Is the concept of race scientifically valid and useful? -- James T vs. Chris Porter

[KnightWhoSaysNi]

This thread has been set up for a formal debate between James T and Chris Porter who will debate the following resolution:

"Resolved: race is a scientifically valid and useful concept."

James T will affirm and Chris Porter will oppose. The debate will have 3 rounds and James T will go first, per the parameters (http://www.iidb.org/vbb/showpost.php?p=4513444&postcount=31).

A Peanut Gallery (http://www.iidb.org/vbb/showthread.php?p=4513519#post4513519) is set up in the Science & Skepticism forum for the rest of us to comment on the debate.

Enjoy the debate!

- KWSN, FD Moderator

[James T]

Race is a scientifically valid and useful concept

Essentially this is going to be a discussion about what race means. Within the scientific community there is a group with a strong desire to remove race from the vocabulary. The problem is that this attempt is contrary to well recognised racial groups within the community and the presence of many real issues where continued use of the term race is required.

These scientists, in an understandable aim to communicate how little race means biologically, have erred by attempting to say that small differences are no differences at all. The end result is that they lower their credibility, undermining any valid criticisms they may have.

My position has always been that race are clearly evident, that these have a foundation in differing allele frequencies that are verifiable (and have been tested) and that it is unreasonable to assume that these differing allele frequencies have no effect on characteristics that are a political minefield. What I am met with is races do not exist hence this formal debate.

What do the scientists think?

Ann Morning (On Distinction (http://raceandgenomics.ssrc.org/Morning/), 7 June 2006) nicely encapsulates the issue:

In 2001 and 2002, I interviewed over 40 university professors in biology and anthropology about their definitions of the term “race.” Their views varied widely. …

Despite their differing conclusions about race, however, the scientists … shared the same fundamental ideas … . They agreed that human beings vary biologically, both genetically and phenotypically (e.g., in surface appearance); that this variation is shaped by evolutionary processes; that the patterns of variation are linked to geographic locale; and that each variant usually shades gradually into the next, without sharp, crisp borders separating people …

Clearly scientists agree on the fundamental genetic characteristics of the human race. The problem is over how to define race.

The Species Problem

The species problem (http://en.wikipedia.org/wiki/Species_problem) is the long-standing failure of biologists to agree on how we should identify species and how we should define the word 'species'. Hey (2001)

How is this relevant to the problem of defining race? There are two relevant issues where I will draw parallels. Firstly the meaning of species and what biologists understand of species has changed form dramatically over time. Secondly, while the meaning has changed and precise definition is considered impossible, biologists are not running about crying that species do not exist.

Both race and species have suffered from initial naïve and invalid concepts of distinct and separate natural kinds. Over time, and increasingly with the development in genetics a more mature understanding has developed, to repeat:

human beings vary … genetically and phenotypically …[are] shaped by evolutionary processes; that the patterns of variation are linked to geographic locale; and that each variant usually shades gradually into the next, without sharp, crisp borders separating people

Now we recognise “What It Means to Be 98% Chimpanzee: Apes, People, and Their Genes”. Yes, the currently accepted figure is closer to 94%. Yes, I am abusing the title of a book written by Jonathan Marks. Now consider a recent statement by R.C Lewontin link (http://raceandgenomics.ssrc.org/Lewontin/):

The remaining variation, about 6% to 10% of the total human variation is between the classically defined geographical races that we think of in an everyday sense as identified by skin color, hair form, and nose shape.

Apparently 6-10% of the available range of human variation between classically defined races is not particularly important. Perhaps Lewontin is correct as to the lack of importance of this 6-10%, however this is an entirely different proposition from stating that no such racially identifiable difference exists.

Genetic Identification of Race

Perhaps though, with our modern understanding of race, it is not possible to use genes to identify race. In this instance it would be reasonable to accept Lewontin’s hand waving argument as a neat summary of the scientific position. Unfortunately for Lewontin this turns out not to be the case.

From “Geography and genography: prediction of continental origin using randomly selected single nucleotide polymorphisms” Allocco et al link (http://www.biomedcentral.com/content/pdf/1471-2164-8-68.pdf).

We used small numbers of randomly selected single nucleotide polymorphisms (SNPs) from the International HapMap Project to train naïve Bayes classifiers for prediction of ancestral continent of origin. Predictive accuracy was tested on two independent data sets. Genetically similar groups should be difficult to distinguish, especially if only a small number of genetic markers are used. The genetic differences between continentally defined groups are sufficiently large that one can accurately predict ancestral continent of origin using only a minute, randomly selected fraction of the genetic variation present in the human genome. Genotype data from only 50 random SNPs was sufficient to predict ancestral continent of origin in our primary test data set with an average accuracy of 95%. Genetic variations informative about ancestry were common and widely distributed throughout the genome.

This is in stark contrast to Lewontin’s selectively stated position expressed in Confusions About Human Races, R.C. Lewontin Jun 07, 2006 link (http://raceandgenomics.ssrc.org/Lewontin/):

A clustering of populations that does correspond to classical continental "races" can be acheived by using a special class of non-functional DNA, microsatellites.

The Alloco et al paper was accepted 10 Mar 2007, perhaps Lewontin was unaware of these findings in 2006. However I am surprised Lewontin did not consider the potential and reserve comment until after the science was done.

A number of analyses have been done on the accuracy of self-identification of race when compared to genetic markers. From Genetic Structure, Self-Identified Race/Ethnicity, and Confounding in Case-Control Association Studies from Tang et al 29 Dec 2004 link (http://www.journals.uchicago.edu/AJHG/journal/issues/v76n2/41839/41839.web.pdf):

Of 3,636 subjects of varying race/ethnicity, only 5 (0.14%) showed genetic cluster membership different from their self-identified race/ethnicity.

So, people can self identify into genetic groups. Perhaps this could be useful to medical practitioners if there are any conditions that occur with more frequency in particular racial groups.

Medicine

Perhaps though … no, there definitely are genetic conditions that occur with more frequency in some racial groups, sickle cell anaemia is but one.

I would be doing a disservice to ignore a more significant medical concern with regard to race. Self identifying racial groups are also customary groups. Customary groups have customary habits and these habits are often the real source of racial medical complaints. Note, science extends far beyond the bounds of genetics.

Summarising the Biological Position

Biologists would like to have us believe that there is no such thing as race. What they really mean is that it is hard to come up with a pat definition of race. They have the same problem in defining species.

Perhaps medicine is not part of biology, this position is correct for customary racial practices but fails to explain the presence of medical conditions with genetic conditions that are strongly associated with convention racial groups.

Where to from here?

So people can reliably self-identify their race. This corresponds accurately with randomly selected SNPs. A number of genetic conditions exist which correspond to race in a medically significant way. Doctors can use self-identified race, as part of a patient history, as an aid to diagnosis without the need to gene-sequence everyone who walks through the door.

All I really need to do is demonstrate one case where race has a scientifically valid use. I have done that. However I’d like to explore a little more in scattered areas through various fields of thought, both scientific (and lay).

Why?

Why does James T get so hot under the collar over this issue?

It turns out that it is not the issue that irritates me. If this degree of political correctness was getting in the way of clear and honest expression on some other issue, then the other issue would be receiving my attention (and some are, though in real life).

This matters because if we can not talk clearly and precisely about the real issue then what hope do we have of addressing it. Race is involved in many social problems and to address these we cannot begin by having to first argue about whether race is even a meaningful term.

Oddly enough, the most effective means for a racist to prevent us from being able to correct racial inequity is to waste our time with questions of the validity of race as a concept.

The Lay View

This made me really uncomfortable when I saw the movie (about 2 weeks ago for the first time). The movie – Domino (2005) – was rubbish, but it illustrates a point I’d like to make. From the script (http://www.imdb.com/title/tt0421054/quotes):

Lateesha Rodriguez: Now, as a blactino woman, I believe we deserve our own race category to forge an identity, Jerry. That's how I feel.

Jerry Springer: Did you just say "blactino"?

Lateesha Rodriguez: Yes, I did. I'm a blactino-American.

Chinegro Woman: Wow. Uh, first of all... first of all, you don't even look latino. You look black. You're... You're black. Second of all, I'm of mixed race, and I've struggled my whole life as to whether I'm Chinese or whether I'm black.

Lateesha Rodriguez: Chinegro! What you are is chinegro!

Chinegro Woman: Chinegro?

Lateesha Rodriguez: Chinegro! There you go!

Chinegro Woman: Chinegro?

Lateesha Rodriguez: You are a chinegro!

Chinegro Woman: What the [bleep] is chinegro?

While many of these terms won’t appear in your dictionary I’d be surprised if many of you misunderstood their meaning. Mixed race is a common excuse for why race is an invalid term. Perhaps when bastardised terminology like this no longer has meaning to you then this excuse will be correct.

Philosophy of Scientific Knowledge

Falsifiability, as addressed by Karl Popper, provides a very basic hurdle for discussions about race to qualify as being valid scientific knowledge. At present this hurdle is met with some ease.

Kuhn’s incommensurability, on the other hand, provides an indication of the process that will eventually result in race being dropped from the scientific vocabulary. Once all fields of science have developed to the point where the paradigm shift, whereby science develops a view of people that is incommensurable with race, then race will fall from the scientific landscape. However, while genetic science is well on the way to this ideal, many fields of science lack the finer grained knowledge to replace the concept of race.

Sociology

One such fields is, of course, sociology. Perhaps you’d like to argue that sociology is not a science. However, until such time as racial groups have parity or have been bred out of existence, there are some inequalities I’d like to see addressed.

My Challenge to Chris

Please, no straw man definitions of race.

[Chris Porter]

First off, I think I need to apologize in advance for some rather offensive claims I will be making here. I am going to reference common beliefs about race, and some of those beliefs may be very offensive to readers. I apologize for this, but believe that in order to accept the debate, I must acknowledge the fully common-sense view of what race is. I am not going to claim race is geographic ancestry, for even though that may be capable of being determined, race has never meant just that.

This debate is important because race is a term with an important hidden premise: people of the same race are fundamentally all the same. What’s true of one black, will be true of another black. Further, that race is not only just skin color and hair texture or nose shape, but race also determines personality, moral integrity and intelligence. This hidden premise imbues the term race with all the power of a stereotype, with all a stereotype’s oversimplification and rigidity, and all its subsequent possible harm. If it can be shown that race is scientifically valid and useful, then by careless osmosis, it will also validate the use of stereotypes. Never mind the fallacy of composition here, more people than not, upon hearing that race is a scientifically valid term, will associate in just the way I say, regardless of such errors. Perhaps that is why so many are not willing to go in the direction of seeking scientific validity for race, seeing that common use of that fact would lead almost inevitably towards misuse.

The Scientific Validity of Race

While I agree with James T that there are scientists that believe that the term race as applied to the human species means so little as to be worthless to discuss, other than to dismiss, neither of us shall go there, as the opinions of individual scientists (or all scientists, actually) are not germane, it’s facts we’re after. I intend to show that if race is indeed meant to be scientifically valid, it must be pinned to the playing field of science, not of social community. It is clear the debate is intended to show that race correlates to some yet-to-be-identified genetic algorithm that affects not only skin color, nose shape and texture of hair, but intangibles such as intelligence, moral character and personality, as has been historically connected to the concept race. In order for race to support such contentions in a scientific manner, so to speak, it must show itself to be amenable to objective criteria of science. That means, that not only must race, within scientific measurements, mean more that skin, hair, and nose shape, but it must be shown to correlate with the intangibles such as intelligence in a way that can best be cross correlated with the adjunct of visibly different skin color, hair texture, and nose shape. We need not only to be able to identify the tangibles, but the intangibles as well, and then to show the high degree of correlation between them. This is what race is, as people who think race is a useful concept believe to be the case. To be scientifically valid, it must better correlate with those characteristics than any other correlations, such as economic opportunity, or country of origin. Further, this concept must then be of use to people, perhaps medically, perhaps economically, socially or politically. Only then can the affirmative “race is a scientifically valid and useful concept” be well defended. I propose that race as is commonly or historically conceived of is not a scientifically valid concept, nor is it useful. I will present my case first, then rebut James.

Race is certainly historically, culturally, economically and socially capable of being defined, but as far as scientifically valid, especially in that area it wishes to excel, biology, I think it fails muster. If, for example, race only meant “from this-or-that continent”, then surely James T is correct, and not only is it scientifically valid, but easily verified by experiments such as -- well, he hasn’t actually presented any scientifically valid experiments like that. His example of “self-identified” subjects showing high correlation to continental genome sequences don’t really show that race is self evident. It shows that people that know their ancestors can guess what continent they came from, and this can be verified by comparing gene sequences from those continents to the test subjects and noting the high correlation. Now, if those test subjects had to identify someone else’s race, and then the tests comparing gene sequences were run and they did correlate by continent, wouldn’t that show that race is scientifically valid? No, that too would not be scientifically valid. Why? because the concept of race is more than just the skin color, hair texture, or other physical features--it’s the personality and intelligence as well. The intangible traits, let us say. Which can’t be identified by pictures, nor can it be done through self-identification. Throughout history, people have tried to correlate images to personality, moral values, and intelligence.

If we are going to scientifically validate race, it really has to be on all fronts. We can’t just say something like “Hey, we can recognize people’s continent of origin from their physiology. That’s race, and it must therefore be scientifically valid”. Because race isn’t used like that in the everyday sense. Making that claim, then trying to use race to also include the personality intangibles spoken of earlier is a fallacy of equivocation. Race is commonly used to mean all the intangibles and tangibles that come for originating on one continent or another. People who want what the concept race offers complain bitterly when race is simply modified to genotypic clines grouped in geographic proximity. That alone can’t really say all that much about the nature of the race, after all, other than it appears dark skin often goes with kinky hair, and slanted eyes with straight hair, and people vary that way from continent to continent, for the most part. It says nothing about intelligence, for example. They want more, they want that geographic designation to correlate to intelligence (smart asians), to personality (lazy blacks), and to moral integrity (pious whites). For that to occur, some studies must be undertaken to show how these traits are genetically coded. I haven’t actually come across any. Most such claims these intangible traits correlate to physical traits seem to stem from 16th or 17th century statements of opinion, and are promulgated to this day. Therefore, I believe the concept of race as a scientifically valid term, using the common definition of race, has yet to be investigated! Making claims at this time as to scientific validity seems incredibly premature. They are too easily shot down on a number of fronts from the above 1) race is more than just genotypic variation in overlapping geographic proximity, to 2) race has yet to be correlated to intelligence in ways that haven’t been carefully rebutted, and 3) race has yet to clearly correlate tangible with intangible traits in any scientifically valid investigations. Surely these issues need to be investigated and facts ascertained before claiming race is scientifically valid.

The Usefulness of Race

The meaning of race in medicine is commensurately different than the use of race in common knowledge. In medicine, it means only something equivalent to “where-your-ancestors-came-from”, and has no connotations of intangible traits relating to issues like moral integrity, intelligence, or personality. In medicine, race might be used to make a note of possible complications, like diabetes in American Indians, prostate cancer in African men, or high-blood pressure in Hispanics. Unfortunately, going solely by visible aspects to determine race, in this instance, is not nearly as useful as taking a careful family history is. Thus, is this truncated, not-really-what-we-mean-when-we-say-race-anyway, really all that useful, when more immediate, and more individualized treatment, is obtained by another method? I think it’s less than useful. And it’s not even what we normally mean by race. R.C. Lewontin, in “Confusions about Human Races” states, For the vast majority of human genetic variations, classical racial categories as defined by a combination of geography, skin color, nose and hair shape, an occasional blood type or selected microsatellites make no useful prediction of genetic differences.” … “The categorization of individuals simply as “white” or “Afro-American” or “Asian” will result in a failure to test for such abnormal hemoglobins because these abnormalities do not characterize the identified “race” of the patient. Even group identities below the level of the conventional races are misleading.Thus it would seem that even taking into consideration that race in medical realms may be only geographic, unlike the common definition of race, it still cannot usefully predict genetic differences well enough to make medical diagnostics as well as a knowledge of family history can. Race, of one definition or another, is not useful here.

Well, if race is less useful than other methods in medicine, perhaps it’s of great usefulness elsewhere? If we use race in the common sense meaning of the term, I know of no reports other than crime statistics that can be considered as potentially useful, since crime is so often associated with lack of moral integrity, or due to laziness, or lack of intelligence, all of which are intangible traits long associated with race. Crime and race seem likely candidates. Or would that be crime and minorities? Or crime and poverty? Or crime and gender (we have a winner!)? Which correlation might be most useful? Surely each of these tools of comparison has problems: the crime rate of women is growing at a far faster rate than men: maybe there isn’t a correlation of gender and crime, but opportunity and crime? Could the same be said of crime statistics and minorities, race or poverty? And what sort of usefulness, other than apparently correlation, can be gleaned? Well, race is not something amenable to change, nor is being a minority, or a gender. But poverty can be addressed by social means, should the society desire. This ability to physically address an issue makes one tool more useful than another, and in this way, we can see that race and crime correlations, should they exist, are not as useful in addressing crime reduction as poverty and crime correlations. Again, race loses out on usefulness to another, as easily aggregated, set of conditions.

And anyway, even if race is a less useful tool of discrimination than other tools, why toss any tool away, even if it doesn’t perform as well as others? Well, I think we can safely toss a tool that can’t be used to aid society. If one can’t change race, and crime is associated with race, then what can be done? Creepy things like eugenics, or racial uplift programs, I suppose. (I have often entertained the thought that the Hawaiian race was carefully (or unthinkingly, hard to tell) being culled from existence by federal and state government programs, btw. But that’s my dreadful secret conspiracy voice speaking.)

I will now address some of James’ points specifically.

What Do Scientists Think

James T quotes Ann Morning here: In 2001 and 2002, I interviewed over 40 university professors in biology and anthropology about their definitions of the term “race.” Their views varied widely. … What follows the ellipsis is this:Almost 40 percent of these academics took what can be called an “essentialist” view: they described races as groups of people who share certain innate, inherited biological traits. In contrast, over 60 percent held a “constructionist” perspective: they argued that races do not correspond to patterns of human biological variation, but rather that racial groupings are “constructed” through social processes that take place in particular historical, political and economic contexts. In other words, the jury was out on the scientific nature of race.Compare this to the statement James makes that:
Biologists would like to have us believe that there is no such thing as race. It appears to me that Meadows clearly held the view that biologists have some sort of definition of race, and the majority of them think it just doesn’t have a biological definition, but rather a social definition.

Genetic Identification of Race

James says of Lewontin’s remarks, Apparently 6-10% of the available range of human variation between classically defined races is not particularly important. Perhaps Lewontin is correct as to the lack of importance of this 6-10%, however this is an entirely different proposition from stating that no such racially identifiable difference exists. and then, Perhaps though, with our modern understanding of race, it is not possible to use genes to identify race. and goes onto explain the nice correlation of genetic markers and geography. I believe that’s a step towards a fallacy of equivocation, for the common sense concept of race has suddenly, radically, altered meaning, such that race no longer means “Hey, you're visibly black, white, yellow, red, or brown, regardless of continent of origin”, but now means continent of origin via genetic markers. It doesn’t depend on visual data humans use to describe race, but depends on the analysis of genetic markers. I’m pretty sure race has never meant “this or that group of genetic markers”, and aside from the following section where James describes the ability of people to identify their ancestry, thus their race, so race must be real, this isn’t what most people think race is. The concept race also is the ability to tell something about someone else, not just yourself. If race now depends on genetic markers, then issues of what continent someone came from are rather a side issue. And still, those genetic markers don’t seem to be able to correlate the tangible genetic markers with the intangible traits long associated with race. Perhaps that is for the future.

The Lay View

Completely irrelevant to the topic of race as a scientifically valid concept. I refuse to address this.

Philosophy of Scientific Knowledge

James believes that falsifiability is well met with the data at hand. However, in order to do that, race as a concept is reduced to the definition of continentally distinct genetic markers--and apparently there are only three races, not 4, 5, or 30. I’m pretty sure even The Lay View would throw a fit over that. Where’s the data showing the validity of the common sense meaning of race? Where’s the data showing that the intangible traits we associate with race? Perhaps that can be addressed in the next essays.

James mentions the Kuhnian paradigm will eventually replace race with...something, possibly genetic markers, and I suspect he is correct. In the meantime, should we find a use for race in its common meaning, or for race in its genetic meaning? Would it ever be more useful than other tools of analysis?

[James T]

Race is a scientifically valid and useful concept

Thanks Chris, for your response.

I see from the peanut forum that some of you have more detailed knowledge on genetics and biology than I do. That's cool, I am not looking at the trees. It would be great if scientists (in fact everyone) would address their concerns directly rather than through the poor proxy of trying to eliminate the word "race". This is a kind of doublethink that I find distasteful.

Defining race

I didn't clearly identify the definition of race that I was happy with. I like Ann Morning's definition, not that she'd call it a definition for "race" per se. However while some are into word games substituting clines where everyone else would say race ... oh well. I have added a piece in bold just to make it clear:

[scientists] agreed that human beings vary biologically, both genetically and phenotypically (e.g., in surface appearance); that this variation is shaped by evolutionary processes; that the patterns of variation are linked to geographic locale; and that each variant usually shades gradually into the next, without sharp, crisp borders separating people … these variants are called races.

We know they are relevant to a degree biologically because people can accurately self-identify in a way that can align with a small set of randomly selected SNPs (link) (http://www.biomedcentral.com/content/pdf/1471-2164-8-68.pdf).

the concept of race is more than just the skin color, hair texture, or other physical features--it’s the personality and intelligence as well.

Chris has chosen to define race to include as much baggage as possible. I don't share this view at all; I doubt scientists who actually study, say, the relationship between race and IQ would either, since they wouldn't then need to measure for IQ, just race.

Chris's Straw Man & Other Fallacies

Actually, I struggled to see a clear definition of race from Chris too, so accusing Chris of Straw Men would be unreasonable, so I won't.

I intend to show that if race is indeed meant to be scientifically valid, it must be pinned to the playing field of science, not of social community.

So, for race to be a valid term it cannot be of the social community, I wonder how many sciences we lose with that statement. The problem here is that social sciences are a science. Racial scientific investigations are based on determining issues in social settings. As such Chris is attempting to begin his her attempt to define based on a contradictory premise.

Perhaps Chris could clarify.

If we are going to scientifically validate race, it really has to be on all fronts.

This is another form of fallacious reasoning. The all or nothing ploy, Chris is appealing to say that race must have validity not just in one or some branches of science, but all. To demonstrate race is valid and useful in science I need demonstrate this for only one branch of science, this is actually a pretty trivial ask. I am surprised to be having this debate because race appears in the title of so many scientific papers the mere thought of the need for this debate surprises me.

the concept of race is more than just the skin color, hair texture, or other physical features--it’s the personality and intelligence as well.

I am repeating this here because this one uses the all or nothing fallacy again, but in a differing form.

Chris's Post

Chris opens with an apology, I feel he she really would rather not use the term "race" at all. Mostly I'd prefer not to either, however it is entrenched in our political process by deniers, affirmative action activists and racial oppressors (racists if you like). These people use these concepts to twist our understanding of government policies with respect to ethnic groups; they lie to us. It behooves us to have a clear and valid understanding of what the real issues actually are. To do this we need to approach the meaning of race honestly.

Chris relies on Lewontin in an attempt to demonstrate that race is genetically meaningless. While I am happy to accept other far more detailed information exists, this is not the same as saying that race describes nothing.

For the vast majority of human genetic variations, classical racial categories as defined by a combination of geography, skin color, nose and hair shape, an occasional blood type or selected microsatellites make no useful prediction of genetic differences.

Lewontin is wrong, the example in my prior post with the 50 SNPs identification addressed this (link) (http://www.biomedcentral.com/content/pdf/1471-2164-8-68.pdf), I wanted to find another way of showing this with some examples of areas where different medical conditions occurred more in some racial groups. The article (http://www.post-gazette.com/healthscience/20020507hgene0507p3.asp) I found in the post gazette would be funny if it wasn't so sad. The dance of political correctness much be exceedingly droll.

The problem is, what constitutes a race might seem obvious to most people, but race doesn't mean much from a biological standpoint.

That's why most scientists say race is a social construct, not a biological one. In other words, social rules determine what races are and what they mean.

Sickle cell anemia, most closely identified in the United States with black Americans, is often found in African and Mediterranean peoples, ...

Conversely, cystic fibrosis is a common genetic disorder in people of northern European descent, but far less so in Africans.

Those are both diseases that are caused by a single mutation, so the genetic link is clear-cut. [sickle cell and cystic fibrosis]

... , African Americans seem to have a higher frequency of one mutation that reduces the liver's ability to break down certain tricyclic antidepressant drugs, ...

Blacks also have a higher frequency of a mutation that increases the speed at which a newer class of antidepressants, such as Prozac, take effect.

About 8 percent of whites carry a mutation that would cause trouble with warfarin, a blood-thinning drug, compared to 2 or 3 percent of blacks.

Last year, the New England Journal of Medicine published a study showing that African Americans with heart failure were less likely than whites to benefit from a type of drug known as an ACE inhibitor.

The above points are in order, however the following one I have pulled out of sequence

"Race is a very crude predictor of any sort of differential response to medication."

For race to exist scientifically and be useful it can and does sit in with any number of other crude predictors. When a doctor does a patient history the fact that sickle cell anaemia occurs more frequently in particular races does not provide the diagnosis, it provides a pointer to help the doctor select between diagnoses on the basis of comparative probability. This is science, this is useful and even the deniers accept it.

Even Chris accepts, or why call race a tool?

Well, I think we can safely toss a tool that can’t be used to aid society. If one can’t change race, and crime is associated with race, then what can be done? Creepy things like eugenics, or racial uplift programs, I suppose.

I liked the (perhaps unintentional) David Brin reference (The Uplift Saga (http://www.davidbrin.com/upliftbooks.html)) these are a damned good read.

(I have often entertained the thought that the Hawaiian race was carefully (or unthinkingly, hard to tell) being culled from existence by federal and state government programs, btw. But that’s my dreadful secret conspiracy voice speaking.)

Creepy, Chris is debating on the premise that "race" is scientifically meaningless, or is he she. Perhaps he she has some inside knowledge of the Hawaiian pogrom.

James mentions the Kuhnian paradigm will eventually replace race with...something, possibly genetic markers, and I suspect he is correct. In the meantime, should we find a use for race in its common meaning, or for race in its genetic meaning? Would it ever be more useful than other tools of analysis?

Chris and I agree here anyway, however until we have pocket gene sequencers incorporating a genetic functional database race will continue to be a proxy for better information. And even when these are available ... we have atomic clocks with incredible accuracy yet counting "one mississippi two mississippi" remains a valid proxy, if used appropriately.

The Hidden Agenda

I think what is behind a desire to eliminate race from the scientific community is a back door effor to have various research programs eliminated. Studies that correlate IQ and race, life expectancy and race, income and race, obesity and race, etc. But herein lies yet another contradiction, many of the same people are entirely happy to turn around and use the very plight of races that these measures quantify as a basis for affirmative action programs.

A Question

How does one have both? Denial of race and the support programs to correct racial inequalities that somehow exist / don't exist :confused:

What's reasonable

The reality is that while the UNESCO statement was entirely reasonable in stating
"biological differences as exist between members of different ethnic groups have no relevance to problems of social and political organizations, moral life and communication between human beings"

Try telling the NZ Maori party, one of our political parties that.

The problem is that moving to a view that no biological differences exist is not credible and at odds with the scientific evidence.

[KnightWhoSaysNi]

Chris Porter, please note that your next statement is overdue. You will have a 3 day grace period to submit it.

Thank you,

KWSN, FD Moderator

[Chris Porter]

I feel like I’ve been roped into a scam, unfortunately, whereby I agree to discuss the issue of whether race is a scientifically valid concept, but my opponent has just made a claim that the definition I use isn’t proper. I tried not to bring a strawman to the argument, to use what I thought was the typical definition of race, one dependent on what people normally construe race to be--a specific mixture of phenotypically obvious characteristics coupled with specific ranges of types of behavior. Put simply, race means more than skin color to most people. The Bell Curve would have you think that not only is it skin color, it’s intelligence. The old Southern Baptist Convention would have you think that not only is it skin color, it’s submissive, slave type behavior (for ‘blacks’). Nor did I think race was an issue of country, as the US Census handles it. In this way, I felt I had grasped the handle of race as it’s normally considered, and then looked at the scientific data to see if there is any way to reconcile this understanding of race, and our knowledge of genetics. I didn’t think there was.


Now, I understand that my definition of race was not what James had in mind when he asked for debate. Instead, he wants to use race as meaning genetic variants associated with a geographic local, and that race does not have crisp borders separating these variants. Believe it or not, I accept this to be a true statement--there are geographic genetic variants. But still there are problems. James uses HapMap data analysis to show that people generally know something about their ancestry, and that knowledge can be verified to current HapMap data. However, I cannot see how HapMap data currently supports James in this endeavor to show race is scientifically valid, because it is in its first stages, and has limited it’s data to only three extremely broad geographic groups. I believe using HapMap data to defend the concept of geographically based genetic variation is a problem, because HapMap wasn’t designed to do the job.
The initial design of HapMap was to provide a tool for testing association between common SNPs and risk of complex disorders, and for this purpose, one does want a sample biased toward more common SNPs. However, when one wants to apply HapMap data for other sorts of population genetic inference, this strongly biased sample needs to be used with caution. (http://www.genome.org/cgi/content/full/15/11/1496)

This means analysis using the HapMap data to compare what is normally considered a race is going to be easily verified, because HapMap has chosen the broadest geographical variants to study first, and most humans being tested against that data come from somewhere in those geographic locals. Imagine if HapMap had started the investigation with American Indians, Pacific Islanders, and Australian aborigines. Then the demonstration showing that people are aware of their own ancestry, and that ancestry is mappable to genetic variation, would show confusing results. Indeed I expect that ten or twenty years down the road, when other geographic units are analyzed by HapMap or something similar, that people will still be showing their knowledge of their family history, and that that history will more often than not, map to geographical genetic data, and we can, or might as well, call that race. So in this way, James is correct, race can be thought of a scientifically valid, give a specific definition of race.

There is still the social science issue that is not addressed, that the the analysis is not showing that people can easily identify race in others. In fact the best that can be said for this sort of definition of race, is that it is independent verification of someone’s stated family history, useful perhaps in some way. Using race this way, is indeed, scientifically valid.


Unfortunately, this sort of definition of race is not the typical understanding of people that use the term race. How can that be reconciled? Do we need to teach the difference? Is that possible, especially when the traditional meaning of race is so well entrenched in so many societys? I don’t know.


There are further problems with this sort of definition of race, scientifically valid or not, and that is that purported scholarly tomes are not using the term in the same way, and are in fact, using ‘race’ in broad brushstrokes reminiscent of traditional meanings of race. Research like “The Bell Curve”, for example, is problematic because the authors did not associate race with genetic variations--they associated race with groups of genetic variations. So someone called all the races of Africa as one race--black. How can we tell if they are right? What if it turns out that all the Zulus tested actually are more intelligent than Asians, but have been lumped into African because they’re on the continent of Africa? Broad brush stroke applications of this continental grouping of genetic variations run a grave risk of being completely inappropriate, and still, there is no current way of looking at the data and making such distinctions. We can’t even tell if we are wrong. To me, that makes the defintions used in scholarly or governmental issues not scientifically valid. As Lewontin points out, and so does James, using race in this way is counterproductive, it’s not finely sorted enough to distinguish the issues some government programs, scholars or doctors wish to address. You say someone is black? So they may have sickle cell anemia. However, some whites also have sickle cell anemia. Issues like medical conditions, or intelligence can’t be determined by obvious phenotypes, and sorting people by these broad means is not scientifically valid. You can’t dependably use race alone as a factor in medical analysis, it needs to be an adjunct to medical history. As such, it’s useless alone, it must be coupled with family history knowledge. Thus, race as a scientifically valid concept being useful is still not yet demonstrated, especially because current genetic markers are not sorting geographic genetic variation finely enough to be useful. So Tobagoans may have specific propensity to an illness? Even the article James quotes says, “There is no African gene, no Caucasian gene, no Asian gene.” Indeed, describing as successful, the aspect that people can self-identify what continent their ancestor came from is not accurate enough in terms of medical issues, when real needs are for genetic markers for far smaller populations than continent wide.

I don’t have a hidden agenda, I just think the research at this point in time is insufficient to provide scientific standing or backup to social policies, and to be useful for medicine. I’m not saying it won’t become more useful in time. But I also think that race is not being used as a scientifically valid concept in government policy, nor is it usefully applied in medicine, except as a verification of potential problem, not as a diagnostic tool.

[James T]

Concluding my position

Race is a Scientifically valid and useful concept

Chris's effort

Chris's defence seems to rest on the ability to define race in such a way as to make all who consider race a valid term to be racists.

Put simply, race means more than skin color to most people. The Bell Curve would have you think that not only is it skin color, it’s intelligence. The old Southern Baptist Convention would have you think that not only is it skin color, it’s submissive, slave type behavior (for ‘blacks’).

And then to express surprise and chagrin as I point out that this is a Straw Man definition.

I was disappointed to see this comment from Chris.
I feel like I’ve been roped into a scam, unfortunately, whereby I agree to discuss the issue of whether race is a scientifically valid concept, but my opponent has just made a claim that the definition I use isn’t proper.

One would think from her post in setting up the debate she understood.
Since part of the scope of the discussion is that race is scientifically valid, I would presume that part of the argument will indeed be definitions, in an effort to demonstrate just that. Then James will probably want to say something about usefulness of race as a concept.

I suppose this should have come as no surprise me. Nor does it surprise me that Chris essentially agrees with my position:
This means analysis using the HapMap data to compare what is normally considered a race is going to be easily verified, ...

The fallacies of denial

Let's look at some of the basic fallacies used in denying race has a valid use.

The most common one is a variation of the Continuum Fallacy where the fact that the concept is fuzzy or open is supposedly sufficient to demonstrate victory. Typically people who use this approach will selectively apply it to concepts they dislike and allow Special Pleading on the concepts they consider acceptable.

A variant of the Continuum Fallacy is famously illustrated in the argument of the beard. A man with one whisker on his face does not have a beard, if he grows a second whisker ... still no beard, a third whisker ... still no beard. So repeatedly adding one whisker does not make a beard.

I showed that even biologists are guilty of this contradiction, both in terms of their treatment of the problem of species and in terms of a specialist subset of their science where gene mapping is able to identify racial groups as aligned to self-identifying groups.

Another approach in denying race is to craft an invalid but reasonable sounding definition. This is the most common approach. Certainly this is the one that Chris Porter is guilty of. Chris's definition clearly also defines any person who believes races exist as racist.

Inconsistency is also a common issue for those denying race. It is common to see both denial of race and support of affirmative action. It's also not uncommon to find people who deny race exists while still maintaining a clear understanding of their own heritage - cultural, familial and genetic - which is after all what race is really used to mean.

The use of Race in Science

There are a number of places where race appears in scientific endeavour. Despite biologists special pleading and straw men, genetics have found a way to identify people in a way that corresponds to their self-identified racial origins.

Medical conditions are more prevalent in certain racial groups, some of these conditions are genetically linked with race, the drug Bidil is instructive in this regard. Doctors don't depend heavily on race for diagnosis, it just one of the many indicators they note. Though one wonders what politically correct terminology they use in their notes.

In social sciences race plays a far more central role in science. There are a wide variety of social indicators that vary with race: life expectancy, obesity, income, imprisonment rates, intelligence. Note that these are mean group figures, it's not really valid to apply these to individuals.

These figures, in the social science arena, are commonly used to aid in identification of where racial bias in society may occur. As such they are one type of shortcut for identifying issues. Provided this information is treated carefully, and the corrective actions are structured appropriately (sorry you are white, there is no aid for you!), this is an effective approach.

So I can claim scientific value with ease in the social sciences and with validity in the medicine and hence, by proxy, in biology.

Finally

You know who you are, where you come from, what your heritage is, familially, culturally and racially. It should come as no surprise to those who recognise the common traits in your family that geneticists are able to identify race from genetic data. It comes as none to me.

Next time you want to address a racial issue consider, if you deny race exists at the outset then shut up. You have already stated you have nothing useful to say on the topic.

[Chris Porter]

I think James wants an overhaul of the term race, from social, cultural, and political definitions to one of geographical genetic variation. I’d like James to think about why that should be necessary. I’ve been playing along with his new definition, now that I know about it, and I don’t see the value in calling this new definition race. In fact, I think from a social and cultural perspective, it’s actually harmful, and would be better of distinguished by some other term, rather than trying to shoehorn a new definition into an old and dubiously valued term. Still let’s soldier on...

Taking quotes out of context can be dangerous, as shown by James, taking part of a quote I made, “ This means analysis using the HapMap data to compare what is normally considered a race is going to be easily verified, because HapMap has chosen the broadest geographical variants to study first, and most humans being tested against that data come from somewhere in those geographic locals.” By quoting only part, “This means analysis using the HapMap data to compare what is normally considered a race is going to be easily verified,” James has made it seem like I agree that HapMap is currently a good indicator of a method of discriminating race. I do not, nor do other researchers believe this to be the case of adequate means of distinguishing race.

Let’s look at HapMap. It’s designed to discriminate different SAP over a set, specific, and above all, narrow group of genetic markers. Narrow, in the sense that it’s data can only correlate to three landmasses of this planet. Narrow, in the sense that it doesn’t have the ability to distinguish individual geographically-closely-tied groups from other groups on that continent. That is, it can’t distinguish Zulu from Khoikhoi. It lacks the any data to distinguish Australian aborigines from Maoris. Narrow, in the sense that, even were we to agree that the new definition of race is a set of geographically tied genetic markers, HapMap still can’t distinguish what we consider to be those races, other than amorphous, immense groups of people. This is not how people would actually use the current definition of race, nor would it be how a person would use the new definition of race as a set of geographically tied genetic markes. How do I know it wouldn’t be used like this? Because I agree Maories could conceivably have genetic markers that distinguish them, and thus would be a race. However, HapMap has no way to distinguish this purported race, nor data. So I don’t agree that HapMap can distinguish race, rather, I think it’s only capable of distinguishing some continental groups of races.

I’m willing to agree there may be genetic markers that distinguish Maori from Germans. I’m willing to call that race. However, I’m not willing to agree that we know what those markers are, that we have the ability to visually distinguish races other than our own, and that we have the need to use race-in any definition that requires the use of genetic markers, in social services, when other means are currently better measured.

Further, I need to bring back into focus something rather important that has yet to be tested: the purported ability of people to distinguish the race of someone else. This has been the age-old method of distinguishing race. But, yet, it’s not all that accurate. Not only do humans fail to accurately identify race, so do biometric programs. If race is easily phenotypical, we shouldn’t have the problem we currently have of distinguishing races. Yet it is not the case--instead, the only method we have is self-identification of family history, and verification of that family history by genetic markers. Not of use when discussing issues of social or cultural needs, because the self-identification alone is surely adequate.

While my hypothetical “Zulus are the smartest race” sounds silly--it’s actually immensely important, in that we don’t currently have the tools we need to distinguish race in a manner that can discriminate “smartest race” from any other race at all. HapMap certainly can’t do so. So books like “The Bell Curve” are using data sets that lacks the ability to discriminate race, yet claims to be all about race as a set of genetic markers. If they are using this new “geographically tied genetic markers” definition to discriminate race, and those markers are too coarse--not fine enough, to discriminate between groups of people in Africa, or Asia, or Europe, then these books are not valuable to us. They equivocate or actually lie about the ability of the data set they use and it’s ability to discriminate by race. Lumping all Africans, or all Asians, or all Caucasians, into one race, is not what ‘geographically-tied genetic markers’ means, if such races as American indians, Maoris, and Australian aborigines are left undistinguished as being races. If these books are to be of value, it has to be shown why lumping races together and calling the groups one race is a valuable tactic to pursue. HapMap doesn’t do that.

Is race scientifically valid enough to be of use, even this new definition of geographically-tied genetic markers? No, not really. Sure one can use specific genetic markers, and find out if people that have those markers are susceptible to this or that disease or medical issue. But since those genetic markers can show up in people who aren’t phenotypically identifiable, and lack other markers of this or that race, calling them by a race seems less than useful. Some whites have sickle cell anemia. Some blacks (quite a lot, actually) don’t have sickle cell anemia. It’s good to have the ability to test for these issues, and it’s good to have the ability to verify if a patient may have propensity to a disease based on their family history. But describing this issue as one of race is surely a misnomer, since it’s actually an issue of what genetic markers a person has, regardless of their family history. Thus using this technology and calling it a useful or valid while at the same time calling it race seems a bit of a stretch.

So what use is race today, if the only accurate methodology is a genetic exam, and those genetic exams are not all that accurate? How is this helpful to our societies? Need we test every individual, find genetic markers, and give them a race label? In what way is that useful? Historically, the US used to buy and sell people of a certain skin color. But skin color is not race, surely, not by this new definition. So there need be no reparations for our past actions by skin color. Instead, social remedy could occur by assessment of family history, or genetic marker. Likewise with the the ‘problem’ of Maoris in the jail population, or the blacks in the US prison population. The issue is far less likely to be one of genetic markers, but historical antipathy toward one social group or another, in other words, simple racism, not race ‘failings’. Thinking that assessing the genetic markers of a population is going to show some cure for the social ills of past discrimination does not strike me as a reasonable expectation for the technology. A more likely issue is one that addresses differences in cultures in a country, rather than addresses genetic markers. In this way, having the ability to distinguish race by genetic markers is not as useful as addressing past discrimination, regardless of race.

Finally

I think you will find that telling people to shut up is not a particularly valuable discussion tactic if you want others to listen to you carefully.

[KnightWhoSaysNi]

The formal debate is now complete. We would like to thank James T and Chris Porter for their participation. Discussion can be continued in the peanut gallery.

KWSN, FD Moderator