--- title: "Annotating Factive Verbs — paper by II and Miyao (2012)" description: "The paper presents an annotation scheme for factivity in English verb–complement constructions. 15 references." authors: ["Alvin Grissom II", "Yusuke Miyao"] year: 2012 venue: "Proceedings of the Eighth International Conference on Language Resources and Evaluation (LREC'12)" source: "https://aclanthology.org/L12-1443/" licence: "CC BY-NC 4.0 (ELRA, LREC proceedings)" licence_url: "https://creativecommons.org/licenses/by-nc/4.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/ii-2012-annotating-factive-verbs/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/ii-2012-annotating-factive-verbs/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2012](https://md.research.elamur.ai/free/en/papers/2012/page/2/) / Annotating Factive Verbs # Annotating Factive Verbs **Authors:** Alvin Grissom II and [Yusuke Miyao](https://md.research.elamur.ai/free/en/authors/yusuke-miyao/) · **Year:** 2012 · **Venue:** [Proceedings of the Eighth International Conference on Language Resources and Evaluation (LREC'12)](https://md.research.elamur.ai/free/en/venues/lrec/) ([LREC 2012](https://md.research.elamur.ai/free/en/venues/lrec/2012/)) · **Publisher:** [European Language Resources Association (ELRA)](https://md.research.elamur.ai/free/en/publishers/european-language-resources-association-elra/) · **Licence:** [CC BY-NC 4.0 (ELRA, LREC proceedings)](https://creativecommons.org/licenses/by-nc/4.0/) **Links:** [Publisher's page](https://aclanthology.org/L12-1443/) · [PDF](https://aclanthology.org/L12-1443.pdf) · [OpenAlex](https://openalex.org/W2250217290) **Citations:** 0 citations (OpenAlex, 2026-10-14) ## Summary The paper presents an annotation scheme for factivity in English verb–complement constructions. Factive verbs such as *regret* or *reveal* presuppose that their complement clause is true. That makes them useful for textual entailment, yet no existing corpus marked them in this way (PropBank and FactBank cover related but different ground). The authors work through the difficult cases: lexical ambiguity (*I'm afraid I don't know*), contextual coercion (*remember* followed by a correction), claim verbs (*say*, *announce*, *admit*), and performatives, where truth depends on timing and on the speaker's authority. From this analysis they build a lightweight XML tag set. It links verb cues to their clauses through id/ref attributes and can also record lemma, nominalization, ambiguity, coercion and agent authority. They applied it to sentences from the PASCAL RTE2 corpus. The paper reports no corpus size or agreement figures. Its contribution is the scheme itself, the reasoning behind each design choice, and a preliminary pattern-mining experiment on UKWAC for discovering factive verbs automatically. It matters because it offers a practical starting point for systems that need to recover implicit facts in a text. ## Key points - **Tag set:** - ``/`` mark a factive verb and its clause. - ``/`` mark a non-factive verb and its clause. - ``/`` mark a claim verb and its claim. - Optional tags are ``, ``, `` (e.g., `type="authority"`) and ``. - Attributes `id`/`ref`, `lemma` and `form` link elements and record nominalization. - **Data and scope:** The annotated data is the PASCAL RTE2 entailment corpus. Only verb–complement pairs are covered; adjectives, adverbs and other presupposition triggers are left out. - **Performatives:** Annotators judge whether the complement is true at the moment of utterance. - *I hereby declare the treaty null* counts as non-factive. - A past-tense report of a performative counts as factive. - A declaration by a speaker without authority counts as non-factive. - **Claim verbs:** Verbs like *admit* may carry both a factive tag and a claim tag. *Announce* is singled out as persistently ambiguous. - **Counter-factivity:** It is rare in entailment data (citing Garoufi 2007), so it is merged into non-factivity. - **Future work:** The authors searched UKWAC (about 2 billion words) for POS patterns such as *have + past participle + that* and *should + verb + that*. They then estimated, per verb, how likely it is to fall in the factive or non-factive group. ## Interesting ideas - Instead of asking annotators to label the verb or the whole sentence, ask one question about the complement: is it true at speech time? This sidesteps tangled performative philosophy and keeps labels consistent. - Treat factivity as a graded default per verb that context can override, and record the override with an explicit coercion flag. A contrastive *but … actually* continuation, for example, cancels the presupposition of *remember*. - Keep "claim" separate from "fact," and add an agent-authority attribute, so that reported speech by officials is neither discarded nor blindly believed. - Allow a verb to carry overlapping labels (factive and claim at once) plus an ambiguity marker, instead of forcing one choice. Unresolved cases then stay visible as data for later learning. - Mine a large corpus for cheap surface patterns around *that*-complements and compute per-verb factive versus non-factive odds. This bootstraps a factivity lexicon without manual labels. ## For Mova How these ideas could serve Mova, our small open language model — its files, stages and tests: - **I1** (frames): The prag crate could decide whether an embedded clause becomes a world fact by checking the complement's truth at utterance time, rather than by trusting the verb's lexical label alone. - **I2** (world): Verb-class knowledge could store a default factivity value, and the world builder could retract a presupposed fact when a later clause contradicts it, logging the retraction as an explainable step. - **I3** (world): Entities' states could carry a provenance tag ("claimed by X, authority=yes"), letting QA answer "who says so" separately from "is it so." - **I4** (learn): Silver data from witness parsers or LLM teachers could keep multi-label and "ambiguous" outcomes as legitimate classes, which also gives natural negative controls. - **I5** (learn): Run similar pattern statistics over UD trees (ccomp with a *that* marker under modal or perfect auxiliaries) to induce initial factivity priors for verb classes. *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 Colorado Boulder](https://md.research.elamur.ai/free/en/universities/university-of-colorado-boulder/) — University, Boulder, United States · 52 works · 49,647 citations - [National Institute of Informatics](https://md.research.elamur.ai/free/en/labs/national-institute-of-informatics/) — Research institute, Tokyo, Japan · 21 works · 3,045 citations 📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 9,086 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 8,889 works · [🎭 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 🧭 Topics (1): [English](https://md.research.elamur.ai/free/en/topics/english/) · 1,373 works ## Cite Cite as: Alvin Grissom II and Yusuke Miyao. 2012. Annotating Factive Verbs. Proceedings of the Eighth International Conference on Language Resources and Evaluation (LREC'12). The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @inproceedings{ii2012annotating, title = {{Annotating Factive Verbs}}, author = {Alvin Grissom II and Yusuke Miyao}, booktitle = {Proceedings of the Eighth International Conference on Language Resources and Evaluation (LREC'12)}, year = {2012}, doi = {10.63317/25yvrsgxemkk}, url = {https://doi.org/10.63317/25yvrsgxemkk} } ``` - [← previous A rule-based approach to unknown word recognition in Arabic](https://md.research.elamur.ai/free/en/papers/cahill-2012-rule-based-approach-unknown-word/) - [next → Arabic Dialects of Central Asia: Internal Development Tendencies of Arabic Language Material](https://md.research.elamur.ai/free/en/papers/chikovani-2026-arabic-dialects-central-asia-internal/) - [↑ back to papers of 2012](https://md.research.elamur.ai/free/en/papers/2012/page/2/) ## References 15 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. J.L. Austin. 1975. How to do things with words. Harvard Univ Pr. [↗](https://doi.org/10.1093/acprof:oso/9780198245537.001.0001) 2. R. Bar-Haim, I. Dagan, B. Dolan, L. Ferro, D. Giampiccolo, B. Magnini, and I. Szpektor. 2006. The second PASCAL recognising textual entailment challenge. In Proceedings of the second PASCAL challenges workshop on recognising textual entailment, pages 1–9. Citeseer. [find](https://scholar.google.com/scholar?q=R%2E+Bar%2DHaim%2C+I%2E+Dagan%2C+B%2E+Dolan%2C+L%2E+Ferro%2C+D%2E+Giampiccolo%2C+B%2E+Magnini%2C+and+I%2E+Szpektor%2E+2006%2E+The+second+PASCAL+recognising+textual+entailment+challenge%2E+In+Proceedings+of+the+second+PASCAL+challenges) 3. D.I. Beaver. 2001. Presupposition and assertion in dynamic semantics. CSLI publications. [find](https://scholar.google.com/scholar?q=D%2EI%2E+Beaver%2E+2001%2E+Presupposition+and+assertion+in+dynamic+semantics%2E+CSLI+publications%2E) 4. D. Clausen and C.D. Manning. 2009. Presupposed content and entailments in natural language inference. In Proceedings of the 2009 Workshop on Applied Textual Inference, pages 70–73. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=D%2E+Clausen+and+C%2ED%2E+Manning%2E+2009%2E+Presupposed+content+and+entailments+in+natural+language+inference%2E+In+Proceedings+of+the+2009+Workshop+on+Applied+Textual+Inference%2C+pages+70%E2%80%9373%2E+Association+for+Com) 5. A. Ferraresi, E. Zanchetta, M. Baroni, and S. Bernardini. 2008. Introducing and evaluating ukwac, a very large web-derived corpus of english. In Proceedings of the 4th Web as Corpus Workshop (WAC-4) Can we beat Google, pages 47–54. [find](https://scholar.google.com/scholar?q=A%2E+Ferraresi%2C+E%2E+Zanchetta%2C+M%2E+Baroni%2C+and+S%2E+Bernardini%2E+2008%2E+Introducing+and+evaluating+ukwac%2C+a+very+large+web%2Dderived+corpus+of+english%2E+In+Proceedings+of+the+4th+Web+as+Corpus+Workshop+%28WAC%2D4%29+C) 6. K. Garoufi. 2007. Towards a better understanding of applied textual entailment. [find](https://scholar.google.com/scholar?q=K%2E+Garoufi%2E+2007%2E+Towards+a+better+understanding+of+applied+textual+entailment%2E) 7. G. Gazdar. 1979. Pragmatics: Implicature, presupposition, and logical form. Academic Press New York. [find](https://scholar.google.com/scholar?q=G%2E+Gazdar%2E+1979%2E+Pragmatics%3A+Implicature%2C+presupposition%2C+and+logical+form%2E+Academic+Press+New+York%2E) 8. A. Giannakidou. 2006. Only, emotive factive verbs, and the dual nature of polarity dependency. Language, 82(3):575–603. [↗](https://doi.org/10.1353/lan.2006.0136) 9. I. Heim. 1990. Presupposition projection. In Reader for the Nijmegen Workshop on Presupposition, Lexical Meaning, and Discourse Processes, Univ. of Nijmegen. [find](https://scholar.google.com/scholar?q=I%2E+Heim%2E+1990%2E+Presupposition+projection%2E+In+Reader+for+the+Nijmegen+Workshop+on+Presupposition%2C+Lexical+Meaning%2C+and+Discourse+Processes%2C+Univ%2E+of+Nijmegen%2E) 10. P. Kiparsky. 1970. Semantic rules in grammar. The Nordic languages and modern linguistics, pages 262–285. [find](https://scholar.google.com/scholar?q=P%2E+Kiparsky%2E+1970%2E+Semantic+rules+in+grammar%2E+The+Nordic+languages+and+modern+linguistics%2C+pages+262%E2%80%93285%2E) 11. R. Nairn, C. Condoravdi, and L. Karttunen. 2006. [Computing relative polarity for textual inference](https://md.research.elamur.ai/free/en/papers/nairn-2006-computing-relative-polarity-textual-inference/). In Proceedings of ICoS-5. Citeseer. 12. R. Prasad, N. Dinesh, A. Lee, A. Joshi, and B. Webber. 2006. Attribution and its annotation in the Penn Discourse Treebank. Traitement Automatique des Langues, 47(2):43–63. [find](https://scholar.google.com/scholar?q=R%2E+Prasad%2C+N%2E+Dinesh%2C+A%2E+Lee%2C+A%2E+Joshi%2C+and+B%2E+Webber%2E+2006%2E+Attribution+and+its+annotation+in+the+Penn+Discourse+Treebank%2E+Traitement+Automatique+des+Langues%2C+47%282%29%3A43%E2%80%9363%2E) 13. J. Pustejovsky, P. Hanks, R. Sauri, A. See, R. Gaizauskas, A. Setzer, D. Radev, B. Sundheim, D. Day, L. Ferro, et al. 2003. The TimeBank corpus. In Corpus Linguistics, volume 2003, page 40. [find](https://scholar.google.com/scholar?q=J%2E+Pustejovsky%2C+P%2E+Hanks%2C+R%2E+Sauri%2C+A%2E+See%2C+R%2E+Gaizauskas%2C+A%2E+Setzer%2C+D%2E+Radev%2C+B%2E+Sundheim%2C+D%2E+Day%2C+L%2E+Ferro%2C+et+al%2E+2003%2E+The+TimeBank+corpus%2E+In+Corpus+Linguistics%2C+volume+2003%2C+page+40%2E) 14. R. Saurı́ and J. Pustejovsky. 2009. FactBank: a corpus annotated with event factuality. Language Resources and Evaluation, 43(3):227–268. [find](https://scholar.google.com/scholar?q=R%2E+Saur%C4%B1%CC%81+and+J%2E+Pustejovsky%2E+2009%2E+FactBank%3A+a+corpus+annotated+with+event+factuality%2E+Language+Resources+and+Evaluation%2C+43%283%29%3A227%E2%80%93268%2E) 15. J.R. Searle. 1969. [Speech acts](https://md.research.elamur.ai/free/en/papers/searle-1969-speech-acts/): An essay in the philosophy of language. Cambridge university press. [↗](https://doi.org/10.1017/cbo9781139173438) *Updated 24 October 2026*