--- title: "Whither Discourse and Speech Acts? — paper by Grosz (1987)" description: "The piece offers no new data or experiments. 8 references. This brief statement opens the TINLAP-3 panel on discourse and speech acts." authors: ["Barbara J. Grosz"] year: 1987 venue: "Theoretical Issues in Natural Language Processing 3" source: "https://aclanthology.org/T87-1018/" 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/grosz-1987-whither-discourse-speech-acts/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/grosz-1987-whither-discourse-speech-acts/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [1987](https://md.research.elamur.ai/free/en/papers/1987/) / Whither Discourse and Speech Acts? # Whither Discourse and Speech Acts? **Authors:** [Barbara J. Grosz](https://md.research.elamur.ai/free/en/authors/barbara-j-grosz/) · **Year:** 1987 · **Venue:** [Theoretical Issues in Natural Language Processing 3](https://md.research.elamur.ai/free/en/venues/tinlap/) ([TINLAP 1987](https://md.research.elamur.ai/free/en/venues/tinlap/1987/)) · **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/T87-1018/) · [PDF](https://aclanthology.org/T87-1018.pdf) · [OpenAlex](https://openalex.org/W1985613933) **Citations:** 0 citations (OpenAlex, 2026-10-14) ## Summary This brief statement opens the TINLAP-3 panel on discourse and speech acts. Grosz, the panel chair, asked panelists to reflect on how their thinking had shifted over roughly five years: which earlier beliefs now look most mistaken, and what has become clearer. The piece offers no new data or experiments. It is a survey of where the field stands. Grosz argues that research has matured and become more focused. Several complete theories of discourse structure now exist, including her joint work with Sidner on attention and intentions. Alongside them are partial accounts of how individual expressions, clauses and links between clauses are interpreted or generated, and early psychological tests of these theories are underway. She notes three shifts. Questions are now sharper. Speech-signal information (Hirschberg) and the mental-state effects of utterances (Perrault) are taken seriously. Intentions and plans are recognized as pervasive, which shows that standard AI planning models are too weak for discourse (Wilensky, Pollack, Kautz & Allen). Her verdict is that the problems are now well understood, but how to solve them is still disputed. ## Key points - The panel had two guiding reflective prompts: what prior belief now seems most wrong, and what is now less confusing. - Competing discourse theories make different predictions about structure, coherence, and the processing of cue phrases, definite descriptions and pronouns. - The central reference theory is Grosz & Sidner's paper treating discourse structure in terms of attention and intentions (Computational Linguistics 12(3), 1986). - Hirschberg's contribution claims that many discourse problems get simpler solutions when the speech signal is not ignored. - Existing AI planning is judged insufficient for discourse. Cited alternatives include Pollack 1986 (actor vs. observer beliefs in plan inference) and Kautz & Allen's generalized plan recognition. ## Interesting ideas - Asking researchers which of their own past beliefs now seems most wrong is a cheap, honest way to audit progress. - Intentions shape interpretation at three levels at once: phrases inside an utterance, single utterances, and long stretches of discourse. - Treating utterances as actions whose main effect is a change in participants' mental states makes speech acts concrete and checkable. - Plan recognition should keep the observer's beliefs separate from the actor's beliefs, as in the cited Pollack model. - Discourse theories can be compared by the different predictions they make for specific expression types such as pronouns, definite descriptions and cue phrases. ## For Mova How these ideas could serve Mova, our small open language model — its files, stages and tests: - **I1** (learn): Periodically review induced repair rules and teacher-tier assumptions with this question, and retire those that negative controls no longer support. - **I2** (frames): Let the prag crate attach intention or speech-act information to each of these levels, not only to whole sentences. - **I3** (world): Model each speech act as an explicit update to a character's beliefs or goals in the text world, so later answers can trace why a character knows something. - **I4** (world): Store separate belief states for characters and for the narrator or reader, which matters for fables built on deception or mistaken plans. - **I5** (frames): Build coreference and discourse-link evaluation sets split by expression type, so each resolution strategy is tested where theories actually disagree. *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 - [Harvard University](https://md.research.elamur.ai/free/en/universities/harvard-university/) — University, Cambridge, United States · 112 works · 42,023 citations 📚 Shelves (3): [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works · [🤔 Philosophy](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/philosophy/) · 284 works 🧭 Topics (2): [🧵 Discourse analysis](https://md.research.elamur.ai/free/en/topics/discourse-analysis/) · 466 works · [💬 Speech act theory](https://md.research.elamur.ai/free/en/topics/speech-act-theory/) · 229 works ## Cite Cite as: Barbara J. Grosz. 1987. Whither Discourse and Speech Acts?. Theoretical Issues in Natural Language Processing 3. The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @inproceedings{grosz1987whither, title = {{Whither Discourse and Speech Acts?}}, author = {Barbara J. Grosz}, booktitle = {Theoretical Issues in Natural Language Processing 3}, year = {1987}, doi = {10.3115/980304.980323}, url = {https://doi.org/10.3115/980304.980323} } ``` - [← previous The New Oxford English Dictionary Project](https://md.research.elamur.ai/free/en/papers/benbow-1987-new-oxford-english-dictionary-project/) - [↑ back to papers of 1987](https://md.research.elamur.ai/free/en/papers/1987/) ## References 8 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. Appelt, D. 1985 Planning English Referring Expressions. Artificial Intelligence 26: 1-33. [↗](https://doi.org/10.1016/0004-3702(85)90011-6) 2. Grosz, B.J., Joshi, A.K., and Weinstein, S. 1983 [Providing a Unified Account of Definite Noun Phrases in Discourse](https://md.research.elamur.ai/free/en/papers/grosz-1983-providing-unified-account-definite-noun/). Proc. 21st Annual Meeting of the Association for Computational Linguistics. Cambridge MAr 44-50 3. Grosz, B.J. and Sidner, C.L. 1986 [Attention, Intentions, and the Structure of Discourse](https://md.research.elamur.ai/free/en/papers/grosz-1986-attention-intentions-structure-discourse/). Computational Linguistics 12(3): 175-204. [↗](https://doi.org/10.5555/12457.12458) 4. Hobbs, J. 1979 Coherence and Coreference. Cognitive Science 3(1): 67-82. [↗](https://doi.org/10.1207/s15516709cog0301_4) 5. Kautz, H. and Allen, J. 1985 Generalized Plan Recognition. Proc. Fifth National Conference on Artificial Intelligence. Philadelphia PAr 32-37 [find](https://scholar.google.com/scholar?q=Kautz%2C+H%2E+and+Allen%2C+J%2E+1985+Generalized+Plan+Recognition%2E+Proc%2E+Fifth+National+Conference+on+Artificial+Intelligence%2E+Philadelphia+PAr+32%2D37) 6. Mann, W. and Thompson, S. 1983. Relational Propositions in Discourse. Tech. Report RR-83-115. Information Sciences Institute, Marina del Rey, CA [find](https://scholar.google.com/scholar?q=Mann%2C+W%2E+and+Thompson%2C+S%2E+1983%2E+Relational+Propositions+in+Discourse%2E+Tech%2E+Report+RR%2D83%2D115%2E+Information+Sciences+Institute%2C+Marina+del+Rey%2C+CA) 7. McKeown, K. 1985 Discourse Strategies for Generating Natural-langauge Text. Artificial Intelligence 27:1-42 [find](https://scholar.google.com/scholar?q=McKeown%2C+K%2E+1985+Discourse+Strategies+for+Generating+Natural%2Dlangauge+Text%2E+Artificial+Intelligence+27%3A1%2D42) 8. Webber, B.L. 1983 So What Can We Talk About Now? in M. Brady and R. Berwick (eds.), Computational Models of Discourse MIT Press, Cambridge MAr 331-371. [find](https://scholar.google.com/scholar?q=Webber%2C+B%2EL%2E+1983+So+What+Can+We+Talk+About+Now%3F+in+M%2E+Brady+and+R%2E+Berwick+%28eds%2E%29%2C+Computational+Models+of+Discourse+MIT+Press%2C+Cambridge+MAr+331%2D371%2E) *Updated 1 November 2026*