> Hindley-Milner type systems – which I here use to refer broadly to type systems like those used by Haskell and OCaml – correspond to propositional logic.
Is this true? HM type systems at least have a limited form of universal quantification (type schemes), which means we're dealing with predicate calculus, not propositional logic.
