Elamur Research⌕
☰
home / papers / 2016 / The Inner Circle vs. the Outer Circle or British English vs. American English

The Inner Circle vs. the Outer Circle or British English vs. American English

Paper · 2016 · Yong-hun Lee and Ki-suk Jun

Summary

The authors ask which split better explains how English varieties differ in their use of the near-synonymous modals can and may. One option is Kachru's division into native (Inner Circle) and second-language (Outer Circle) varieties. The other is the American-versus-British lineage. They took four comparable ICE components (Britain, USA, India, Philippines), extracted modal sentences with NLPTools, and randomly drew 1,000 per corpus. Each sample was hand-annotated for twenty morphological, syntactic and semantic factors. The framing comes from the Competition Model: the two modals compete, and contextual cues decide which one wins.

They ran two analyses in R:

  • a logistic-regression GLM with backward stepwise selection, to find which factors interact with variety;
  • Behavior Profile clustering, to see which varieties group together.

Indian and Philippine writers favour can more than native writers do. Three factors behave differently across varieties. The clustering pairs India with Britain and the Philippines with the USA. This suggests that colonial lineage (AmE vs. BrE) matters more than native/non-native status. For corpus studies, it shows that variety effects can be measured multi-factorially rather than only through word lists.

Key points

  • Data: the British, USA, India and Philippines components of the ICE corpus. Each has about 1M tokens and a near-identical design. 1,000 randomly sampled can/may sentences were taken per component.
  • Twenty ID tags were manually coded using the annotation scheme of Deshors and Gries (2014), which builds on the earlier work of Deshors (2010). Examples include subject morphology, sentence type, clause type, modal sense (epistemic/deontic/dynamic), Vendler class, animacy and voice.
  • Model: a binary logistic GLM with interactions. Terms were removed one at a time, and each nested pair of models was compared with ANOVA.
  • Significant interactions with CORPUS were found for SUBJMORPH, SENTTYPE and VENDLER:
    • may is likelier in declaratives;
    • may is likelier with accomplishment and state verbs;
    • can is favoured with achievement and process verbs.
  • The BP analysis used multiscale bootstrap clustering with AU p-values. It yielded {{Britain, India}, {Philippines, USA}}. This contrasts with a simple association plot that had hinted the Inner/Outer split was sharper.

Interesting ideas

  • I1. Model the choice between two near-equivalent words as a competition, decided by a small set of interpretable context cues and estimated with an inspectable regression rather than a black box.
  • I2. Before clustering, write down which grouping each competing hypothesis predicts, so the dendrogram acts as a decisive test instead of a picture to interpret afterwards.
  • I3. Attach bootstrap stability scores (AU/BP values) to every cluster, so that groupings that would not survive resampling are not trusted.
  • I4. Prune a full interaction model one term at a time, keeping a factor only if removing it significantly worsens the fit.
  • I5. Lexical aspect (Vendler class) proved to be a meaningful predictor of modal choice, which supports storing aktionsart as an explicit verb property.

For Mova

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

  • I1 (frames): Modal sense labelling (epistemic vs. deontic vs. dynamic can/may) could rely on explicit cues such as Vendler class, sentence type and subject form, each with a visible weight that explains the decision.
  • I2 (learn): When deciding which treebanks or text genres to pool for training, state the expected groupings in advance and check them against clustering of parser behaviour profiles.
  • I3 (learn): Clusters of parser errors used to propose self-induced repair rules could be kept only if they stay stable under resampling, which acts as a guard against spurious rules.
  • I4 (learn): Candidate repair rules or conditions in rule-based modules could be simplified the same way, dropping any condition whose removal does not change accuracy on held-out data.
  • I5 (world): Adding a Vendler class to compiled verb knowledge would help both modal interpretation and reasoning about event duration and order in the world of the text.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Fields, topics and concepts

📚 Shelves 3The field of science this work belongs to (OpenAlex fields).Arts and Humanities · 17,450 worksLanguage and Linguistics · 17,296 worksLiterature and Literary Theory · 541 works

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

Cite

Cite as Yong-hun Lee and Ki-suk Jun. 2016. The Inner Circle vs. the Outer Circle or British English vs. American English. Proceedings of the 30th Pacific Asia Conference on Language, Information and Computation: Posters.

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

@inproceedings{lee2016inner,
  title = {{The Inner Circle vs. the Outer Circle or British English vs. American English}},
  author = {Yong-hun Lee and Ki-suk Jun},
  booktitle = {Proceedings of the 30th Pacific Asia Conference on Language, Information and Computation: Posters},
  year = {2016},
  url = {https://aclanthology.org/Y16-3004/}
}

References

24 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Beryl Atkins. 1987. Semantic ID Tags: Corpus Evidence for Dictionary Senses. In Proceedings of the Third Annual Conference of the UW Centre for the New Oxford English Dictionary, 17-36. find
  2. Braj Kachru. 1992. The Other Tongue: English across Cultures. University of Illinois Press, Urbana, IL. find
  3. David Crystal. 2003. The Cambridge Encyclopedia of the English Language. Cambridge University Press, Cambridge. find
  4. Deshors, Sandra. 2014 Constructing Meaning in L2 Discourse: The Case of Modal Verbs and Sequential Dependencies. In Glynn, Dylan and M. Sjo..lin (eds.) Subjectivity and Epistenicity: Stance Strategies in Discourse and Narration, 329-348. Lund University Press, Lund. find
  5. Elizabeth Bates and Brian MacWhinney. 1982. Functionalist Approaches to Grammar. In Eric Wanner and Lila Gleitman (eds.) Language Acquisition: The State of the Art, 73–218. Cambridge University Press, Cambridge. Elizabeth Bates and Brian MacWhinney. Functionalism and the Competition Model. In Brian MacWhinney and Elizabeth Bates (eds.) The Cross-linguistic Study of Sentence Processing, 3–73. Cambridge University Press, Cambridge. find
  6. Pam Peters. 2004. The Cambridge Guide to English Usage. Cambridge University Press, Cambridge. ↗
  7. Peter Collins. 2009. Modals and Quasi-modals in English. Rodopi, Amsterdam. ↗
  8. Peter Trudgill and Jean Hannah. 2013. International English: A Guide to Varieties of Standard English. 5th Edition. Routledge, London. find
  9. R Core Team. 2016. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna. find
  10. Ronald Langacker. 2000. Grammar and Conceputalization. Mouton, Berlin. find
  11. Sandra Deshors and Stefan Greis. 2014. A Case for the Multifactorial Assessment of Learner Language: The Uses of May and Can in French-English Interlanguage. In Dylan Glynn and Justyna Robinson (eds.), Corpus Methods for Semantics: Quantitative Studies in Polysemy and Synonymy, 179-204. John Bejamins, Amsterdam. ↗
  12. Sandra Deshors. 2010. Multifactorial Study of the Uses of May and Can in French-English Interlanguage. Ph.D. dissertation, University of Sussex. find
  13. Sidney Greenbaum. 1996. Comparing English Worldwide: The International Corpus of English. Claredon, Oxford. find
  14. Stefan Gries and Naoki Otani. 2010. Behavioral Profiles: A Corpus-based Perspective on Synonymy and Antonymy. ICAME Journal 34:121-150. find
  15. Stefan Gries. 2003. Multifactorial Analysis in Corpus Linguistics: A Study of Particle Placement. Continuum Press, London. find
  16. Stefan Gries. 2010a. Behavioral Profiles: A Fine-grained and Quantitative Approach in Corpus-based Lexical Semantics. The Mental Lexicon 5(3):323-346. find
  17. Geoffrey Leech. 1969. Towards a Semantic Description of English. Indiana University Press, Bloomington. find
  18. Stefan Gries. 2010b. Behavioral Profiles 1.01: A Program for R 2.7.1 and Higher. find
  19. Gunnel Tottie. 2002. An Introduction to American English. Wiley- Blackwell, Oxford. find
  20. Tom McArthur. 2002. The Oxford Guide to World English. Oxford University Press, Oxford. find
  21. Jennifer Coates. 1983. The Semantics of the Modal Auxiliaries. Croom Helm, London. find
  22. Yong-hun Lee. 2007. Corpus Analysis and Their Applications Using NLPTools: Applications to Study of Language, English Education, and Development of English Textbooks. Cambridge University Press, Seoul. find
  23. John Algeo. 2006. British or American English? Cambridge University Press, Cambridge. ↗
  24. Orin Hargraves. 2003. Mighty Fine Words and Smashing Expressions. Oxford University Press, Oxford. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/lee-2016-inner-circle-vs-outer-circle/