by the sources cited for the material you removed. Please support your suggested edit with reliable sources. — ArtifexMayhem (talk) 09:40, 19 July 2015 (UTC)[reply]
Well of course it isn't. That's because the sources are transparently false. How can they then be "reliable sources"? Zhang500 00:07, 20 July 2015 (UTC)
Without, reliable sources there is nothing to discuss. — ArtifexMayhem (talk) 01:48, 20 July 2015 (UTC)[reply]
Sources which are transparently false are clearly not reliable sources. Zhang500 03:23, 20 July 2015 (UTC)
Who has decided that the sources are 'transparently false'? AndyTheGrump (talk) 03:27, 20 July 2015 (UTC)[reply]
Don't you see that they are false? Zhang500 08:29, 20 July 2015 (UTC) — Preceding unsigned comment added by Zhang500 (talkcontribs)
[edit]

Hello fellow Wikipedians,

I have just modified one external link on Human Genetic Diversity: Lewontin's Fallacy. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

  • If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
  • If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 12:40, 8 November 2017 (UTC)[reply]

non-biologist critics

[edit]

This article gives undeserved weight to critics who aren't even biologists, let alone experts in the field.Jwray (talk) 09:12, 3 February 2018 (UTC)[reply]

Yeah, that would make sense to me considering that this is strictly a biological concept. Take for example species classification; sure there are arbitrary definitions of species. The difference between a wolf and a coyote is arbitrary. But this is not how one makes an argument, you see? One does not attack the definition and how it is arbitrary, otherwise words will lose all meaning. To bring my point home further, the FST (fixation index) between a European and sub-Saharan African are the exact same (1.53) 2600:4040:5584:1A00:49CE:2B4B:810E:EAB7 (talk) 01:54, 23 September 2022 (UTC)[reply]

Disputed edit

[edit]

There's nothing massively wrong with this edit, but I do think it's a bit of a WP:COATRACK (introducing content not directly pertinent to the present article). It's also just less clear for the reader than the previous summary was. Wikipedia is meant to be primarily composed of WP:SUMMARY of the relevant literature. I urge IP 2001:861:4B40:C8F0... to engage here rather than reverting to restore their preferred content. Generalrelative (talk) 16:40, 31 May 2023 (UTC)[reply]