Circular arguments

GrumpyMrGruff wrote:

otseng wrote:However, the question is what can explain that all other female lines disappeared? In terms of the probability that only one female is the progenitor of all, I’m going to ask Zeeby in my next post about that.

As Zeeby pointed out, in stochastic birth-death models like this the number of progeny left by each member of the initial population will inevitably go to zero or one after a sufficient amount of time (dependent on model parameters). The correct question is not if one female’s offspring will eventually dominate the population, but how long (on average) it will take for this to occur. To answer that question, we need to know things like population structure, generation time, population size, etc.

Zeeby never modeled it going back to only one.

Also, if what you say is true, out of all the females that are living now, only one female will account for all the descendants at a particular time in the future, though it may be a long way off.

Figure 1: Haplogroup A Biogeography Circa 1500 (Before Modern International Transit)

Your images say “Circa 1500”. How exactly could they have determined the genetic distribution around 1500 AD?

Because BT, CT, and CF are all hypothetical and not found in any human, it is entirely possible that F can be located nearer to the root of the tree.

You lost me here. This sounds like a non sequitur. yAdam is hypothetical in exactly the same way that BT, CT, and CF are hypothetical: Their markers, though shared by many males (the default marker set in the case of yAdam) are not found in isolation in any living male.

I’m not saying because that because the haplotypes are hypothetical that they could not have existed. I’m suggesting that because they are hypothetical that the known haplotypes can be in a different tree structure.

The phylogenetic analysis sets branch length proportional to amount of change between samples.

Could you provide a reference for this? Are you saying for instance, that there are more variations within the A haplogroup than the F haplogroup? And if so that it is necessarily older?

GrumpyMrGruff wrote:Note that every phylogenetic tree we make assumes homology.

Would even phylogenetic trees based on genetics also assume homology?


Then would not using phylogentic trees as evidence of homology be a circular argument?