Hello There, Guest!  
 Previous 1 26 27 28 29 30 40 Next   

[OBG] Nature of Race Full

(2015-Jul-20, 19:51:10)Krom Wrote: And here's the Brace paper:


So Brace writes:

Quote:Evidently, traits that are distributed in conjunction with the graded intensity oftheir controlling selective forces will be poor indicators of population relationships
(Darwin, 1859). This is the logic behind Livingstone’s classic phrase, “There are no
races, there are only clines” (Livingstone, 1962:279). The use of a characterization
of a single trait that is under selective force control to generalize about any particular human population can only create confusion. This then will be the inevitable consequence of the use of a description of skin color to say anything about the general nature of human biological variation...

Figure 2 might be construed as providing support for the hoary folk belief that modern Homo sapiens can be sorted into three convenient “races”: “caucasoid,” “mongoloid,” and “negroid.” When the number of separate regional representatives is multiplied, however, it becomes clear that the ties between adjacent twigs on the dendrograms are simply indications of the extent to which geographically adjacent people are genetically related to each other rather than the extent to which they reflect anything that could be called a “racial” essence. Large geographic regions obviously will have many resident and related populations, and an assessment of their trivial traits will automatically produce adjacent twigs on a dendrogram which by definition constitute a cluster.

So, Brace's results support the idea that humans can be divided up into divisions delineated by propinquity of descent, by not ones divided by "essences". Ok, but location in genealogical space can constitute relational essences. So, in fact human can be divided into essentialist group, jut not intrinsic essentialist ones.

Quote:Remarkably, Blumenbach’s consideration of Egyptian form in the perspective of what he knew about the worldwide spectrum of human biological variation was more sophisticated than the crude, categorical “eitherior” treatment of his 19th and 20th century successors. He identified three “varieties in the national physiognomy of the ancient Egyptians:” an “Ethiopian cast,” “one approaching to the Hindoo” and a “mixed, partaking in a manner of both the former” (1794:191, original emphasis).His use of the term “mixed,” however, did not refer to the actual mixing of separate populations. Instead, it was a purely descriptive expression. He concluded that “the Egyptians will find their place between the Caucasian and the Bthiopian,” where he was using the term “Ethiopian” to refer to all of sub-Saharan Africa (1794: 193).

I would of course agree that Blumenbach’s considerations were sophisticated. I would disagree that later ones were not. Though we would have to look on a case by case basis.

Quote:The possibility of a connection between South Asia and Egypt emerged once
again as a result of the metric exercises carried out by some of the proteges of Karl Pearson in London. While there was a continuing effort to see something “negroid” in the Predynastic Egyptians (Morant, 1935, 19371, the use of the Coefficient of Racial Likeness managed to provide a quantitative dimension to Blumenbach’s assessment. Cranial similarities were shown between Predynastic Egyptians and “the primitive Indian, the Dravidian and the Veddah,” at the same time that a clear-cut separation from the “Negro type” was noted (Stoessiger, 1927:147)... Both discriminant function (Fisher, 193613, 1938) and D2 (Mahalanobis, 1930, 1936, 1949) have been accepted as useful approaches to population comparisons that are not plagued by the problems of the Coefficient of Racial Likeness (Howells, 1973), and we have used discriminant functions to produce the values in Table 5.

So the method Brace uses is just a more sophisticated version of the one used by biometric typologists.

Quote:[Blumenbach's] successors (except Prichard) adopted an increasingly categorical and essentialist view of the nature of human biological variation where a population was either one thing or another-or a literal mixture between them. Blumenbach, however, saw human form as grading without break from one region to another (Blumenbach, 1865). The continuum could be cut however one might choose to suit one’s convenience.

It wasn't just Blumenbach, but other race theorists such as Buffon and later Darwin. But the question is whether our mysterious "typologists" thought so too. Regarding Aristotlean essentialism, classics scholar James Lennox commented that Aristotle's essentialism “is at once sophisticated and remarkably unlike what passes for "Aristotelian essentialism" in modern philosophy”. He is absolutely correct; the same, I imagine, holds for 19th/20th century biometric typology.

Quote:The old-fashioned chimerical concept of “race” is hopelessly inadequate to deal with the human biological reality of Egypt, ancient or modern. But neither the use of clines nor clusters alone can present a complete account. An assessment of both is necessary before we can understand the biological nature of the people of the Nile valley.

Ok, but Braces "clusters" are Buffon's, Blumenbach's, and Darwin's -- and I would argue Hooton's -- races. So, in the end he is saying that we need both "cline" (character gradient) and "race" (correlated ensemble of character which index relationship) concepts. It's funny how the stuff you cite never really supports the stuff you say, when you look at what is discussed, not the words.

But you say:

Quote:The "race types" were an ideal/eidos, this is why. This view did not disappear after Darwin, it continued.

Again, I would like to see a specific references. I am 100% sure that you are confusing issues; a platonic eidos was a transcendent form, an Aristotelian essence was an immanent one or structural design/genetic program. Species realists believed in something like the latter and believed that these did not change over generations.

Now, the problem with your claims is that "types" and "essences" in the sense of eidos refer to species realist essences. So for example:

Quote:In this century, the systematist Ernst Mayr (eg, [1970]) has championed the view that what he calls 'typological thinking' has been abandoned by modern biologists in favour of what he calls 'population thinking'. Typology is the view that there are 'types' - unchanging forms that are what makes a species what it is. It derives from the philosophy of Plato, who claimed that true knowledge is knowledge of the Idea (Greek eidos ).

By definition these could not have been races in the sense of intraspecific divisions which diverged from a species common stock (Buffonian/Darwinain races). When you claim that races were thought of this way you are using "race" in the sense of "species" (as used by Morton and Nott) -- when doing so you are discussing a different concept. To be clear, insofar as our biometric typologists were discussing intraspecific race (divisions of a species), they were not discussing edios.

But what were they discussing. Let's refer to Hooton's from Up from Ape and finish that discussion:

On the evolution of races and racial classifications

(Pith: Populations becomes isolated and differentiate due to natural selection on individual variants; they can be divided into primary races, primary subraces, secondary races, secondary subraces. Comment: this is clearly a Darwinian frame as we would expect; no ever-present edios.)


On the measure of races

(Pith: Races are diagnosed by quantitative and qualitative differences. Comment: quantitative differences necessitates individual variation; qualitative differences expressed in frequencies do likewise.)


One variants in races and homogeneity

(Pith: With isolation and inbreeding races become increasing homogenous. Comment: Hotoon speaks of "more or less" and "relative" homogeneity; this implies individual variation -- both in his primary and secondary races -- as does the occurrence of novel mutations and ability of primary races to subracinate and species to racinate.)


(See also his discussion of somatotypes.)

What we can say is that Hooton did not emphasize individual variation. But he clearly recognized it and it shows up in his metrical data.
(2015-Jul-21, 04:59:19)Chuck Wrote: But what were they discussing. Let's refer to Hooton's from Up from Ape and finish that discussion:

In case you missed it, here's the point: Hooton was a Darwinian. His primary races differentiated from a common species population because selection acted on individual-variants and because populations were isolated. In the same way his primary subraces differentiated within primary races. Since these primary races included primary subraces and individual variation which allowed for subracianation, they were necessarily somewhat heterogenous. So you are down to arguing that he over-estimated the relative homogeneity of primary races -- calling them along with some sibships and some family stocks "types" or molds. Well, how shall we evaluate that position? "Type" versus "population" sounds a lot like descriptive emphasis to me.

Oh, I hope that you didn't miss this passage:

"...indeed, there are no such profound morphological and physiological differences between Mongoloids, Negroids, and Whites, as would seem to justify the implication of any very great and geologically ancient genetic discontinuities between these groups, it would seem preferable to recognize them simply as different races. Then because the differences between the various subgroupings within the three great or primary races are obviously more recent and quantitatively and qualitatively less than exist between the primary races respectively, it becomes necessary to relegate them to the subrace category"
You're mistaken about genetics:

"race concept isn't useful because there is more variation in populations than between them" which precludes the accurate classification of humans into racial (i.e., genealogy-based) groups."

This is not false (which you claim). Craig Venter and James Watson for example are more genetically similar to Seong-Jin Kim, than to each other:

[Image: pics.png]

"Among the first genomes completely typed were those of James Watson and Craig Venter, two U.S. geneticists of European origin; they share more alleles with Seong-Jin Kim, a Korean scientist (1,824,482 and 1,736,340, respectively) than with each other (1,715,851). This does not mean that two random Europeans are expected to be genetically closer to Koreans than to each other, but certainly highlights the coarseness of racial categorizations. On average, nearby populations tend to resemble each other more than distant ones, but individual members of the same population, Watson and Venter in this case, can be very different. In short, if races are defined as subspecies, there is no such thing in humans. The best way to know what is in a person’s DNA is to study that person’s DNA." (Barbujani & Pigliucci, 2013)
(2015-Jul-22, 20:38:19)Krom Wrote: You're mistaken about genetics:

This is not false (which you claim). Craig Venter and James Watson for example are more genetically similar to Seong-Jin Kim, than to each other:

Quote: (Barbujani & Pigliucci, 2013)

Naturally, I discussed the matter:

Quote:II-H. Genomic-Genealogical Complications

This suggests that, on the individual level, genealogical and genomic similarity need not always correspond. However the results are not clear cut. Tal (2012), discussing the example of James Watson, Craig Venter, and Seong-Jin Kim, noted:

"Our model also facilitates the assessment of results from analysis of complete genome sequences. 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. 2 E). 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)."

In short, it seems quite unlikely that Venter actually shares more genetic information with Kim than with Watson. Tal (2012) found the following, based on a hypothetical infinite number of slightly informative loci:

"The probability that a random pair of individuals from the same population is more genetically dissimilar than a random pair from distinct populations is primarily dependent on the number of informative polymorphic loci across genomes from the total population pool. This probability asymptotically approaches zero with a sufficiently large number of informative loci, even in the case of close or admixed population."

The reason for the disagreement between the results of Tal (2012) and Witherspoon et al. (2007) concerning close and admixed populations is not clear. Tal's (2012) results fit with those of Gao and Martin (2009) who found almost complete differentiation between populations when using a large number of loci. Whatever the case, we have to take into account possible instances of such discordance. When/if this occurs what does one do? There are two possibilities...

I solicited comments from both Barbujani and Pigliucci but after an initial exchange they did not follow up, presumably because they had nothing to say. It's funny that they cite this example (based on data from 2008), since whole genome data has become quite common -- if Barbujani & Pigliucci's position were true, the discordance would have been obvious in the 1000 Genomes sample. Also, it's odd because Watson's data has been updated, so the low coverage hypothesis could readily be disconfirmed.

Whatever the case, since I am an imaginative fellow, I did entertain the possibility:

Quote:Since we are advancing a general race concept, we will not decide which is the better method of defining "overall" genetic similarity in the case of genomic and genealogical discordance. We would suggest going with genomic similarity, though. If two horses begat, in the natural way, a genomic and phenotypic human, most people would probably classify the genealogical horse genomic human as a human. That is, we imagine that most people would classify by genomic phenotypic similarity, and not pedigree, in cases of discordance. So, when it comes to racial groups, doing the same would seem to be reasonable. Whatever the case, we leave the issue undetermined. These are just different formulations of a basic conception. In practice, this is not a pressing matter since, on the individual level, the correspondence between the two forms of genetic relatedness is extremely high when dealing with non-trivially-differentiated populations

In principle genomic similarity can be discordant with genealogical and genotypic similarity (the latter meaning similarity in coding DNA). In these cases, one would have to specify how to define race, species, etc.. You see this problem presently in taxonomy at least on higher order levels, where classifications delineated strictly by descent (i.e., cladistic ones) differ from those delineated genomic similarity. In practice, this is a non-issue especially when dealing with individual assignment to divisions. With enough ancestry informative markers you can without exception assign individuals their pedigree defined populations (or if the individuals are admixed, to some higher order or admixed population).

Anyways, in regards to our earlier discussion, I found the following passage in Darwin's "The Variation of Animals & Plants Under Domestication":

"Pallas maintained, and he has had some followers, that variability depends exclusively on the crossing of primordially distinct forms...In an early part of this chapter it was stated that Pallas and a few other naturalists maintain that variability is wholly due to crossing. If this means that new characters never spontaneously appear in our domestic races, but that they are all directly derived from certain aboriginal species, the doctrine is little less than absurd; for it implies that animals like Italian greyhounds, pug-dogs, bull-dogs, pouter and fantail pigeons, etc., were able to exist in a state of nature. But the doctrine may mean something widely different, namely, that the crossing of distinct species is the sole cause of the first appearance of new characters, and that without this aid man could not have formed his various breeds. As, however, new characters have appeared in certain cases by bud-variation, we may conclude with certainty that crossing is not necessary for variability."

I would like to see what Pallas actually had to say. Can you look for his work? He seems to be cited on account of the unusuality of his position.

I wonder if he thought that "races" were hybrid Linnean species (i.e., creator made ones) -- from what I recall, Linnaeus conjectured something like this regarding "constant varieties" -- if so, he would not have been discussing races in the sense of divisions of a species which descended from a common stock. The trick, for you, is to find someone who actually believed that races as "divisions of a species which descended from a common stock" were homogenous.

I only found one case and the author was a historian, not natural scientist. He also only seemed to be semi-serious when proposing his model.
Can't we just download their genomes and see? Are they not public? Piffer knows how to use the software to calculate the distance metrics.
Barbujani & Pigliucci (2013) do clarify "on average, nearby populations tend to resemble each other more than distant ones, but individual members of the same population, Watson and Venter in this case, can be very different."

So typically (but not always) two individuals from a population will be more genetically similar, than an individual from another population.

I don't get though what you think this has to do with race. The small amount of genetic variation between populations is best captured by an "isolation by distance" model, i.e. spatial gradients of allele frequencies.

"Therefore, there is no reason to assume that major genetic discontinuities exist between different continents or races."

Now it might be argued that it is convenient to arbitrarily divide this genetic continuum, is this your stance? I see this as invalid. My reasoning is as follows:

"Imagine a species that lives on a continuous plane, but has low dispersal rates, so that allele frequencies vary continuously across the plane. If we sample at K distinct locations, we might infer the presence of K clusters, but the inferred number K is not biologically interesting, as it was determined purely by the sampling scheme. All that can usefully be said in such a situation is that the migration rates between the sampling locations are not high enough to make the population act as a single unstructured population." - Pritchard, Jonathan K., Matthew Stephens, and Peter Donnelly. 2000. "Inference of Population Structure Using Multilocus Genotype Data." Genetics 155(2):945-59
In response to the forensics -

The "counting methods" are anthroposcopic analyses used by forensic scientists, but they also use Giles and Elliot's (1962) discriminant function. There are many studies showing all these methods produce poor results, e.g. Birkby (1966) tested Giles and Elliot's method (which is supposed to be superior to the "counting methods") and yet found only < 65% of aboriginal crania were correctly identified as "American Indian" (racial category):

"The study shows that non-deformed American Indian crania are racially misclassified as American White and Negro in 35.6% of the cases when using this metrical method."

Henneberg (2015) in his study did note contrary to your criticism:

"Authors of each method gave somewhat different names for the ‘racial categories’ into which their method is supposed to classify and individual. For purposes of comparing the methods we have grouped their results into three general classes; Black, White and Other. ‘Black’ includes any determination pertaining to Sub-Saharan African ancestry, ‘White’ includes any determination pertaining to European ancestry, while ‘Other’ includes any determination that is pertaining to Asian, Amerindian, Indigenous Australian and Oceanian ancestry."

So I don't see the problem. Only (8) and (9) don't have an "other" category, but what is revealing is that some of the same skeletons are considered "South African White" by one of these but "South African Black" by another, when both these methods are using the pelvis.

Here is another study showing the same thing:

"The claim of 90% accuracy that has been reported for “race”- determination methods is unsubstantiated. Despite the relatively high-allocation accuracies (often more than 80%, but rarely more than 90%) and the strength of the statistical significance that are noted when various methods are first described, the comprehensive independent tests of “race”-determination methods consistently result in low-allocation accuracies." - Albanese, John, and Shelley R. Saunders. "Is It Possible to Escape Racial Typology in Forensic Identification?". (2006). In: Forensic Anthropology and Medicine. Humana Press: 281-316

In response to anthropologists (Hooton's students) who abandoned race for clines, Birdsell is covered in the following -

Littlefield, A., Lieberman, L., Reynolds, L. T., Azevêdo, E. S., Beals, K. L., Brace, C. L., ... & Wolfe, L. D. (1982). Redefining race: The potential demise of a concept in physical anthropology [and comments and reply]. Current Anthropology, 641-655

[Image: clines.png]
(2015-Jul-23, 04:26:23)Krom Wrote: So typically (but not always) two individuals from a population will be more genetically similar, than an individual from another population.

Given the genetic divergence values, the probability is infinitesimal. Given this probability and the reported problems with the watson data, occam's razor counsels that we discount the data point for the time being. If at a future point your agreement to my position turns on the matter, I will find someone to re-analyze the data, which is readily available: http://ftp.ftp.ensembl.org/pub/current_v...o_sapiens/

Quote:I don't get though what you think this has to do with race. The small amount of genetic variation between populations is best captured by an "isolation by distance" model, i.e. spatial gradients of allele frequencies.

Naturally enough, I explicitly formulated the concept such to be consistent with divergence due to isolation by distance (e.g., page 41) -- which is to say a continuum due to primary intergradation. This is what distinguishes it from e.g., Shiao's "clinal class" one (see page 64).

Now you quote:

"Therefore, there is no reason to assume that major genetic discontinuities exist between different continents or races."

You know that I know the history of this debate at least as well as you do, so I don't see why you would cite the above. Shiao (2014) addressed this in his defense of his clinal class concept. But of course, you are right, my stance is:

(a) one could "arbitrarily divide" a "genetic continuum" into races.

Of course, I anticipated objections to this position and discussed them lengthy. Which parts of my discussions didn't you find convincing? Do you want me to refer you to the specific sections or to elaborate on specific points?

Now, what did we decide about our typologists? The more evidence I find in defense of your homogeneous-race argument, the worse it looks. Consider the following statement by Manouvrier in "La détermination de la taille d'après les grands os des membres":

Quote:D. Enfin les variations ethniques des proportions du corps seront dans le même cas que les précédentes. Il y a des races macroskèles et des races microskèles, comme il y a des indivi dus de ces deux sortes, et les variations individuelles sont bien plus grandes que les variations ethniques les plus accusées. Or les coefficients moyens des os de grande longueur tendant à abaisser la taille et ceux des os de faible longueur tendant à l'élever, il s'ensuit qu'il sera tenu compte dans une certaine mesure de la macroskélie des races comme de celle des indi vidus dont les os seront absolument longs et de la microskélie des races comme de celle des individus ayant des os absolu ment courts. Il est vrai que, par exception, la microskélie peut exister avec des membres absolument longs ou au moins de longueur moyenne. Cela ne peut arriver, évidemment, que pour des races d'une stature très élevée. Tel est, vraisemblablement, le cas des Polynésiens.

Roughly: "D. Finally ethnic variations of body proportions will be in the same situation as the previous. There has macroskèles microskèles races and races, as there are individuals to both kinds, and the individual variations are much larger than most claimed ethnic variations. Now the average coefficients of greater length of bones tending to lower the size and those with low bone length tending to rise, it follows that it will be reflected to some extent the macroskelia breeds such as that of individuals whose bones will be long and absolutely microskélie breeds such as that of individuals with absolute bone ment courts.It is true that, exceptionally, the microskélie can exist absolutely with long limbs, or at least of medium length. This can only happen, of course, only for races of a very high stature. This, presumably, if the Polynesians."

Commenting on this statement, Karl Pearson, who is said to be a "typologists" noted:

Quote:If we admit for the moment, which I should not be prepared to do generally, that the individual variations in a local race are greater than the "ethnic variations" or divergences between means of local races, M. Manouvrier's conclusion by no means follows.

Pearson's comment suggests that he was overestimating the relative magnitude of between group differences -- though it is not clear as it's not clear if Pearson was discussing multivariate distance (his coefficient of likeness, which was a primitive form of Mahalanobis distance) or average differences. Whatever the case, the very occurrence of this discussion calls your claim of a general view of homogenous races into question. And once you admit that variability was recognized, you are forced to double down on your claim that biometric typologists felt that variability within populations arose from the admixture of distinct homogenous groups. Yet it is clear that they didn't believe in originally distinct homogenous races. Thus Pearson notes:

Quote:Of course, assuming the origin of man to be monogenetic, "Association" and "Divergence" are only relative terms of a continuous grade of relationship, indicating only the greater length of differentiated ancestry. [/b]But they are convenient terms. It is proper to look upon Anglo-Saxons and modern English as associated races, but on Chinese and English as divergent races, if we only mean by this that the forerunners of Chinese and English diverged much earlier from a common ancestry than English and Anglo-Saxons. ("On the Coefficient of Racial Likeness")

If they diverged from a common stock in continuous degree and now contain enough variability such that a " Coefficient of Racial Likeness" is needed when and how were they ever homogenous groups in the sense you mean?
Real genetic similarity cant be derived from overall loci in common, since loci can have no effect at all and different loci can have the same effect and also non additivity. What counts is expression of the gene and thus phenotype especially in complex traits and in this case people from different races are frequently more alike to one another than to members of their own race and also in many cases by very large margins.

Its technically misleading to use overall loci in common to determine genetic similarity.
(2015-Jul-23, 06:03:13)Krom Wrote: In response to the forensics -

To preclude confusion, let's distinguish issues:

(1) whether or not the natural division concept of race is coherent
(2) whether or not such and such characters are reliable indexes of evolutionary relationship (and thus reliable indexes of racial membership)

Recall this discussion on reliability

Could you clarify which point you wish to discuss in relation to the forensic data?

As for reliable identification see if you can find a meta-analysis or something.

You say:

Quote:Here is another study showing the same thing:

I will look at it later.

You say:

Quote:In response to anthropologists (Hooton's students) who abandoned race for clines, Birdsell is covered in the following -

Quote:By 1975, however, his position had changed: "The use of the term race has been discontinued because it is scientifically undefinable and carries social implications that are harmful and disruptive."

Where does it say that he renounced the concept. This is what I get when I search his "Human Evolution: An Introduction to the New Physical Anthropology (1981)"

His statement on page 379 that "Race in the world of evolutionary biology is a valid taxonomic unit, the equivalent of species... and his discussion of "the great Negroid race or subspecies of humankind" on page 348 doesn't seem to perfectly support your claim. I will have to look more into the matter later.
 Previous 1 26 27 28 29 30 40 Next   
Forum Jump:

Users browsing this thread: 4 Guest(s)