LINGUIST List 26.2445

Tue May 12 2015

Sum: 26.2105, Qs: Epistemic

Editor for this issue: Anna White <>

Date: 10-May-2015
From: Seiichi Myoga <>
Subject: 26.2105, Qs: Epistemic
E-mail this message to a friend

26.2105, Qs: Epistemic

Let me first express my gratitude to And Rosta, Bruce Despain and Carolin Baumann.

Part I: Purpose and Result

The original purpose is to see if my assumption is valid that under the interpretation of (2), May is what receives the primary stress in (1):
(1) May John be leaving?
(2) Do you agree that there is a possibility of John’s leaving? (Anderson 2011:155)
In my previous post, I actually tried to kill two birds with one stone.
Both Palmer (1987) and Declerck (1991) agree that epistemic may is a positive polarity item (or an assertive form/item in their terms), but only the former accepts this modal in the environment of Yes/No questions. As for (4a) Palmer (1987:105) says, “Here the speaker expresses a rather more positive attitude towards the statement being questioned.”
(4) a. May John be in his office? (Palmer 1987:105)
b. *May he be in his office now? (Declerck 1991:404)

I wondered (and still wonder) why Declerck (1991) judges (4b) as unacceptable. Is it because epistemic may appears in biased Yes/No questions or in default Yes/No questions (or whatever)? I had expected the position of the primary stress would give us the answer, or at least a clue to the answer.

We have received three replies, two of which came from native speakers. After all, my assumption was neither verified nor refuted. Surprisingly for me, though, both native speakers prefer to interpret (1) differently. Their comments suggest it is possible for some native speakers of English to use (1) to ask if there is a possibility of John’s leaving. To my knowledge, no mention of this possibility is made in the literature. I tentatively assume that native speakers of English who accept (1) as a default Yes-No question may take it as expressing two exhaustive possibilities (just as an alternative question does) rather than a single possibility.

I don’t know which comes first, but I think we could reasonably say that we can use epistemic may in alternative questions:
(5) I may hate the prospect of an increased air force as much as any of you do. But I have to ask myself - may this or may it not make the difference? May it or may it not save the lives of those people who put me in power to do my duty? (J. Dobson, Why Do the People Hate Me So? my emphases)
Note that this was in 1935 and the speaker is Stanley Baldwin, a British prime minister of the time.

Part II: Comments We Have Received

Speaker A: I think it would most naturally fall on ''leaving'', but perhaps I've failed to understand the import of paraphrase (2). I assume ''receive stress'' means ''bear the *primary* accent in the sentence''.
> The other question is, Do you agree that (1) works in the following imaginative dialogue?
> I'm wondering if a previous mention of possibility helps improve the acceptability of (1).
Rather, it's the previous mention of John leaving that allows it to be given information and hence deaccented.
Speaker B: The ''may'' is not stressed. The question is about the probability of the event taking place, The question asks about whether the hearer has thought about it being possible, maybe likely, maybe 80% probability, maybe anything > 50% probability.

The weird dialogue is possible (in Japan?), but the ''may'' should not even be stressed in this context, except possibly to emphasize the idea that the hearer didn't hear it in its first utterance. But, I assume you didn't take it to have been uttered beforehand. The dialogue is in fact realistic for an occurrence of the question, though the weird epistemic attitude of the hearer being expressed is not expected.

There does not seem to be a situation I can think of that would take the hearer out of his subjective role in judging the existence of the possibility. Why would the question even come up? The objective possibility would have to have a much more unusual context. ''Is it at all possible that John may be leaving?'' would ask this more unusual question, not the modal auxiliary ''may''. (probability > 0.0%)
In (3)c there is no stress on ''is''; the word order is enough to mark it as a yes/no question. (3)a the stress is contrastive, meaning that his leaving is not in question, just the day of his leaving. That is why (3)b is asking about the time of his leaving, not the fact that he plans to leave. (wh-question, the wh-word is usually stressed.)

Speaker C: I’m not sure, if I can help you with your question, but I wanted to answer, because I have problems with the distinction between subjective and objective epistemic modality, too. Although no native speaker of English (but concerned with German modals in my PhD) I think, what your example (1) means is, no matter in what context, something like this:
Is John leaving? I know, you don’t KNOW, but what do you think (subjectively)? (I guess that fits well with your given context: The simple question demands the knowledge of the hearer, what has to be rejected – the simple question replaced by a epistemically modalized.) Whereby “may” is an element calling for the subjective epistemic estimation of the hearer. By using “may” as an epistemic marker the speaker shows that he does not expect a definite answer, but an epistemic estimation.
I think there is no linguistic relevant distinction between subjective and objective epistemic modality.

Linguistic Field(s): General Linguistics

Subject Language(s): English (eng)

Page Updated: 12-May-2015