Pharyngula

the subject of race was discussed on the September 5th episode of The Skeptic's Guide To The Universe. I thought Steven Novella had some interesting things to say on the subject.

You can download it here:

http://www.theskepticsguide.org/skepticsguide/podcastinfo.asp?pid=111

Posted by: Brian W. | September 16, 2007 1:28 PM

Basically, that article is a lot of handwaving. Basically, if the theory of evolution is true, humans living in different environments must have evolved differences. Basically, for some people when science conflicts with their ideology, they choose ideology.

Posted by: James T. Smith | September 16, 2007 3:22 PM

Basically, if the theory of evolution is true, humans living in different environments must have evolved differences.

Given enough time without interbreeding, sure.

Posted by: MartinM | September 16, 2007 3:26 PM

Bingo.

"Race" is a cultural construct. There's a difference between "race" and cline. How would the asian eyelid fold mean more than red hair as a marker of "race"? How does skin colour mean anything of that sort when the ancestors of some groups with black skin left Africa before the ancestors of the so-called Caucasians?

Posted by: Graculus | September 16, 2007 3:33 PM

somebody's race (i.e. what their ancestral ethnicity

Conflating distinct concepts.

Posted by: MAJeff | September 16, 2007 3:57 PM

This sure seems like bad science to me, of the type usually attributed to "the other side". It's clear that they are looking for a conclusion that satisfies their ideology, even if if means wearing blinders.

We know that race affects outward appearance, whether in terms of skin color or eyelids or hair texture or nose shape. We know it can affect other things, such as alcohol tolerance or tendency toward sickle cell anemia / tolerance toward malaria. Whether these things are "irrelevant" or not is totally subjective, but denying they are determined by genetics, and those genetics come from their ancestry, is just silly.

Posted by: rob | September 16, 2007 4:37 PM

'Race' is an event in which people try to go faster than each-other, typically starting from the same location and endeavoring to reach a finish line.

Posted by: Christian Burnham | September 16, 2007 5:10 PM

Obviously, looking for a strictly biological basis for the notion of race is going to fail, but that shouldn't lead us to conclude that race doesn't exist any more than we should reject notions like "hot" or "cold", or "Christian" or "Buddhist". Race isn't something that's been defined from strictly biological grounds, but rather by taking traits commonly shared by and also largely particular to people within certain a particular geographical region as the defining characteristics. No, those traits aren't going to be nicely demarcated from the rest of the traits shown by the rest of the people the world over, but that isn't grounds for rejecting the notion altogether -- not any more than I should reject notions like "green" or "blue" because there are hundreds of paint chips between the two, and I can't pinpoint where they stop being green and start becoming blue. That's how the human brain works -- it is an analog computer that does nothing but develop loose conceptual associations for the purpose of classifying the world around it. And it works. That's why things like weight and hair color aren't every bit as good for defining race as facial structure or skin tone, they have to traits which are generally common and particular to a group of people who come from a geographical region which, for whatever reason, we consider to be a region worth applying the label to.

It makes exactly as much sense to dismiss the notion of distinct religions. What makes the difference between a schism and a completely different religion? It isn't genetics and it isn't usually any one ideological trait. But we don't make the argument that any ideological trait could be used as well as any other to demarcate one religion from another. Rather, we simply make a quick mental rundown of percieved similarities and differences and decide, with respect to the system in question, which ones stand out as the best markers for classification.

I think dismissing the notion of race is going to garner knee-jerk reactions from people who think that they have a good idea of what race is (that's nearly everyone). Everyone, even the ones who pretend not to, makes decisions based on nothing more than fuzzy conceptual associations, so it is better to challenge whether "race" is a classification of value, rather than a classification at all. In any case, there was more than biology that went into the classification, so simply suggesting that biology doesn't by itself support that classification isn't going to dispel the notion of "race".

Posted by: OhNoDustinDidn't! | September 16, 2007 6:15 PM

I should make my stance more explicit -- I think (as the post was claiming) that race is an irrelevant classification. I'm taking issue with the "race doesn't exist" line of attack. It should be a "'race' is poorly defined, based on geography and historical tensions, and has no utility" line of attack instead.

Posted by: OhNoDustinDidn't! | September 16, 2007 6:19 PM

I'm a sociologist. Race has social origins not biological ones.

Posted by: MAJeff | September 16, 2007 6:22 PM

so Christian, what is race biologically?

Posted by: MAJeff | September 16, 2007 6:52 PM

rob's attempt to say that race affects skin color etc, is exactly backwards.

Posted by: rob | September 16, 2007 7:03 PM

Our being visual primates, it's not especially surprising that they focus on dramatically visible phenotypic features (skin color and such), and downplay the more subtle population variations.

True, but that never stopped us classifying all other species into races, varieties, subspecies and whatnot.

Posted by: windy | September 16, 2007 7:23 PM

We know that race affects outward appearance, whether in terms of skin color or eyelids or hair texture or nose shape. We know it can affect other things, such as alcohol tolerance or tendency toward sickle cell anemia / tolerance toward malaria.

These all vary independently. For example, sickle-cell anemia* occurs only in a strip of Africa between then Sahara and the rainforest (and obviously in descendants of people from there); people just south of there look identical but lack the allele.

Take hair color and the rhesus factor. Almost all of the variation in these two occurs within people that have are tightly clustered around one extreme each in skin color, eyelids, hair texture, and nose shape.

* You can't have a "tendency toward" that. Either you have the allele, or you lack it.

rob's attempt to say that race affects skin color etc, is exactly backwards. Those phenomena have been used to define race.

And not always in the same way! I keep being amazed at the US definition of "black", which is "any visible amount of African ancestry", so that it includes even Colin "Paleface" Powell. Not only is he not considered "white", he's not even considered "mixed"! That wouldn't happen in Brazil, as far as I've read.

Posted by: David Marjanović | September 16, 2007 7:38 PM

David Marjanović,

The Colin Powell/Brazil issue you raise is interesting. I've seen estimates of between 49 and 54% of the Brazilian population having African ancestry. In the US, we declare 12% of the population "black" but that depends on factors such as how many of your great-great-great-grandparents had ancestry in Africa. If one of them is "100% black," you're black; if that person is "50% black" you're white (at least in Louisiana where 1/32 "black blood" officially designates one as black--upheld by SCOTUS)

When my students ask me how many races there are, I tell them "However many a society defines there to be." Yeah, it's inexact, but we're dealing with an ephemeral concept conjured into existence.

Posted by: MAJeff | September 16, 2007 7:48 PM

We know it can affect other things, such as alcohol tolerance or tendency toward sickle cell anemia

Oh, lovely. tell me, what "racial" characteristic indicates a higher chance of sickle cell anemia? I'll tell you which one it isn't, and that's skin colour.

When Cavalli-Sforza produced his genetic map, he was asked why he had 256 "races". His response was that the computers would only let him use 256 colours.

The genetics blend seamlessly, the concept of "race" (as opposed to clines) is a social one, and is not useful in science.

Posted by: Graculus | September 16, 2007 7:49 PM

Greg Laden's article is an excellent article on the lack of correlation between skin color and running ability.

In other words, I think what's being contested is the idea that [...] constitute a "suitcase" of traits and that possession of the first two or three is a reliable predictor of the others--not the idea that each of these individual traits is hereditary or has a genetic basis.

Bingo.

On another note, the cats (Felidae) and the dogs + foxes (Canidae) are traditionally considered families (that's why their names end in -idae) belonging to different suborders (Caniformia, Feliformia; Caniformia also includes the bears, weasels, seals, and so on) of the same order (Carnivora), while Bacteria and Archaea are considered different domains (each containing several kingdoms). But don't confuse the clades with their ranks: the clades are real, the ranks are 100 % subjective; the clades exist in nature, the ranks are a convention (and a rather unstable one).

Posted by: David Marjanović | September 16, 2007 7:55 PM

My best friend would be considered "Asian" considering that her mother is from Korea. However, her father's ancesters come from Norway. What is she?

I'm saying the continent of origin has no inherent meaning or importance, and only takes on such under social conditions. Are you going to dispute that?

Posted by: MAJeff | September 16, 2007 8:15 PM

let's take it a step further, rob, what's Tiger Woods?

Posted by: MAJeff | September 16, 2007 8:18 PM

And, moving beyond what race is, why is it so important to define people's race, rob?

Is it clear whether a polar bear and a grizzly bear are the same or different species? Not really, because they can and do interbreed in the wild and produce fertile offspring.

I suppose a case could be made, then, that the word "species" should be tossed out the window, because it is meaningless. Honestly, I don't have a strong opinion on that one. As with the word "race". They are just words, and I'm not interested in the words. You keep implying I am, but I'm not.

But if you then try to argue that the only difference between this animal and this animal is "cultural", I'm gonna call BS.

Posted by: rob | September 16, 2007 8:32 PM

Rob wrote:

The only reason is because there are genetic differences which happen to result in differences to their appearance. Are you seriously arguing otherwise? Seriously???

People do have genetic differences that make us look different from one another. That isn't really the issue. The issue is which traits do we select to define race? Skin color turns out to be not terribly useful from a genetic or biological standpoint.

Part of the problem is that race is typically defined by what is most obvious to us, like skin color, but what is most obvious to us may not be the most meaningful biologically trait. More importantly, the differences that are so obvious and easily identifiable to us are dwarfed by the genetic similarities we share.

Posted by: Leni | September 16, 2007 8:37 PM

More importantly, the differences that are so obvious and easily identifiable to us are dwarfed by the genetic similarities we share.

Posted by: rob | September 16, 2007 8:51 PM

race is typically defined by what is most obvious to us, like skin color, but what is most obvious to us may not be the most meaningful biologically trait.

However, all of these things taken together DO give us a fairly good means of estimating what part of the world someone's ancestry is mostly from. Sometimes the information more reliable than at other times, of course, but with people from european vs asian vs african descent, it is generally pretty reliable. Obviously, the more "pure" their ethnic background, the more reliable it is.

And that comes from the fact that people whose ancestors come from different parts of the world would be expected to have more genetic differences than people with ancestors from similar parts of the world. Some of those would be expected to be superficial appearance differences, some might be other differences like immunities and such.

This doesn't mean some people are "superior" to others based on their ancestry, or that people fit into neat little categories. If you read what I write and assume I think those things, read it again and again until you are convinced otherwise, please.

Posted by: rob | September 16, 2007 9:04 PM

#33: My best friend would be considered "Asian" considering that her mother is from Korea. However, her father's ancesters come from Norway. What is she?

When populations mix and interbreed, the early-generation individuals of mixed ancestry are likely to be identified by their differences from the majority group. You gave a good example of the strong cultural bias (at least in America) to classify "mixed" individuals as members of minorities. That bias is obviously purely cultural, with zero biological basis.

It follows logically that in a highly "mixed" society, such as present-day North America, any biological significance race might have had when populations were separated, have lost most of their significance.

But this line of argument does not help to answer the question how much significance there was originally.

Posted by: bullfighter | September 16, 2007 9:19 PM

Here, rob:

What race is this man

What can you tell me about him from this picture (geographic origin, diseases, anything)

Posted by: Graculus | September 16, 2007 9:37 PM

However, all of these things taken together DO give us a fairly good means of estimating what part of the world someone's ancestry is mostly from.

Since I moved to Europe, I have met quite a few people here who are completely surprised I am an American. "Wow, but you don't look American!" Evidently, they have some ideas about what Americans looks like, and somehow I do not fit into that. With piqued curiosity, I have asked for an explanation, but I have never managed to get anything coherent out of them. Rob, you sound like these people.

Sometimes the information more reliable than at other times, of course, but with people from european vs asian vs african descent, it is generally pretty reliable.

Rob, I am sure you're a nice person and all, but this kind of talk just doesn't come across well. Maybe you think it sounds smart, but I would ask you to follow your own advise and re-read your posts... and keep re-reading them until you see why people might not think that you're coming up with any new insights.

Posted by: j.t.delaney | September 16, 2007 11:04 PM

Rob wrote:

What does that even mean? We share 50% of our genetics with a banana. Does that mean we are "equally similar to a banana as we are different"? Relative to what?

It means that the difference in, for example, skin color or eye fold phenotypes are relatively minor considering the vast amount of similarities we share, genetically speaking.

Here is an explanatory snippet from the same text I mentioned in my earlier post:

"Anthropologists and geneticists have computed these percentages based on data on indigenous populations from various large geographic regions of the world...that have often been suggested by proponents of the race concept as representing different major "racial" groupings. Studies of blood group and other genetic polymorphisms show that roughly 10 percent to 12 percent of the total variation in our species exists between these regions, with the remaining 88 percent to 90 percent existing within regions... The same pattern also applies to cranial and facial measurements...The bottom line is that geographic race explains only 10 to 12 percent of the total variation in the human species."

Emphasis in original. In terms of our genetics, the typical markers that we ascribe to race just aren't all that important.

Further, it's harder than it seems (as was already pointed out) to identify a person this way. I mean, you're just guessing in cases where it isn't immediately obvious, which means your tool isn't very good. What is this young woman's ancenstry?

Yes, populations of people sometimes look very different. But sometimes they don't. And since there is usually a gradient, demarcating groups of humans this way isn't exactly the best tool you could use.

However, all of these things taken together DO give us a fairly good means of estimating what part of the world someone's ancestry is mostly from. Sometimes the information more reliable than at other times, of course, but with people from european vs asian vs african descent, it is generally pretty reliable. Obviously, the more "pure" their ethnic background, the more reliable it is.

They give us a crude and often unreliable method, to be sure. They just don't tell us very much about our species as a whole. Which is why the concept is more meaningful from a cultural perspective.

In any case, which traits do we pick? You have skin color and nose shape listed, but how to we tell the difference between, for example, people of West African heritage and those with East African just by looking at them? The racial concept is almost worthless in cases where the typical "giveaway" features vary more subtly. It's not just fuzzy, sometimes it is useless or even worse it's downright misleading.

Posted by: Leni | September 16, 2007 11:07 PM

Race is how one group tends to assert itself and classify outsiders. Remind me how the Anglo-Saxons are racially superior to the Celts, or how the Nordic races are superior to the Alpine races. Oh, sorry, we don't indulge in that sort of nonsense ay more. How about the relative superiority of the Punjabi warriors, over the plodding Bengali peasants. The British were great at this sort of stuff, classifying everybody ad coming up with reasons why one group would be better at task than another, then moving populations round, as labourers, administrators, etc. Hence we have Tamils in Sri Lanka, their cousins in Fiji, etc.
We differ in too many ways , most of which are hidden from view, but we classify people on their appearance, or at times on their ancestry even when their appearance gives no information. Octoroons could be slaves, because that small amount of blood made them inferior.
Humans have a lot of social malleability, but that often gets confounded with physical markers, so we blame race for what we see of SES. I reviewed a lot of Phil Rushton's work on IQ and brain size when it came out, and it was and is pernicious nonsense. We might expect group differences if we had different demands between groups, but one of the major roles of intelligence is to allow us to manipulate others and avoid being manipulated ourselves, a demand that occurs in all social situations.
Race' has a social reality, but correlates poorly with what we actually think of when we mean race.

Posted by: stewart | September 16, 2007 11:43 PM

You people should take it up with real scientists who are doing real race based research. Notice that Laden and PZ don't - because they can't.

Snark aside, that was an interesting article you referred us to. However, I don't think it definitively refutes MAJeff's and other's point in quite the way you seem to think it does.

p. 2:

Although risk factor associations [here, "race"] do not usually imply direct causal links (emphasis mine)

is a long way away from rob's assertion that race causes phenotype.

p. 2 again:

When direct causal factors are identified, risk estimates on both an individual and population basis can be made much more precise. Before such identification, however, the use of cruder surrogate factors (again, emphasis mine) can still provide valuable input for prevention and treatment decisions, even while acknowledging the latent heterogeneity within strata defined by such variables.

That sounds a lot like Leni's point that

They give us a crude and often unreliable method, to be sure.

As for that "latent heterogeneity within strata defined by such variables" (p. 3):

In summary, populations outside Africa derive from one or more migration events outside of Africa within the last 100,000 years. The greatest genetic variation occurs within Africans, with variation outside Africa representing a subset of African diversity or newly arisen variants.

In other words, "black Africans" can be more different from each other than they are from Caucasians or Asians or other "races". That's a fairly elastic definition of "difference", when within-group variation can regularly exceed between-group variation.

And it's reminiscent of Graculus' point that

How does skin colour mean anything of that sort when the ancestors of some groups with black skin left Africa before the ancestors of the so-called Caucasians?

Still on p. 3:

The terms race, ethnicity and ancestry are often used interchangeably, but some have also drawn distinctions. For the purpose of this article, we define racial groups on the basis of the primary continent of origin...The continental definitions of race and ancestry need some modification, because it is clear that migrations have blurred the strict continental boundaries.

So they have to modify the commonly-accepted definition of "race" to make it even crudely useful.

p. 6:

The persistence of genetic differentiation among these US racial groups (as defined by the US Census) has also been verified recently in a study of nearly 4,000 SNPs in 313 genes. These authors found distinct clusters for Caucasian Americans, African Americans and Asian Americans; the Hispanic Americans did not form a separate cluster but were either grouped with Caucasians or not easily classified.

In other words, race works as a classification, except when it doesn't.

pp. 6-7:

Consider the group they labeled Caucasian, consisting of Norwegians, Ashkenazi Jews and Armenians. Their genetic cluster analysis lumped these three populations together into a single (Caucasian) cluster/ Yet numerous genetic studies of these groups have shown them to differ in allele frequencies for a variety of loci. For example, the hemochromatosis gene C282Y has a frequency of less than 1% in Armenians and Ashkenazi Jews but of 8% in Norwegians. Thus, in this case, self-defined ethnicity provides greater discriminatory power than the single genotype cluster obtained by Wilson et al.

In other words, race works as a classification, except when it doesn't.

p. 7:

The true complication is due to the fact that racial and ethnic groups differ from each other on a variety of social, cultural, behavioral and environmental variables as well as gene frequencies. leading to confounding between genetic and environmental risk factors in an ethnically heterogeneous study. (emphasis mine).

So even differences in outcomes among races can be environmental rather than genetic.

Continuing on p. 7:

For example, with respect to treatment response, "An individual's response to a drug depends on a host of factors, including overall health, lifestyle, support system, education and socioeconomic status - all of which are difficult to control for and likely to be affected, at least in the United States, by a person's 'race'".

Which I believe was Jeff's point to begin with, and the antithesis of rob's, about the direction of causality.

Thank you for an interesting article, but I don't think it provides quite the devastating critique of PZ, Greg, MAJeff, and others that you seem to think it does.

Posted by: thalarctos | September 17, 2007 12:39 AM

Re: #55: Said Stewart, "We might expect group differences if we had different demands between groups"

Like maybe vastly different climates? Or are the environments of, say, Norway and Somoa socially constructed, too?

"Race' has a social reality, but correlates poorly with what we actually think of when we mean race"

That's completely wrong, read the link provided in comment #54.

Posted by: James T. Smith | September 17, 2007 12:44 AM

thalarctos wrote:

In other words, race works as a classification, except when it doesn't.

Those were pretty much the same thoughts I had when I read those parts.

I'm not opposed to looking at groups of people based on race or anything, so I let it go. It's ok if it doesn't work in every single instance, if they can learn something useful and do it in a way that doesn't neglect large swathes of people because of the color of their skin, I say go for it.

But it looks like an absolute minefield at first blush, and I can't help but be suspicious that it's all going to end up as a marketing ploy for cholesterol medication.

Posted by: Leni | September 17, 2007 1:01 AM

James T. Smith wrote:

Yes. If you can sort people genetically into related groups then race exists. And of course you can, and these genetic sortings correlate with percieved racial categories.

Ok. Except no one said race didn't exist or that it didn't have a biological basis. I merely pointed out that the concepts of race that we use in everyday life don't have nearly as much genetic significance as we seem to give it. MA Jeff points out that it is culturally far more significant, and I don't think your article refutes that.

One is far more likely to encounter social and cultural issues regarding race (particularly in this country) than they are health ones. I imagine. I admit that is a guess, but it seems like a plausible one to me. Course, that isn't to say there are no race-related health issues. But calling them race related seems to emphasize their connection to racial traits in a way that is probably misleading.

It's genetic, and only incidentally racial because those characteristics are also genetic and an easy way to pick out groups of different people. Emphasizing race just seems odd and implies a kind of causality that doesn't really exist, and it doesn't address variation within the group. (I don't think that article was suggesting race causes illness or anything- it just seems overly broad and arbitrary.)

Posted by: Leni | September 17, 2007 1:29 AM

rob,

And how is that [race] different from any other sort of classification, such as species, subspecies, or whatever?

Sometimes those classifications are inappropriate as well. "Species" doesn't mean much for a lot of asexual organisms. "Subspecies," which is pretty close to "race," doesn't work well if the characters you're interested vary clinally rather than discontinuously, and/or have very different patterns of distribution.

It's not that humans couldn't in principle have well-defined races or subspecies--but they seem to be among the species which don't.

Posted by: Anton Mates | September 17, 2007 2:29 AM

Anton said, "Filipinos and Indonesians often don't have it; on the other hand, Bushmen and Lapps do have it.

That's one of the problems with the race concept--the characters used to define it don't actually line up with one another geographically."

If you look at single traits. But if you look at the correlation between sets of traits, then you can sort people into racial groups. It's why despite having eye folds, no one would ever mistake a Bushman for a Lapp. Unsurprisingly, the genetic distance between Bushman and Lapp can also be determined, and percieved genetic backgrounds do line up geographically. Denying the reality of race is putting politics and ideology over science, nothing more.

Posted by: James T. Smith | September 17, 2007 2:43 AM

The sociological approach exemplified (from #55 above)

'Race' has a social reality, but correlates poorly with what we actually think of when we mean race