Re: the E1b hypothesis: We need to remember that we're talking about two different European E1b clades, E1b-M81 (probably came over from North Africa) and E1b-M78 (probably came over from the Near East). Now we're pretty confident that E1b-M81 came over well before the Moors, possibly as long ago as the Neolithic (I'm not familiar with STR dating of this but that's what I've read). So if we're talking about migrations out of pre-R1b Iberia, there's a very good chance that they had a lot of I2a1a, G2a, and E1b-M81... in fact, the spread of E1b-M81 into France is very likely explainable by Beaker Culture. I don't see a similar pattern for E1b-M78 though, so I'm not sure it solves our E1b problem.
There are much more than two subclades of E1b1b. The more I think about it the less sense it makes that E1b1b expanded from the Near East, mainly because all subclades of E1b1b are found in East Africa, even the so-called European ones like V13. I posted
a new hypothesis last month about the possibility that the lush green Paleolithic and Mesolithic Sahara was once inhabited by a great number of E1b1b people, who crossed the Mediterranean in search of more fertile land when the Sahara started to desertify.
- E1b1b1
a (E-V68) => found in Sardinia and Africa but not in the Middle East (clearly pointing to a direct migration from North Africa to Sardinia)
- its subclade E1b1b1a1 (E-M78) => found in East Africa, North Africa, the Middle East and Europe. E-V13 is one of its numerous subclades and is found all over North Africa as well as in Europe and the Middle East. All the other subclades are mostly East African (V12, V22, V32) or North African (V65), but not Middle Eastern.
- E1b1b1
b (E-L19/V257) => the parent haplogroup of E-M81, recently discovered by
Trombetta et al. (2011). It is so far found in Kenya, Morocco, Corsica, Sardinia, and Spain, once again showing a direct Africa to Southwest Europe movement. E-M81 itself is found all over the Sahara, as far south as Burkina Faso, and as far east as Sudan.
- E1b1b1
c (E-M123) => extremely wide distribution, throughout East Africa, North Africa, Europe, the Middle East and Central Asia. No clear pattern of distribution so far, but some pockets here and there with higher frequency due to the sparsity of tests.
The lack of clear pattern and the very old age of these haplogroups and their surprising diversity for their age (a sign that they descend from a large Paleolithic population) all agree with a pan-Saharan distribution in the late Paleolithic (from circa 20,000 years ago) until the last desertification of the Sahara roughly 6,000 years ago.
It is a mistake to try to comprehend the world as it was 10,000 or 15,000 years ago by looking at how it is now. Just before the end of the last Ice Age, the world was a totally different place. Northern Europe was uninhabitable, southern Europe was had a climate closer to modern Scandinavia, and North Africa was a vast Eden, teeming with game and fruit trees that could support a very large hunter-gatherer population (certainly more than all Ice-age Europe put together). The Middle East, as far east as Afghanistan and Pakistan, was also much greener than today and filled with big mammals of all kind. That's why most of the late Paleolithic haplogroup diversity was to be found from the Sahara to South Asia, and haplogroups such as R1a and R1b probably arose in such a region too.