1
$\begingroup$

Let $\Sigma$ be any set of propositional sentences. If $\Sigma$ is finitely satisfiable, then for every propositional sentence $\theta$, either $\Sigma\cup\{\theta\}$ is finitely satisfiable or $\Sigma\cup\{\neg\theta\}$ is finitely satisfiable.

Can someone please tell me the error in my proof?

Suppose $\Sigma$ is finitely satisfiable, and $\theta$ is an arbitrary sentence. Let $\Pi\subset\Sigma$ be any finite subset, then there exists a truth evaluation $e$ such that $e(p)=1$ for every $p\in\Pi$. Now either $e(\theta)=1$, or $e(\neg\theta)=1$, so either $e$ satisfies $\Pi\cup\{\theta\}$ or $e$ satisfies $\Pi\cup\{\neg\theta\}$.

But $\Pi$ was chosen arbitrarily, hence the result follows.

$\endgroup$
2
$\begingroup$

The problem is this, suppose that for $\Pi_1$ and $\Pi_2$ you have two assignments, $e_1$ and $e_2$ which give "True" and "False" to $\theta$ respectively.

You have no control over this, because these were taken arbitrarily. So you haven't shown that $\Pi_2\cup\{\theta\}$ is satisfiable. What you have shown is that every finite set of propositions can be satisfied with either $\theta$ or its negation, which is trivially true.

What you need here is to argue that there is a way to satisfy all the formulas in $\Sigma$ at once. And luckily, the condition you have is exactly enough to argue that. Compactness is your only friend.

$\endgroup$
  • $\begingroup$ Thanks a lot! I never thought it through until now. $\endgroup$ – Sid Caroline Aug 6 '17 at 11:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.