This is not a question about genetics or biology. I am interested in understanding the meaning of race from a sociological perspective. Please focus your answer on the case I give for ancestry and race being the same. I do bring up genetic variation below, but only to dispel the argument around it so that answers don't focus on it. In fact, the reason I posted this question is because I can't find an article on the subject that doesn't focus on genetics. Here is the question:

It seems to me they are the same, but it has come to my attention that the official position in academia is that they are distinct. In fact a quick internet search of the question brings up articles (citing research) on how they are distinct.

These articles argue that there is more genetic variability within each race than between them (or more precisely, within/between each geograpical area that would correspond to each race e.g. Africa, Europe, Asia etc.) and therefore race can't be based on heritage. But this is a total non sequitur. It may mean race is not a very useful classification, but it doesn't prove that race can't be based on heritage (nor does it prove that heritage can't be determined by genetics, as it is well known that certain genetic markers can predict heritage).

My reason for thinking they are equivalent, and a more precise reformulation of my question, is drawn from common intuition about two individual races, white and black:

Isn't having African heritage necessary and sufficient to be considered black? And isn't the same true for Western European heritage and being white? If inclusion in each race is determined 100% by biological heritage, how can the two be distinct?

  • Comments are not for extended discussion; this conversation has been moved to chat.
    – Geoffrey Thomas
    Commented Aug 17, 2021 at 7:27
  • 1
    @Thomas some of those comments were the answer to the question... Particularly Colinfold's
    – Jonah
    Commented Aug 17, 2021 at 13:03
  • I had to follow the site rules and move +20 comments to Chat. This was not a personal decision I made. I have to follow the rules like everyone else. The comments are still accessible as indicated above.
    – Geoffrey Thomas
    Commented Aug 17, 2021 at 14:31
  • I don't see how this is off topic. Is there a better stackexchange for questions about sociology? I made it clear the question wasn't about genetics or biology, and I can't control what the answerers choose to focus on. If you ask me, Ted made this about genetics/biology then requested to close the question because of it, despite me telling him his answer is off topic many times!
    – Jonah
    Commented Aug 19, 2021 at 13:53
  • @Jonah: The meaning of the term 'race' has not been fixed. See washingtonpost.com/news/volokh-conspiracy/wp/2017/03/22/… It was considered a major issue to be of a different heritage like Irish or Italian in the English speaking world, where now we barely distinguish. Religious strictures like the Protestant-Catholic divide kept groups separate, in a way they no longer do. Religious intermarriage barely raises eyebrows, but bot long ago could see excommunication. It's disingenuous to see it as a fixed term
    – CriglCragl
    Commented Aug 19, 2021 at 23:11

3 Answers 3


As a former Academic, I can tell you that one should be very careful in relying upon the definitional incoherence of Social Science Professors with regard to the topic of Race. The phrase, "Race is a social construct", is loaded with inaccuracies and defies scientific reality and objectivity-(which is just fine with many Social Science Academics, considering the fact that many of them tend to have Post-Modern sympathies). Sadly, throughout most of the Contemporary era, many Social Science Academics and Theorists, have successfully converted and transformed Biological and Anthropological Science, into a highly subjective discourse whereby the topic of race, is positioned front and Center. Social Science Academics took and still take great liberties with what was and is still, a term, that is primarily related to Genetics, as well as Anthropology and deeply politicized it, in order to promote a newer (and I believe, a vacuous) intellectual narrative.

While my background is in History, I have taught introductory courses in Anthropology and focused on Ethnography. My courses NEVER examined "race as a social construct", but instead, closely examined the relationship between race and ethnicity-(namely from taxonomic and historical perspectives). For example, when I discussed, "White peoples", I asked the class to define the word, "White"-(or "Caucasian") in ethno-geographical, as well as ethno-historical terms...few were able to do so. I also asked my class if the word, "Hispanic" or "Latino", was racially or ethnically etymological. And with regard to the word, "Asian", there were numerous misunderstandings of its ethnic and racial connotations among many of my students. In the case of African peoples, my goal was to teach students, the diverse ethno-racial composition of the African continent, stretching from Egypt to Mozambique. Yet, throughout my lectures and discussions over the years, the phrase, "race as a social construct", was never uttered.

The word, "White" or "Caucasian", according to many Anthropological texts, Dictionaries, Encyclopedic texts and the U.S. Census, is typically defined as, "any person(s) who are of European, North African and Middle Eastern descent". According to this mainstream definition, Arabs, Berbers and Persians/Iranians, could be identified...as "White" or "Caucasian". The word, "Asian", is typically defined-(usually by the same above mentioned sources), as "any person(s) who are of East Asian and South Asian descent". This definition means that any person who has ancestry from the Indian subcontinent, somehow, belongs to the same ethno-racial category as someone from China or Japan. The word, "Hispanic" or "Latino", is usually defined as, "any person(s) from a Latin American country who is either White or Non-White", which definitionally speaking, has its own limits. The word, "African" or "Black", is usually defined as, "any person(s) who has ancestry in the African continent"-(however, this definition would exclude North African Berbers, Arabs and Egyptians).

While these ethno-geographical definitions have some legitimacy and validity, they tend to fall short in certain areas. For example, take the word, "Asian"; technically, many ethnic Arabs have ancestry in WEST ASIAN lands, stretching from Lebanon to Oman. Iran, as well as neighboring countries, such as Uzbekistan and Tajikistan are also...Asian lands. If this is geographically true, then why use the word, "Asian" to only include peoples from East Asia, Southeast Asia and the Indian subcontinent?-(the mainstream's definition, is, at best, incomplete). In the case of the word, "White" or "Caucasian", is a person from Egypt, the same race as someone from Sweden? Again, what exactly does the word White or Caucasian really mean? In the case of the word, Latino or Hispanic, there are the words, "White" and "Non-White" Hispanic, which are probably based on historical demographics whereby so-called, "White" Hispanics were originally from Spain and Spanish speaking indigenous "First Peoples/Nations" of the Western Hemisphere are largely, "Non-White". What about the centuries old Spanish speaking peoples of Argentina? Are they identified as, "White" Spaniards? How about the Mestizos/as-(who comprise, a sizable percentage of Mexico's population), what would be their classification? In the case of "Africans" and "Blacks", how would one identify East Africans?-(who are actually of mixed ethno-racial descent, due to centuries of intermixing with neighboring Egyptians and Semitic peoples across the greater Middle East). While East Africans are certainly "Black" and "African", their genetics reveal that they have non-African ancestry as well. How then would they be accurately classified when using the mainstream's definition?

So, as you can see, the topic of race, is far more complex than Genetics; it encompasses historical, ethnographic and geographical components as well. However, the idea that Race, has been and is still, the Center of ALL Reality, due to a supposed systematic design that was and is still, enacted by Powerful Elites of a particular race for the SOLE purpose of disenfranchising and disempowering another race-(while partially true), is by no means, Universally true. The Post-Modern Theory of Race, erroneously disregards (and conveniently omits), the above mentioned components that accurately define the complexities of race. In doing so, the Post-Modern Racialists radically politicize a concept, which for millennia, has been largely-(though not entirely)..apolitical.

  • 3
    Look at the history of use of the term: "The term was first used to refer to speakers of a common language and then to denote national affiliations. By the 17th century the term began to refer to physical (phenotypical) traits. Modern science regards race as a social construct, an identity which is assigned based on rules made by society" en.wikipedia.org/wiki/Race_(human_categorization) How do you deal with people that have heritage of 4+ groups? What about merging of individuals from other races into a wider genepool? The idea of a 'pure race' is incoherent
    – CriglCragl
    Commented Aug 16, 2021 at 16:31
  • 1
    This is by far the best answer because it compares and contrasts race and heritage, and gives numerous examples of where race and heritage don't match up. It seems heritage is much more granular than race, and race is only a subjective crude heuristic for geographical genetic heritage. Race has a genetic component, but it is distinct from heritage.
    – Jonah
    Commented Aug 16, 2021 at 20:00
  • Many thanks for the comment; it is greatly appreciated.
    – Alex
    Commented Aug 16, 2021 at 20:18
  • 1
    Thank you for the comment; it is much appreciated.
    – Alex
    Commented Aug 16, 2021 at 20:31
  • 2
    This is by far the best answer because it tells OP what he wanted to read... Yet it just demonstrate that race is less about genetics than how people see and classify each other, hence a social construct.
    – armand
    Commented Aug 20, 2021 at 21:52

At heart, this issue involves a misunderstanding of statistics. Let's start by laying out a simplified thought experiment. Assume we do the following:

  1. Design a random number generator that produces a uniform distribution of integers between 1 and 100: i.e., that every integer in that range has the same likelihood of being generated
  2. Use that random generator to create sets of 30 randomly generated numbers, which we'll call 'units': i.e., we run the generator 30 times, group the results together as a unit, and repeat that process a large number of times (until we have as many units as we want, or our undergraduate research assistants start threatening to set themselves on fire in protest).
  3. Sort the units as follows:
    • every unit that contains the number 12 is type A
    • every unit that contains the number 43 is type B
    • every unit that contains the number 67 is type C
    • every unit that contains none of these numbers is type D

Obviously some units might end up in more than one category...

Now, here's the question: Do these four 'types' constitute statistically distinct groups? The answer, of course, is 'no'. They may be functionally different: we may want to color A red, B green, C blue, and D yellow, for instance. But each unit is drawn from the same statistical population by the same mechanical process, with typing occurring after the fact. I mean, the distribution of each type will be slightly different (small shifts in the mean and variance because one of the numbers is now fixed, not random). But that would be quite difficult to demonstrate analytically through sampling.

This thought experiment is a decent analog for racial genetic testing. In racial genetic testing we:

  • Predetermine that there are racial groups
  • Examine people whom we have assigned to racial group X to find a few genetic sequences (out of the many, many genes humans have) which they exclusively have in common
  • Assert that anyone who shares those genetic sequences can be assigned (to some extent) to racial group X

That's all well and good, as far as it goes. But nothing in this suggests that racial group X is drawn from a statistically different population than other racial groups. I mean, we know that a unit of type B contains the number 43, even though it's drawn from the same population as all the other types; we know that because we designed it that way. Likewise, we began with a preconceived notion of races and found genetic markers that correlate with them (akin to our numbers 12, 43, and 67). But the presence of those markers does not imply or demonstrate that the genetics of these racial groups are fundamentally different or distinct.

Biological ancestry is determinate: we each have a certain set of genes that necessarily came from our ancestors. But the fact is that we could pick arbitrary gene sequences and generate arbitrary, artificial racial groups, merely by asserting that the presence of this arbitrary sequence is a 'marker'. Unless one can show that the genetic markers for conventional racial distinctions are somehow less arbitrary than the genetic markers for such an artificial race, then we have to acknowledge that race isn't biological at all, but merely an imposed sociological distinction.

  • 1
    @Jonah There is no genetic marking possible that can be aligned with geography. Depending on which statistical distinctness you want to allow, there is either a single human race or literally thousands, and they do not add up well with geography. That is what Ted illustrated here: Sure, you can put individuals (numbers) into different groups, but these types are going to be arbitrary. The studies looked at much more markers than ancestry webshops do.
    – Philip Klöcking
    Commented Aug 17, 2021 at 16:23
  • 1
    @Philip Klöcking a model can be (and has been by many "ancestry webshops") created to infer the place of origin of a given (never before seen) DNA sample with high accuracy. Granted, the model will look at many markers, not just one, but if place of origin were an arbitrary category this wouldn't be possible.
    – Jonah
    Commented Aug 17, 2021 at 17:38
  • 1
    Hair color, blood type, alcoholism, etc. are all causally linked to DNA, location on earth is not! That's what makes it special. You could try to group people by something truely arbitrary and like their astrological sign, but you wouldn't be able to predict that from their DNA. At best, your toy model successfully refutes a straw man (nowhere in the question did I imply the misunderstanding of statistics that you claim), at worst it's a poorly thought out and misleading analogy of ancestry testing.
    – Jonah
    Commented Aug 17, 2021 at 21:48
  • 1
    @Ted Wrigley They're correlated strongly enough for ancestry tests to be possible, and for people to be able to accurately infer it by appearance alone (I'm pretty sure everyone on earth can tell a South African from an East Asian with 100% accuracy). Your eyes are not deceiving you, the correlations exist. And yes, race is heritable, but it's not causal in relation to geograpic origin. An extremely unlikely set of genetic mutations could make you look like (or actually be?) a different race, but it could not change your family tree.
    – Jonah
    Commented Aug 18, 2021 at 22:39
  • 1
    For the 10th time, @Ted Wrigley, this isn't a question about whether race is a valid taxonomic class for humans. Read the title. I've only been trying to explain that your model misrepresents heritage testing. And no, one cannot always find correlations that generalize well. I suggest you look into the basics of machine learning.
    – Jonah
    Commented Aug 19, 2021 at 14:12

Race is a social construct. Skin colour is a terrible proxy for actual heritage. Lineage is the biological reality. Every human has African heritage, because hominids originated in Africa.

Humans are unusually uniform, because of a bottle-neck 70,000 years ago (reasons unclear, was thought short & linked to a volcanic eruption, now thought longer & linked to climatic shift). More variation within single breeds of dog, than among all humans. The broad categorusations, African European & Asian, likely mean homo sapiens only, small percentage neanderthal, and small percentage neanderthal and denisovan, respectively (the Americas are a mix, Australians likely migrated directly from Africa). Neanderthal genes provided disease protection & low-light adaptions (critical in ice age), and denisovan genes gave disease resistance and greater altitude tolerance, essential to crossing Tibetan plateau (Malaca Strait & disease made a barrier to early hominids, to sea route to East Asia).

More variation between some black African groups, than between some black African groups & white European groups.

Genes for lactose tolerance have spread very rapidly. Significantly useful genes spread outside originating populations. Gene impacts on IQ for instance, are very close to the margin of error, vs massive impact of the motivation of test takers, effects of nurture etc. Things like facial features are just not a good guide to gene inheritance. Especially in a globalised world, & melting-pot countries like the USA & UK.

See other discussions here eg.

Are there philosophically serious moral arguments against eugenics?

Is identity a construct or something real?

Is it ethical to research potentially harmful topics?

  • 1
    Could you please explain why race being a social construct is relevant here? Aren't all taxonomic classes social constructs? Also, I already stated that there is little variability between races. Im more interested in how, from this, you can logically conclude that race and heritage are distinct.
    – Jonah
    Commented Aug 15, 2021 at 20:15
  • 1
    And of course we are all from Africa if you go back far enough, but your 23andMe test results don't just say "Africa" for everyone
    – Jonah
    Commented Aug 15, 2021 at 20:20
  • 2
    This answer doesn't seem relevant to the question. Either race is related to biological ancestry or it isn't. How society views racial differences such a skin color is irrelevant. The fact that if you go back far enough all humans are related is irrelevant. The question of whether it is ethical to research certain topics is irrelevant. Commented Aug 15, 2021 at 20:41
  • 1
    @CriglCragl, I don't agree that culturally, race is mostly skin color. You can think because of skin color that someone is white, and then meet his biological parents, and change your mind and decide he's black. East Asians are generally considered a race, but they have a wide variety of skin colors from Swedish pale to dark brown. Skin color is nothing but a rhetorical substitute for race; race is about ancestry. You can't change your race by changing your skin color. Commented Aug 16, 2021 at 2:38
  • 1
    @CriglCragl, race may be described (taking some liberties) as a heuristic, but if so, it is a heuristic involving ancestry, not visual appearance. I suspect that you are either trying to change the definition of the word "race" in a misguided notion that this will reduce racial animosity (it won't) or that you have been mislead by teachers who were trying to do that. Commented Aug 16, 2021 at 17:09

Not the answer you're looking for? Browse other questions tagged .