Re: Multi-Regional vs. Replacement, was This might be an ev

J. Moore (j#d#.moore@canrem.com)
Wed, 24 May 95 10:27:00 -0500

Jo> >It does not, however, tell you anything about admixture or the lack of
Jo> >it, as the MtDNA proponents had claimed, and as is still claimed by
Jo> >people using it as support for the replacement theory. It tells you
Jo> >about one side of a family.

Jo> >In the same way, knowing that someones's last name is "Moore", which
Jo> >is English (although possibly changed from the Scottish [or Irish?]
Jo> >form of the surname) doesn't tell you what their ancestry consists
Jo> >of. In my case, for instance, it's mostly Danish. My name, though
Jo> >handed down through one side of the family, just as MtDNA is handed
Jo> down
Jo> >through one side of the family, doesn't show the whole picture about my
Jo> >ancestry.

Jo> Your example uses a sample size of 1. Pretty sus statistics!

It's not statistics at all, as I explain at the end.

Jo> On the other hand, if *everyone*'s name was Moore (known to be an
Jo> English name), this would be at least pretty good evidence for the
Jo> hypothesis that there were no (or at least very few) Danish ancestors
Jo> around.

Would it? Leaving aside the literalism in your complaint about sample
size (yet somehow not evident in your not accepting that last names are
often changed in form;-), it wouldn't tell you a damn thing about their
ancestry, except that there was at least *ONE* Moore in it. That's
actually ALL it can tell you for sure. In this analogy, everybody
except that one person could've been from somewhere else entirely, and
the amount of English ancestry could be infinitesimally small. Yet that
name could persist.

Now this is deliberately an extreme example, but I did it to show that
your conclusion does not necessarily hold up. What we are seeing with
MtDNA, as with surnames, is a single line being traced back until it
peters out. It tells us something, but it does not, as it supporters
seemed to claim, tell us everything. And it doesn't even disprove the
multi-regional hypothesis, as we would expect to find lineages going
back to somewhere (they have to, really), and it could as well be Africa
as somewhere else. But it doesn't show that *everybody* came from
there. Although it isn't likely that it's showing that only *one* person
came from there, that could in fact be the case.

Jo> I do have my doubts about the Replacement theory. But a valid
Jo> statistical refutal of the validity of the MtDNA argument must go *much*
Jo> further than just saying "It doesn't say anything about the male line."
Jo> I believe such valid (quantitative) refutals have been attempted, with
Jo> considerable success.
Jo> John

I am not doing a "statistical refutal"; I am merely showing the
inherent limits of what this worthwhile and important data can show us.

Jim Moore (j#d#.moore@canrem.com)

* Q-Blue 1.0 *