Is $(\phi\implies \psi) \iff ((\phi\land \psi)\lor \neg \phi) $ in the following hilbert calculus without contraposition or reductio ad absurdum? If so, how would I go about proving it, and if not, how would I prove that it is not derivable?
Inference Rules: Modus Ponens
Axioms:
$\phi \implies (\psi \implies \phi)$
$(\phi \implies (\chi \implies \psi)) \implies ((\phi \implies \chi) \implies (\phi \implies \psi))$
$(\phi \land \psi)\implies\phi; (\phi \land \psi) \implies \psi $
$\phi \implies (\psi \implies (\phi \land \psi)) $
$\phi \implies (\phi \lor \psi); \psi \implies (\phi \lor \psi) $
$(\phi \implies \psi) \implies ((\chi \implies \psi) \implies ((\phi \lor \chi) \implies \psi)) $
Biimplication is defined as usual.
$\phi\iff\neg\neg\phi $
$\neg (\phi \land \psi) \iff (\neg \phi \lor \neg \psi) $
$\neg (\phi \lor \psi) \iff (\neg \phi \land \neg \psi) $
$\phi \lor \neg \phi $
Here is my current attempt at proving one side of the conditional
Best Answer
The following tables satisfy the above axioms:
Tables
{1, 2} are the designated values.
Counterexample at 2, 3. ((ϕ ∧ ψ) ∨ ¬ϕ) is designated there but (ϕ⟹ψ) is not. Therefore the other side of the biconditional is not provable in this logic. □