Grant
You are quite to quick to make sweeping statements about Somali y-dna when the scientific evidence is limited to say the least.
For example, you keep throwing around the idea that Somalis are 77% E-V32 carriers when it has not been scientifically established.  One of the earliest studies (Sanchez et al) on Somali y-dna which is often quoted by individuals such as yourself to support their erroneous claim did not use a representatiave sample:
The very high frequency of the E3b1 cluster 
	
	
	
		
		
		
		
	
	
 in our Somali population sample could be due to ascertainment bias or special clan or family relationships in the present sample of Somali immigrants to Denmark. No reliable information on geographic origin or clan relationship in the present Somali population sample was available. 
http://www.nature.com/ejhg/journal/v13/n7/full/5201390a.html
No study has also been conducted into Dir subclans that reside in Somalia proper, NFD and the Southern regions of K5 so it is too soon to assume that haplogroup T is the definitive marker of the Dir clan.  The evidence from private test results and the Djiboutian/Dire Dawa studies suggest that HG-T might be the dominant Dir ancestral haplogroup but wihout further research, you are clutching at straws.  Furthermore, how representative of the Ciise and the Samaroon subclans are the Djiboutian Dir samples?
The same goes for other Somali clans.  You state that Hawiye and other Samaales are predominately E-V32 when there is no academic study that backs up your statement.  Only one academic study the tested Tuuf Garres from Marsabit County can be referenced to support an argument that some Samaales are E-V32.  Private test results on 23andme partially corroborate your argument as several Habargidirs, Abgaals and a Gaaljecel Samaale are E-V32.  However, no study has been conducted among Somaliland Isaaqs and your claim that they are predominately E-V32 has no supporting academic evidence.  Only 9 individuals were tested and the Isaaqs that have a historical presence in Djibouti are the Habar Awal Sacad Muuse. They are predominately E-V32 based on private test results.  In contrast, more than ten Garxajis clan members that have been tested privately on 23andme and FTDNA are haplogroup T.  Both sets of private test results and the study you have referenced are not conclusive because they are not necessarily representative of the subclans of the above clans.
You also seem to ignore the fact that E-V32 has been stimated to be around 7000 years old.  We do not know how closely the different E-V32 Somalis are paternally related until y-full tests are conducted but they are too expensive for academic researchers to conduct them on a widescale.  Only then will we be able to consider confirming or rejecting your theory that non T haplogroup Somalis split from Oromos in Southern Ethiopia.  Some might be related to Northern Beja, Tigre E-V32 carriers whilst others might be related to Southern Ethiopian Oromo, Sidamo E-V32. Lord Only Knows Mr Grant.
In regard to Samaale, have you ever considered it is a confederacy united by a common Xeer system and not a genetically uniform ethnic group?  Dir have a historical presence in Northern Somalia and they are by Somali tradition considered to be the prototype Samaale/Somalis.  If as it seems that they are unique in being predominately haplogroup T, it still does not support of dismiss your South to North migration theory because it has not yet been etsbalished whether the Haplogroup T subclade that they belong to is Middle Eastern or African in origin.  Even if it turns out to be Middle Eastern in origin, we are not talking about a recent migratory event such as the Arab/Iranian Haplogroup T presence in coastal Southern Somalia.
		
 
Anonimo,
I don't think I said much about V-32. We are agreed that studies out of context have little value. This study includes but does not rely on the Denmark bit that you quoted:
https://en.wikipedia.org/wiki/Haplogroup_T-M184
 
Population Language Location Members/Sample size Percentage Source 
Notes
Dir (clan) Issa/
Gadabuursi Somali (East 
Cushitic) 
Djibouti 24/24 100% 
[17] The main sub-clans of the 
Dir (clan) in Djibouti are the 
Issa and 
Gadabuursi. Also, T1a-M70 has been found in only 1 sample belonging to a member of the 
Hawiye clan (1/1) and in 0/9 (0%) samples belonging to the 
Isaaq clan.
Northern Dir tribes Northern Somali (East 
Cushitic) 
Dire Dawa 14/17 82.4% 
[16] Dir sub-clans of Dire Dawa are 
Issa, 
Gurgura and 
Gadabuursi.
Somalis (Djibouti) 
Somali (East 
Cushitic) 
Djibouti 30/40 75% 
[196] This sample is only 25% E-M215. Clan not specified.
Anteony Antemoro (Plateau Malagasy) 
old Antemoro Kingdom 22/37 59.5% 
[197] The Anteony are the descendants of aristocrats, from whom the Antemoro king is chosen. Can be grouped into the Silamo, because they have the right to undertake the ritual slaughter of animals (Sombily)
Northern Dir tribes and 
Afars Northern Somali and 
Afar(
Cushitic) 
Djibouti 30/54 56.6% 
[198] Mixed sample of Somali and Afar individuals.
Somalis (Ethiopia) 
Somali (East 
Cushitic) 
Shilavo (woreda) (
Ogaden) 5/10 50% 
[17] The geographic location of this Ethiopia sample as seen in Fig.1.
Aushi Aushi Zambia 1/2 50% 
[199] 
Toubou Toubou Chad 
31% 
[200] All belonging to the T1a-PF5662 subclade
Afars Afar language (East 
Cushitic) 
Djibouti 5/20 25% 
[17] 
Akie Akie people (
Nilotic) 
Tanzania 3/13 23.1% [Hirbo et al.] Akie people have remnants of a Cushitic language
Somalis Somali (East 
Cushitic) 
Jijiga (
Ogaden) 19/83 22.9% 
[16] Jijiga Somalis.
Arabs from 
Somalia Somali (East 
Cushitic) immigrants in Yemen 7/33 21.2% 
[201] 
Lemba Venda and 
Shona (
Bantu) 
South Africa 6/34 17.6% 
[2] Exclusively belong to T1a2* (old T1b*). Possible recent founder effect. Low frequency of T1a2 has been observed in Bulgarian Jews and Turks but is not found in other Jewish communities. Y-str Haplotypes close to some T1a2 Armenians.
Rangi Rangi Language (
Bantu) 
Tanzania 5/32 15.6% [Hirbo et al.] 
Multiple ethnicity 
- Somalia 15/105 14.3% 
[202][203] 
Iraqw Iraqw language (
Cushitic) 
Tanzania 6/47 12.8% [Hirbo et al.] 
Wachagga Kichagga (
Niger-Congo) 
Dār as-Salām 3/24 12.5% 
[164] Mixed with Rift Southern Cushites.
Somali Somali (
Cushitic) immigrants to 
Norway 12/104 11.5% 
[204] 
Bench Bench(
northern Omotic) 
Bench Maji Zone 14/126 11.4% 
[16] 
Kores (
Cushitic) 
SNNP 2/18 11.1% 
[16] 
Oromo Afaan Oromo language (
Cushitic) 
Oromiyaa 1/9 11.1% 
[205] 
Fulbe Fula northern 
Cameroon 3/27 11.1% 
[206][207] 
Gorowa Gorowa language (
Cushitic) 
Tanzania 2/19 10.5% [Hirbo et al.] 
Somali Somali (
Cushitic) immigrants to 
Denmark 21/201 10.4% 
[208][7] 
Upper Egyptians Egyptian Arabic Luxor Governorate 3/29 10.3% 
[23][209] 
Kontas Konta language (
Omotic) 
Konta special woreda 11/107 10.3% 
[16] 
Rendille Rendille language (
Cushitic) 
Marsabit County 3/31 9.7% [Hirbo et al.] 
Datogs Rendille language (
Cushitic) 
Tanzania 3/31 9.7% 
[210] 
Gewadas Gewada language (east 
Cushitic) 
SNNP 11/116 9.5% 
[16] 
Antalaotra Antemoro (Plateau Malagasy) 
old Antemoro Kingdom 4/43 9.3% 
[197] The Antalaotra are in charge of the magical and religious domains; they have the ability to read and write Sorabe. Can be grouped into the Silamo, because they have the right to undertake the ritual slaughter of animals (Sombily)
Upper Egyptians Egyptian Arabic Aswan Governorate 1/11 9.1% 
[211] 
Subiya Subiya/Kuhane (
Bantu) 
Zambia 1/11 9% 
[199] 
Upper Egyptians Egyptian Arabic Assiut Governorate 6/70 8.6% 
[211] 
Konsos (
Semitic) 
Konso special woreda 2/24 8.3% 
[16] 
Somali Somali (
Cushitic) immigrants to 
Sweden 12/147 8.2% 
[212] 
Arabs and Berbers 
Egyptian Arabic and 
Siwi Lower Egypt 12/147 8.2% 
[18] 
Upper Egyptians Egyptian Arabic Sohag Governorate 4/52 7.7% 
[211] 
Egyptians Erythraic (
Cushitic) 
Egypt 7/92 7.6% 
[203][205] If the K* sample is M184+ then 8.7%
Oromo (
Semitic) 
SNNP 2/28 7.1% 
[199] 
Tigrayans Tigrinya (
South Semitic) 
SNNP 2/30 6.7% 
[16] 
Dirashas Dirasha (
east Cushitic) 
Dirashe special woreda 5/79 6.3% 
[16] 
Canarians Canarian Spanish Tenerife 11/178 6.2% 
Omo Valley Omotic languages Ethiopia 6/98 6.1% 
[199] 
Kordofanians Kordofanian Kurdufan 4/69 5.8% 
[193] 
Upper Egyptians Egyptian Arabic Qena Governorate 3/52 5.8% 
[211] 
Tuareg Tuareg (
Berber) 
Gorom-Gorom 1/18 5.6% 
[213] 
Afars Afar (East 
Cushitic) 
Afar Region 6/111 5.4% 
[16] 
Ethiopians Ethiopian languages Ethiopia 4/74 5.4% 
[173] 
Mashiles Mashile language (
Cushitic) 
SNNP 7/130 5.4% 
[16] 
Gurages Gurage languages (
South Semitic) 
SNNP 6/118 5.1% 
[16] 
Canarians Canarian Spanish Gran Canaria 4/78 5.1% 
[199] 
Oromo Afaan Oromo language (
Cushitic) 
Oromiyaa 4/78 5.1% 
[199] 
Oromo Afaan Oromo language (
Cushitic) 
Adis Abeba 2/40 5% 
[199] 
Turu Nyaturu (
Bantu) 
Tanzania 1/20 5% 
[210] 
Moroccan Jews Haketia (
Romance) 
Israel 1/20 5% 
[214] 
Gedeos Gedeo (east 
Cushitic) 
SNNP 6/122 4.9% 
[16] 
Wairak Iraqw (
Cushitic) 
Tanzania 2/41 4.9% 
[18] 
Western 
Libyans Libyan Arabic (
Semitic) 
Tripoli region 7/142 4.9% 
[215]
[216]
Tunisians Tunisian Arabic (
Semitic) 
Sfax 5/105 4.8% 
[217] 
Libyans Libyan Arabic (
Semitic) 
Tripoli area 3/63 4.8% 
[218] 
Kanuri Kanuri Cameroon 1/21 4.8% [Hirbo et al.] 
Iraqw[219] Iraqw (
Cushitic) 
Tanzania 2/43 4.7% 
Yems Yemsa (
Omotic) 
SNNP 5/107 4.7% 
[16] 
Jews (
Semitic) 
Ethiopia 1/22 4.5% 
[17] 
-----------------------------------------------------------------------------------------------------------
You will notice multiple Somali studies, along with sample sizes and locations. The 77-85% in Somalis is E1b1b1a1 (M78), not V-32, which is a specific subclade. The clans are clearly confederacies, with the Isaaq  even split between T and E.
There appear to have been multiple admixture events bringing T from the north, but the most recent work places one at 3000 YA, associated with the beginnings of Semitic languages in the Horn.
http://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1004393
https://en.wikipedia.org/wiki/Somalis
"According to 
Y chromosome studies by Sanchez et al. (2005), Cruciani et al. (2004, 2007), the Somalis are paternally closely related to other 
Afro-Asiatic-speaking groups in 
Northeast Africa.
[167][168][169] Besides comprising the majority of the Y-DNA in Somalis, the 
E1b1b (formerly E3b) 
haplogroup also makes up a significant proportion of the paternal DNA of 
Ethiopians, 
Sudanese, 
Egyptians, 
Berbers, 
North African Arabs, as well as many 
Mediterranean populations.
[168][170] Sanchez et al. (2005) observed the M78 (E1b1b1a1) subclade of E1b1b in about 77.6% of their Somali male samples.
[167] According to Cruciani et al. (2007), the presence of this subhaplogroup in the Horn region may represent the traces of an ancient migration from 
Egypt/
Libya.
[169]
After haplogroup E1b1b, the second most frequently occurring 
Y-DNA haplogroup among Somalis is the West Asian 
haplogroup T (M184).
[171] The clade is observed in more than 10% of Somali males generally,
[167] with a frequency peak of 82.4% among Somalis in 
Dire Dawa.
[172] Haplogroup T, like haplogroup E1b1b, is also typically found among other populations of Northeast Africa, the 
Maghreb, the 
Near East and the Mediterranean.
[173][174]"
--------------------------------------------------------------------------------------------------------------
The background T in the large number of southern and non-Somali groups is typically around or less that 10% and probably does represent an ancient migration or influx from coastal merchants.  But scores in the 82-100% range would seem to be something else. I think it is time to look a lot more closely at T, which is not likely to have been part of the migration from Egypt/Libya at these high levels, and is clearly concentrated in the area of Aksumite/Himyarite influence in the North.
IMO it is not at all too early to start looking at this stuff. New research areas are already indicated.