This lineage would be about 3000 years old:
http://www.anthrogenica.com/showthread.php?9370-Where-did-haplogroup-T-M184-originate
"Smh Traders/Merchants never had a significant impact on Somalis.Majority of these Somali clans that have this haplogroup are also inland pastoralists with no recent foreign admixture lol.Also this haplogroup is also found among other cushite groups like the Iraqw of Tanzania.
It makes much more sense that this haplogroup entered Africa from the Levant maybe during the neolithic and was spread southward by migrating Cushites."
Note that the Khoekhoe carried both Levantine and Cushitic DNA and Levantine sheep all the way to South Africa..
Grant
Most of what is being discussed in online forums about the origin of Somali haplogroup T is conjecture. As I have previously stated, the subclade/subclades that Somalis belong to needs to be discovered before we can seriously discuss whether T arrived from the Levant or from Arabia via the Red Sea.
In the Anthrogenica thread, Vettor raises a valid point in regard to the subclade/subclades that Somalis belong to. If the Somali haplogroup T subclade is Y16897, then it is possible that their ancestor came to Africa via the Red Sea route because this subclade is found among Gulf Arabs and Ashkenazi Jews as the map posted illustrates, and as confirmed by private test results on FTDNA. As things stand, the CTS2214 subclade appears to be the most common haplogroup T1a1a subclade found in Africa but Y16897 is as rare as a hen's teeth, hence why Vettor wants Somali T carriers to do the Big Y test so he may determine the population that they are closely related to. Check the following STR Haplogroup T DNA test results and you will observe that the Habar Yoonis Somali haplogroup T carriers are grouped with Y16897/Z19971:
https://www.familytreedna.com/public/Y-Haplogroup-K2/default.aspx?section=yresults
I should add that belonging to Y16897 does not necessarily guarantee that Somalis HG-T carriers migrated to the Horn via the Red Sea because as more populations get tested, it might be discovered among other African populations in the North or Southern part of the continent. For example, it was recently discovered that the Tobou of Chad possess haplogroup T and R in large numbers. Could their HG-T possibly be Y16897 too?
http://www.cell.com/cms/attachment/2073 ... 8/mmc2.pdf
Nevertheless, whatever migratory route the ancestor/ancestors of Somali haplogroup T carrier took does not change the probability it has an old presence among Somalis. It is most likely the result of a founder effect in the Somali population that occurred 2k to 2.5k ago as hypothesised by that Danish study. However, the T samples they used might not be representative of all Somali HG-T carriers and we do not know the clan breakdown of the people tested.
As for what Qabils are haplogroup T, the jury is still out but there are indicators that it is high among the Northern Ciise, Isaaq and Samaroon. It is also high among the South Central Surre Dir subclan. Among the Isaaq, only the dominant haplogroup in only two subclans can be hypothesised, the Sacad Muuse (E-V32) and the Habar Yoonis (T). Bare in mind that it is important to tests sub sub sub clans within all clans to obtain accurate data. That has not been done yet.