The research was funded by the Wenner Gren Foundation, grant number 5777; the
National Science Foundation, grant number SBR-94–08655; and the Department of
Anthropology at the University of California at Davis. I thank Peggy Barlett, William
G. Davis, Leslie Gray, Johann H. Hesse, Michael Kevane, G. William Skinner, and
Anita Spring for their critical advice. A preliminary version of this chapter, entitled
“Recent Research in Ghana and Zimbabwe, ” was given at the conference sponsored
by ODI, in London, 26–27 June 1997. Many of the participants provided valuable
Mampurugu is the land of the Mamprusi people, who live in the northern
section of the Northern Region of Ghana.
The Ghana Cotton Company measures out planting areas in units of 75
yards by 75 yards, or half a hectare. This unit is known as an acre in the local language. Other cotton companies operating in the region use units of different sizes
(for example, the major competitor to the Ghana Cotton Company uses units of 60
yards by 60 yards). This makes comparisons problematic. The units referred to in
the text are Ghana Cotton company units, equivalent to half a hectare.
See Poulton (1998) for more details about the various cotton companies operating and the nature of the contracts. His macrolevel study, which essentially
looks at the problem from the cotton companies' perspective, complements this village-level study. We reach similar conclusions, except that Poulton found few
women growing cotton. His sample was taken from the Dagomba area, closer to
Tamale, which differed in both ethnicity and population density from my study area.
The mean income per unit was 43,000 cedis in 1994, equivalent to approximately U. S. $43. A bag of maize or millet equal to one month's staple starch
for a conjugal monogamous household cost about 20,000 cedis in December 1994.
The mean income was calculated using financial data for 184 farmers who contracted with the Ghana Cotton Company. The distribution is skewed toward the low
end, with a standard deviation of 26,000.
See Porter (1995) for the folly of this decision.
The question of how the Mamprusi have maintained control of their land
can only be answered by considering the history of the region. Such a discussion is
outside the scope of the chapter, but it should be noted that the population of this
region was reduced by the demand for slaves, and slave-raiding continued until the
end of the nineteenth century. Thus people, not land, were a source of power. Further, colonial control of the region did not occur until the beginning of the twentieth century, when political support for expatriate plantations was waning (Hogendorn 1995:53). The harsh climate and the plain landscape was unappealing and
frequently fatal to the British, and by 1931, out of 107 non-Africans living in the
entire northern territories, only three lived in Mampurugu (Kuczynski 1948:419).
There was neither scarcity nor colonial alienation of land.
See Wilson (1998) for a full description of the study village and methods.
Cotton was not the main focus of the research, but residence in a farming village
fosters interest in farming.
Using the cotton company tractors also frees the bullock teams for other
work, either for the owner or for hire.
There is also the problem of “ghost farmers, ” that is, signing up farmers
who do not exist for the purpose of siphoning off inputs. Such fraud, however, is
perpetrated by the agents, not the farmers. Agents sign up individual farmers, not
households, and will only sign up those who are able-bodied and have land. Few
farmers are permitted to contract for more than one or two units, limiting the inputs
they receive (see Table 13.4).
Net payment data were not made available by the Ghana Cotton Company.