Pragmatic information extraction from subject ellipsis in informal English

What it is about

The paper looks at sentence-initial subject ellipsis in informal English ("(I'd) love a coffee", "(Would you) like a coffee?") and asks what it signals beyond the sentence. Three small corpora are used: family conversation transcripts, three TV drama transcripts and the Switchboard telephone corpus. Of the 181 elided subjects counted in total, 26.0% are first-person "I" and 7.2% are second-person "you", so second-person ellipsis is rare. The referent is recovered mostly from the verb and the sentence form: epistemic or private verbs (love, feel, hope) take a first-person subject, and fixed expressions such as "gonna be long" fix person and polarity. The author describes five implicatures that ellipsis can carry (interruptive, polite, intimate, friendly, less determinate), with examples, and gives a short three-step procedure: detect the ellipsis, discard imperatives, then extract the implicature. The work is qualitative; the prosodic tests that would decide the implicatures are left for future work.

Ideas

• I1. Person-recovery table for elided subjects, kept as a BTreeMap from the verb (and the verb's frame) to the person and the clause form. Rules: epistemic and private verbs give first person; a fixed phrase fixes the person; an interrogative with "like" gives second person. Measure the referent accuracy on dialogue lines of the school courses (world/src/knowledge.rs) where the subject is dropped.
• I2. Build a negative control from the paper's minimal pairs (1a/1b, 3a/3b, 7a/7b, 8a/8b, 13a to 13d): the rule must accept the first member and reject the second. Any accepted second-person ellipsis counts as an error.
• I3. Imperative versus elided subject, as a rule before tagging (en/src/tag.rs): no past-tense imperatives, and the object decides ("Tell you" is "(I'll) tell you", "Tell me" is an imperative). Test on the paper's (14) to (16) and (11) to (12) pairs.
• I4. Keep the five implicatures as candidate labels in prag, each with the cue that supports it (fast speech, object presence, intimacy marker), and report them as candidates only. The paper says prosody decides, so Mova should not claim a single label without a prosodic cue.

For Mova

The chapters are 37. Register and variation (informal register and ellipsis), 33. Implicature and presupposition (the five implicatures), 32. Politeness and hedging (indirect and polite subjects) and 38. Dialogue, discourse markers and reported speech. The code is en/src/tok.rs and en/src/tag.rs for the dropped-subject detection and the imperative check, en/src/ptag.rs for the tagging, and prag/src/infer.rs for the candidate implicature. The measure is the person-recovery accuracy on dialogue lines, and the minimal-pair control, run twice to confirm it is identical.

Topics: Implicature

Cite

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

@inproceedings{nariyama2006pragmatic,
  title = {{Pragmatic information extraction from subject ellipsis in informal English}},
  author = {Nariyama, Shigeko},
  booktitle = {Proceedings of the Third Workshop on Scalable Natural Language Understanding},
  year = {2006},
  url = {https://aclanthology.org/W06-3501/}
}

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. Haruo Aoki. 1986. Evidentials in Japanese. In Chafe & Nichols (eds). Evidentiality: the linguistic coding of epistemology. Norwood: Ablex 223-238 find
2. Penelope Brown and Stephen Levinson. 1987. Politeness: some universals in language usage. Cambridge: Cambridge University Press. find our card
3. Cote Sharon. 1996. Grammatical and Discourse Properties of Null Arguments in English, Ph.D. Dissertation, University of Pennsylvania. find
4. Barbara Fox. 1996. Studies in anaphora. Typological studies in language 33. Amsterdam: John Benjamins. find
5. Gary Gilligan. 1987. A cross-linguistic approach to the pro-drop parameter, University of Southern California Dissertation. find
6. Paul H. Grice. 1975. Logic and conversation. In P. Cole & J.L. Morgan eds., Syntax and semantics 3: Speech Acts. 43-58, New York: Academic Press. find
7. M.S.K. Halliday and R. Hasan, 1976. Cohesion in English. London: Longman. find
8. Shalom Lappin and Herbert Leass. 1994. An algorithm for pronominal anaphora resolution. Computational Linguistics, 20(4): 535-561. find our card
9. Geoffrey Leech. 1983. Principle of pragmatics. London: Longman. find
10. Stephen Levinson. 1983. Pragmatics. Cambridge: Cambridge University Press. find
11. Liliane Haegeman and Tabea Ihsane. 1999. Subject ellipsis in embedded clauses in English, English Language and Linguistics, 3(1): 117-145. find
12. Marjorie McShane. 2005. A theory of ellipsis. Oxford University Press. find
13. Shigeko Nariyama. 2003. Ellipsis and Reference-tracking in Japanese. Studies in Language Companion Series 66, Amsterdam: John Benjamins Shigeko Nariyama. 2004. Subject ellipsis in English, Journal of Pragmatics, 36(2): 237-264. find
14. John Swales. 2002. “Any last minute thoughts on this particular search?” The occurrence of sentence-initial ellipsis in research speech. In Proc. of the 4 th North American Symposium on corpus linguistics and language teaching. find
15. Jo Taylor. 2002. The quasi-modals gonna, gotta and hafta in Australian English. paper presented at Australian Linguistic Society Annual Meeting at Macquarie University Sydney. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/nariyama-2006-pragmatic-information-extraction-subject-ellipsis/