Friday, May 18, 2012

Nilo-Saharan Homeland Proposed

There is a firm linguistic based proposal for the place of origin of the Nilo-Saharan languages (with the reasoning behind it explained in the linked material), a language family most prominently represented by the Maasi of Kenya, who are known for a predominantly pastoralist lifestyle, particularly in East Africa.

The proto-Saharo-Sahelian society can therefore also be placed somewhere in the combined regions of northwestern Sudan and northeastern Chad.
From "The Archaeology of Africa" (1993) via the Linear Population Model blog.

The location in question included the Blue Nile basin near where it merges with the White Nile, and Lake Chad's endoheic basin, and probably dates to a time of increasing aridity in that region.

As my post yesterday noted, an expansion from this source may have been as recent as 1000 B.C.E., at the dawn of the Mediterranean basin's Iron Age, or perhaps a couple of centuries earlier, around the time of Bronze Age collapse.

The accomplishment of locating this homeland is somewhat muted by the fact that the Nilo-Saharan languages are probably the youngest of the major African language families, that it is one of the smaller major language families of the world, and that it is mostly fairly compact in its geographic range (with one notable branch of the language family providing an exception that proves the rule). This makes the task of discerning the relationships between the family's languages phylogenetically, and their geographic origins considerably easier than in the case, for example, of the far older and more widely dispersed Afro-Asiatic macro-language family that includes, for example, the Berber, Hebrew and Arabic languages. On the other hand, the Nilo-Saharan languages were not literate languages until very late in history, and that impedes inquiry into its historical origins, so the task was not entirely without its challenges.

The Origins Of Neighboring Language Families In A Nutshell

Much of the overall narrative of African pre-history hinges on how one resolves the places of origin and sequencing of linguistic expansions that gave rise to the current linguistic landscape of Africa.

The general vicinity where the Nilo-Saharan languages arose also probably overlaps or is adjacent to the homeland for the Chadic languages, associated with pastoralists who roam the African Sahel (most notably the Hasua people) within the Afro-Asiatic language family.

Marnie quotes Tishkoff who brings autosomal genetic insight to the potential relationships between the two language families established from other sources:

Our data support the hypothesis based on linguistic, archeological, mtDNA, and Y chromosome data, that the Sahel has been a corridor for bi-directional migration between eastern and western Africa.  
We observe the highest proportion of the "Nilo-Saharan AAC [ed. ancestral autosomal genetic component]” in the southern/central Sudanese populations (Nuer, Dinka, Shilluk, Nyimang), with decreasing frequency from northern Kenya (e.g. Pokot) to northern Tanzania (Datog, Maasai). From K = 5-13 [ed. refering to the number of ancestral components chosen by the analyst into which the autosomal data are arbitrarily resolved by a mathematical algorhythm], all Nilo-Saharan speaking populations from Kenya, Tanzania, southern Sudan, and Chad cluster with west-central Afroasiatic Chadic speaking populations.   
These results are consistent with linguistic and archeological data, suggesting a possible common ancestry of Nilo-Saharan speaking populations from an eastern Sudanese homeland within the past ~10,500 years, with subsequent bi-directional migration westward to Lake Chad and southward into modern day southern Sudan, and more recent migration eastward into Kenya and Tanzania ~3,000 ya (giving rise to Southern Nilotic speakers) and westward into Chad ~2,500 ya (giving rise to Central Sudanic speakers).   
A proposed migration of proto-Chadic Afroasiatic speakers ~7,000 ya [5000 B.C.E.] from the central Sahara into the Lake Chad Basin may have caused many western Nilo-Saharans to shift to Chadic languages. Our data suggest that this shift was not accompanied by large amounts of Afroasiatic gene flow. Analyses of mtDNA provide evidence for divergence ~8,000 ya [6,000 B.C.E.] of a distinct mtDNA lineage present at high frequency in the Chadic populations and suggest an East African origin for most mtDNA lineages in these populations."
Since Tishkoff's dates are heavily driven by mutation rate methodologies that are often too unreliable to the level of accuracy necessary to make them useful in distinguishing alternative narratives of pre-history, they must be taken with a grain of salt although the relative sequence of events suggested by the mutation rate dates, at least, is probably the most probable possibility absent evidence to the contrary.

While Chadic linguistic populations tend to have generally East African mtDNA lineages (but see, in particular, this paper regarding distinctively Chadic mtDNA lineages), their most distinctive and defining Y-DNA haplogroup, R1b-V88 (which is a branching basal to the European branches of Y-DNA haplogroup R1b) has a likely origin somewhere in far Southeast Europe or Western Central Asia near the likely point of the Y-DNA haplogroup R1a v. R1b divide.  This is very far afield from most of the rest of the African genetic legacy for a population with its ethogenesis possibly taking place in the vicinity of Lake Chad, two thousand years or so before the world's first written historical records, and hence, at a time when what we know about the civilizations that existed is pretty much limited to pots and bones dated to that era.

The Chadic mtDNA paper linked above also notes that:

Within the Afro-Asiatic language phylum, the Chadic branch is linguistically close to the East African Cushitic branch although they are separated by ~2,000 km of territory in which different Semitic and Nilo-Saharan peoples live today. We show that only northern Cushitic groups from Ethiopia and Somalia are genetically close to Chadic populations. Thus, the archaeologically and linguistically supported route of proto-Chadic pastoralists via Wadi Howar [the remnant of the ancient Yellow Nile, a tributary of the Nile from about 8000 to 1000 BCE] to the Chad Basin may have genetic support.
Both language families probably expanded sometime after the initial expansions of the Niger-Congo language family, and the Chadic languages probably expanded no earlier than and quite possibly considerably later than the Cushitic language family.

The homeland of the Niger-Congo language family isn't terribly controversial, and was probably more or less smack dab in the midddle of West Africa at a confluence of some its major rivers; it has only a few remote members not attributable to Bantu expansion (the most notable is the Kordofan language family of Nuba Mountain in Sudan). The Niger-Congo language expansion in Africa, could pre-date the Afro-Asiatic language expansion in Africa, but could also plausible have taken place not long after Afro-Asiatic languages started to expand.

The Niger-Congo languages seem to be associated with locally domesticated Sahel crops, while the Afro-Asiatic languages seem to be associated with Fertile Crescent food production packages. A determination of which language family came first is deeply intertwined with the question of whether Sahel agriculture was developed independently of the Fertile Crescent package, or was only made possible because of animal domestications and concepts that had their source in the Fertile Crescent.

Afro-Asiatic Linguistic Origins

But, the Cushitic language family's origins are a much hotter issue. This is because the status of the Cushitic languages as a source or sink within the Afro-Asiatic languages, which in turn goes to the controversial question of whether the Afro-Asiatic languages have an African or West Asian place of origin.

A "First Farmers" paradigm would suggest a probably origin for the language family in the Levant that expanded and diversified with the Fertile Crescent Neolithic, or perhaps a secondary origin somewhere in the Nile Basin as a language of early adopters of the Fertile Crescent Neolithic in Africa who expanded elsewhere in Africa (with Semitic arriving in the Levant via Egyptian cultural influence).

The main argument for the African origin of Afro-Asiatic languages is that five of the six subfamilies of the Afro-Asiatic languages appear to have origins in Africa (Berber, Chadic, Cushitic, Omotic, and Coptic) and are exclusively African, leaving only the Semitic languages, whose origins can be traced quite a distance back in time through historic and linguistic means (the Ethio-Semitic languages arrived in Ethiopia as a single proto-language probably sometime in the Bronze Age from the Levant; the family tree of the West Semitic languages are reasonable well established in both time and place).

The argument for an African origin of the Afro-Asiatic languages is also bolstered by the fact that the speakers of the Afro-Asiatic languages are not united by a genetic signature that can be rooted in Levatine peoples. Some distinctively Semitic genetic ancestry markers are largely absent in other Afro-Asiatic populations, but the phylogeny of the most widely shared genetic ancestry markers of Afro-Asiatic language speakers have roots that are widely, although not universally, believed to have African origins. But, the genetics are not conclusive, because there is no one genetic ancestry marker that is clearly a distinctive component of the gene pool of all six of the constituent language families of the Afro-Asiatic languages.

In a West Asian origin theory, proto-Afro-Asiatic was imposed on Egyptians and perhaps separately upon Berbers by an early Neolithic colonist superstrate from the Levant made up of people who had enough power to induce language shift in the subject peoples despite not having much of a demic impact visibile in the gene pool, a phenomena documented more than once historically.

In an African origin theory, the Semitic languages, whose speakers are genetic outliers relative to the African language families in the Afro-Asiatic macro-language family, has its roots in an Egyptian superstate population that caused a language shift in the Levantine substrate peoples. We know, historically, that the Levant did have an Egyptian superstrate for more than a thousand years.  But, we don't know what kind of influence Egyptians had in the Levant centuries prior to the earliest attested West Semitic languages which is when the proto-Semitic language would have flourished somewhere in West Asia. 

As a footnote, it is also worth observing that, if there was an Egyptian origin, it would have had to have taken place at least 1500 years or so before Hebrew ethnogenesis which is relatively well dated historically and linguistically, since the West Semitic Akkadian languages and related Northwestern Semitic languages are attested long before the Hebrew people came into being as a cultural construct.  The origins of the Semitic languages in West Asia were most definitely not in a Hebrew Exodus from Egypt.

Of course, even though both a sub-Saharan African origin for the Afro-Asiatic languages and an Egyptian origin for the Afro-Asiatic languages would both probably have the same Egyptian mediated origin for the West Semitic languages (since almost all road to West Asia from Africa with any demographic must pass through Egypt), the case for the Cushitic or Coptic languages (or for that matter the Berber or Omotic languages) being the most basal root of the Afro-Asiatic linguistic tree within Africa is not at all settled (and hence the controversy over the location and timing of the Cushitic homeland).

For what it is worth, while I am personally agnostic on the question of whether the Afro-Asiatic languages arose in Egypt or the Levant (most plausibly Jericho, the first Levantine city, given the tendency of languages to snowball into being predominant in a way very sensitive to the initial conditions when the expansion starts), but I believe that there is a very strong case that the Afro-Asiatic languages expanded from Egypt within Africa with the expansions of the Fertile Crescent Neolithic package of crops and animals, rather than from Ethiopia, for example, which is where the Cushitic languages are mostly spoken today, or some other location. And, there is a very solid archaeological and inferential case that the Fertile Crescent Neolithic package first established itself in Africa in the Nile Basin.

In my view, to the extent that the genetic patterns don't unequivocally show this pattern it is a result of Epipaleolithic (or earlier) population migrations that have a well established archaeological basis, and non-demic language shift events. But, the pattern of languages spreading with societies that have adopted food production methods to the detriment of the existing languages of hunter-gatherer populations is too close to universal to make it anything other than the possibility that is by far the most likely. Clearly there were at least one instance, and really there is a good case that there were two or three instances of predominantly non-demic language shift in the origin of the major Afro-Asiatic language families, given the evidence from the population genetics of the modern speakers of these languages, so genetics can't easily resolve the question of the relationship between the main Afro-Asiatic language families and other evidence must be brought to bear to answer it. Given that there is considerable disagreement amongst linguists on this question given purely linguistic evidence, this also isn't conclusive.


Etyopis said...

There is one really easy litmus test to verify if Academic/cultural bias is involved in the 'debate' of the Afroasiatic urehimat.
Just think of a language group that has almost all of its branches spoken in Asia, and one relatively young branch spoken in Africa, and ask yourself, would any one argue for an African origin of such a hypothetical language group with a straight face?

The question of the urehimat of Afroasiatic languages is tri-dimensionally loaded, it is loaded with respect to religion, the language group of the Torah, Gospels and Koran, it is loaded with respect to race, no explanation needed and it is loaded with respect to culture, as the areas the Aa languages are spoken in are largely inhabited by people that are the most vehemently ethnocentric in the world.

Heck the very script I am using to type this message originated with Afroasiatic speakers.

I think the fact that the origin of Aa languages has tremendous implications in the historical course that humanity took is a big part of the reason for an unsolicited and scientifically thorough analysis of it, that and also the fact that Indo European gets all the attention, another loaded language group.

andrew said...

The speakers of the Kordofan languages are genetically identical to Nilo-Saharan language speakers, indicating a probably language shift.