---
title: "Every ambiguity isn't syntactic in nature: Testing the Rational"
description: "The authors test the probabilistic ambiguity-resolution model of Scontras and Pearl (2020), an extension… 18 references, cited by 1 paper here."
authors: ["Sherry Yong Chen", "Bob van Tiel"]
year: 2021
venue: "Proceedings of the Society for Computation in Linguistics 2021"
source: "https://aclanthology.org/2021.scil-1.23/"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/chen-2021-every-ambiguity-isn-t-syntactic/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/chen-2021-every-ambiguity-isn-t-syntactic/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2021](https://md.research.elamur.ai/free/en/papers/2021/page/7/) / Every ambiguity isn't syntactic in nature: Testing the Rational Speech Act model of scope ambiguity\\ensuremath\*

# Every ambiguity isn't syntactic in nature: Testing the Rational Speech Act model of scope ambiguity\\ensuremath\*

**Authors:** [Sherry Yong Chen](https://md.research.elamur.ai/free/en/authors/sherry-yong-chen/) and [Bob van Tiel](https://md.research.elamur.ai/free/en/authors/bob-van-tiel/) · **Year:** 2021 · **Venue:** [Proceedings of the Society for Computation in Linguistics 2021](https://md.research.elamur.ai/free/en/venues/scil/) ([SCiL 2021](https://md.research.elamur.ai/free/en/venues/scil/2021/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2021.scil-1.23/) · [PDF](https://aclanthology.org/2021.scil-1.23.pdf) · [OpenAlex](https://openalex.org/W3133745795)

**Citations:** 2 citations (OpenAlex, 2026-10-14) · [1](https://md.research.elamur.ai/free/en/papers/chen-2021-every-ambiguity-isn-t-syntactic/cited-by/) in our library, of the 141,840 papers whose bibliographies we read

## Summary
The paper asks how much of the ambiguity in sentences like "Every horse didn't jump over the fence" comes from syntax and how much from pragmatics. Such sentences have a 'none' reading and a 'not-all' reading, traditionally tied to surface versus inverse scope parses. The authors test the probabilistic ambiguity-resolution model of Scontras and Pearl (2020), an extension of the Rational Speech Act (RSA) framework. In that model, the scope parse, the question under discussion (QUD) and prior beliefs about the world are separate inputs. Three experiments measured, for the same set of short stories, how salient each QUD is, what listeners expect a priori, and how they actually interpret the ambiguous utterance. The measured QUD and state priors were then fed into the model. The fit suggests that variation in interpretation is explained almost entirely by pragmatics, with syntax playing only a marginal part. This challenges classic accounts, such as the child–adult difference, that attribute reading preferences to children's limited ability to build parses.

## Key points
- **Model inputs:** two messages (the 'every-not' utterance and an always-true null message), three states (none, some-but-not-all, all) and three QUDs (all?, any?, how-many?).
- **Listener levels:** a literal listener L0 answers the QUD given a message and a scope. A soft-max speaker S1 has rationality α, standardly 1. A pragmatic listener L1 jointly infers state, scope and QUD using priors over each.
- **Experiments:** 72 short stories manipulated prior probability and QUD bias. Exp. 1 measured QUD salience, Exp. 2 measured prior expectations about states, and Exp. 3 measured interpretations.
- **Scope priors:** these cannot be measured directly, so several assumptions about them were compared in model evaluation.
- **Caveat on alternatives:** exploratory runs that added 'no' and plain 'every' utterances as alternatives gave syntax a larger role, so the set of alternatives matters. Materials and data are on GitHub (linguistsherry/qud-rsa-scil2021).

## Interesting ideas
- Instead of assuming context priors, measure them independently in separate tasks (QUD salience, world priors), then plug them into the model and compare against interpretation data. This avoids fitting everything to the same responses.
- A surface parse alone can still yield a 'not-all' meaning when the QUD collapses the none/some distinction, so meaning arises from parse plus question, not from parse alone.
- Treat the scope parse as a latent variable with its own prior that constrains but does not fix the interpretation, and infer it jointly with state and QUD.
- The conclusions depend heavily on which alternative utterances the model assumes. Making the alternative set explicit and testing variants (null, 'no', positive counterpart) is a useful robustness check.
- When reading an 'every…not' sentence, keep the world states (none / some / all) as candidate states weighted by prior plausibility, rather than committing early.

## For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

- **I1** (learn): Evaluation could estimate context priors from separate probe sets before scoring a disambiguation module, keeping the test independent of what was tuned.
- **I2** (frames): The prag crate could compute the reading of negated universals relative to an explicit QUD, rather than branching into two scope trees at parse time.
- **I3** (tree): Instead of forcing gparse to output a scope decision, the tree could stay scope-neutral, with scope left as an explainable, scored hypothesis downstream.
- **I4** (frames): Implicature reasoning in prag could list its competing alternatives explicitly in its explanation trace and report how conclusions change when 'no X' or the positive sentence is added.
- **I5** (world): The world-of-the-text could store quantity states as a small distribution over none/some/all, resolved later by questions in the answer stage.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [Massachusetts Institute of Technology](https://md.research.elamur.ai/free/en/universities/massachusetts-institute-of-technology/) — University, Cambridge, United States · 159 works · 65,157 citations

📚 Shelves (6): [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 9,296 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 9,206 works · [💭 Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/) · 798 works · [💭 Developmental and Educational Psychology](https://md.research.elamur.ai/free/en/shelves/psychology/developmental-and-educational-psychology/) · 260 works · [🧠 Neuroscience](https://md.research.elamur.ai/free/en/shelves/neuroscience/) · 95 works · [🧠 Cognitive Neuroscience](https://md.research.elamur.ai/free/en/shelves/neuroscience/cognitive-neuroscience/) · 95 works

## Cite

Cite as: Sherry Yong Chen and Bob van Tiel. 2021. Every ambiguity isn't syntactic in nature: Testing the Rational Speech Act model of scope ambiguity\\ensuremath\*. Proceedings of the Society for Computation in Linguistics 2021.

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{chen2021every,
  title = {{Every ambiguity isn't syntactic in nature: Testing the Rational Speech Act model of scope ambiguityensuremath*}},
  author = {Sherry Yong Chen and Bob {van Tiel}},
  booktitle = {Proceedings of the Society for Computation in Linguistics 2021},
  year = {2021},
  doi = {10.7275/h3rp-m711},
  url = {https://doi.org/10.7275/h3rp-m711}
}
```

- [← previous Do UD Trees Match Mention Spans in Coreference Annotations?](https://md.research.elamur.ai/free/en/papers/popel-2021-do-ud-trees-match-mention/)
- [next → Familiar vs. unique in a diachronic perspective. Case study of the rise of the definite article in North Germanic](https://md.research.elamur.ai/free/en/papers/piotrowska-2021-familiar-vs-unique-diachronic-perspective/)
- [↑ back to papers of 2021](https://md.research.elamur.ai/free/en/papers/2021/page/7/)

## References

18 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Mark Baltin. 1974. Quantifier-negative interaction. In New ways of analyzing variation in English, pages 30–36, Washington, DC. Georgetown University Press. [find](https://scholar.google.com/scholar?q=Mark+Baltin%2E+1974%2E+Quantifier%2Dnegative+interaction%2E+In+New+ways+of+analyzing+variation+in+English%2C+pages+30%E2%80%9336%2C+Washington%2C+DC%2E+Georgetown+University+Press%2E)
2. Guy Carden. 1970. A note on conflicting idiolects. Linguistic Inquiry, 1:281–290. [find](https://scholar.google.com/scholar?q=Guy+Carden%2E+1970%2E+A+note+on+conflicting+idiolects%2E+Linguistic+Inquiry%2C+1%3A281%E2%80%93290%2E)
3. Anastasia Conroy, Scott Fults, Julien Musolino, and Jeffrey Lidz. 2008. Surface scope as a default: The effect of time in resolving quantifier scope ambiguity. Poster presented at the 21st CUNY conference. [find](https://scholar.google.com/scholar?q=Anastasia+Conroy%2C+Scott+Fults%2C+Julien+Musolino%2C+and+Jeffrey+Lidz%2E+2008%2E+Surface+scope+as+a+default%3A+The+effect+of+time+in+resolving+quantifier+scope+ambiguity%2E+Poster+presented+at+the+21st+CUNY+confer)
4. Judith Degen, Michael Henry Tessler, and Noah D Goodman. 2015. Wonky worlds: Listeners revise world knowledge when utterances are odd. Proceedings of the 37th Annual Conference of the Cognitive Science Society, pages 548–553. [find](https://scholar.google.com/scholar?q=Judith+Degen%2C+Michael+Henry+Tessler%2C+and+Noah+D+Goodman%2E+2015%2E+Wonky+worlds%3A+Listeners+revise+world+knowledge+when+utterances+are+odd%2E+Proceedings+of+the+37th+Annual+Conference+of+the+Cognitive+Scienc)
5. Alex Drummond. 2013. Ibex Farm. [https://spellout.net/ibexfarm/](https://spellout.net/ibexfarm/). [↗](https://spellout.net/ibexfarm/)
6. Michael C. Frank and Noah D. Goodman. 2012. [Predicting pragmatic reasoning in language games](https://md.research.elamur.ai/free/en/papers/frank-2012-predicting-pragmatic-reasoning-language-games/). Science, 336(6084):998. [↗](https://doi.org/10.1126/science.1218633)
7. Andrea Gualmini. 2004. The ups and downs of child language. New York, NY: Routledge. [find](https://scholar.google.com/scholar?q=Andrea+Gualmini%2E+2004%2E+The+ups+and+downs+of+child+language%2E+New+York%2C+NY%3A+Routledge%2E)
8. Andrea Gualmini, Sarah Hulsey, Valentine Hacquard, and Danny Fox. 2008. The question–answer requirement for scope assignment. Natural Language Semantics, 16:205. [↗](https://doi.org/10.1007/s11050-008-9029-z)
9. Laurence R. Horn. 1989. A Natural History of Negation. CSLI, Stanford, CA. [find](https://scholar.google.com/scholar?q=Laurence+R%2E+Horn%2E+1989%2E+A+Natural+History+of+Negation%2E+CSLI%2C+Stanford%2C+CA%2E)
10. Nirit Kadmon and C. Roberts. 1986. Prosody and scope: The role of discourse structure. In Proceedings of the Annual Meeting of Chicago Linguistics Society, pages 16–28. Chicago Linguistic Society. [find](https://scholar.google.com/scholar?q=Nirit+Kadmon+and+C%2E+Roberts%2E+1986%2E+Prosody+and+scope%3A+The+role+of+discourse+structure%2E+In+Proceedings+of+the+Annual+Meeting+of+Chicago+Linguistics+Society%2C+pages+16%E2%80%9328%2E+Chicago+Linguistic+Society%2E)
11. Julien Musolino. 1998. Universal grammar and the acquisition of semantic knowledge. Ph.D. thesis, University of Maryland. [find](https://scholar.google.com/scholar?q=Julien+Musolino%2E+1998%2E+Universal+grammar+and+the+acquisition+of+semantic+knowledge%2E+Ph%2ED%2E+thesis%2C+University+of+Maryland%2E)
12. R Core Team. 2017. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. [find](https://scholar.google.com/scholar?q=R+Core+Team%2E+2017%2E+R%3A+A+Language+and+Environment+for+Statistical+Computing%2E+R+Foundation+for+Statistical+Computing%2C+Vienna%2C+Austria%2E)
13. Craige Roberts. 2012. Information structure: Towards an integrated formal theory of pragmatics. Semantics & Pragmatics, 5:1–69. [find](https://scholar.google.com/scholar?q=Craige+Roberts%2E+2012%2E+Information+structure%3A+Towards+an+integrated+formal+theory+of+pragmatics%2E+Semantics+%26+Pragmatics%2C+5%3A1%E2%80%9369%2E)
14. KJ Savinelli, Gregory Scontras, and Lisa Pearl. 2017. Modeling scope ambiguity resolution as pragmatic inference: Formalizing differences in child and adult behavior. In Proceedings of the Annual Meeting of Cognitive Science Society. [find](https://scholar.google.com/scholar?q=KJ+Savinelli%2C+Gregory+Scontras%2C+and+Lisa+Pearl%2E+2017%2E+Modeling+scope+ambiguity+resolution+as+pragmatic+inference%3A+Formalizing+differences+in+child+and+adult+behavior%2E+In+Proceedings+of+the+Annual+Meet)
15. KJ Savinelli, Gregory Scontras, and Lisa Pearl. 2018. Exactly two things to learn from modeling scope ambiguity resolution: Developmental continuity and numeral semantics. In Proceedings of the 8th Workshop on CMCL, pages 67–75. [find](https://scholar.google.com/scholar?q=KJ+Savinelli%2C+Gregory+Scontras%2C+and+Lisa+Pearl%2E+2018%2E+Exactly+two+things+to+learn+from+modeling+scope+ambiguity+resolution%3A+Developmental+continuity+and+numeral+semantics%2E+In+Proceedings+of+the+8th+Wo)
16. Gregory Scontras and Lisa Pearl. 2020. [When pragmatics matters more for truth-value judgments: An investigation of quantifier scope ambiguity](https://md.research.elamur.ai/free/en/papers/scontras-2021-when-pragmatics-matters-more-truth/). Lingbuzz/005287. [↗](https://doi.org/10.16995/glossa.5724)
17. Kristen Syrett, Georgia Simon, and Kirsten Nisula. 2014. Prosodic disambiguation of scopally ambiguous quantificational sentences in a discourse context. Journal of Linguistics, 50:453–493. [↗](https://doi.org/10.1017/s0022226714000012)
18. Peter Wason. 1972. The context of plausible denial. Journal of Verbal Learning and Language Behavior, 4:7–11. [find](https://scholar.google.com/scholar?q=Peter+Wason%2E+1972%2E+The+context+of+plausible+denial%2E+Journal+of+Verbal+Learning+and+Language+Behavior%2C+4%3A7%E2%80%9311%2E)

*Updated 24 October 2026*