@Christy: Considering linear evolution (I’ll deal with branching below), in part you’re saying the same thing I’ve been saying, although not entirely.
The exact dividing line, as noted, is arbitrary but exists within broad parameters. There are humans today and have been for about 4.3 million years. There were nonhuman ancestors immediately before that and long before that. The nonhuman ancestors evolved into humans, and did so gradually, perhaps with as little as one molecule changing at any time. Even if every individual who lived 4-4.5 million years ago had been preserved under ideal conditions and we had everyone’s DNA and age and place and identified some as human and some as nonhuman, we still would be left with some individuals who could be either. Those who could be either were in what I called an in-between state and what I think a biologist above referred to as hominins; any of them might have been a human or a prehuman ancestor but we don’t know which and if we could time-travel back and meet them and conduct full medical examinations we still wouldn’t know. We could classify the genes and chromosomes so that every single body had to be one or the other, but that would, at the in-between stage, be arbitrary, in that it would be so precise as to have only weak scientific underpinning, at least as of now (that could change in the future as we develop good reasons for tighter taxonomic boundaries but I think we haven’t yet). So there was a community that we’re unsure about; they could have been either human or nonhuman.
But the community’s members were so similar in terms of what defines a species that they could reproduce with each other. Thus, it is possible that individuals who were among the first humans might have reproduced with individuals who were among the last prehuman ancestors. (If someone wanted to be gross and toss accuracy out the window, they might say, “you mean that beautiful lady in the red dress had sex with a hairy gorilla?” Remove the exaggeration and what’s true is that I’m saying that probably a human had sexual intercourse with a prehuman and thereby had a baby and probably that happened many times with no heightened medical risk or social disapproval.) They wouldn’t have known the difference even if today’s scientists had time-travelled back, arrived with precise classifications, proceeded to say, “you, standing by the tree, you’re human” and “you, wading in the brook, are a prehuman ancestor”, and taught the individuals all of our modern science. Even so, the two individuals could have looked at each other with no idea how to confirm or deny the modern visitors’ findings or declarations. If those two individuals were adults of different sexes and chose to attempt reproduction, they’d have had no unusual difficulty (beyond that faced by any two who are clearly of the same species). The baby might be either human or prehuman, then could grow up and reproduce the same way within the same paradigm of species-definition uncertainty, repeating the cycle until evolution had advanced enough that there’d have been no doubt that the resulting children were all human and any attempt by any of them to reproduce through a primate’s service would eventually fail, because the genetic differences would have become large enough to make certain that they were of different species.
In other words, human females from Group A did not have to wait for human males from Group B to save humanity. Group A could have reproduced within itself and Group B could have reproduced within itself, each Group being composed of members at an in-between stage and thus any member could have been newly human or an almost-human ancestor, with all of the members of one Group being similar enough genetically that they could reproduce together. If environmental conditions continued to pressure evolution in favor of humans, babies in both groups would have tended toward being human until humans predominated and then would have evolved enough that they could no longer reproduce with nonhuman primates.
There may be a way in which the first humans, whom I understand numbered at least 10,000, did not have sexual intercourse with any individuals who were prehuman ancestors. That would be if the first 10,000 included both males and females who were adults at the same time and in the same geographic community/ies and in sufficient numbers of each sex to generate enough offspring to lead to today’s 7,000,000,000+. Possible, but that’s less likely. If evolution from sexually-reproducing ancestral species required creating both sexes temporally that close together when they had not been doing so before, then that would be one more reason why most mutations fail: the two sexes not being even nearly simultaneous would end the evolutionary attempt. But if a new-species adult of either sex could reproduce with an immediate nonparental nearly-lateral ancestor because they’re similar enough to support the possibility, then that would make evolution easier, and I think that possibility is the more plausible of the two.
You wrote, “[i]ndividuals don’t change into a different species than their parents. Populations diverge over many many generations. At some point, the entire population is reckoned a separate species from the other descendants of an ancestor population.” I take the second and third sentences as correcting or counterbalancing the absolutism of the first. In that case, we agree. I’ve already said that change is gradual. But change happens. If some number of generations occur between what is clearly one species until there is clearly another species, and all of the generations in between changed in the same direction, we don’t say that the in-between individuals are not of any species. Either they’re in-between and thus of one or the other species (although we don’t say which) or we arbitrarily assign a species to each.
You wrote, “[y]ou cannot have a population where the males are a different species than the females. That doesn’t make any sense.” Yes, it does, and yes, we can have that population, provided the members who are of one sex and one species are able to reproduce with their immediate ancestral neighbors of the other sex, unless there are even more failed mutations because both sexes did not evolve into existence at about the same time and place. But because of the uncertainty of species-boundary definition, it’s likely that the genetically extremely similar could reproduce together across the arbitrary boundary.
Branching became a subject in this topic but I think I may have confused things there. The primate species from which humans evolved still, I think, exists, albeit itself having evolved, too. Humans branched off from a species of primate. But if two species branch from one so that three species result, the observation of that makes a more compelling case for recognizing and naming all of the species involved and for attempting to describe the species precisely enough to make identification of more members easier. But that’s a trivial point (meaning me who’s being trivial), because it doesn’t change that there’s a vaguely-defined area in between at each junction where one species evolves into another, where some individuals could be either the “before” species or the “after” species but we don’t know which.
When scientists draw a tree of evolution, I think humans come into existence from an ancestral species at only one locus on that tree. The tree is a summary, a scientifically valid summary but still a summary, and does not map every individual, but summarizes what’s known about species. Whether the number of prehuman-to-human evolutionary loci (on that conceptual tree) is limited to only one is what I was asking about at the beginning of this topic and that seems to be the consensus answer, unless I hear otherwise.