222 Nino Luraghi
anthropologists see as the divides that separate from one another the various dominant
methodological paradigms such as evolutionism, functionalism, structuralism, and so
on (Amselle 1990). The continuity of ethnological reason is expressed precisely by its
ongoing fixation on boundaries and on cultural difference, which in its turn is made
necessary by the inherently comparative nature of the anthropological approach. In
Amselle’s words (quoted from Amselle 1998: x), “[I]n order to be integrated within
comparative processes, whether inductive or deductive, the contents of anthropological
categories must be abstracted from their historical contexts. Comparativism assumes
diversity, and thereby guarantees that diversity will be found.” In this focus on cultural
difference as opposed to cultural continuity, Amselle and other anthropologists see,
not without reason, a heritage of the colonial system, for which it was necessary to
find ways of subdividing the subjects in order to integrate them into the colonial
administrative structures: “The invention of ethnic groups is the joint work of colonial
administrators, professional ethnologists, and those who combine both qualifications”
(Amselle 1998: 11).
With a striking feedback effect, the categories produced by the Western social sciences
have been appropriated by the colonized themselves, who in some cases turned them into
the foundation of their own political agenda, in a political framework typically derived
from first-world nationalism. Perhaps the most striking case of this feedback effect is
the emergence of ethnic boundaries in Rwanda (Newbury 1987). Before becoming part
of German East Africa at the end of the nineteenth century, Rwanda was a highly cen-
tralized kingdom, homogeneous in terms of language and religion. Its population was
roughly divided in three groups based on wealth and main occupation, the one being,
as is often the case, a function of the other: the Twa, semi-nomadic hunter–gatherers
inhabiting the mountainous areas; the Hutu, a Bantu name that means “the ruled, the
subjects,” who were peasants; and the Tutsi, the wealthier cattle-owners who formed
the ruling elite of the kingdom. It is fairly clear that growing or declining economic
fortunes and changes in life style could move families from one group to the other. In
other words, the three groups are best described as social classes. Both the Germans
and later the Belgians, who took over Rwanda from the Germans after World War I,
exploited this social system by ruling the country vicariously by way of its indigenous
elite. However, in order to make sense of the social stratification in the framework of
their own episteme, the Germans and the Belgians started speculating that the Tutsi,
who seemed to them more civilized than the Hutu, had to be of a different racial origin.
As Amselle would put it, once ethnologists started looking for differences, they found
them: phenotypical stereotypes were developed, in which the Tutsi scored high in all the
features that distinguished the Europeans from the Africans, and theories were devel-
oped that made of them immigrants of Kamite stock, or even Aryans. When the Belgians
moved from indirect control to direct colonial administration, they literally reified these
ethnic speculations, treating Hutu and Tutsi as two discrete groups and enforcing the
distinction: the Belgian census of 1930 classified the people of Rwanda according to
their ethnic origin, distributing identity documents that defined them, once and for all,
as Hutu or Tutsi, even though the criteria for ascription were so flimsy that, in the end,
the colonial administrators, incapable of telling Hutu from Tutsi and not trusting the
self-description of the actors, identified as Tutsi whoever owned more than 10 heads