With verbs like 'feel', and 'think' where negation can be transferred from the dependent clause to the main clause, e.g. "I don't think it will rain" can mean "I think it will not rain," how do we interpret questions like:

"Do you think it will rain?"

Is one asking whether

a) You think it will rain vs "Not" You think it will rain.
b) You think it will rain vs You think it will not rain.
c) Is it ambiguous between both a and b.

I see many examples of negative raising in the references of https://en.wikipedia.org/wiki/Negative_raising e.g. Horn, Laurence R. (2001). A natural history of negation. but couldn't find any examples on questions like the above where negation may be considered implicit in the question.

  • 2
    Just to complicate the question: You could also ask "Don't [do not] you think it will rain," or "Do you not think it will rain," both with the same meaning, and both with only very slight connotational shift from the original example. All three sentences boil down to, "Look, in your opinion, is it gonna rain or not," with the only difference being that "Don't you think" has an element of "leading the witness." (And we will not confuse things by discussing "Do you think it will not rain"... or "Don't you think it won't rain..." :) ) Oct 21, 2021 at 18:40
  • 2
    If someone couldn't judge or didn't know if it was likely to rain, they would reply to "Do you think it'll rain" by saying "I don't know", not by saying "no". This suggests the question is about weather, not about thought processes.
    – Stuart F
    Oct 21, 2021 at 19:42
  • 1
    A similar question is "Do you think John will come to the party?" Here you say "yes" if you think John will come, "no" if you think John will not come, and "I don't know"/"I'm not sure"/"maybe" if you don't know.
    – Stuart F
    Oct 21, 2021 at 19:44
  • 1
    I feel like some are having trouble zeroing in on the nature of the question. Would this be an accurate clarification?: 1) Assumption: the binary question holds a potential positive and negative simultaneously. 2) So then, is the hypothetical negative subject to negative-raising? Oct 21, 2021 at 19:59
  • 1
    And personally, I'm having trouble getting on board. I'm not sure that the "potential negative" held within a Schrödinger's-question is sufficient to imply anything about syntax. I feel a distinction between implied words in a sentence, like the "you" subject in an imperative, and potential words, which are simply concepts contingent upon present concepts. Which I guess makes my answer "C": To figure out whether the negative has been raised, Schrödinger has to open the box and answer the question. Oct 21, 2021 at 19:59

1 Answer 1


In "Do you think it will rain?", you are asking an open question that can be answered by "Yes" or "No."

The addition of or not [rain] adds nothing to the meaning other than a virtual (as opposed to real) greater psychological liberty to answer in the general negative.

However, in current Modern English, it is not possible to answer "Do you think it will rain or not?" with “Yes” or “No”.

The addition is therefore pointless as it prevents a direct unambiguous answer. (In earlier forms of Modern English, this was possible to answer such questions by the response of “Yeah” or “Nay”.)

The constructions parallels “May I go out?” in which there is an apparently implied “or not”. However, as it is unspoken, (in the same way that "Do you think it will rain?"), the question is taken at face-value and can be answered “Yes” or “No.”

The "or not", nevertheless, has an effect: as it cannot be answered monosyllabically, it encourages the responder to explain the reasons for his opinion.

Absent further, and less common, context, in questions like: "Do you think it will rain?", one is asking for an opinion – no more, no less. The reply - "Yes" or "No" creates a conversation that is the minimum required for an exchange of information and, as such is neither ambiguous or inefficient.

Not the answer you're looking for? Browse other questions tagged or ask your own question.