On maps the Sahara looks like a dangerously empty space, a gap between places. But the world’s largest hot desert is in fact a place of transit, from nomadic pastoralism through cigarette smuggling to oil transportation; the latest example is Colonel Gha-daffi’s “Great Man-Made River,” an underground pipeline delivering - so far - three million cubic meters of water a day 1,200 km from the desert aquifers in southern Libya to the thirsty coastal belt in the north. This is nothing new: travel and contact across the desert between the Maghrib and Egypt goes back at least as far as the thirteenth century bc, when “Libyans” appear in tomb frescoes at Amarna, and Libyan tribes are reported to attack settlements in the Nile Delta (Brett and Fentress 1996: 22). Like the sea, the desert has always been a gateway between North Africa and the rest of the world rather than a barrier between them (Liverani 2000a).
As on the coast, organized networks for the exchange of trade and technology across the Sahara seem to date from the early first millennium bc, and they coincide with the beginnings of urbanism and agriculture in the desert. As David Abulafia has noted “[d]eserts, like seas, have their islands, or oases” (Abulafia 2005: 65), and one of the best examples of this is the Fazzan region of southern Libya, inhabited in antiquity by people known to the Greeks and Romans as Garamantes. Far from the picture painted of these people in sources such as Herodotus, who sees them as so perverse, uncivilized and unlike “us” that even their cows graze backwards (4.183), a major recent archaeological project there has revealed a powerful Saharan state based on cities, settled agriculture and long-distance trade (Mattingly 2003). It has often been thought that the evidence for intensive ancient use and occupation of the desert means that the region must have been wetter then, but new research on climate change has made it clear that the move to urbanism, agriculture and trade in the Fazzan coincides instead with the final desertification of the region. It seems that the change in lifestyle from pastoralism to sedentary agriculture was due not to a kinder climate, but precisely to the need to respond to an increasingly hostile one (Liverani 2000a: 19). This creative response made good use of contacts through and beyond the desert.
The first permanent settlements in the Fazzan date from the ninth century BC. These are easily defended hillforts overlooking the valley of the Wadi al Ajal, later the heart of the Garamantian state. Although there were probably still springs in the valley at this time (Brooks et al. in Mattingly 2003: 67, 73), wells helped to water new winter crops (wheat and barley) characteristic of the Near East. Paleobotanical investigation now suggests these were imported as a package from the Nile Valley (R. Pelling 2005: 401).
This link with the Nile Valley was not limited to imports of agricultural technology; commerce quickly became a crucial component of the local economy. The Fazzan has the largest group of oases in the desert and has long been recognized as a crucial crossroads in medieval and early-modern Saharan trade routes, but it now seems likely that this role started much earlier. In the fifth century BC Herodotus described the peoples of inland Africa, including the Garamantes, as living at ten days distance from each other in places with palm trees, springs and salt mines, along a ridge of sand stretching from Thebes in the Nile Valley to the Pillars of Hercules (4.181-85). Notwithstanding the ethnographic fantasies which decorate the account, Mario Liverani has recently forcefully re-put the case that this is a schematic description of a contemporary caravan trading route leading south-west from the Nile valley. Rejecting earlier assumptions that the later stages were fictional, he has traced Herodotus’s oases at least as far as the Niger Bend (Liverani 2000b). This was doubtless not the only caravan route in operation: Herodotus also mentioned connections between southern oases and the Mediterranean coast, specifically between the Garamantes and the “Lotus Eaters” of Tripolitania (4.183), a 20-30-day caravan ride to the north (Mattingly 2003: 7).
What was being traded along this route.? Liverani suggests that (as in the medieval period) the basic exchange to the south was salt for gold and slaves from the subSaharan kingdoms such as Ghana which were developing at much the same time. These were in turn were exchanged to the north for olive oil and luxury items from the Mediterranean (Liverani 2000a: 26, 2000b: 507-8). Although this model is not universally accepted, Herodotus’s account certainly emphasizes the importance of salt mining in the oases, and he mentions Garamantian slave-hunts among the Ethiopians to the south (4.183). Garamantian links to Tripolitania also point to a slave trade: Lepcis Magna was later heavily associated with slaving (di Vita 1982: 588-95, Braconi 2005), and there is a hint in Homer that such trade began early on the coast, with the tale of a Phoenician attempt to sell Odysseus as a slave in Libya (Od. 14.287-300). Trade in gold is also likely: Herodotus does discuss Phoenicians trading for gold somewhere on the Atlantic sea-coast of Africa (4.196), and gold artifacts have been found in Garama (Mattingly 2003: 360). The gold mines of ancient Ghana were certainly exploited by the fourth century ad, when the amount of gold in African coinage increases dramatically and the Romans begin to demand taxes in gold (Garrard 1982, A. Wilson 2007), but the small-scale export of gold may have started long beforehand.
Whatever the exact nature of trans-Saharan trade, the Garamantes did well out of it: they moved their settlements down into the valleys in the fourth century, suggesting that they no longer needed to fear attacks from elsewhere, and founded the impressive oasis city of Garama, which remained their capital until the final decline of their civilization c.500 ad. Punic and some Italian pottery of the last centuries bc is found at Garamantian sites, and soon Mediterranean goods including glass and faience are very common in the Fazzan, as well as amphorae which would have carried oil and wine; amphorae from Tripolitania are found as far south as the oasis of Ghat (Mattingly 2003, Liverani 2000b).
Technological innovation as well as trade continued to travel along the east-west caravan route: the foggara, a complex irrigation technology originally developed in Persia which reached the Fazzan via Egypt in the fourth or third centuries bc (Wilson in Mattingly 2003: 261-65), and new summer crops from sub-Saharan Africa including sorghum and pearl millet were introduced in the late first millennium, both enabling an intensification of the agricultural regime. It seems that the Fazzan was not simply a consumer of technology: Andrew Wilson has suggested that Roman-period foggaras in southern Tunisia, the Aures mountains, and the Touat region of Algeria are the result of technological diffusion northwards and westwards from the Fazzan (Wilson 2005, 2006). Similarly, pearl millet did not reach the Nile valley until the early first century ad (R. Felling 2005), perhaps via the same route.
As we have seen above, trade and technology are not all that travel across seas and deserts; networks of culture and religion often map the same routes (Abulafia 2004). Zeus Ammon, for instance, was worshipped at Thebes in Egypt, and by the mid-sixth century he had a famous shrine at the oasis of Siwa, one step along the caravan route (Hdt. 2.55 for the traditional explanation). A step further again the cult of Ammon persisted in Augila until the time of Justinian (Frocop. Aed. 6.2.16), and in the first centuries bc/ad. Roman sources firmly associate him with the Garamantes as well (Verg. Aen. 4.198, Sil. Pun. 2.58, Luc. 9.511-12, with Mattingly 2003: 89). Beyond religion, the eclectic architecture of the temples and tombs of the Garamantes, with Ionic and Corinthian columns and engaged pilasters (Mattingly 2003: 20, 189-92), recalls that of the Hellenistic Mediterranean as well as that of carthage of the same period.
Overall, the new excavations among the Garamantes highlight the extent to which this region, apparently beyond the ken of classical ethnographers, was, like the Funic coast, a booming center in intensifying long-range networks along which people were exchanging not only commercial but also technological and cultural capital.
World History
