---
title: "Signalling the Interpretation of Indirect Speech Acts — Cohen 1980"
description: "Cohen argues that the hearer's later wording can show which interpretation she actually chose."
authors: ["Philip R. Cohen"]
year: 1980
venue: "18th Annual Meeting of the Association for Computational Linguistics"
source: "https://aclanthology.org/P80-1008/"
doi: "10.3115/981436.981446"
licence: "CC BY-NC-SA 3.0 (ACL Anthology, before 2016)"
licence_url: "https://creativecommons.org/licenses/by-nc-sa/3.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/cohen-1980-signalling-interpretation-indirect-speech-acts/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/cohen-1980-signalling-interpretation-indirect-speech-acts/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [1980](https://md.research.elamur.ai/free/en/papers/1980/) / Signalling the Interpretation of Indirect Speech Acts

# Signalling the Interpretation of Indirect Speech Acts

**Authors:** [Philip R. Cohen](https://md.research.elamur.ai/free/en/authors/philip-r-cohen/) · **Year:** 1980 · **Venue:** [18th Annual Meeting of the Association for Computational Linguistics](https://md.research.elamur.ai/free/en/venues/acl/) ([ACL 1980](https://md.research.elamur.ai/free/en/venues/acl/1980/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY-NC-SA 3.0 (ACL Anthology, before 2016)](https://creativecommons.org/licenses/by-nc-sa/3.0/)

**Links:** [Publisher's page](https://aclanthology.org/P80-1008/) · [PDF](https://aclanthology.org/P80-1008.pdf) · [DOI 10.3115/981436.981446](https://doi.org/10.3115/981436.981446) · [OpenAlex](https://openalex.org/W2164124540)

**Citations:** 1 citations (OpenAlex, 2026-10-14)

## Summary
This short ACL 1980 panel paper asks how plan-based theories of speech acts can be checked against real dialogue. Such theories model how a hearer infers the illocutionary force of an utterance, including indirect requests, by recognizing the plan behind it. Cohen argues that the hearer's later wording can show which interpretation she actually chose. The setting is task-oriented dialogue in which an expert guides a novice through assembling a toy water pump, with experiments over teletype and telephone. In the example, the expert says a declarative about needing a bent tube, and the novice replies "OK, it's done." Two inference routes could have led her to act. On the "helpful" route, she hears a statement of need and helps on her own initiative. On the "intended" route, she recognizes that the expert meant her to do it. Her pronoun "it" refers to an action, which requires that action to be mutually believed and in focus. Only the indirect-request reading supplies that shared focus. Anaphora therefore becomes empirical evidence of how intent was interpreted.

## Key points
- The context is a task-oriented expert–novice dialogue about assembling a toy water pump, with analogous phenomena observed in both teletype and telephone modes.
- The paper builds on the Perrault and Allen plan-based model. In that model, mood-dependent "surface" speech act operators are turned into illocutionary acts such as requests by plan-recognition inferences.
- For a declarative stating a joint need, the theory offers two paths, "helpful" and "intended". Perrault and Allen's heuristics favour the intended path when the relevant act is mutually believed.
- The paper relies on the view that definite and pronominal or pro-verbal reference needs mutual belief that the referent is in focus (Clark & Marshall; Grosz). It also cites Robinson's implemented system for resolving reference to actions.
- Cohen proposes decomposing primitive surface speech acts so that they include reference acts. Planning those acts to help the hearer might allow cooperative answers without appeal to Gricean maxims.

## Interesting ideas
- Use the hearer's follow-up wording, especially anaphora, as a probe that reveals which reading of an earlier utterance was taken, instead of guessing intent from the utterance alone.
- Keep two competing inference paths for one utterance, the hearer acting on her own helpfulness versus acting on the speaker's recognized intent, and choose between them using mutual belief and task context.
- Pronouns and pro-verbs can refer to actions, not just entities, and only actions placed in shared focus are available as referents.
- Treat communicative acts like any other plan operator whose effects change mental states, so one reasoning mechanism serves both talk and physical action.
- Plan reference expressions to support the listener's likely plan, as a route to cooperative answers that does not depend on Gricean maxims.

## For Mova

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

- **I1** (frames): When prag picks between a literal inform and an indirect-request reading, it can check whether later pronouns like "it" or "do so" resolve to the requested action. That check gives the choice an explainable piece of evidence.
- **I2** (frames): The speech-act layer could store both readings with their justifications and let context such as expected task goals select one, so the final decision can be traced.
- **I3** (world): The world of the text could keep a focus list of mutually expected or requested events. Coreference could then link "it's done" to a concrete event, which also updates the event order and state.
- **I4** (world): Speech acts could be modelled as events that alter beliefs and obligations of characters, using the same state machinery as physical events in fables and tales.
- **I5** (back): During generation, the system could choose referring expressions and extra details by checking what the reader needs for their inferred goal.

*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

- [University of Illinois Urbana-Champaign](https://md.research.elamur.ai/free/en/universities/university-of-illinois-urbana-champaign/) — University, Urbana, United States · 209 works · 71,394 citations

📚 Shelves (2): [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 17,450 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 17,296 works

🧭 Topics (1): [💬 Speech act theory](https://md.research.elamur.ai/free/en/topics/speech-act-theory/) · 229 works

## Cite

Cite as: Philip R. Cohen. 1980. Signalling the Interpretation of Indirect Speech Acts. 18th Annual Meeting of the Association for Computational Linguistics.

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

```bibtex
@inproceedings{cohen1980signalling,
  title = {{Signalling the Interpretation of Indirect Speech Acts}},
  author = {Philip R. Cohen},
  booktitle = {18th Annual Meeting of the Association for Computational Linguistics},
  year = {1980},
  doi = {10.3115/981436.981446},
  url = {https://doi.org/10.3115/981436.981446}
}
```

- [← previous Computer-Aided Grammatical Tagging of Spoken English](https://md.research.elamur.ai/free/en/papers/svartvik-1980-computer-aided-grammatical-tagging-spoken/)
- [next → On FROFF: A Text Processing System for English Texts and Figures](https://md.research.elamur.ai/free/en/papers/abe-1980-froff-text-processing-system-english/)
- [↑ back to papers of 1980](https://md.research.elamur.ai/free/en/papers/1980/)

*Updated 29 October 2026*