Insisting that "races are real" is a self fulfilling and overt racist act. So stop it now, please.

Greg

I have to commend you -- I say this all the time - yet my friends don't seem to understand that using phrases like "driving while Asian" is racist!

They exhibit confirmation bias (a bad 'white' driver is a bad driver, a bad 'asian' driver is an 'asian driver:: all asian drivers are bad')!

Thanks for providing the link - I'm sharing it with lots of people!

I couldn't agree more. I noticed something odd happen after we had kids, as our family is multiracial. I started having trouble identifying people racially. It was the weirdest thing. I'd be looking at someone and I could identify the person individually but damned if I could remember what the race was supposed to be. I thought it was odd and silly and never mentioned it to anyone until I heard about that study. I think it's a brain thing.

I wonder when Ethylene will be chiming in to say that geographic boundaries cause races?

Yes, this is an interesting study. I also liked the part where they put the volunteers through 10 hours of "training," in which the study subjects were asked to consider personal aspects of the people in the photographs, such as what they like to eat. That also reduced the racial bias.
I took it as a reminder to bridge any gaps I can, whenever I can, between myself and "other," whether another race, viewpoint on religion, culture, whatever. You should see all the comments over on Pharyngula's blog right now, dehumanizing Christians. Scary how we do that. Happens in war, too.
We could have an amnesty day where we reach out to people who are different from us ... Hey! That reminds me! It's blog amnesty day, where you're supposed to link to a blog that gets less traffic than yours, and I have just the place: anneminard.com, which contains my experimental new blog called "100 Days of Science" ... Consider sending some new readers my way!

This makes sense, you're going to learn to identify the data you train on. I've always suspected the largest reason we have trouble telling people of other races apart is a) we have less experience judging the relevant features so we just latch onto the obvious features, and b) rather than learning the nuances we start off with the obvious, for a minority their minority status is memorable, for someone with a big beard we latch onto that.

I've noticed that whenever I become fairly familiar with an individual I tend to forget their racial category. I suspect that has to do with my brain learning to pick out the more individual features and throwing away the obvious but less individualized racial characteristics.

I'm not sure how to use this to reduce racism other than finding ways to mix races more. If you have a group with 5 Caucasians, 1 Asian, and 1 Black person you're going to recognize the race for the latter two because that's the easiest way to identify them. Maybe the key is to group minorities together. TV shows like to have one or two minority members, maybe instead of making those separate minorities they make it multiple members of a single minority and force the viewers to discriminate on the individual features rather than the race. (of course the goal of TV shows is just to promote equality, not actually cause it).

I have pretty much, on my own "trained" myself out of identifying people "racially". For one thing, it's often impossible to do this, and for another, to me, it doesn't matter all that much. The problem comes into play in that there are a lot of people to whom it does matter, and then I have to go toe to toe with them, and explain, for example, that there is a ton of diversity, among, say, sub-Saharan Africans, and I've seen much of this with my own eyes. Frankly, I don't really understand why all of this is so important to some people.
Anne G

Anne wrote:

Pharyngula's blog right now, dehumanizing Christians. Scary how we do that. Happens in war, too.

Yes, those comments were truly an orgy of blood. I trust the spleen-venting you've indulged in there has repaired your delicate constitution.

While there certainly aren't any genes that belong to only and all members of a given race, there is obviously some geographic clustering of traits. If you meet someone with several obvious traits common to a given region, it's likely they'd have other less visible traits common to the same region. It still leaves you with a fairly nebulous definition of race. Individual variation is still far more interesting.

Can't wait to see my wife's response to this, I used to rail at her all the time for slipping into the "they all look the same" trap. She didn't mean it, she really didn't, but it annoyed the hell out of me whenever it happened. I'll get no small amount of pleasure from rubbing her face in it with this :-)

Funny story: I once mentioned the increasing diversity of my home town (Edinburgh) to some "Afro-Caribbean" friends from London. They immediately asked whether the new arrivals were mainly Caribbean or African. When I replied that I couldn't tell the difference just by looking, I got the riposte "So we all look the same to you, do we?"

I was gobsmacked. ;)

Greg,
I just don't get this "race doesn't exist schtick." I meant to comment on this post of yours a while back about it, but didn't have the time then. I'll get to commenting on that in a moment.

For this post, I have no disagreement with the political, sociological and pedagogical problems with discussing race. I especially agree with the interpretations of on cognitive bias. But to put up a title like this post is not based on anthropology in the slightest, but on some kind of ideology.

But to the basis of your schtick. Back in November you (rightly) described the term "race" in the context of biology:

A species that can be divided into races demonstrates two additional features without which the term 'race' is really not applicable. First, there are usually distinct boundaries between the races. These boundaries must apply to the features that are used to define the racial differences (which may themselves vary by race ... the different races need not be distinguished by variations in the same exact set of features) and we do not expect the boundaries to be perfect. Indeed, among the members of a given species, some of the boundaries may be outright vague. But if there are no or almost no clear boundaries, there are no races.

The second feature that technically need not be true but almost always applies is that there should be some correlation between traits. In other words, if you take one trait and use it to define races, then there should be another trait that, if races are defined on its basis, you get a very similar set of races. And there should typically be several such traits that all, if analyzed independently, define almost the same exact set of races. An exception to this is where races can be arguably defined on the basis of a single trait. Were this to be the case, then obviously there could be no correlation.

Now, in the past few hundred years, the reality of these criteria has been breaking down. (that's great politically, but here I'm just stating a fact - massive human migrations are taking place more frequently in recent centuries).

Prior to that however, we had little travel between continents. Human populations on those continents were separated for 10,000 years or more in many cases. Clear morphological characteristics did arise that distinguish people on these continents, which incidentally, are what most people's bias-machinery focuses around with this "Other-Race Effect" concept.

So race existed, even by your own definition. (I say kudos to gene flow though.)

There is no physical evidence to support your assertion that prior to some point in time in the past that well defined races existed and have since gone away. Indeed, the physical evidence suggests the opposite in some cases.

This "races used to be firm and real concepts" trope is not really new, but its increased use in this discussion is increasing. I'm not sure why.

My statement here about the irrelevancy of race and the sociological damage the race concept causes is not ideological. But I do wonder about the insistence that race is a real, valid and viable concepts in humans as possibly ideological. I tend to think this whenever I see the premise to an argument as: "Your version is ideological/pc/political, mine is based on science" followed by an incorrect pseudoscientific statement or a simple error in fact or wacky assumption.

Prior to that however, we had little travel between continents. Human populations on those continents were separated for 10,000 years or more in many cases.

A highly questionable assertion, unless you regard the whole of Europe, Asia and Africa as one continent. The language we're currently using is part of the Indo-European group.

This "races used to be firm and real concepts" trope is not really new, but its increased use in this discussion is increasing. I'm not sure why.

I would have thought it was obvious - people are psychologically invested in the concept of race, but it's becoming increasingly obvious that it's bullshit, so they have to try and salvage it somehow. Making stuff up about periods of history you know sod-all about (and assume that nobody else knows any better) is one obvious strategy.

There is no physical evidence to support your assertion that prior to some point in time in the past that well defined races existed and have since gone away.

So at no point has it ever been non-difficult to distinguish asians from europeans, or either from native africans? Who are you kidding?

My statement here about the irrelevancy of race and the sociological damage the race concept causes is not ideological.

I never said that that race was relevant to sociology, in fact I agreed quite explicitly that it is harmful in the sociological context. What I said, if you care to reread my first comment, was that claiming that race is irrelevant to anthropology was ideologically based.

Let me clarify what I'm suggesting very clearly - claiming that race doesn't exist in the context of anthropology and population migrations over the past 50,000 years appears to be an ideologically-based extension of (well-founded) criticisms of misuse of the term "race" in sociology.

Dunc,

A highly questionable assertion, unless you regard the whole of Europe, Asia and Africa as one continent. The language we're currently using is part of the Indo-European group.

I'm referring to the geographic separation between the continents. For instance, in the case of Europe and Asia, the big separator being of course the Himalayas. In the case of Africa, you have sub-Saharan populations separated from the North even, by different geographical features.

I'm referring to the geographic separation between the continents. For instance, in the case of Europe and Asia, the big separator being of course the Himalayas. In the case of Africa, you have sub-Saharan populations separated from the North even, by different geographical features.

Nah, come on. The Himalayas can be got round, which is why the South Asian sub-continent has been invaded from the north west every hundred years or so since records began. Go tell the Mughal Babur from Samarkand they're the big separator. Or you can go north and come down into China one way and Iran the other. Like Genghiz.

The separation of sub-Saharan populations in Africa is always exaggerated too. Take the coast road; walk down the Nile; follow the trade route from Morocco to Timbuktu. If, as I suspect, you're of recent European origin, I'd bet you a comfortable sum that if you walked into a room full of Ethiopians and Yemenis, you wouldn't confidently identify which people were which until they started talking to you.

All look the same, see. But some of them are Africans and some of them are Arabs.

Okay, I'll take all comers. ;-)

Chris,
The Himalayas can be got around, but they just weren't often enough to allow for more than a trickle of gene flow at best.

If, as I suspect, you're of recent European origin, I'd bet you a comfortable sum that if you walked into a room full of Ethiopians and Yemenis, you wouldn't confidently identify which people were which until they started talking to you.

Maybe, maybe not. What does that rather badly chosen example prove? That people close enough to have some degree of gene flow between them in the past might be more similar than, say, people from Tanzania and Iran (different ends of the same respective regions).

That would confirm exactly what I'm talking about.

So at no point has it ever been non-difficult to distinguish asians from europeans, or either from native africans? Who are you kidding?

Seriously. I'm not kidding.

Dan, regarding the continents, the Himalayas do not separate Asia from Europe. Northern and "sub saharan" Africa are not separated by anything. (The Sahara is a relatively ephemeral feature, comes and goes, and for much of the last several tens of thousands of years was a well watered grassland).

How it works:

http://xkcd.com/385/

For instance, in the case of Europe and Asia, the big separator being of course the Himalayas.

The Himalayas separate Europe from Asia? Have you ever even seen a map?

Greg,
How about addressing the main substance of my comments - that of gene flow versus gene drift - and the geographical context that existed up until 500 years ago.

Dunc, (and Greg),
Okay, after your last comment Dunc, I see that neither of you get the Himalaya bit. Let me explain.

I'm not contesting that there has, for a long time, been gene flow between the various Indo-European groups. And these extend into Asia. Most people generally group these people into the "Caucasian race". So when I said there hasn't been much gene flow up until 500 years ago between Europe and Asia, I was talking about the Far East of Asia.

Sheesh, I'd have thought that was obvious.

No, it is very much not obvious. It only looks obvious if one already agrees with your premise on the subject of race and interprets the geography in a completely unconventional way to suit that argument. I don't group anybody into the "Caucasian race", because I don't believe the term has any real meaning. When I use the terms "Europe" and "Asia", I'm using them in the conventional geographic sense.

Anyway, the Himalayas don't separate the Far East from the rest of Asia either. The "Indo" part of the term "Indo-European" refers to the Indian subcontinent - you know, the place that actually is on the other side of the Himalayas from Europe.

From Dan

The second feature that technically need not be true but almost always applies is that there should be some correlation between traits. In other words, if you take one trait and use it to define races, then there should be another trait that, if races are defined on its basis, you get a very similar set of races. And there should typically be several such traits that all, if analyzed independently, define almost the same exact set of races.

Then we all are, simply, waiting for you to just identify the races and the traits that define them.

Those who push race, as a science, always end up saying a version of this "You already know the races and what separates them, so I don't have to objectively define anything..." if they say anything at all.

Please, if racial groupings are real just give us something to work with.

Dunc,
You really should read Cavalli-Sforza's The History and Geography of Human Genes.

Actually, has anyone here read Cavalli-Sforza's book, or one along those lines? From the various responses, I'd say no. Go check it out, there is plenty there for you to work with (tif, I'm looking at you).

Dan, I'm asking you to intelligently defend your position. You are choosing to not do so.

I'll give you another try... Perhaps if there is something of relevance to our discussion in Cavalli-Sforza's book you can summarize it right now?

The same Cavalli-Sforza who said ""The classification into races has proved to be a futile exercise", and specifically the book in which the sixth section of the very first chapter is titled "Scientific failure of the concept of human races"? I'd love to, when I can find the time. However, I suspect you might need to read it again (assuming you have actually read it already), and perhaps a little more carefully this time.

Tif,
Sounds like someone needs to cool off. Come back and we'll talk the finer points of gene flow and drift, generally or within specific context of the human species, when you're not offended by my ideas challenging your sensibilities.

Third time...

Dan, please just defend your position. Your remarks about my demeanor is not an intellectual defense. However, I think you already know this.

Dunc,
The same. Oh, I didn't say that classification is easy, or even possible in the final analysis, for some circumstances. You'll always have "lumpers" and "splitters" when it comes to classification.

Cavalli-Sforza himself has also said things (which I agree with) to the effect that it is arbitrary at many levels to separate groups, based on tribe, allied tribes, etc., up on to regional groups. Which group do you cut it off at and say, "that grouping is significantly separated to call it a separate race"? Well that's a very fishy thing to do, I agree.

However, this is Cavalli-Sforza's description of the map that is the capstone of his half century of labor in human genetics:

The color map of the world shows very distinctly the differences that we know exist among the continents: Africans (yellow), Caucasoids (green), Mongoloids ⦠(purple), and Australian Aborigines (red). The map does not show well the strong Caucasoid component in northern Africa, but it does show the unity of the other Caucasoids from Europe, and in West, South, and much of Central Asia.

"Very distinct differences."

I think the resulting diversity is a beautiful thing, and from a biologist's perspective, reflects gene flow and drift amazingly.

tif,
No, but my remarks about your demeanor do explain why I do not take you seriously. Let Dunc or Greg speak for you, they're not thinking like a guy with road rage.

(1) Your ideology is blinding, but I see everything clearly. 6:08 a.m.
(2) How about you address what I actually said. 8:28 a.m.
(3) W00t! Look at how unpopular I am. + Your single exposure of the gaping holes in my argument is not enough to defeat my sweeping assertions. 9:03 a.m.
(4) No, no. Talk about what I want you to talk about. Ignore everything else. 9:23 a.m.
(5) If you don't agree with me, you're just not getting it. 9:27 a.m.
(6) Homework assignment. 9:50 a.m.
(7) Keep the tone civil, people, no matter how much I insult you. 10:08 a.m. and 10:21 a.m.
(8) Quote mangling. 10:19 a.m.

Come on, someone's got to have a bingo by now!

10:21: Triangulating among assumed allies.

BINGO!!!!

Dan - the quote you present does not actually contain the words you subsequently highlight in quotes. "[S]hows very distinctly the differences ..." does not equal "shows very distinct differences". Nobody here is arguing that there aren't any differences in gene distribution across various geographies, we're just arguing that (a) those differences do not add up to the existence of "races"; and (b) those differences do not map well onto "races" as they are commonly conceived.

However, the quote you have provided does appear to refute your earlier statements to the effect that the populations of various continents (specifically Europe and Asia) were effectively geographically segregated until recent times.

Greg,
What can I say? I guess you and your honchos know how to bully a guy. Way to go.

From tif

Those who push race, as a science, always end up saying a version of this "You already know the races and what separates them, so I don't have to objectively define anything..." if they say anything at all.

Dan, it's clear to me that you are trying to win a discussion by disqualifying your detractors as unworthy (on a science blog no less) but remember that I won't be the only one to ask you this question.

So maybe you need to re-evaluate why you have beliefs you are are unable to quantify or defend. Ciao.

Damn. "You're all a bunch of meanies for arguing effectively with me" was going to put me over the top, too. Congratulations, Greg.

I guess you and your honchos know how to bully a guy.

Quick, somebody call a waaahmbulance!

Listen, I know bullying, and you're not being bullied. You're being asked to defend your arguments in a cogent manner. Not. The. Same. Thing.

Dan: Let me explain. This is all in reference to a recently posted analysis of trolling behavior. Somehow you managed to make yourself a poster-boy for that post, like you were following instructions almost.

Take this as a learning opportunity and you'll be fine. See this and this.

Oh, wait, a glimmer of substantive conversation from Dunc, making good points. Thank you Dunc for making good comments, even if the host of the blog is unwilling:

Nobody here is arguing that there aren't any differences in gene distribution across various geographies, we're just arguing that (a) those differences do not add up to the existence of "races"; and (b) those differences do not map well onto "races" as they are commonly conceived.

This is a vexing point. While we agree on differences in gene distribution across various geographies, we clearly set the boundary of what is enough to suggest that races exist at different points. I think it's asinine of Greg to call me a racist (well not me personally, but see the title of the post) for suggesting "look at all that variation, there must be a lot of genetic drift behind it, and I'll loosely call those most distinct groups races."

As you've noted, it's not easy to determine exactly where to draw boundaries. (Biology is never clear cut, as I'm sure all here know).

I especially like point (a), because it makes me wonder, well, what do the differences between peoples (that we agree on) add up to, if not races? Ethnicities, perhaps, but I wouldn't know how to graph that term onto a discussion of human biology. Any other ideas?

Or maybe you just wouldn't call it anything?? Why not? "[D]ifferences in gene distribution across various geographies" is a bit heavy of a phrase to toss around when talking about the conventional "race" term.

Greg,
Oh.

Sorry. Point taken. (but I really AM interested in continuing this latest bit with Dunc!)

Dan: This is your problem. Watch carefully. This is not a magic trick ...

"I especially like point (a), because it makes me wonder, well, what do the differences between peoples (that we agree on) add up to, if not races?"

Peoples? What peoples? You have started with the assumption of 'peoples' (= groups, ethnicities, etc.). THAT is the invalid starting point, yet you insist on it. You've even invoked some pretty strange geography to make the argument that 'peoples' existed as distinct entities (races) before the magic date of 500 years ago.

Stop thinking boxes and start thinking clines. Many people argue that race is an appropriate shorthand for clines, but the point of this post is that it is not a cost-free term. It has a cost that white guys like you don't have to pay. So stop making others pay for it.

Oh, wait, a glimmer of substantive conversation from Dunc

A "glimmer of substantive conversation"!? That's it - I'm stopping now before I really do start bullying you. I'll let Greg deal with this from here on in, it's his specialism.

"Glimmer of substantive conversation" indeed...

Specialism? Are you calling me a Specialist?!?!!??

Ah, sorry, I take it back. ;)

Dan:

I'm mostly here to offer you moral support as I've been in a similar position to you on this blog :)

I don't agree with you entirely, I think you significantly underestimate the degree of variation within the generally defined races. My opinion is there are a lot of different subpopulations that roughly correspond to historical geographical and/or maybe political boundaries. I'm not sure exactly what these boundaries were, or how sharp they were, but they certainly existed as there are some clear divisions.

I also think that the broad characteristic of skin colour has been used historically to define races but it's a little deceptive as there are significant subpopulation variations within groups with similar skin colour, Africa is probably the best example of this. But at least in some cases different skin colour is an indicator of reduced gene flow and there likely are some other non melanin related genes that differ as well. What they are and what significance they have I don't know but they likely exist.

On the greater meta-topic (which I reluctantly bring up since I don't really want this to degenerate). I'm always reluctant to disagree with Greg since he is a domain expert and obviously has a lot more knowledge in these areas. However, it does seem to me that Dan is correct is feeling persecuted. Greg's post is titled with an assertion that anyone disagreeing with the premise is a racist and he hints that Dan is racist in two latter posts (really, how can you impartially evaluate a position once you claim it's racist/sexist/evil?). Dunc has concentrated more on nitpicking Dan's argument rather than attacking it directly, and as for any degeneration in the level of discussion I feel this is due far more to Stephanie's and Dunc's discourse than Dan.

From what I've read everyone here is a good person who wants to find the truth, so why not raise the level of dialog rather than pushing people into a corner so they feel they have to lash out?

If you focus on categorizing people into races at the expense of recognizing variation within these alleged racial groups, you will a) get 'good' at categorizing races, b) get bad at recognizing individual differences within the "other" races (other = not you), and c) become more racist.

Why is this mutually exclusive? Individuals of other backgrounds, are unique, but their background is also part of their identity.

Why does one need to deny the the obvious to be 'correct'? People have characteristics unique to them as individuals, we also have characteristics that give a hint, if not more, about our ancestry.

Racism (mistreating people based on their presumed origin) is evil. Accepting the the human species has a lot of variation, some of it regional, is not.

False premise: "I think you significantly underestimate the degree of variation within the generally defined races."

Unsupported pseudofactoids: "My opinion is there are a lot of different subpopulations that roughly correspond to historical geographical and/or maybe political boundaries. I'm not sure exactly what these boundaries were, or how sharp they were, but they certainly existed as there are some clear divisions."

Rabbit being pulled out of hat: "skin colour has been used historically to define races but it's a little deceptive as there are significant subpopulation variations within groups with similar skin colour, Africa is probably the best example of this. But at least in some cases different skin colour is an indicator of reduced gene flow and there likely are some other non melanin related genes that differ as well. What they are and what significance they have I don't know but they likely exist."

Inverse reasoning: "From what I've read everyone here is a good person who wants to find the truth, so why not raise the level of dialog rather than pushing people into a corner so they feel they have to lash out?"

I have a feeling I'm reading a blog. Looks like a blog. Smells like a blog. Must be a blog.

jay: The statement you quote is merely the conclusion of the paper, not a random assertion.

Aaron: Perfectly nice people who want to pursue the truth can be and often are racist, and for the most part, would prefer not to be. My assertion here, and I've made this elsewhere, is simply that race is a crappy, diffuclt to support and washed out way to characterize humans. The argument that it is a handy shortcut or partly true so why not use it, etc., is a poor argument given that the use of the race concept is not cost free.

Yatzee: You're being a Snark, but I essentially agree. These assertions are all the same thing we've seen before. "I believe this thing is real so I'm going to say it is real over and over again, in a number of different ways."

Cold is not a thing. A vacuum is not a force. The human species is not neatly divisible into races. Become accustom to the realities revealed to us by science, grasshopper.

You're being a Snark

my pleasure

Dunc has concentrated more on nitpicking Dan's argument rather than attacking it directly

What, pointing out that the primary source cited in defence of an argument explicitly rejects that argument is "nitpicking" now?

Greg,
With all my talk of gene flow and drift, I actually think that I'm describing something in between boxes and clines - boxes describe some drift that may occur over time, and clines representing flow between populations. The two reflect in a nice conceptual way the full range of diversity in humans.

But this is the really asinine thing: you insist that a white guy who might talk about an ethnic heritage other than his own with admiration, as I might, is racist. That's just f'ed up. It stifles open discussion. As jay says:

Racism (mistreating people based on their presumed origin) is evil. Accepting the the human species has a lot of variation, some of it regional, is not.

Be an intellectual, Greg, and don't throw the term "racist" around so foolishly.

Dan: You have lost your grip on the argument. Go back and review. Try to explain, maybe just to yourself, what the argument you think I'm making is. Also, work on your definition of racist. In order to understand what this is, you need to reconcile this fact: It is possible to have a racist argument that is not evil.

Greg,
To quote the title of your post, I think that you're saying that "Insisting that 'races are real' is a self-fulfilling and overt racist act."

Yep, asinine.

Be an intellectual, Greg, and don't throw the term "racist" around so foolishly.

Concern trolling.

Dan: I need you to understand that insisting on something that is a weak construct at best but that has a negative social implication, which you are then imposing on others with no direct cost to yourself, is indeed the asinine thing. I'm telling you to stop doing it. Ignore me if you like, but it makes you an ass. Don't be an ass. Just do what I say, quietly, and appreciate the lesson.

(Don't you hate when people talk to you that way? I do. But it gives the kind of sting I want you to feel so you have a vague idea of the pain you cause in others by maintaining the racist trope.)

Yatzee:

I made broad general statements because that's the data I have. But I challenge you to contradict those broad general statements or the derived conclusions rather than attacking me for bringing out figures when they're not necessary.

Greg: You rightly mention the cost of throwing around labels of races. But what about throwing around the label of racist? How do you expect someone to react rationally when you label them as racist?

"It is possible to have a racist argument that is not evil."

True but that doesn't circumvent the emotional charge of the word.

Yes, Aaron. It's so terribly rude of me to point out when people are obstructing the conversation they purport to be wanting to have, as when I suggest that you'd rather worry about whether someone gets their privilege bruised by use of the word "racist" than explore what the word actually means. Ever so sorry.

Even if the human population is clinal, the allopatric impact is slight, since there is no genetic isolation at the extremes of the cline. That in itself refutes Dan's assertion that gene flow is less than genetic drift (founder effect on Pitcairn Island aside).

Mike

Stephanie:

Your only contribution to the conversation was to mock Dan, do you really expect that to bring him around to your side of the argument?

And nothing in my argument suggested you don't use racist for fear of making someone feel bad, I suggested you don't use it because that devolves the level of the conversation. If someone calls you a name are you really motivated to respond with "ah yes, I see you are correct". The fact is you're not concerned with convincing Dan of anything but merely with attacking him.

In related news, NPR had a segment on this week on professor Nina Jablonski's work (head of the Penn State Department of Anthropology) on skin color. She says that for many families on the planet, if we look back only 100 or 200 generations (that's as few as 2,500 years), "almost all of us were in a different place and we had a different color."

"Our original estimates were that [skin color changes] occurred perhaps at a more stately pace," Jablonski says. But now they're finding that a population can be one color (light or dark) and 100 generations later â with no intermarriage â be a very different color.

It's "a blink of an eye," she says.

(quotes from NPR's web page)

NPR reporting is my level of understanding of the topic, so I usually just lurk.

Aaron: The first step in combating racism is to recognize racism. Please don't tell me Dan is not a racist. Please don't tell me you are not a racist. Let's get real. Racism is pervasive in our society. Nobody escapes it, but we can all reduce it.

Aaron, no, I was not interested in trying to convince Dan of anything. However, that was because he was throwing off every indication that he was completely uninterested in engaging with anyone who was trying to convince him. I thought this was useful information for anyone who was trying to engage him. As you can see from the thread, two other people independently came to the same conclusion and decided to disengage on their own.

As for mocking, I mock. You scold. Is there a particular reason one should be appropriate and the other not?

Still, if you prefer scolding: knock it the hell off with trying to insist that discussions of racism be polite and intellectual. There is nothing polite and intellectual about the topic or its effects. Claiming that only people who can approach the discussion in a calm, detached matter should be allowed to participate is a recipe for excluding everyone to whom the topic makes a difference. It is a classic maneuver of those who have a vested interest in maintaining the ugly, hurtful status quo. Don't be that person.

So you are saying you yourself are a racist, Dr. Laden.

Yatzee,
It would appear so. Yay, we're all racists!

No one is exempt. Not even you, Mr. Dice Man.

Stephanie:

It's hard to disengage but the fact remains if you want to have a rational discussion and reach proper conclusions emotions never help. You can't put someone on the defensive then complain that they won't listen to reason.

Greg:

Isn't there a better term? I, and I think most other people, really perceive racism as having an active desire to look down on other races and I don't think that's what you're describing.

"...I actually think that I'm describing something in between boxes and clines..." -- Dan

Isn't that like saying there is something in between digital and analogue?

"In parallel, other-race faces are more difficult to differentiate relative to own-race faces" on a differant take...

The article did make me think how hard it is for me read some "other-race faces" unless I'm quite close. Very dark faces are really difficult this way, my being pink and all...

And yet, Aaron, you eventually decided that the people who were pissed off were the people who had their facts straight. In other words, nice theory, but it doesn't work that way just because you want it to. Human impact is a real and important factor in decision-making. The fact that you have a bias against it doesn't change that.

In my admittedly unscientific opinion, trying to classify people into the three races I was told existed while I was going to school in the '60s is a fool's game. I've always wondered, for instance, where you'd place Polynesians within those groupings. To me, they have features of all the so-called races.

Africa has by far the most genetic diversity. How you could even claim there was an "African" race is beyond me.

There are certainly ethnic groups, with individual traits that are prevalent within a population, but races? I think you'd spend more time trying to shoehorn individual groups of people into those categories than is worth it.

Stephanie:

I naturally think I'm right, as do you and everyone else here, virtually no one intentionally holds a wrong opinion. The difference is civility. I'm sorry if I come across as scolding but that is not my intention, my intention is to effectively communicate what I feel the issues are, if you have a suggestion as to how I can improve in that regard I'm all ears (well eyes I guess).

My meta-discussion had very little do to with who was right vs. who was wrong, it has to do with respecting your opponent. If you in fact believe you are right it's advantageous to keep respect because then your ideas can win on their own merits. Dan struck me as someone who if they're wrong, is wrong because of poor facts, not ideology. If you attack him on ideology and emotion who miss him argument and merely alienate him.

Greg:

Just a curious side-note. Does your racism definition include only people who assume genetic differences or people who assume only cultural/sociological differences as well? Additionally, would you classify someone as racist who believed in races but didn't assign specific characteristics to those races (ie, just used them as arbitrary identifiers but didn't think they implied anything positive or negative)?

I would say that race is a fuzzy classifier, in the sense the real existence of "race" is in the manner it is used by people to subdivide people to groups.

Race does not create strict equivalent classes, as interracial reproductions happens all the time, leading to people (such as our current President) who exist at the "border" of two or several races.

The phrase "alleged racial groups" seems strange to me. I would suggest that if you got took 100 ethnically Japanese men, 100 ethnically Indian men, 100 ethnically West African men, and 100 ethnically Irish men, your typical person could easily sort the people into four groups based on "alleged race" without difficulty.

The problem is not that races don't "exist", but rather that they are associated with divisive and perjorative stereotyping and discrimination. And while I understand that there are racists out there who are quite eager to make the jump from "racial divisions exist" to "the existence of racial divisions support my social agenda X", I don't see how anything is gained by pretending racial divisions do not exist.

Or, more to the point, from an information theory standpoint, I don't see the purpose here. It seems like an exercise in throwing away information.

The bit in the paper about being able to learn how to recognize people of different races (or, as you put it, "alleged races"), is very interesting.

Anne wrote: "I took it as a reminder to bridge any gaps I can, whenever I can, between myself and "other," whether another race, viewpoint on religion, culture, whatever. You should see all the comments over on Pharyngula's blog right now, dehumanizing Christians. Scary how we do that. Happens in war, too."

This is politically-correct nonsense.

Racism and sexism (and any sort of class bias, really) are "bad" because the statements that constitute them have no universal validity (think "blacks are . . ." or "gays are . . ."). Hatred of Christianity as an institution, however, has quite a lot of validity. Historically, morally, etc. it is one of the worst institutions ever.

Are you seriously claiming that you work towards "bridg-[ing] any gaps I can, whenever I can, between myself and 'other,' whether another race, viewpoint on religion, culture, whatever"? In my terms, you try to elide class differences of *all* sorts; all differences *whatever*? So you try to bridge the (presumed) "gap" between yourself and, say, torturers? How about serial rapists and pedophiles?

Distinction (criticism, differentiation) is not the same thing as discrimination (bigotry, "dehumanization"). In fact, one of the greatest powers humans have is the power to distinguish--our very ability to exercise moral judgement depends on it.

And Pharyngula's P. Z. Myers never--to my knowledge--"dehumanizes Christians" as a class; on the contrary, he demonstrates 2 things (nearly daily):

--that Christianity consistently brings out the worst in many *individuals*, individuals whom he specifies.
--that Christianity is not intelligent or decent as an *institution*.

These are truisms, as far as I can see.

"...I actually think that I'm describing something in between boxes and clines..." -- Dan

Isn't that like saying there is something in between digital and analogue?

When, in fact, there is. Digital is not clear cut on/off, but it really works out to the "on" signal being in the range of a certain value and the "off" signal being near another certain value.

Analog is not so smooth at people imagine. Everything you see comes from the relfection of photons, which are discreet bits of light.

This is a good analogy for the discussion, because "race" is not an effective classification for each individual. Many people do not fit exactly into one race or other. At the same time, as RickD notes above, we can clearly see that people have different appearances, and the source of these differences is the good old genetic drift.

If an American racist wanted to categorize my kids, she would classify them as being closest to the Japanese in RickD's groups above. If a Japanese racist wanted to categorize my kids, she would group them with the Irish. But someone who has learned to recognize people with both "asian" and "caucasian" features would immediately see the genetic manifistations of my kids' ancestory.

To say that people will or must behave a certain way, that they will have a particular behavioral destiny, because of their skin tone, hair crinkle and nose size is easily disproven by those from the morphological group who behave differently. But to deny that genetics has had the exact same effects on human features as it has on those of other animals is also absurd. So if a scientist feels the need to group people abased on their appearance and geographical origins, what term should she use, if not "race"?

I'm still in the camp that weighs environment, particularly social training, as having such a big effect on behavior that the genetic components of behavior are mostly negligible. There is no "shy gene", nor is there a "crime gene" nor even a "white man's burden gene."

Erik A: Spot on, man, spot on. Way to knock down Anne's touchy-feely crap.

(Of course, it's possible that Anne was just writing anything she could think of that was vaguely on topic so she could slip in the plug for her own blog.)

Or, it coul be that Erik A also has his head in his posterior and overgeneralized Ane's position.

I don't know any religionists, personally, whom I'd classify as an enemy. If I were as whacked-out against religion as many of the posters I see on SciBlogs, I would have a lot fewer friends than I do.

Gaps are worth bridging, to pretend that Anne means she'd be friendly with ax-murderers just because she said she'd "bridge any gaps" she could is pure straw-man illogic. It's that kind of argmentation where you say, "I can use unfounded assertions, but YOU have to qualify everything you wrote to try to stop me from deliberately misinterpreting your meaning -'cause I'm too lazy to think for myself and to much of an asshole to attempt to bridge the gaps between me and you."

Y'all are just mad 'cause she dissed your homey, PZ, who for some reason has a cultlike following among the self-proclaimed "godless liberals" who post on his blog.

I love to watch the anti-racist argument and all of that hypothetical altruistic academic speak turn into the slobber that it is when it comes to these scenarios:

1) The look on the whitewomans face if catches her man sidling up to, or banging the black booty, or the ever popular asian arse.

2) The look on the the whiteman face when his baby comes out of his mommie dearest, but doesn't look 'racially' like him at all.

That's when all of the racionalizations begin;-)

Dude, you have some issues that have nothing to do with race.

I think we are in a state of quandary with regard to race in America. As the study points out, the more we focus on determining a person's race, the better we get at doing just that (self-fulfilling). However, it really does not cut it to go "color-blind" in today's society, either; we can't pretend that skin color and other racial markers play no part in the amount of privilege people have in society. For instance, we must recognize that skin does not have a default pigmentation as bandages or crayons would have you believe.

Somehow we have to recognize the race of others to take into account all the baggage that comes with being a non-White person in the U.S. and the privilege that comes with being a White person in the U.S. while also moving past stereotypical or even bigoted thinking attached to the notion of whichever race a person happens to be. That seems to be awfully difficult, especially for mainstream media.

Thanks for the summary and details of the study, Greg.

The phrase "alleged racial groups" seems strange to me. I would suggest that if you got took 100 ethnically Japanese men, 100 ethnically Indian men, 100 ethnically West African men, and 100 ethnically Irish men, your typical person could easily sort the people into four groups based on "alleged race" without difficulty.

And if you took 100 men over 6.5 feet tall and 100 men under 5.5 feet tall, people could sort them into two groups without difficulty. So what?

To paraphrase Greg's earlier statements on the definition of biological races, if you took a stroll from India to West Africa and found that a) there was some fairly sharp boundary region over which people stopped looking "Indian" and started looking "West African" and b) all the ways in which they changed from looking "Indian" to "West African" varied at roughly the same rate, then you'd have grounds to consider "Indian" and "West African" two different races.

Otherwise, all you're saying is that people show various genetic traits which vary geographically. Which, of course, they do. But "races" are a poor way to model that variation unless it satisfies the conditions above.

I have one question for the defenders of the term "race" here: if you're genuinely interested in a sophisticated scientific understanding of genetic variation in human populations, why are you defending the use of a hopelessly imprecise and unscientific term which mainly serves as a placeholder for prejudice?

"Race" is not a precise synonym for "sub-population", nor is it a neutral term. Words have both denotational and connotational meanings, and you don't get you unilaterally redefine them when it suits you.

Dunc,

if you're genuinely interested in a sophisticated scientific understanding of genetic variation in human populations, why are you defending the use of a hopelessly imprecise and unscientific term which mainly serves as a placeholder for prejudice?

For myself, I started reading up on race and human variation a year or more ago. Asking generally about "race" was the starting point for me as a newcomer to the topic as a sophisticated topic in anthropology and biology. Yes I've come to recognize that top-down categorizations involving race are inherently flawed. But bottom-up groupings of relationships (the gene flow and drift that I keep harping on) is not. Re-reading some literature and notes, I came back across the bottom-up definition of race as "extended families isolated from pan-mixing." That's exactly what I've been talking about, and despite all the clines that Greg mentioned yesterday, I still can't get past the point of Lewontin's Fallacy and the clustering of related groups based on not just a handful of traits, but by a myriad of hundreds of subtle and complex traits.

This isn't redefining the term race, it's seeking the correct definition that matches the variation that we agree is there.

Dan, you're confusing traits with DNA sequences, and "variation" is a perfectly good word for variation. As for Lewontin's Fallacy, there's a perfectly good three-bullet critique of the idea on the Wikipedia page for it. However, I expanded on those problems (as well as a few others with the concept of biological race) one of the more recent times the topic came up. Since the point of that post was to keep from having to answer the same stock defenses of race, I'll just post it here.

The racists brought up studies that showed that, starting from a knowledge of the region of origin of a test population's ancestors, researchers could find genetic markers that, in combination, could sort the test subjects into clusters by region of origin. This was given as evidence of the underlying genetic validity of race. There are three problems with these studies.

The first is a sampling problem. Remember that game of telephone? One of the racists in the thread suggested sampling whites from Sweden, blacks from Nigeria, and Asians from China. Researchers in one study cited excluded subjects who gave "other" for their race. If you cut out the parts of your sample that don't unambiguously fit into your racial mold, it's much easier to point to the remaining subjects as supporting it.

The second problem is related to how the genetic data was chosen. Researchers used a program for analysis that searched for sites among the hundreds sampled on the genome that could be used to sort the population into a specified number of groups. That means that sites that didn't covary weren't used in the analysis. Variation was again discarded on the path to finding distinct populations, and even then, the one of these studies cited in its entirety was not able to differentiate between two Asian groups without restricting their data further. Other studies that haven't discarded data have found that there is much more variation within races than between them.

Third, nothing about these studies suggested that there were any real-world correlates of any note to the genetic sites used to separate the populations, thus failing to demonstrate any importance in the differences that were found. Relying on self-reported data on origin, they did not even show that the genetics they were testing correlated to any traits typically used to sort people into races.

Aaron: I'm talking about the biological concept of race, and yes, the way race is used is not relevant to the fact that it is used. I feel that it is very important to understand that labeling a particular belief or behavior as benign or un-evil does not exempt it from being part of a race-based model.

For instance, if I think that all black people are cute because of their natural sense of rhythm, and I add that black people are also really good at sports, those are all (one could argue) nice, positive things to say. It does not make the belief that a) there is this biologically distinct group of humans with b) specific characteristics that I can d) detect on the basis of a discernible trait such as skin color which in turn e) allows me to predict what individuals will do, not do, be good at, be bad at, whatever, and f) I can extend this prediction to their children and ancestors, etc. etc.

From Dan

But bottom-up groupings of relationships (the gene flow and drift that I keep harping on) is not. Re-reading some literature and notes, I came back across the bottom-up definition of race as "extended families isolated from pan-mixing." That's exactly what I've been talking about, and despite all the clines that Greg mentioned yesterday, I still can't get past the point of Lewontin's Fallacy and the clustering of related groups based on not just a handful of traits, but by a myriad of hundreds of subtle and complex traits.

Forth time...

Dan, what are those traits and what are those groups?

Yes, it is redefining the term "race". Words are defined by common usage, not by what you personally think they should mean. When the vast majority of people use the term, they are not using it to mean "extended families isolated from pan-mixing". You're trying to use the Humpty Dumpty defense - "When I use a word, it means just what I choose it to mean, neither more nor less." Sorry, but it just doesn't work like that.

I agree that it's possible to imagine some other cultural context in which the term "race" is both precise and neutral, with a meaning such as you propose - but we most definitely do not live in such a context right now. We live in a cultural context where people are routinely discriminated against (in all sorts of ways ranging from mild social ostracism to suffering extreme violence) on the basis of other people's perceptions of their "race", in the old-fashioned, unscientific, imprecise, racist sense. We live in a world where a million people can be butchered with machetes, or six million people can be exterminated in concentration camps, for being believed to belong to the wrong "race". And that indisputable fact is the key point of this post.

In the end, all those who love the idea of race will refuse to objectively define or quantify what they are talking about. They will refuse to add substance to their claims by actually naming the groups (races/etc) and why they grouped in this way.

Without objectively defining things your definition for race pans out to: "Race means whatever I want it to mean, for whatever group I want it to mean to." This is not science. Just let the 18th century racial paradigm go.

Dunc,
Come on, terms used in science are evolved and refined all the time. In this case, it's a word: "race". What we know about race, and the pattern of variation it represents, changes. So when I say that race doesn't mean what it used to, I mean that at some point between then and now we learned something about what it is and isn't. It's still the same pattern of variation though, and I (and most people) call that race.

Dan: Define race. Be precise, complete, and accurate. Give us the definition we would need for a two paragraph entry in a dictionary/encyclopedia. No waffling, no ifs/and/ors/buts. Just the definition.

You need to do this because you are making the claim that people who disagree with you have got this wrong. So help us out.

Clarification: Biological race only, please.

5. Biology
a. An interbreeding, usually geographically isolated population of organisms differing from other populations of the same species in the frequency of hereditary traits. A race that has been given formal taxonomic recognition is known as a subspecies.
b. A breed or strain, as of domestic animals.

http://www.thefreedictionary.com/race

"Race" is not a scientific term, and if you really believe "most people" use it in a precise scientific sense to describe multi-axis genetic variation across geographically segregated sub-populations, I've got a bridge to sell you.

You want to discuss the science, use scientific terms.

I'm working with the definition of race that I was alluding to just a moment ago: "a racial group is an [extremely] extended family that is partly inbred."

Actually, I've been alluding to this since my first comment on this thread yesterday - the principle component being the relative isolation (or mingling) with other groups (see gene drift and flow). This yields clines along geographic distances, since mixing with tribes and nation-states across the world was impossible up until ~500 years ago.

Yes, but nobody else is. That's not what "race" means in our current cultural context. You're unilaterally redefining the term to suit your own arguments.

mixing with tribes and nation-states across the world was impossible up until ~500 years ago.

I don't get what you mean here, please expand/explain, maybe give examples.

This just in: The World Kennel Club has stopped categorizing dogs by breeds.

"We recently read a scientific study where people trained to distinguish between individual dobermans faired far better at doing so than people asked only to distinguish between dobermans and dachshunds" said the president. "Obviously 'breed' is unscientific".

"Exactly" chimed in another man. "Unless you can find a sharp dividing line between mixes of dobermans and dachshunds, then 'breed' is a poor way to identify those differences. I don't care how people actually use the term."

"I've gotten so good at ignoring breed I can no longer tell a chihuahua from a mastiff" cried one excited woman. "I feel so progressive. Hopefully one day I won't feel the need to call my pets 'dog' and 'cat' any more!"

"I don't buy it" objected another man. "People have been using breed categorizations effectively for hundreds of years. It doesn't necessarily have negative or unscientific connotations."

"That's easy for you to say" replied the president. "You've never paid the price for being called a doberman. You're just a breedist. We're all breedists!"

"That's right" chimed in another. "Unless you can specifically and objectively define and quantify 'breed', it is just an imprecise and unscientific term which mainly serves as a placeholder for prejudice."

So science avenger says that the biological definition of "race" is the same as "breed" for dogs. Personally, I'm fine with that.

If dog breeds are real races, then why are "Caucasian" "Negroid" and "Mongoloid" real races?

In that dog breeds are something that don't occur without a great deal of work and regulation and that the concept doesn't apply to the vast majority of dogs, Greg?

The concept applies very well to the vast majority of dog breeds and to a couple of wild group but not most individuals.

So I would say that Canis lupus is a species. Two or three subspecies of wild wolf can be defined and distinguished ... they have good geographical barriers, etc. but some of the reported subspecies are probably not good. Then there are a thousand or so breeds of dogs, and each of these is a candidate for a race. One can move the threshold back and forth to group breeds together more or less, but I think almost everyone would agree on a minimal number of several hundred breeds.

Then there are the other dogs. Some of these actually map on to breeds, even though the actual kept dogs in those breeds are a small subset. Like the basenji, for instance. But a lot of the other dogs ... perhaps a majority of Canis lupus ... are what we would call domestic mutts, and placing them in their own "race" would be biologically absurd. They are the non-race dogs.

Science Avenger, do you want to pick the slack for Dan and name the groups (races/etc) and why they grouped in this way?

Or are you also comfortable with the working definition: "Race means whatever I want it to mean, for whatever group I want it to mean to."

Your critics are eager to know...

Sorry, for those just tuning in, regarding the dogs: race = breed = variety (plants) = subspecies.

Dunc,
Says who? I'm getting the precise definition that I'm using from a comment here, and it also apparently matches what the Free Dictionary says (thanks Elizabeth). That, plus, if you've forgotten your Modern Evolutionary Synthesis notes, everything in evolution at the species level has to do with population isolation and/or gene flow between populations (see Dobzhansky; Mayr).

Greg,
The gradation of human variation isn't anything new, it took somewhere between 5 and 50 thousand years to accumulate, since we ventured out of Africa (right?). And as I've been saying, everything relates to isolation of extended families (tribes, cities, etc.) and the frequency of interruption by mixing with neighbors, either by trade or war or just relocation. Not to mention that neighboring groups separated off from each other more recently in, say, Europe, than those same groups separated from their kin in, say, Oceania. This is being traced by the Genographic Project financed by National Geographic, by the way.

Now it's difficult to separate races by what characteristics that they have today, true. But with genetics and archaeology, it's not so difficult to group people by their ancestral routes getting to Europe, the Far East, Oceania, the Americas, etc. That is, grouping people by their historical and genetic relationships.

I also referenced Lewontin's Fallacy at one point. Check it out, and while your at it, GNXP has a link-laden analysis as well.

But by referencing "before 500 years ago or so", I'm talking about the fact that it was only the very rare explorer that traveled more than 500 miles, much less around the globe. Europeans rarely ventured to Africa, Asia, and hadn't yet discovered Australia, much less the many islands in the Pacific. Time was aplenty for genetic drift. And without mixing populations, they truly were isolated from each other. Isolation=races.

All that's going away, albeit slowly, over another 100 generations of population mixing. I need some reading of these concepts, but I've also heard that Sewall Wright's neighborhood model, Wright's island model and the stepping stone model apply to this discussion.

I'm working with the definition of race that I was alluding to just a moment ago: "a racial group is an [extremely] extended family that is partly inbred."

In other words: "Race means whatever I want it to mean, for whatever group I want it to mean to."

BTW that last blockquote is attributed to Dan.

I have one question for the defenders of the term "race" here: if you're genuinely interested in a sophisticated scientific understanding of genetic variation in human populations, why are you defending the use of a hopelessly imprecise and unscientific term which mainly serves as a placeholder for prejudice?

I'm not a defender of the usage of the term race, but an observer of the usage of the term race, and its meaning does not appear to be what Greg says it means. That happens with language. You don't get to impose definitions from a position of authority.

"Race" is not a precise synonym for "sub-population", nor is it a neutral term. Words have both denotational and connotational meanings, and you don't get you unilaterally redefine them when it suits you.

Posted by: Dunc | February 5, 2009 5:09 AM

Who's "redefining" anything?

Sorry, for those just tuning in, regarding the dogs: race = breed = variety (plants) = subspecies.

Posted by: Greg Laden | February 5, 2009 9:59 AM

Not really...

I'm sorry that the way the word "race" is used doesn't fit neatly into a biological sense - at least not in modern life except for the boundaries of true racists who still decry miscegenation. My concern is that the insistence that people blind themselves to race results in a loss of information.

Because, you know, black people are black. They not only have shared skin color and other physical characteristics, they have shared genetic and cultural history, not to mention (in America at least) a shared history of how they have existed in society. In the 70s it was fashionable to pretend race didn't exist at all. As a white man with an adopted black brother, my experience is that this attitude is detrimental to understanding a damned thing about what's really going on in society. It'd be nice to move to a utopian society where race didn't matter at all, but race matters here and now, and pretending that it doesn't exist at all is, from my viewpoint, profoundly idiotic.

And I'm sorry it doesn't exist in a way that divides into neat little equivalence classes, and I'm sorry that many people who discuss race do so in a faux scientific manner, but it is completely counterproductive to declare by fiat that the term is off-limits.

It makes about as much sense as trying to discuss vision without referring to colors. Anybody with even a moderate exposure to the physics of light understands that colors do not exist in discrete units ("red", "blue", "green", "yellow", etc.) and yet these terms are useful as categories in spite of that.

RickD, I'm not sure who or what you're arguing with. None of the people here saying that race is not biologically valid are saying anything other than that it's a social construct. Yes, social constructs are a critical facet of people's experience of the world, but that doesn't have much to do with what the study was looking at.

Who's "redefining" anything?

Dan is. He's redefining "race" to mean "extended families isolated from pan-mixing", rather than "a largely arbitrary classification of people based on how their outward appearance is perceived after being modulated by the prejudices and prior experience of the classifier".

I'm sorry that the way the word "race" is used doesn't fit neatly into a biological sense - at least not in modern life

Sigh. It has never fitted neatly in a biological sense. We've been through this already.

It'd be nice to move to a utopian society where race didn't matter at all, but race matters here and now, and pretending that it doesn't exist at all is, from my viewpoint, profoundly idiotic.

Of course race matters - that's why we're having this argument about it. The point is that there is no biological basis for "race" as it is currently understood in our society - it's a cultural construct, not a scientific observation. That's the whole frakkin' point.

Dunc:

That's a good point about the definition of race, the definition that myself or Dan uses may not be the same definition that the average person may use.

It reminds me of the whole hacker/cracker debate among programmers. We believe that hacker is still a noble term but if the rest of the world has a different definition and it's not really their fault if they misinterpret what we say when we say we're a hacker.

Greg,

mixing with tribes and nation-states across the world was impossible up until ~500 years ago.

I don't get what you mean here, please expand/explain, maybe give examples.

I left a response about an hour ago that got stuck in moderation, perhaps due to the fact that I put a slew of reference links in.

Beyond what I said an hour ago (I hope it made sense), I'm not sure how to elaborate or explain the obviousness of that statement. Populations in Europe didn't mix up until the last few hundred years, and didn't know each other existed at all prior to Marco Polo, if my history isn't off much. Thus you had mixing of populations on local scales for the 50,000 years prior to that, and drift over time of populations separated by greater distances. Plenty of time for both clines to get set up, as well as discontinuities that one such as I might call putative racial boundaries.

More on the comment in moderation...

There you go with a completely non-standard use of the term "Europe" again. China is not in Europe.

Alexander the Great made it as far as India in 326 BCE. The Mongols invaded Europe in 1241 CE, uniting an empire that spread from the Korean peninsula to the Danube. Much of European history is defined by inter-continental conflicts - these are just two of the most obvious examples. Heck, elements of the La Tene material culture and the Celtic language group spread across the whole of Europe and sizeable chunks of Asia Minor in the early Bronze Age.

In short, your history is off. Way off. There has been long-distance cultural and genetic interchange since at least the late Mesolithic.

Dunc,
Way to miss the forest but see the trees. I am sorry though, the comment was written extremely quickly and that sentence could've been written better indeed. To your examples though: just how much did the Mongol and European populations mix? Alexander's army and the Indians? The Celts and people in Asia Minor? Not much? No, I didn't think so. Way to go, history major.

Dunc,
For follow-up and comparison, a recent study showed traceable genetic elements of the Phoenicians in 1 in 17 people from countries surrounding the Mediterranean. Knowing how much the Phoenicians shared culture and genetics throughout trade routes of the Mediterranean for hundreds of years, and yet only 1 in 17 retain a trace. And that's considered a lot.

How much trace do you think the Mongols, the Celts, or the Macedonians left during their excursions?

RickD and some others: We are talking about biological race here, and yes, it can be defined, and yes, there are ways to limit or operationalize a definition.

No one is arguing from authority. And subspecies = race = dog breed.

.. going now to see about freeing comments in moderation. ...

Dan, you lost me. The genetic diversity around the Mediterranean, which has been dominated by one culture after another for centuries (including several post-Phoenicians), supports your view of races how?

Dan: OK, now I get what you are saying. And you are committing a fallacy (but don't feel bad, everybody gets this wrong).

The world you are imagining is one that has certain characteristics including limited movement of people and large spaces between certain populations more or less all the time and forever, followed by a world that changes dramatically 500 years ago. You are telling us that we can understand this world and define the races that developed in this world by using genetics and archaeology.

BUt that would involve actually using genetics and archaeology and using proper models (including null models). The archaeology tells us a very different story than the one you are describing. For instance, in various parts of the world between about 5 and 50 thousand years ago, there were periods of very long distance trade and exchange, and in some cases, demonstrable movement of people over very very large areas. In very recent times (lat 10,000 years) you have mainly a slowing down of movement, a phenomenon some archaeologists call "circumscription" whereby much less movement happens, but there are very few gaps between populations, so you have continuous gene flow over the centuries across very large areas with less movement of culture groups, etc.

It is also almost certainly the case that circumscription and reduced movement of people would have occured at other times in the more distant past in some areas. We also know that certain areas became totally unoccupied, and that a) there are certain points in time when many areas, and rather vast areas, of human habitation became disoccupied, but there were other, longer periods of time when the disoccupied regions fluctuated and alteranted considerably. So, for instance, at the same time that the Sahara was continuously occupied and doing the exact opposite of being a barrier, the southern arid regions of Africa were seemingly not occupied by anybody. All those lands in which the famous primordial bushmen lived in Angoal, Botswana, South Africa and Namibia have not one trace of archaeological evidence of humanity ... nada, zip, zero ... during this period, and the Sahara is loaded with sites.

The actual history of humanity is incredibly complex. "isolated groups" emerge and then merge with other 'isolated groups'. There are periods when there are vast areas of steady gene flow across the landscape more or less 'full' of people.

There are many areas of the world where you can do this: Dig up the dead and look at their skulls for racial characteristics and see what happens over thousands of years. What happens is that the indigenous race, using modern standards of typical race classification, fluctuate across all known races in some characteristics, and totally non-existent races emerge and disappear. This is because these skull-based forensic racial characteristics are fluctuating labile characteristics (yet these are the very features that form the basis for the modern race models), because people change as this dynamic history to which I refer above happens, and because people are moving a LOT. Place you can do this? The Amazon. The andes. The Central Asian Highlands. Southeast Asia, the Pacific Northwest, Australia, Europe, and Japan. And elsewhere.

Stephanie,
Think about gene flow and the definition of race that I'm working with: "a racial group is an [extremely] extended family that is partly inbred."

Gene flow (or group mixing) is incomplete across geographical regions and especially across the world. It sets up situations of both diversity and relatedness - something pulling and something pushing back and forth between separateness and drift, and unity and mixing.

In the case of the Mediterranean, you get the spectrum of diversity that is obviously there. The people of this region have obviously, over the same time period of the Phoenicians and since, mingled less with people of other regions further way, and even less with people virtually on the other side of the world. What you won't get is as much mixing of Mediterranean diversity with regions outside the Med as you are getting just with intra-Med mixing.

Remember, this is all about heredity and ancestry. How many of people in the Mediterranean are descended from Phoenicians? 1 in 17 apparently. How many of them are descended from Celts or Mongols? Any?? Maybe 1 in 10,000, prior to the rise of trade coming from Great Britain in the Middle Ages, but I'm just guessing.

Recap: Heredity and ancestry, with population mixing as the blender.

Greg,

For instance, in various parts of the world between about 5 and 50 thousand years ago, there were periods of very long distance trade and exchange, and in some cases, demonstrable movement of people over very very large areas. In very recent times (lat 10,000 years) you have mainly a slowing down of movement, a phenomenon some archaeologists call "circumscription" whereby much less movement happens, but there are very few gaps between populations, so you have continuous gene flow over the centuries across very large areas with less movement of culture groups, etc.

Really? Is there any data on this continuous mixing over centuries?

Rick D. : "Dude, you have some issues that have nothing to do with race."
Yeah, you might be right. One of those issues is one hit wonders like yourself with exponential nothing substantive to say--while plugging your own dull blog.

Tell me boy-genius: what do you suspect my issues might be?

Dan: What are you asking here? I don't think you quite understand the magnitude of all of this.

Start with Bill Howell's monographs (published by the Peabody Museum). Get back to me after you've read those and any other supplementary materials you might need to understand that material. Then I'll turn you on to another set of studies. After a few years (if you can devote, say, quarter time or half time) you'll have the beginning of a handle on the physical anthro literature, then we can start on the genetics.

Then later, the archaeology.

Cuckholded guy: The Korma is quite good.

Guess you better let National Geographic know that they're Genographic Project is doomed then. I mean, if there's been continuous population mixing of any appreciable degree, tracing human migrations back to the paleolithic is impossible, right?

I took my anthro capstone class (and a couple digs) with a guy named Dan and he too argued with me about race. I think he showed a more complex level of understanding than this Dan but one can't be too sure. I don't understand the reason for fighting so hard to keep race a part of our daily lives. Race is not embedded in the bones nor in our bodies. The only thing that makes us look different is the climate we try to survive in and that is incredibly fickle. If ancient humans did not have population mixing then we would not have one species all across the world, and we would not be able to intermarry with anyone all across the world. I think what we get trapped into thinking is that ancient people were very minimalistic in everything they did. We think that today we are so magnanimous compared to our ancient brethren that we undermine their capabilities. Guess what? Columbus was not the first one to arrive in the Americas. Ancient humans did that a hell of a long time ago and they may have done so in more ways than just walking over an ice bridge.

Greg: Glad you liked it! When were you there? Hope you used "the name" ;-)

The Korma, huh? Just for fun, I will tell you a secret: the Shahi King Korma is sweet and mellow in texture and form...the the Crore Queen Vindaloo is '10,000' times hot, and more powerful...kind of an inside joke about gender roles;-) The menu is full of those inside jokes--my personal piece towards an egalitarian world.

So, I am thrilled that you had the korma: it suits you!

We will soon be opening a kabob bar next door with more vegetarian entree's/kabobs, and hopefully a beer/wine license! I will send a grand opening invite to you for that.

BTW, love this discussion--"can't we all just get it wrong?" and 'stay in our own race'...if I had more time I would troll it but very BZ BZ BZ right now...

From the nonexpert, again:

Dan: ..."since mixing with tribes and nation-states across the world was impossible up until ~500 years ago."

NOVA had an interesting show on the Takla Makan mummies:

"In the late 1980's, perfectly preserved 3000-year-old mummies began appearing in a remote Chinese desert. They had long reddish-blond hair, European features and didn't appear to be the ancestors of modern-day Chinese people. Archaeologists now think they may have been the citizens of an ancient civilization that existed at the crossroads between China and Europe."

The upshot of the episode was that this was a major trading post, much older than your 500 year cutoff.

btw, what's up with all the PZ envy?

Although I do not subscribe the concept of race, there is something to be said for the reluctance of populations to "mix." The Han Chinese lived and traded side by side with numerous Turkic and Mongol tribes, yet never accepted them as full citizens nor intermarried beyond isolated episodes. The Taklamakan mummies (Sogdian?) may be evidence of long range trade and exchange, but do not point to anything other than sporadic contact.
Much ado about very little, but then again that is the bread and butter in field of anthropology.

Dan: Guess you better let National Geographic know that they're Genographic Project is doomed then. I mean, if there's been continuous population mixing of any appreciable degree, tracing human migrations back to the paleolithic is impossible, right?

You keep referring to "continuous mixing" an I'm not sure that I know what you mean. I can tell you this: You do not need races to have history or even to have genetic history.

Greg:

"Dan: OK, now I get what you are saying. And you are committing a fallacy (but don't feel bad, everybody gets this wrong).

[long data filled explanation follows]
"

Thanks for that. I'm still not 100% convinced but that comment went a very long way to convincing me.

One other thing that seems to get conflated here is that "trade" means populations come in contact with each other. Not necessarily so... Marco Polo's trip was renowned for the distance he traveled and the length of stay, he was the exception not the rule. There must be some recognition that merchants are a very small portion of society, and smaller than that were merchants that traveled far. Case in point the Sogdians set up trading posts on the distant fringes of Central Asia; however, those permanent settlements were not populated by their women, the job was restricted to men. These men collected, stored, patronized production, then returned home after their tour of duty. A village of a dozen men is hardly going to introduce enough DNA in the native population surrounding them to make any difference. Trade was a point to point exchange, rarely expanding beyond 500 km until the advent of sea travel. Remember, bactrian camels were not domesticated until 1000 BCE, and they were all but required for trade across the Karakum and other arid regions.
The real difference would be slave trade. The manumitted soldiers brought down during Seljuk expansion, the Slavs brought down by the Golden Horde, and other episodes were more likely to "mix it up." However, that status also brought the stigma of servitude. Then there is the Ottoman practice of castrating those brought into the Janissary, bit hard to pass on the DNA after that event (even if it was not widely done).
Although race is a social construction, the reluctance of ethnic populations to intermingle cannot be denied. It was that reluctance that likely produced these phenotypical divergences.

Onkel Bob, which phenotypical divergences?

IRC, blood type is the standard marker for phenotype divergence.

I like Razib's approach to the question of biological race based on actual allele frequency. If you look at a chart like this: http://scienceblogs.com/gnxp/2009/01/how_ashkenazi_jewish_are_you.php#m… or this: http://bulgarsandgreeks.blogspot.com/2007/07/hla-genes-in-macedonians-a… and note that there are clusters, is that racist? If you use ethnic, regional, or political terminology that vaguely maps to these clusters in order to identify the clusters, is that racist? The fact is, there are a whole bunch of genes for which the continua of allele frequency do cluster, and they sometimes correspond with other clusters, and they sometimes correspond with a social label.

I'm not sure why the fact that these clusters are very fuzzy would indicate a moral failing in admitting that they are still clusters.

You can talk about the deadly and irrational social history of racial categorization, and you'd likely find little disagreement among the commenters on this page. It's interesting and important, but a separate topic. Allele clusters do or do not exist, regardless of the sordid social history of racial categories. Data indicate that fuzzy clusters do exist.

If you point out that social race is a severely insufficient indicator of genotype, then I'm with you.
If you favor a term other than "race" to label pockets of allele frequency, then you're discussing semantics, and probably shouldn't be throwing around heavy terms like "racist" at anyone with purely semantic disagreement.
If you deny that allele pockets exist, you need to look at more cluster graphs of allele frequencies.

this article is pretty good too: http://scienceblogs.com/gnxp/2009/01/of_clines_and_clusters.php#more

Bob:

Indeed, trade, exchange, and different kinds of movements or migrations are all contrasting modalities of human geographic interaction, and there is not a clear way to understand exactly what is happening in the archaeological record with respect to these things. However, it is probably true that long distance movement of lithcs, which is well documented at certain times and places (such as the HP of the Middle Paleolithic, ca 60K, in South Africa, or the obsidian trade around Anatolia and adjoining regions of West Asia, etc.) probably indicate reasonably large interaction spheres across which genes were likely moving not in great saltational leaps but in many acumulative baby steps.

The fact that there was a lot of change, movement of genes, and transformations is clear, however, from the contrasting physical morphology of people in a given region as one looks over time where there is such a record. There are very few places in the world where the basic in situ morphology of people was not different ... more different than any two extant races are ... across spans of time in the one to five thousand year range. This is in contrast to, and falsifies, a model of people not moving, changing, and interacting in a way that is fairly dynamic over spans of space of hundreds of km to thousands, and spans of time of centuries.

The basic five (plus/minus) race model, with these races sitting in place and cooking away ... becoming normatively asian, african, etc. not only ignores the stark reality of a much more complex diversity of living and recently documented people, but it also ignores the reality that when we can see a broad geographical picture for some reasonable time slice in the past, the world, from a 'racial' point of view, simply looks utterly different from time to time in a way that cannot be accomodated by that simplistic model.

It is impossible to have even a vague understanding of the physical anthropology of H. sapiens over the last several tens of thousands of years and accept the race model as typically conceived.

To clarify, Greg - I think you're assuming that your opponents in this discussion are defending the "basic five (plus/minus) race model" and I'm not sure that's an accurate assumption.

One might have general concepts of race in which everyone is divided into a small handful of basically African "races."

Or one might have a finer-grained concept where one distinguishes, say, Greek and Cretan "races" as genomically distinguishable subpopulations.

In an academic discussion, one can discard the clusters of social perception of race and separately discuss clusters in genetic data.

I'm sure we're quite aware that social perceptions and their repercussions are not so easily ignored in the real world.

To clarify, Greg - I think you're assuming that your opponents in this discussion are defending the "basic five (plus/minus) race model" and I'm not sure that's an accurate assumption.

It's close enough.

One might have general concepts of race in which everyone is divided into a small handful of basically African "races."

Or one might have a finer-grained concept where one distinguishes, say, Greek and Cretan "races" as genomically distinguishable subpopulations.

What you are doing here is recreating what actually happened in the history of thinking about race. There were a small number of races to begin with. Then as researchers did more field work, it was fairly easy to document an extra new race here or there, and the late nineteenth century through WW II monographs are full of people describing new races. When the number of races started to surpass five hundred or so, a handful of very insightful biological anthropologists thought further about this and declared that we were seeing a pattern of variation not well described with the race concept. So, actually in the mid 1920s, the race concept died among a particular strand of anthropologists (but it was kept alive by others).

Greg,

Can you tell us the exact number of "cultures" on earth?

Define "culture" ...

People who study culture know you can't actually answer this question, but there are ways to approach it.

An archaeologist has the concept of "archaeological culture" which is probably very different from the standard ethnographic one, which often relies on language. Archaeological cultures, when they are even close to some actual cultural concept, are probably larger than (more inclusive of) linguistically defined culturesl.

So each language is a culture by one approximation. But people speaking different languages may share a culture. Sometimes such a phenomenon is called a "complex" or "culture complex" or "X complex" where "X" may be "cattle," for instance, or "maritime" or whatever.

The number of known languages spoken today is just shy of about 7,000.

Greg,

Is "culture" a scientific concept?

State your point, man.

I don't know your position yet. Answer the question. Is "culture" a scientific concept?

What is a "scientific concept"?

That's a good question. We are constantly told that race is a "social construct," not a "scientific concept." Given your comments in this thread, I am assuming this is your position, so perhaps you can elaborate for us.

What interests me though is why "culture" is not held to the same standard as "race." "Cultures" cannot be precisely defined either. There is no exact point where one "culture" becomes another. The diversity of views within a "culture" is often greater than the average distance between "cultures." The boundries between "cultures" are abitrarily set.

"Culture" is vaguer and harder to define than "race." It is harder still to measure. Why aren't the same objections leveled against "cultural anthropology" that are made against racial classifications?

Where Culture = a group of people with whatever whatever, the concept of "culture" has been jettisoned long ago. No one who studies culture (which is a real thing and which can certainly be studied scientifically) sees 'culture' as a valid category of much use when applied in this sense. That does not mean that you can't find examples of a well defined group of people exhibiting some set of cultural traits etc. etc. But across humanity the concept is, as you suggest, roughly as useless as the race concept.

I like Razib's approach to the question of biological race based on actual allele frequency. If you look at a chart like this: http://scienceblogs.com/gnxp/2009/01/how_ashkenazi_jewish_are_you.php#m… or this: http://bulgarsandgreeks.blogspot.com/2007/07/hla-genes-in-macedonians-a… and note that there are clusters, is that racist? If you use ethnic, regional, or political terminology that vaguely maps to these clusters in order to identify the clusters, is that racist? The fact is, there are a whole bunch of genes for which the continua of allele frequency do cluster, and they sometimes correspond with other clusters, and they sometimes correspond with a social label.
I'm not sure why the fact that these clusters are very fuzzy would indicate a moral failing in admitting that they are still clusters.
You can talk about the deadly and irrational social history of racial categorization, and you'd likely find little disagreement among the commenters on this page. It's interesting and important, but a separate topic. Allele clusters do or do not exist, regardless of the sordid social history of racial categories. Data indicate that fuzzy clusters do exist.
If you point out that social race is a severely insufficient indicator of genotype, then I'm with you.
If you favor a term other than "race" to label pockets of allele frequency, then you're discussing semantics, and probably shouldn't be throwing around heavy terms like "racist" at anyone with purely semantic disagreement.
If you deny that allele pockets exist, you need to look at more cluster graphs of allele frequencies.
this article is pretty good too: http://scienceblogs.com/gnxp/2009/01/of_clines_and_clusters.php#more
Posted by: Spaulding | February 5, 2009 7:08 PM

Precisely.

To clarify, Greg - I think you're assuming that your opponents in this discussion are defending the "basic five (plus/minus) race model" and I'm not sure that's an accurate assumption.

It's close enough.

Posted by: Greg Laden | February 5, 2009 8:25 PM

So you can make assumptions about our "race models" but we can't question your calling all of us biological traditionalists "racists?" FWIW, my first and enduring definition of race was as a systematic level for populations (geographic or otherwise) below the level of species, a term no harder or softer than any of the other levels of systematic nomenclature. (And applying, of course, essentially to all kingdoms of biology; you appear to be only looking at one primate in one order in one class...etc. And if you think my POV is bizarre and that humans are a special case biologically, that would seem to be most species-ist of you.) Since then advances in genetics have only helped define such levels, which are still plastic and arbitrary, as are most things systematic. (& biological, for that matter...) Whatever you want to call them, "fuzzy clusters" do exist, and they need to be identified if you want to do science. (And for the most part, the kind of science you want to do determines what level of clustering below the species level you need to make distinctions. In botany this could be clones.)

The botanical literature has no problem with this...from a quick google search of "botanical races," here's one of the first refs that does not deal with cultivated varieties:
http://www.ou.edu/cas/botany-micro/ben/ben313.html

Please don't tell me I'm missing the fraught connotations of the popular (non-scientific) concept of race, esp. your "basic five" model, in US history. I was around to read first-hand accounts of the Freedom Riders. And everything that's gone down since then.

I could see you making a case for being careful of how we use the term "race" in any context other than purely academic ones--for press interviews, for instance, it would probably be best to talk about "interesting clusters of allele frequencies," or some such, thus leaving John Q Public and most reporters too baffled to object. (Maybe even for limiting its use in regards to any of the "soft sciences," such as anthro.) But to be as strident as you're being is enormously offensive. By your logic we shall next have to start saying there are no such things as theories, because of the way the common usage and the scientific usage are conflated by the enemies of science in their attempts to discredit evolution.

--Diane G

I loerv learning about anthro from these discussions.

Greg Laden, you are pretty much The Reason I "don't believe" in races anymore. (Clines, people! It's about clines!)

More anthro! I know now what is the *wrong* way to think about these things - now help us see what is the RIGHT way!

I know you have been doing just that. So tanks!

I loerv learning about anthro from these discussions.

I think that might be from a Woody Allen movie.

Greg,

You keep referring to "continuous mixing" an I'm not sure that I know what you mean. I can tell you this: You do not need races to have history or even to have genetic history.

Huh? Genetic history is a record of extended family relationships. Hence by my definition of race (which I traced back from where I found it to Steve Sailer), that race is "an [extremely] extended family that is partly inbred."

Genetic history IS race, in a manner of speaking.

Steve Sailer? THAT's your source of the working definition of race that you are using?

HAHAHAHAHAHAHAHAHAH!!!!1111!!1!11!!1!11!

Why do I laugh? He's an ignoramus. No wonder, Dan, that you have such a shallow and limited view of history. E.G.:

Also, keep in mind that the near-absolute ban on black male-white female couplings until recent decades means that 50% of the part black-part white individuals were locked out of almost any chance of mating with pure-whites.

This is so typical of the limited, Appalachian/Idaho log cabin view of the world, and the history of humans, that it should be placed in Wikipedia as an exemplar of the phrase "Kick me, I'm a dumb fuck."

HHAHAHAHAAHAHHA!1!!!1!!

Whoa man, don't give this kind of crap to me during my first cup of coffee.... gotta clean my keyboard ...

Greg,
That's a good point. I'd've left that bit if idiocy out if I were Sailer. Still, I gotta go with the logic of bottom-up extended families, overall.

Greg,
And keep in mind, I explicitly have made it clear that my thought process has not been: I found this essay by Steve Sailer that convinced me. NO --- I found a definition of race that fits with what I know about diversity and biology, and traced it back to that essay.

But no, you'd rather pin me on association to bad ideas of a guy simply because I inadvertently quoted one of his good ideas. Way to dodge the issue of races as extended families and find something else to hit me with.

And just to be clear though, on just about all other matters after looking him up, I find Sailer repugnant too.

Racialist: Race exists. It's real.
Race denier: Ok, what are "the races" then?
Racialist: You just can't define "the races." It's shorthand for whatever gradient of any variation (genetic or physical) I happen to arbitrarily notice in whatever arbitrarily defined geographic populations I happen to want to group together.
Race denier: Well, this doesn't work. You take any group of people you deem are "a race" then look for any similarity (genetic or physical) which would justify your grouping.

To do this would seem very subjective because you can move/redefine the requirements/size/location/etc (willy-nilly) of any arbitrarily selected "race"; all the while ignoring everyone (inside and outside your sample) who doesn't fit into your (now ad-hoc) arbitrarily defined "race". How can this be science! This doesn't even meet up with the #1 science tenant of falsification.

You assume you are right and then look for any evidence supporting yourself, all the while, using an infinitely sliding scale.
Racialist: Yes, that's race.
Race denier: So, France is a race, Paris is a race, Zulu is a race, my brother and sister are also races, Gingers are a race, Blacks are a race, upper and lower Egyptians are races, etc...
Racialist: Gene clusters exist. Face it man.
Race denier: What genes? What populations? Define the clusters? What sampling did you do? What is the genetic gradient? What is the universal measuring stick?!! I want to know!!!
Racialist: I know I've offended your left-wing sensibilities, this is the price I must pay for telling the truth. In time the world will realize that politics have not a place in science.
Race denier: AHHHHHHHHHHHHH!!!!!!

Dan, part of the reason a person has to take a good look at your sources is that your definition of race is meaningless, and every time you're asked to explain what it means in practical terms, you use the same definition again. Hell, I have a known genealogy with some nonbranching spots (cousin marriage) a few generations back. Does that make my mom's side of the family a race? If so, how far out can I trace the connections and still be in the same race--all the way back to mitochondrial Eve, or are you suggesting I stop somewhere short? And why?

For those who insist on race actually existing, please explain... what are the distinguishing characteristics of the race to which Chinese people belong? what are the historical boundaries for their race? what are they today? I would love to know.

Stephanie,

Dan, part of the reason a person has to take a good look at your sources is that your definition of race is meaningless, and every time you're asked to explain what it means in practical terms, you use the same definition again.

Obviously I don't think it's meaningless. And the key reason being is that it reflects the biological concept of what a species is, a la my hero, Mayr:

Species: The biological species concept defines a species as members of populations that actually or potentially interbreed in nature, not according to similarity of appearance. Although appearance is helpful in identifying species, it does not define species.

Racial Group: an [extremely] extended family that is partly inbred.

Look, they fit together.

The key thing here is that we define how much population mixing has been going on since, say, the paleolithic (or prior to that even). I don't know the answer, and I don't think that any of you do either. But that would really address what I've been talking about. Does anyone have any leads?

... what are the distinguishing characteristics of the race to which Chinese people belong? what are the historical boundaries for their race? what are they today? I would love to know.

1) They are all Chinesee and shit.
2) The place where Chinese-ness ends and non-Chinese-ness begins.
3) Chinese.

This is easy.

Dan, they fit together so beautifully that you can't tell me the difference between the two. You've been challenged several times to say what your definition of race means in any practical sense--something related to the real world--and haven't been able to do it. What is the point in breaking down the species into smaller units?

Dan, how am I dodging an issue? The extended families idea is nothing. Of course there are extended families, this is evident. The fact that every person has an extended family does not in any way predict or require a race concept. If anything it suggests that the null model is a continuum of allelic variation across almost all of the landscape. Like mutts.

Regarding Mayr: Indeed this is a bit tragic. The guy was pretty smart (though also rather stuck up, if you must know) but his concept of race is classic racialism and easily reconstructed. Fortunately most of his work is not about this sort of thing.

You seem to be sticking with the argument that races were established and existed in some form in the "old days" (which you are now calling the paleolithic) and then things started to change at some later time. What about the paleolithic? What about the physical remains of humans found form the Paleolithic, Neolithic, etc. that show wildly dramatic change over time in a given place, and changes in clinal patterns across space, in your racial characteristics? The actual record of the paleolithic simply does not accord with your pop-sci made up version of history.

If there are races, then there must be good reasons for knowing someone's race. What can we do on an individual basis to prove someone's race? It's cute and fun to talk about "fuzzy clusters" but are what are ways to prove an individual's race. Will there be genetic makers in a "black" American man as opposed to a "white" American man?

Dan: This is the crux of the matter. Your definition of race could be thought of like a hypothesis. But you've created an untestable hypothesis. Rigor it up, man, and maybe you'll have something.

Race, subspecies, breed, variety, type, population. The term "race" has a lot of bad baggage. "Subspecies" and "breed" just sound ugly if applied to human beings. Personally I would avoid those terms.

As long as anyone recognizes some kind of identifiable set of populations, even if they accept that these are fuzzy, mixed, clinal etc. these can be labeled "races" (or "subraces" or "supraraces" depending on the scale of populations and subpopulations recognized) and we have another "race" concept.

So race is ultimately semantic. It's a culturally resonant word with reifying implications. And therefore best avoided.

One can support the scientific race concept (which one?) without being a racist. Some forensic anthropologists do.

Likewise, someone can have the most sophisicated, complex understanding of human population variation, with clines and admixture aplenty, and still be a racist.

These things are not inherently racist:

- Accepting a scientific race concept.

- Believing that some cultures or at least specific cultural features are more conducive to human health and happiness than others. (And if one condemns National Socialism on principle and not just on the historical record or supports any political worldview, one really believe this, regardless of whatever culturally relativist anti-imperialist pieites one intones.)

Here's the crucial question determining whether someone is racist: do they believe that some (relatively large - not some isolated inbred village) populations are significantly genetically stupider and/or prone to violence than others? The genetic part is key, because otherwise we're back to culture or other environmental developmental factors.

Race, subspecies, breed, variety, type, population. The term "race" has a lot of bad baggage. "Subspecies" and "breed" just sound ugly if applied to human beings. Personally I would avoid those terms.

Hiding the baggage in the trunk does not make it go away. The most divergent of the races, in a species that can be divided into races, are the most likely to produce a new species, all else being equal.

One can support the scientific race concept (which one?) without being a racist. Some forensic anthropologists do.

That is only if your definition of "racist" is "bad" ... but there are many many areas where people will disagree on the "badness" of a particular racialized statement. Better to call all of it what it is: Racist, or racialized, whatever. But the moral judgment can't be part of the definition because that is too vague.

I'm assuming Greg is equally opposed (yeah right) to the use of the malicious and exclusionary "family" concept which suffers from the exact same definitional problems of "race" and "culture."

There, I have said it: families don't exist. They are figments of our imagination. Families are just too ambiguous for us to identify.

Note: I'm still holding my breath here waiting for "cultural anthropology" to be condemned as a psuedoscience.

Stephanie,

What is the point in breaking down the species into smaller units?

Guess you missed my point agreeing that "breaking species down", i.e. top-down approaches to race, are futile. Try thinking bottom-up groupings of people by relationships.

Greg,
Extended families as a way for describing relationships in biology is everything. It wasn't until Darwin and the early evolutionists stopped trying to classify by division, as every naturalist from Linnaeus to Cuvier had been doing, that studying common ancestry was of any help. After Darwin started looking at species as related from the bottom-up, everything fell into place.

Likewise with race. You're committing a falacy for dismissing approaches in human diversity by grouping people by familial and ancestral relationships.

Basically what I'm saying is Greg: Get your head out of the "Everyone thinks top-down" box. It's clouding your thinking.

Dan, it is, as I have said, axiomatic that familial relationship is fundamental to biological relationships, populations, etc. Duh. What I'm asking you is why this obvious fact leads you to the inescapable conclusion that a particular species is divisible into sensible, meaningful distinct races?

Your arguments have deteriorated into a rather pitiful state, quite honestly. This "top down" gambit is going nowhere.

Godless: You need to make more of an effort to make yourself clear. Your side long comments are not working (for me, anyway).

How can Cultural Anthropology possibly be condemned as "Pseudoscience"?????

Judging by the comments above, it looks like "nations," "ethnicities," and "states" have to go too; all figments of our imagination which are too ambiguous to recognize or classify.

Let's continue further down this road: obviously we have to dispense with "yellow" and "green," along with all the other colors, beyond that, I can see "young" and "old," "child," "teenager," and "senior citizen" being problematic, and then "hot" and "cold, the "Atlantic," "Pacific," "Indian" and "Artic" Oceans seem too mysterious to comprehend, all the states of the union (Mississippi doesn't exist), "ice," "water" and "gas," the continents ...hmm, isn't all life whatsoever just a cline?

Greg,

Where is the exact point where the Gulf of Mexico becomes the Caribbean Sea? Where is the exact point where the Caribbean Sea becomes the Atlantic Ocean?

Awwwwww! Don't throw a fit just because you can't think of a way to determine a race through tests. Race is just an ill conceived notion and getting rid of this idea is not going to hurt our comprehension of the world. As a materialist, you sure are throwing a bunch of material out all because race is being thrown out? I guess I'll be throwing out my hopes of studying Caddoan speaking people since race doesn't exist.

Well, isn't the central organizing concept of "cultural anthropology" - "culture" - a social construct that suffers even more so from the same definitional problems as "race"? How can something as vague and undefinable as "culture" (or how about "luminous aether") be studied scientifically?

What we're really arguing about is the definition of "racist" rather than "race." I would argue that a racist is someone who believes that some races (or clinal fuzzy populations, whatever) are genetically inferior to others. (Even if they protest that being stupider or more violent is not necessarily inferior.) It's about ranking in addition to merely making distinctions. As for "racialist," it's been tarnished by "racist." So I would say that someone who recognizes races is "race-recognizing." He or she may or may not be racist/racialist; that is, assert a ranking of these races.

OK, fine, I'll acccept "we're all racist" if you want to play it by the Whiteness Studies, postcolonialist, critical sociology handbook. But c'mon. We know that there is difference between between some schmo who is racist by virtue of by being human (or white or Western or privileged or whatever the definition) and an honest to gosh racist.

Aside: I have a fondness for anything that mucks up nice, neat categories - not just "race" but also species and higher taxa: hybridization, horizontal gene transfer, symbiogenesis.

Godless Materialist, did you miss the distinction between social construct and biologically valid?

Dan, guess you missed the point of this post; i.e., that treating race as a valid biological concept has some undesirable social consequences, such that there is a distinct burden upon those who want to use the concept to show that it has usefulness that offsets the downside. Stop trying to condescend to everyone who tells you to take up that burden or drop the concept. As has been pointed out to you repeatedly, there are plenty of ways to describe relationships that don't require the concept of biological race. Clinging to race as to a security blanket just tells the world that you're insecure.

Where does Israel end and Jordan begin? Where does someone's blackness become whiteness?

I'm unaware of the existence of any "test" that can give us the exact so-called "age" of a decomposed skeleton. Forensic anthropologists still have methods though that can produce reliable estimates.

"Age" seems to be an "ill conceived notion" and getting rid of this idea is not going to hurt our comprehension of the world. Some skeletons appear to be more "aged" than others, but the notion of "age" (which often leads to "ageism") is just so hopelessly imprecise that it is of no utility whatsoever.

I would say "age" is in the eye of the beholder. I might be 75 years old, but I feel "youthful."

That's an interesting question Chas, since Arabs were considered white for US census purposes for a long time, Lebanese Americans are generally viewed as white, and light skinned Arabs have long perceived themselves as "white" and are labeled as such by some scholars of sub-Saharan African descent. So this notion of Arabs as automatically belonging to a category called "people of color" or "brown people" (as in the left wing meme "war on brown people") is really pretty novel. Edward Said has more than a little to do with it. So in addition to 'How the Irish became white' (see Race Traitor) there is also the question of how Arabs became nonwhite.

"Undesirable social consequences" is a value judgement. Natural science has nothing to say about "good" and "evil."

Godless Materialist, you win, hands down, for the poster who has provided the most bluster in a single blog thread (in under then 45mins no less! we must have touched a raw nerve). Keep it up you might win an award...

Godless, do you do children's parties?

Godless Materialist, are you affiliated with or a fan of this shindig:

http://preservingwesternciv.com/speakers.html

On the subject of semantics, two euphemisms for scientific racism I'm already quite sick of: 'race realism' and 'human biodiversity.' The first is just wimpy (if you're a racist, sin boldly) and the second is already taken.

Now that I think about it, the division of time into so-called "seconds," "minutes," "hours," "days," and "weeks" is arbitrary and varies across cultures, so we might want to get rid of these pesky units. Also, why do we call it the "Northern Hemisphere"? Couldn't the "Southern Hemisphere" just as easily be the "Northern Hemisphere" relative to our sun and Sagittarius A? Is it really the year 2009? Why is our calender based on the birth of Christ as opposed to some other date?

Colugo,

I believe in the existence of races, yes. I also believe in the concept of "age." It is sometimes difficult though to pinpoint the exact age of a given person in a lineup. I think terms such as infant, toddler, adolescent, teenager, young adult, adult, and senior citizen are useful and valid concepts - even though they blend into each other at the margins.

Godless - have you completely missed the repeated attempts in this thread to distinguish between cultural constructs and biological realities? All of your examples are of the former. No-one here disagrees that race exists as a cultural construct. We just disagree with the idea that the prevailing cultural constructions of race are an accurate reflection of the underlying biological reality.

You're doing an amazing job of completely missing the point.

Good you stopped trying to scream, bluster and confuse people long enough to hobble together a coherent argument.

From Godless

I believe in the existence of races, yes. I also believe in the concept of "age." It is sometimes difficult though to pinpoint the exact age of a given person in a lineup.

...and I think race is a real social concept (at least in the US where there is a very real and very strong racialized history), however what this has to due with the tea in China, I just can't find out...

BTW, the point of this little exercise has been to illustrate that the same infantile objections which are made against the race concept could be made against any number of other categories that 1.) exist along a continuum and 2.) the borders of which are socially defined. Yet they are never made; all of these other categories, in particular "culture," are accepted without controversy.

Now, why is that? The answer is quite obvious: the concept of race is offensive to the liberal/humanist political and moral beliefs of people who believe dispensing with the concept is some kind of social service.

And he's doing a great job of making our points for us. Why do we call ourselves a part of the northern hemisphere"

Dunc,

Is "age" a "biological reality" or a "cultural construct"?

When you start describing it using "terms such as infant, toddler, adolescent, teenager, young adult, adult, and senior citizen", then it's obviously a cultural construct.

Colugo, that is a amazing website you just linked. Did you check out their purpose section?

We believe that Americaâs Judeo-Christian heritage and European identity must be defended. Today, our glorious Western civilization is under assault from many directions. Three such threats will be discussed at this conference.

They go on to list #1. "Third-World immigrants" #2. "Islam" #3. "guilt feelings" (...that whites feel due to "the blacks", which in turn "undermine Western morale and deter us from dealing sensibly with the other threats").

It reads like any other right-wing paranoid xenophobic nut-fest that you can see on all "race-realist" websites. As always the usual cast of characters are there. Good job, we should start a blog. ;)

Greg,

What I'm asking you is why this obvious fact leads you to the inescapable conclusion that a particular species is divisible into sensible, meaningful distinct races?

Just the previous sentence you agreed that familial relationships are essential for understanding evolutionary groups. What the heck is this crap then? I'm explicitly NOT trying to divide species, I'm trying to establish whether familial relationships can be extended to establish human genealogical clades. And if those clades exist at all, which I think they just might, how much have they remained distinct despite "continuous mixing since the mesolithic" as you said a while back up this thread. Why the hell do you think I was asking about degrees of mixing?

I'm GIVING you the grounds to get me to concede that I'm wrong: show me data on population mixing in the mesolithic that you mentioned!

Godless, did you miss the part upthread where Greg said that culture is not a clear-boundary concept? Why try to still insist that someone's saying it is?

Now, one more time, because you've been too busy trying to be clever to listen, biological race is a problem concept because using it leads to harm. Not some nebulous political positioning--actual harm to actual individuals and societies. If you can tell me what harm it does to use day and night or northern and southern hemispheres as arbitrary constructs, you might have a point. If you can tell me what the concept of biological race buys us that is worth the harm, you might have a point. Until then, you're just generating noise.

tif: "the usual cast of characters are there."

Yup. And including, a little disconcertingly, Henry Harpending. Which reminds me that I'm looking forward to Greg's review of Harpending and Cochran's new book. (To our host: Sorry, I don't mean to sound like a noodge!) Harpending and Cochran can be quite brilliant when not indulging their racialist hobgoblin.

Dan: Just a little mixing on the peripheries of populations could transfer positively selected genes from one side of the globe to another in little time. So even populations that had never met until recent historical times could share genes due to mixing.

Tif, your summary is pretty accurate.
I'm still waiting for the part where someone explains why it's a problem to have terminology simply because borders are indistinct.

What genes? What populations? Define the clusters? What sampling did you do? What is the genetic gradient? What is the universal measuring stick?!! I want to know!!!

Genes that have not reached fixation. The more you pick, the more the patterns cluster. Leewontin showed that relying on just one gene wouldn't be useful in identifying regional subpopulations. Edwards showed that relying on several makes identification fairly easy. The universal measuring stick is the frequency of alleles. The size and number of the clusters are a product of the arbitrarily chosen resolution. The maximum resolution is determined by the # of individuals and # of genes charted. Obviously that varies by study. Feel free to check out the ones that I liked to earlier as a starting point.

My position is that "race" is as good a term as any to describe these data clusters, especially if one chooses a low resolution. Given a high enough resolution, I don't think your example of a "French race" is absurd. For privacy reasons, I hope we don't ever have sufficiently detailed data to start talking about something like a "Parisian race". Furthermore, the linguistic precedent for race tends to avoid such fine-grained distinction, but that's a purely semantic point.

I'll also note the obvious point that many individuals will not neatly fit one of the major data clusters.

The existence of outliers or indistinct boundaries are not valid arguments against either the existence of clusters or the use of terminology for clusters.

Stephanie Z,

Thanks for illustrating my point. Your objection to concept of race is based on liberal/humanist ideology, not scientific considerations.

Note: The "Northern Hemisphere" corresponds to an underlying geographical reality.

I'm reading the book now with great interest.

As someone who has traveled throughout the DRC, I find it difficult to believe that Greg is unable to distinguish between Ituri Pygmies and Katangan Bantus.

Spaulding, I just did that.

Godless, thanks for pointing out that you're indifferent to harm.

Colugo,

Dan: Just a little mixing on the peripheries of populations could transfer positively selected genes from one side of the globe to another in little time. So even populations that had never met until recent historical times could share genes due to mixing.

"Could" or "do"? (just out of curiosity.)

Stephanie,

Why not dispense with the concept of sex too? After all, women have lived under the oppression of patriarchial societies for centuries.

Godless, there's a much clearer basis for describing who belongs to what sex, although there are still border cases that I'm perfectly happy to not classify. Gender, of course, is another matter. Are you really stuck with that poor an analogy?

Dan: ""Could" or "do"? (just out of curiosity.)"

Out of sheer scientific conservatism, I'm going to stick with "could" (perhaps someone knows better than I). But also keep in mind that absence of evidence is not evidence of absence. Just because our instruments currently do not detect a signal of something happening does not prove that it did not happen. (And that's really important in archaeology and paleontology too.)

But even if the operative word is "do" does this mean that we cannot speak of distinguishable human races, I mean, populations? We have to accept some categorical fuzziness in a whole lot of things, not just in biology. So I'm on your side in principle but I've already given my reason for avoiding the word "race." It's bad mojo.

Stephanie,

You are arguing that we should ignore hate facts on the basis of the harm such division has caused in the past. Well, patriarchy was certainly far more common than race-based societies across history.

No, Godless. I'm arguing that there needs to be some underlying basis in reality for continuing to use social concepts that cause harm. Otherwise, we're embracing harm because we're scared to change our thinking.

And you're still having trouble differentiating sex from gender.

Me:

To clarify, Greg - I think you're assuming that your opponents in this discussion are defending the "basic five (plus/minus) race model" and I'm not sure that's an accurate assumption.

Greg:

It's close enough.

Then you and I are interpreting comments in very different ways, and I fear you are arguing against a strawman.

Would anyone reading this who is specifically defending the biological accuracy of the social "basic five (plus/minus) race model" please raise your hand? Or, you know, explicitly state this in a comment.

"Race" corresponds to the genetic clusters mentioned above. "Age" corresponds to objective differences in skeletal morphology. Forensic anthropologists determine the approximate "race" and "age" (and "sex") of skeletons all the time for law enforcement agencies. This is mundane work.

The same people who insist "race doesn't exist" (who are motivated by a PC liberal/humanist ideological agenda) seem to have no difficulty in making traditional racial distinctions in everyday life (he's black, he's white, he's Asian, etc).

The term is not as problematic as you are making it out to be. It is no more mysterious than the Atlantic Ocean or longitude and latitude. Sure, the various races blend together at the edges, but so do any number of categories.

What good comes from knowing what race someone belongs to? What can "black" people do to that would be beneficial to their otherly bodies that "white" people don't have to? What makes them different?

Oh, and you obviously don't know the differing perspectives among bio-anthropologists pertaining to "racing" bones. Men and women play along the rules of sexual dimorphism so it is a lot easier to determine sex but race... that is something that law enforcement requires from forensic anthropologists and is by no means mundane. Finding the race to bones would really have to be determined by the socio-economic status of the individual and from there you can guess at the race. But, the bones don't reveal race. Sorry, try again. (And, dude, this is stuff from my bachelor's degree, so maybe you shouldn't post such lies on a blog managed by a bio-anthropologist. He might think that you're a... you know... liar)

Stephanie Z, maybe I'm missing a specific comment that you're referring to, but from what I read you've presented a lot of arguments from consequences: i.e. We should stop using the term "race" not because it is false but because the social harm outstrips the utility of the term. This may very well be an accurate appraisal, and is likely to become even more true as globalization continues to encourage gene flow and personalized medicine overtakes population-based risk assessment.

However, while an argument from consequences may bear relevance to how we are morally obliged to behave, it does not bear relevance for the accuracy of our measurement and categorization systems. It's a separate conversation entirely. Though I find both conversations to be of interest, I am (and I think some of the other commenters are) trying to isolate questions of accuracy.

In other words, I'm not attempting to dispute your views of the social phenomenon of race and its undesirable consequences. I'm trying to discuss the biological grounding of subpopulation labels.

Either A: human genetic variation does not exist or
B: human genetic variation does not at all map to phenotype or
C: human genetic variation is a smooth continuum or
D: human populations form a bunch of big and little clusters of allele frequencies

A and B are stupid. C and D can be distinguished by the data. D is not an all-or nothing statement, but a matter of confidence intervals. Fuzzy edges and outliers would be expected for D. If the data is much more like C than D, we can scrap D. The data I have seen suggests D is valid.

It's interesting to examine these clusters because, as would be expected from a globe that is less than panmictic, populations linked by geography, trade, or recent shared ancestry, etc. tend to have closer allelic overlap than more discreet populations. And then sometimes you have a population that shows closer relationship with a more geographically distant population - wow, we learn something about the history of those populations! It's exciting and informative, and gives us information that is impossible if we make it taboo to speak of human subpopulations.

And informative large and small-scale subpopulation groupings need not be static, monolithic, non-porous, or based on hierarchical, self-serving old school categories. (Though yes, they often are socially treated that way, and yes, we should call shenanigans on that approach.)

Godless, which races correspond to which genetic clusters? The studies which purport to show genetic clustering use the same kind of data analysis that caused the authors of this study to suggest that cell phones will give you vertigo but cure your Alzheimer's. Read the critique of that study. Try not to let your agenda and ideology blind you to the implications for the studies you're relying on to construct some kind of biological underpinnings for race.

Then, start working on better analogies. Seriously. You keep giving me half analogies. Where is the harm caused by agreeing to the arbitrary borders of oceans or arbitrary reference points on the globe?

Chas,

Thanks for your input. Here's a real forensic anthropologist:

http://www.pbs.org/wgbh/nova/first/gill.html

As pointed out in a recent 2000 edition of a popular physical anthropology textbook, forensic anthropologists (those who do skeletal identification for law-enforcement agencies) are overwhelmingly in support of the idea of the basic biological reality of human races, and yet those who work with blood-group data, for instance, tend to reject the biological reality of racial categories. . . .

First, I have found that forensic anthropologists attain a high degree of accuracy in determining geographic racial affinities (white, black, American Indian, etc.) by utilizing both new and traditional methods of bone analysis. Many well-conducted studies were reported in the late 1980s and 1990s that test methods objectively for percentage of correct placement. Numerous individual methods involving midfacial measurements, femur traits, and so on are over 80 percent accurate alone, and in combination produce very high levels of accuracy. No forensic anthropologist would make a racial assessment based upon just one of these methods, but in combination they can make very reliable assessments, just as in determining sex or age. In other words, multiple criteria are the key to success in all of these determinations.

Spaulding, see the comment to Godless, just above, as well as yesterday just after 8 a.m. I am arguing both that the data is not there and that, in the absence of data, utilitarian considerations rule.

I agree that it would be fascinating to discover that populations that have been largely believed to be unrelated are, in fact, related beyond their basic humanity. Do you know of any cases where this has been revealed by looking at frequency of alleles? All the tracing of lineage that I know of relies on specific mutations in mitochondrial DNA for the direct maternal line and in Y chromosomes for the direct male line.

Godless, that is an interesting link. Just in case you're accused of cherry picking, here's the main page. Seems to me that there's room to agree to disagree on this subject. While I disagree with the "harm" argument (just how does one do a cost benefit analysis) and find Dr. Gill's argument persuasive, I disagree with the pejorative politically correct that is batted about there and here. Furthermore, Dr. Braces' argument has merit when he speaks of common geographic ancestry, yet he too, invokes pc to justify his argument. It's a dark day when Safire's "nattering nabobs of negativism" can be applied equally to both sides.

All the tracing of lineage that I know of relies on specific mutations in mitochondrial DNA for the direct maternal line and in Y chromosomes for the direct male line.

Yup, that's within the category of allele frequency. E.g. the recent Etruscan discoveries based on mtDNA allele frequencies.
If you're implying that recent novel alleles indicate shared ancestry more reliably than ratio of common variants, I certainly agree. For example, two populations that start with a full range of pigment alleles could evolve to be predominantly dark or predominantly light based on their climate. If they display convergence, it need not be evidence of recent shared ancestry.

However, convergence among numerous allele ratios, especially those not evidently environmental adaptations, would rise to the level of evidence.

I posted a couple of links in a comment above to works on allelic distinction of ashkenazi jews as well as distinction of greeks/macedonians/other mediterranean groups. These are interesting because they are groups that have had relatively little geographical barrier to intermixing, yet their genomes remain statistically distinguishable.

Chas,

Nice to see you moving the discussion to the non-scientific realm of "good" vs. "evil." That's where it properly belongs. It has been apparent for quite some time now that the jihad against the concept of race is motivated by the ideology of liberal humanism.

Re: Stephanie

See the work of geneticist Neil Risch. He was able to make racial classifications on the basis of genetic clusters that are 99%+ accurate. This alone belies the argument that races are too ambiguous to classify.

http://med.stanford.edu/news_releases/2005/january/racial-data.htm

Godless, I've seen the study. See the cell phone critique.

And while you're engaged with arguing the data with me, stop trying to suggest that people are only arguing on the basis of politics. It looks like you can't keep your arguments straight from comment to comment.

Dan:

Hence by my definition of race (which I traced back from where I found it to Steve Sailer), that race is "an [extremely] extended family that is partly inbred."
I found a definition of race that fits with what I know about diversity and biology, and traced it back to that essay.

Sure Steve's definition seems to fit with what we know about biology, but we already have a name for that sort of thing - population.

But for those who are saying that races don't exist: do populations exist?

Godless, which races correspond to which genetic clusters? The studies which purport to show genetic clustering use the same kind of data analysis that caused the authors of this study to suggest that cell phones will give you vertigo but cure your Alzheimer's.

It's not the same kind of data analysis.

All you have talked about is politics. PC this, PC that. What makes the races different? What does it mean for someone with a dark complexion vs. a light complexion? What does it tell us? What does it mean, when people have a complexion that is hard to categorize (when their parents have easily placed races)? From my own observations, people with who fall somewhere between white and black are almost always considered black. Why? How should we treat black people as opposed to white people if it is so important to categorize them?

windy, these are studies that throw in a very large number of variables and look for variables that correlate with the predetermined population breaks. What are the differences that would effect the fact that one would expect a certain percentage of the variables to show a statistically significant correlation even if the distribution is actually random?

Dr. Gill could not stop throwing out the moniker "race denier" for those who feel that races are a social construct. That is a political maneuver and something that abortion politics use.

windy, these are studies that throw in a very large number of variables and look for variables that correlate with the predetermined population breaks.

That's not how cluster analysis works. Your cell phone example would make sense as a critique of genome-wide association studies, not so much for clustering.

Godless materialist: That is actually an invalid study for the purpose of asking this question. The study involved comparing samples of people from populations where the geographic point of origin for each population is small, and the distance between the geographic clusters is very large. If every allele in the human genome was randomly distributed in independent even clines across the planet (which would be the exact opposite of a racial configuration) you would get the results obtained in this study.

As pointed out in a recent 2000 edition of a popular physical anthropology textbook, forensic anthropologists (those who do skeletal identification for law-enforcement agencies) are overwhelmingly in support of the idea of the basic biological reality of human races, and yet those who work with blood-group data, for instance, tend to reject the biological reality of racial categories. . . .

Yeah, and they've been wrong about that for years.

Start with a continent of Native Americans. Add Europeans, mainly Germans (that's the largest group of Europeans up through recent years). Add West AFricans (not central, not east, not northern, now southern, but west and even then from only a relatively small area). Add chinese (from only some areas of chinese).

For several generations (a small period of time in the broader evolutionary history of things) these people will mostly not interbreed. They come from vastly different geographical areas. They look (falsely) like races, so everybody gets on board with the race idea. Of coruse, forensic anthropologists who need to tell the cops if the body is a black guy or a white guy, etc. can do so with some degree of certainty.

But not nearly the certainty that you attribute, or think, is the case. And it is often wrong. And as people from other areas of the world ... southeast asia, other parts of Africa, other parts of eruope, south asia, etc. etc. move into the region the ability to tell people apart based on the previously somewhat useful but always limited racial characteristics becomes more and more difficult until finally we reach a point where the methods pretty much become useless.

"I would argue that a racist is someone who believes that some races (or clinal fuzzy populations, whatever) are genetically inferior to others. (Even if they protest that being stupider or more violent is not necessarily inferior.)"

See, this is what puts people's backs up. I think that there may well be differences in intelligence among different "clinal fuzzy populations" resulting from different average frequencies of certain alleles. To use but one example, my own "clinal fuzzy population" - southern Europeans/Italians, consistently scores about 5-10 points lower on IQ tests than the "clinal fuzzy population" most people would recognize as Ashkenazi Jews, on average. I think there's a good chance that this is due to Ashkenazi Jews having a higher frequency of IQ-boosting alleles. A fairly persuasive theory explaining why we should expect Ashkenazi Jews to have a higher-than-average freequency of IQ-boosting alleles - and even some hypotheses as to where some of those alleles might be found in the genome - has been put forth here:

http://homepage.mac.com/harpend/.Public/AshkenaziIQ.jbiosocsci.pdf

Now, does this make me a racist? Against Italians? I guess by your definition it does. But when I think racist, I think at worst the KKK or the Nazis and, at best, the guy at a recent Prohibition party I went to who got drunk and called the black girl I brought with me a racial epithet. I am neither of these things. I do not hate people and I do not hurt people. Nor do I appreciate being lumped in with those who do because I find it implausible that there are no average, genetic-based differences in personality, intelligence, athletic ability, creativity, etc... between groups of people that have been relatively, though incompletely, isolated from each other for thousands of years, adapting to different cultures and different physical environments.

The idea just strikes me as very unlikely, as though the brain, where so many thousands of genes are expressed, has somehow been immune to the effects of drift and selection.

resulting from different average frequencies of certain alleles.

Mark. Explain how exactly specific genes determine the brain wiring in a way that has to do with intelligence. Discuss details. Be thorough in your description of the ontogeny of the tissues, and changes in the brain tissues over time, and mention the actual genes that are involved in the process and discuss the known allelic variations in those genes.

Or is it just that it must be true because you feel it must be true?

Stephanie,

I agree with Windy re: your cell phone critique. Cluster analysis studies don't use prior information about the groups when doing the cluster analysis. It's only *after* the analysis is completed and genetic clusters have *already emerged* that these clusters are looked at to see if they correspond with traditional racial or ethnic groups, as defined sociologically. In all the studies that I've seen, the clusters have correlated with traditional racial groupings, which is pretty convincing evidence that there is a genetic basis for much of what people have traditionally termed race.

I also think it's silly - sorry - to say that because people have done horrible things in the name of racism, we shouldn't discuss or investigate these differences. It's like saying we should shut forbid the study of economics because so many people died as a result of Marxism.

Greg,

Are you saying there is no genetic basis for intelligence? Because we know from twin studies that it is highly heritable...

As for the genes themselves, Cochran and Harpending do a much better job of presenting the case than I could, and you are apparently reading the book yourself. Their discussion on the specific genes begins on page 213, if you care to read ahead, and peaks on page 220, which, for the benefit of those without the book, reads as follows;

"The sphingolipid mutations [abnormally common among Ashkenazi Jews], in particular, have effects that could plausibly boost intelligence. In each, there is a buildup of some particular sphingolipid, a class of modified fat molecules that play a role in signal transmission and are especially common in neural tissues. Researchers have determined that elevated levels of those sphingolipids cause the growth of more connecting neurons, the basic cells of the central nervous system..."

There's much more, and I'm not doing the book justice, but I don't want to be seen as avoiding your question.

Mark: see my comment above to godless materialist regarding the self fulfilling prophecy of such studies.

BTW, one thing about cluster analyses: They make clusters whether they are there or not. In many common forms of analysis used with genetic data, there are measures that indicate the 'strenth' (sometimes 'treeosity') of the resulting trees. You need this when using a technique that makes trees whether trees should be there or not.

When comparing many human data trees to other species, the human trees can be very weak. Nothing you would go to the bank with.

Yet, if you really really really need to believe in race, then you will go to the bank with it. But these studies are the bad mortgages of science.

Mark,

In all the studies that I've seen, the clusters have correlated with traditional racial groupings, which is pretty convincing evidence that there is a genetic basis for much of what people have traditionally termed race.

These studies have typically had large geographical gaps in the sampling though. Therefore at this point the amount of clustering in the total human population is an unsettled question, IMO.

Greg:

In many common forms of analysis used with genetic data, there are measures that indicate the 'strenth' (sometimes 'treeosity') of the resulting trees. You need this when using a technique that makes trees whether trees should be there or not.

We are not talking about trees.

Windy: Yes we are.

Greg,

Regarding those clusters, the fact that humans races - or "races" if you prefer - are less genetically differentiated than, say, breeds of dogs doesn't change the fact that those differences exist, and doesn't tell us anything about what those differences are.

Also, regarding genes for intelligence, I recently reviewed practically all the literature on candidate gene association studies looking for genes that code for higher intelligence. The strongest candidate gene identified was the succinate semialdehyde dehydrogenase (SSADH) gene, which is necessary for the degredation of y-aminobutyric acid (GABA), the central nervous system's primary inhibitory neurotransmitter. A condition known as semialdehyde dehydrogenase deficiency leads to variable degrees of cognitive impairment in sufferers, likely due to neurotoxicity resulting from accumulation of y-aminobutyric acid.

Anyway, the T allele on the functional SNP rs2760118 in the succinate semialdehyde dehydrogenase (SSADH) gene codes for a protein (tyrosine) with only 85% of the enzymatic activity of the protein (histidine) encoded by the more common C allele. The C allele on rs2760118 has been associated with both increased full scale IQ as measured by the Weschler Adult IQ test (Plomin et al, 2003) and, interestingly, increased longevity and cognitive performance in the elderly(De Rango et al, 2008). The full scale IQ increase is estimated at 1.5 points per copy of the C allele in the general population, and was confirmed in both between-family and within-family analyses.

When you look at the three Hapmap populations, you find that the C allele is most common among East Asians, where it reaches a frequency rate of 90% and 88% among the Chinese and Japanese, respectively, is somewhat less common among northern Europeans, where it reaches a frequency rate of 71%, and least common among the Yoruba of Nigeria, where it reaches a frequency rate of 55%.

Now, this is just one gene. But the point of citing the information on it is to highlight the fact that there's no scientific reason to expect alleles relating to brain function and higher intelligence to be evenly distributed among all human population groups.

Regarding those clusters, the fact that humans races - or "races" if you prefer - are less genetically differentiated than, say, breeds of dogs doesn't change the fact that those differences exist

No, the fact that they pretty much don't exist is the reason they don't exist, and also the reason why treeosity values in most cases are so low.

I'm not especaily impressed with genetically broken brains as models for "interracial" intelligence.

Windy: Yes we are.

Well, maybe you are. Are you talking about the 'trees' that are constructed as part of a hierarchical clustering process? Not all clustering methods work like that.

And 'treeosity' you just made up. I think there's another word for it. Treeitude? Treelikeness?

Windy: I'm talking about the algorithms that are used by evolutionary biologists to construct trees ... and it is trees that we call them ... and treeosity is in fact a statistic that really exists. It is not necessarily used by everybody but yes, it's real!

There are all different ways to make trees, of course.

Windy: I'm talking about the algorithms that are used by evolutionary biologists to construct trees ... and it is trees that we call them ...

Who's 'we' kemo sabe? I thought you were an anthropologist (not that there's anything wrong with that)

Mark and Stephanie and I were talking about clustering methods. While there's some overlap in the methods, I'm not sure why you are suddenly talking about phylogenetics as if it's the same thing.

and treeosity is in fact a statistic that really exists. It is not necessarily used by everybody but yes, it's real!

Is it used by more than one person? I still think you are confused about the name of the statistic, at the least. There are no meaningful hits for 'treeosity' anywhere. Or care to give some references?

A lesson in treeosityhttp://www.youtube.com/watch?v=s8iEBcIS1Vk

Wait, all this race realism stuff doesn't make sense. I'm "white" (Lithuanian dad, Croatian mom), yet all the local donair shop proprietors give me discounts because they think I'm middle Eastern. What's the deal with that? And am I, on average, smarter or dumber than someone from Turkmenistan or not? I'd hate to be smarter or dumber than my race would dictate because of some silly allegiance to political correctness.

What Dan, Godless, Mark (probably because he was too busy transcribing the names of SNPs), and others here fail to understand is that of course one can pick any trait and find different frequencies of that trait among different populations. Unfortunately for the 'race realists', you'll find different distributions depending on what traits you look at. Take blood groups for instance, and map them. (Oh, whaddaya know, it's already been done.) See anything there that conforms to your idea of 'race'? Given that, what makes the traits you think you've based your category of 'Caucasian', 'Asian', and 'Negroid' more real, besides the fact that blood type isn't immediately visible?

It's been said more than a few times up thread (and more succinctly, too), but at the risk of being repetitious, but given how poor 'race' is an indicator of, well, almost anything biological, and the grievous harm the concept has had and continues to have, what the hell are you all defending it so staunchly for?

(And Godless Materialist: seriously, look up the difference between sex and gender sometime, as Stephanie Z suggested. Then pull that anti-liberal stick out of your ass. You must be virtually buried in chiropractic bills.)

Windy, I am a biological anthropologist. And I know what cluster analysis is.

I am a biological anthropologist. And I know what cluster analysis is.

Then it's a shame that you choose to post such obscure comments about it. And I'd still be sincerely interested in those references to papers that use the 'treeosity' statistic?

Windy, it is a shame that you have chosen to judge my credentials. I spent a fair amount of time writing an about page (not too snarkily, but a little). One click away.

Treeosty is obscure. I was used only briefly and in a few papers, but it has such a cool name it is not forgettable. In the field of phylogenetics there is a huge literature on methods and statistics.

Start here: http://tinyurl.com/b5nnpt

Windy, it is a shame that you have chosen to judge my credentials.

No, silly, I found it amusing that you chose to lecture me about what statistics "we" evolutionary biologists use, when I actually work with phylogenetics and have never heard anyone use 'treeosity'. But it's entirely possible that it has been used before my time so it would be fun to see a link to those papers.

Treeosty is obscure. I was used only briefly and in a few papers, but it has such a cool name it is not forgettable. In the field of phylogenetics there is a huge literature on methods and statistics.

Then please give me a link to a paper that uses 'treeosity' not to 'phylogenetic statistics'! Sheesh!

Oh shit...poor greg...windy and colugo stopping by and leaving dropping samples...if I remember right, windy is a humorless PZ sycophant; don't waste a definition of treeosity on such a self-sniffing beast--it will only turn into fartsmoke, in the elevator to PZ's dungeon.

Then please give me a link to a paper that uses 'treeosity' not to 'phylogenetic statistics'! Sheesh!

for what it's worth, not having seen the term "treeosity" before myself, I believe what Greg was trying to do was point out that it belongs to the realm of statistics used to analyze phylogenetic trees, hence the link.

A bit more explanation of how it is used would be nice, but my impression seeing the term used here is that "treeosity" is a kind of "best fit" analysis?

could be totally wrong, but that was the impression I got.

... as to the comment just before mine...

wtf?

Just spent quite a bit of not-grading-quizzes time reading this thread, and it was a very interesting way to procrastinate. Human genetic variation shows clines and clusters. The specific clines and clusters observed depend on which loci and alleles are analyzed. Still, many clusters and clines correlate geographically. (This is all unsurprising, and is doubtless true for any species with a large geographic range relative to dispersal ability.) Humans are exceedingly visually oriented animals, and we notice visible features of other people. We use these visible features to identify and distinguish among family members, friends, acquaintances, celebrities, enemies, and complete strangers. In addition to these categories, our brains are capable (and, arguably, hardwired) to use visible features to categorize other people in other ways. One of those ways, in contemporary multicultural societies, is by the probable geographic origin of another person's ancestry. Empirically, it usually works. I live and teach in probably the most ancestry-diverse place in the world, and on the first day of class as I circulate around to start meeting students, names (derived from languages) and faces (derived from genes and ancestry) are easy to match in many many cases. None of this is controversial.

Do races exist? Are races purely a social construct? Geographically correlated variation in phenotypic traits that are visible with clothes on does exist--that's not a social construct. It's the categorization and labelling of such variation as "races" that is the social construct. To me, far more insidious than the abstract concept of "races" is the semantic labelling of socially perceived races. What could be more polarizing than the ubiquitous terminology of "black" vs. "white"? That taxonomy is empirically incorrect*, connotation-laden, and absurdly insufficient, but is used by nearly everybody nevertheless. Even our blog-host, perhaps the most vociferous and persuasive opponent of the race concept on the blogosphere, has no compunctions about calling somebody a "white guy."

Not sure I've added anything substantive to the discussion but there you have it anyway.

p.s. windy is anything but "humorless;" nobody on the internet has made me laugh harder, Brownian and a few others possibly excepted.

*An anecdote remembered from a long-ago discussion on the Mike Douglas or Merv Griffin show:
1: Why do you call yourself "black"? Your skin is brown.
2: What about you? Is your skin "white"?
1: No, you're right--my skin is flesh-colored.
2: So's mine.

Dan writes:

For follow-up and comparison, a recent study showed traceable genetic elements of the Phoenicians in 1 in 17 people from countries surrounding the Mediterranean. Knowing how much the Phoenicians shared culture and genetics throughout trade routes of the Mediterranean for hundreds of years, and yet only 1 in 17 retain a trace. And that's considered a lot.

... and then:

Remember, this is all about heredity and ancestry. How many of people in the Mediterranean are descended from Phoenicians? 1 in 17 apparently. How many of them are descended from Celts or Mongols? Any?? Maybe 1 in 10,000, prior to the rise of trade coming from Great Britain in the Middle Ages, but I'm just guessing.

But the article he links to says:

Scientists reported Thursday that as many as 1 in 17 men living today on the coasts of North Africa and southern Europe may have a Phoenician direct male-line ancestor.

In any random generation of my ancestors, I will have exactly 1 direct male-line ancestor, and 2^n other ancestors. One in 17 means the Phoenicians have contributed about 1/17 of all the genetic material in that population, and by now everyone will have multiple Phoenician ancestors. The only exceptions will be recent immigrants.

That study doesn't help your point in any way at all.