Joined: 07 Feb 2007 {Posts: 1829 } Location: Lookin DC Metro, Feelin Geneva
Posted: Sat 14 Feb 2009 02:44 Post subject: Question About Race
Frank said during my last suspension that I could ask questions as to specific peer reviewed articles and studies that mention "race" and an experts on this site would teach me.
Quote:
Nature Genetics 36, S28 - S33 (2004)
Published online: ; | doi:10.1038/ng1435
Genetic variation, classification and 'race'
Lynn B Jorde & Stephen P Wooding
Department of Human Genetics, University of Utah School of Medicine, Salt Lake City, Utah 84112, USA.
Correspondence should be addressed to Lynn B Jorde lbj@genetics.utah.edu
New genetic data has enabled scientists to re-examine the relationship between human genetic variation and 'race'. We review the results of genetic analyses that show that human genetic variation is geographically structured, in accord with historical patterns of gene flow and genetic drift. Analysis of many loci now yields reasonably accurate estimates of genetic similarity among individuals, rather than populations. Clustering of individuals is correlated with geographic origin or ancestry. These clusters are also correlated with some traditional concepts of race, but the correlations are imperfect because genetic variation tends to be distributed in a continuous, overlapping fashion among populations. Therefore, ancestry, or even race, may in some cases prove useful in the biomedical setting, but direct assessment of disease-related genetic variation will ultimately yield more accurate and beneficial information.
I bolded a section because I made very similar statements based on a New York Times article in the past and was suspended partially due to them.
I would like someone who Frank considers an 'expert on this site' to explain why this statement is false based on the data in the study.
As the rules and explanations state, a science is a process and scientist pay little attention to conclusions but look at the validity of the data and how it was accessed.
Fair enough.
Why are these people wrong based on the data they have presented.
Posted: Sat 14 Feb 2009 04:14 Post subject: Re: Question About Race
Dragon Horse wrote:
I would like someone who Frank considers an 'expert on this site' to explain why this statement is false...
Explain why which statement is false?
Do you mean the statement: "Clustering of individuals is correlated with geographic origin or ancestry. These clusters are also correlated with some traditional concepts of race, but the correlations are imperfect because genetic variation tends to be distributed in a continuous, overlapping fashion among populations."?
If so, who said it was false? If not, what statement are you referring to?
Dragon Horse wrote:
Why are these people wrong based on the data they have presented.
What people? Do you mean Jorde & Wooding in 2004? If so, who said that they were wrong? If not, who are you referring to?
Joined: 07 Feb 2007 {Posts: 1829 } Location: Lookin DC Metro, Feelin Geneva
Posted: Sat 14 Feb 2009 20:42 Post subject: Re: Question About Race
fwsweet wrote:
Dragon Horse wrote:
I would like someone who Frank considers an 'expert on this site' to explain why this statement is false...
Explain why which statement is false?
Do you mean the statement: "Clustering of individuals is correlated with geographic origin or ancestry. These clusters are also correlated with some traditional concepts of race, but the correlations are imperfect because genetic variation tends to be distributed in a continuous, overlapping fashion among populations."?
Yes this statement which came from Jorde & Wooding (2004).
If so, who said it was false? If not, what statement are you referring to?
Dragon Horse wrote:
Why are these people wrong based on the data they have presented.
What people? Do you mean Jorde & Wooding in 2004? If so, who said that they were wrong? If not, who are you referring to?
So for the sake of this conversation I will use "race" as a substitute for "continental based population clusters" found in this article since they heavily overlap with the notion of race we currently use in America.
Most whites in America have ancestry predominately from Europe or a border region of Europe.
Most "Asians" have predominant ancestral lines to the Asian continent.
Most "blacks" are predominately of subSaharan African ancestry...although due to social convention (one drop rule) there are those who are predominately or intermediate admixture of European ancestry who are "black" (African American) in the United States.
SNIP
The NY Times and the scientist in the article seem to support this notion that humans do have genetically based population clusters and that is (not perfectly) but fairly interchangeable with population notions of race, as we know it in the U.S. in 2008.
Your response was:
Quote:
They do not. The only line in the article that suggests such a bizarre thing was, "But some say the genetic clustering into continent-based groups does correspond roughly to the popular conception of racial groups." This statement (given away by the weasel words "some say") was obviously inserted by the ignorant newspaper reporter. Scientific writing does not use weasel words.
If DH wants to ask this forum's genetics experts for an analysis of the ignorance displayed in the grotesque NY Times article, he should post his request as a new thread.
You were referring to the NY Times article and the chart in the article I referenced.
They said the same basic thing that the Lynn B Jorde & Stephen P Wooding (2004) said above.
I also said in the thread:
Quote:
When I said race, I was not using the term as in "subspecies", this does not exist in any human population..
You said this was incorrect, right?
Why? It is not just a NY Times article with some opinions and interviews with researchers, this study came to the same conclusion as the writer in the NY Times article...(although I can't say the NY Times writer used the same methodology to arrive at that conclusion).
So why is the line of thinking in the study above (and the NY Times article wrong)?
This is the main question.
A subquestion I already asked in the other thread, but I will ask it here since it is more on topic.
The rule say that:
Quote:
3.3.1 race (in biology) or bio-race — A sub-division of a species that is identifiable by a cluster of traits that vary together geographically. For example, the key lime and the Mexican lime are two races of Citrus aurantifolia that differ in several traits including peel thickness and color. Synonyms are: subspecies, variety, breed. No cluster of geographically co-varying traits (bio-race) has ever been found in Homo sapiens.
Do you consider (meaning this site) lactose tolerance, skin color, immuno response genes, brain genes, etc. as co-varying traits that cluster fairly strongly with different population groups (that tend to have different phenotypes)? (as referenced in the NY Times article and the article I posted here:
If you do, then how many co-varying traits make a "breed, bio-race, subspecies"?
Actually before we define subspecies, what is a species? Based on my reading in biology a species is a vague concept that has no firm agreement, this has been called the "species debate" or "species problem".
Usually it is defined as "A species is often defined as a group of organisms capable of interbreeding and producing fertile offspring. "
Still there are many "species" that can interbreed and produce offspring, which is the source of the problem...
It is similar to what is a "language" in linguistics...
I guess to make it short. How do you define species. Then, how does this site define subspecies, breed, or bio-race specifically if you accept there are co-varying traits that cluster with different geographic populations? Do you require perfect distinct populations with no genetic overlap in thousands of years? Groups that don't interbreed normally but can? (lions and tigers), (dogs and wolves) or groups that can freely interbreed but have vary different phenotypes due to history (dog breeds)...
All these things interconnect and are important because I could imagine an experiment where one could argue that based on one allele like Lactose tolerance in East Africans, if you took a random population of people from Kenya, Somalia, Gaulic French, Irish, Japanese, Chinese, and Indians...
You could predict who was a Sub-Sahara African with a good about of certainty.
You could do the same with that Chinese brain gene that was mentioned in the NY Times article that is present in most Chinese (I'm going to guess a significant portion of neighboring East Asians) but very rare to absent in most European and African populations.
Etc.
This is using one gene each.
If I use 20 genes that heavily cluster with one population over another (say Europeans) I could identify Europeans most of the time.
Now this will break down in places like Latin America or even with African Americans...other "mixed populations"...likely also break down someone in Central Asia...although you might be able to identify the person as "non-subSahara African" even if you could not identify them as "white or Asian"...
If you use enough alleles you could easily say (as you often do Frank) that every person is a race.
If you use a rare enough gene you could say a "village is a race" or a "family" is a race.
The point is...is this just a question of "resolution".
If you use the right "resolution" you could approximate common notions of "race" (keeping in mind 'black' in America has a "one drop rule" and Hispanics are not considered a race but a cultural group of any "race" according to the U.S. census).
So yes the approximation of clusters would not be perfect but "predictive" with a reasonable amount of confidence...is this correct? If not why not?
There might be a odd Indian or European who would pop up.
I have transferred this topic from "Molecular Anthropology and Genetics" because it deals with the attempt to match up genetic diversity with ethnopolitical groups.
1. You said this [NY Times article] was incorrect, right? Why?
2. why is the line of thinking in the study above ... wrong? This is the main question.
3. Do you consider lactose tolerance, skin color, immuno response genes, brain genes, etc. as co-varying traits that cluster fairly strongly with different population groups (that tend to have different phenotypes)?
4. how many co-varying traits make a "breed, bio-race, subspecies"?
5. what is a species? Based on my reading in biology a species is a vague concept that has no firm agreement, this has been called the "species debate" or "species problem".
6. I guess to make it short. How do you define "species"?
7. Then, how does this site define subspecies, breed, or bio-race specifically if you accept there are co-varying traits that cluster with different geographic populations?
8. Do you require perfect distinct populations with no genetic overlap in thousands of years?
9. [Do you require] Groups that don't interbreed normally but can?
10. [Do you require] groups that can freely interbreed but have vary different phenotypes due to history (dog breeds)...
11. So yes the approximation of clusters would not be perfect but "predictive" with a reasonable amount of confidence...is this correct?
12. If not why not?
The above flurry of questions was response to my request for clarification as to who said that an opinion by Jorde and Wooding was wrong. The most charitable interpretation that I can put on this stream of questions (many of which are based on false assumptions) is that Dragon Horse has discovered a sudden personal interest in phylogeography but lacks formal training. (A less charitable interpretation is that he just likes argue.) Assuming the charitable interpretation, I suggest that Dragon Horse:
1. Carefully re-read pages 9-27 of L. L. Cavalli-Sforza, Genes, Peoples, and Languages, 1st ed. (New York: North Point Press, 2000), which he said he owns. Most of his questions regarding human variation are answered there. This popularization was written at a sixth-grade level and should be intelligible to any reasonably educated adult.
2. Better yet, if Dragon horse still has or can acquire a copy of the paperback abridged edition of Luigi Luca Cavalli-Sforza, Paolo Menozzi, and Alberto Piazza, The History and Geography of Human Genes, trans. Sarah Thorne (Princeton: Princeton University, 1994), which he said he was required to study in college, he should carefully re-read pages 16-20, especially page 19. This will answer most of his questions in even greater detail.
3. As he indicated in another post, Dragon Horse may have difficulty understanding some of the advanced material above. So I would urge him first to become familiar with the basics of physical anthropology by self-study. I happen to like Robert Jurmain and others, Introduction to Physical Anthropology (Belmont CA: Wadsworth, 2000), but any current college-level textbook will do. Again, I said "college-level textbook," not popularization. Go to a college bookstore and buy whatever is assigned for physical anthropology 101.
4. Best of all would be to sign up for an introductory course in physical anthropology at a local college. Believe me, a teacher's help would greatly accelerate the process of acquiring the basic skills needed to grasp the concepts essential to studying geographic variation in our species.
5. Finally, I am willing to help Dragon Horse learn the basics of human geographic variation, but he must limit himself to one question per day. (Not twelve in a single post!) And each question must be in reference clarifying something he read in one of the books mentioned above or in another college-level textbook. (The reason for this restriction is that Dragon horse will draw little useful knowledge out of technical research reports until he develops a better feel for clarity of definition, falsifiability, replicability, and drawing conclusions from findings.)
6. Some of Dragon Horse's questions (e.g.: "What is a species?") are best answered at the high school level, since (Sonia may correct me here if I am wrong) most college teachers assume that students are already familiar with the concept of "species," I suggest that Dragon Horse seek this knowledge elsewhere.
Joined: 07 Feb 2007 {Posts: 1829 } Location: Lookin DC Metro, Feelin Geneva
Posted: Mon 16 Feb 2009 23:00 Post subject:
Quote:
1. Carefully re-read pages 9-27 of L. L. Cavalli-Sforza, Genes, Peoples, and Languages, 1st ed. (New York: North Point Press, 2000), which he said he owns. Most of his questions regarding human variation are answered there. This popularization was written at a sixth-grade level and should be intelligible to any reasonably educated adult.
From the book above (excuse any misspelling typos)...
I will start with pg. 21 and go from there, the stuff about common known variation and blood groups (before pg. 21) was interesting but I've read that many times before.
Pg. 21, starts the discussion on Genetic Distances between populations.
On pg. 22 (last two paragraphs)
Using the Fischer Formula (one of many formulas to calculate this), Cavalli-Sforza said he takes several genes and calculates the distance between population and then averages those distances to get a value of overall distance. He also found through the work of some other mathematicians that genetic distance tends to increase with geographic distance, but this is not a linear increase, but a curve and he produces charts of this on each continent and New Guinea (pg. 24) The only genes that seem to deviate are immunoglobulin.
Pg. 25, goes specifically into "race".
he states clearly that he can likely prove a statistically significant genetic difference between any two populations, even neighboring villages, and specifically names Pisa and Florence, Italy.
He then says:
"Classifying the world's population into several hundreds of thousands or a million difference races would, of course, be completely impractical" (first full paragraph, pg. 26).
Then he goes on to say you might be able to scientifically seperate people into "races" using a method used by Barbujani and Sokal (1990) however, "a 'genetic barrier' would naturally be chosen in an arbitrary manner." There is no math to explain their calculations in the text but it is basically saying that you could 'analyze of discontinuities in the surface of gene frequencies generated on a geographic map....looking for local increases in the rate of change of gene frequencies per unit of geographic distance...Obstacles to migration or marriage could create these local increases' (that is a slight paraphrase of paragraph 3 on pg 26).
He said the method is still useful for identifying the geographic location of genetic boundaries (end of pg. 26).
he goes on to say that this is still difficult because there has been no barriers that create enough lack of gene flow to truly isolate people, but for some island populations, which he says "could be classified as a race".
For a nation like the U.S. he says it can not.
On pg. 29, he says that Classification based on Continental Origin could furnish a first approximation of "racial divisions" but he says Asia, Africa, and the Americas are very heterogeneous and even the Europe (more homogeneous) you could still divide into several subdivisions.
He goes on, on pg. 29, to give some basic statistical examples of genetic distance between people and divergent geographic populations and how they are all quite more related than not related....for example, random individuals from two different places in the world the variation between them is only about 85% (last full paragraph pg.29).
On pg. 37, he does say that the easiest way to talk about population genetic distance is by continent since that usually yields the largest divergence and he gives a chart. He gets the continental data by averages results from 15 populations (3 per continent).
He goes on to explain things as to why distance of separate geographically or time-wise does not always reflect in "genetic distance.
He eliminated Western Eurasia (Middle East and Central Asia I'm guessing) and North Africa due to the fact they are historically "admixed" (his language, at the top of pg. 40). In the end he ignores almost 1/4 of the worlds population (pg 40) to get his results.
Starting, Pg. 49, "Genetic Variation Between Populations" is also a very interesting section.
I found it interesting that on pg. 52, the author says "The genetic distance from Europe is anomalously low. North Africa is populated with CAUCASOID people like Europeans."
pg. 89 also had a nice map of genetic distances between given regional groups (these are averages of various samples collected from these areas).
-----------------------------------------------------------------------------
Anyway...QUESTION
I have never maintained on this site or anywhere else there are "pure biological races or subspecies" of humans.
I agree with Cavalli-Sforza that the idea of "pure races" is nonsensical.
That being said, Cavalli-Sforza makes it clear that using various markers he can show genetic distance between different populations, he also talks about how some markers don't exist or exist in such low frequencies in some populations but are more frequent in others.
This is how DNA admixture (autosomal) DNA testing can predict "continental admixture", "predict" being a key word. There are still many markers that have not been identified and so there is often difficulty telling the difference. Some Europeans will show Native American ancestry, and telling the difference between Native American and "Asian" is very difficult.
Still if we can make rough approximations between different continental groupings and also do fairly predictive (although imperfect) predictions about a individuals % of continental ancestry I'm having trouble understanding Franks response to the NY Times article and why Lynn B Jorde & Stephen P Wooding (2004) (which comes to a very similar if not identical conclusion about "social race" correlating loosely with continental clustering is wrong.
Well Frank said the NY Times had choose words from the NY Times article but did not comment on Jorde and Wooding (2004) at all.
Saying "clustering correlates" is not the same as saying "biological race exists" that might be too much of a nuanced argument for some, but I don't see a conflict because one can say that there are no pure racial groups and gene flow between all humans has always occurred but due to the fact it has not been linear (as Cavalli-Sforza stated).
Cavalli-Sforza also stated that due to various reasons, bottlenecks, selective pressures, and natural genetic drift populations vary and the further they are apart, the more they tend to vary, continents usually showing the greatest variation (once again this is quite small all things considered but enough to chart a difference).
If one can identify these markers as being common in people we call "whites" but absent in people we call 'East Asian" or "SubSahara Africa" or so low as to be statistically insignificant than one might not be able to identify a Chinese person 100%, but you have a very high chance of identifying someone as "East Asian'.
Cavalli-Sforza spoke about African Americans in some detail on pg. 74.
He acknowledged the "one drop rule". It is obvious the idea of "pure race" for Hispanics and African Americans do not exist and even separating out who is African American or Hispanic by genetic test would be very difficult (I would say fairly impossible for Hispanics, and even the U.S. census does not recognize them as a "race", but a cultural grouping).
Still if someone had 80% SubSaharan African admixture or even 60% you would probably feel certain they are "black" if they are in American, but that would not be true in Brazil. This is the result of the one drop rule. If they are 50, 30, 15% well there you get to the point where just due to blood quantum the chance that their phenotype will be in the range of a "black African" is low so the less likely they will be considered or identify as 'black' in America (but you can't be sure, there some African Americans who likely have very low admixture with subSahara Africa so the prediction rate breaks down at this point).
I'm not sure what is wrong with these statements and despite asking a few times I have got no clear answer from any of the "teachers" on the board but "read this". Okay, I read the selection and even more, am I missing something between Cavalli-Sforza, the NY Times Article, and Jorde and Wooding (2004) because they seem to be saying very similar things. What is wrong with my conclusion?
Please use an example, you can even use mathematics equations, I've taken math up to Cal I as well as grad level stats, I think I can understand any formula put out.
The only question mark in all the above follows the phrase, "What is wrong with my conclusion? By "my conclusion," you apparently mean that Cavalli-Sforza, the NY Times Article, and Jorde and Wooding "seem to be saying very similar things."
Is that your question? Are you asking what is wrong with your claim that Cavalli-Sforza, the NY Times Article, and Jorde and Wooding "seem to be saying very similar things"? (That genetic distances among continents match up with U.S. ethnopolitical groups.)
Joined: 07 Feb 2007 {Posts: 1829 } Location: Lookin DC Metro, Feelin Geneva
Posted: Tue 17 Feb 2009 13:11 Post subject:
fwsweet wrote:
The only question mark in all the above follows the phrase, "What is wrong with my conclusion? By "my conclusion," you apparently mean that Cavalli-Sforza, the NY Times Article, and Jorde and Wooding "seem to be saying very similar things."
Is that your question? Are you asking what is wrong with your claim that Cavalli-Sforza, the NY Times Article, and Jorde and Wooding "seem to be saying very similar things"? (That genetic distances among continents match up with U.S. ethnopolitical groups.)
Yes.
I outlined what I read so you had an idea of where I was looking in the book, also I want to avoid someone assuming what I mean or the underlying reasoning, which is a common thing it seems.
Actually "match up" in this situation would imply that we are talking about something discreet...I would prefer the word "correlate".
Question: What is wrong with saying that L. L. Cavalli-Sforza, Genes, Peoples, and Languages (New York: North Point Press, 2000), a NY Times Article, and Jorde and Wooding "seem to be saying very similar things"? (That genetic distances among continents correlates with U.S. ethnopolitical groups.)
Let us consider each item individually, and then compare and contrast them.
The NY Times article makes no mention of the United States at all, much less of U.S. ethnopolitical groups. Only two sentences come close. One sentence by Wade reads, "But some say the genetic clustering into continent-based groups does correspond roughly to the popular conception of racial groups." The other, a quotation attributed to Jonathan Pritchard reads, "There are difficulties in where you put boundaries on the globe, but we know now there are enough genetic differences between people from different parts of the world that you can classify people in groups that correspond to popular notions of race." Let us apply the four tests of scientific thinking to those sentences: clarity of definition, falsifiability, replicability, and drawing conclusions from findings.
Both sentences fail the first test. What could possibly be meant by "popular conception of racial groups" or "popular notions of race"? European professional anthropology and sociology journals today routinely refer to "the Polish race," "the Italian race," "the "German race." Latin Americans refer to the "Latino race" and the "Anglo race". Orthodox Jews refer to "the Jewish race" and "the Gentile race". Surely, these cannot be what was meant.
U.S. Popular culture lacks consensus as to whether Turks, East Indians, or Bengalis are White. But it does consider Australian Aborigines, Papua-New Guineans and Andaman Islanders to be Black. Yet these folkloric beliefs are contradicted by the only list of genetic clusters published in the article, which separates Australians from Africans and lists people of the Indian subcontinent as "Caucasians."
U.S. federal regulations are even less consistent; the EEOC insists that Hispanics are a separate and distinct "race" while the Census Bureau insists that they are not.
Please do not misunderstand. My point is NOT that the correlation between genetic groups and U.S. ethnopolitical groups is weak. My point is that the article presents no definition of the terms "popular conception" or "popular notion," and that it is therefore hopeless to guess what those terms might possibly mean to the authors because they lack consensus. Had the article merely listed what it considered to be the "popular conception of racial groups" we could at least begin to consider its accuracy, but it does not. Again, I am not saying that the sentences are wrong. They are far, far worse than merely wrong—they are unintelligible.
Having failed the test of clarity of definition, it is pointless to pursue the other tests. Nevertheless, the article offers no explanation as to how to disprove its catalog of "popular conception of racial groups" because it provides no such catalog. Furthermore, the article produces no findings to be replicated as to precisely what are the "popular conception of racial groups." Nor does it base its conclusions about the "popular notions of race" on its findings as to just what those popular notions are (there are no findings).
In short, the two sentences fail the first smell-test (clarity of definition), rendering them meaningless.
The Jorde and Wooding article makes one mention of U.S. ethnopolitical groups in its introduction, one more in its findings, and three more in its conclusions.
In the introduction it says, "Forensic databases in the US are typically organized according to traditional racial and ethnic categories (e.g., African-American, European-American, Hispanic)." The parenthetical phrase provides a list of what they mean by "traditional racial and ethnic categories." In my view, the parenthetical phrase passes the first test (clarity of definition).
In the findings it reports that, "hypertensive African-Americans show less response, on average, than hypertensive European-Americans to angiotensin-converting enzyme (ACE) inhibitors." It then gives findings that this average difference is less than one-third the standard deviation within the groups. These findings cite replicated and peer-reviewed sources. In my view, these findings pass the second and third tests (falsifiability and replicability).
That same paragraph then concludes that, due to the three-times greater measured intra-group variation than inter-group variation, "many African-Americans would respond better to ACE inhibitors than would many European-Americans" and that "To conclude, on the basis of population averages, that ACE inhibitors are ineffective in African-Americans could deny many people a powerful and appropriate drug treatment." This conclusion—that basing therapy on ethnic self-identity comprises medical malpractice—is supported by the findings, thereby passing the fourth test.
The first mention of U.S. ethnopolitical categories in the article's conclusions says that ancestral factors "should not be confused ... with ... ethnicity or race." This conclusion—that ancestry is different from ethnicity or race and to confuse the two can also lead to malpractice—is also supported by the findings. (For example, the figure on page 529 shows far more genetic diversity within sub-Saharan Africa than across the rest of the planet combined, suggesting that there are a dozen or so distantly related groups (neighbor-joining genetic "races," if you wish) within sub-Saharan Africa and a single closely-related group (neighbor-joining genetic "race") comprising the entire rest of the globe.
The second conclusion mentioning U.S. ethnopolitical groups emphasizes even more strongly that to confuse ethnopolitical groups with genetic ancestry could be disastrous. "Ignorance of the shared nature of population variation can lead to diagnostic errors ... or to inappropriate treatment or drug prescription. The general public, including policy-makers are easily seduced by typological thinking and so they must be made aware of the genetic data that help to prove it wrong." This conclusion fails the fourth test. No findings were presented to show that confusing genetic admixture with ethnic self-identity in treating sickle-cell or cystic fibrosis (the examples given) results in malpractice. Obviously, the authors feel strongly about this, but no findings were cited.
The final mention of U.S. ethnopolitical groups is the single sentence, "Considering this extraordinary complexity, the idea that variation in the frequency of a single allele could explain substantial population differences in behavior would be amusing if it were not so dangerous." Given the findings cited, this sentence passes all four tests.
In short, with one minor exception, the article passes all four tests. It mentions U.S. ethnopolitical groups five times. At each mention, it vigorously attacks the notion that U.S. "ethnic or racial" groups correlate with genetic ancestry, and it concludes that anyone making such a correlation for medical reasons is engaged in serious malpractice.
L. L. Cavalli-Sforza, Genes, Peoples, and Languages (New York: North Point Press, 2000)
This book mentions the United States only twice. On page 28 it says that U.S. races "are extremely simplistic. The U.S. census recognizes Whites, Blacks, Native Americans, Asians, and Hispanics. This last category has almost no biological meaning." [The author errs, by the way. "Hispanic" is not a census "race."]
On pages 200-201 it attributes Americans' extreme religiosity to cultural founder effect.
Nowhere does this book correlate genetic distances among continents with U.S. ethnopolitical groups. Indeed, the author expends a great deal of effort attacking the notion that genetic distances among continents correlate with ethnopolitical groups anywhere.
And so, in conclusion
Two sentences of the newspaper article correlate genetic distances among continents with "popular notions of race" but the article fails say what the quoted phrase means, or even if it refers to the United States at all.
The serious journal article mentions U.S. ethnopolitical groups in five places. Each place attacks the notion that U.S. ethnopolitical groups correlate with genetic distances among continents.
The popularization by Cavalli-Sforza makes no mention of U.S. ethnopolitical groups in this context, but opposes the idea that genetic distances among continents correlate with ethnopolitical groups in general.
With these observations in mind, I would not agree that the three sources "seem to be saying very similar things." (That genetic distances among continents correlates with U.S. ethnopolitical groups.) Only two sentences in the newspaper article say something like this, although without mentioning the U.S., and they say it in a way that fails all four tests of scientific thinking (clarity of definition, falsifiability, replicability, and drawing conclusions from findings). The journal article attacks the notion. The book does not address U.S. groups at all.
Joined: 07 Feb 2007 {Posts: 1829 } Location: Lookin DC Metro, Feelin Geneva
Posted: Wed 18 Feb 2009 01:35 Post subject:
fwsweet wrote:
Question: What is wrong with saying that L. L. Cavalli-Sforza, Genes, Peoples, and Languages (New York: North Point Press, 2000), a NY Times Article, and Jorde and Wooding "seem to be saying very similar things"? (That genetic distances among continents correlates with U.S. ethnopolitical groups.)
Let us consider each item individually, and then compare and contrast them.
The NY Times article makes no mention of the United States at all, much less of U.S. ethnopolitical groups. Only two sentences come close. One sentence by Wade reads, "But some say the genetic clustering into continent-based groups does correspond roughly to the popular conception of racial groups." The other, a quotation attributed to Jonathan Pritchard reads, "There are difficulties in where you put boundaries on the globe, but we know now there are enough genetic differences between people from different parts of the world that you can classify people in groups that correspond to popular notions of race." Let us apply the four tests of scientific thinking to those sentences: clarity of definition, falsifiability, replicability, and drawing conclusions from findings.
Both sentences fail the first test. What could possibly be meant by "popular conception of racial groups" or "popular notions of race"? European professional anthropology and sociology journals today routinely refer to "the Polish race," "the Italian race," "the "German race." Latin Americans refer to the "Latino race" and the "Anglo race". Orthodox Jews refer to "the Jewish race" and "the Gentile race". Surely, these cannot be what was meant.
U.S. Popular culture lacks consensus as to whether Turks, East Indians, or Bengalis are White. But it does consider Australian Aborigines, Papua-New Guineans and Andaman Islanders to be Black. Yet these folkloric beliefs are contradicted by the only list of genetic clusters published in the article, which separates Australians from Africans and lists people of the Indian subcontinent as "Caucasians."
U.S. federal regulations are even less consistent; the EEOC insists that Hispanics are a separate and distinct "race" while the Census Bureau insists that they are not.
Please do not misunderstand. My point is NOT that the correlation between genetic groups and U.S. ethnopolitical groups is weak. My point is that the article presents no definition of the terms "popular conception" or "popular notion," and that it is therefore hopeless to guess what those terms might possibly mean to the authors because they lack consensus. Had the article merely listed what it considered to be the "popular conception of racial groups" we could at least begin to consider its accuracy, but it does not. Again, I am not saying that the sentences are wrong. They are far, far worse than merely wrong—they are unintelligible.
Having failed the test of clarity of definition, it is pointless to pursue the other tests. Nevertheless, the article offers no explanation as to how to disprove its catalog of "popular conception of racial groups" because it provides no such catalog. Furthermore, the article produces no findings to be replicated as to precisely what are the "popular conception of racial groups." Nor does it base its conclusions about the "popular notions of race" on its findings as to just what those popular notions are (there are no findings).
In short, the two sentences fail the first smell-test (clarity of definition), rendering them meaningless.
The Jorde and Wooding article makes one mention of U.S. ethnopolitical groups in its introduction, one more in its findings, and three more in its conclusions.
In the introduction it says, "Forensic databases in the US are typically organized according to traditional racial and ethnic categories (e.g., African-American, European-American, Hispanic)." The parenthetical phrase provides a list of what they mean by "traditional racial and ethnic categories." In my view, the parenthetical phrase passes the first test (clarity of definition).
In the findings it reports that, "hypertensive African-Americans show less response, on average, than hypertensive European-Americans to angiotensin-converting enzyme (ACE) inhibitors." It then gives findings that this average difference is less than one-third the standard deviation within the groups. These findings cite replicated and peer-reviewed sources. In my view, these findings pass the second and third tests (falsifiability and replicability).
That same paragraph then concludes that, due to the three-times greater measured intra-group variation than inter-group variation, "many African-Americans would respond better to ACE inhibitors than would many European-Americans" and that "To conclude, on the basis of population averages, that ACE inhibitors are ineffective in African-Americans could deny many people a powerful and appropriate drug treatment." This conclusion—that basing therapy on ethnic self-identity comprises medical malpractice—is supported by the findings, thereby passing the fourth test.
The first mention of U.S. ethnopolitical categories in the article's conclusions says that ancestral factors "should not be confused ... with ... ethnicity or race." This conclusion—that ancestry is different from ethnicity or race and to confuse the two can also lead to malpractice—is also supported by the findings. (For example, the figure on page 529 shows far more genetic diversity within sub-Saharan Africa than across the rest of the planet combined, suggesting that there are a dozen or so distantly related groups (neighbor-joining genetic "races," if you wish) within sub-Saharan Africa and a single closely-related group (neighbor-joining genetic "race") comprising the entire rest of the globe.
The second conclusion mentioning U.S. ethnopolitical groups emphasizes even more strongly that to confuse ethnopolitical groups with genetic ancestry could be disastrous. "Ignorance of the shared nature of population variation can lead to diagnostic errors ... or to inappropriate treatment or drug prescription. The general public, including policy-makers are easily seduced by typological thinking and so they must be made aware of the genetic data that help to prove it wrong." This conclusion fails the fourth test. No findings were presented to show that confusing genetic admixture with ethnic self-identity in treating sickle-cell or cystic fibrosis (the examples given) results in malpractice. Obviously, the authors feel strongly about this, but no findings were cited.
The final mention of U.S. ethnopolitical groups is the single sentence, "Considering this extraordinary complexity, the idea that variation in the frequency of a single allele could explain substantial population differences in behavior would be amusing if it were not so dangerous." Given the findings cited, this sentence passes all four tests.
In short, with one minor exception, the article passes all four tests. It mentions U.S. ethnopolitical groups five times. At each mention, it vigorously attacks the notion that U.S. "ethnic or racial" groups correlate with genetic ancestry, and it concludes that anyone making such a correlation for medical reasons is engaged in serious malpractice.
L. L. Cavalli-Sforza, Genes, Peoples, and Languages (New York: North Point Press, 2000)
This book mentions the United States only twice. On page 28 it says that U.S. races "are extremely simplistic. The U.S. census recognizes Whites, Blacks, Native Americans, Asians, and Hispanics. This last category has almost no biological meaning." [The author errs, by the way. "Hispanic" is not a census "race."]
On pages 200-201 it attributes Americans' extreme religiosity to cultural founder effect.
Nowhere does this book correlate genetic distances among continents with U.S. ethnopolitical groups. Indeed, the author expends a great deal of effort attacking the notion that genetic distances among continents correlate with ethnopolitical groups anywhere.
And so, in conclusion
Two sentences of the newspaper article correlate genetic distances among continents with "popular notions of race" but the article fails say what the quoted phrase means, or even if it refers to the United States at all.
The serious journal article mentions U.S. ethnopolitical groups in five places. Each place attacks the notion that U.S. ethnopolitical groups correlate with genetic distances among continents.
The popularization by Cavalli-Sforza makes no mention of U.S. ethnopolitical groups in this context, but opposes the idea that genetic distances among continents correlate with ethnopolitical groups in general.
With these observations in mind, I would not agree that the three sources "seem to be saying very similar things." (That genetic distances among continents correlates with U.S. ethnopolitical groups.) Only two sentences in the newspaper article say something like this, although without mentioning the U.S., and they say it in a way that fails all four tests of scientific thinking (clarity of definition, falsifiability, replicability, and drawing conclusions from findings). The journal article attacks the notion. The book does not address U.S. groups at all.
Frank that was excellent.
I think you misunderstood me before. I do not doubt your ability to speak on these issues. I am the type of person who can not accept someone's "word" on issues like this. I want to know "why". This is all I wanted and you took the time to do this, which I thank you for as I'm sure it took a bit of time to put it together.
Quote:
My point is that the article presents no definition of the terms "popular conception" or "popular notion," and that it is therefore hopeless to guess what those terms might possibly mean to the authors because they lack consensus. Had the article merely listed what it considered to be the "popular conception of racial groups" we could at least begin to consider its accuracy, but it does not.
You make a excellent point here. Let's expand this though. My wife is Japanese (not just ethnic Japanese but a national). Her concept of race is completely different from the American one. She considers Southeast Asians to be a completely different race from her and she is not sure that Chinese are the same race, although she says that Mongolians and Koreans are the same. Some Japanese would disagree with this assessment.
My point is exactly the point you make. There is no standard in the literature because their is not scientific process to replicate findings. It is simply random selection of boundaries based on sociological concepts, not biological ones. My wife would be 100% correct in saying that Thai people are more genetically distant (on average) from Japanese than Koreans and therefore Japanese and Koreans cluster. What does that mean? Not much of anything outside of that context. Someone else might draw a bigger cluster and say "well they are all the same race".
I guess this goes back to my point some weeks ago that biological race does not exist but people can use genetic science to at least approximate their racial ideology but drawing arbitrary boundaries, but as the study above pointed out this is seriously problematic even if there are correlations. As you correctly point out this is not really "science" because there is no standard definition, no replication across regions or international borders, etc. Still people present this as if it is scientific law. (sigh).
My next question is...
How can Jonathan Pritchard and those like him get away with these statements? When I say those like him, I would also add the people quoted in the other article I posted who specifically said "race" in regard to genetic data?
I would like to know though what by North Africans, having been for thousands of years, Caucasoid. I mentioned this above as it stood out to me and he never used "negroid" or any of the other classical racial physical anthropological categories (that I know are no longer in use by the vast majority of scientist).
I'm going to guess he was just trying to speak about a phenotypical range and did not have a better way to express it that would be easy to understand for the reader (as this is more a pop science book). That's my guess anyway.
It is somewhat disheartening that even trusted and internationally respected news sources like the NY Times write things like this so sloppily.
Hell, last year, not sure if you know, they even referenced "race realists" like Half Sigma and "Gene Expression" as "science blogs".
I wrote a letter of complaint to the editor siting various racist language on both blogs in regard to blacks and Hispanics and the fact that none of the people who write the blog are in any way scientist or have jobs related to any science, let alone "experts". Razib (Genex) is a web developer and Malloy an "artist".
This was in the NY Times!!! They never printed a retraction of anything but for getting Henry Louis Gates name wrong.
Last edited by Dragon Horse on Wed 18 Feb 2009 13:01; edited 1 time in total
Posted: Wed 18 Feb 2009 02:11 Post subject: New York Times
I would guess that The New York Times, as the semi-official newspaper of record, is trying to promote a certain narrative about "race." The official "racial" categories, the "one drop" rule, etc. are all part of their narrative. People who point out facts that contradict that narrative are never given a forum to present their views. The readers are supposed to believe, for example, that "everyone" accepts the "one drop" myth because they are never exposed to contrary views.
The ODR-loving Brent Staples and Henry Louis Gates, Jr. can say anything they want without fear of being challenged.
How can Jonathan Pritchard and those like him get away with these statements?
First, I have no way of knowing that Pritchard actually said those words. As Powell implied, the NY Times has an agenda. I once caught them in deliberate falsification of known historical fact (regarding the content of Mary Chesnut's diary), wrote to them, and was ignored. I then posted a fuming letter to a listserv for professional historians. A few of them tut-tutted, "what can you expect of newspapers?" (Ironically, one historian said that deliberate falsification was okay as long as it was in a good cause.)
Second, the most logical of Americans can become irrational when their racialism is challenged, even scientists. There are many hypotheses as to why this is demonstrably so (Cosmides, Tooby, Krzban, Aboud, Hirschfield, Holmes, Katz, Phinney, Van Ausdale). Most hypotheses point to cognitive dissonance (Festinger) or analogize to language-learning. Fortunately, you can tell when a scientist falls into this hole because he/she suddenly forgets the four basic rules of the scientist's craft (definition, falsifiability, replicability, conclusions based on finding), usually the first. Well-edited journals and textbooks catch and correct such stumbles before publication. Magazines and popularizations often miss them (sometimes deliberately, to help sales). And, as mentioned, the NY Times has not hesitated in the past to manufacture them.
Regarding Cavalli-Sforza's usage of "Caucasoid" to refer to North Africans, I believe that he was using the terminology of craniofacial anthropometry. (Before DNA, skull-bone measurements were the only method of classifying people, that followed the four rules of science.) The following map shows the span of "Caucasoid" peoples according to craniofacial anthropometry. As you can see, North Africa is the center of the Caucasoid band.
Why would the man use a precise scientific term from 50 years ago that has lost its meaning today? The man is 87 years old. He is a great scientist and everyone who uses human variation to understand migrations is forever in his debt. But, through no fault of his own, he has not been able to remain current with modern findings nor terminology. He still writes that sexual selection may be a cause of the distinctive depigmentation of northern Europeans (a disproven hypothesis), that African Americans average 25 percent Euro DNA (the number is more like 17 percent), and he sometimes uses archaic scientific terms that have become meaningless.
'
FAQ
Search
Register
Log in
