We discuss the role of counter-factual meaningfulness (a weaker cousin of "counter-factual definiteness") as a premise in the derivation of the Bell and CHSH inequalities. The basic question motivating the discussion is this: can the CHSH inequality, unlike the original Bell inequality, be derived without making a hidden-variables (or equivalent counter-factual definiteness) assumption? We answer, somewhat tentatively, in the negative, and suggest that an appropriately-modified version of the EPR argument is needed to rigorously establish that the empirical violation of Bell-type inequalities can only be blamed on the failure, in nature, of local causality.