An active sentence entails its passive counterpart, and vice versa. Thus, the two sentences John likes Joan and Joan is liked by John mutually entail each other. My question in this regard concerns the extent to which predicate-argument analyses of such sentences should be similar or the same. For instance, is it correct to posit that the two sentences just produced allow the same one predicate-argument analysis, something like (1)?

(1) like (John, Joan)

Or perhaps the analyses should be similar but not the same. They might assume the same arguments but allow for a more surface-oriented understanding of predicates, e.g.

(2) a. likes (John, Joan) (2) b. is liked by (John, Joan) or (Joan, John)

Consider this same issue with respect to light verb constructions, e.g. John hugged Joan vs. John gave Joan a hug. These two sentences are arguably almost synonymous; they seem to mutually entail each other. Should they therefore receive the same or similar predicate-argument analyses, something like (3)?

(3) hug (John, Joan)

Or, again, assuming the same arguments but allowing for a more surface-oriented understanding of predicates, something like (4a-b):

(4) a. hug (John, Joan) (4) b. give a hug (John, Joan)

In short, then, my question is looking for guidance concerning the (potential) connection between entailment and predicate-argument analyses. All comments welcome, including suggestions pointing to the relevant literature.

  • 2
    In a derivational theory, entailment has nothing to do with syntactic structures like active and passive (or any other two sentences related by a transformation), since transformations don't change semantics. Different syntactic approaches may answer the question differently, or they may regard it as meaningless.
    – jlawler
    Commented Jul 21, 2022 at 13:39
  • 1
    Thanks for the comment. I do not quite understand, though. I think you are stating that there is indeed mutual entailment across an active sentence and its passive counterpart, but that there can be only one predicate-argument analysis underlying them because predicate-argument analyses are distinct from derivational syntax and not affected by derivational syntax. Is that correct?
    – Buffoon
    Commented Jul 21, 2022 at 15:30
  • 2
    Yeah, pretty much. The systems are partially synchronized -- most adjective and noun predicates are intransitive, most 3-place predicates have Source-Trajector-Goal semantics, most transitive predicates have agent or experiencer (aka human) subjects, etc. But they're not the same system and syntax can warp everything to fit, as needed. Oh, and of course entailment is not the only pragmaticological relation possible. Presupposition is much more important.
    – jlawler
    Commented Jul 21, 2022 at 18:32


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.