ca. 3500 and 1500 (Clark 1988; Fattovich 1988). It is possible that there were some independent attempts to domesticate local plants, as groups living in eastern Ethiopia may have been pre-adapted to food production since the late Pleistocene epoch, about 11,000 years ago (Clark and Prince 1978), and ensete (false banana) might have been domesticated locally in southern Ethiopia at an early date (Brandt 1984).

At least two major centers of peopling arose as a consequence of the new adaptive strategy of food producing. These were located along the northwestern Ethiopian-Sudanese lowlands and on the eastern Ethiopian plateau. The Ethiopian-Sudanese lowlands were occupied by pastoral and agropastoral peoples with an indigenous cultural tradition partly connected with the traditions of the middle Nile Valley (Fattovich 1990a). The eastern Ethiopian plateau was occupied by a pastoral people with Afro-Arabian traditions, as documented by Ethiopian-Arabian rock drawings. By the late third millennium b.c., this population had spread progressively toward southern Ethiopia, northern Somalia, and Eritrea (Cervicek 1971). Rock drawings in the Ethiopian-Arabian style also suggest that peoples with similar cultural traditions were living in eastern Ethiopia and central Arabia from the mid-third to the second millennia b.c. (Cervicek 1979).

An Afro-Arabian interaction sphere most likely began in the seventh millennium b.c. as a consequence of the obsidian trade from Africa to Arabia. In the mid-third to the mid-second millennia, it apparently included the whole region stretching from the northwestern Ethiopian-Sudanese lowlands to the eastern Rift Valley, Djibouti, Aden, and northern Yemen. At this time, the groups living along the Ethiopian-Sudanese lowlands acted as intermediaries between the northern Ethiopian plateau and southern Arabia and the Nile Valley, connecting the Horn of Africa to Egypt. In the mid-second millennium b.c., the interaction sphere was intensified by the development of an Afro-Arabian cultural complex along the opposite shores of the Red Sea, from the southern Saudi coast to the Eritrean coast and Aden. At present, the archaeological evidence suggests that the southernmost limit of Egyptian trade, Punt, was located in the western lowlands and on the northern Ethiopian plateau (Fattovich 1993b). In the early first millennium b.c., with the rise of the southern Arabia states, Eritrea and northern Ethiopia (Tigray) were included in the Sabaean area of influence, and an Ethiopian-Sabaean culture appeared in the region (Anfray 1990; Fattovich 1990b).

From the archaeological evidence, four stages can be distinguished in the development of complex societies in northern Ethiopia (Fattovich 1988, 1994). Chiefdoms appeared in the northwestern Ethiopian-Sudanese lowlands and perhaps on the Eritrean plateau in the late-third to second millennia b.c. as a consequence of the inclusion of these regions in the interchange circuit between Egypt and the Horn of Africa. An Ethiopian-Sabaean state arose on the plateau in the mid-first millennium b.c. when the region was included in the area of southern Arabian political and economic expansion. After the decline of the Ethiopian-Sabaean state in the third to second centuries b.c., petty kingdoms most likely emerged on the plateau, but any archaeological evidence of them is still very scarce. Yet it is known that a form of urban society probably survived in central Eritrea. Finally, a new kingdom, with its capital at Aksum, arose on the plateau at the end of the first millennium b.c. and dominated the region up to the eighth and ninth centuries a.d. Christianity became the formal religion of the kingdom in the fourth and fifth centuries a.d.

So far, no real effort has been made to investigate the dynamics of state expansion in the Middle Ages using archaeological evidence, and scholarly interest in that period has focused mainly on the study of the rock-hewn churches (Anfray 1990). These monuments suggest a cultural continuity between the Aksumite and post-Aksumite periods, up to the thirteenth century a.d., and the distribution of the monuments enables scholars to follow the progressive expansion of Christianity southward in the first half of the second millennium a.d. Development of the Ethiopian state during the fourteenth to seventeenth centuries is still totally unexplored archaeologically.