This paper takes into account the Wittgenstein’s idea of family resemblance as a particular crisp relation between some fuzzy sets, that is, between some representations of predicates use from its use. It is shown that all uses of the same predicate actually do have some kind of family resemblance, that some pairs of predicates cannot, and how the predicates migration between different universes of discourse can be related with family resemblance. A possibility of sketching all that through Galois Connections is also considered.