@Jay313 at some point you are going to have to read the article. You seem to argue that it is an absurd conclusions. It is not. It is a plausible estimate based on what we know of that time. With an extensive discussion of the archaeological evidence that bears on this. I would request you be less dismissive of science of which you little knowledge.
In the paper, we show with one plausible assumption, it is likely that we have very recent UGAs (e.g. 6 kya UGA for by AD 1) . Remember "plausible + likely => plausible", not a "string of maybe" as you put it.
However, most models do not need 6 kya Adam. For Adam 10 kya becoming UGA by AD 1 (before seas rise btw), the assumption becomes more plausible and the estimate becomes overwhelmingly likely.
We show why the convergence times are actually very conservative, and are expect in a better simulation to converge more quickly. The farther back in time, the quicker we expect convergence (because land bridges are still intact, and population is smaller).
That is why Jeff Hardin (in consultation with @JeffSchloss and @DarrelFalk) recently endorsed recent UGAs a "plausible" (not unlikely) assertion consistent with the evidence. http://biologos.org/blogs/deborah-haarsma-the-presidents-notebook/on-geniality-and-genealogy This is important. We are talking about well established estimates, based on a single plausible assumption. Perhaps @Sy_Garte can comment further.
Here is one of the figures from the paper. Notice how we say that science cannot discriminate between the two options? That is because to deny recent UGAs you have to make a different scientific assumption. We face a problem of "perfectly complementary total negatives." EITHER
- There was absolutely no mixing for at least one population for thousands of years (an absolute negative)
- No absolute isolation of any population for thousands of years (an absolute negative)
Both cannot simultaneously be true (in the way we construe them), at least one has to be true, and neither can be determined with evidence.
This is actually an interesting conundrum from an epistemological point of view. Science cannot establish absolute negatives of these kinds (which would require truly comprehensive knowledge of the distant past). Essentially, the ultimate question requires total knowledge to adjudicate, so it is outside of science to know. Looking at the data, however, we see mixing everywhere we look; not isolation. So though we cannot say for sure, recent UGAs are certainly plausible.
Still, the longer the time period, the more plausible #2 is, which is why one would struggle to find a population geneticist that did not think the identical ancestor point was earlier than 15 kya.
As I write in the paper, reaching the limits of science, there is "flexibility in the scientific account."