Proof about substitution in well-formed formulas.

logicproof-verification

I’ve completed a proof for the question below (taken from H.Enderton’s ‘A Mathematical Introduction to Logic’), but I don’t feel confident about it. Unfortunately my university doesn’t offer a course in mathematical logic, so I’ve resorted to self-study. Hopefully someone can check the validity of the proof:

We prove (a) by induction on the number of sentence symbols in an arbitrary wff ф.

For the base case we have $n=1$, which is a single sentence symbol $A_1$, and from the defintion of our valuation we have $\overline{u}(\phi) = u(A_1) = \overline{v}(\alpha _1) = \overline{v}(\phi ^*)$.

Assume the hypothesis holds for any wff containing $\leq n$ sentence symbols.

Next suppose we have a wff $\gamma$ that features $n+1$ sentence symbols. Clearly it could have two types of predecessors: either a negation, such that $\gamma = \epsilon _\lnot (\beta )$, or a binary connective, such that $\gamma = \epsilon _\ast (\beta , \alpha )$, where $\ast\in \{ \land ,\lor ,\to ,⇔\}$ and $\alpha ,\beta$ are also wff’s. In the latter case we know that the number of sentence symbols in $\alpha$ and $\beta$ will be $\leq n$, respectively, so the inductive hypothesis applies and $\overline{u} (\alpha ) = \overline{v} (\alpha ^*)$ and $\overline{u} (\beta ) = \overline{v} (\beta ^*)$. It then follows trivially that $\overline{u} (\gamma )$ will equal $\overline{v} (\gamma ^*)$ since we are not adding any new sentence symbols, but appending the two using the same binary connective, thus the valuation will be preserved. In the case where $\gamma = \epsilon _\lnot (\beta )$ we simply consider the two predecessors of $\beta$ since the negation just flips the values of the valuation and is thus consistent, with the procedure identical to that described above. With no other cases left to consider, this concludes the proof.

enter image description here

Best Answer

You certainly have the right idea ... but there is just one small problem with the case of the negation. You say to consider the 'two predecessors' of $\beta$ ... but what if $\beta$ is a negation itself?

Fortunately, the proof is easily fixed ... instead of performing induction on the number of sentence symbols, simply perform induction on the length of the statements, i.e. the number of sentence symbols plus the number of logical operators that are in the expression.