ABSTRACT
Cross-cultural research of any kind cannot afford to ignore the problems posed by semantic differences between languages. These problems are particularly pertinent for psychology, given that information about other people’s mental states is inevitably mediated by language. Unfortunately, however, social scientists often regard the problem of translation as a mere methodological nuisance—as something to be “gotten around” so that they can move on to implementing familiar research techniques, rather than as a profound epistemological and conceptual issue deserving of sustained and focused attention. At the same time they underestimate both the scope of semantic variation between ethnopsychological lexicons, and the hazards of uncritically using English as the metalanguage of cross-cultural description.
