I disagree on both counts.
If the definitions were wholly arbitrary the results would be meaningless, as you originally said, and which I find a highly doubtful statement. But if you accept that the definitions are only imprecise, no it doesn’t follow that such results (test score differences) are necessarily ‘useless’. Agreement won’t be reached I’d venture, but better IMO and you’d appear more logical if you’d broken it off as ‘no the definitions of race in social measures aren’t imprecise, they are meaningless’. (they are fairly obviously not entirely meaningless, but it doesn’t seem quite as stubbornly political to insist that they are as to insist that an imprecise definition necessarily leads to useless results).
The definitions are wholly abitrary, biologically speaking – the definitions are based on culture, history, and society, and not genetic ancestry.
The conspiracy lives on…
Funny, I view as a litmus test of reasonableness when one refuses to recognize that a personal agenda of popularizing perniciously harmful and unethical claims on the violent and intellectual nature of certain people, based on historical racial groupings, using a systematic distortion of cutting edge bio-research, would clearly be disapproved by the scientific community.
…But as the old saying goes “you can lead a horse to the conclusions of the leading group of practicing population geneticists, but you can’t make him think.”
I’m well aware of all of this (check out my occupation in my public profile ;))
My point was that while statistical significance is a necessary condition in order to accept the racialism view point, it is not a sufficient condition. It is the nature of large data sets that that with enough samples even very small correlations can be proven. With enough data we could probably demonstrate that ones hair color was correlated with the likelihood of choosing a hamster as a pet (don’t ask me which direction). Similarly it is likely that due to random genetic linkage that there may be some differences in distribution between the races in any given genetic characteristic. However just because there is a difference that can be shown statistically, doesn’t mean that the difference is large enough to provide any utility for consideration of race when trying to estimate the likelihood that a given individual has that characteristic.
As an example it is entirely possible that due to genetic linkage people with black hair will, all else being equal, have an IQ on average 1 point higher than people with blond hair. But that doesn’t mean when trying to determine whether or not a person (or group of people) is intelligent one should take their hair color into consideration.
I’ve made this point in the Pit thread:
If you believe that test scores, like IQ tests, for various populations groups in the United States, are an accurate reflection of average genetic intellectual potential, then you believe that it’s just a coincidence that the two groups treated most abominably by America over its history, African Americans and Native Americans, also just so happen to have the lowest test scores (and therefore, according to your belief, the most inferior genes for intellect).
That seems pretty far-fetched to me.
Oh :o
[Nonetheless, I think you misspoke in talking about statistical significance given “potentially billions of people”, since there’s no possiblity that anyone has actually tested billions of people and discovered some tiny correlation.]
I agree. But whether you should or shouldn’t take it into consideration as a practical matter is another issue entirely (and in the case of intelligence in particular, has moral implications as well).
But suppose that was not the question. Suppose you weren’t contemplating making any sort of judgement about any individual based on their race. No, you’re just some guy wasting time discussing things on the internet. And the issue being discussed is that scientists have run a test on billions of people and have discovered that “people with black hair will all else being equal IQ on average 1 point higher than people with blond hair”. And a bunch of people full of righteous fervor start multiple threads insisting that this must be due to some environmental factor, and that anyone who suggests that it might be due to “genetic linkage” is some crank or anti-blond bigot.
What side do you take?
It’s a good thing that no such threads with no such assertions exist anywhere except in the imaginations of a few strange folks!
Yes, dividing people into the natural definitions of race makes as much sense biologically as dividing the animal kingdom by the number of legs the walk on, so that humans, birds and certain lizards are in one group while dogs and alligators are in a second group.
Quick quiz: Is this woman more closely related to this man, or this man?
How about taking this quiz. The quiz is down two or three pages. The author is very strongly anti-rascist, but does not blind herself to the extremes of liberalism.
No. We are discussing science here. Science has a default setting of 95% confidence in your correlation. Show us that or you’ve got nothing.
It’s also quite obvious to everybody that when you speak of a correlation between skin colour and sickle cell, that you mean the trait is positively correlated with darker skin colours. Attempting to weasel that into a negative correlation is fooling nobody.
Which is nothing but an argument from ignorance. Show us the evidence that the correlation even exists. You made the claim in GD, show us your evidence.
Show us the evidence for these claims please.
First, show us the evidence that a dark skinned person is more likely to be African than Australian, Melanesian, Indo-Malayan, Tamil, Indian or American Indian. Because on the face of it that seems like nonsense. The population of India alone is larger than that of Africa.
Then show us how you calculated that you had a >10% chance of being right about whether a Tamil has sickle cell based upon the shade of her skin.
The author has an undergraduate degree in Film Theory, just not very useful for this subject.
And defending Jason Richwine is also not useful, except on pointing out that you either have trouble finding good support or that there are no good pickings left among the ones that would support what you claim.
I’ve mostly been trying to give an accurate summation of Wade’s argument, not my own view. But since if I don’t declare my POV, I’m going to be assumed to agree with every word that both Wade says: I don’t need anthropologists to help me see the flaws in the second part of the book; if the science in the first part is in fact correct, the speculations in the second part are reasonable enough … but not any more reasonable than a good hard scifi story that bases the whole plot one on single bit of science. He’s critical (rightly) of Jared Diamond’s theory of geographical determinism, but veers too close to genetic determinism himself. Even if we take as a given the science in the first half of Wade’s book, a more probable reading is that genetics played a role in shaping culture, along with geography and all sorts of other things, and that we’re still a long ways from identifying them all, let alone figuring out the relative importance of each.
However, I do need biologists and hard scientists to help me evaluate the claims in that first half, which is why I thanked the OP for compiling some reviews from them.
Having read them, as well as some of the book’s favorable reviews, my general take is that the consensus among the scientists is that yeah, “race” or something akin to it does exist, and it does matter, though how much and in what ways is still TBA.
I’ll let let Wade explain what he means, taken from the links in the OP:
[QUOTE=Wade]
“So if you wanted you could say there are seven human races instead of five. But there’s no need to get hung up on numbers, race is a very fuzzy (?) concept. It doesn’t matter how many there are, the fact that people may differ on a number of races is a matter of definition and not of the fact that races exist.” …. “Biological variation is a continuum,” he wrote, “so where you want to cut this continuum is kind of arbitrary, and population geneticists have developed measures.” By one such measure, FST, humans did have enough variation to have subspecies. Wade said that a claim to the contrary was “a political judgment.”
He continued: “There’s enough variation for you to say these are subspecies or races within the single human species if you wish to, and exactly how many of them there are is really a question of not very much interest. The whole system is continuous, an unbroken spectrum from races to the ethnicities between races, which all depends on how many genetic markers you wish look at.”
[/QUOTE]
[QUOTE=Wade]
“Wade claims that the latest genomic findings actually support dividing humans into discrete races.” In fact, I say the exact opposite, that the races are not and cannot be discrete or they would be different species, but it’s easier to attack an invented statement.
The human genome points to the overriding unity of humankind. Everyone has the same set of genes, so far as is known. Genes come in the alternative versions known as alleles, so one might expect next that races would be demarcated by alleles. But even this is not the case. In fact, the races are not demarcated at all. They differ only in relative allele frequency, meaning that a given allele may be more common in one race than in another. How that translates into the familiar differences in physical appearance between human races is a matter I explain in my book.
Because of these characteristic differences in allele frequency, geneticists can analyze the genome of someone of mixed race – an African American, say – and assign each segment to an African or European ancestor, an exercise that would be impossible if races did not exist. Also because of differences in allele frequency, researchers analyzing human genetics around the world have found in surveys dating back to 1994 that people cluster in groups that coincide with their continent of origin.
Raff and Marks take issue with one of these surveys, Rosenberg et al. 2002, which used a computer program to analyze the clusters of genetic variation. The program doesn’t know how many clusters there should be; it just groups its data into whatever target number of clusters it is given. When the assigned number of clusters is either greater or less than five, the results made no genetic or geographical sense. But when asked for five clusters, the program showed that everyone was assigned to their continent of origin. Raff and Marks seem to think that the preference for this result was wholly arbitrary and that any other number of clusters could have been favored just as logically. But the grouping of human genetic variation into five continent-based clusters is the most reasonable and is consistent with previous findings. As the senior author told me at the time, the Rosenberg study essentially confirmed the popular notion of race.
[/QUOTE]
Thus, he is not arguing that there is a precise, neat division into five or seven racial groups, nor is he arguing that “races” are dramatically and entirely different.
To analogize, if we were to look at the number sequence
1 2 4 5 5 7 9 10 18 21 33 45 47 51 55 106 107 109 114 133 148 155 207
it would be true to say that there is no obvious pattern or definitive taxonomy. But there are groupings we can make based on the observed data – the “undelined under 10s,” the “over 100 in boldface,” and the “11-99s without formatting.” The categories blur, and the boundaries are not clear, and you could argue that there are two categories rather than three (the bold-and-over-100 grouping seems more clearly distinct). What you cannot do, unless you are being willfully obtuse, is look at that series of numbers and deny that general groupings do exist.
He does point out that some of the more easily identifiable groupings do emerge when the genetics are studied, and that that they do have some correspondence with geography. These are the “ancestry informative markers” that forensic scientists use to dig up a skeleton and state with a high degree of confidence what color that person’s skin was. This, to take it from some of the critics cited in the OP, is not controversial:
“When Henry Louis Gates Jr. sends a sample of his DNA off to find out how much is of African versus European origin—and then acts as host of a PBS miniseries in which he broadcasts the results—it seems hard to maintain that educated people deny that DNA sequences differ subtly among continents.”
“Human genetic diversity does mirror geography, as does much phenotypic diversity.”
http://www.michaeleisen.org/blog/?p=1609#sthash.FBQDkbGy.dpuf
“First, let’s examine Wade’s many straw men. He argues against the claim that all the people of the world have essentially similar genetic makeup (a position not actually held by any credible geneticists or anthropologists)”
http://election.princeton.edu/2014/07/08/race-and-mental-traits-nicholas-wades-third-error/
Note that that last is almost insulted by the implication – and yet the precise assertion (that all the people of the world have essentially similar genetic makeup) is one I’ve seen proclaimed loudly on the SDMB. Similarly, I’m pretty sure that if Chief Pedant were to say “educated people can’t deny that DNA sequences differ among continents,” or “Human genetic diversity does mirror geography,” he’d be met with howls of derision.
The gist I take from the critics w/r/t Wade’s take on the hard science – and I’ve only read the critiques from hard scientists, not from the sociologists – is that Wade oversimplifies and jumps to too many conclusions – but most of them do not argue the point, and indeed some explicitly confirm it – that there are indeed identifiable regionally-based population differences. (All those reviews are generally negative of the book as a whole, but some are reasonably ok with the first part, and they do not all agree with each other) I don’t get the impression Wade is overly invested in the nomenclature, and would say that whether you call them “groupings” or “races” or “populations” or “subspecies” or “Larry” is not the point; and certainly it doesn’t matter to me as a layman. The scientists themselves seem divided on whether or not they should use “race,” and one of the ones that does reject it nonetheless asserts that his view is the minority one:
"Indeed, the current predominant belief amongst human geneticists and biomedical researchers is that the socially-defined races of U.S. society are biological races, and that genetic differences between these groups have important biological and medical consequences (Graves 2011)."http://www.thisviewoflife.com/index.php/magazine/articles/book-review-great-are-wades-errors-in-a-troublesome-inheritance-genes-race
For me, the main takeaway from both the book and the criticism is that that there are observable groupings, tied to geography, and that it’s not merely an issue of “some numbers are boldface, some aren’t, no patterns, nothing to see here.” Certainly the variation between groups is much less than the differences among individuals; but the suggetion that we’re just all unique and that there are no patterns whatsoever seems to me pretty untenable.
To clarify the point I was making:
- We can and sometimes do know Gene #1 is linked to trait A. (in my example, it was a gene linked to aggressiveness).
- We can and sometimes do know that group X is much more likely to have gene #1 than group Y.
AFAICT, these are facts not in dispute.
- We cannot conclude, however (contra, I think, some that have argued on this site), that group X is more likely to have trait A, because
[ul]
[li]There are other genes, #2 and #3, that are also linked to trait A, whose incidence in group X and Y we don’t know.[/li][li]There are many, many, many genes whose functioning we do not understand and likely will not understand for many years, and it is likely that many of them (#4, #5, #6, #73 …. ) are also linked to trait A – perhaps, for all we know, much more strongly linked, and which may be more present in group Y than in group X.[/li][/ul] - On the other hand, it is highly improbable that if/when we were indeed able to decode the entire genome (in the year 3000, perhaps) we would find out that when we account for every gene, every sequence, and every allele that – mirabile dictu! – each and every group had exactly the same tendency to trait A all along. That’s only slightly more likely to be the case than to suggest that each individual has exactly the same genetic tendencies (and if that were true I’d argue it was pretty strong evidence for intelligent design.)
Ergo: we know that there is an overwhelming probability of group differences between X and Y, even though we don’t currently know what those tendencies are, and may never.
Hope that’s clearer.
Why on earth would you look at that data and make those groupings if you had no prior reason to? There’s several alternative groupings (with 3 groups) that make a lot more sense and don’t involve any blurred lines - “all underlined”/“all bold”/“unformatted” being one. Numbers with 1/2/3 digits being another. x<30<y<100<z is another. I’m sure more no-blurred categories exist.
The only reason (I can see) to come up with the 3 groups you did, would be the assumption you didn’t state - that the groupings had to be continuous and unique on the numberline. i.e they had to be a “geographic” cluster.
Can you see how if you apply that logic to human populations, you’re assuming the existence of races before you even start?
This, so much all of this.
Good post.
You are confusing the degree of confidence that you have that there is a correlation, with the degree of correlation itself.
Of course. This is obvious.
I’m not sure just what it is that you’re confused about here. But I guess it could be anything.
[This kind of thing is actually pretty instructive for me. A lot of these discussions focus to one degree or another on technical aspects of genetics, and I tend to stay out of that angle, because I don’t feel I know enough about genetics to have an definitive opinion about these matters. But I do have this suspicion that a lot of the people posting vociferously on those subjects are also ignorant and have just picked up some technical jargon that they dish out. (I base this primarily on the regularity with which they accuse people who do seem to have qualifications in the field of being ignorant fools - usually that’s the sign of a deluded fool.) But I don’t know.
But I do know statistics. And what I see in discussions of statistics is exactly what I suspect is happening in genetics - people with vague and misformed understanding of fundamental concepts just tossing around jargon. And some of these are the same people. So my suspicions about the genetics discussions is strengthened.]
A good point, completely true, and I thank you for making it.
I don’t see any reason we have to choose one versus the other, however. We all belong to multiple categories, and the fact that there are obvious no-blur ones doesn’t mean the ones that are more subtle are meaningless. If the boldface in that numberlist represents Horrible Incurable Genetic Disorder, it seems to me worth knowing that HIGD does not just randomly occur, that some numbers are high-risk and some are low, etc.
An eight-foot-tall female Catholic diabetic with chronic flatulence has things in common with other people who are tall, female, Catholic, diabetic, and/or with chronic flatulence. The tallness is immeditely visible, the diabetes definitely isn’t, the others may or may not be depending on social expression. It would be wrong to assume that her tallness or her gender define who she is or are the most important things about her … but it’s equally silly to say things like “height doesn’t matter,” when we know there are a list of medical risks associated with it.
The argument – not Wade’s and certainly not mine – isn’t that race is the only thing that matters, nor that is the most important thing … merely that recent finding seem to indicate that it isn’t meaningless or unimportant.
No. The fact that there are other distinctions matter, too, doesn’t make the geography irrelevant.
Again: Wade’s critics (and again, I only read the hard scientists), agree that geography matters, and even say that no educated person can think otherwise. See the quotes above.
One thing I find odd in these discussions is that on the one hand many anti-racial people claim “everyone is the same” or something to that effect, and deny that “races” exist, but at the same time, one study that is constantly being trumpeted as refuting the race-genetics hypothesis is completely predicated on the notion that not only can you tell whites and blacks apart via blood tests, but that you can measure the percentage of white versus black ancestry via blood tests.
These seem highly inconsistent to me (although I haven’t bothered tracking down whether it’s the same people making the two claims).
Again, going by the numbers you showed, the only numbers you can say are at high risk are the ones that actually get the “bolding disease” unless you assume some geographic commonality.
It is meaningless - actually, it’s negatively meaningful - when it obscures the actual relevant groupings, the way using skin colour or even “race” as a sickle cell proxy is wrong and misleading.
Not geography, exactly - there’s nothing wrong with smaller geographic groupings. But continent-level groupings have been consistently shown to be useless for any real science work. They obscure cross-continental associations and inter-continental dissociations(like calling sickle-cell an African disease obscures both the Med/Arabia cross-occurrence and the non-occurrence in South Africa and the Horn)
Geography matters, geographic-based races don’t. If someone told me something (like a disease, or a twitchy muscle gene, or a skin colour) had a “West African” occurrence, I’d have no problem with that level of grouping. If they told me it was “African”, I’d laugh in their faces.