Consider the following FOL sentence:
phi=existsxforallyexistsz((x=y)lor(P(x,y,z)landlnotP(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