Recapitulation: 23,000 BP ~ 21,050 BC — The Ogolian aridity hits the North-to-equator section of Africa; much of North Africa and the Sahara are characterized by adverse weather conditions, with much of the region turning arid. The Sahara at this time, extends south beyond its current boundaries to a certain point, possibly a little beyond the Niger bend.
Arid conditions extend all the way to the "horn" coast of the African Horn region, possibly encouraging populations to reside more inwards—away from that horn-shaped coastal region; rather, likely towards the region straddling southern Sudan, Ethiopia, Kenya and Uganda or even further—region straddling Uganda, Kenya, and Tanzania...
— PN2 clade (E3) bearers in the vicinity of the Sudanese-Central African Republic -Ugandan-Kenyan general region give rise to E3a ~ between 21 and 18 ky ago [pending additional or new info]; E3b-M35* would have likely arose relatively earlier than E3a* [as evidenced by its near absence in some the populations that carry this], sometime prior to the Ogolian and the LGM period. At this time, it was likely the M78 derivative that came about ~ between 19 and 15 ky ago. It was also likely during this period, that some E3b-M35 variants spilled over to the "southwest Asia", which would be identified as E-M34. The E-M78* likely arose somewhere in the bidirectional-migration route between Northeast and sub-Saharan East Africa; this location was likely in the region straddling upper Egypt and Sudan of the eastern Sahara, amongst earlier E-M35 migrants from sub-Saharan East Africa. These M78 bearers were increasingly pressured to move further south due to progressive aridity, possibly as far as Uganda-Kenya and/or Tanzanian general region...
...between 15ky and 13ky ago. During this period, as the Saharan aridity began to gradually slacken, some E-M78 bearing proto-Afrasan speaking nomads likely made their way into the Levant via the Sinai corridor.
Of course range of subsequent habitat of the early south-bound migrating E-M78 bearers, as hinted on above, would have been from the African Horn to the Tanzanian general region. Whereas during the intense periods of the Ogolian aridity when the aridity spread to as far as the horn-coastal areas of the African Horn, and due to which populations would have been discouraged to set up permanent settlements in those coastal areas, by around ~12ky ago and thereof, the aridity retracted back further north, i.e. to areas north of the said coastal areas during the wet phase of the Sahara, encouraging populations which had been living relatively inwards—i.e. away from the horn-shaped coastal region—to repopulate those horn-shaped coastal areas. During this period, all sorts of activities were going on in the Sahara—allowing populations which were taking refuge below the earlier arid regions of the Sahara to then spread northward. During this period, with the associated filling up of the river systems of the continent above the equator due to the Monsoon rains, tools like pottery became handy, and hence the development pottery spread across the Sahara. We see early examples dating back to 11 ky ago in west/central-Sahara [where Niger lies] and ~ 10 ky ago or so in eastern Sahara, and at about this period more or less, evidence of cattle domestication in the Sahara appears.
But then by c. 5ky ago, re-aridification of the Sahara takes hold again, although not as far south as it did during the intense aridification periods of the Ogolian aridity. During this period, settlements in northeast Africa hug the valleys adjoining the Nile River, resulting in increasing social differentiation and organization along the Nile Valley. As just noted, while the aridity didn't extend as far south as it did in the pre-Holocene Sahara, it appears that the *northern* section of the African Horn territory was to some extent eventually affected by the aridity of the Sahara, leading to semi-desert conditions in that region, which might have factored into the following:
...My current hypothesis is that during the late 3rd millennium BC, in response to a drying climate, people were on the move. Some settled on Dahlak island. The people who settled in al-Midamman **crossed the Red Sea and settled in the Tihama** where they found a window of opportunity for life as result of the **massive flooding that was emanating from the highlands**, from a landscape out of control. When checks and balances were put in place in the highlands, as part of the landscape stabilisation for which Yemen became synonymous, the people at the coast were forced to move on. Groups may have found their way into the Jawf, and the Hadramawt. They retained some of their specific lithic technology, but generally otherwise **became integrated** with the rest of the South Arabian populations. — Edward Keall
In any event, by c. 5 ky ago, movement of groups from east Africa onto south Arabia occurred, as evidenced by the following archaeological indicators:
c. 2400-1800 BC
The earliest cultural record from al-Midamman is an ephemeral presence defined by the surface recovery of stone projectile points and scrapers belonging to a Neolithic culture, say, from before 4000 BC. The first substantial and monumental phase of the site starts in the 3rd millennium BC. It involved the setting up of **giant stone markers**. Certain slender pillars were once set up with infants buried beneath them, yet without grave goods; an isolated stone marked the grave of an adult male. Hypothetically, these burials pre-date the setting up of giant stones, an act dated roughly to 2400-1800 BC by the cache of copper-alloy tools and a core of obsidian found buried beneath one of the megaliths.
All of the stone used had to be imported from at least 50 km away. A later phase of the activity involved recycling the stone. Yet there is no evidence that this was a destructive act. Rather, it appears to suggest reverence for the past….
…From al-Midamman there is one bull's head in relief from a pottery vessel; two human figurines in pottery; incense burners of pottery; and an example of alphabetic letters scratched into a pottery vessel. But pottery items are very rare within the corpus of finds, representing four out of 4000 recorded (and diagnostic) fragments. As for the pottery itself, it is far superior to anything from classic South Arabia. Though hand-built, it is well produced from good quality clay. It is often burnished and decorated with punctate designs that call to mind Fattovich's Afro-Arabian cultural complex theories regarding the punctate incised pottery from Kassala in the Gash delta of southeastern Sudan. And upper Nile-area specialists will no doubt think of so-called wavy-line punctate pottery associated with the C-Group people. Yet, the one striking absence, which cannot be overlooked, is that Kassala does not have the same kind of obsidian record as al-Midamman where there is a clearly definable assemblage of obsidian microliths. It arrives **fully developed** as a lithic tradition, and it does not evolve out of the Arabian bi-facial tradition. Numerous *antecedents* can be found in East Africa. Our expedition has also observed obsidian of exactly the same technological tradition on the island of Dahlak Kabir, offshore from the Eritrean mainland. — Edward Keall
The preceding Neolithic culture, according to the Edward Keall article in question, is associated with 'stone projectile points and scrapers' dating back to prior to the 4th millennium BC:
The earliest cultural record from al-Midamman is an ephemeral presence defined by the surface recovery of stone projectile points and scrapers belonging to a Neolithic culture, say, from before 4000 BC. The first substantial and monumental phase of the site starts in the 3rd millennium BC. — E J Keall
The article is *not* associating this Neolithic culture with the giant stone burial structures. Those "markers" or structures appear at a phase dated later, specifically to some time in the 3rd millennium BC. Hence, as noted earlier, this would indicate 'differentiation' of culture with time. It is this 3rd millennium phase wherein Keall draws archaeologically substantiated links, via the type of stone used, pottery, as well as obsidian microlithic assemblage. So the implication here, is that those "giant burial markers", were likely a tradition brought by East African immigrants some time in the 3rd millennium BC. This could not be more clearer in the following:
*The first substantial and monumental phase of the site starts in the 3rd millennium BC. It involved the setting up of giant stone markers.
*Certain slender pillars were once set up with infants buried beneath them, yet without grave goods; an isolated stone marked the grave of an adult male. Hypothetically, these burials **PRE-DATE** the setting up of giant stones, an act dated roughly to 2400-1800 BC by the cache of copper-alloy tools and a core of obsidian found buried beneath one of the megaliths. — E. Keall
Keall makes this estimation, presumably from the finding that the said burials lacked grave goods, whereas later burials—as is the case with those maintained with "recycled stones" or "re-used stone"—had been furnished with grave goods. The question then becomes, why would people at this phase of stone monument appearances pay tribute to burial sites of the past, which were very likely not marked as such in their original state? Well, that's where one can assume *cultural sympathy* to be the likely motivator, if not a link, between the folks of the earlier archaeological phase and the later one. So, while no specific mention of migration in the Neolithic period is made, it is suggested however, that "people on both sides of the Red Sea may have had a common ancestry, and their cultural expressions emerge from that common background."
Furthermore, to the extent that cultural *differentiation* occurs between the mentioned Neolithic phase and the 'monumental' phase, involving people from a common ancestry, Keall's mention of this might give one a clue:
It is hypothesised that this did not involve the **mass movement of people, with their cultural baggage complete**.
Further differentiation is noted even in 'monumental phase' of the site in question, as exemplified by the later 'recycling of stones', yet consistency in pottery design, and in many cases [though obviously, not in all], reappearances of obsidian tools, make it clear that there was socio-cultural continuity here, and hence:
The idea of different phases of the occupation has always been present in the eyes of the excavators. The idea of newcomers supplanting the old ways has always been a possibility. **The most recent work has demonstrated this to be untenable.** Finding only the same kind of pottery in both the domestic, the funerary and commemorative areas implies that the **SAME PEOPLE** were involved throughout the site's life. Yet clearly their cultural habits did evolve. — by Keall, E.
This sums up the present author's take on the article in question.
Contact across the Red Sea (between Arabia and Africa) in the 2nd millennium BC: circumstantial evidence from the archaeological site of al-Midamman, Tihama coast of Yemen, and Dahlak Kabir Island, Eritrea — by Edward J. Keall