Elamur Research⌕
☰
home / papers / 2012 / Towards a model of formal and informal address in English

Towards a model of formal and informal address in English

Paper · 2012 · Manaal Faruqui and Sebastian Padó

Summary

Modern English no longer marks formal versus informal address (the T/V distinction) through pronouns. This gap hurts tasks such as translating English into languages that must choose between forms like du and Sie. Faruqui and Padó ask whether the distinction can still be recovered from English text.

They built a labelled corpus without manual annotation. Using parallel 19th–early 20th century novels from Project Gutenberg and Project Gutenberg-DE, they tagged German sentences as T or V with high-precision pronoun rules. They then transferred each label to the aligned English sentence.

A human study showed that people can label English sentences as T or V reasonably well, but only when they see surrounding context. Supervised classifiers trained on the projected labels reached about 67–68% accuracy. Plain word features worked best, while semantic classes and features derived from Politeness Theory barely beat a frequency baseline.

The analysis found an asymmetry. Formal speech relies on conventional cues such as titles and set phrases. Informal speech is signalled by cues specific to each novel. This matters for MT and for extracting social relations.

Key points

  • Corpus: 110 stories and novels available in both English and German (about 0.5M sentences per language). The pipeline used the EUROPARL splitter/tokenizer, TreeTagger, Gargantuan for sentence alignment and Giza++ for word alignment.
  • German labelling rules:
    • du was labelled T, except where it acts as a French nobiliary particle (detected by neighbours with an "unknown" lemma).
    • Only capitalized Sie outside utterance-initial position was labelled V.
    • ihr was dropped entirely.
    • About 14% of German sentences received a label (37K V, 28K T), with an estimated precision above 99%.
  • Projection: labels were transferred at sentence level rather than through word alignment, because pronoun alignment was unreliable. This yielded 25K V and 18K English sentences. The 255 sentences labelled both T and V were discarded. The data was split by novel: 74 novels for training, 19 for development and 13 for test (26K / 9K / 8K sentences).
  • Human annotation (200 sentences):
    • Agreement between the two annotators rose from 75% (κ .49) without context to 79% (κ .58) with three sentences on each side.
    • When the annotators agreed with each other, they matched the projected labels 79% of the time.
  • Classifiers:
    • The best cross-novel model used words with an 8-sentence direct-speech context and reached about 67–68% accuracy.
    • Models trained on individual novels reached 80% (micro average over 7 novels) but generalize worse to other novels.

Interesting ideas

  • I1. Labels for a feature that English leaves unmarked can be harvested from a language that marks it overtly, using precision-oriented rules and accepting low coverage.
  • I2. Projecting at sentence level instead of through noisy word alignment avoids errors with function words, and it works because address rarely changes within a sentence.
  • I3. Testing human raters with and without context shows how much a pragmatic label depends on discourse rather than on the sentence alone.
  • I4. Formal register is conventional and transfers across texts, while informal register depends on the particular text and characters. The authors suggest modelling V as a single class and treating T as the default.
  • I5. Address is fixed per pair of interlocutors rather than chosen sentence by sentence, so it is better modelled as a property of a relationship.

For Mova

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

  • I1 (learn): German or Ukrainian translations could supply silver labels for formality or addressee status on English sentences, without hand annotation.
  • I2 (learn): when building silver data from parallel text, Mova could project coarse, stable properties at a larger unit and check consistency within that unit as a filter.
  • I3 (frames): speech-act or politeness decisions in prag could be evaluated in both conditions, to show whether a frame is justified locally or needs the surrounding dialogue.
  • I4 (frames): a formality detector could fire on explicit conventional cues such as titles and formulaic phrases, defaulting to informal otherwise, which keeps each decision traceable to a concrete cue.
  • I5 (world): store formality as an attribute of the speaker–addressee relation among the text's entities, and use it to constrain every turn between that pair.

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

🎓Heidelberg UniversityHeidelberg · Germany39 works · 41,356 citations🎓Indian Institute of Technology KharagpurKharagpur · India25 works · 714 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksArtificial Intelligence · 17,965 worksArts and Humanities · 17,450 worksLanguage and Linguistics · 17,296 works

🧭 Topics 1Research subjects and languages the work is about.English · 2,379 works

Cite

Cite as Manaal Faruqui and Sebastian Padó. 2012. Towards a model of formal and informal address in English. Proceedings of the 13th Conference of the European Chapter of the Association for Computational Linguistics.

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

@inproceedings{faruqui2012model,
  title = {{Towards a model of formal and informal address in English}},
  author = {Manaal Faruqui and Sebastian Padó},
  booktitle = {Proceedings of the 13th Conference of the European Chapter of the Association for Computational Linguistics},
  year = {2012},
  url = {https://aclanthology.org/E12-1064/}
}

References

35 references, 6 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. John Ardila. 2003. (Non-Deictic, Socio-Expressive) T-/V-Pronoun Distinction in Spanish/English Formal Locutionary Acts. Forum for Modern Language Studies, 39(1):74–86. ↗
  2. John A. Bateman. 1988. Aspects of clause politeness in Japanese: An extended inquiry semantics treatment. In Proceedings of ACL, pages 147–154, Buffalo, New York. find
  3. Luisa Bentivogli and Emanuele Pianta. 2005. Exploiting parallel texts in the creation of multilingual semantically annotated resources: the MultiSemCor Corpus. Journal of Natural Language Engineering, 11(3):247–261. ↗
  4. Adam Bermingham and Alan F. Smeaton. 2009. A study of inter-annotator agreement for opinion retrieval. In Proceedings of ACM SIGIR, pages 784– 785. find
  5. Philip Bramsen, Martha Escobar-Molano, Ami Patel, and Rafael Alonso. 2011. Extracting social power relationships from natural language. In Proceedings of ACL/HLT, pages 773–782, Portland, OR. find
  6. Fabienne Braune and Alexander Fraser. 2010. Improved unsupervised sentence alignment for symmetrical and asymmetrical parallel corpora. In Coling 2010: Posters, pages 81–89, Beijing, China. find
  7. Roger Brown and Albert Gilman. 1960. The pronouns of power and solidarity. In Thomas A. Sebeok, editor, Style in Language, pages 253–277. MIT Press, Cambridge, MA. find
  8. Penelope Brown and Stephen C. Levinson. 1987. Politeness: Some Universals in Language Usage. Number 4 in Studies in Interactional Sociolinguistics. Cambridge University Press.
  9. Alexander Clark. 2003. Combining distributional and morphological information for part of speech induction. In Proceedings of EACL, pages 59–66, Budapest, Hungary.
  10. J. Cohen. 1960. A Coefficient of Agreement for Nominal Scales. Educational and Psychological Measurement, 20(1):37–46. ↗
  11. David Elson, Nicholas Dames, and Kathleen McKeown. 2010. Extracting social networks from literary fiction. In Proceedings of ACL, pages 138–147, Uppsala, Sweden. ↗
  12. Rong-En Fan, Kai-Wei Chang, Cho-Jui Hsieh, Xiang-Rui Wang, and Chih-Jen Lin. 2008. LIBLINEAR: A library for large linear classification. Journal of Machine Learning Research, 9:1871–1874. ↗
  13. Manaal Faruqui and Sebastian Padó. 2011. “I Thou Thee, Thou Traitor”: Predicting formal vs. informal address in English literature. In Proceedings of ACL/HLT 2011, pages 467–472, Portland, OR.
  14. Jenny Rose Finkel and Christopher D. Manning. 2009. Nested named entity recognition. In Proceedings of EMNLP, pages 141–150, Singapore.
  15. Joseph L. Fleiss. 1981. Statistical methods for rates and proportions. John Wiley, New York, 2nd edition. find
  16. Alexander Fraser. 2009. Experiments in morphosyntactic processing for translating to and from German. In Proceedings of the EACL MT workshop, pages 115–119, Athens, Greece. find
  17. Jerry Hobbs and Megumi Kameyama. 1990. Translation by abduction. In Proceedings of COLING, pages 155–161, Helsinki, Finland. ↗
  18. Rebecca Hwa, Philipp Resnik, Amy Weinberg, Clara Cabezas, and Okan Kolak. 2005. Bootstrapping parsers via syntactic projection across parallel texts. Journal of Natural Language Engineering, 11(3):311–325. ↗
  19. Hiroshi Kanayama. 2003. Paraphrasing rules for automatic evaluation of translation into Japanese. In Proceedings of the Second International Workshop on Paraphrasing, pages 88–93, Sapporo, Japan. find
  20. Philipp Koehn. 2005. Europarl: A Parallel Corpus for Statistical Machine Translation. In Proceedings of the 10th Machine Translation Summit, pages 79–86, Phuket, Thailand. find
  21. Heinz L. Kretzenbacher, Michael Clyne, and Doris Schüpbach. 2006. Pronominal Address in German: Rules, Anarchy and Embarrassment Potential. Australian Review of Applied Linguistics, 39(2):17.1– 17.18. ↗
  22. Alexander Künzli. 2010. Address pronouns as a problem in French-Swedish translation and translation revision. Babel, 55(4):364–380. ↗
  23. Zhifei Li and David Yarowsky. 2008. Mining and modeling relations between formal and informal Chinese phrases from web corpora. In Proceedings of EMNLP, pages 1031–1040, Honolulu, Hawaii. find
  24. Christopher D. Manning, Prabhakar Raghavan, and Hinrich Schütze. 2008. Introduction to Information Retrieval. Cambridge University Press, Cambridge, UK, 1st edition. ↗
  25. Andrew Kachites McCallum. 2002. Mallet: A machine learning for language toolkit. http://mallet.cs.umass.edu. ↗
  26. Roberto Navigli. 2009. Word Sense Disambiguation: a survey. ACM Computing Surveys, 41(2):1–69. ↗
  27. Eric W. Noreen. 1989. Computer-intensive Methods for Testing Hypotheses: An Introduction. John Wiley and Sons Inc. find
  28. Franz Josef Och and Hermann Ney. 2003. A Systematic Comparison of Various Statistical Alignment Models. Computational Linguistics, 29(1):19–51. ↗
  29. Lance Ramshaw and Mitch Marcus. 1995. Text chunking using transformation-based learning. In Proceeding of the 3rd ACL Workshop on Very Large Corpora, Cambridge, MA.
  30. Michael Schiehlen. 1998. Learning tense translation from bilingual corpora. In Proceedings of ACL/COLING, pages 1183–1187, Montreal, Canada. find
  31. Helmut Schmid. 1994. Probabilistic Part-of-Speech Tagging Using Decision Trees. In Proceedings of the International Conference on New Methods in Language Processing, pages 44–49, Manchester, UK. find
  32. Doris Schüpbach, John Hajek, Jane Warren, Michael Clyne, Heinz Kretzenbacher, and Catrin Norrby. 2006. A cross-linguistic comparison of address pronoun use in four European languages: Intralingual and interlingual dimensions. In Proceedings of the Annual Meeting of the Australian Linguistic Society, Brisbane, Australia. find
  33. Ralf Steinberger, Bruno Pouliquen, Anna Widiger, Camelia Ignat, Tomaž Erjavec, and Dan Tufis. 2006. The JRC-Acquis: A multilingual aligned parallel corpus with 20+ languages. In Proceedings of LREC, pages 2142–2147, Genoa, Italy.
  34. David M. J. Tax and Robert P. W. Duin. 1999. Support vector domain description. Pattern Recognition Letters, 20:1191–1199. ↗
  35. David Yarowsky and Grace Ngai. 2001. Inducing multilingual POS taggers and NP bracketers via robust projection across aligned corpora. In Proceedings of NAACL, pages 200–207, Pittsburgh, PA. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/faruqui-2012-model-formal-informal-address-english/