User talk:Middayexpress/Archive 18

Rendille
Hi Middayexpress,

I just read your comment Rendille have mixed with and assimilated many Nilotes. What do you mean by this? Yes they have adapted certain cultural traits from Nilotes but they are still genetically overwhelmingly native Cushitic and not Nilo Saharan.

Here are their autosomal dna results from the Tishkoff et al. 2009 study. The first result includes non-African references the second is exclusively African thus the results may differ due to references. However, both results show them being overwhelmingly Cushitic.

I see no reason why they shouldn't be listed, please don't show pseudo scientific outdated articles claiming they are mixed. They are arguably much closer to Somalis geographically, genetically as well as ethno-linguistically than distant groups such as the Beja, Amhara and Tigray who are listed. Listing groups like Semitic Ethiopians completely contradicts your own argument about adapted cultures and languages.

This is the name of the study: The Genetic Structure and History of Africans and African Americans by Sarah Tishkoff

Here is a link, download the supplementary data to check the results: http://www.sciencemag.org/cgi/content/abstract/1172257

(Page 90 of supplementary data) Proportion of AACs from the Global Unsupervised Structure Run at K=14
 * 0.2% Fulani
 * 3.3% Nilo Saharan
 * 0.2% Oceania
 * 2.8% Chadic
 * 0.2% Khoesan
 * 4.2% Niger-Kordofanian
 * 0.5% East Asian
 * 0.2% Hadza
 * 0.6% Sandawe
 * 3.6% Indian
 * 74.3% Cushitic
 * 0.3% Pygmy
 * 8.9% European
 * 0.6% Native American

(Page 95) Proportion of AACs from the Africa Structure Run at K=14
 * 0.1% Mbugu
 * 4.3% Chadic
 * 75.8% Cushitic
 * 0.5% Eastern Bantu
 * 7.2% Nilo Saharan
 * 3.3% Saharan
 * 0.5% Fulani
 * 1.0% Non-Bantu NK
 * 0.4% Khoesan/Mbuti
 * 4.6% Niger Kordofanian
 * 0.4% Sandawe
 * 1.3% Central Sudanic
 * 0.2% Hadza
 * 0.4% Pygmy

Let me know what you think. Mazi99 (talk) 03:18, 20 September 2010 (UTC)

I also want to add that their genetic results didn't differ much from Ethiopian Oromos in that study. Especially the amounts of the Cushitic cluster which we are the most interested in.

I hope you reconsider your previous assumption about the Rendille.

Mazi99 (talk) 03:41, 20 September 2010 (UTC)


 * Thanks for the info. I have managed to track down a full copy of that Tishkoff study, and it turns out that she did not study Somalis or any other Cushitic or Semitic speaking group from the Horn except for the Beja and the Beta Israel. Her "Cushitic cluster" does not seem to be really representative; while it is generally made up of Cushitic-speaking groups from south eastern Africa that migrated to the area a while back from the Horn (she says that "these populations originated from southern Ethiopia and migrated into Kenya and Tanzania within the past ~4,000 years"), many of said groups have since mixed with neighboring Nilotic and/or Bantu groups. She also indicates that there is a large "Cushitic" component in the very Nilotic groups that the Rendille and other Cushitic-speaking groups from the area are known to have mixed with: "Many Nilo-Saharan speaking populations in East Africa, such as the Maasai, show multiple cluster assignments from the Nilo-Saharan (red) and Cushitic (dark purple) AACs, in accord with linguistic evidence of repeated Nilotic assimilation of Cushites over the past 3,000 years (31) and with the high frequency of a shared East African-specific mutation associated with lactose tolerance". If you look at row 2 on Figure 3 on page 10 of the study itself, it's also easy to see the very significant Nilotic/Bantu influence on the Rendille since the Rendille show an even lower 'blue' component than the Nilotic Samburu despite the fact that the Samburu got that blue component from assimilating Cushitic groups in the first place. If you look again at the last row, which appears to show the greatest degree of detail, almost half of the Rendille's genetic makeup appears to be from Bantu groups since it is the characteristic 'orange' color that seems to typify the Niger-Congo speaking populations in the study. Only about 35% of the remaining column seems to link with the Cushitic cluster, with the remainder belonging to the Nilotic red cluster.


 * When I wrote that the modern (not ancient) Rendille are less closely related to Somalis than are many other Cushitic-Semitic speaking groups from the Horn, I was referring to the Rendille's long-standing practice of intermarrying with Samburu Nilotes, in particular:


 * "Inter-tribal marriage occurs more frequently between the Rendille and the Samburu than between the Rendille and other tribes (Borana, Gabra, Somali and Sakuye). As a reflection of this complexity, the lineage groups include those which are said to derive from the Somali, Gabra, Borana, Sakuye and Samburu. With such diverse origins, the Rendille lineage groups are included in 15 clans, but among these only nine are included in the moiety system. These nine clans may be called the "Rendille proper" [Spencer 1973]. Of those six clans excluded from the moiety system, one is a mixed group of Rendille and Borana, and the remaining five are mixed Rendille-Samburu groups. Those clans which have come strongly under the socio-cultural influence of the Samburu are called "Ariaal Rendille"."


 * "The extensive immigration of Rendille men to the cattle economy is matched by an extensive marriage of Rendille girls to the Samburu, so that the Rendille are in a sense losing women as well. The extent to which this is not counteracted by an equivalent degree of marriage of Samburu girls to Rendille may be judged from Table 6."


 * "The Rendille and Samburu have had close ties for many generations. They intermarry and exchange cultural ideas even though they live in separate areas, speak different languages, and have very different life-styles. Marriage between the Rendille and Samburu is subject to differences in customs. Young Rendille men are allowed to marry only at certain times, but the Samburu are not limited. Thus Samburu men often choose Rendille wives when the Rendille men are not allowed to marry. Both the Samburu and Rendille practice polygamy, which means that a man can marry more than one wife."


 * This intermixture has also been noted in mtDNA studies, with the Rendille showing far fewer of the markers that are typical of Somalis and other Cushitic/Semitic Horn populations and a high frequency of markers that are typical of Nilotic and/or Bantu groups. By the same token, one can also see how the Rendille and other Cushitic-speaking groups in the south eastern region have influenced the neighboring Nilotic and Bantu groups' own gene pool by the presence of typical Cushitic markers in some of the latter populations (c.f. ).


 * Also note that, while Horn groups such as the Amhara and Tigray may today speak Semitic languages, many are actually of Agaw descent. This is why a) there is a Cushitic substratum in, among others, the Amharic, Harari, Tigrinya and Gurage languages, (i.e. the sign of Cushitic speakers that switched over to speaking a Semitic language but retained certain features of their old language), b) the E1b1b Y-DNA haplogroup is still the most common Y DNA amongst many of these peoples , and c) the Ethio-Semitic groups' mtDNA profiles so closely resemble those of the area's Cushitic groups . In short, irrespective of what Afro-Asiatic languages these Cushitic and Semitic speaking groups from the Horn may speak, they generally tend to biologically cluster closely together . Middayexpress (talk) 00:18, 21 September 2010 (UTC)

Sarah Tishkoff did indeed not have any Somali samples, the main reason is that the samples were actually taken during fieldwork in Africa and we all know Somalia is not the safest place to go with expensive lap equipment due to the civil unrest. However, she sampled Borana Oromos who are highly similar to Somalis and clustered with Somalis in previous studies such as Sanchez's yDNA study from 2005 on Somali Danish immigrants. The notion that the Cushitic cluster does not seem to be really representative is false, if genetics is able to cluster people of Northeast African descent than it truly is a Cushitic cluster. Many either speak Cushitic languages or have known ancient admixture from Cushitic groups. It doesn't randomly show up in ethnicities we know have nothing to do with Northeast Africa.

The Cushitic cluster is not found in significant numbers outside of East Africa (some Saharan people and Arabians posses it in minimal ammounts per the Tishkoff study), suggesting it being native to Northeast Africa. About the Maasai you are spot on, they are a mixture of multiple clusters due to their geographical proximity to Bantus, Nilotes and Cushites.

I could not find anything about Rendille being less 'blue' than Samburu. Figure 3 is about the ratio of genetic diversity not clustering or did you mean figure 3 on the regular page (STRUCTURE analysis of the global data) and not the supplementary data? If so what you said is not true:

The 'blue' what you are seeing is West Eurasian ancestry, the exact percentages being:

Rendille: 8.9% Samburu: 2.7%

Again per (Page 90 of supplementary data) Proportion of AACs from the Global Unsupervised Structure Run at K=14

It is clear that you confused some other group for the Rendille sample (likely the Mbugwe from Northern Tanzania), as Rendille hardly have any significant Bantu or Nilo Saharan admixture. The figures I gave you in my previous message are the results of the Rendille people.

You should pay some attention to figure S19 (on page 45 (supplementary data). Rendille remain clustered along with the Oromos and Beta Israel while distant groups such as the Southern Cushitic speaking Iraqw and all the Nilotic Kenyans cluster away at K set to 10 ancestral groups. This suggest Rendille, Oromos and Beta Israel share recent common ancestry. It is only more than likely that Somalis would have had similar results. Rendille and and Nilo Saharan Kenyans might have intermarried but the Tishkoff study shows this only had a minimal influence on Rendille overall genetic makeup. Modern Rendille (this study is only a mere year old) are still overwhelmingly Cushitic genetically.

About the mtDNA study, thank you for sharing. I examined the results of the Rendille and didn't find any non-typical results. Most of their mtDNA is similar to other Horn Africans (mostly being haplogroup L0, M1 and L3 which arose somewhere around what we now call Ethiopia).

By the same token, one can also see how the Rendille and other Cushitic-speaking groups in the south eastern region have influenced the neighboring Nilotic and Bantu groups' own gene pool by the presence of typical Cushitic markers in some of the latter populations (c.f. 

I find this quote to be inaccurate, Rendille are a modern ethnicity of recent Southern Somali origins as their language clearly suggests (classified as a Lowland East Cushitic Omo Tana language just like Somali is). Most of the Cushitic influences in Nilo Saharan and Bantu populations of Kenya is ancient and predates the modern ethnic groups of today. I am well aware of the Cushitic substratum in Habesha languages and that they are genetically quite similar to Cushitic people. However, they have obviously been influenced by Semitic cultures in the same way Rendille have been influenced by Nilotic cultures. Listing Habesha groups but not allowing Rendille people on the Somali ethnicity page is a contradiction in itself

Mazi99 (talk) 09:43, 21 September 2010 (UTC)

To sum it all up:

1) We know from genetic studies that Rendille have minimal non-Northeast African affinities. 2) The Rendille language is the closest language to Somali.

So who is pure anyway? There is no such thing as purity, this is an outdated pseudo scientific term.

Mazi99 (talk) 09:52, 21 September 2010 (UTC)


 * Borana Oromos (at least the ones in Kenya, which are the Borana that Tishkoff actually studied) are not highly similar to Somalis. They are pretty much in the same boat as the Rendille, El Molo, Iraqw, etc., as they too have mixed with neighboring Bantu/Nilotic peoples. And this as well is quite well-documented. From the Kenya Institute of Education :


 * "The Cushites interacted with Kenyan communities and this led to intermarriage between them (.e.g. Pokomo and Borana)."


 * One also cannot use the Tishkoff study to speculate on whether or not the Rendille are more related to Somalis than are other Cushitic or Semitic-speaking groups from the Horn since of course Somalis weren't even studied in it nor were most other Cushitic and Semitic-speaking Horn populations -- groups that are comparatively less mixed than the Bantu/Nilo-Saharan-influenced Iraqw, Rendille, Konso, El Molo, etc. that Tishkoff did actually study. As I've already pointed out, the only such Horn groups that were studied were the Beja and the Beta Israel, and the Rendille clustered with neither. Even the Cape Coloured group of South Africa and the Dogon of West Africa were closer (see Figure 1 on page 8 of the study proper for evidence of this), and this specifically has to do with the considerable level of the 'blue' component all of these groups possess. Refer to Figure 5A of Tishkoff's study (not the supplementary material), where the authors actually compare the southern Cushitic-speaking groups with the northern ones in the Horn and indicate a clear genetic differentiation between north and northeastern Africa on the one hand, and the areas inhabited by the Rendille, El Molo, and other similar Cushitic-speaking groups from more southerly areas of eastern Africa on the other. Here's how they explain it:


 * "We incorporated geographic data into a Bayesian clustering analysis, assuming no admixture (TESS software) (24) and distinguished six clusters within continental Africa (Fig. 5A). The most geographically widespread cluster (orange) extends from far Western Africa (the Mandinka) through central Africa to the Bantu speakers of South Africa (the Venda and Xhosa) and corresponds to the distribution of the Niger-Kordofanian language family; possibly reflecting the spread of Bantu-speaking populations from near the Nigerian/Cameroon highlands across eastern and southern Africa within the past 5,000-3,000 years (25, 26). Another inferred cluster includes the Pygmy and SAK populations (green), with a noncontiguous geographic distribution in central and southeastern Africa, consistent with the STRUCTURE analyses (Fig. 3). Another geographically contiguous cluster extends across Northern Africa (blue) into Mali (the Dogon), Ethiopia and northern Kenya. With the exception of the Dogon, these populations speak an Afroasiatic language. Chadic and Nilo-Saharan speaking populations from Nigeria, Cameroon and central Chad, as well as several Nilo-Saharan speaking populations from southern Sudan comprise another cluster (red). Nilo-Saharan and Cushitic speakers from the Sudan, Kenya and Tanzania, as well as some of the Bantu speakers from Kenya, Tanzania and Rwanda (Hutu/Tutsi) constitute another cluster (purple), reflecting linguistic evidence for gene flow amongst these populations over the past ~5,000 years (27, 28). Finally, the Hadza are the sole constituents of a sixth cluster (yellow) consistent with their distinctive genetic structure identified with PCA and STRUCTURE."


 * As the quotes I produced yesterday clearly show, the modern Rendille are in no way representative of 'typical' Cushites. They are a mixed Cushitic/Nilotic population, with the attendant mtDNA sequences to show for it. As can clearly be seen in Figure 2 of the Castri study, their most common mtDNA lineage is L0 (~23%), which is an archaic clade that is not at all common among Somalis. It is, however, extremely common amongst Nilotes. In fact, the highest frequency of L0 chromosomes in the study were observed in the Nilo-Saharan speaking Luo group. The maternal lineages the Rendille do actually share with Somalis (such as M1) are in the minority, not the majority. And this is specifically a consequence of their practice of marrying Nilotic women, as I've pointed out. Likewise, the notable frequencies of typical Cushitic maternal markers in the Samburu Nilotes in particular are due to their intermarriages and otherwise symbiotic relationship with the Rendille specifically (re-read that passage above about how the Rendille preferentially inter-marry with this particular Nilotic group over even other Cushitic-speaking populations from the area).


 * As I've already pointed out before, while the Amhara, Tigray, Gurage, etc. may today speak Semitic languages (which are Afro-Asiatic languages, like Somali and, incidentally, may have even originated in Ethiopia itself), those were, for the most part, only recently adopted, as most of said peoples are actually of Agaw descent. This is why their languages all have strong Cushitic substrata (i.e. they've retained features of their original Cushitic languages), and they are biologically very closely related to other Cushitic groups from the Horn such as the Somali:


 * "Blood samples from members of the Oromo and Amhara ethnic groups of central Ethiopia were tested for 10 erythrocyte protein systems: ACP1, ADA, AK1, CA2, ESD, G6PD, GLO1, HBβ, PGD, and PGM1. Differences between the two samples were relatively slight and not statistically significant. Gene frequency distributions were then analyzed in the context of the genetics of the African and Arabian peoples. Considering the erythrocyte enzyme data, the Oromo and Amhara appear quite similar to Europoids (particularly to the South Arabians) and considerably different from the Negritic peoples. There is evidence for close genetic affinity among the Cushitic- and Semitic-speaking population groups of the Horn. Admixture between Europoid and Negritic populations seems to have been the main microevolutionary factor in generating the present day Cushitic (and Semitic)-speaking group of eastern Africa. The results are consistent with the hypothesis, supported by historical and linguistic evidence, for a common origin of these groups from a Cushitic-speaking group living in eastern Africa."


 * The modern (not the ancient) Rendille, on the other hand, are not nearly as closely related to the modern Somali, as they have mixed with neighboring Nilotic/Bantu groups for generations, irrespective of the language they speak (the Rendille language is also not the closest language to Somali, btw; it is actually the most divergent of the "Sam" sub-group ):


 * "The Rendille and Samburu have had close ties for many generations. They intermarry and exchange cultural ideas even though they live in separate areas, speak different languages, and have very different life-styles. Marriage between the Rendille and Samburu is subject to differences in customs. Young Rendille men are allowed to marry only at certain times, but the Samburu are not limited. Thus Samburu men often choose Rendille wives when the Rendille men are not allowed to marry. Both the Samburu and Rendille practice polygamy, which means that a man can marry more than one wife." Middayexpress (talk) 00:25, 22 September 2010 (UTC)

Looking at colorful maps is meaningless, I suggest you to actually take the time to study the figures in the supplementary data. I wrote this earlier, you didn't quite get this part it seems:

Pay some attention to figure S19 (on page 45 (supplementary data). Rendille remain clustered along with the Oromos and Beta Israel while distant groups such as the Southern Cushitic speaking Iraqw and all the Nilotic Kenyans cluster away at K set to 10 ancestral groups.

Click here to directly see this image.

I am not completely denying that they have mixed with neighboring populations, so those quotes on intermarriages are not really necessary. I could play this little game as well and show many documented relationships between certain Southern Somali clans and Bantus (would that make them any less Somali anyway?), but I find this tactic to be pseudoscientific and I rather focus on facts (genetics).

All I am saying is that this so called 'Nilo Saharan admixture' has been largely overestimated by you as shown by the Tishkoff study where they showed minimal Nilo Saharan (7.2%) and Niger-Congo (4.6%) admixture. Your notion that Borenas and Rendilles are heavily mixed with non-Northeast Africans is disproved by the study. Somalis would have had similar results to Borenas, as we know from previous yDNA and mtDNA studies that Somalis have little to no West Eurasian admixture that could possible diverge them from this group. There are theories that suggest an East African origins of haplogroup M1, so don't bother suggesting that. []. The study you showed that Amhara and Oromo are related doesn't provide us anything new that the Tishkoff study didn't cover.

Ethnicities like the Beja have assimilated groups from the Arabian peninsula as is clear in their elaveted levels of haplogroup J1 (36% in Beja compared to only a mere 2.5% in Somalis). On top of that their inclusion in the Cushitic language family has even been questioned.. The Habesha despite being predominately Northeast African had their fair share of Semitic influences. mtDNA L0a is ubiquitous in the Horn of Africa. For instance, 10.7% of Ethiopian Oromos belong to this clade A localized founder effect could've increased the frequency of this haplogroup without the necessary admixture from Nilotic groups. Most of the Rendille mtDNA is quite similar to Somalis. Somalis are not completely M1, their majority haplogroup is L3, which originated in Northeastern Africa. Rendille mtDNA is not divergent from other Horn African populations.

About the linguistic data, the Rendille language is still closer to the Somali language than any of the currently listed ethnic groups on the Somali page. Proven by this image based on Figure S32 (page 69) from the Tishkoff study: showing linguistic relationships among populations and trees of language divergence constructed. It's evident that the Rendille language is linguistically close to Somali.

This simple fact should have just been enough to show shared ancestry between Rendille and Somali peoples. It's clear you do not want Rendille to be listed as they are too African for Somalis. I thought that showing you rational scientific data would convince you but apparently you still can't accept it. I contacted an admin about this issue and let him review it as our discussion is going in a vicious circle.

Mazi99 (talk) 21:42, 22 September 2010 (UTC)


 * I don't play "games" nor do I particularly need to since the quotes and studies plainly indicate exactly what I have stated. Namely, that the Rendille have mixed heavily with Nilotic groups, to the point where they now actually biologically cluster with other mixed Nilotic-Cushitic groups (like the Samburu) in their own separate cluster, not with northeast African groups (including Somalis). Remember, it's you that contacted me and brought up genetics and the Tishkoff study, so there's no point now in crying foul or getting upset when neither support what it is you are claiming. The fact is, Tishkoff's purple "Cushitic cluster" -- which you have repeatedly indicated the Rendille belong to -- does not even characterize Somalis or many other groups from the Horn. The author and her co-workers explicitly state that the cluster groups together Nilo-Saharan and Cushitic speaking groups from south eastern Africa and Sudan (not the Horn), and specifically as a consequence of admixture:


 * "Nilo-Saharan and Cushitic speakers from the Sudan, Kenya and Tanzania, as well as some of the Bantu speakers from Kenya, Tanzania and Rwanda (Hutu/Tutsi) constitute another cluster (purple), reflecting linguistic evidence for gene flow amongst these populations over the past ~5,000 years (27, 28)."


 * By contrast, she indicates that north and northeastern Africa (including the Somali-inhabited parts of Somalia, not the Bantu-inhabited ones) form a separate, distinct 'blue' cluster of their own:


 * "Another geographically contiguous cluster extends across Northern Africa (blue) into Mali (the Dogon), Ethiopia and northern Kenya. With the exception of the Dogon, these populations speak an Afroasiatic language."


 * This is what your own study quite plainly indicates in the Figure 5A (the only place it actually addresses Somali areas) that the text above is describing; see.


 * Your suggestion that the Beja are not particularly related to Somalis due to the fact that about 36% of them belong to Y haplogroup J1 is also somewhat misleading since according to that same study the majority (52%) appear to belong to the E1b1b or E3b haplogroup that is common in Somalis, and they too belong to the 'blue' areas in Tishkoff's genetic map. It is also misleading because Somalis themselves aren't all E1b1b members; only about 77% are, the remainder belong to mainly Eurasian Y clades per this study . Like the Beja, Semitic-speaking Ethiopian groups are mostly E1b1b bearers (reflecting their Agaw origin) and their mtDNA likewise closely matches those of their Cushitic speaking neighbors in the Horn . I've also only managed to track down one study which indicates that the frequency of E1b1b in Oromos exceeds 62% (this study's estimate, which is not that far off from the Beja's 52%). That was in a very small sample of only 7 Boranas , which of course is not at all representative of that entire population.


 * While it's true that Somali mtDNA is not all made up of M1, much of it is. It is also generally very different from the Rendille's mtDNA, the latter of which, as I've already pointed out, has as its most frequent lineage L0 (~23%), which is an archaic haplogroup that is common in Nilotes but not at all in Somalis. Similarly, according to Castri, the Rendille's second most frequent mtDNA haplogroup is the Eurasian haplogroup I, which is not that common in Somalis either. Per Harich et al. 2010, there is also a definite north-south cline that is observable in the Somali mtDNA, with the northern and central areas of Somalia showing a considerably lower frequency of all L haplogroups, including L3 (see this colorized version of the study's Figure 2b ). The mtDNA profile of northern Kenya, where the mixed Rendille and Kenyan Borana groups actually live, is completely different.


 * But all this genetic talk is ultimately academic and does not obscure the basic fact that the modern Rendille (which is what the 'related ethnic groups' parameter pertains to) are engaged in a symbiotic relationship with Nilotic groups (the Samburu, in particular); a relationship characterized by frequent, preferential intermarriages and culturual exchanges, irrespective of what language the ancestors of the Rendille -- like the Cushitic-speaking ancestors of the similarly Bantu and/or Nilotic-influenced Iraqw, El Molo, etc. -- or the Samburu may have spoken. They are both very mixed peoples now. Insisting otherwise is what is really misleading. Middayexpress (talk) 00:26, 23 September 2010 (UTC)

Again you are dodging this graph, which is an intra-East African structure run. The Rendille do not cluster with Samburus or other Nilotic Kenyans. Rather they cluster with Beta Israel and Oromos when the K is set to 10 inferred ancestral populations (this purple cluster is not same as the one used in earlier graphs but a new cluster in itself). Notice how El Molo and Samburu are made up of entirely different clusters compared to Rendille. Harich et al. 2010 did not study any Somalis, his maps are based on surrounding populations who have been tested (shown here ). He is assuming Northern Somalis have more Eurasian haplogroups merely based on geographical guesswork and not based on data gathered locally, which is meaningless in itself. Looking at mtDNA results of Ethiopian Cushites we can only assume Somalis are majority L3, since there aren't any modern studies on Somali mtDNA explicitly stating haplogroup frequencies.

I find this genetics talk a bit redundant, the Rendille language is closely related to the Somali language. This should be enough - end of story. If these are your final remarks I still see no reason why they should not be allowed, so does Gyrofrog.

Mazi99 (talk) 01:11, 23 September 2010 (UTC)


 * Language doesn't necessarily mean much, as populations mix and adopt new tongues all the time. Many modern Oromos, for example, are actually assimilated Omotic peoples, as Ethiopianist scholars among other have pointed out . The Rendille are a mixed people, as their mtDNA and many documented cases of intermarriages with Nilotes clearly show. And no, the Rendille do not cluster with the Beta Israel. That link you posted is a structure run, not a relational tree which actually shows clustering in unambiguous terms. When I say that even the Cape Coloured group of South Africa and the Dogon of West Africa were closer to the Beta Israel or the Beja, I mean that Figure 1 on page 8 of the study proper includes such a tree that explicitly shows this. In terms of mtDNA, Somalis are also not most commonly L3; they are mostly M1:


 * "We analysed mtDNA variation in ~250 persons from Libya, Somalia, and Congo/Zambia, as representatives of the three regions of interest. Our initial results indicate a sharp cline in M1 frequencies that generally does not extend into sub-Saharan Africa. While our North and especially East African samples contained frequencies of M1 over 20%, our sub-Saharan samples consisted almost entirely of the L1 or L2 haplogroups only. In addition, there existed a significant amount of homogeneity within the M1 haplogroup. This sharp cline indicates a history of little admixture between these regions. This could imply a more recent ancestry for M1 in Africa, as older lineages are more diverse and widespread by nature, and may be an indication of a back-migration into Africa from the Middle East." Middayexpress (talk) 01:24, 23 September 2010 (UTC)

That study doesn't state any frequencies. Only the fact that M1 is found in relatively high frequencies but not explicitly how much. Ethiopians are an indication how Somali mtDNA would look like:. Haplogroup L3 and M1 both arose in Eastern Africa so why are we even discussing this? M1 is essentially L3 (it just means L3m1).

Mazi99 (talk) 01:50, 23 September 2010 (UTC)

The study above indicates a frequency range of well over 20% of M1 in both the Libyan and Somali samples. If L3 were the most common clade in Somalis, then the authors wouldn't cluster them with the North African sample as they do nor would they emphasize M1 or indicate that there was "a history of little admixture between these regions". For the rest, the scholarly consensus is that M1 arose in Asia, and arrived in northern Africa via a back-migration. M1 is also not essentially L3. It is a relatively recent sub-clade of macrohaplogruop M, which appears to be the most common mtDNA lineage in Asia. It directly descends from South Asian lineages, not Ethiopian ones. That link you produced also leads to an original user-made map, not an official one. The actual official mtDNA data (i.e. this study ) generally indicates that the most common mtDNA clade in most of the studied Cushitic and Semitic speaking groups alike in the Horn is M1, not L3. Middayexpress (talk) 02:20, 23 September 2010 (UTC)

An Asian origin of M1 is not conclusive. M1 is a sub-clade of L3 (I'm surprised you didn't know this). It isn't a 'recent' M haplogroup either as it is estimated to be over ~50,000 years old. Over 20% M1 means nothing, why couldn't they just stated the frequencies like all other studies do? It leaves room for a lot of L3.

You are hilarious, your own sources consistantly contradict yourself. M1 vs L3 frequencies per

Cushitic Speakers M1=14% L3=24% Amhara	M1=16%	L3=20% Gurages	M1=14%	L3=15% Tigrais	M1=17%	L3=12% Ethiopians M1=17% L3=16.4% (only a mere 0.6% difference)

Frequencies vary per study and place were the samples were gathered, but a general consensus is that both L3 and M1 (which IS a L3 sub-clade) make up most of the mtDNA in the Horn.

Mazi99 (talk) 02:51, 23 September 2010 (UTC)


 * Nah; the modern scholarly consensus is that M1 originated in Asia (ex. ). And this specifically has to do with the fact that M1 is younger or more downstream (that is what I mean by 'recent') than the many Asian sub-clades of its parent macrohaplogroup M, the latter of which is almost exclusively found in that continent. You can also rationalize the Holden study and its actual statements on M1 however you like, but it too unfortunately won't change what the authors have actually indicated. Not that this really matters either, but it's actually typically your "own sources" (your misreading of Tishkoff, in particular) that have contradicted you. This one time, though, I will concede that I did not notice that many of the listed L sub-clades in that study belonged to L3. Oops. Only >50% of all the studied Ethiopian samples (including the Cushitic speakers) are made up of Eurasian mtDNA clades . My bad. I'll also ask you to stop writing me. The Rendille thing is settled, so we have nothing further to discuss. I'm also not particularly interested in idle chat. If I had been, I would've indulged those personal questions you asked me earlier, but of course I didn't. Middayexpress (talk) 03:17, 23 September 2010 (UTC)

It's evident that you are a novice when it comes down to population genetics. Thank you for the wonderful discussion on the Rendille, this is my final message to you. Cheers!

Mazi99 (talk) 03:08, 23 September 2010 (UTC)


 * Whatever you say man. Middayexpress (talk) 03:41, 23 September 2010 (UTC)

Recent edits
Hi, I've noticed an unfortunate trend of edits like this. Could you please stop these edits. They're disrupting the flow of the article completely. It's just sentence after sentence in a really robotic fashion, rather than good writing, where it flows. The rest of the changes are good; it's just that these types of edits are harming the readability. Thank you.  Paralympiakos  (talk) 02:14, 21 September 2010 (UTC)
 * That was a run-on sentence. But your point is taken. Middayexpress (talk) 02:15, 21 September 2010 (UTC)
 * Cheers.  Paralympiakos  (talk) 02:17, 21 September 2010 (UTC)