Drennan G, Levett A, Swartz L
Department of Psychology, University of Cape Town, Rondebosch, South Africa.
Cult Med Psychiatry. 1991 Sep;15(3):361-81. doi: 10.1007/BF00046543.
This article examines the social relationships involved in the production of a Xhosa version of the Beck Depression Inventory (BDI). The South African sociopolitical context and the imperatives of the biomedical context are discussed. These play an important role in how the problem of semantic equivalence between translations, and the issue of the imposition of a colonialist, Western psychiatric framework are addressed. After a brief but critical outline of recommended translation techniques, the particular combination of back-translation and committee method employed in this study is presented. An analysis of the social relationships revealed a complex set of interrelated factors affecting the process. The power differential between researcher and back-translators, compounded by their different perceptions of the task, impeded discussion about the instrument. Translators had ostensibly markedly opposed claims to competence to speak for the "true Xhosa language". At a deeper level however, these claims, and indeed the entire translation enterprise, could be seen to be reproducing apartheid structures, regardless of the intention of the participants. The dynamics we describe are particularly salient in the South African context but may also be operative in other contexts.
