John Harshman: Bottlenecks and Trans-Species Variation

180 thousand years. We already looked at the data.

1 Like

The analysis will be done and the results will be surprising and unexpected.

1 Like

I agree. Maybe the team will even include @John_Harshman, @glipsnort, or myself. Real scientists stalk these forums.


I’d say it is, especially the massive mosaic being discussed (all human diversity). Of course the genetic mosaic idea does dispose of any problems with Eve being a clone; that would no longer matter. Still, a massive genetic mosaic is still a pile of extra miracles.

It does remind me of the early notion of preformation, in which the homunculi of all Eve’s descendants are present, like nested Russian dolls, within her body.


It is fantastic, and I’m not personally comfortable with it.

How can genealogical sole-progenitorship be supported? It’s even more impossible than genetic sole-progenitorship, which is at least theoretically possible given a miracle of coalescence. And in fact genealogical sole-progenitorship would entail genetic sole-progenitorship, though the reverse is not true. You have lots of genealogical ancestors at any time-period, lots more than you have genetic ancestors, as you yourself have demonstrated. Unless you mean something by “sole” other than sole, this doesn’t work at all.

Have you forgotten the MHC alleles already?

1 Like

Not definitive.

And I had such hopes for you. Why not? If one sums the alleles shared between humans and chimps, humans and gorillas, humans and orangutans, etc., it comes to more than 4.

But my more urgent question is on the meaning of genetic and genealogical sole-progenitorship, particularly the latter. Either they don’t really mean “sole” or you have a scenario I haven’t seen yet.


We discussed this at length. One extra allele could be convergent. We need to do intron study. Until we do it is not clear.

I’m really disappointed. This is not how the discussion ended. And what is “intrinsic study”?

I’m also disappointed that you have twice ignored appeals to explain “sole-genealogical progenitor”.

Was gonna start a thread just for that. Just give me a moment. Not ignoring you. relax.

1 Like

I want to settle this HLA issue first here, then I’ll start a thread on genealogical progenitorship.

I meant the “intro study” we discussed here:

First, off it does appear I ow you a bear…

I need to settle precisely how many trans-species alleles there are in this studies. I haven’t had a chance yet, and will do so soon. I’m expecting you are going to be right @John_Harshman, though I haven’t settled it firsthand yet.

So this how I see the evidence stacking up…

  1. Genome-wide divergence: 500 kya (very strong)
  2. Introgression: 500-700 kya (very strong)
  3. HLA exon diversity: uninterpretable because of balancing selection
  4. HLA introns: uninterpretable because of draft and balancing selection
  5. HLA exon inter-species variation: possibly in 1 case up to 5 alleles in one locus, and could have been convergence.
  6. HLA intron inter-species variation: not currently in literature.

I think that #6 is the most important experiment to do, and I encourage it. Even then, however, it would only be one line of evidence. Still, depending on the strength of the results, I’d call it enough. That analysis has not yet been done. The best number, then, is 700-500 kya, with the common ancestors of Sapiens, Neandertals, and Denisovans.

So, my assessment is that there is strong evidence against a bottleneck more recent than 500 kya. More ancient, the evidence has been overstated in the past (see Ayala and Venema) and it is much less clear. It is possible that further work will rule out Gauger’s preferred sole-genetic progenitor scenario. I welcome work to settle this question for good, but in the mean time we do not know from evidence yet. We have testable hypothesis, and a clear research plan. That is it.

I want to add @John_Harshman that I want to get the science right. This is actually a big reason I even have this forum. I want to resolve all objections by legitimate scientists like yourself. I’m committed to taking all your scientific objections very seriously, by either successfully making my case or changing my position.

Feel free to complain if I don’t get to a legitimate point after a week or so, but please don’t jump to conclusions after less than a day, especially late at night when conversations are ending. That is not a reasonable standard.

Doesn’t any sole-genetic progenitor scenario require a deceptive God who makes it look as though humans are related to other mammals? Or are you (I mean Gauger, really) talking about a 2-person bottleneck in a previously large population?


I believe its a two person bottleneck from a previously large population. One of the big criticisms from @glipsnort was what kind of natural disaster could cause that sort of bottleneck and leave the necessary conditions to repopulate. Then you also have to take into account this two person bottkeneck has to be male and female and able to breed. Why im not fan of this model. A mature male and infant boy arent going to do you much good. If the bottleneck only happened to
A small subset of the total population, what prevented that pair from reconnecting with other populations? What kind of geographical barrier? One thing we know about hominins is that they sure do move around. It’s not even genetics that makes me strongly question that model.


Yes, and one should also consider how unlikely it is for a two-person sample of a population to have maximum genetic variation in, apparently, every single gene. And how unlikely it is for a population to survive a bottleneck that extreme. Of course with enough miracles anything can happen.


That is actually a very significant issue. This is going to be hard to explain to Creationists, but the right number to look at is TMR3A. I’ve done the simulation work on this but haven’t put it out yet. That pushed the date back to about 700 Kya.

I’ll say though that the mosaic loophole is already were people are heading. It seems the Noahic Bottleneck is going to be more important.

Whatever the case, AIG style YEC has no place to hide. RTB will likely have to make some adjustments but might survive if they do. In my view, however, none of this makes any Theological sense. The Genealogical Adam view is far more coherent with everything. I’m only engaging the bottleneck question because people ask about it and I want to be rigorous.

1 Like

Remind me what TMR3A is. Does it deal with genes under frequency-dependent selection like MHC etc.? There seem to be a fair number of them.

More important for what? Falsifying YEC? If so, don’t see the point. YECs won’t believe it, and nobody else needs to be told.

1 Like

I’ve focused on Time to Most Recent Four Alleles (TMR4A). TMR3A is just Time to Most Recent Three Alleles. This is computed as a genome-wide median. The justification is explained in depth elsewhere: Heliocentric Certainty Against a Bottleneck of Two?.

I do aim to clarify this point eventually. It is going to be a harder explain to creationists, as population genetics is already just so non-intuitive. Still, I’ve wanted to understand this, and will be likely publishing on this soon too. It has has taken a lot of simulation work that I have not yet put in public. In the end, it is not that consequential, because it just pushes back the date from 500 kya to about 700 kya (or so).

There are many types of YEC. I know you don’t believe me. Just trust that I am not crazy, and wonder if I might know my own community better than an atheist. That just possibly might be the case, right? In the end, time will tell. Let me make my case, and let’s see what happens. Certainly let me know of any scientific errors you find. I need this to be above reproach. On what YECs will or won’t do, that is a different question. Let me make my case without having to guard my flank.