Alternative questions: Distinguishing between negated and complementary disjuncts
Paper · 2020 · Andrea Beltrama, Erlinde Meertens and Maribel Romero
Summary
The paper asks why questions that seem logically equivalent behave differently in conversation. Earlier work contrasted Polar Questions ("Is the light on?") with Negative Alternative Questions that end in "or not". The authors add a third, little-studied type: Complement Alternative Questions (CAQs). These name the opposite option with a positive antonym or phrase instead of negation ("on or off"). NAQs and CAQs share the same Hamblin-style denotation {p, ¬p}.
Three sets of rating experiments tested the three forms in Bolinger-style special uses (invitations, rhetorical questions, checking an inference) and in the two parts of the Cornering Effect.
- PQs and NAQs patterned uniformly in the special uses, while CAQs varied by context.
- NAQs were confirmed as odd when they open a discourse, but CAQs were acceptable there.
- Neither alternative-question form was strictly limited to closing the discourse.
These results show that no existing account covers the whole paradigm. The authors suggest revisions based on speaker goals, a ban on repeating failed moves, and focus licensed by the Question under Discussion.
Key points
- The paradigm is a triple with the same resolutions: PQ "Is the light on?", NAQ "…or not?", CAQ "…or off?".
- Experiment 1 had three sub-studies (1A invitation, 1B rhetorical, 1C inference). Each used 4 conditions: PQ, NAQ, CAQ and an unrelated control question.
- Each sub-study used 24 items in a Latin Square across four lists, plus 24 fillers.
- Participants were 48 native English speakers from Amazon MTurk. They rated naturalness relative to the speaker's goal on a 1–7 scale.
- The analysis fit mixed-effects models in lmerTest, treating Question Type as the fixed predictor and including random intercepts for both subjects and items.
- Cornering Part 1 (NAQs are bad discourse-initially) was supported. Part 2 (NAQs must be discourse-final) was challenged by Experiment 3.
Interesting ideas
- I1. Same meaning, different form. Holding the denotation fixed while varying only the surface form (negation vs. antonym) isolates what the wording itself contributes pragmatically.
- I2. Unrelated baseline condition. Adding an off-topic question gives a floor for the ratings, so judgments of the real conditions can be read against a known-bad option.
- I3. Ban on repeating failed moves. A follow-up may be odd not because of its form, but because the same move was already tried without success and dropped; the same holds for repeated commands.
- I4. CAQs fit when both options are useful. Positive-alternative questions work when each named option serves the conversation, for example offering a polite way out ("or would you rather go alone").
- I5. Focus must match the QUD. Stressed negation in an NAQ needs a prior question to license it, which would explain why NAQs fail when they open a discourse.
For Mova
How these ideas could serve Mova, our small open language model — its files, stages and tests:
- I1 (frames): Speech-act and stance detection in prag should treat a trailing "or not" as a marker of insistence or pressure, rather than collapsing it to the plain yes/no question.
- I2 (learn): Evaluation suites could include deliberately irrelevant candidate answers or parses as negative controls, to check that scores actually separate signal from noise.
- I3 (frames): The dialogue tracker could record failed moves and flag a repeated question or command as escalation or irritation instead of a fresh request.
- I4 (frames): When a disjunctive question names two desirable options, prag could tag it as an offer or politeness strategy rather than a neutral information request.
- I5 (world): Keeping an explicit stack of open questions in the text world would let Mova check whether a focused or "or not" question fits the preceding context.
Institutions
United States📚 Shelves 3 Arts and Humanities · 9,296 works Language and Linguistics · 9,206 works Philosophy · 211 works
Cite
Cite as Andrea Beltrama, Erlinde Meertens and Maribel Romero. 2020. Alternative questions: Distinguishing between negated and complementary disjuncts. Semantics and Pragmatics.
@article{beltrama2020alternative,
title = {{Alternative questions: Distinguishing between negated and complementary disjuncts}},
author = {Andrea Beltrama, Erlinde Meertens and Maribel Romero},
journal = {Semantics and Pragmatics},
year = {2020},
doi = {10.3765/sp.13.5},
url = {https://doi.org/10.3765/sp.13.5}
}References
- Bartels, Christine. 1999. The intonation of English statements and questions. New York, NY: Routledge. find
- Biezma, Maria. 2009. Alternative vs polar questions: The cornering effect. Semantics and Linguistic Theory (SALT) 19. 37–54. https://doi.org/10.376 5/salt.v19i0.2519. ↗
- Biezma, Maria & Kyle Rawlins. 2012. Responding to alternative and polar questions. Linguistics & Philosophy 35(5). 361–406. https://doi.org/10 .1007/s10988-012-9123-z. ↗
- Biezma, Maria & Kyle Rawlins. 2017. Or what? Semantics and Pragmatics 10(16). 1–44. https://doi.org/10.3765/sp.10.16. any reference to the QUD. However, under the latter approach, the contrast between (i.b) and (ii.b) is unexpected. For discussion of examples like (iii), which seemingly do not license focus via the QUD, see Roberts’ (2012: 53 ff.) suggestion that metalinguistic use of prosody is involved. (i) a. b. Who called Amy? Did JOHN 𝐹 call Amy? Who called Amy? Did JOHN 𝐹 or PAUL 𝐹 call Amy? (ii) a. #Who called Amy? Did John call AMY 𝐹 ? b. #Who called Amy? Did John call AMY 𝐹 or SUSAN 𝐹 ? (iii) An AMERICAN 𝐹 farmer was talking to a CANADIAN 𝐹 farmer. Alternative questions Bolinger, Dwight. 1978. Yes—no questions are not alternative questions. In Henry Hiż (ed.), Questions, 87–105. Dordrecht: Springer. https://doi.org/1 0.1007/978-94-009-9509-3_3. ↗ our card
- Brown, Penelope & Stephen C. Levinson. 1987. Politeness. Cambridge, MA: Cambridge University Press. find
- Caponigro, Ivano & Jon Sprouse. 2007. Rhetorical questions as questions. Sinn und Bedeutung 11. 121–133. https://doi.org/10.18148/sub/2007.v11i 0.635. ↗
- Goodhue, Daniel. 2018. On asking and answering biased polar questions. Montreal, QC: McGill University dissertation. ↗
- Hamblin, Charles L. 1973. Questions in Montague grammar. Foundations of language 10(1). 41–53. find
- Han, Chung-hye & Maribel Romero. 2004. Disjunction, focus and scope. Linguistic Inquiry 35(2). 179–217. https://doi.org/10.1162/002438904323019 048. ↗
- Kuznetsova, Alexandra, Per B. Brockhoff & Rune Haubo Bojesen Christensen. 2017. lmerTest package: Tests in linear mixed effects models. Journal of Statistical Software 82(13). 1–26. https://doi.org/10.18637/jss.v082.i13. ↗
- Roberts, Craige. 2012. Information structure in discourse: Towards an integrated formal theory of pragmatics. Semantics and Pragmatics 5(6). 1–69. https://doi.org/10.3765/sp.5.6. van Rooy, Robert & Marie Šafářová. 2003. On polar questions. Semantics and Linguistic Theory (SALT) 13. 292–309. https://doi.org/10.3765/salt.v13i0 .2887. ↗ our card
- Wilder, Chris. 2013. English ‘emphatic do’. Lingua 128. 142–171. https://doi.org/10.1016/j.lingua.2012.10.005. Andrea Beltrama Department of Linguistics University of Pennsylvania 3401 Walnut Street — 300C Philadelphia, PA, USA 19104 beltrama@sas.upenn.edu Erlinde Meertens Department of Linguistics University of Konstanz Universitaetsstrasse 10, Room H146 Konstanz, Germany, 78457 erlindemeertens@protonmail.com Maribel Romero Department of Linguistics University of Konstanz Universitaetsstrasse 10, Room G222 Konstanz, Germany, 78457 maribel.romero@uni-konstanz.de ↗
Updated
This page in Markdown → https://md.research.elamur.ai/free/en/papers/beltrama-2020-alternative-questions-distinguishing-between-negated/