Introduction to the deduction theorem

In mathematical logic, deduction theorem is a general name for meta-theorems that state something on the kind:

If \(\Gamma, A \vdash B\) then \(\Gamma \vdash A \to B\).

This is read as: If \(B\) can be proved from a set of premises \(\Gamma\) and \(A\) (i.e. the premises include \(A\)), then there is a (different) proof of “\(A \to B\)” (\(A\) implies \(B\)) from the same set of premises \(\Gamma\) without \(A\).

Very informally a deduction theorem states that “implies implies implies”. This is a pun to the fact that the material implication logical operator \(\to\) is closely related to the deduction yields (\(\vdash\)), either of them is called implication (usually when it’s clear from context which one is meant).

The reason for being a general name is that words like “there is a proof”, represented by the symbol \(\vdash\), are relative to a formal system e.g. “there is a proof using this specific formal system”. A specific formal system has its own theorem that we call the deduction theorem for that formal system. For some formal systems a deduction theorem does not exist/does not make sense, while in others, as we’ll see, it comes with additional constraints/requirements to the basic form above.

The plan in this series of articles is to cover two examples of deduction theorems: a simple one for propositional calculus and a more complicated one with additional constraints for predicate calculus.

Will cover:

References

Stephen Cole Kleene: Introduction to Metamathematics (Ishi Press: 2009 reprint)

Elliott Mendelson: Introduction to Mathematical Logic - sixth edition (CRC Press: 2015)

Encyclopedia of Mathematics: Deduction Theorem

Euclid: The Thirteen Books of the Elements / Translated and commentary by Sir Thomas L. Heath / Vol.1 (Books I and II)