The heliocentric quote, which I thought was the object of your concern, is about humans (Homo sapiens). When I’m speaking about our lineage leading up to humans at 200KYA I use “lineage” or similar.
The other two quotes remain valid. Does “it seems” sound like I’m saying this is as certain as heliocentrism? That would be quite the understatement. That is a summary statement of all the lines of evidence in the literature to date that do not provide support for a bottleneck below ~10,000 at any time in the last 18MY (which remains the case).
"All methods employed to date agree that the human lineage has not dipped below several thousand individuals for the last 3 million years or more – long before our lineage was even remotely called “human”.
This quote also remains valid. There are no studies in the literature that support a lower bottleneck, and several that support large Ne values over this timeframe (PSMC and LD studies, for example). If there were (perhaps if I had missed one somewhere?) I’m sure you would point it out if you were aware of it.
So: “heliocentric certain”: humans. Pretty darn certain: lineage leading to humans over the last several hundred thousand years (say back to ~500,000 years ago). Confident but not as definitive: lineage over the last few million years. Survey of literature to date: no evidence of a bottleneck greater than thousands anywhere, regardless of time.
You suggest that perhaps the data in Zhao could go back to 4 haplotypes in 178,000 years. How certain are you about that value? You have to (a) pick the very lowest value within the 95% CI and then (b) assume that 1/4 of that is reasonable in this case. One quarter of the mean value is 339,000 years ago, which pretty much any scientist on the planet would say is more accurate than cherry-picking the lowest value. The upper bound (528,000 years ago) is just as probable as the lower bound. I could pick that value with the same confidence which which you pick the lowest one.
If that’s what counts as a “win” that’s a pretty thin “win”, don’t you think?
Ok, so that we can move this conversation forward to the stronger data: I agree that Zhao (2000) does not support the case I make in Adam and the Genome, in that it might be statistically possible to have the variation in their dataset come from 4 haplotypes less than 200,000 years ago.
Done and dusted. Shall we move on?
Are you ever going to answer my questions about common ancestry and what you think are reasonable speciation times? They are directly relevant to determining mutation rates.