I think we're talking in the context of defining the symbolic language, rather than talking about actual sets just yet. So the 'constants' here are the symbols which have a specific fixed meaning in the definition of the language - the membership symbol, the empty set symbol, the equality symbol and the logical symbols.
Any other symbols - x, P, v1 and so on - are (in a sense) variables: their specific meaning must be defined in use, whether they represent set-theoretic entities, predicates or formulae.
Yes - (iii) is saying the formula Q may contain the standard logical notation symbols ("logical constants"), those basic set-theoretic symbols ("primitive"), and any symbols for which proper definitions have already been established ("previously defined"). It looks like a sensible definition for a function of n variables, except I would have expected w also to be allowed as a free variable of Q - otherwise I don't see how (iv) makes sense.
Comments 3
Reply
Any other symbols - x, P, v1 and so on - are (in a sense) variables: their specific meaning must be defined in use, whether they represent set-theoretic entities, predicates or formulae.
Reply
(The comment has been removed)
It looks like a sensible definition for a function of n variables, except I would have expected w also to be allowed as a free variable of Q - otherwise I don't see how (iv) makes sense.
Reply
Leave a comment