frames the journeys of the Oromo captives. The country of Ethiopia boasts as many as half of the highest peaks in Africa—including Badda, a volcanic peak in the Bale mountain range soaring to 4,200 meters (13,650 feet). The country also lays claim to some of the lowest-lying land on earth, including the below–sea level desert on the edge of the Afar or Danakil Depression, through which all the children passed on their way to the entrepôts. The Bale Mountains are separated from the former Abyssinian highlands by the Great Rift Valley. The land of the central plateau, intersected diagonally by the Great Rift Valley, falls away—in places sharply—to the lowlands of the north, west, south, and east.
The lateral escarpments of the rift drop down and diverge to the northeast and northwest, where they transect the Afar or Danakil Depression toward the Red Sea coastal regions, including the children’s destination entrepôts of Araito (Rahayta) and Tajurrah (Tadjoura).3 The Danakil Depression constitutes one of the hottest and driest places on earth and is Ethiopia’s lowest point at 120 meters (393.70 feet) below sea level. Archaeologists consider the entire rift area one of the cradles of humankind.
Mapping the captives’ places of origin against the topography of Ethiopia shows that an overwhelming majority came from highland regions, with their homes dotted along the ridges of the mountain ranges (see map 4.1).
Symbols have been attached to the girls’ homes to highlight gender differences in the children’s geographic and orographic distribution. Note that with only two exceptions, the girls’ homes lie to the west of the escarpment, whereas the boys’ homes for the most part follow the mountain ridges on either side of the main Ethiopian rift. These mountain ridges, with their rugged terrain, provided a degree of natural protection from all comers, but they were not sufficient to deter the invading cavalry from the north. Only the homes of Isho Karabe (see appendix B; narrative 23), who lived in the village of Imo in the southeast; Hawe Sukute (narrative 56), from the village of Gani in Garjeja country in the west; and Turungo Tinno (narrative 62), from Saate, a village in the Kaffa country in the south, lay at altitudes of less than 1,000 meters (3,281 feet). Two boys (Amaye Tiksa [narrative 4] and Badassa Wulli [narrative 6]), came from the village of Badda, standing at 3,820 meters (12,415 feet) on the slopes of Mount Badda, a volcanic mountain peaking at 4,200 meters (13,650 feet).
An unanticipated finding was the pronounced gender difference that emerged in respect to altitude. While the average altitude of the boys’ homes was 2,156.52 meters (7,008.69 feet), the girls’ average was almost 400 meters (1,290 feet) lower, at 1,759.33 meters (5,717.82 feet). The boys had a wider altitude range, from a low of 441.36 meters (1,434.42 feet) to a high of 3,819.97 meters (12,414.90 feet). The girls, on the other hand, ranged from 653.83 meters (2,145 feet) to 2,745.22 meters (9,006 feet). The full explanation of this gender difference related to altitude remains obscure.
There may be virtue in remembering that young Oromo girls realized the highest prices in the external slave markets, so traders might have sought out girl children in places that were easier to access than those living in the higher mountainous regions of the highlands. Ease of access meant a quicker journey to the Red Sea. The children’s evidence shows that fewer girls than boys were sold into local servitude initially. Nearly three-quarters (73.8 percent) of the boys were enslaved locally, compared with only half of the girls (50 percent). Some boys spent years—one as long as nine years—in local servitude before being sold into the external network headed for the coast (see pages 84–94 for further discussion).
Graphs 4.1 and 4.2 are based on two different measures of central tendency. The first, a bar graph, is based on the mean, or average, altitude; while the second, a box and whisker plot, is based on the median, or midpoint. The bar graph shows the descending altitude at which each child was captured. This view, not possible with the otherwise effective box plot, provides a statistical cross section of the topography of capture in which the statistically significant gender differentiation is clear.
GRAPH 4.1. Altitude at which each child was captured (source: Sandra Rowoldt Shell).
The complementary box and whisker plots (graph 4.2) give a better idea of the gender differences in altitude range. Here, the median, as the midpoint, is unaffected by the outliers at either end of the boys’ range.
Note that 50 percent of the girls fall within a discernibly lower second and third quartile range than the boys, with no extreme measures at either high or low altitudes. Using both graphs allows us to visualize the altitude range in two different ways and to appreciate the gender discrepancies more clearly.
The Homelands of the Oromo Children
Most of the children (84.88 percent) remembered some details of where they had been living at the time of their capture. The remaining 15.12 percent could not give the name of their village, town, region, or even their country of origin. The linked pie diagram (graph 4.3) represents the home countries of the Oromo children.
As the diagram illustrates, the children originated in twenty-six different principalities. The majority of the children were concentrated in six of these “countries” (as they termed them). The rest of the children were sparsely distributed at a rate of one or two per principality. For clarity, these have been aggregated in the pie within the category of “Other” principalities, which compose a significant geographic spread, as the map showing the children’s places of origin and domicile demonstrates (see map 4.1). This multiplicity of principalities is indicative of the extent of subinfeudation in the region. Some of the monarchies were more prominent and powerful than others; the kingdoms of Jimma, Enarya, Goma, and Guma coexisted, often uncomfortably, in what some might regard as a retrospective federation of principalities.
MAP 4.1. Places of domicile of the Oromo captives (source: Sandra Rowoldt Shell and GIS Laboratory, University of Cape Town).
Further, while there was a limited amount of country-level congruence, with small numbers of children coming from each of six countries, the same congruence was not reflected at the more localized level. With only two exceptions, the children all came from different towns and villages. However, in one case, Amaye Tiksa (see appendix B; narrative 4) was carried off by marauding raiders after a battle with his people in November 1887; while Badassa Wulli (narrative 6) was kidnapped from his home in Badda three years earlier, in May 1884. Similarly, two girls and one boy came from Sayo. Kintiso Bulcha (narrative 25) was snatched from his home after Sidama raiders emerged victorious after having invaded his country and engaged in battle against his chief in October 1888. The two girls, Asho Sayo (narrative 44) and Soye Sanyacha (narrative 60), were both taken as spoils of war by invading Sidama raiders in November 1887. This is possibly the same raid, but each child had a different captor and followed a different route thereafter. The Oromo children had no group experience as captives until they neared the coast.
GRAPH 4.2. Altitudes of places of origin by gender of child (source: Sandra Rowoldt Shell).
