The general aim of the article is to show how translations can be used to study polysemy. The particular research problem is to investigate the extent to which see has extended its meaning into new semantic domains on the basis of its translations into other languages. The data are taken from the English-Swedish Parallel Corpus. At a general level, the English verb see and the Swedish verb se are similar to each other as might be expected from the fact that they are etymological and semantic cognates. Both refer to the situation of perceiving with one's eyes but the verbs diverge in their further developments. It is shown that see extends its meaning into semantic domains such as evidentiality, aspect, causation and discourse, and that the development depends to a large extent on the context, in particular complementation.