The clash of ontologies and the problems of translation and mutual intelligibility

We face interlocking questions: (a) of translatability/mutual intelligibility and (b) ontology/reality—that is, what is there to be understood. On (a) I shall take a positive line, combining the following points: (1) no translation is ever perfect and complete, all are provisional and revisable; (2) there is indeed no perfect, complete, mutual understanding, even when all interlocutors share the same natural language. On the other hand, (3) some understanding is always possible, even across divergent systems, and even across incommensurable paradigms, even if (4) there is no neutral vocabulary in which it can be expressed. This depends (5) on allowing that the terms in any language exhibit what I call “semantic stretch.” I shall illustrate these points with ancient Greek and Chinese examples and spell out the implications for our current anthropological and philosophical debates on my second group of questions, (b).