Picking up from this thread, at thispost.
Not your anecdotes, or logic, or rationalization. Just provide the definition like this:
The human races are ____, ____, … , and ____
You can determine the race of somebody by examining _________
From that determination of race you can determine ________
If your contention is true, it can’t be that hard to do.
I started this thread a while ago, which is obviously similar.
But the position I ended up with is more of a gray area than you seem to be acknowledging. I don’t think that you can divide humanity into a fixed and finite number of discrete groups based on purely biological characteristics that will correspond with the generally accepted notions of race. At the same time, I don’t think those groupings are completely orthogonal to biology. That is, if you took all the people in the US and divided them “white, white, black, asian, not sure, not sure, latino, black, white” into groups based using the traditional societal racial groupings based entirely on external physical characteristics, and you handed those groups you’d made to a biologist and said “do these groupings make any sense”, the biologist would come back and say “I see some patterns within the groups, if you give me some more people I can make guesses as to which group additional people would be sorted into, and would succeed better than chance”. Whereas if you did the same thing dividing by last-digit-of-phone-number or something like that, the biologist would not be able to do so.
So I’d say, using informal language, that while races are not really biologically defined, and are likely not biologically useful, they are not completely biologically meaningless.
But we’re not talking about the US. We’re talking about the entire species. The US is a little easier to get your arms around because not all the populations of the world are well represented here, and you lose the clinal nature of the species.
I like to pose the question this way. Start in Beijing and walk to Berlin. Tell me what the races are on that axis and where you draw the boundaries and why. Show me the place where people on one side of the line are objectively different from people on the other side of the line. (And not statistically, but individually.)
I feel like what I said still applies. Again, bear in mind how weak my claim is. I’m just saying “it’s not TOTALLY AND COMPLETELY AND UTTERLY BIOLOGICALLY MEANINGLESS”, not that it’s particularly meaningful or useful or anything.
I have no answer to that, and have never claimed I did.
I’m really not ever sure what the means. There are certainly times when it is meaningful to talk about certain populations have statistically more or less propensity for “x” than other populations. But those “populations” are fluid, and can easily change depending on what one happens to be studying.
We might say “people whose ancestors come from the Levant are more likely to be XYZ than those from outside the Levant”. But that generally means that as you move from the center of that population away from it, XYZ becomes less prevalent. It’s not like the Levant is some geographically isolated area with a boundary preventing genetic flow.
I never claimed you claimed you did. 
That comment was not directed at you, but rather as a way of re-framing the OP.
OK, the human races are the 100m, the 200m, the 400m, the 800m, the 1k, the 5k, the 10k, the half-marathon, the marathon, and the ultramarathon.
What next?
Soul. N.B.: The “One Drop Rule” is unscientific; to qualify as black, you need a soul count of 4.8 millibros per cc or higher.
I think my first post in this thread sums it up nicely… take a bunch of people, group them into categories, and then take blood samples (which contain DNA) (or something else which doesn’t allow inspection of appearance) and send those groups off to a super-brilliant but naive biologist (maybe an alien, so they have no preconceptions about what is and is not important), and ask them two things:
(a) do these groupings in any way correspond with groupings you might have come up with yourself, if you were hierarchically sorting all these samples on any criteria you find useful or interesting
(b) have you noticed enough patterns inside each group that you would be able to predict with some accuracy (not necessarily 100%, but greater than chance) into what group future samples would go
If the grouping was by gender, or by “people with Downs Syndrome vs everyone else”, then I expect the answer to both (a) and (b) would be a resounding yes. If it was by last digit of SSN, or astrological sign, then I would expect it to be 100% no (barring some possible contrived corner cases that might give very very slight patterns). I think groupings by perceived stereotypical race would be between those two extremes.
Is that a really fair and applicable way to define race though? If a population is clinal, as humans are, then it’s certainly fair to separate out type localities. Take for example, The Boa Constrictor. The ones from Columbia are nearly twice the size of the ones from surinam, darker in color, and have a heavy pattern. Surinam boas are smaller, lighter, have little to no patterning and have bright red saddles on their tails. They are the same species, and between the two population types there are a lot of intergrades. Yet we have no problem with identifying and separating them out for useful purposes. Why should humans be any different?
A theoretical alien collector of humans would certainly be able to distinguish visually between a H.sapiens (Equatorial, African) and a H. sapiens (Arctic, Inuit). Likewise it would probably recognize most of us as intergrades from between areas of population homogenization.
You’ll have to be more specific about what you mean when you say “we have no problem with identifying and separating them out for useful purposes”. Because we do that with humans all the time, too. It’s just that there isn’t any one objective way to separate out populations and we’ll do it differently depending on what we’re trying to do. We might talk about “East Asians” or we might talk about “Koreans and Japanese” or we might talk about “Japanese” or we might talk about “Okinawans”.
Good point.
Space, arms, rat, and for the cure.
The Susan G. Komen Foundation’s lawyers will be calling…
What is it exactly that you think you are proving here?
So what? We could do that, and it would probably work pretty well. We could do that all over the world, too. But the point is that you could chop up the species into any arbitrary number of populations, and you wouldn’t find clear boundaries anywhere. Then we’d be left fighting over whether there were 5 or 500 races. You could make the case for almost any arbitrary number of classifications you would want.
We could go back to the days when we talked about an “Irish race”. We could have an Icelandic race and a Sicilian race and a Maltese race and a dozen Vietnamese races. Who is going to decide which classification is the correct one?
Well, that’s kind of MaxtheVool’s point (if I may be so bold . . .). All he’s saying (and I agree with him) is that the term “race” can be used to meaningfully put people into groups using their biological characteristics. The OP of this thread seems to disagree with that (although he’s a little fuzzy on exactly what he believes at this point).
Are the breeds of dogs a meaningful biological definition?
How does one draw the line between a beagle and another breed of hound? What might the “races” of dogs be? How many “breeds” of humans are there?
You may be so bold. 
Just because we can devise a scheme to classify people using DNA, doesn’t mean that classification tells us anything about that group biologically. Let’s took at look at one of the problematic classifications (and this is just one like this that you will find around the world): Hispanics (what Max calls “latino”).
Three hypothetical people are in this group: Juan, Pedro and Maria. They all are an admixture of European, African and Amerindian ancestry. Our ET biologists have figured that, for some reason, we Earthlings like to put people with almost any of that admixture into a “race”. But Juan is 80% European, 5% African and 15% Amerindian; Pedro is 70% Amerindian, 15% European and 5% African; Maria is 75% African, 20% Amerindian and 5% European.
It makes no sense, biologically, to put them in the same group if you want to make some prediction about what their biology is. You would really want to put Juan in with Europeans, Pedro in with Amerindians, and Maria in with some African population (which is more diverse, genetically than any other continental group of people), if you wanted to make the best prediction about their biology.
But then the ETs are going to start asking: what does it mean to be European? Where does Europe end and Asia begin, genetically? And when we say “African”, does it make sense to lump in people from Camaroon with people from Ethiopia? Not really.
It’s a mess. You can classify it any number of ways, but you can’t avoid the fact that it’s a mess all over the place.
Why would you think up a scheme that omitted the vast, vast majority of dogs? Dog breeds are artificially kept in genetic isolation. They are purposely bred for certain characteristics. There is no analog in the human species.
There are no “breeds” of humans.
Yep, I agree with that.
But that doesn’t mean that if you examine Barack Obama and Bill Clinton you can conclude only that “there’s no meaningful way to differentiate between these two people based on biology alone.” It doesn’t mean that if you examine Chelsea Clinton you’d have no way of predicting whether Bill or Barack is her dad with a probability greater than chance.
As I’ve said in other threads, I think many people here have essentially let the racists win. They’ve gotten so fed up with racists that they’ve taken the silly position that “there’s no such thing as race.” I think it’s possible to say that the racists are wrong but it’s still possible to differentiate among different races based on biology alone.