What do you mean by cladistics? Aren't molecular phylogenies fundamentally cladistic?
I find your example of maniraptorans and birds interesting. For one thing, "birds" were recognised as a separate group long before we had any inkling of non-avian theropods. Had we known about, say, cute fuzzy tyrannosaurs millennia ago, would the line now be drawn somewhere else? I agree it's meaningful to create taxonomic entities based on shared traits. But the tree of life is a fractal; which branches we highlight and what traits we deem interesting enough is a function of our perceptions.
Second, while we recognise birds as an entity, we *also* recognise that "maniraptorans minus Aves" is not a monolith but contains a lot of diversity. By the same token, it makes sense to recognise and study the diversity within the paraphyletic group of "Sub-Saharan Africans minus other people". (Maybe it's already being done, I don't know, I don't follow this area at all.)
(Also, the analogy may not be the best because non-African populations are commonly subdivided. Having Africans vs Europeans and East Asians and Australians seems more like maniraptorans vs penguins and ratites and songbirds than maniraptorans vs birds. And no, I have no clue about avian interrelationships, but I hope you get the general idea )
I'm pretty sure the "African" in African-American is understood to mean "Sub-Saharan".
Africa has immense genetic diversity, so if you're going to divide humanity into races, Sub-Saharan Africa should definitely get more than one. I mean, if "Hispanic" is a different race from "Caucasian"...
Of course, that's what you get when "Caucasians" do the classification.
I do find it funny that most of the "Caucasian" population of the US doesn't originate from anywhere near the actual Caucasus. "Western European" would be a far better description.