sci.logic
[Top] [All Lists]

Re: Proof of finite axiomatizability

Subject: Re: Proof of finite axiomatizability
From: Aatu Koskensilta
Date: Fri, 27 Oct 2006 06:20:01 +0300
Newsgroups: sci.logic
William Elliot wrote:
On Wed, 25 Oct 2006, Atreides wrote:

F is a set of sentences: { A_1, A_2, ...} s.t. {A_1 .. A_i} |/= A_i+1 (
A_i+1 is not a logical consequence of previous A_i's).

|- means logical consequence, provable.

T |= A means that A is a logical consequence of T, i.e. A is true in all models of T. T |- A means that A is provable from T. In case of first order logic these two notions coincide.

As David already pointed out, from Atreides's assumption it doesn't follow that F is not finitely axiomatizable. For example, let c be some fixed constant symbol and let A_1 be the conjunction of the usual axioms for the successor function and let A_i for i > 1 be SS...S0 =/= c with i-1 S's. The finite set of sentences consisting of the usual axioms for the successor function together with

 0 =/= c
 for all x(Sx =/= c --> SSx =/= c)

then has the same deductive closure as {A_1, A_2, ...} even though {A_1, ... , A_i} |/- A_i+1 for all i.

How to prove that F is NOT finite axiomatizable?

{ (Ex)~(x = x) } is a finite collection of statements that will prove
every statement in F.

Certainly. But it isn't an axiomatization of F unless F is inconsistent.

--
Aatu Koskensilta (aatu.koskensilta@xxxxxxxxx)

"Wovon man nicht sprechen kann, daruber muss man schweigen"
 - Ludwig Wittgenstein, Tractatus Logico-Philosophicus

<Prev in Thread] Current Thread [Next in Thread>
Privacy Policy