Monday, 1 December 2008

lo.logic - Models for the given FOL statement

Consider the following FOL sentence:



$phi = exists x forall y exists z ((x=y) lor (P(x,y,z) land lnot P(y,x,z) ) $



It can be proven that for any natural number n > 0 there exits a model of size n for the above sentence. (Please correct me here if I am wrong. This should be provable using induction.).



Now imagine a FOL sentence that does not use = (and similar) predicate. And if such a sentence has a model of size n can I claim that the sentence will essentially have a model of size n+1 ?

No comments:

Post a Comment