The Environmental and Social Costs of Improvement 29works rapidly deteriorate, accelerating erosion instead of reducing it. If perform-
ance is measured over long periods, the results have been extraordinarily poor for
the amount of effort and money expended (Shaxson et al, 1989; Hudson, 1991;
Reij, 1991).
It is well established that poorly designed structures cause erosion. Yet through-
out Africa, little account has been taken of how more terracing can lead to more
erosion. In the early 20th century, erosion in Lesotho was not a serious problem in
cultivated fields, because grassed field boundaries were well developed and main-
tained (Showers, 1989). Despite this indigenous practice, contour banks were
installed. Local people did not approve, because they reduced the size of fields, and
either breached or the outlets developed into gullies. The administration attrib-
uted these gullies to ‘unusual weather’ (Showers and Malahleha, 1990). Elsewhere
in southern Africa, the first anti-erosion measures introduced in the early 1930s
were large ridge terraces and bunds. But these imported measures disturbed natu-
ral patterns of drainage and permitted storm water to break through at vulnerable
points. Careless construction made them susceptible to bursting and locals came
to believe that ‘gully erosion was caused by the government’ (Beinart, 1984).
Narrow-based terraces were introduced into Kenya from the US in 1940
(Gichuki, 1991). For 15 years they were widely used. By 1947, some 4000 hectares
were being protected each year and this rate continued until 1956–1957. But these
terraces filled up with sediments too quickly, were impossible to maintain and
even began to aggravate erosion. And so, by 1958, the number falling into disre-
pair was exceeding new construction. By 1961, some 20,000ha had fallen into
disrepair. Eventually, the authorities recognized the problems and L. H. Brown,
the chief agriculturalist, issued a memorandum in 1961 saying that ‘narrow-based
terraces should be abandoned as policy... we should move to strips of vegetation,
preferably grass’ (in Wenner, 1992).
Bad contour ridging in the 1960s was worse than none at all in Zimbabwe
(then Rhodesia), where farmers say the compulsory construction of ridges caused
siltation of rivers. The ridges connected whole fields and drained in a single drain-
age line, so during large storms, concentrated water into powerful and fast-moving
bodies that caused great damage (Wilson, 1989). The same thing has occurred
with cut-off drains in Kenya. Their function is to intercept and divert storm water,
but many were constructed in a way that caused erosion. ‘The most severe mis-
takes were that cut-off drains were laid and constructed on the wrong sites. They
were designed with steep gradients... The water is discharged into gullies which are
deepening. The channel ridges were bare and cut-off drains were not supported by
other structures below them... All these factors have made the structures more
dangerous than useful. More problems were created. Gullies have widened, soil
was eroded and crops destroyed’ (Hunegnaw, 1987).