Suppose that the generality-particularity ordering is the comprehensive ordering on inference. Compare:

  1. What is the first integer after 2?
  2. What is the first prime integer after 2?

(2) would seem more particular than (1). However, I would be uncomfortable with a strict "erotetic deduction" of (2) from (1). Also, I would be more comfortable inferring (1) from (2), so to say. And all this even though (1) and (2) have, apparently, the very same answer.

Or consider:

  1. Where did you go yesterday?
  2. You were gone for about an hour yesterday, and I saw that you didn't take the car, and all of your friends were busy, so you had to walk wherever you went.
  3. So, where did you go yesterday, that you could have gone there and gotten back in an hour, on foot?

I'm not sure, then, that the generality-particularity ordering, on questions, is an inferential order at all, or if it is, it seems somewhat "deviant." Does this testify against using that ordering to characterize assertoric logic (e.g. induction as "inference of generalizations from particular cases") too, or does it just testify against talk of "erotetic inference"?

  • 1
    They took an Uber? Seriously though, there is an exception to every rule. Particularity violates generality and generality over-rules particularity. It's like the old Holism - Reductionism thing.
    – Scott Rowe
    Commented Sep 3, 2022 at 17:19
  • @ScottRowe, I would have to have the second sequence be more like "probabilistic inference," maybe, although how probabilistic inference differs from strict inference of individual probabilities, IDK. Of course, the more qualifiers you add in, or subtract out, the easier or harder it will be, in different ways, to satisfy the description... Commented Sep 3, 2022 at 17:36
  • 1
    The standard view as expressed by Belnap is in agreement with you:"Absolutely the wrong thing is to think it is a logic in the sense of a deductive system, since one would thus be driven to the pointless task of inventing an inferential scheme in which questions, or interrogatives, could serve as premises and conclusions." See Wiśniewski, The Logic of Questions as a Theory of Erotetic Arguments for an alternative inspired by Hintikka's interrogative games.
    – Conifold
    Commented Sep 3, 2022 at 22:23
  • 2
    If deductive logic is the mark of vanilla machines then erotetic logic is the mark of artificial intelligence as conversation is mainly in Q&A dialogue format, thus Fregean sense (sinn) and Kantian modality take dominance here compared to mere logic and even a slight seemingly non-deviant contingent substitution may quickly become deviant. For example, Does Tom know that Cicero denounced Catiline? could be totally different from another question that is Does Tom know that Tully denounced Catiline? if Tom doesn't know Cicero aka Tully at all... Commented Sep 6, 2022 at 3:07


You must log in to answer this question.

Browse other questions tagged .