Archiver > GENEALOGY-DNA > 2009-07 > 1247414722

From: adam bradford <>
Subject: Re: [DNA] L21 oriigns
Date: Sun, 12 Jul 2009 12:05:22 -0400
References: <><><><>
In-Reply-To: <>

I am still suspicious about that heavy concentration on the Rhine. It could
be more a reflection of the sources for emigration to America than it is a
reflection of the haplogroups' actual concentration in Europe, whether 4
years ago or 4000 years ago.

I am reminded of this passage from Bernard Bailyn's Peopling of British
North America (a must read):

"A thousand local rivulets fed streams of emigrants moving through nothern
and central Europe in all directions. . . A sampling of the sources of the
first sizable German migration to British territory, which scattered
ultimately to Ireland and New York - that of 1709-1710 - shows a remarkable
range of communities from which the migrants came. Approximately 42 percent
of the immigrants came from the Palitinate proper, west of the Rhine and
south of the Mosel. But as many came from east of the Rhine . .. .* This
was no concentrated exodus with a singular impetus, but a kind of seepage
from an otherwise almost invisible but continuous and normal flow of people
from village to village, from village to town, and from town to city. The
Palitinate, it has been suggested - long before America was thought of in
such terms - was a melting pot of European peoples*."

I think most of the modern day distribution and frequency of haplogroups
today are going to be the result of this "almost invisible but continuous
and normal flow of people" that Bailyn writes about. We tend to always
assume that the distribution has to be due either to some world historical
event or something that's going to show up in the archaeological record.
That is not necessarily the case.

I think we also have to be careful not to trust too much in the big premise
on which most of our haplogroup speculations are based - that people tended
to remain where they were in the vast period that precedes the point where
we begin to be able to see them moving all over the place.

For these reasons, I would tend to put more stock in whatever geographical
correlations might exist within a haplogroup phylogeny, like the clear SE-NW
cline in R1b and descendant clades, and a lot less stock in frequency and
even modern day (last 500 years) distribution. Those are just snapshots of a
point in time, and it is dangerous to infer too much from the past based on
how things look in the present.

On Sun, Jul 12, 2009 at 10:32 AM, Alan R <> wrote:

> Given the limitations of local intraclade dating of a clade outlined in my
> previous post and that it is essentially impossible to date geographical
> spread of a clade by only looking at the clade itself, I think its
> unlikely the origins of L21 or any other clade will be resolved much more
> than they are now unless of course L21 is broken down into subclades that
> straddle both the isles and the continent (within-isles ones are of no real
> use for this purpose) So, unless the latter happens then our tools for
> this job are pretty well what we have now. Essentially to date a clade and
> infer direction of spread you have to look at the sibling and upstream
> clades in terms of date, phylogenetic position and modern distribution.
> There are no other tools for the job and no within-clade techniques that can
> help.
> For L21, our only tools that really help for spread, date and origin are
> looking at the interclade dates between it and sibling/cousin clades and
> modern distributions of both. We can also look at similar a step upstream
> to look at slightly older distribution and dates of ancestral forms. it is
> so obvious that the heavy amount of L21 in the isles has been seized upon
> (probably by people from a genealogy angle looking for certainty) as an
> 'isles marker'. While statistically it may serve a purpose of heavily
> weighting the likely point of origin of new world ancestors of unknown
> origin towards the isles (purely in terms of sheer probability based on
> modern numbers) this does not make it an 'isles originated marker'. That
> is something quite different.
> In reality if you compare L21 and S28's distributions (maps on the FTDNA
> clades projects are very useful) its so obvious that they share an identical
> continental core area all along the Rhine (I suspect under-sampled France
> too) from the Alps to the North Sea. Yes they differ outside this core with
> L21 having a more northerly weight and S28 more southerly and easterly one
> but there is no denying the shared core along and near to the Rhine axis
> (and I suspect in France) is where an awful lot of the continental L21 and
> S28 (and S21) has been located so far. The differences in distribution have
> been grossly exaggerated in discussions. I mean seriously go to the FTDNA
> projects pages and print out the L21, S28 and S21 projects maps. You will
> see what I mean. They look like Rhine clades that have spilled out of that
> core is different directions at the extremities.
> Add to this the identical phylogenetic positions of L21 and S28 and add the
> almost identical dates arrived at through interclade variance dating (now
> pushed back by Tim towards the earlier Neolithic) and its clear that it is
> almost perverse to argue that the story of L21 and S28 differs to any huge
> degree. As far as I can see the only difference is that :L21 had some sort
> of founder effect when R1b reached the isles (a founder effect is very
> likely when settling an island) while at some point S28 spread a little
> south into Italy and a little east and S21 spilled a little eastwards from
> the mouth of the Rhine. Look at the maps and tell me that is not how it
> looks! The core continental distribution and the very close cousin
> relationship, the very similar interclade dating and the phylogenetic
> positions all scream that the early story of S28 and L21 is the same. The
> differences seem to me to be largely down to chance and founder effects.
> I have heard all the attempts to write off the non-isles L21 as
> out-of-isles migration but (perhaps with the exception of Norway) this just
> does not convince. It relies (given the similar intraclade dates for L21 in
> each area) on a bizarre scenario of a spread of the newly appeared S116 from
> the continent (where L21s sibling clade S28 remained) followed by an almost
> immediate return back to source from the isles after the L21 mutation. I
> think people fancied the Beaker and subsequent Bronze Age trade network as a
> vehicle for this boomerang scenario but Tim's new dates for L21, S28 etc
> make a beaker/Bronze Age isles origin for L21 impossible. That is
> significant because in pre-Beaker Neolithic times trade and cultural contact
> between the isles and the continent was hugely more limited than in the
> Bronze Age. Its important for those not familiar with isles prehistory that
> no-none (and I mean no one) in modern times has suggested any intrusion
> into the isles between the Early Neolithic c. 4000BC and the beaker period
> c. 2500BC. That was a period of pretty strong insularity when contact was
> largely just between the isles. So, if we rule out the beaker phase or
> later then the next stop back is the early Neolithic spread. There is
> (especially in the isles) no in-between option.
> I really think its time to open our eyes, look at the continental
> distributions of the main western European R1b clades and stop all this
> L21=isles, S21=Germanic and S28=Alpine Celtic labelling. It just does not
> appear to stand up to scrutiny if you look at the FTDNA projects maps with
> an open mind.
> Alan
> -------------------------------
> To unsubscribe from the list, please send an email to
> with the word 'unsubscribe' without the
> quotes in the subject and the body of the message

This thread: