Last weekend I mentioned a paper, The Genetic Structure and History of Africans and African Americans, which had the best coverage of disparate African populations we’ve seen so far. The map to the left shows the various ancestral population clusters inferred from the samples they had. Really the only failing is that they didn’t have samples from Angola, Zambia, Zimbabwe and Mozambique. Unfortunately, that’s not totally trivial. These are regions which were effected by the Bantu Expansion, with southern Angola in particular still having remnants of Khoisan language speakers which likely attest to the pre-Bantu populations. Luckily for us innovation and scientific ingenuity are such that minor questions can quickly be answered because of how cheap the basic methods have become. A new paper in The European Journal of Human Genetics tackles Mozambique in particular, and discerns a heretofore unknown possible population cluster. A genomic analysis identifies a novel component in the genetic structure of sub-Saharan African populations:
Studies of large sets of single nucleotide polymorphism (SNP) data have proven to be a powerful tool in the analysis of the genetic structure of human populations. In this work, we analyze genotyping data for 2841 SNPs in 12 sub-Saharan African populations, including a previously unsampled region of southeastern Africa (Mozambique). We show that robust results in a world-wide perspective can be obtained when analyzing only 1000 SNPs. Our main results both confirm the results of previous studies, and show new and interesting features in sub-Saharan African genetic complexity. There is a strong differentiation of Nilo-Saharans, much beyond what would be expected by geography. Hunter-gatherer populations (Khoisan and Pygmies) show a clear distinctiveness with very intrinsic Pygmy (and not only Khoisan) genetic features. Populations of the West Africa present an unexpected similarity among them, possibly the result of a population expansion. Finally, we find a strong differentiation of the southeastern Bantu population from Mozambique, which suggests an assimilation of a pre-Bantu substrate by Bantu speakers in the region.
The main value-add of the research were the 279 individuals from Mozambique, who they plugged into previous data sets (e.g., HGDP, HapMap3). It must also be noted that they limited their genetic survey to ~2800 SNPs.This is sufficient for their purposes. Below are the figures of interest from the paper. Note immediately how Mozambique separates out at K = 4 in the first image. The subsequent figures are from PCA. The axes represent components of variation. The last panel shows a PCA plot transposed onto a map. In this case, PC 1 & PC3.t
The first figure is important because it suggests population structure we hadn’t known of in the Bantu Expansion. This doesn’t mean that it should be surprising. With Africa’s current level of genetic variation it seems implausible that the carriers of the Bantu culture would not have assimilated other groups along the wave of advance. In fact, as a cultural movement gains steam through positive feedback loops different societies may become co-opted into them, and spread the culture in their own turn. As an American example, I will give the Irish American Catholic hierarchy’s campaigns against German language parochial school instruction in the 19th century. Old English aside the language of the Irish was originally not English, but by the early 19th century apparently English had already become dominant among the Roman Catholic peasantry of Ireland. So they brought English, not Gaelic, to the United States. Similarly, the spread of Islam in India occurred predominantly under the ageis of Turks and Afghans, not Arabs, while the spread of Islam in Southeast Asia was promoted by South Asian Muslim merchants in their turn. So you have Arab cultural forms in eastern Indonesia thanks to cultural expansions at two removes from the original Arab source (in fact, it could be argued that the Turks and Afghans were Islamicized through a Persian intermediate as well).
But it is the PCA plots which are of more curiosity for me. They note that it is the third component of variation which maps well onto geographic distance. In the paper they say:
This is the PC that is mostly correlated with geography…and the fact that it is the third rather than the first component, as would be expected if isolation by distance was the predominant force shaping genetic diversity…implies that directional population movements (such as the Bantu expansion) and barriers to gene flow (such as that between food producers and hunter gatherers) are more relevant than geographic distance to understand the genetic landscape of sub-Saharan Africa….
There were folk migrations in Africa. They might simply not have been the ones we are aware of, at least in our sparest conceptions. Those folk migrations were very recent, within the last ~2,000 years or so. Which is why the distinctive correlations between language and genes persist, especially on the outer edge of the wave of advance in southern Africa (in contrast, the Pygmies of the Congo have lost their native language, and the western Pygmies are highly admixed with their neighbors).
Addendum: The life of Shaka may give us a clue as the disturbances which pushed the Bantu ever outward.
Razib Khan’s degrees are in biochemistry and biology. He has blogged about genetics since 2002, previously worked in software development, is an Unz Foundation Junior Fellow and lives in the western US. He loves habaneros.
August 29th, 2010 at 2:51 am
[...] This post was mentioned on Twitter by razib khan, Al Poe, Maggie, World Amazing Things, Sains & Teknologi and others. Sains & Teknologi said: People, not pots, in Africa | Gene Expression: Last weekend I mentioned a paper, The Genetic Structure and History… http://bit.ly/a2XM9U [...]
August 29th, 2010 at 4:32 am
in fact, it could be argued that the Turks and Afghans were Islamicized through a Persian intermediate as well
That is something certain. Turkic and Afghan Islam are direct descendants of the Khorasani variety of the Persian Islam of the early centuries of Islam. Hanafism was then the dominant madhhab of Khorasani Persians (like most of Persians then), and directly from them Islam and Hanafism spread eastwards to western Central Asian Turkic peoples (including Oghuz, Karluks and some of Kipchaks), Afghans and some north Indo-Pakistani peoples. Turkic, Afghan and Subcontinental languages have many New Persian (Islamic-era Persian) loan words and the overwhelming majority of the Arabic loan words in these languages show clear New Persian influence (thus they were borrowed through New Persian), their religious terminologies are descended from the Persian Islamic terminology and their Arabic scripts are descended from the Perso-Arabic script.
Western Iranic peoples like Kurds, Zazas and Lurs also have Persian (not necessarily the Khorasani variety) Islamic culture instead of Arabic. In fact, all Iranic, Turkic, Caucasian, Subcontinental, Balkan and Chinese Muslims have Persian Islamic culture instead of Arabic. Even Iranian and Iraqi Arabs have strong Persian Islamic cultural influence. I don’t know the Islamization of Southeast Asia much, so I can’t say anything about their Islamic culture.
August 29th, 2010 at 7:43 am
Here’re a couple of peculiarities in this paper. First, Nilo-Saharans (represented by the Maasai) appear to be the most divergent population in Africa (“The first PC (Figure 2a) and STRUCTURE with K=2 (Figure 3) separate the Nilo-Saharan-speaking Maasai from all other populations…” and “by all other populations” the authors mean both Niger-Congo, including Pygmies, and the San). This is a good example of how language affiliation mirrors genetic distance. But then all other genetic studies claim that the San are the most divergent. Then, their global PCA in Fig. 1 show East Asians (rather than American Indians) as the most divergent from Africans. This is unlike any other study, I’m aware of.
August 29th, 2010 at 12:19 pm
In fact, all Iranic, Turkic, Caucasian, Subcontinental, Balkan and Chinese Muslims have Persian Islamic culture instead of Arabic. Even Iranian and Iraqi Arabs have strong Persian Islamic cultural influence. I don’t know the Islamization of Southeast Asia much, so I can’t say anything about their Islamic culture.
in general i agree with this (even if someone might quibble with the term ‘persian islamic culture’). but, i will enter in one exception: the muslims of kerala in south india are part of the south arabian islamic culture. unlike sunnis in the rest of the subcontinent they are not hanafi but shafi. southeast asians are shafi too btw, and generally they prefer to claim arab antecedents for their religion. though as i said, there tends to be an assumption that the eastern half of the indian ocean trade in the bay of bengali to indonesia was run by south asian muslims, not arabs. though they may have been shafi muslims from south india who were therefore culturally connected with south arabia, and not central asia.
August 29th, 2010 at 12:26 pm
german, interesting points.
August 29th, 2010 at 1:44 pm
Reading through Wikipedia, it seems that the Islamization of Southeast Asia is more of a mystery than the Islamization of most regions of the Muslim world. Their dominant madhhab (Shafi’ism) only increases the confusion.
August 29th, 2010 at 1:56 pm
re #6, the initial period, somewhat. the later period, when the islamic sultanates defeated majapahit, less so. i think part of the issue is that the indian ocean trade network is the critical nexus. it looks likely that indian merchants brought buddhism and hinduism to southeast asia. when those indian merchants became muslim, they brought islam. but, the nature of long distance trade during this period is somewhat murky because navigation and destination was somewhat a ‘trade secret.’ there were persians and arabs involved in the western indian ocean though, as well as jews. in fact, the portuguese found italian traders in malacca and its environs when they first arrived. it was clear that the italians had simply hidden the extent of their contacts or the length of their lines from other europeans so as to monopolize the flow of luxury goods (apparently the italian colony was decimated immediately by the enraged portuguese).
August 29th, 2010 at 2:15 pm
I understand, but how many Shafi’is were there in India* (there aren’t that many even today, though things may have been different in the past) to have such a big religious impact on SE Asia?
* I am including Sri Lanka and the other nearby islands in India.
August 29th, 2010 at 2:22 pm
onur, the turks and afghans had no impact on SE Asia. they were land powers. the indians who had an impact would be south indian or gujarati merchants. the europeans who came to india in the medieval and early modern period note that the international merchants of kerala were all muslim. the balance of hanafi and shafi are then irrelevant to this question, because those indians who entered into commerce in southeast asia were not a random distribution (in any case, the demographic prominence of north indian islam probably dates to the mughal period anyway, after the islamicization of southeast asia).
also, know gujarati muslims used to be much more religiously diverse than today. they were homogenized forcibly by aurangzeb in the 17th century.
August 29th, 2010 at 2:45 pm
Razib, I am already not expecting the Turkic and Afghan rulers or any other non-Shafi’i Muslim rulers or communities of the Subcontinent to have had any significant religious influence on SE Asia given the fact that SE Asian Muslims are predominantly Shafi’i while the Turkic and Afghan rulers of the Subcontinent were Hanafi. In my previous post I was just contrasting the fact that the overwhelming majority of the population of India, Sri Lanka and the other nearby islands were and still are Hindu and the fact that it was Islam which would ultimately come into dominance in a large part of SE Asia.
August 29th, 2010 at 2:51 pm
onur, sri lanka is overwhelming buddhist, not hindu. second, the musilms of sri lanka (who call themselves “moors”), are ~10% of the population, are shafi. and, they are concentrated on the eastern coastal regions facing sumatra. third, maldives is muslim, and was converted from buddhism, and is shafi. fourth, the coastal trading communities of western india which are muslim (e.g., mapilla and ismailis) are far less uniformily hanafi than the muslims of the gangetic plain or pakistan (those of kerala are shafi). fifth, european observers noted (often with anger) that indian muslims dominated international maritime trade in south asia by 1500 (this was noted by cheng ho’s fleet in the early 15th century too).
i think with those facts i’ve entered into the record the confusion should be mitigated some.
August 29th, 2010 at 3:21 pm
sri lanka is overwhelming buddhist, not hindu
Sorry, I accidentally wrote “Hindu” while I should have written “non-Muslim”.
the musilms of sri lanka (who call themselves “moors”), are ~10% of the population, are shafi
I know, but they are in the minority in the total population.
maldives is muslim
I know, but the Maldives is negligible because of its low population.
the coastal trading communities of western india which are muslim (e.g., mapilla and ismailis) are far less uniformily hanafi than the muslims of the gangetic plain or pakistan (those of kerala are shafi)
I know them too, but, again, they are in the minority in the predominantly non-Muslim (overwhelmingly Hindu) coastal India.
european observers noted (often with anger) that indian muslims dominated international maritime trade in south asia by 1500 (this was noted by cheng ho’s fleet in the early 15th century too)
This I didn’t know, though still it doesn’t change the fact that there were many non-Muslim Subcontinental (including Sri Lanka and beyond) sailors and maritime traders, and they were very probably in much bigger numbers than their Muslim Subcontinental counterparts even around the year 1500 given the demographics of the Subcontinent.
August 29th, 2010 at 4:22 pm
This I didn’t know, though still it doesn’t change the fact that there were many non-Muslim Subcontinental (including Sri Lanka and beyond) sailors and maritime traders, and they were very probably in much bigger numbers than their Muslim Subcontinental counterparts even around the year 1500 given the demographics of the Subcontinent.
you continue to operate under the assumption that there’s random sampling in these sorts of activities. so you dismiss the moors, even though i pointed out that they’re located on the east coast. the moors have traditionally been fisherman and traders, while the buddhists have been farmers, in their region. since you won’t accept the likelihood of non-random sampling by religion/caste (continuing to point out the total proportions as if that’s relevant in your inferences), i guess we’re not going to go anywhere here. thanks for the exchange, it’s done.
August 29th, 2010 at 4:34 pm
OK, you may be right about Sri Lanka, but what about the coastal India? It was and still is predominantly Hindu. Weren’t there enough sailors and maritime traders among the coastal Hindus to rival their Muslim counterparts there during the 15th-16th centuries?
Also Kerala is located on the west coast, so relatively more distant from SE Asia.
August 29th, 2010 at 4:53 pm
OK, you may be right about Sri Lanka, but what about the coastal India? It was and still is predominantly Hindu. Weren’t there enough sailors and maritime traders among the coastal Hindus to rival their Muslim counterparts there during the 15th-16th centuries?
the point though is that at some point between 1000 and 1500 it looks like muslims took over the long distance indian ocean trade networks. we see the network mostly on the back end, when the europeans challenged it. there was a period when hindu states sponsored trade missions to southeast asia. the cholas of tamil nadu did for example. but not beyond a certain point in time. the muslims had co-religionists all across the indian ocean, from madagascar to malacca. the hindus did not. when the europeans took over the indian ocean it opened the way again for hindu merchants, who reappear in arabia.
Also Kerala and Gujarat are located on the west coast, so relatively more distant from SE Asia.
the two are not analogous. kerala is only on the “west coast” in a technical sense. socio-culturally it’s an extension of tamil country on the other side of the ghats. so kerala had access to arabia, and, it was a major entrepot for spices from the east indies via the bay of bengal.
but again, as i keep telling you, you can’t assume random sampling from south asians when it comes to trade. in east africa most south asians are gujarati merchants, hindu patels, or muslim ismailis. they’re not gujarati peasants. when cheng ho and the portuguese arrived they noted that hindus ran the army, christians were tolerated and own land, and muslims controlled international trade.
August 29th, 2010 at 5:30 pm
Razib,
Did the Muslim trade from South Asia extend to Vietnam, China and other place North?
Were they in contact with Chinese Muslims of the period?
August 29th, 2010 at 5:34 pm
Also getting back to the original paper, it’s fascinating that Mozambique Bantus are different than other Bantu groups. I wonder who they are most related to, in a wider context.
I’ve always though that if Austronesians could get to Madagascar, then they could obviously get to the East Coast of Africa – so I wonder if this is reflected in these results?
August 29th, 2010 at 5:40 pm
Did the Muslim trade from South Asia extend to Vietnam, China and other place North?
Were they in contact with Chinese Muslims of the period?
canton had a large muslim population which was massacred around 840 by a tang emperor, and a minority of the modern hui probably derive from maritime, not inland, contacts. but from what i have read these were much more likely to be assimilated into han society than the hui of the northwest (who later migrated to beijing). generally the modern chinese muslims are assumed to have derived from the central asians who came with the mongols in the 13th century, but there are plenty of records of muslims in china in the 700s.
whether they were south asian or not is debatable. the chinese had a hard time distinguishing muslims of various kinds, and sometimes muslims from jews. what is clear is that the muslim maritime hegemony went no further than the arc between central vietnam and the philippines, and even then it was contested. the chams, a malay people of vietnam and cambodia, converted, and there were muslims on manilla bay when the spanish arrived. most historians who have ventured to guess assume that the phillipines, which is ethno-culturally similar to the rest of maritime southeast asia, would have been islamicized if the europeans did not intervene.
also, i think we should be careful about assuming the ethnic identity of the southeast asian traders. these things are fluid. even south asian muslims referred to themselves as “moors” (in the phillipines “moros”) it makes sense that people who assumed they were arab. btw, the spread of the shafi sunni school:
http://upload.wikimedia.org/wikipedia/commons/8/8d/Madhhab_Map2.png
endcap on the india ocean trade
August 29th, 2010 at 6:32 pm
[...] did this happen? This became the subject of discussion below. There aren’t any hard & fast answers here. Much of what I know is from a few books, After [...]
August 29th, 2010 at 6:37 pm
Just as I guessed in the beginning, there are lots of unknowns regarding the Islamization of SE Asia. What we see are mainly the end results. Your scenario is plausible, as Shafi’ism (the dominant Islamic madhhab of SE Asia today) was (and still is) the dominant Islamic madhhab of the main Indian Ocean Basin, but the rise to dominance of the Muslim merchants in the Indian Ocean trade networks is hard to track due to the paucity of documentation.
August 29th, 2010 at 6:51 pm
part of the issue is the lack of documentation from humaid tropical asia. stuff doesn’t preserve if it’s not stone. we’ve got OK middle eastern and european records.
August 30th, 2010 at 9:57 am
As I was reading the paper, I recalled a first-hand account by Ludovici di Varthema, an early 16th-century Italian traveler to Mozambique, of what sound like cavern-dwelling click speakers in a mountainous region of the mainland. More on that here (in the interest of full disclosure, it’s my own blog).