Wednesday, June 13, 2012

The Oldest Domesticated Cattle Bones In Africa

A 2005 report provides the oldest solid evidence of the use of domesticated cattle and associated Neolithic transition in Egypt (no later than 7000 BCE (9000 BP), and no earlier than 7300 BCE (9300 BP) in the Middle Nile) (and for that matter in Africa), although the local variety of wild aurochs from which domestic cattle were domesticated were present (and presumably hunted) at least six thousand years earlier according to newly dated petroglyphs from the same general region.

The Advent Of Middle Nile Neolithic Is Established To Within About 300 Years

Marnie has some notable data points on the archaeology of domesticated cattle in Egypt which he has translated from a Francophone report of a Swiss research team. An excerpt from his translation is below:
The difference between the tombs attributed to the Mesolithic and the cemeteries of the Neolithic are fundamental. . . . In a millennium, Nubian society had completely transformed their social organization. This transformation must have taken place on account of the introduction of domestic cattle, the oldest findings of which are in the Middle Nile at Nabta Playa in Egypt and Kerma. . . . [examination of the Northern Sudanese Middle Nile site of] El Barga . . . confirmed our hypothesis that El Barga is a Neolithic site: a man's grave next to which was deposited the skull of a domestic cattle buried just above the burial of a child. Two carbon 14 dates have given results of approximately 5750 BCE [7762 bp] that make this the oldest Neolithic site in the Nile Valley. . . .

[T]he cemetery shows technical and social transformations that had already taken place and one can't doubt that the introduction of pastoralism had occurred at an earlier time. The discovery of two sites located five kilometers from El Barga confirmed this hunch. . . . The sites . . . contain the bones of domestic cattle. Some of the remains were dated by radiocarbon dating to approximately 7000 BCE [9000 bp].
The study classifies a nearby site 300 years earlier as Mesolithic (7300 BCE; 9300 BP), pin pointing the Mesolithic to Neolithic transition at this location well down the Nile, at least insofar as cattle herding is concerned (but also in the overall cultural transformation that results from the transition), quite precisely.

Marnie also notes that in another post that the dating of "Egyptian rock art drawings of aurochs at Qurta on the Upper Nile . . . . [which] indicate that aurochs (bos primigenius) were present south of the First Cataract, earlier than 15,000 years ago."

To the best of my knowledge, the oldest archaeological evidence for the domestication of cattle (or for that matter, the oldest reliably dated evidence of the Neolithic revolution generally) in Africa is in Egypt, so this very precisely establishes the latest possible date for the arrival of the Neolithic in the Middle Nile, while also bracketing out the earliest possible date for its arrival there.


The study, as translated, correctly on this point, I think, states: "The sites . . . contain the bones of domestic cattle. Some of the remains were dated by radiocarbon dating to approximately 7000 BCE [9000 bp]."  This has some weasle wording, implying that domestic cattle probably go back to 7000 BCE in continuity with the remains at the site generally as part of a common culture, but not saying anything nearly that definitive.  This language wouldn't be inconsistent with a reading in which the domestic cattle bones in the neighboring sites really back pack only to 6000 BCE, and that the earlier remains were associated with a continuous somewhat advanced for a Mesolithic society proto-Neolithic culture (a goats and sheep without cattle culture, for example), and that it is premature to state that the Neolithic transition in the Middle Nile dates to 7000 BCE rather than 6000 BCE.  I think that the authors intend a stronger reading that that, which their juxtaposition of findings and strong implications suggest, but that they don't feel that their evidence is so definitive that they can truly commit to a more explicit conclusion yet.

I couldn't locate information on error bars on the dates cited in the translation in the original, but perhaps an informed commenter could provide information on how accurate carbon-14 dating is at this time depth or call attention to the pertinent language in the paper or its supplemental materials.

Also, I have assumed, but not confirmed that the carbon-14 dates cited in the paper are calibrated dates rather than crude ones, a distinction that is material to the tune of something on the order of low single digit centuries at this time depth. At the very least, one ought to assume that the dates discussed in the paper, whether calibrated or not, were at least compared with dates in the same units from the literature given the claim that this is the oldest such site.

Implications For The Big Neolithic Transition Picture

The study does not appear to distinguish the domesticated cattle bones found in the Middle here from Near Eastern domesticates from a thousand years earlier in the Upper Euphrates Valley. As Marnie noted in another post citing Zeder (2008):

"Domestication and early agriculture in the Mediterranean Basin: Origins, diffusion and impact" has these domestication dates BP: sheep 11,000; goats 11,000; pigs 10,500; cattle 10,000 (Upper Euphrates Valley). However, she states that morphologically altered domestic cattle are not found in Central Anatolia until after 8,500 BP.
Thus, this new data is consistent with the most widely accepted (although, of course, not universally accepted) narrative in which the primary domestication of cattle in the Fertile Crescent ca. 8000 BCE (10000 BP) which has spread to the Middle Nile in Upper Egypt and Northern Sudan roughly 800 to 1000 years later via migrating pastoralists rather than a local domestication in Egypt.

Open Questions Requiring Further Investigation

This analysis still leaves open questions and room for controversy (although some questions left open by this report's conclusion and mentioned below may have actually already been resolved with other research):

Why Did Cattle Domestication Leap Frog From The Fertile Crescent To The Middle Nile?

You can't get to the Middle Nile without going through the Sinai, and either the Lower Nile or the Eastern Sahara immediately to the West of the Nile River Valley (which had a climate more favorable for cattle herding then than it does now). So, where isn't aren't their older domesticated cattle skulls in the Lower Nile, for example, than there are in Middle Nile?

Marnie has suggested that there may have been an independent cattle domestication from Egyptian aurochs, something that may have been favored everywhere aurochs were found at the time by large scale climate shifts in the entire region, and perhaps mere culture transmission of the idea of domesticating aurochs in general from travelers who did not bring the animals and the entire cultural package with them in a demic migration. He notes in support of this the somewhat distinctive character of domesticated cattle in Africa. I tend to disfavor this interpretation, mostly because a technology transfer scenario is consistent with the known dates and has been documented fairly persausively for most of the rest of the world where the Fertile Crescent package (or other Neolithic packages) arrived.

Other possibilities are that as of ca. 7800 BCE to 7000 BCE:

(1) archaeological preservation conditions then and in the millennia that followed may not have been as good in Lower Nile as in the Middle Nile, destroying the evidence of an earlier Lower Nile Neolithic transition (this could be due, for example, (A) to continuously high populations densities in the Lower Nile with later populations defiling older remains, (B) to greater practical use and recycling of bones by people in the Lower Nile, (C) to more dramatic flooding, (C) to higher groundwater levels, (D) to temperatures more favorable to decomposition, or (E) to poorer soil, moisture and temperature conditions for preservation enhancing mud packs and clays), or

(2) conditions for cattle herding may have been better in the Middle Nile than the Lower Nile, perhaps due to elevation or localized flooding concerns (the even later appearance of domesticated cattle in Central Anatolia which was closer to the site of the earliest known auroch domestication, however, tends to disfavor a scenario in which highland river basins are favored ecology for cattle herding; but the fact that cattle seem to have been domesticated in the Upper Euphrates Valley rather than the Euphrates Delta, suggests that maybe Lower Egypt wasn't an optimal ecology for cattle herding; perhaps it one goes too far into the highlands, such as Central Anatolia and Ethiopia, the evenings get too cold in the winter or the grasses don't get enough moisture to be as thick and reliable), or

(3) existing populations may have posed less of a barrier to the introduction of cattle herding in the Middle Nile than the Lower Nile, either because:

(A) the Lower Nile people were so prosperous in hunting and gathering at the time that they resisted encroachment and innovation from cattle herders, or because

(B) a society involving semi-sedentary fishing, pre-farming of wild crops and undomesticated animals, pottery making, and short range hunting and gathering in a confined and abundant territory gave rise to a society more capable of rapidly and voluntarily (or with minimal coercion from a thin superstate class) integrating Fertile Crescent cattle herding into their way of life even though this earlier collection of intermediate innovations towards sedentary society was less transformative and less dynamic.

The (3)(B) scenario is quite similar to Marnie's hypothesis, but would attribute modern domestic cattle distinctiveness in Egypt to admixture of imported domesticated cattle with wild aurochs over the millennia until the wild aurochs were extinct in the region, or to the introgression and genetic influence in Indian cattle in the Bronze Age, rather than to an independent domestication.

Of course, for either of the claims in possibility (3) to hold up, one must first establish from the archaeological record that there were strong cultural distinctions between the Lower Nile and Middle Nile in the 8th millennium BCE.

What Else Accompanied Domestic Cattle To The Middle Nile?

To what extent was the Neolithic transition in the Middle Nile a population replacement phenomena and to what extent was it was culturally adoption phenomena, and how gender biased was the demic impact that did result whatever its magnitude? The male and female uniparental and the autosomal genetics of North Africa and East Africa all make clear that there were one or more demic back migrations from the Near East to that region. What particular components of the modern population genetic mix arrived in this cattle herding wave relative to Upper Paleolithic back migrations and later back migrations? What did the Middle Nile look like from a population genetic perspective in the Mesolithic before Fertile Crescent food production techniques arrived?

Were the dramatic societal changes observed in cemeteries home grown or do they represent adoption of a Fertile Crescent cattle herding culture?

At what time did sheep, goats, and Fertile Crescent farming crops arrive at which places relative to the arrival of domesticated cattle?   Did they arrive all at once after consolidating into a complete Neolithic package in the Near East, or did they arrive in multiple closely successive waves?  See, e.g., here and here. See also, here (citing Dillon (2007) which argues that "The domestication of sorghum has its origins in Ethiopia and surrounding countries, commencing around 4000–3000 BC.")

It is generally accepted that the donkey was domesticated in Egypt. The usual chronology is to assume that this happened sometime after Fertile Crescent domesticated animals arrived (I've seen 6000 B.C.E. proposed as a domestication date). Where did this happen and is that the right chronology? What evidence supports that conclusion?

When do we start to see cultural innovations from the Fertile Crescent package, in general, in the Nile, and what particular innovations do we see? Do those innovations show continuity from a Mesolithic substrate in Egypt, or are they something genuinely new under the sun?

Taking it as a given that the oldest evidence for the Fertile Crescent Neolithic in Africa is found in the Middle Nile at this location, do subsequent dates from elsewhere in Africa allow us to develop an empirically grounded picture of when and where it spread from that African epicenter? To the extent that this picture is consistent with one or more of multiple possible scenarios for the expansion and spread of the Afro-Asiatic and Nilo-Saharan languages, those scenarios are favored relative to the alternatives.


This is a bit of a raw data dump, but is included for convenience and future reference. The narrative part of the post ends above. This is one big footnote.

In general, what is notable is that the currently available evidence suggests that the domestication of plants used in Sahel agriculture was surprisingly recent (with pearl millet only just barely domesticated in Africa before it was exported to India), that the evidence for an independent African domestication of cattle is shallow, that sheep and goats arrived in African, if anything after if not simultaneously with, cattle herding, rather than before it, and that the human presence in East African was intermittent with modern humans frequently banished from the region for centuries by prolonged arid periods.

Domestication dates and climate in North Africa

Selective domestication and climate data points from here (citing "Are the early Holocene cattle in the Eastern Sahara domestic or wild?" Fred Wendorf & Romuald Schild (Evolutionary Anthropology 3(4), 1994)):

Eastern Sahara

The Eastern Sahara near Upper Egypt has been repeatedly abandoned and repopulated in response to shifting climate conditions during the Holocene era. It provides a latest possible introduction date for sheeps and goats which quite possibly post-date the widespread use of cattle as a food source.

[T]here is no evidence of human presence [in the Eastern Sahara near Upper Egypt] before 9,500 B.P. except for a radiocarbon date of around 10,000 years ago from a hearth west of Dakhla. The earliest sites with large bovid remains are imbedded in playa sediments that overlay several meters of still older Holocene playa deposits. . . [of] Middle Paleolithic artifacts[.]. . .

These assemblages have been classified as the El Adam type of the Early Neolithic. Several radiocarbon dates place the complex between 9,500 and 8,900 B.P. There is no evidence that there were wells during this period. It is assumed, then, that these sites represent occupations that took place after the summer rains and before the driest time of the year when surface water was no anger available. Three of these sites . . . have yielded, through excavation, more than 20 bones and teeth of large bovids that have been identified as Bos. These occurred along with several hundred bones of gazelle (Gazella dorcas and G. dama) and hare (Lepus capensis); a few bones of jackal (Canis aureus), turtle (Testudo sp.); and birds (Otis tarda and Anas querquedula); the large shell of a bivalve (Aspatharia rubens), probably of Nilotic origin; and various snail shells (Bulinus truncatus and Zoorecus insularis).

After a period of aridity around 8,800 years ago, when the desert may have been abandoned, the area was re-occupied by groups with a lithic tool-kit that emphasized elongated scalene triangles. The grinding stones, scrapers, and rare pieces of pottery that are present characterize the El Ghorab type of Early Neolithic and have been dated between 8,600 and 8,200 B.P. Oval slab-lined houses occur during this phase. all of them located in the lower pans of natural drainage basins. However, there are no known wells, suggesting that the desert still was not occupied during the driest part of the year. Faunal remains are poorly preserved in these sites and. indeed, only one bone of a large bovid was recovered from the four sites with fauna. in these sites the Dorcas gazelle is the most numerous, followed by hare, together with single bones of wild cat (Felis silvestris), porcupine (Hystrix cristata), desert hedge-hog (Paraechinus aethiopicus) an amphibian, and a bird.

Another brief period of aridity between 8.200 and 8,100 B.P. coincides with the end of the El Ghorab type of Early Neolithic in the desert. With the return of greater rainfall about 8,100 B.P., a new variety of Early Neolithic, the El Nabta type, appeared in the area. . . . Radiocarbon dates place the El Nabta sites between 8,100 and 7,900 B.P. . . . Beside each house was one or more bell-shaped storage pits; nearby were several deep (2.5 m) and shallow (1.5 m) water-wells. This site, located near the bottom of a large basin, was flooded by the summer rains. The houses were repeatedly used, probably during harvests in fall and winter. Several thousand remains of edible plants have been recovered from these house floors. They include seeds, fruits, and tubers representing 44 different kinds of plants, including sorghum and millets. All of the plants are morphologically wild, but chemical analysis by infrared spectroscopy of the lipids in the sorghum indicates that this plant may have been cultivated. Of the four El Nabta sites that have yielded fauna, two contained bones of a large bovid identified as Bos. The faunal samples from the other two sites are very small.

Another brief period of aridity separated the El Nabta Early Neolithic from the succeeding Middle Neolithic, which is marked by the much greater abundance of pottery. In addition, each piece of pottery is decorated over its entire exterior surface with closely packed comb- or paddle-impressed designs. Some of the pots are large, and analysis of the clays indicates that they were made locally. There were also some changes in lithic tools. More of them were made of local rocks, but there was sufficient continuity in lithic typology to suggest that the preceding Nabta population was also involved.

Radiocarbon dates indicated an age for the Middle Neolithic between 7,700 and 6,500 B.P. The sites from the early part of this period range from one-or-two house homesteads in some of the smaller playas to multi-house villages in the larger basins. There is also one very large settlement along the beach line of the largest playa in the area, as well as, small camps on the sandsheets and the plateaus beyond the basins. . . . All of the sites have large, deep walk-in wells and storage pits. Except for the small camps, most of the sites appear to have been reused many times, with new house floors placed on top of the silt deposited during the preceding flood.

Excavations at five Middle Neolithic sites have yielded more than 50 bones from large bovids. Most of these bones came from the large “aggregation” site at the margin of the largest playa in the area and from the early Middle Neolithic site, dated before 7,000 B.P., which is located on a dune adjacent to another large playa. Each of the other three Middle Neolithic sites yielded only one to three large bovid bones.

Around 7,000 B.P., the remains of small livestock (sheep or goats) appear in several Middle Neolithic sites at Nabta. Because there are no progenitors for sheep or goats in Africa, these caprovines were almost certainly introduced from southwest Asia. . . . the paucity of the fauna and the absence, except for cattle and small livestock, of animals that require permanent water suggests a rather poor environment, most likely comparable to the northernmost Sahel today with about 200 mm of rain or less annually.

The Middle Neolithic was brought to an end by another major but brief period of aridity slightly before 6,500 B.P., when the water table fell several meters and the floors of many basins were deflated and reshaped, The area probably was abandoned at this time. . . .

With the increase in rainfall that began around 6,500 years ago. human groups again appeared in the area, but this time with ceramic and lithic traditions that differed from those of the preceding Middle Neolithic. This new complex, identified as Late Neolithic, is distinguished by pottery that is polished and sometimes smudged on the interiors. This pottery resembles that found in the slightly later (about 5,400 or, possibly, 6,300 B.P.) Baderian sites in the Nile Valley of Upper Egypt. It seems likely that an as yet undiscovered early pre-Badarian Neolithic was present in that area and either stimulated or was the source of the Late Neolithic pottery in the Sahara. It is unlikely, however, that this hypothetical early Nilotic Neolithic will date much earlier than 6,500 B.P. There are terminal Paleolithic sites along the Nile that are dated to around 7,000 B.P. and it is highly improbable that two such different life-ways could co-exist exist for long in the closely constrained environment of the Nile Valley.. . . The end of the Late Neolithic in the Eastern Sahara is not well established.The period may have tasted until around 5,300 B.P. when this part of the Sahara was abandoned.

Due to poor preservation faunal remains in Late Neolithic sites are not as abundant as those from the Middle Neolithic. However, the Late and Middle Neolithic samples generally include the same animals suggesting that the environment was also generally similar during these periods. Although large bovids are also present in three Late Nealithic sites, and more frequently than in the faunal assemblages of the preceding period, they still are a minor component of the sample.

The Late Neolithic Nabta is marked by interesting signs of increased social complexity, including several alignments of updght slabs (2 x 3 m) imbedded in, and sometimes almost covered by, the playa sediments. Circles of smaller uptight stabs may calendrical devices. Stone-covered tumuli are also present; two of the smaller ones contain cow burials, one in a prepared and sealed pit. . . .

Even the earliest of these early Holocene Eastern Sahara sites have been attributed to cattle pastoralists. It is presumed that these Early Neolithic groups came into the desert from an as yet unidentified area where wild cattle were present and the initial steps toward their domestication been taken. This area may have been the Nile Valley between the First and Second Cataracts, where wild cattle were present. Moreover, lithic industries were closely similar to those in the earliest Saharan sites. . . . It is assumed, because of the apparrent absence of wells at the earliest sites, that the first pastoralists used the desert only after the summer rains, when water was still present in the larger drainage basins. After 8,000 years ago, when large, deep wells were dug, the pastoralists probably resided in the desert year-round.

Other Parts of North Africa:

The antiquity of the known domestic cattle elsewhere in North Africa does not offer much encouragement with regard to the presence of early domestic cattle in the Eastern Sahara.

Gautier recently summarized the available data, noting that domestic cattle were present in coastal Mauritania and Mali around 4,200 years ago and at Capeletti in the mountains of northern Algeria about 6,500 years ago. At about that same time, they may have been present in the Coastal Neolithic of the Maghreb. Farther south in the Central Sahara, domestic cattle were present at Meniet and Erg d’Admco, both of which date around 5,400 years ago, and at Adrar Rous, where a complete skeleton of a domestic cow is dated 5,760 +/- 500 years B.P ].

Domestic cattle have been found in western Libya at Ti-n-torha North and Uan Muhuggiag, where the lowest level with domestic cattle and small livestock (sheep and goats) dated at 7,438 t 1,200 B.P. At Uan Muhuggiag, there is also a skull of a domestic cow dated 5,950 +/- 120 years. In northern Chad at Gabrong and in the Serir Tibesti, cattle and small livestock were certainly present by 6,000 B.P. and may have been there as early as 7,500 B.P. We are skeptical, however, about the presence of livestock at Uan Muhuggiag and the Serir Tibesti before 7,OO0 B.P., when small livestock first appear in the Eastern Sahara, if we must assume that these animals reached the central Sahara by way of Egypt and the Nile Valley. This also casts doubt on the 7,500 B.P. dates for cattle in these sites.

The earliest domestic cattle in the lower Nile Valley have been found at Merimda, in levels that have several radiocarbon dates ranging between 6,000 and 5,400 B.P. and in the Fayum Neolithic, which dates from 6,400 to 5, 400 B.P. These sites also have domes-tic pigs and either sheep or goats. In Upper Egypt, the earliest confirmed domestic cattle are in the Predynastic site of El Khattara, dated at 5,300 B.P. However, domestic cattle were almost certainly present in the earliest Badarian Neolithic, which dates before 5,400 B.P. and possibly were there as early as 6,300 B.P. Farther south, in Sudan near Khartoum, the first domestic cattle and small livestock occurred together in the Khartoum Neolithic, which began around 6,000 B.P.

It is probably significant that none of the early Holocene faunal assemblages in the Nile Valley from the Fayum south to Khartoum that date between 9,000 and 7,000 H.P contains the remains of cattle that have been identified as domestic. It is this ab-sence of any evidence of recognizable incipient cattle domestication in the Nile Valley or elsewhere in North Africa that cautions us to consider carefully the evidence of early domestic cattle in the Eastern Sahara. . .

Equus, even in the Late Paleolithic, seems to have been confined to the Red Sea Hills and the east bank of the Nile. . . .

[Did] the first steps toward cattle domestication began in the Nile Valley, perhaps during the Late Pleistocene, when there is so little faunal evidence to support that hypothesis? The answer may lie in the identification of the cattle remains found in the Late Paleolithic sites in Sudanese and Egyptian Nubia. It has been suggested that it would be very difficult to separate the bones of the incipiently domestic cattle from those of wild cattle.

More On African Cattle Domestication

See also here:

The problem of cattle domestication in Northeastern Africa is considered and hopefully ‘‘solved’’ in the light of new mtDNA evidence which suggest an early late Pleistocene split between African, Asian, and Eurasian wild Bos populations.

The Grigson’s study concluded cattle from all periods at Nabta Playa were morphologically wild (2000).

Smith’s study: morphologically wild prior to and including the El Nabta/Al Jerar Maximum (7050 – 6150 BC), but domesticated from the Ru’at El Ghanam phase (5900 – 5500 BC) onward.

More on Sorghum Domestication With A Barley Footnote

From the same source:

One of these sites yielded charred seeds of wild millet and two varieties of legumes (Wasylikowa, report to F. Wendorf 1996)

It also has a reference to possible early domesticated sorghum. Although again the case is bit weak. There’s more reference to it here. The seeds don’t appear to resemble any kind of cultivated sorghum though. They did seem to be harvesting and storing them in large amounts; some of the houses had storage pits for the grains.

Preliminary chemicalanalyses by infrared spectroscopy of the lipids in the archaeological sorghum show closer resemblance to some modern domestic sorghum than to wild varieties (Wasylikowa et al. 1993)

In a later publication (97) Wasylikowa describes the Sorghum as more likely to be wild, after another study of the seeds showed them to be typically wild seeds.

Smaller grain size and the lack of any spikelets containing attached branchlets of the inflorescence or rachis fragments suggest that the material harvested and eaten at the Nabta Playa site were of a wild type.

This sorghum doesn’t seem to ‘spread out’, as farmers tend to expand massively into their hunter gatherer neighbours very rapidly. The expansion of domesticated sorgum doesn’t seem to begin until the expansion of the domesticated donkey, which parallels it’s spread into Asia quite well, and the donkey seems to have been domesticated about 6,000 BP.

The barley recovered from this site during the 1977 excavations (Hadidi in Wendorf and Schild 1980: 347) is regarded as intrusive. . . .

Around 8000 cal B.P. there was an important new addition to the food economy of the Middle Neolithic. Domestic caprovids, either sheep or goat, or both, were introduced from Southwest Asia, probably by way of the Nile Valley (although the oldest radiocarbon dates now available for the Neolithic along the Nile are about 500 years later)

See also here (citing Dillon (2007) (who in turn seems to be replying on the book C. Wayne Smith, Richard A. Frederiksen, "Sorghum: Origin, History, Technology, and Production" (2000) in the vicinity of pages 26-39, which in turn cites Murdock (1959) (Mande in West Africa ca. 4500 BCE and then to Sudan ca. 4000 BCE); Ehret (1988) (Lake Chad 4000 BCE); Shaw (1976) citing Clark (Adrair Bous 2000 BCE); J.R. Harlan (1976; 1989; 1995) (3000 BCE in Africa since in India by 2000 BCE or even as early as 3000 BCE; Doggett (1988) (3000 BCE); Mann (1983) (6000 BCE); Fred Wendorf (1984)(not 8000 BCE); Stemler (Sahara and Nile 1000 BCE); Wigboldus (1990) (after 0 CE); Rowley-Conwy 1997; 1999) (relatively recently); A. A. Magid (1989; 1995) (relatively recently in Nile Valley); Barnett (1997) (relatively recently in Ethiopia); R. Haaland (1995; 1998) (wild sorghum introduced to India, domesticated there and reintroduced to Africa; notably, none of the book's sources argue for an Ethiopian sorghum domestication ca. 4000 BCE-3000 BCE):

Arthropological evidence suggests that hunter-gatherers consumed sorghum as early as 8000 BC. The domestication of sorghum has its origins in Ethiopia and surrounding countries, commencing around 4000–3000 BC. Numerous varieties of sorghum were created . . . . These improved sorghum types were spread via the movement of people and trade routes into other regions of Africa, India (approx. 1500–1000 BC), the Middle East (approx. 900–700 BC) and eventually into the Far East (approx. AD 400).

By the time sorghum was transported to America during the late 1800s to early 1900s, the diversity of new sorghum types, varieties and races created through the movement of people, disruptive selection, geographic isolation and recombination of these types in different environments would have been large. . . . Sorghum is the closest cultivated relative of sugarcane.

The book notes (at 30) that wild types are now found in the Eastern Sahara-Nile area, Lake Chad, and the inland Niger delta, and that the earliest uncontested evidence of crop cultivation in North Africa comes from Mauritania (Amblard 1996); the Fezzan 2600 BCE-2500 BCE of Libya (van der Veen (1995) neither of which has sorghum. Wendorf and Schlid (1980); A. Gautier (1980, 1984), Kuper (1981); and Pachur and Roper (1984) found wild sorghum in the Nile Valley in anticipation of the creation of the Lake Nassar Reservoir from 8000 BCE to 9000 BCE. Domesticated sorghum has been found in Southern Sudan at the confluence of the Blue Nile and Atbara Rivers at the Shaqadud cave complex which are dated to 2173 BCE +/- 83 years and 2109 +/-65 years by L. Constantini (1983) (1500 years before Stemler (1976) and Haaland (1987). Klee and Zach (1999) claim 4000 years of domestication of grasses and tending of wild grasses in Northeast Nigeria at early Neolithic sites. Neumann (1999) argues for the spread of African domesticate agriculture from Saharan out-migrations starting 3000 BCE. G. Connah (1981) documents native versions of modern African domesticated crops and useful trees in savannas of the Sudan, the Sahel and Lake Chad. A. E. Close and F. Wendorf (1992) discuss some of their relevant excavations. Rowley-Conwy has reported samples of domesticated Sorghum from 1000 BCE that become abundant by 800 BCE, and historic attestations from the 700s BCE by Assyrians in Egypt. The Middle Nile Valley has no date from about 3000 BCE to 1000-500 BCE, but older finds appear consistent with wild type sorghum in high frequency but low volume. Neumann (1999) states that as late as 2700 BCE, the residents of the Lake Chad basin relied on wild millet and other wild grasses rather than domesticated cereals with gathering of wild plants forming a dominant part of the economy that also produced pottery impressed with wild grass patterns. Dates of Arabian and Indian finds as far back as 4000 BCE (Qurm, wild); Hili (3090 BCE (repudiated by authors) and 2500-2400 BCE, wild), Pirak (1600 BCE, domesticated). Inamgao in Deccan India ca. 1600 BCE-700 BCE has a variety of foreign plants some African, but not sorghum and only in a minor way per Kajale (1977, 1991). Rojdi, India has samples of both sorghum and pearl millet as far back as 2000 BCE in some abundance although Weber (1991) and Meadows (1996) have somoe doubts about some species identifications. Hulas, India has both African and Near Eastern crops in association with Late Harappan (1700 BCE to 1000 BCE) cultural finds.

The book recites at 64-66 that Zhijun Zhao (2000) reports sorghum finds in China from 5000 BP (Donghuishan, Ganzu Province), 3000 BP (Niazipoo, Shaanxi Province) and the Western Zhou dynasty 1100-771 BCE (Sanlidun, Jiangsu Province) while Underhill (1997) makes a non-definitive claim for sorghum in China as early as 3500 BCE to 3800 BCE in the East Central Yellow River Valley The early sites all involve Northern China's Yellow River and not the Southern Chinese Neolithic. To find an African crop in China so early, and before any parallel instances in India is remarkable.

The book also notes that R.H. Meadow (1996) finds African domesticated sorghum in the Indus River Valley in the late third millennium and early second millennium. Kajale (1991) discusses a possible later arrival; A. de Maigret (1986); Mercy (1991); and Clieuziou and Tosi (1989) discuss a possible late third millenium arrival. Doggett (1988) suggests a route from Somolia and East Africa via Aden to Baluchistan continuing Sumerian trade routes in place in 2800 BCE. Constantini (1984) and Potts (1994) document evidence of trade between Yemen and Ethiopia/Sudan in the Bronze Age from African origin botanical remains. Edens and Wilkinson (1998) see a Bronze Age complex civilization in the Yemen highlands ca. 3600 BP-2600 BP possibly as a Sabaean civilization precursor with strong ties to the Horn of African and Ethiopia and Punt, even pre-Iron Age. Pickersgill (1983) (based on historical sources) likewise argues for a pre-second millenium B.C.E. trade route connecting Yemen to East Africa and on to India. Fattovich (1997) argues that the trade is at least as old as the Iron Age. Harappan trade with the Sumerians is documented prior to 2500 BCE.

The book reaches some overall analysis at 70+ with domesticated sorghum speculated to have been used in the Nile earlier than any finds to date, and a possible back migration of a second domesticated version from India also considered a provacative, but real possibility. The Tarim basin is identified as a fruitful place to look for sorghum trade with China. The Sahara may have been more of a barrier than it is now prior to camel domestication. A rapid demic diffusion of sorghum cultivation is offered as an explanation of why it is hard to localize its geographic origins between the West Africian Sahel and Sudan. New discoveries about Bronze Age Yemen and Southern India that are older than current finds are anticipated. F. Allard (1998 Int. J. Hist. Archaelogy, 1999 Antiquity) is looking at these nearly contemporaneous with India connections from China to Africa that could have been by land or sea with a land route somewhat favored. The book predates ancient DNA evidence from these plants or even really definitive modern phylogenies of old strains. At 74, the book references (without citation on that page) an old Indian tribal story about an African queen or king sending a group of people to colonize and trade with the Deccan pennisula.

More On Ethiopian Agriculture

More on early Ethiopian agriculture here (citing Blench, Enset culture and its history in highland Ethiopia):
Cultural and linguistic evidence concerning the origin and distribution of enset culture seem to point generally in the same direction. Enset was part of a widespread and ancient system of cultivation of vegetative crops formerly distributed much more widely through the Ethiopian highlands. The main cultivators of enset were Omotic-speakers, though it was probably adopted early by some groups of Cushitic-speakers.

However, when the Ethio-Semites entered Ethiopia bringing seed agriculture and the plough, enset and other root crops such as yams (Dioscorea spp.) and the Labiates (Coleus spp.) were pushed into residual cultivation, except where the terrain was so highly dissected that ploughing was effectively impossible. In this situation, notably in the southwest, the Gurage Semitic-speakers adopted enset and it became central to their production system, permitting the expansion of population to levels such that no other crop would support comparable densities in similar terrain.

Blench's contention that Cushitic has origins in the Neolithic is also referenced, although the purely linguistic argument (the existence of a shared proto-word for sheep and goats) is rather flacid. Among other reasons, a loan word could easily have accompanied the introduced domesticated animal where ever it went, something observed in the cases, for example, of tea and bananas. Also, if the original domestication of the loan word was into one Cushitic language, and the original borrower of the domesticated animal modified the word in a characteristically Cushitic fashion and was the source of the domesticated animal's spread to with its Cushitically adapted name to other Cushitic populations, the common loan word would even look Cushitic in a particularlized way, and secondary dispersion of ideas and technologies is likely to occur preferrentially with people who speak languages related to one's own.

More On Pearl Millet Domestication

Dorian Fuller (2007) has written the definitive review of what science knows about pearl millet domestication in West Africa, one of the most important African specific domesticates, given current knowledge (comprehensive citations and references to omitted figures not included):

Sites with early pearl millet are numbered: 1, Dhar Tichitt sites; 2, Dhar Oualata sites; 3, Djiganyai; 4, Winde Koroji; 5, Karkarichinkat; 6, Ti-n-Akof; 7, Oursi; 8, Birimi; 9, Ganjigana; 10, Kursakata. Historical sites with pearl millet metrical data: 11, Arondo; 12, Jarma; 13, Qasr Ibrim.

Pearl millet domestication is inferred from two sets of evidence. First, there was loss of natural seed shedding, which is linked to the shift from sessile involucres to development of a non-dehiscent peduncle. This shift is already evident from ceramic impressions of pearl millet chaff by 1700–1500 BC in Mauretania, and slightly later in Nigeria. Pearl millet shows a subtle but clear change in grain shape, becoming apically thicker and more club-shaped than its wild counterpart, i.e. an increased thickness/breadth ratio. However, a major increase in seed size appears delayed. . . .

Early West Africa averages (Birimi, 1700–1500 BCE; Kursakata, 1500–800 BCE) fall in the wild zone although ranges extend into the larger domesticated zone. The earliest finds in India (Surkotada, approx. 1700 BCE) are close to these as are Early Historic (200 BCE–300 CE) Nevasa in southern India. North Indian Narhan (1400–800 BCE) shows a marked shift towards larger sizes comparable with modern domesticates, as does early medieval Qasr Ibrim (Egypt, approx. 450 CE: this find is preserved by dessication and has been reduced to be comparable with carbonized material). Jarma in south-west Libya may show an apparent shift towards somewhat larger grains during the early first millennium CE, comparable with the size found in medieval Senegal at Arundo. Later Medieval Jarma has shifted back towards to near wild size range.

Of note is that early West Africa populations, from the second and first millennia BC, have their averages firmly in the wild size range, although there are long tails of variation that extend into the larger size range (e.g. at Birimi). One of the earliest finds of pearl millet from India comes from Surkotada, Gujarat, approx. 1700 BC, which can be seen to fall with these early domesticated African populations. By contrast, a rather later, Gangetic population from Narhan is markedly larger, suggesting selection for larger-grained pearl millet. In Africa larger grained populations appear in the first millennium AD, represented by finds from Nubia and Libya, as well as Medieval Senegal. However, the continued small-grained populations in Early Historic South India (Nevasa) and apparent reversion in later Medieval Libya suggests that there may be factors that work against gigantism in pearl millet, and in the absence reinforcing selection populations may tend towards the smaller size ranges. . . .

As both Libya and South India lack wild populations, this cannot be attributed to cross-pollination with wild types. There may be some constraints particular to this crop, as one experiment indicates that optimal germination occurred under higher temperatures that resulted in lower average grain weights. In addition, pearl millet involucres are polymorphic in grain count with the vast majority producing two grains, a large minority with one larger grain, and a further minority producing 3–9 grains, which are necessarily smaller. Thus, selection for higher grain counts, and more reliable germination, might conflict with selection for larger seed sizes. Nevertheless as a working hypothesis, I would propose that, as with pulses, there is a deeper burial threshold that selected for gigantism in pearl millet at some times and in some locations. In that regard it might be noted that the larger grain populations in Libya and Nubia, like that in Gangetic India, are associated with more intensive plough cultures, whereas ards were not present in West Africa and may have declined in post-Garamantean Libya. Thus, we can hypothesize that large-grained varieties evolved under plough systems and then dispersed back to West Africa at a later date. If so, this would imply separate events of grain enlargement in India and north-eastern Africa. While initial cultivation must have selected for non-shattering, and slight changes in grain weight and shape (the club shape), serious gigantism may have required a stronger selection pressure and therefore evolved later: a millennium or more later in India, and two millennia later in Africa. . . .

[W]e would predict that pearl millet cultivation began by 3200–2700 BC.
Even if pearl millet was cultivated as a crop a thousand years before the date estimated, this pillar of locally domesticated crops in Sahel agriculture is still domesticated long after the Fertile Crescent package of crops arrived in the parts of Africa where they were viable, and probably after Sorghum as well. Fuller's account also suggests that Pearl Millet may have made made its way to India almost as soon as it had been domesticated in Africa, rather than in a transfer of a crop which had long been cultivated there, although the possibility of a reversion to a wild type under selection pressure in an increasingly arid Sahel environment is also a possibility.

No comments: