I suspect that if we studied and classified humans in the same way we studied a species of bird or beetle, we'd have no trouble identifying tens of different subspecies based on fairly distinct phenotypes and geographic distributions.
I'm sure a proper taxonomist would be able to speak more authoritatively, but I rather doubt it. Subspecies tend not to interbreed for whatever reason, whether geographic isolation, sexual selection, or other factors, whereas humans rather enthusiastically behave in the opposite fashion. I don't think one could convincingly argue that humans are geographically isolated, either. You might be able to make an argument that human races are akin to pet breeds, but my understanding is that zoologists don't really consider pet breeds as subspecies.
>whereas humans rather enthusiastically behave in the opposite fashion.
What are the worldwide rates of interracial marriage? I see in the US (a country where the majority race makes up about 60% of the population) that it's now around 10% for new marriages, and has historically been much much lower. If your statement were true, wouldn't we expect to see much higher rates of interracial marriage than 10% in the US? Wouldn't we expect that the historical rate was also much higher?
Please don't confuse interracial marriage (a social construct, and prevented by law in many places in the US until 1967) with interracial breeding (the thing that would matter for biological definitions). They can have very different rates, and surely did in the US in the 18th and early 19th century.
I'm not suggesting they're the same thing at all. They're just the best approximation I could quickly find any decent statistics for.
What were these higher rates you refer to? Were they as high as would be expected if humans 'enthusiastically' interbred, or even as high as would be expected if humans had no preference one way or the other?
According to that document, in 1890 (I chose that year because the document actually clearly describes the classification criteria for that year, unlike other years it mentions), the "Negro population" was 7.5 million, of which 1.1 million were "Mulatto" and 6.3 million were "Black" (there are some rounding errors here; see document for more exact numbers). The definition used in that census was that "Mulatto" only included people with at least one white grandparent; this would obviously include the ones with a white parent as well. This doesn't directly tell you anything about interbreeding rates per se because there may have been systematic differences in the numbers of children different types of couples had. But it does indicate that even in the 19th century a fair number of children who had at least one black parent also had a white parent. Somewhere between 1/14 and 1/7, if I'm running the numbers right (the former corresponds to a "mulatto" having one "white" grandparent and three "black" ones, while the latter corresponds to having a "white" parent and a "black" parent).
1/7 to 1/14 is incredibly low for a race that had been in close proximity to another for over a hundred years. If there were no preference, or if there were a positive preference for interbreeding, you'd expect by that time nearly all descendants of the initial minority race to trace most of their ancestry back to the majority race. That 6/7 to 13/14 of the members of the race can trace back the vast majority of their heritage to the initial minority race is proof that not much interbreeding was taking place.
I don't have exact numbers, and no one does that I've seen. Due to social disapproval people tended to keep such activities somewhat hidden.
That said, genetic analysis of African-American populations shows a significantly higher percentage of European-origin genes than can be accounted for by marriage rates, yes?
> Were they as high as would be expected if humans 'enthusiastically' interbred
What rates would you expect? To make this more specific and avoid questions of lack opportunity vs lack of enthusiasm, what rate of slaveowners having children with at least one of their slaves would you expect under the "enthusiastic" hypothesis?
(There's the complicating factor that in that situation at least one of the parties here might or might not have taken measures to avoid pregnancy even if sexual intercourse did occur; it's not clear to me how to account for that.)
>That said, genetic analysis of African-American populations shows a significantly higher percentage of European-origin genes than can be accounted for by marriage rates, yes?
Yes, and a much lower percentage of European genes than you'd expect if there were no preference one way or the other for interracial breeding, and a much much lower percentage than you'd expect if humans enthusiastically interbred.
>What rates would you expect? To make this more specific and avoid questions of lack opportunity vs lack of enthusiasm, what rate of slaveowners having children with at least one of their slaves would you expect under the "enthusiastic" hypothesis?
In the early 1800's, blacks were about 20% of the population, and whites were most of the rest. If there were no preference one way or the other, we'd expect 20% of children of white women to be fathered by black men, and 80% of children of black women to be fathered by white men. If the actual rates are lower than that, then some kind of self-segregation is happening.
Pointing out that slavery or other social institutions limited the rate of interbreeding is not an argument in favor of the idea that humans enthusiastically interbreed. If when living in close proximity they tend to set up systems that limit interbreeding, that is evidence that they don't enthusiastically interbreed.
>(There's the complicating factor that in that situation at least one of the parties here might or might not have taken measures to avoid pregnancy even if sexual intercourse did occur; it's not clear to me how to account for that.)
We don't have to account for that because we're discussing rates of interracial breeding, not rates of interracial sex.
> blacks were about 20% of the population, and whites were most of the rest
Yes, but this ignores the geographic inhomogeneity of both populations. For example, in Boston blacks were a much smaller fraction than 20%.
Obviously if you're white and you live in a village in which everyone else is white too, your chances of having kids with someone who is not white are pretty low. This is why I suggested the more specific metric.
> then some kind of self-segregation is happening.
You're assuming either homogenous population distribution or a lot more population mobility than was happening at the time, I think. That is, you can conclude there is segregation in mating, but the extent to which it's _self_-segregation is a much harder question.
Also, there were non-racial and non-geographic sources of segregation, like class/income. Note the comments in the wikipedia link below about black/Irish couples being a thing because the Irish were very poor.
I'm pretty sure it's possible to account for these effects and discover that there were still racial self-segregation effects, but they will not be nearly as large as a naive 80-20 calculation like yours would suggest. They would, I expect, be larger than the (also existing!) self-segregation effects amongst white people of different ethnicities.
An interesting thought experiment here is what you would expect to happen if you drop two (cis, straight) people of opposite gender but different races on a deserted island for a few years and how this compares to what happens when members of different subspecies are placed in proximity to each other with no other available mating partners. I don't know the answer to the latter, unfortunately.
> Pointing out that slavery or other social institutions limited the rate of interbreeding
I'm not doing that, precisely for the reasons you describe. I'm suggesting that limiting to my specific scenario removes issues of access due to population distribution effects and asks a specific question: given a white male who has the option of having children with black females (yes, the "enthusiasm" is very not symmetric here), what is the chance he will do so? Yes, this ignores the white-female-black-male case (though according to https://en.wikipedia.org/wiki/Mulatto#Antebellum_era claims this was also going on; I'd have to look up numbers on "free people of color in the 1790–1810 censuses in the Upper South" to have an idea of how common it was).
> We don't have to account for that because we're discussing rates of interracial breeding
Well, we're discussing level of enthusiasm. Which as I said above was probably not symmetric, for reasons not necessarily related to race per se. I expect male white masters of female white indentured workers were also more enthusiastic about having kids than the workers were.
>Yes, but this ignores the geographic inhomogeneity of both populations. For example, in Boston blacks were a much smaller fraction than 20%.
The fact that the races end up living in different areas is strong evidence that when put in close proximity, different races don't want to interbreed.
>You're assuming either homogenous population distribution or a lot more population mobility than was happening at the time, I think. That is, you can conclude there is segregation in mating, but the extent to which it's _self_-segregation is a much harder question.
The vast majority of blacks came over as slaves. That means they all started out living in white areas. If they later chose to move away from white areas, for whatever reason, I don't see how anyone can possibly argue that they interbreed enthusiastically. When they find themselves in close proximity, they choose to segregate by race.
>Also, there were non-racial and non-geographic sources of segregation, like class/income. Note the comments in the wikipedia link below about black/Irish couples being a thing because the Irish were very poor.
The reason for segregation is not relevant, as long as it's not a one-off occurrence. If the segregation occurs because the dominant race prevents the subordinate race from succeeding economically, segregation is still occurring.
>what is the chance he will do so?
This doesn't give us any insight whatsoever into the question. We'd have to compare this to how white men behave when they have white women as slaves.
>claims this was also going on
There's no doubt that some interracial breeding was occurring. There's also no doubt that it's far less than would occur if there were no preference one way or the other for interracial breeding, and far far less than would occur if humans enthusiastically interbred, rather than self-segregating when given the opportunity.
>Well, we're discussing level of enthusiasm.
I don't think we're talking about how emotionally happy humans are when doing something, I think we're talking about how likely a certain event is to occur. The overall topic is whether humans can be considered sub-species, and the argument was that they shouldn't be because they interbreed enthusiastically. I'm pointing out that they don't actually end up interbreeding nearly as much as you'd expect if, for some reason, they had no way of distinguishing members of their own race from members of other races.
> The fact that the races end up living in different areas is strong evidence that when put in close proximity, different races don't want to interbreed.
That's true, but (1) that's not what I was talking about; I was talking about geographic inhomogeneity in original migration patterns and (2) this argument applies to ethnic/language/religious groups in general, not just races. More on this below.
> That means they all started out living in white areas.
Sort of. https://en.wikipedia.org/wiki/List_of_U.S._states_by_histori... shows that in 1850 South Carolina had 668k people of whom 385k were enslaved (hence not choosing where they lived, exactly). I didn't look up the number of free blacks, but I expect it was quite small given the laws about re-enslavement at the time. Anyway, more like the white people were living in a black area there. ;)
But the real point is that for a while there various parts of the country did _not_ have appreciable black populations because slaves had not been brought there while your typical plantation had a much larger black population than white population. Voila, _involuntary_ geographic inhomogeneity.
> The reason for segregation is not relevant
That really depends on the question you're trying to answer.
> We'd have to compare this to how white men behave when they have white women as slaves
Sure. But the worst case for the "enthusiastic" hypothesis is that in this situation they always end up with kids, right? There may be value in comparing to this worst case, even.
> There's also no doubt that it's far less than would occur if there were no preference one way or the other for interracial breeding
Sure. I don't think anyone reasonable is arguing there was no preference on the matter.
But if we're going to talk about sub-species, exact frequencies and cutoffs do matter if the whole thing is to make sense other than as a mood-affiliation labeling exercise ("_these_ two groups are different sub-species, but these two other groups are just less likely than average to intermarry/interbreed"). There is good data showing that breeding across religious and language boundaries is less common (and was even less common in the past) than breeding within those boundaries. And yet very few people nowadays will argue that, say, the French and the Germans are separate sub-species, or were separate sub-species 100 years ago. Yes, I realize that 100 years ago people were in fact likely to make just such an argument. :)
Heck, inter-political-party marriage is a lot less common than within-political-party marriage in the US. And the rate of inter-political-party marriage, in the US, is dropping, while the rate of interracial marriage is rising. According to http://fivethirtyeight.com/features/how-many-republicans-mar... the former is sill more common than the latter but not _that_ much more common. I expect the lines to intersect within the next decade or two if current trends continue...
I guess we could argue that sub-species boundaries are malleable and they can merge or separate over a timescale of a few generations. That would be a valid take on the whole thing, I guess, but then I question the usefulness of the "sub-species" labeling; it seems to not have a useful definition in this situation from my point of view.
>this argument applies to ethnic/language/religious groups in general, not just races
OK, but that doesn't take away from the fact that it applies to races.
>Sort of. https://en.wikipedia.org/wiki/List_of_U.S._states_by_histori.... shows that in 1850 South Carolina had 668k people of whom 385k were enslaved (hence not choosing where they lived, exactly). I didn't look up the number of free blacks, but I expect it was quite small given the laws about re-enslavement at the time. Anyway, more like the white people were living in a black area there. ;)
Haha fair enough, but in this case we'd expect interracial breeding rates in certain areas to be extremely high (approaching 50%) but we see nothing even close to that.
>But the real point is that for a while there various parts of the country did _not_ have appreciable black populations because slaves had not been brought there while your typical plantation had a much larger black population than white population. Voila, _involuntary_ geographic inhomogeneity.
If the rates of interbreeding are diminished because one race purposefully holds the other race in small confined geographical areas, that's still an example of interbreeding being limited despite close proximity (i.e. less than a few miles).
>Sure. I don't think anyone reasonable is arguing there was no preference on the matter.
Someone was arguing that there is a preference for interbreeding. That's what this discussion started over -- a claim that humans 'enthusiastically' interbreed.
>And yet very few people nowadays will argue that, say, the French and the Germans are separate sub-species, or were separate sub-species 100 years ago.
You're missing the crucial first criteria for subspecies, which is that they must be "separate groups that are clearly distinct from one another". Frenchmen are not clearly distinct from Germans. White republicans are not clearly distinct from white democrats. White catholics are not clearly distinct from white protestants. White people are clearly distinct from black people.
Really? You think that walking up and trying to talk to one won't make it obvious which is which within about 30 seconds?
If you really prefer, replace French and Germans with Swedes and Italians. Even the visual appearance there is not that far off from the visual appearance difference between "white" and "black" as the terms were defined in the US in the 19th century. In both cases you can find some people that can "pass" as the other group, but the typical members look quite different.
>Really? You think that walking up and trying to talk to one won't make it obvious which is which within about 30 seconds?
I'm clearly talking about ethnic Frenchmen and ethnic Germans. I would not be able to tell an ethnic German raised in France from an ethnic Frenchman raised in France.
>replace French and Germans with Swedes and Italians.
Northern Italians I think you'd have trouble telling from Swedes. Southern Italians can trace a lot of their ancestry back to non-Europeans and so look different.
Actually, I have a historical example I'd like your opinion on. 18th century Warsaw, Jews and Poles. Clearly distinct groups, near-zero interbreeding. Separate sub-species?
In what way were they clearly distinct? Was it their culture, i.e. how they spoke and how they dressed, or was it their genetics, i.e. how they looked?
Could you describe to me how they looked different? I'm not looking for cultural body modifications, but descriptions like skin color, skull shape, and cranial capacity.
Typically somewhat different shape of facial features. E.g. the long nose stereotype is there for a reason.
The fact that you think cranial capacity is a "looks different" that would affect people's choice of sexual partners is ... odd. It's not something you can tell from the outside, typically.
> E.g. the long nose stereotype is there for a reason.
Is there actually any science behind that? I'm under the impression that it's not actually possible to identify any ethnic groups based on the stereotypical jew nose.
>The fact that you think cranial capacity is a "looks different" that would affect people's choice of sexual partners is ... odd.
I don't think that. You're asking me if Jews and Poles were different sub-species. Different sub-species must have different patterns of attributes. It doesn't have to be an attribute that people sexually select for, or that can be observed from the outside. I'm asking whether scientists could find such different patterns of attributes. My guess is that the answer is that there wouldn't be anywhere near sufficient differences to warrant considering them a sub-species, but I don't have near enough information to say.
According to Wikipedia[1], one way a species can be considered monotypic (not having multiple subspecies) is with the following criteria:
> The variation among individuals is noticeable and follows a pattern, but there are no clear dividing lines among separate groups: they fade imperceptibly into one another. Such clinal variation always indicates substantial gene flow among the apparently separate groups that make up the population(s). Populations that have a steady, substantial gene flow among them are likely to represent a monotypic species, even when a fair degree of genetic variation is obvious.
That sounds like humans to me, so I think we would be considered monotypic. Perhaps before globalization this would not be so.
I think I take issue with that whole line of logic. Red and yellow fade imperceptibly into one another via orange but they are still distinct colors. Whether there is a sharp dividing line or not this is still the case. It would not be logical to claim the distinctness of A and B depended on a third entity C existing. You shouldn't need to know anything about C to say whether A and B are or aren't distinct.
Claiming that orange makes red and yellow one and the same is the continuum fallacy. It does not follow that an intermediate population must be evidence of substantial gene flow between populations, rather simply that contact exists. What does substantial mean anyway?
Moreover most human ethnic groups probably do not have "substantial" gene flow between one another, else how did they arise in the first place? With no data to back the claim up I guess less than 5% of people are "racially mixed".
If an idea is obvious enough to occur in an armchair biology discussion on HN, it's likely those who do this sort of thing as a profession have considered it already. Indeed, https://scholar.google.com/scholar?q=human+subspecies+taxono... --- the outcome appears to be: "Human 'races' are below the thresholds used in other species, so valid traditional subspecies do not exist in humans.".
It's also possible their interpretation is shaped by a desire to promote social justice and anti-racism, and oppose harmful stereotypes, taking precedence over strict scientific consistency. This is understandable, and ethical; I'd do the same thing if I thought scientific claims could cause harm. Differentiating humans into biologically distinct groups has contributed to scientific racism and genocides in the last century.
Before suggesting that such results are wrong, you'd need to read some of the papers and find something specific to criticize. Otherwise, that's just empty ideologically colored speculation.
