Abstract

I consider the influence of the language used in anthropological analysis (the metalanguage). If in principle there are at least as many anthropologies as there are languages, then we must allow the possibility of seven thousand or so more or less incommensurable anthropologies. However, incommensurability need not follow: not only can sparse theory aid comparison but it can also help establish partial (incomplete) commensurability. The parsimony of sparse theory allows for clarity even when dealing with vague ideas. If to achieve this some fashionable terms of analysis have to avoided, this is a price worth paying. In sum: we should all say less more clearly.