---
title: "Classifying Speech Acts using Verbal Response Modes (2006)"
description: "The authors describe this as the first automatic VRM classifier. 79.75% accuracy. 26 references, cited by 6 papers here."
authors: ["Andrew Lampert", "Robert Dale", "Cécile Paris"]
year: 2006
venue: "Proceedings of the Australasian Language Technology Workshop 2006"
source: "https://aclanthology.org/U06-1007/"
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/lampert-2006-classifying-speech-acts-verbal-response/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/lampert-2006-classifying-speech-acts-verbal-response/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2006](https://md.research.elamur.ai/free/en/papers/2006/page/2/) / Classifying Speech Acts using Verbal Response Modes

# Classifying Speech Acts using Verbal Response Modes

**Authors:** Andrew Lampert, [Robert Dale](https://md.research.elamur.ai/free/en/authors/robert-dale/) and [Cécile Paris](https://md.research.elamur.ai/free/en/authors/cecile-paris/) · **Year:** 2006 · **Venue:** [Proceedings of the Australasian Language Technology Workshop 2006](https://md.research.elamur.ai/free/en/venues/alta/) ([ALTA 2006](https://md.research.elamur.ai/free/en/venues/alta/2006/)) · **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/U06-1007/) · [PDF](https://aclanthology.org/U06-1007.pdf) · [OpenAlex](https://openalex.org/W777401746)

**Citations:** 26 citations (OpenAlex, 2026-10-14) · [6](https://md.research.elamur.ai/free/en/papers/lampert-2006-classifying-speech-acts-verbal-response/cited-by/) in our library, of the 149,439 papers whose bibliographies we read

## Summary
The authors want to help email users track open requests and commitments. To get there, they need a domain-independent way to label the communicative function of each utterance. Existing dialogue act schemes tend to be tied to particular tasks, so they adopt Stiles' Verbal Response Modes (VRM). VRM derives its eight categories from three binary principles: whose experience is the topic, who makes presumptions regarding that experience, and whose viewpoint serves as the frame of reference. Each principle takes the value speaker or other. Every utterance gets two codes, one for its literal form and one for its pragmatic intent, which captures indirect speech acts in a systematic way.

This work tackles only the literal dimension. The classifier draws on two kinds of features:
- cues taken from Stiles' annotation manual, such as pronoun person, imperative verbs, inverted word order and small word lists;
- statistical features, such as n-grams, the first word, the last token and utterance length.

Using SVM on a corpus of 1368 utterances, the full feature set reaches 79.75% accuracy. The manual-derived cues alone reach 60.82%. The authors describe this as the first automatic VRM classifier. They intend it as a building block for linking speech acts across email threads.

## Key points
- VRM defines 8 literal modes: Disclosure, Edification, Advisement, Confirmation, Question, Acknowledgement, Interpretation and Reflection. Crossing literal and pragmatic codes gives 64 form–intent pairs, of which 8 are "pure" and 56 are "mixed". For example, "Can you pass the sugar?" is coded QA.
- Results on 1368 utterances:
  - SVM: 79.75% with all features, 60.82% with only Stiles' cues, 74.49% without Stiles' cues.
  - Bayes Net: 78.51%, 60.16% and 75.68% on the same three feature sets.
  - Baseline: about 49%.
- Morphological tags and imperative detection come from the Connexor parser. N-grams are kept only if they occur at least 5 times.
- The authors built each hand-made lexical list (permission verbs, interrogatives, yes/no variants, evaluation terms) in under an hour, expanding seed words with online thesauri.
- Interpretation and Reflection score worst (recall 0.453 and 0.514) and are often confused with each other. Larger classes such as Disclosure and Question score best (precision 0.905 and 0.839).

## Interesting ideas
- Build the label set by crossing a few binary principles (whose experience, who presumes it, whose viewpoint) instead of listing acts ad hoc, so the scheme needs no "other" bucket and stays comparable across domains.
- Code every utterance twice with the same inventory, once for surface form and once for intent, so an indirect request becomes a clean "form-in-service-of-intent" pair.
- Turn human annotation guidelines directly into features (pronoun person, imperative mood, subject–verb inversion), then check how much shallow statistical features add on top.
- Ablate feature families (only guideline cues / everything except them / all) to show which source of knowledge carries the signal.
- Error analysis tied to the taxonomy's structure: the most-confused pair (Interpretation vs Reflection) differs on just one principle, and some Reflections echo the previous turn, which isolated-utterance classification cannot see.

## For Mova

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

- **I1** (frames): crate prag could compute speech-act labels as a triple of speaker/other values read off the clause, which makes each label explainable by naming the principle that decided it.
- **I2** (frames): prag could store literal and pragmatic acts as two slots on the same frame, which would let implicature and irony be represented as a mismatch between the slots rather than as a separate mechanism.
- **I3** (frames): the guideline cues map naturally onto UD features (Person, Mood=Imp, word order in the tree), giving rule-based act detection whose gap to a statistical model can be measured explicitly.
- **I4** (learn): the same three-way ablation could serve as a control when evaluating hand rules versus learned components, revealing whether the rules contribute beyond lexical shortcuts.
- **I5** (frames): when acts are confused, check which single principle separates them and add discourse context (e.g., coreference to the prior turn) specifically to resolve that dimension.

*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

- [Macquarie University](https://md.research.elamur.ai/free/en/universities/macquarie-university/) — University, Sydney, Australia · 44 works · 5,778 citations
- [Commonwealth Scientific and Industrial Research Organisation](https://md.research.elamur.ai/free/en/government/commonwealth-scientific-and-industrial-research-organisation/) — Government body, Canberra, Australia · 17 works · 2,064 citations

📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [🎭 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: Andrew Lampert, Robert Dale and Cécile Paris. 2006. Classifying Speech Acts using Verbal Response Modes. Proceedings of the Australasian Language Technology Workshop 2006.

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

```bibtex
@inproceedings{lampert2006classifying,
  title = {{Classifying Speech Acts using Verbal Response Modes}},
  author = {Andrew Lampert, Robert Dale and Cécile Paris},
  booktitle = {Proceedings of the Australasian Language Technology Workshop 2006},
  year = {2006},
  url = {https://aclanthology.org/U06-1007/}
}
```

- [← previous Measuring annotator agreement in a complex hierarchical dialogue act annotation scheme](https://md.research.elamur.ai/free/en/papers/geertzen-2006-measuring-annotator-agreement-complex-hierarchical/)
- [next → Coreference Resolution with and without Linguistic Knowledge](https://md.research.elamur.ai/free/en/papers/uryupina-2006-coreference-resolution-without-linguistic-knowledge/)
- [↑ back to papers of 2006](https://md.research.elamur.ai/free/en/papers/2006/page/2/)

## References

26 references, 4 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Jan Alexandersson, Bianka Buschbeck-Wolf, Tsutomu Fujinami, Michael Kipp, Stephan Koch, Elisabeth Maier, Norbert Reithinger, Birte Schmitz, and Melanie Siegel. 1998. Dialogue acts in VERBMOBIL-2. Technical Report Verbmobil-Report 226, DFKI Saarbruecken, Universitt Stuttgart, Technische Universitt Berlin, Universitt des Saarlandes. 2nd Edition. [↗](https://doi.org/10.22028/d291-25326)
2. Victoria Bellotti, Nicolas Ducheneaut, Mark Howard, and Ian Smith. 2003. Taking email to task: The design and evaluation of a task management centred email tool. In Computer Human Interaction Conference, CHI, Ft Lauderdale, Florida, USA, April 5-10. [find](https://scholar.google.com/scholar?q=Victoria+Bellotti%2C+Nicolas+Ducheneaut%2C+Mark+Howard%2C+and+Ian+Smith%2E+2003%2E+Taking+email+to+task%3A+The+design+and+evaluation+of+a+task+management+centred+email+tool%2E+In+Computer+Human+Interaction+Conferen)
3. Jennifer Chu-Carroll, 1998. Applying Machine Learning to Discourse Processing. Papers from the 1998 AAAI Spring Symposium., chapter A statistical model for discourse act recognition in dialogue interactions, pages 12–17. The AAAI Press, Menlo Park, California. [find](https://scholar.google.com/scholar?q=Jennifer+Chu%2DCarroll%2C+1998%2E+Applying+Machine+Learning+to+Discourse+Processing%2E+Papers+from+the+1998+AAAI+Spring+Symposium%2E%2C+chapter+A+statistical+model+for+discourse+act+recognition+in+dialogue+intera)
4. William W Cohen, Vitor R Carvalho, and Tom M Mitchell. 2004. [Learning to classify email into ”speech acts”](https://md.research.elamur.ai/free/en/papers/cohen-2004-learning-classify-email-into-speech/). In Dekang Lin and Dekai Wu, editors, Conference on Empirical Methods in Natural Language Processing, pages 309–316, Barcelona, Spain. Association for Computational Linguistics.
5. Mark Core and James Allen. 1997. Coding dialogs with the DAMSL annotation scheme. In AAAI Fall Symposium on Communicative Action in Humans and Machines, pages 28–35, Cambridge, MA, November. [find](https://scholar.google.com/scholar?q=Mark+Core+and+James+Allen%2E+1997%2E+Coding+dialogs+with+the+DAMSL+annotation+scheme%2E+In+AAAI+Fall+Symposium+on+Communicative+Action+in+Humans+and+Machines%2C+pages+28%E2%80%9335%2C+Cambridge%2C+MA%2C+November%2E)
6. Mark Core. 1998. Predicting DAMSL utterance tags. In Proceedings of the AAAI-98 Spring Symposium on Applying Machine Learning to Discourse Processing. [find](https://scholar.google.com/scholar?q=Mark+Core%2E+1998%2E+Predicting+DAMSL+utterance+tags%2E+In+Proceedings+of+the+AAAI%2D98+Spring+Symposium+on+Applying+Machine+Learning+to+Discourse+Processing%2E)
7. Simon H Corston-Oliver, Eric Ringger, Michael Gamon, and Richard Campbell. 2004. Task-focused summarization of email. In ACL-04 Workshop: Text Summarization Branches Out, July. [find](https://scholar.google.com/scholar?q=Simon+H+Corston%2DOliver%2C+Eric+Ringger%2C+Michael+Gamon%2C+and+Richard+Campbell%2E+2004%2E+Task%2Dfocused+summarization+of+email%2E+In+ACL%2D04+Workshop%3A+Text+Summarization+Branches+Out%2C+July%2E)
8. John J Godfrey, Edward C Holliman, and Jane Mc-Daniel. 1992. SWITCHBOARD: Telephone speech corpus for research and development. In Proceedings of the IEEE Conference on Acoustics, Speech and Signal Processing, volume 1, pages 517–520, San Francisco, CA, March. [find](https://scholar.google.com/scholar?q=John+J+Godfrey%2C+Edward+C+Holliman%2C+and+Jane+Mc%2DDaniel%2E+1992%2E+SWITCHBOARD%3A+Telephone+speech+corpus+for+research+and+development%2E+In+Proceedings+of+the+IEEE+Conference+on+Acoustics%2C+Speech+and+Signal+Pr)
9. Susan L Kline, C L Hennen, and K M Farrell. 1990. Cognitive complexity and verbal response mode use in discussion. Communication Quarterly, 38:350– 360. [↗](https://doi.org/10.1080/01463379009369772)
10. Anton Leuski. 2004. Email is a stage: discovering people roles from email archives. In Proceedings of Annual ACM Conference on Research and Development in Information Retrieval, Sheffi eld, UK. [find](https://scholar.google.com/scholar?q=Anton+Leuski%2E+2004%2E+Email+is+a+stage%3A+discovering+people+roles+from+email+archives%2E+In+Proceedings+of+Annual+ACM+Conference+on+Research+and+Development+in+Information+Retrieval%2C+Sheffi+eld%2C+UK%2E)
11. Karen J McGaughey and William B Stiles. 1983. Courtroom interrogation of rape victims: Verbal response mode use by attourneys and witnesses during direct examination vs. cross-examination. Journal of Applied Social Psychology, 13:78–87. [find](https://scholar.google.com/scholar?q=Karen+J+McGaughey+and+William+B+Stiles%2E+1983%2E+Courtroom+interrogation+of+rape+victims%3A+Verbal+response+mode+use+by+attourneys+and+witnesses+during+direct+examination+vs%2E+cross%2Dexamination%2E+Journal+of+)
12. Ludwein Meeuswesen, Cas Schaap, and Cees van der Staak. 1991. Verbal analysis of doctor-patient communication. Social Science and Medicine, 32(10):1143–50. [find](https://scholar.google.com/scholar?q=Ludwein+Meeuswesen%2C+Cas+Schaap%2C+and+Cees+van+der+Staak%2E+1991%2E+Verbal+analysis+of+doctor%2Dpatient+communication%2E+Social+Science+and+Medicine%2C+32%2810%29%3A1143%E2%80%9350%2E)
13. Diana S Rak and Linda M McMullen. 1987. Sex-role stereotyping in television commercials: A verbal response mode and content analysis. Canadian Journal of Behavioural Science, 19:25–39. [↗](https://doi.org/10.1037/h0079873)
14. Norbert Reithinger and Martin Klesen. 1997. Dialogue act classifi cation using language models. In Proceedings of Eurospeech ’97, pages 2235–2238, Rhodes, Greece. [find](https://scholar.google.com/scholar?q=Norbert+Reithinger+and+Martin+Klesen%2E+1997%2E+Dialogue+act+classifi+cation+using+language+models%2E+In+Proceedings+of+Eurospeech+%E2%80%9997%2C+pages+2235%E2%80%932238%2C+Rhodes%2C+Greece%2E)
15. John 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)
16. John R Searle. 1979. Expression and Meaning. Cambridge University Press. [↗](https://doi.org/10.1017/cbo9780511609213)
17. William B Stiles, Melinda L Au, Mary Ann Martello, and Julia A Perlmutter. 1983. American campaign oratory: Verbal response mode use by candidates in the 1980 american presidential primaries. Social Behaviour and Personality, 11:39–43. [↗](https://doi.org/10.2224/sbp.1983.11.1.39)
18. William B Stiles. 1992. Describing Talk: a taxonomy of verbal response modes. SAGE Series in Interpersonal Communication. SAGE Publications. ISBN: 0803944659. [find](https://scholar.google.com/scholar?q=William+B+Stiles%2E+1992%2E+Describing+Talk%3A+a+taxonomy+of+verbal+response+modes%2E+SAGE+Series+in+Interpersonal+Communication%2E+SAGE+Publications%2E+ISBN%3A+0803944659%2E)
19. Andreas Stolcke, Klaus Ries, Noah Coccaro, Elizabeth Shriberg, Rebecca Bates, Daniel Jurafsky, Paul Taylor, Rachel Martin, Marie Meteer, and Carol Van Ess-Dykema. 2000. [Dialogue act modeling for automatic tagging and recognition of conversational speech](https://md.research.elamur.ai/free/en/papers/stolcke-2000-dialogue-act-modeling-automatic-tagging/). Computational Linguistics, 26(3):339–371. [↗](https://doi.org/10.1162/089120100561737)
20. Pasi Tapanainen and Timo Jarvinen. 1997. [A non-projective dependency parser](https://md.research.elamur.ai/free/en/papers/tapanainen-1997-non-projective-dependency-parser/). In Proceedings of the 5th Conference on Applied Natural Language Processing, pages 64–71, Washington D.C. Association for Computational Linguistics.
21. Jan M Ulijn and Maurits J Verweij. 2000. Question behaviour in monocultural and intercultural business negotiations: the Dutch-Spanish connection. Discourse Studies, 2(1):141–172. [find](https://scholar.google.com/scholar?q=Jan+M+Ulijn+and+Maurits+J+Verweij%2E+2000%2E+Question+behaviour+in+monocultural+and+intercultural+business+negotiations%3A+the+Dutch%2DSpanish+connection%2E+Discourse+Studies%2C+2%281%29%3A141%E2%80%93172%2E)
22. Nick Webb, Mark Hepple, and Yorick Wilks. 2005. Dialogue act classifi cation based on intra-utterance features. In Proceedings of the AAAI Workshop on Spoken Language Understanding. [find](https://scholar.google.com/scholar?q=Nick+Webb%2C+Mark+Hepple%2C+and+Yorick+Wilks%2E+2005%2E+Dialogue+act+classifi+cation+based+on+intra%2Dutterance+features%2E+In+Proceedings+of+the+AAAI+Workshop+on+Spoken+Language+Understanding%2E)
23. Steve Whittaker and Candace Sidner. 1996. Email overload: exploring personal information management of email. In ACM Computer Human Interaction conference, pages 276–283. ACM Press. [find](https://scholar.google.com/scholar?q=Steve+Whittaker+and+Candace+Sidner%2E+1996%2E+Email+overload%3A+exploring+personal+information+management+of+email%2E+In+ACM+Computer+Human+Interaction+conference%2C+pages+276%E2%80%93283%2E+ACM+Press%2E)
24. Terry Winograd and Fernando Flores. 1986. Understanding Computers and Cognition. Ablex Publishing Corporation, Norwood, New Jersey, USA, 1st edition. ISBN: 0-89391-050-3. [find](https://scholar.google.com/scholar?q=Terry+Winograd+and+Fernando+Flores%2E+1986%2E+Understanding+Computers+and+Cognition%2E+Ablex+Publishing+Corporation%2C+Norwood%2C+New+Jersey%2C+USA%2C+1st+edition%2E+ISBN%3A+0%2D89391%2D050%2D3%2E)
25. Susan Wiser and Marvin R Goldfried. 1996. Verbal interventions in signifi cant psychodynamic-interpersonal and cognitive-behavioral therapy sessions. Psychotherapy Research, 6(4):309–319. [find](https://scholar.google.com/scholar?q=Susan+Wiser+and+Marvin+R+Goldfried%2E+1996%2E+Verbal+interventions+in+signifi+cant+psychodynamic%2Dinterpersonal+and+cognitive%2Dbehavioral+therapy+sessions%2E+Psychotherapy+Research%2C+6%284%29%3A309%E2%80%93319%2E)
26. Ian Witten and Eiba Frank. 2005. Data Mining: Practical machine learning tools and techniques. Morgan Kaufmann, San Francisco, 2nd edition. [find](https://scholar.google.com/scholar?q=Ian+Witten+and+Eiba+Frank%2E+2005%2E+Data+Mining%3A+Practical+machine+learning+tools+and+techniques%2E+Morgan+Kaufmann%2C+San+Francisco%2C+2nd+edition%2E)

*Updated 1 November 2026*