of conquest by the army of Menelik of Shawa in the late nineteenth-century which is by far the most critical and perverse event in Sidama history.”23
Not only family members took advantage of the more vulnerable of the children. Fayissa Murki (see appendix B; narrative 14), another child rescued off the later dhows, started off in a normal and secure family environment, living with his parents and two sisters. His father owned a small piece of land, about twenty head of cattle, and some goats in the village of Alle in the country of Danno. However, this familial stability was nonetheless fragile. While playing near his home one afternoon, a neighbor approached and asked Fayissa to accompany him to his home nearby. Fayissa complied, but once there, he must have been terrified when the neighbor detained him in his house. That night, the neighbor took Fayissa to a nearby slave market and sold him to a group of Atari merchants, who, in turn, took him to a place called Dalotti in Tigre country, where he became entrapped in the remorseless slave traffic destined for the Red Sea ports. We can presume the neighbor knew that the newly enslaved person would never be seen again.
Establishing household size, including the average family size among the general Oromo population in 1888 or 1889, is a challenging task, and estimates of the average number of children per family in sub-Saharan Africa during precolonial times have a broad range. Economic historian Gwyn Campbell, in his article on the precolonial historical demography of Madagascar, cites an average of “between 4.9 and 5.25 children per household estimated for sub-Saharan Africa in general during pre-colonial and recent times.”24 While scholars like Elisée Reclus have provided approximate population totals for the region in general and for the Oromo population in particular for 1885 (see map 1.2 and discussion on page 19),25 detailed demographic data for the Oromo in the late nineteenth century are elusive. Geographer R. T. Jackson, in his research on the influence of population size on market size and density in southern Ethiopia, estimated an average family size of five among the inhabitants of this region bordering present-day Oromia.26 This figure at least gives a rough yardstick by which to suggest a regional estimate of trends relating to family size. The average number of children per Oromo family in this study is snugly commensurate with this figure.
GRAPH 2.3. Family sizes of boys and girls (source: Sandra Rowoldt Shell).
The histogram above shows the family sizes of all the Oromo children (see graph 2.3). Gender differences in family size patterns were minor.27 Most of the boys and girls (80.2 percent) lived in families of between three and nine family members. A small proportion of children (3.5 percent) either had no memory of any other family members (Gilo Kashe, Faraja Jimma, and Galgalli Shangalla; see appendix B; narratives 19, 12, and 54, respectively) or were full orphans with no siblings (Meshinge Salban and Isho Karabe; see appendix B; narratives 59 and 23, respectively). These are included as single-member families.
At the upper end of the scale, 15.1 percent of the children had families of between ten and fourteen family members. However, even the Oromo families with larger than the average number of siblings in the home did not provide the sort of protection to an individual child that might have been imagined. Safety in numbers carried little weight against the forces that were driving the internal and export slave trade.
Orphanhood
UNICEF defines an orphan as “a child who has lost one or both parents,” so those Oromo children who lost either their father or their mother were ranked as orphans.28 Nearly one-fifth (19.77 percent) of the children (11.63 percent girls and 8.14 percent boys) either could not remember one or both of their parents or did not mention them in giving details of their family composition. These figures tell their own story. Of the majority 81.2 percent who could supply these details, as many as 12.8 percent of the children were full or double orphans when they were captured—with boys accounting for 3.5 percent of that total and girls nearly three times that proportion at 9.3 percent.29
These figures compare negatively when ranked against comparable statistics from twenty-first-century Ethiopia, Oromia, or even South Africa, where the death toll of men and women between the ages of nineteen and forty-four—many of them victims of the massive HIV/AIDS pandemic—is distressingly high. Graph 2.4 shows the clear discrepancy between the percentage of orphans among our group of Oromo children and the prevalence of orphanhood in any of the regions selected for comparative purposes.
A total of 12.8 percent of the Oromo slave children were full orphans, 17.4 percent were paternal orphans, and 1.2 percent were maternal orphans. In aggregate, 31.4 percent of the children had lost either one or both parents. This figure falls just short of three times the aggregated national Ethiopian orphan prevalence percentage of 11.9 percent in 2005, more than three times the aggregated national Ethiopian orphanhood total in 2007, and 3.27 times the aggregated orphanhood total in Oromia in the same year. The prevalence of orphanhood among the sixty-four Oromo children is therefore significantly higher than might be anticipated.30
Fred Morton believes that the deaths or absence of the parents of the East African slave children of his study certainly rendered them unprotected and at greater risk.31 Raiders probably regarded orphaned Oromo children as easier quarry without the usual protection of a full family-unit complement. They may also have regarded orphans as generally more acquiescent and less likely to run away, given that they had no families to which they could return.
Orphanhood in Ethiopia is an old and still-growing phenomenon. Laura Camfield, a social anthropologist, has written recently that parental mortality, more specifically maternal mortality, is increasing in Ethiopia. She ascribes this not only to the growth in the incidence of HIV/AIDS, but also to high maternal mortality, acute illness, and the effects of drought, famine, displacement, migration, and conflict.32 The conditions experienced by the Oromo children in the late 1880s were not dissimilar—acute illness, drought, famine, displacement, migration, and conflict.
Of the fifty children rescued aboard the Osprey in September 1888, 28.4 percent were orphans. This figure contrasts starkly with the 47.4 percent of orphans aboard the dhows who were rescued and liberated eleven months later on 5 August 1889. While it is not possible to draw a straight line of causality, the considerably higher prevalence of parental mortality in the later, non-Osprey group could be considered consistent with the increased impact of the drought and famine by the end of 1888.
GRAPH 2.4. Orphanhood of the Oromo children vs. children in South Africa, Ethiopia, and Oromia (source: Sandra Rowoldt Shell).
Were children also at risk if they lived with their mother but there was no father in the household? According to Abbi Kedir and Lul Admasachew, “There is a great degree of stigma attached to children who are raised without a father figure” in Ethiopia.33 Without the paterfamilias, the children were clearly vulnerable. There was only one example of maternal orphanhood. That all other Oromo orphans were paternal or full suggests the significance of the presence of the father for the protection of the family. Nearly one-fifth (17.4 percent) of all the Oromo children were paternal orphans, a total that was distributed more or less evenly between the boys (8.1 percent) and the girls (9.3 percent). Male relatives (uncles and elder brothers) were again quick to intervene following the deaths of the fathers of six of these paternal orphans, five boys (Badassa
