Jul 22 2016

Does Race Exist?

World_Map_of_Y-DNA_HaplogroupsIs Pluto a planet or a dwarf planet? Are these two categories even meaningful? The reality is that objects orbiting our sun occur on a continuum from asteroids to planetoids, dwarf planets, and full planets.

Humans like to categorize, however. It helps us wrap our minds around complexity, gives us convenient labels to help sort our knowledge, and hopefully the categories reflect some underlying reality.

Categories often begin as purely observational. We label diseases by what they look like (their signs and symptoms), and then later may have to recategorize them once we know what causes the diseases.

Prior to Darwin, taxonomists categorized all of life according to superficial characteristics. These categories sometimes, but not always, matched the underlying reality of evolutionary relationships. We now have a different system of taxonomy called cladistics, which is purely evolutionary. That’s why birds are now dinosaurs. 

Issues of Race

There is rarely any political implications to categorizing the subpopulations of deer that inhabit North America. When we start categorizing humans, then suddenly the political and social implications are huge, and the very process of categorization comes under scrutiny.

There are many who claim, for example, that human race is a social construct without any underlying scientific validity. For example, Michael Hadjiargyrou, Chair of the Department of Life Sciences, New York Institute of Technology, wrote a commentary in 2014 (which is making the rounds again on social media) in which he claims:

It is history, not science,that reveals how the concept of different human “races” arose, how the term has become widely misused, and how it continues to pervade our planet. In fact, the word race has come to symbolize the division of humanity into segments, divisions that often lead to conflicts.

While I understand where this position is coming from, it has never sat right with me. Perhaps it is partly due to my background as a physician. Within medicine we routinely consider a person’s ethnic and racial background to help determine their risk of various diseases. Medical studies are often stratified by race, and no one thinks twice about it. In fact I have often heard criticism that the failure to do so is a disservice to minorities who are then underrepresented in clinical studies, which means the results might not be broadly applicable to them.

I completely agree with Hadjiargyrou when he writes:

We all evolved from the same ancestors and are, indeed, all virtually genetically identical to each other, making us a single race.

Although I would substitute the word “species” for “race.” Humans are all one species. While we are an outbred species with a great deal of genetic variation, resulting from the fact that we range widely around the world and have been for thousands, perhaps even hundreds of thousands of years, there are no subspecies of humans. We share 99.9% of our DNA. In short there is much more that unites than divides us.

And of course I fully agree with the political point here – every human is equally deserving of dignity and respect and should be treated as an individual, not as a member of some arbitrary group. It is also true that group identity tends to be divisive. We are tribal and tend to think in terms of in-groups and out-groups.  As we become more global it is important to emphasize how similar, rather than how different, all humans are.

What makes me uncomfortable is subjugating the science to a political goal, no matter how noble that goal is. There are two big problems with this. First, it’s important to get the science right. Distorting scientific thinking is pernicious. Second, if we make a valid political position dependent upon a scientific position, then the political position is vulnerable to new evidence, and it almost forces you to distort the science.

Regardless of what we discover about the genetic variation of humans, it is clear that all humans should be treated as equal as an ethical principle. This principle is not dependent upon the scientific fact that race does not exist.

So does race exist?

The scientific question of whether or not race exists is, in my opinion, not completely objectively answerable. It depends. Is Pluto a planet or dwarf planet? Astronomers can reasonably disagree about where to draw the line.

What we can say is that genetic variation within the over 7 billion humans on the planet is not homogeneously distributed. There are populations where certain genetic traits tend to cluster. There is no question about this. This clustering of genetic variation is determined by ancestry, which largely reflects geographical origin. The biggest differences tend to reflect the continent of origin.

We tend to refer to continent of origin as race, and further subdivisions as ethnicity. These are arbitrary dividing lines, just as the line between tall and short is arbitrary. That does not mean they have no basis in reality – that is the false continuum logical fallacy.

Hadjiargyrou makes a couple of specific scientific arguments. Perhaps his most compelling argument is this:

Genetically speaking, studies have shown that there is much greater genetic variation within a given human population (e.g., Africans, Caucasians, or Asians) than between populations (Africans vs. Caucasions), indicating that human variation cannot be subdivided into discrete races.

I agree with the premise but not the conclusion. Genetically speaking, all vertebrates are fish. There is much greater genetic variation within the fish clade than there is between fish and other vertebrates. Land dwelling vertebrates represent a tiny twig on the vast fish genetic tree.

In the exact same way, there is much more genetic variation within Africans, then between Africans an all other human populations. This simply reflects the fact that humans lived in Africa for a long time, evolving extensive genetic diversity, and the population that migrated out of African represents a tiny twig on the African genetic tree. We are all Africans in the exact same way that we are all fish.

The point is that looking at genetic variation within and between groups can be deceiving. It also means, however, that the concept of race can be superficial and likewise deceiving. Lumping all Africans into one race greatly underrepresents the genetic diversity within Africans, while counting Europeans as one race greatly overplays the genetic diversity between African and Europeans.

So Hadjiargyrou has a point – race is scientifically a bit arbitrary and gives us a distorted view of genetic variation. That does not mean, however, that there are not discrete subpopulations of genetic variation within the human population, roughly reflecting continent of origin. This is further complicated, of course, by exchange of genetic material among all human groups. But this has not occurred to the point that the groups themselves have vanished.

Hadjiargyrou’s second point is much weaker.

Biologically speaking, one clear example is that most diseases afflict all of us — diseases like cancers and cardiovascular and neurological disorders, as well as viral, microbial and parasitic infections.

As I stated above, any physician would recognize this statement to be highly misleading. There are genetic disease that exist exclusively in some races or ethnicities and not in others. Africans develop sickle cell anemia; Europeans and Asians don’t.

It would not be prudent to erase all consideration of genetic background from medicine. In fact, if anything such considerations are increasing with our increasing knowledge of genetics and ability to sequence DNA.

Conclusion

While I agree with the political and social point, I disagree with the strong statement that race does not exist or that it is purely a social construct. This is clearly an overstatement of the reality.

To be clear, I am not taking the strong position that race does exist. Rather I am saying that the question is inherently arbitrary, as are all questions of categorization. You have to make choices about which characteristics are meaningful, and where to draw lines. Nature tends to exist on a continuum, and so any dividing lines will have to be arbitrary to some extent.

The social construct of race is distorted from the scientific reality, with that I agree also. The underlying scientific reality is more complex, and the tree of human genetic variation is probably very different from what most people would imagine based upon social and historical categories.

I simply would not go so far as to conclude that scientifically race does not exist. Humans do exist in discrete genetic subpopulations that roughly reflect geographical origin. These genetic subpopulations are useful when it comes to medicine and predicting risk of various diseases. I predict that the medical profession will go on completely ignoring the social debate about race, and treating it as if it exists, because it is a useful medical construct.

Rather than making strained scientific arguments that are not completely valid, it is far better to simply divorce the scientific details from the ethical principles. Even the fact that humans are all one species is not necessarily important to the ethical principles. What if a subspecies of humans existed? Wouldn’t they deserve respect and dignity as sentient beings?

The lesson is clear, and I think Steven Pinker was the first to articulate this (as far as I know) – don’t tie a valid ethical principle to the details of science. It forces you to distort the science, and makes the ethical principle vulnerable to false refutation.

70 responses so far

70 thoughts on “Does Race Exist?”

  1. Cathy Newman says:

    Thanks for sharing this. I’ve been struggling with this issue for a while as a population geneticist and also a Southern white woman who regularly thinks hard about racism. Your arguments here are the first I’ve come across that I feel reflect the reality of biology without ignoring the historical social baggage forever linked to the concept of human race.

  2. mumadadd says:

    “Rather than making strained scientific arguments that are not completely valid, it is far better to simply divorce the scientific details from the ethical principles.”

    Yep. I think there are a lot of people who confuse the two. E.g. stating that there are differences, statistically on the average, between male and female brains means you are endorsing social reinforcement of gender roles.

  3. Bill Openthalt says:

    mumadadd —

    E.g. stating that there are differences, statistically on the average, between male and female brains means you are endorsing social reinforcement of gender roles.

    This might be a variation on the “appeal to nature” argument. It has, for example, been used to legitimate homosexuality (if there is a gene for it, then it’s OK). Some vegans argue that humans are not omnivores to shore up a moral concept (that one should not eat moving critters).

    With whom to have sex, what to eat, and whether or not there should be distinct gender roles etc. are moral/societal issues, and cannot be settled by science.

  4. mumadadd says:

    Bill,

    I agree — this does seem like an appeal to nature fallacy. Something tangential that I’ve noticed (and I’m happy to be shot down here because this is my own observations/musings alone): in the public consciousness, whatever is accepted by society is deemed ‘natural’; e.g. homosexuality is genetic/biological and part of the natural range of sexuality, whereas paedophilia is not biological, but down to environmental factors such as abuse in childhood etc. People (or social psychologists at least…) seem to have difficulty accepting that someone could be ‘naturally’ sexually attracted to children, or that environmental factors could play a part in sexual orientation in the ‘acceptable’ range.

  5. mumadadd says:

    I’m thinking my example was a little off. Let me just say this: ‘natural’ has a halo effect, and I think people will, in addition to believing that ‘natural’ = good, also believe that bad = ‘unnatural’. In the case of human behaviour, that means attributing good behaviour to internal causes and bad behaviour to external causes.

  6. That is the naturalistic fallacy, also confusing “is” with “ought.”

    It really doesn’t matter socially and politically whether or not there is sufficient and distinct enough genetic variation to justify racial categories. It’s a complex and interesting scientific question. I think its reasonable scientifically to say that races exist, although their nature is probably different than what most people think.

    It is misguided to argue that race does not exist because you don’t like the alleged social and political implications. Rather, regardless of what the science says about race, we can take the ethical position that we are all people deserving of equal respect and dignity.

    Now, historical and social thinking about race is yet a different issue, which is also interesting, although I wasn’t even discussing that. It is also distinct from the scientific question, although certainly historically the two things have influenced each other. I agree that social constructs of race have influenced scientific thinking about race, especially in the past, and to a lesser extent today. This is basically pseudoscience in the service of a political agenda.

  7. I agree with the sentiment that we shouldn’t base ethical principles on current scientific conclusions, but not doing so does lead to some interesting quandaries. For example, how should we treat animals? Should we value sentient life, or intelligent life, or life with some other ineffable human quality? If so, of course, then we are at the whims of data.

    But if not, it’s hard to have an a priori ethical principle other than valuing all animal life, or possibly all life period. Most people wouldn’t be on board with that, because it would involve not eating meat, or weird paradoxes about what do with carnivores, or questions of utilitarianism that are icky. Instead we might adopt a principle such as, “Treat all life with dignity,” where dignified treatment is a little more nebulous and doesn’t preclude killing things when we want to.

  8. Bill Openthalt says:

    Ori Vandewalle —

    Valuing life doesn’t mean not eating it — plants are also life, and we have to eat something. Humans extend the benefits of humanity to what is dear to them. Many a dog-owner would choose their pooch over an unknown human anytime. Rabid animal rights activists will murder humans to save cows, like rabid pro-lifers will murder doctors who perform abortions. The (arbitrary) moral norms of this moment and this location will determine what can be eaten, and how it can be grown and killed (do you eat horse-meat, do you use Round-up…). Our run-ins with the anti-GMO crowd show how difficult it is to have science influence (let alone override) moral convictions.

  9. Well, yeah. Shying away from the science in the direction you describe gets you another problem, which is massive internal inconsistency.

  10. Ivan Grozny says:

    Your main point that the biological boundaries between various “races” are arbitrary agrees rather well with a proposition that you seem to question: that the very category of “race” is “socially constructed”. How otherwise we came to associate race with the skin colour, rather than say, eye colour, if not by social construction? Why is it so self-evident that a blue-eyed Norwegian and dark-eyed Italian from Sicily are of the same race, rather then a dark-eyed Sicilian and an equally dark-eyed Nigerian? It has to do with culture, history, religion, rather than with any biological fact. The very principle of classification (skin color) is socially constructed.

  11. mumadadd says:

    Ori,

    “I agree with the sentiment that we shouldn’t base ethical principles on current scientific conclusions, but not doing so does lead to some interesting quandaries.”

    I don’t think the principle was meant to be generalised beyond how we treat other people based on race, sex etc. I certainly wouldn’t generalise it beyond: treat people as individuals rather than the statistical averages of the various categories they belong to.

    If we’re talking about treatment of other species, and whether it’s ethical to kill, eat perform medical testing on them, then we should absolutely be informed by the science — for example on their capacity to suffer, their psychological experience of pain etc.

  12. Okay, it’s only a principle for how we treat humans, not non-humans. But how do we tell humans from non-humans, and thus know when to use the principle? The concept of species is a scientific one, and biology tells us that there is no sharp, diving line between species. Yes, we can clearly place humans and fish into different categories, but how are we to tell with closer cases (our descendants, genetically engineered hybrids, cyborgs, etc.)?

    If we attempt to draw such a line based on scientific data, then we run the risk of applying our principle to humans when we’re not supposed to. If we don’t judge human from non-human based on empirical data, then we’re left with (a) hunches or (b) philosophical principles. Hunches are notoriously inconsistent, and philosophical principles are notoriously boring.

  13. dohashi says:

    I feel like Steve has missed the point a bit here. I don’t think may people
    question if “race” exists. The question is: are races defined biologically or
    culturally. Those who claim races are biological would say that the specific
    races have unique genetic traits that make the races unique (in some
    meaningful way). Those who argue that race is cultural would say that you
    would not find genetic justifications for races. So ultimately, does
    someone’s genetics justify creating a “racial” category for them.

    Steve is kind of going the other way: given an individual’s race, does that provide
    useful information about them, in particular their biology? I suspect most people would agree it does, but only in a statistical sense, and really only because we don’t have access to someone’s genetics directly.

    I generally fall on the “race is a social construct” side of
    things and so there are a few points that I would like to bring up. First,
    do we have a definition of race that is not, in some way, cultural? Race is
    generally taken to be people who have certain genetic traits in common, due
    their ancestors living in a particular geographic location. However which
    traits? Which locations? Sometime in the past, fairly arbitrary lines were
    drawn on a map to create the locations that we now use to try to define races.

    Steve’s example of sickle cell points this out, he states “Africans develop
    sickle cell anemia, Europeans and Asians don’t” however “Sickle cell trait is
    an inherited blood disorder that affects 1 million to 3 million Americans and
    8 to 10 percent of African Americans. Sickle cell trait can also affect
    Hispanics, South Asians, Caucasians from southern Europe, and people from
    Middle Eastern countries” (from
    http://www.hematology.org/Patients/Anemia/Sickle-Cell-Trait.aspx). The
    “African” lines on the map don’t actually line up very well with sickle cell
    genetics.

    Another issue that points to race being a social construct is the fact
    that different cultures define races differently. In the past groups like
    “Irish” were considered a race. Consider what constitutes races in Brazil:
    https://en.wikipedia.org/wiki/Race_and_ethnicity_in_Brazil#IBGE.E2.80.99s_racial_categories

  14. mumadadd says:

    dohashi,

    I think the response to your post is the whole of Steve’s post, excluding the point about sickle cell anemia if that’s inaccurate.

  15. mumadadd says:

    If it’s a social construct then it’s intrinsically non existent. If you try to answer the question based on stuff that doesn’t go away if people stop believing it, you get to Steve’s point about genetic sub populations.

  16. JohnW says:

    Southern Europeans and south Asians do get sickle cell anemia in fact. In genetic diagnostics, questions of self-identified “race” can be somewhat useful but also misleading. More useful is family ancestry and geographical origin when known.

  17. Cheryl says:

    Steve — you consider a patients’ race and ethnicity when considering their health. But I think that is fraught with problems, too. How much black does one need to be to be at risk for sickle cell anemia? Especially since many folks might not be aware of their actual racial heritage. And most Hispanics identify as Caucasian, so how does knowing their ethnicity come into play? Is this perhaps a short-hand question for something else, maybe socioeconomic status (which, again, could be terribly misleading)? Aren’t you assuming a lot based on race and ethnicity? I would think you’d want to specifically ask, for example, if a patient is or is descended from Ashkenazi Jews. Why beat around the bush?

  18. dohashi says:

    mumadadd,

    Its interesting because I don’t think that I actually disagree that much with what Steve has written, instead my point was I don’t think he is actually addressing the question that he has posed in his title.

    “Does Race Exist?” lets assume this means “Does there exists a biological (genetic) justification to group people into races?”. What I think Steve is actually arguing is that given that people have been divided into races, is it useful to look at their race for certain purposes, say medical diagnosis. He thinks so, and I can agree, so long as everyone is aware of the limitations of such a “genetic proxy”. However that does not address is if there is/was any scientific justification for the creation of those particular racial categories to begin with. If there isn’t, then the races that we recognize today are just cultural artifacts with no meaningful biological bases.

  19. mumadadd says:

    “What I think Steve is actually arguing is that given that people have been divided into races, is it useful to look at their race for certain purposes, say medical diagnosis.”

    Disregard the “given” and you’re there.

  20. TheGorilla says:

    I’m going to assume that Dr. Novella never intended to address anything more than whether or not race is an instrumentally successful concept in medical science, since that’s clearly what the post aimed to accomplish; however, I’m going to make some comments “as if” this were intended as an *actual* response to the question asked in the title.

    To attempt to answer the question of whether or not race exists by looking at the physical makeup of human beings is to make an assumption about what race actually is. In this case it would be to treat race as a static concept, the validity of which depends on whether or not it picks out actual physical features of the world. This is the same approach as when people talk about a ‘gay gene.’ The problem is not just that race is a “social construct” but that race is not a static thing; racialisation is a dynamic process that is continually unfolding, and, in this light, searching the genetic code or epidemiological data for “race” is wholly misguided — it’s searching for a static marker(s) of a dynamic process.

    On top of that, racism is about social relationships of oppression and, here’s another very important part, historically oppression has been justified by race, not prescribed by it (which, when you consider the history of Western imperialism, would sit nicely with racial categories matching up with some different inclinations to illness). The examine the genetic code attitude is totally useless in advancing this conversation.

    Anyways I’m sorta rambling and this is sorta ugly due to trying to not write an essay, but hopefully the point comes across. And let me reiterate I do not believe Dr. Novella was attempting *this sort of answer.*

  21. bachfiend says:

    There were originally 4 human races, and they all lived in Africa (I’m artificially restricting ‘human’ to members of Homo sapiens – Homo erectus, homo neanderthalensis, denisova and floresiens were all human too).

    And then around 70,000 years ago members of one human race left Africa to populate the rest of the world, so everyone outside of Africa were one race – Europeans, East Asians, Melenesians, Australian Aborigines, American Indians, etc. there’s less genetic variation between Scandanians and Papuan highlanders separated by 20,000 kilometres and around 40,000 years than there are between populations of common chimpanzees living several hundred kilometres apart.

    The external differences being due to several dozen genes.

    And then within the last 500 years, members of one human race came to West Africa to abduct members of another human race to bring to the Americas as slaves. Which is where Steve’s comments about race being real in America, with Afro-Americans having different disease patterns and different responses to pharmaceuticals.

    Whether there’s any biological basis for lower success of Afro-Americans since emancipation or whether it’s all social due to lack of opportunity is a separate question. I think that it’s lack of opportunity.

  22. Daniel Hawkins says:

    Steve,

    You have an error in your cladistic analogy to fish. The underlying issue is that the term “fish” is not a proper grouping in systematic biology. According to Wikipedia, a fish is any member of a paraphyletic group of organisms that consist of all gill-bearing aquatic craniate animals that lack limbs with digits. In particular, it excludes all tetrapods.

    Essentially, before we understood genetics, we lumped together a bunch of disparate clades into a single group we called “fish”, so it artificially has enormous diversity. However, even if that problem didn’t break the analogy, it fails for a different reason. Genetic diversity within species is qualitatively different than genetic diversity between species. Speciation is a bit like a phase transition; different rules start to apply. Before speciation, there may be subpopulations for which certain traits are different on average, but for many or most of these traits there is a great deal of continuous overlap. In other words, the standard deviation of the bell curves is such that while you can make statistical statements about populations which are meaningful (and medically actionable from a public policy standpoint), the variation makes their applicability to individuals negligible.

    On the other hand, once speciation occurs, the populations start to overlap less and less for many traits, until there is essentially no overlap between the groups, and the between-group variation is bigger than the within-group variation.

    To give a concrete example, let’s say there is some new disease which only afflicts men less than 61 inches tall. According to the CDC, the average height for white males aged 20-39 is 70.4 inches tall, +/- 2.57 and for black males aged 20-39 is 70.1 inches tall, +/- 2.66 (http://www.cdc.gov/nchs/data/nhsr/nhsr010.pdf). That means that 61 inches corresponds to a z-score of -3.66 for whites, and -3.42 for blacks. That in turn means that the proportion of whites males 20-39 below 61 inches is 1.27 * 10^-4, and the proportion of black males 20-39 below 61 inches is 3.11 x 10^-4. So blacks are 2.45 times more likely to be afflicted by this fictional disease, yet the race is of extremely poor diagnostic value for any single individual, since over 99.9% of both races would not be afflicted. Far better to simply measure the height!

    I’m not claiming that the race is not a useful risk factor to take into account, especially when discussing public health policy. But often these multiplicative risk factors are simply stated, but aren’t put into a meaningful context. If they’re merely a proxy for other measurable values that are different between populations, then race isn’t medically useful in those contexts. I certainly am not familiar enough with all the research to say that most diagnostic applications of race are invalid, but my point is that the numbers can be misleading.

  23. Ranthropologist says:

    I wrote about this on my blog last year, because I get tremendously irritated at the misunderstandings people have about genetic ancestry testing. I agree with a lot of what Steve has written here, and I understand that genetic clustering that occurs geographically can lead to certain genetic conditions being more common in some populations than others, but I still don’t think that that qualifies as a biological basis for race, as most people understand it. http://ranthropologist.com/2015/02/25/ranthropology-class-the-myth-of-race/

  24. ToddfromBC says:

    Another interesting read. I believe good articles make you think and make you ask questions…
    Although I did take some Biology classes three decades ago at University, I now mainly build robots and teach other people how to design and build them.
    Why do we use the word race for humans? In other animals what do we use? I remember something about populations?
    I am curious if its easy to tell someone’s race with a DNA test? Also with (and I hate this expression) so many children of mixed races(my own children), how do they fit in all of this, can you easily tell their race?
    So we know there are certain diseases and conditions related to the genetics of a race,. does the same apply to culture or ethnic groups? Also are there genetic markers for culture or ethnicity?
    If you can identify the difference with science then it is a thing…I guess the difficult part about this is using this for knowledge and not as a weapon? If that makes any sense?.

    And as I went to post this the phrase hybrid-vigor popped into my head. Does that exist in humans?

  25. mumadadd says:

    Dohashi,

    Actually, I can’t fault your logic. Given that racial categories existed long before we understood any genetic basis for them, and along lines that aren’t reflected by by the science, I think you’re right. 🙂

  26. SimonW says:

    The Dr is using it as a pointer. If I as someone with no recent African heritage turned up as a patient he would immediately (and correctly) assume I’m at low risk for sickle cell purely on the basis of my colouring (skin, hair, eyes).

    There are broad statistical differences in medical characteristics between what historically has been termed ‘race’. So medicine finds these useful because your appearance usually reveals heritage.

    Probably doctors also use other correlations which are not politically correct, and say look more critically at the homeless for signs of mental disorder or drug misuse.

    I did want to put “historically” before “geographical” in the article, since we are all moving much more. These assumptions may be less safe with time, but then so will the visual clues we associate with race likely also be less clear cut.

    I suspect within my lifetime, barring it being cut short unexpectedly, we’ll reach the point when near patient medical testing will eliminate much of the need for such guesswork, and thus the utility of using racial characteristics for diagnosis. However those differences will remain, even if they lose their medical utility.

  27. I did not say that race is not socially constructed. My point is that I am not convinced it is entirely socially constructed.

    I did specifically make the point that sharing of genetic material complicates any categorization. In the US this mixing of genetic material is quite extensive.

    My main point is that some argue that race has no scientific existence based in genetics. I disagree with this hard position, because it is based in the false continuum fallacy. This is the notion that categories have to be absolute, unique, with no blending, no fuzziness at the borders, no overlap in order for the categories to be “real.” We simply do not apply those standards in science generally. I think some people are trying to make an exception for the concept of race (whether they realize it or not) because of the political implications.

    If we were considering another species, with the exact same pattern of genetic variation, geographical subpopulations, genetic mixing, etc., I think there would be absolutely no controversy over separating that species into identifiable (if fuzzy) subpopulations.

    Keep in mind, the concept of species itself is fuzzy. There is no clear dividing line between one species and the next, and often scientists debate about this when there are many subpopulations with an amount of disparity that is on the borderline of considering them different species. So, even at the species level we use labels to identify islands of genetic material that we think represent a population that represents an identifiable genetic history.

    If we’re talking about birds, scientists identify the subtlest of variations as subpopulations. Interestingly, when the politics are reversed, these subtle differences are sometimes used to argue that the subpopulation is distinct and deserves protection. The arguments completely flip when the political motivations reverse.

  28. TheGorilla says:

    I think the issue here is that you acknowledge race as a social construct yet, in comparing it to subpopulations of other species, completely strip the social aspect from the equation; nobody (generally, there are exceptions to everything) takes issue with the existence of roughly identifiable subpopulations of human beings, whether this is based on genetic predisposition to certain illnesses or predispositions to obesity based on class, or etc. The difference is not political motivations but theoretical motivations.

    Take what you said from your latest comment as an example here where you refer to the question of whether “race has no scientific existence *based* in genetics.” This question, though it can easily slip under the radar, in effect denies the possibility of race as social construct. Race, understood socially, is not based in genetics — even if there were obvious, rigid genetic differences (meaning a harder line than the actually existing continuum) between races it is only correlated with those genetic differences.

    I think this paragraph from David Camfield’s 2016 paper ‘Elements of a historical-materialist theory of racism’ makes this distinction quite clear:

    “What Charles Mills argues about race is also true of gender and sexuality: these social relations
    are not ‘eternal, unchanging, necessary’ but rather have a ‘contingently deep reality that structures
    our particular social universe, having a social objectivity and causal significance that arise out of
    ourparticular history’ (Mills, 1998: 48). Racial identity can be seen as ‘a
    function of one’s location on a racialized social terrain’ (Taylor, 2009: 188). Race, Mills (2003:
    168) notes, ‘unlike class … roots itself inthe biological, insofar as its identifiers move us to invest
    the physical with social significance’. Thus ‘it is on the body that race is inscribed’; with racism
    ‘political domination becomes incarnated … biologized/naturalized’.”

    Notice that race is *inscribed* upon the body rather than *birthed* from the body.

    Sorry to go full academia here but the root of this misunderstanding really is academic. Race as a social concept is understood in the above way, but looking to genetics for a scientific *basis* for the existence of racism is merely to treat race as, like you mentioned, we treat the question of species subpopulations. This is what is meant when someone says race is only a social construct (process) — not that there is not correlation between race at the genetic level but that race has any *basis* in the genetic level. The question being asked is a sociological one and, to be answered appropriately, needs to use a sociological definition of race.

    It’s not that people are making an exception for race based on political motivations (if this were the case I suspect there would be quite a lot of protest about using race as a medically relevant category?) but that people are talking past each other. Once the terminological issue is resolved the problem itself should disappear. We obviously have no disagreement on the empirical facts.

  29. BBBlue says:

    I think explaining the nature of “race” has more to do with the problem of using labels in general than it does defining differences among subgroups. Steve hit the nail on the head when he said “We are tribal and tend to think in terms of in-groups and out-groups.” That is the context in which most people consider race. However, as Steve has also described, there are functional groups that overlap our cultural definitions, and so the cultural definition may become a kind of shorthand for describing the functional group even when it is a bit imprecise. There are lots of labels that mean different things within different disciplines and misunderstandings often arise when context is not given. To many, as soon as “race” is mentioned, it triggers an alarm that some cultural point will be made or that some group of people are about to be disparaged.

    Some functional groups are appropriately defined based on behavior or culture. Muslims are a cultural subgroup, and within the Muslim world, Jihadists represent a subgroup that is responsible for most deaths due to terrorism in recent years. Is it wrong to point out that fact? Is it wrong to say that, statistically speaking, there is good reason to subject that subgroup to additional scrutiny when evaluating risk? Is the risk of “subjugating science to a political goal” so great in this context that we must not go there? Seems very similar to the discussion about biologically-based functional groups; the baggage that usually accompanies discussions of race, a failure to recognize context, and the spectre of being accused of political or racist motives often inhibits people from engaging in sober, reasonable discussions on the subject.

  30. Marcus_Morgan says:

    Doc

    Its hard to know what to make of the mess of weak facts and logic you write here. You are slipping and sliding around, to make it look like a slippery and slidey issue with the lines mostly if not entirely blurred between races within and between geographical areas.

    You wrote- “What we can say is that genetic variation within the over 7 billion humans on the planet is not homogeneously distributed. There are populations where certain genetic traits tend to cluster. There is no question about this. This clustering of genetic variation is determined by ancestry, which largely reflects geographical origin. The biggest differences tend to reflect the continent of origin.

    We tend to refer to continent of origin as race, and further subdivisions as ethnicity. These are arbitrary dividing lines, just as the line between tall and short is arbitrary. That does not mean they have no basis in reality – that is the false continuum logical fallacy.”

    Doc, its not as slippery and slidey as you make out. You conveniently left out any reference to the most fundamental fact of all. Humans are one species from one geographical location in East Africa, and they migrated over tens of thousands of years to your “geographical areas” and the supposed diversities within those areas.

    It is way too convenient to avoid an overview that shows clear division extending East to Asia as Asian races, a remainder in Africa as Black races, an extension North to Europe as White races, and to the Middle East as mixed races. That’s all you need, and an explanation for it, and the clear medical differences between those races.

    Just stick with a subject embedded in each of those “real areas” and tell us their differing “risks”. You won’t be able to ascribe reasons for the differences, and in fact you will find very few of the real differences, for the same reason you won’t be able to ascribe reasons for the differences, ignorance.

    You have gone nowhere on the issue, and it is really a bad attempt at advertising-style copywriting to distort reality. Include all the facts and use proper logic. Better still, if you want to see how an argument is presented, including race, read my free book without delay http://1drv.ms/1tnKM6f

    Otherwise you will just continue to embarrass yourself.

  31. Marcus_Morgan says:

    I won’t be mean. I will give you a clue to the missing ingredients. Given that is it one species, and it migrated for Asian, White, and Middle Eastern characteristics away from Black, we are talking about genetic variation by migration. Are we talking about differing temperatures, light levels, seasons, or what?

    What affects genetic variation? Is it Epigenetic? Is it subject to adaptation directly to an environment when splicing by Epigenetics, to give specific variations a better chance to become set as mutations?
    Or would you rather close the book by Central Dogma and say its undirected, random based on random successes over eons? They need their heads examined closing a book like that.

    Doc, you are obviously out of your depth on genetics, and generally in your meanderings into lots of areas relying only on a “supposed” capacity for fact and logic. So, read more about the bases for mutation, and whether they are related to some or all of the factors I mention here, and others. You just take an awful approach with no future in your little piece here.

  32. BillyJoe7 says:

    Morg

    Please stop embarrassing yourself.
    Your “book” is unreadable.
    You can find yourself and your book described here:

    https://en.wikipedia.org/wiki/Crank_(person)

    “Doc, you are obviously out of your depth on genetics”

    I take it you have read every post he has written about evolution/genetics.
    Otherwise you are talking out of your ass.
    So instead of passing wind, please busy yourself with the following 86 blog posts:

    http://theness.com/neurologicablog/index.php/category/evolution/

    And please don’t embarrass yourself further by commenting until you have read every one.

  33. BillyJoe7 says:

    Crank =

    Cranks overestimate their own knowledge and ability.
    Cranks underestimate the knowledge of experts.
    Cranks insist that their alleged discoveries are urgently important.
    Cranks never acknowledge any error no matter how trivial.
    Cranks love to talk about their own beliefs.
    Cranks tend to be bad listeners.
    Cranks seriously misunderstand the mainstream opinion.
    Cranks stress that they have been working out their ideas for many decades.
    Cranks compare themselves with luminaries in their chosen field.
    Cranks claim that their ideas are being suppressed.
    Cranks claim that the mainstream are terrified of their revolutionary insights becoming widely known.
    Cranks appear to regard themselves as persons of unique historical importance.
    Cranks exhibit a marked lack of technical ability.
    Cranks misunderstand or fail to use standard notation and terminology.
    Cranks ignore fine distinctions which are essential to correctly understand mainstream belief.

    = Morg

  34. Insomniac says:

    Indeed the logic behind divorcing scientific statement from ethical principles is featured extensively in The Bank Slate, and I don’t know about a similar discussion prior to Pinker’s work.

    Thanks for this careful analysis of this baking hot debate. I’ve come to similar conclusions but as always the reasoning and conclusions are more explicit and well formulated here. Bushes were already cleared but now the path is even more obvious.

  35. EmilKarlsson says:

    I wrote a detailed analysis of the arguments about genetic clusters and discrete genetic races as well as racial medicine at Debunking Denialism in a post called “Genetic Clusters, Racial Medicine and Fishes”.

    There is also a great post about this issue at The Panda’s Thumb called “Continuous geographic structure is real, ‘discrete races’ aren’t” by Nick Matzke

    Here is a summary of the arguments:

    (1) Genetic clustering algorithms artificially overemphasize differences and downplays similarities by the method used to calculate distance measures. This has an especially high impact in low diversity species like humans (0.1% SNP difference between individuals on average). Thus, there is a methodological ‘bias’ built into these methods that magnifies differences. Great in many applications, can be misleading in others (such as this one).

    (2) Genetic clustering algorithms are also often confounded by low sampling density and geographical distance. Thus, what you see is isolation by distance, not discrete genetic races. Thus, together with human tendency to see patterns, this is probably the biggest misconception among those who use genetic clusters to argue for discrete genetic races.

    (3) Taking these things into consideration, an analysis by Serre and Pääbo reveal that human genetic diversity is largely clinal in nature, not discrete races.

    (4) Information about ethnicity or ancestry can be informative in medicine, but it is confounded by disparities in health care, unconscious biases by providers, discrimination and differences in income, education and unemployment.

    (5) The percentage point differences in allele frequencies are often modest.

    (6) Racial medicine considerations has been shown to lead to some misdiagnosis of diseases like sickle-cell anemia, thalassemia and cystic fibrosis. Thus, there are also some pragmatic drawbacks.

    (7) Some icons of racial medicine are not as well-supported as commonly believed.

    (8) Due to the crudeness of ethnic background as a proxy for knowing the gene variant and the advances in technology, we might as well sequence the individual and skip the group level.

    (9) Fishes are technically not a clade.

    For more details, including references to the scientific literature, check out the two posts mentioned in the first part of this comment.

  36. BBBlue says:

    Emil,

    “Taking these things into consideration…”

    These things = algorithms? Have there been no large human genome mapping studies that have revealed genetically distinct groups which could be considered “races.” The conclusion that genetic diversity is largely clinal is based on mathematical models?

    While humans may be a low-diversity species with modest differences in allele frequencies, that does not mean significant genetic differences don’t exist, correct? Is there no evidence to support using “race” as a term in a purely material context to describe genotypic variations among groups?

  37. edamame says:

    Emil very interesting points at your site. I’d say that rather than race being hypostasized fixed with clear boundaries, it is more a continuous trait with unclear boundaries, a multidimensional trait with some clear cut cases (e.g., short-statured bushman from the Kalahari versus a tall blonde Swede) but tons of variability.

    This means that socioeconomic status and other complicating factors will have to be taken into account when doing things like evaluating probability of getting sickle-cell anemia, diabetes, etc.. Ultimately, clearly certain genotypes tend to go with certain diseases. Sure, epigenetics and such are important, and we need to be wary of racial bias.

    There is a fallacy you should be wary of making here, I’m not saying you have made it, but I think you are flirting with it. Namely, because there are lots of insignificant-seeming gradations that could lead you from property X to property Y, there is no important difference between X and Y. But this is just false. E.g., the number 1 and 100. In more realistic case, look at species (whales and hippos).

  38. Fair Persuasion says:

    Margot Minardi has studied the development of race in the 1700s. Race organizes people into categories and explains differences. Assumptions: Physical differences matters. These differences in our bodies cannot change because they are inherited. Each group has a distinct level of intellect and morality, which is naturally and un-changeably ranked. Have we philosophically evolved in our views?

  39. BillyJoe7 says:

    Based on what’s been written above, I’m now starting to question whether children and adults actually exist! I mean there’s no sharp dividing line.

  40. EmilKarlsson says:

    Edamame, I think that your discussion conflate at least three different issues:

    (1) are there genetic differences between people?
    (2) is “discrete genetic races” (or traditional racial categories) a valid description of global human genetic variation?
    (3) can genetic differences between people be important?

    Clearly, (1) and (3) can be true, and is true, without (2) being true. In fact, mainstream science thinks (1) and (3) is true, but that (2) is outright false.

    Here is the quote from the Serre and Pääbo paper I referenced in the blog post to flesh out this distinction a bit more:

    “In the light of these results, and in agreement with extensive studies of classical genetic markers (Cavalli-Sforza et al. 1994), it seems that gradual variation and isolation by distance rather than major genetic discontinuities is typical of global human genetic diversity. Obviously, this does not imply that genetic discontinuities do not exist on a more local scale, for example, between people from different linguistic groups (e.g., Barbujani and Sokal 1990; Sokal et al. 1990). It also does not mean that no differences whatsoever exist between continental groups. In fact, what Rosenberg et al. (2002) have shown is that given enough markers and the extraordinary power of Structure, the tiny amounts of genetic differences that exist between continents can also be discerned. However, this should not obscure the fact that on a worldwide scale, clines are a better representation of the human diversity than clades, and that continents do not represent more substantial discontinuities in such clines than many other geographical and cultural barriers.”

    Notice how Serre and Pääbo does not deny that there are genetic differences between people and that they do not deny that there are genetic differences between continental groups.

    What they do claim, with reference to empirical data, is that ‘genetic races’ is a very inaccurate description of human genetic variation.

    When we understand…

    (i) the very low genetic diversity among humans
    (ii) that differences between continental groups are tiny compared with within continents or between individuals even when looking at ~650k SNPs
    (iii) the misunderstanding, mischaracterization and even abuses of genetic clustering algorithms by race realists
    (iv) that a proper analysis reveals that human genetic variation is largely clinal

    …the notion that there are human genetic races is not supported, but rather contradicted, by empirical evidence.

    It also seems very peculiar to focus on the small fraction of human genetic variation (~5-15%) attributable to between continental groups of an already tiny total genetic variation between humans (0.1%).

    It is almost as if the race realist position is “basically pseudoscience in the service of a political agenda.”

    I have written extensively (~20 posts) about the scientific reasons why race realism is pseudoscience on Debunking Denialism. A good entry to that is the post “Mailbag: Modern High-Throughput Genomics Versus Race Realism”, its references and the subsequent comments.

  41. EmilKarlsson says:

    BBBlueon,

    The “human genome mapping studies that have revealed genetically distinct groups which could be considered races” you speak of are precisely those studies that use genetic clustering methods on low-density data.

    When you, as a layperson, see those clusters, you think “this must be biological races!”. When someone who understands the methods and sampling strategies look at them, they think “method intrinsically inflate the importance of differences while downplaying similarities” and “the analysis is obviously conflated by geographical distance” and perhaps “sigh, now this study will be abused by race realists as well”.

    The conclusion that human genetic diversity is largely clinal is based on empirical data: DNA samples taken from humans with a high-density sampling strategy.

    There are genetic differences among humans, but the global pattern of human genetic variation is not genetic races, but mostly clinal. The exceptions being e. g. oceanic islands (since people typically do not live in the ocean itself).

  42. EmilKarlsson says:

    BillyJoe7,

    The scientific arguments I made do not make the continuum fallacy. Neither I nor my scientific references deny that there are genetic differences between humans and populations or that differences can sometimes be useful for a variety of purposes, such as medical or understanding (parts of) ancestral migration.

    Rather, the claim is that since global human genetic variation is accurately described as mostly clinal, genetic races is not an accurate description of reality.

    Retorting that “well there are genetic differences” is largely an irrelevant objection, since it (1) doesn’t actually address the claim or the evidence of clines, (2) and the statement that “human genetic variation is largely clinal” does not deny existence genetic differences. It is just that those genetic differences do not look the way you think they do.

    The larger claim beyond this is that race realism itself is based on pseudoscience and uses typical denialist tactics, such as abusing scientific concepts and research results. But we can cross that bridge if or when we get to it.

  43. SteveA says:

    BillyJoe

    MM embodies a sub-species of Troll that I call a ‘Pigeon’: he flies in, coos and struts around for a bit to absolutely no effect, does a poop (i.e. leaves a link to his ridiculous book), then flies off again.

  44. BillyJoe7 says:

    EK,

    “The scientific arguments I made do not make the continuum fallacy”

    “Rather, the claim is that since global human genetic variation is accurately described as mostly clinal, genetic races is not an accurate description of reality”

    I’m possibly misunderstanding something here, but isn’t the first quote above contradicted by the second quote.

    A cline is defined as “a gradual change in an inherited characteristic across the geographic range of a species…”
    (http://www.dictionary.com/browse/clinal)

    If you agree with that definition, then it seems to me you are denying the existence of race on the basis that the genetic change is gradual across the geographic range of our species.
    That does sound like the “Continuum Fallacy”.
    As does the following:

    “discrete genetic races”

    What about race as a fuzzy entity?

  45. BillyJoe7 says:

    Steve,

    “MM embodies a sub-species of Troll that I call a ‘Pigeon’: he flies in, coos and struts around for a bit to absolutely no effect, does a poop (i.e. leaves a link to his ridiculous book), then flies off again”

    😀

    I used to call these guys “drive-by shooters”, but I see I’ve been giving them too much credit (unless you imagine them firing blanks instead of bullets).

  46. Bill Openthalt says:

    BillyJoe7 —

    Blanks? More like paint-ball pellets, leaving blotches of drivel all over the blog comments.

  47. Bill Openthalt says:

    EmilKarlsson —

    Rather, the claim is that since global human genetic variation is accurately described as mostly clinal, genetic races is not an accurate description of reality.

    Where would the existence of “human genetic races” justify a different approach?

    Where would a well-published scientific consensus that “human genetic races” do not exist change the attitude non-scientists have towards people with a recognisably different appearance from the group they identify with?

    Which reality? That of the population scientists (where you’re no doubt correct), or that of the population?

    The larger claim beyond this is that race realism itself is based on pseudoscience and uses typical denialist tactics, such as abusing scientific concepts and research results. But we can cross that bridge if or when we get to it.

    Is there such as thing as “race realism”? Reality seems to be that humans do see differences where science shows there are none. There might be more genetic differences between two Luxembourgers from my village than between either of them and a Syrian refugee, but no-one, including you, would mistake them for each other (and in actual fact, most of my fellow villagers would tell you Luxembourger #1 isn’t really from the village because his grandparents came from somewhere in the south of the country 🙂 ).

    Humans prefer those who, for whatever reason, they believe to be closest to them. If this is questioned, they will come up with an “acceptable” reason, where “acceptable” is socially determined. Today it would be security, yesterday it was racial superiority, or the right of the conqueror, or whatever.

    When refugees are present, the distinction between locals and a (fairly) recent acquisition from the south of the country is blurred. Absent the refugees, it re-acquires its significance. Our brain seem to have the ability to manage about 150 inter-human connections, which might be why it has to apply hard-and-fast rules about which connections are most useful. The same mechanism is useful for risk minimisation — because we don’t care much about those not close to us, they are a bigger risk to us (and we to them).

    Sadly, as with GMOs, nuclear energy, the intelligent universe, etc., emotions will trump reason any day.

  48. All,

    Thanks for the excellent feedback on this complex issue. This is one where my own thinking is a bit in flux, and I do not take a hard position (as I tried to make clear). I just have problems with the way some people defend the hard position that race does not exist.

    Gorilla – We mostly agree, but I think you are missing my point. You are saying that race as a social construct and a genetic reality are mutually exclusive. I disagree. You further claim that no ones denies the genetic, but the specific essay I was responding to does just that.

    To clarify – I am not disagreeing with, or even addressing, the social and historical arguments that race is socially constructed. I am disagreeing with the scientific argument that there is no genetic or scientific basis to the idea of race in humans. Some people use the scientific arguments to bolster the social/historical/ethical arguments, and I think they make subjective judgments or even distortion of facts to do so.

    Granted, many people (perhaps most) have a distorted view of race in the other direction (as I explicitly stated). I also said the question has no objective answer. It all depends on when a pattern of genetic variation is sufficient to justify identifying subpopulations. You can make a reasonable argument either way, and people will generally make a subjective decision based on their social and political views and agenda.

    I am mostly just trying to identify the scientific and logical components at play here.

    Emil – I here what you are saying, and your point that human variation is mostly clinal is valid, but that does not change the points I just made above. It is a continuum fallacy to an extent also. Let’s use the terrain example. Natural environments tend to vary in a clinal fashion also, with prairie blending into forest, blending into grasslands then into desert, etc. This does not mean that forests and deserts don’t exist. That is the continuum fallacy.

    Most species follow a clinal distribution, because they are adapting to local terrain which is clinal (plus genetic drift). When environments are discontinuous then species become discontinuous, which you also acknowledged about island populations.

    Also, if you look at a mitochondrial haplogroup map (https://en.wikipedia.org/wiki/Human_mitochondrial_DNA_haplogroup) I think you see less of a clinal distribution and more of a distribution that represents the migration of specific populations. Correct me if you think I am wrong on this.

    What does all of this mean? I agree it means little (again, as I explicitly stated). We should essentially ignore human genetic variation when it comes to social and political issues. They are of some value to medicine, but of course they can be misused and oversimplified (as all valid concepts in medicine are). This does not mean they are not real and useful. Should we replace this with individual genetic evaluation? Sure, that will happen when it comes. It’s not always practical now, so we use ancestry as a rough estimate of likely genetic risk.(such proxy markers in medicine are common) Asians, for example, may have a gene variant that makes them susceptible to a severe side effect of a specific medication. If you ignore Asian heritage in prescribing this medication you are literally guilty of malpractice.

  49. EmilKarlsson says:

    BillyJoe7,

    It isn’t a continuum fallacy since mainstream science does not deny that people and groups differ genetically. This isn’t an issue of whether people or groups differ genetically or not. Had I claimed that there were no genetic differences between populations because the genetic variation was clinal, you would have a good point. But I am not.

    Rather, it is about what description is most accurate for human genetic variation on the global scale. This turns out to be mostly clinal in nature, not discrete or fuzzy races. Places were human genetic variation is not that clinal is not between continents, but in areas with isolation due to very large geographical distances, such as oceanic islands. So the fuzzy alternative doesn’t quite work either.

    In particular, Serre and Pääbo concluded that “clines are a better representation of the human diversity than clades, and that continents do not represent more substantial discontinuities in such clines than many other geographical and cultural barriers.”

    So an emphasis on continental groups as the basis for race just doesn’t make sense from a genetic perspective. Although it _seems intuitive_ that different continental populations are very different from each other (and I also fall prey to this intuition sometimes), this simply is not the case on the genetic level.

    The reason it seems so intuitive is probably because most humans have not experienced the full genetic variation of a continental group, are not fully aware of the low genetic differences between individuals in the first place and our human tendency to detect patterns and put things in boxes.

    But if we have learned one thing from science, it is that intuitive things are not always accurate.

  50. EmilKarlsson says:

    Dr. Novella,

    Your analogy with terrain doesn’t quite work because (1) there is a large difference between, say, desert and lush rainforests, while there are very small differences between humans and human populations, (2) there is probably more variation between lush rainforests and deserts, than within deserts, (3) there is no clear phenotype/genotype equivalent in terrain like there is in humans (phenotypes can look very different, but the genotype does not have to be equally different).

    Again, neither I nor mainstream science deny that populations differ genetically (i.e. no one is denying that forests and deserts do not exist). The main question is this: when we look at global human genetic variation, is it mostly clinal or does it show evidence of genetic races? The answer to this is that it is clinal and offers little in the way evidence for genetic races and continents are not more important than many other cultural/geographical barriers.

    We agree that there are genetic differences between individuals and populations. The question is rather about how is global genetic variation most accurately described.

    Mitochondrial DNA analysis only tells you about a single one of your ancestors, since it almost always moves from mothers to offspring. It can certainly tell you about migration patterns of your great-great-great-great…grandmother’s lineage (and only that lineage), but it is not a useful method for this question. Just like genetic clustering algorithms, it is useful for some kinds of analyses, but can be deeply misleading for others.

    In particular, mtDNA analyses only looks at a small fraction of your genome, about 0.0005% (mitochondrial genome is about 17k base pairs and the human genome is about 3.3 billion base pairs) and ignores sexual recombination. For comparisons, the studies I have referenced looks at ~650 000 SNPs and ~600 microsatellites with high-density sampling. These studies are by no means perfect, but I think we can be reasonably confident that they have the upper hand when it comes to this particular research question.

  51. Bill Openthalt says:

    EmilKarlsson —

    The reason it seems so intuitive is probably because most humans have not experienced the full genetic variation of a continental group, are not fully aware of the low genetic differences between individuals in the first place and our human tendency to detect patterns and put things in boxes.

    It’s intuitive because it works (or used to work a couple of hundred of generations ago). How can humans “experience” the full genetic variation of a continental group? Humans have senses and a brain that look for differences and similarities between individuals, so any “experience” would have to result from reading scholarly articles or doing research. Like other social animals, humans need to be able to assess their relationship with other humans and the rest of the world quickly — hence our ability to recognise faces, discern probable closeness based on physical characteristics, behaviour, etc. Humans started to put the world in conceptual boxes because of survival — edible/inedible, friend/foe, male/female, safe/dangerous etc. are vital distinctions all moving critters need to make.

    The differences we perceive and act upon aren’t scientifically sound, but behaviourally they are/used to be very sound. Obviously, humans can, and do, use their rational mind to modulate their basic behaviour to suit the social situation, or we wouldn’t be able to build societies more complex than those of genetically identical insects. Obviously, the human tendency to identify with groups isn’t limited to perceived “race” distinctions — football hooligans are good examples of how people modulate their behaviour based on the group they identify with at a specific moment.

  52. Emil – I agree the terrain analogy is not perfect in every respect, but it is valid in the one respect I used – smooth transitions does not mean categories are not “real.” It does put them in perspective, however.

    You still seem to be dancing around my main point – the criteria that we choose to use in order to answer the question, are races “real”, are arbitrary. They have to be. Different criteria give you different answers, and it is interesting to think about which criteria are more definitive.

    I am still looking into the claim that human genetic variation is entirely clinal, without any relative discontinuities. That has not been my reading of genetic variation. First, of course all discontinuities will be relative and fuzzy. Humans migrate and interbreed. We are one interbreeding species – there is no question about that.

    Here is a review from Nature published in 2004 (http://www.nature.com/ng/journal/v36/n11s/full/ng1435.html)

    These paragraphs sums up what I thought:

    “When 100 Alu insertion polymorphisms and 60 short tandem repeat (STR) polymorphisms were used, all Europeans, East Asians and Africans were correctly placed according to their respective continents of origin.”

    So – you can slot individuals to their correct continent of origin (these three continents) by looking at their genes. But:

    “The picture that begins to emerge from this and other analyses of human genetic variation is that variation tends to be geographically structured, such that most individuals from the same geographic region will be more similar to one another than to individuals from a distant region. Because of a history of extensive migration and gene flow, however, human genetic variation tends to be distributed in a continuous fashion and seldom has marked geographic discontinuities19, 42. Thus, populations are never ‘pure’ in a genetic sense, and definite boundaries between individuals or populations (e.g., ‘races’) will be necessarily somewhat inaccurate and arbitrary.”

    Middle east and Indian subcontinent are not as easily sorted because they represent much more genetic migration and mixing.

    So yes, 86-90% of genetic diversity is within continent, not between. This largely reflects the much longer genetic history of Africans, but also genetic mixing over historical time. At the same time, I can look at your genes and tell you what continent your ancestors came from. Which of these facts is more important in answering the question of whether or not “race” exists? That is a judgement call. It’s arbitrary.

    I also think your objections to the data from mitochondrial DNA reflects your choice in the former criterion (amount of difference), rather than than the latter (identifiable groups).

    So, to summarize, while I agree that human genetic distribution is somewhat clinal, I am not convinced it is totally so. You seem to be dismissing individual sorting studies, but I don’t see why.

    Second, there is no objective way to relatively value the fact of a difference with the amount of difference in terms of which criterion is definitive.

  53. EmilKarlsson says:

    “Emil – I agree the terrain analogy is not perfect in every respect, but it is valid in the one respect I used – smooth transitions does not mean categories are not “real.” It does put them in perspective, however. ”

    My claim is not that there are categories, but smooth transitions between them. Rather, it is that categories are something that we largely, but not completely (because 5-15% of variation occurs between continents), impose on the mostly (but not completely) clinal data by inflating and overemphasizes genetic differences between continents but downplaying the vast majority of human genetic variation that lies elsewhere.

    “I am still looking into the claim that human genetic variation is entirely clinal, without any relative discontinuities. That has not been my reading of genetic variation.”

    I have never claimed it is entirely clinal. On the contrary, I have given examples of where it is not clinal at all, such as remote islands in the ocean and similar situations. What I have claimed is that it is largely clinal and that continents are not privileged compared with many other cultural/geographical boundaries.

    “So – you can slot individuals to their correct continent of origin (these three continents) by looking at their genes. ”

    Indeed, between 5 and 15% of human genetic variation is between continents and you can tease this out. Never claimed anything else. But, as the figure themselves tell you, it is just a tiny percentage of the total human genetic variation, which itself is very small (0.1% SNP difference between individuals). It is very difficult to look at this genetic data and conclude that there are human genetic races (fuzzy or discrete).

    “At the same time, I can look at your genes and tell you what continent your ancestors came from”

    You can look at my mtDNA and tell me what continent _one_ of my ancestors come from. What about the millions of others?

    When you look at nuclear genome studies, you can indeed analyze those 5-15% of the total human genetic variation and make rough ideas about geographic ancestry if your computational methods are powerful enough. That is also the case mentioned in the Serre and Pääbo quote I gave about. The fact that there are methods powerful to detect the small between-continent genetic variation, however, does not disprove the largely clinal shape of global human genetic variation.

    “You seem to be dismissing individual sorting studies, but I don’t see why. ”

    I am not dismissing them at all! For instance, mtDNA studies can be used to find the ancestry of one of your lineages (great-great-…-great grandmother). The reason this is possible is because it ignores sexual recombination. When trying to find the ancestry of a single lineage, sexual recombination is an irrelevant nuisance factor and is excluded by looking at mtDNA (or Y chromosomes). These studies are great and useful in their contexts.

    But when you want to look at global human genetic variation, sexual recombination is not an irrelevant nuisance factor. On the contrary, it is of considerable importance because, well, humans have sex. Thus, mtDNA studies are not appropriate to address this particular question.

    So the take-home message is that different genetic analyses has different methodological assumptions, sampling strategies and are often specialized for specific scientific questions. Using one method to investigate different question for which it is unsuitable for can give misleading results, such as the examples genetic clustering of low-density data that mistake geographical distance for genetic races.

  54. OK – it seems like we are agreeing on all the facts. The only difference is in how you interpret the fact that genetic populations exist, but the differences are just small.

    Also, I think the concept of “continent” itself is fuzzy. Why are Asia and Europe different continents?

    I think a better way to summarize the genetic landscape of humans is that it reflects geographical distance and relative barriers to movement and interbreeding – in other words, relative isolation. So yes, the island populations are more genetically distinct. I would also include, however, subsaharan Africa. Northern Africa tends to blend into the Mideast, which blends into Europe and Asia. Native Americans were also relatively isolated, but not for much time so their differences are less. Australians more so.

    The bottom line is that the genetic map nicely mirrors an evolutionary map of migration and isolation vs intermixing.

    I maintain my ultimate position, however, that the question of whether or not race exists is not objectively answerable. I agree with the authors of the paper I linked to who said that the position that race is “biologically meaningless” is an overstatement and oversimplification. But, at the same time, clusterings of genetic diversity are fuzzy, incomplete, and complicated.

  55. leonet says:

    @ EmilKarlsson

    I agree wholeheartedly that it is extremely important to be aware of the methodological assumptions in computational studies of large-scale genomic data.

    @ Steven Novella

    When you say:

    “Second, there is no objective way to relatively value the fact of a difference with the amount of difference in terms of which criterion is definitive.”

    This don’t think this is entirely true in this context. I believe that most of the studies that show a difference between subgroups of humans show that the differences in questions are extremely small and are primarily in non-coding regions of the genome with relatively high mutation rates (microsatellites etc. – which a lot of these studies use). If these statements hold true, I think it’s scientifically legitimate to argue that the genetic “difference” in question cannot be the cause of an apparent social difference that people are seeking to ground in biology.

  56. BTW – there is another analogy that may be helpful, the spectrum of visible light. There is no frequency at which yellow ends and green begins. There is no privileged clustering of frequencies. There is a smooth continuum. It is also true that perception of color is partly socially constructed. (Some cultures don’t have two words for blue and green).

    But, I would not conclude from that that color is biologically or physically meaningless. I can still meaningfully refer to yellow, green, and blue.

  57. leonet says:

    And . . . I just wanted to add that my argument above would only necessarily apply to all differences; it would have to be studied on a case-by-case basis.

  58. leonet – but this discussion has never been about grounding social race in biology. I never disputed that race is socially constructed, and I discussed the fact that, even if you think races exist, they are not what most people think for various reasons.

    The fact that you can accurately predict the relative geographical location of someone’s ancestors by looking at their genes, because their genetic variations tend to cluster, means that the clusterings are real. They predict something. They are not biologically meaningless.

    At the same time I can conclude from the evidence that these different clusterings are small in absolute difference and of no social or political importance. They are insignificant in terms of what it means to be human.

    However, they are not biologically meaningless, because I think the evidence also clearly shows that they are medically useful. I disagree strongly with the position that they are meaningless because they are not simplistically determinitive, or because some people use them wrong.

    As I said, if you fail to consider a patients’ Asian heritage when prescribing certain medication you are literally guilty of malpractice.

  59. leonet says:

    Yes, it’s true and unsurprising that differences in functional loci have consequences that are biologically meaningful and that we must take into account. But here, we are still only using ancestry as a marker for a difference in a particular biological pathway because we don’t know any better (yet). As genomic knowledge grows, we hope that those pathways become known so we can improve care for someone who, for example, is in the middle of one of those clines.

    However, what “racial realists” often argue is that some combination of genomic markers that can reliably define a race. What I’m arguing is that particular claims of this nature can be rightly described as pseudoscience because the methodologies and markers that “clustering” studies rely upon are very unlikely to explain the phenotypic differences that we associate with our socially-defined races.

  60. Bill Openthalt says:

    Mutatis mutandis, this discussion also applies to gender/sex. The same social prejudices against being perceived as “racist” apply to being perceived as “sexist”, so people will look for support for the idea that neither “race” nor “gender” are “real”.

    Men are, on average, taller than women, which doesn’t mean a man is always taller than a woman. This might sound a truism, and it is, but in these socially and morally sensitive areas it is often forgotten, witness this excerpt from Theresa May’s first allocution as PM:

    That means fighting against the burning injustice that if you are born poor, you will die on average nine years earlier than others. If you’re black, you’re treated more harshly by the criminal justice system than if you are white. If you’re a white, working-class boy, you’re less likely than anyone else in Britain to go to university. If you’re at a state school, you’re less likely to reach the top professions than if you were educated privately.

    If you are a woman, you will earn less than a man. If you suffer from mental health problems, there’s not enough help to hand. If you’re young, you’ll find it harder than ever before to own your own home.

    “. Almost every sentence is statistical nonsense, but oh boy, does it sound lofty.

    However, trying to fix social ills through science is like trying to teach a horse to sing. Scientifically proving there is no genetic basis for human races has just as much chance of changing Anders Breivik’s mind as proving there’s almost no genetic difference between dog races would convince my mother-in-law to swap her poodle for a great dane.

  61. BillyJoe7 says:

    EK,

    I’m still not quite understanding this.
    It seems now that your main point is not that the differences are mainly clinal (gradual rather than discrete or discontinuous) but that the differences across the human genetic landscape is not great – in fact, that it is “very small”:

    “total human genetic variation…is very small…0.1% SNP difference between individuals…It is very difficult to look at this genetic data and conclude that there are human genetic races”

    However, consider the following:

    The genetic variation between humans is 0.1%
    The genetic variation between humans and chimpanzees is 1%
    The genetic variation between your liver cells and your brain cells is 0.0%

    The major difference – at least in the last two examples – is the silencing and the activation of genes by regulatory genes via regulatory proteins. In view of this, do you think that perhaps we need to look at epigenetics for the difference. Maybe looking at just the genotype is insufficient.

    Or maybe our intuitions ARE wrong as you say.
    (I’d actually say that our intuitions, as far as science is concerned, are nearly always wrong)

  62. wwaynemarlow says:

    I favor Loring Brace’s notion of clines because attempting to categorize groups by skin color, hair texture, and facial features requires ignoring unseen differences that cross socially constructed racial boundaries. While melanin follows a predictable pattern north and south, other clines spread out from specific points. As one example, Belgians and Ugandans have very different skin color, but when it comes to the distribution of the ABO blood group, they are closer to each other than either are to the Chinese.

  63. I agree that using superficial features to define race is socially constructed. These traits are not privileged in any way as markers of subpopulations. They are just evident to casual observation.

    This does not impact, however, the fact that when we look more thoroughly at genetic markers we can still reliably sort people according to their genetic heritage, which is unsurprising.

    The fact is – there are genetically defined (relative) subpopulations of humans. The only question is – are the differences enough to warrant any kind of designation? This is inherently a subjective question without a definitive objective answer. This also depends on context.

    I only maintain that in the medical context it is useful to consider ancestry. This provides useful statistical information about certain genetic risks. This is uncontroversial and the medical profession essentially universally accepts it and ignores the entire discussion about race. It is probably true that this will become entirely obsolete, however, as we make advances in testing for and understanding individual genetics. Future probable obsolescence, however, is irrelevant to the argument of current utility.

    In the context of social justice it is tempting to emphasize that the differences are small and relative, but I would emphasize that this doesn’t matter. All humans deserve equal dignity, respect, and rights regardless of the patterns of genetic diversity.

  64. BillyJoe7 says:

    Relevant anecdote:

    I have an acqaintance (he lives in the next street and we met on a morning run many years ago) who is Chinese. He found a small lump in his neck and attended his GP. His GP used the fact that he was Asian to suspect a nasopharyngeal cancer spreading to a lymph node in his neck. He did in fact have a very small tumour. According to his surgeon, the tumour was so small that he could easily have missed it if he wasn’t meticulous in excluding this possibility in a man of his background. Because it was picked up and treated in its early stages, he is still alive ten years later.

    https://en.wikipedia.org/wiki/Nasopharynx_cancer

    NPC is uncommon…representing less than 1 case per 100,000 in most populations but is extremely common in southern regions of China, particularly in Guangdong, accounting for 18% of all cancers in China. It is sometimes referred to as Cantonese cancer because it occurs in about 25 cases per 100,000 people in this region, 25 times higher than the rest of the world.

  65. Slipper says:

    Hey Stephen,
    I listened to your podcast The Skeptics’ Guide to the Universe #577, July 30th. You dealt with two topics: the word “cool” and the question of whether “race” exists or not.
    The simplest way I can express my view is to say that perhaps you could have dealt with the word “race” in the way you dealt with the word “cool.” In other words, by looking at how the word has been used historically.
    Words and meanings are always changing, as they are cultural phenomena, and both denotation and connotation are in a slow state of flux. The idea of looking for the “scientific” definition of the word “cool” is, of course, absurd. With “race” it is not quite as absurd, but almost.
    I enjoyed your discussion of individual differences being as great as category differences and your discussion of how physicians need to consider all of a person’s characteristics when diagnosing a patient.
    However, as the podcast indicates, all the speakers seemed to conflate race with a person’s ethnic origin, with geographic origin, and with national origin. Is Italian a race? Is Jewish a race? Is Asian a race? Is African? The woman who spoke about Caucasiod, Negroid, and Mongoloid came closest to using the word in the way we were taught in school, but as you quickly pointed out, that definition of the term race is not one that can be supported anymore.
    Just thought I’d make those comments.
    Generally, I enjoyed the podcast.
    Very thought stimulating.
    Cheers

  66. Johnny says:

    “What makes me uncomfortable is subjugating the science to a political goal, no matter how noble that goal is. There are two big problems with this. First, it’s important to get the science right. Distorting scientific thinking is pernicious. Second, if we make a valid political position dependent upon a scientific position, then the political position is vulnerable to new evidence, and it almost forces you to distort the science.

    Regardless of what we discover about the genetic variation of humans, it is clear that all humans should be treated as equal as an ethical principle. This principle is not dependent upon the scientific fact that race does not exist.”

    I just want to say that I completely agree with this, and this point tends to be overlooked in the debate about this. I’m glad you state it.

    I don’t know how to formulate what ethical principles that should guide society (help from anyone with that would be appreciated), but even though it should be informed by science, it can’t be determined by it.

  67. bobbruer says:

    Already a long-in-the-tooth Canadian health-care professional, 2004-06 I took a couple years CE leave, moving to St. Louis to study psychiatric epidemiology methodology at WUSTL. I recall some surprise upon arrival that Race was such an every-day variable in the US data-sets we commonly considered. This type of classification seemed not only odd, but from my tradition, somewhat distasteful.

    In Canada to be honest, we do stuff sort of like that, commonly delineating people by language (English/French/Other), First-Nation status and immigrant/Canada-born. But not Race. Much of our social classifying I feel informs treatment refinements (e.g., translator resource requirements or determining payment options within our varied health funding protocols).

    I suspect what causes me upset most in the US around Race, was the frequent use of ‘African-American’ as a poorly understood proxy for things better explained by low SES or high criminologic risk.

    And let’s be honest, I think it is time to put away the Sickle cell example. Your NIH has called for universal screening of SCT… inferring that “African-American” has little practical utility in this area. SCD prognosis is so serious and genetic risk present at meaningful levels in all races… just do it.

    Isn’t it more meaningful for CDC-type organizations to put race aside more often, and try measure/analyze things like max level of education, employment history, social engagement (e.g., volunteering, sports), family stability, diet, exercise?

    Admittedly, as a researcher, I’ll data-mine as much as you give me including race, provided I have theory to establish prior plausibility estimates. Give me a million records with race and SCT test results and I’ll play quietly in the corner for hours. But in the Canadian mosaic where I work, Race seems less meaningful than what you take in your Tim Horton’s coffee and how many times have you skated in the last year.

    Huge fan… just that I see differently on this issue and I think many of my fellow Canadian see it my way also.

  68. LaPalida says:

    Thank you for tackling this issue Steven.

    Finally someone is talking about this in rational terms. A lot of people balk at talking about this topic because it’s so politically charged. It’s extremely difficult to have a discussion about the facts when people get very indignant and accuse you of racism straight out. If you mention anything about race in a positive or even neutral light you must be a racist Nazi.

    A while back I used to believe that race was fiction and that it was simply invented to divide and conquer people based on the inherent human tribalism and fear of the outsider. Then I started reading a bit more and actually giving my political stance some thought rather than just repeating the mantra and trying to appease the group I was part of in order to fit in. Learning about how evolution works inevitably lead me to make some inevitable conclusions.

    I just want to point to a blog post which goes into much more detail on how this works:

    https://whyevolutionistrue.wordpress.com/2012/02/28/are-there-human-races/

    What I found fascinating here is that they fed the genetic data that they had into ‘a clustering algorithm that sorts individuals into a pre-specified number of groups’ and behold ‘the genetic clusters “corresponded largely to major geographic regions.”’

  69. Erika Butler says:

    >Africans develop sickle cell anemia; Europeans and Asians don’t.

    I am glad to see this error was corrected further up the comments page. This is why I think scientists coming out, using their own definitions, and flatly saying, “Science proves race exists” is so fraught. Yes, there is important genetic diversity based on ancestry and geographical origin, and at times it’s often helpful to divide humans into “races” for the purpose of testing.

    The problem with saying this in an American context, however, is that people here are going to assume such proof of the existence of “race” means that science has proven Americans’ messed-up view of race, even though scientists have a different definition in mind. The very fact that someone with the stature of Steven Novella (whom I personally respect a lot), who is an American, makes such a blatant error shows how careful we must be when using the word “race” in a scientific context.

    Personally, I think “race” is a loaded word, and another word like “clines” might be more helpful. Otherwise, white/black/etc. nationalists are going to engage in equivocation that pivots between the 2 definitions of “race” I discussed above, to further entrench the messed-up views of race most Americans have.

  70. eepobee says:

    “Relevant anecdote:
    I have an acqaintance (he lives in the next street and we met on a morning run many years ago) who is Chinese. He found a small lump in his neck and attended his GP. His GP used the fact that he was Asian to suspect a nasopharyngeal cancer spreading to a lymph node in his neck. He did in fact have a very small tumour. According to his surgeon, the tumour was so small that he could easily have missed it if he wasn’t meticulous in excluding this possibility in a man of his background. Because it was picked up and treated in its early stages, he is still alive ten years later.”

    Anecdotes can be helpful, but they also, unfortunately, can be misleading. Using racial categories to aid in clinical decision making can have dire consequences:

    https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1989738/

Leave a Reply