Universal Dependencies in a galaxy far, far away... What makes Yoda's English truly alien

What it is about

Levshina takes about 2,000 words of dialogue from five Star Wars films, annotates them in UD, and normalizes them to standard English word order, in order to test three efficiency tendencies from typology. The tendencies are the link between rigid word order and poor morphology, the minimization of dependency lengths, and the late placement of the verb. The character's order puts a non-finite predicate, object or oblique first, then the subject, then the finite verb or auxiliary. The subject-object order entropy is high in every film (0.76 to 0.99, 0.96 overall), where English is near zero. Dependency lengths are longer than in the normalized version, significantly so in Episode III (mean 2.24 against 1.94), and the cause is mostly auxiliaries and copulas separated from their predicates. The paper concludes that this variety is less efficient than English but still works as a stylistic device.

Ideas

• I1. en/src/lin.rs: build a word-order stress test from EWT dev by fronting an object or oblique and moving the finite verb or auxiliary to the end (a construction, not the film lines), and measure the LAS drop of en parse-eval on it.
• I2. Compute subject-object order entropy per treebank in en ud-eval, and compare the EWT dev value with the orders that lin.rs generates; the generator should not produce more order freedom than the treebank has.
• I3. Use the mean dependency length as a score for lin.rs: compare the generated orders with the treebank mean, using the paper's 2.24 against 1.94 as the scale of the difference.
• I4. In prag, check that the link from an auxiliary to its subject and lexical verb survives into the event frames when the auxiliary comes last, which is the case the paper says causes the longer dependencies.

For Mova

The word-order generator en/src/lin.rs and the order statistics are the components that the paper speaks to most directly. The ideas touch 29. Ellipsis, word order and information structure, 21-clause-structure.md and 26-questions.md (auxiliary inversion). The film scripts are copyrighted and are not to be used; only the construction and the measures carry over.

Topics: Universal Dependencies

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.18653/v1/w19-8005

@inproceedings{levshina2019universal,
  title = {{Universal Dependencies in a galaxy far, far away... What makes Yoda's English truly alien}},
  author = {Levshina, Natalia},
  booktitle = {Proceedings of the Third Workshop on Universal Dependencies (UDW, SyntaxFest 2019)},
  year = {2019},
  doi = {10.18653/v1/w19-8005},
  url = {https://doi.org/10.18653/v1/w19-8005}
}

References

19 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Barry J. Blake. 2001. Case . Cambridge University Press, Cambridge, UK. find
2. Matthew Dryer. 2013. Order of Subject, Object and Verb. In: Matthew S. Dryer and Martin Haspelmath (eds.), The World Atlas of Language Structures Online. Max Planck Institute for Evolutionary Anthropology, Leipzig. (Available online at https://wals.info/chapter/81, Accessed on 2019-04-25.) ↗
3. Mariya Fedzechkina, Elissa L. Newport and T. Florian Jaeger. 2016. Balancing effort and information transmission during language acquisition: evidence from word order and case marking. Cognitive Science , 41(2): 416–446. find
4. Ramon Ferrer-i-Cancho. 2017. The Placement of the Head that Maximizes Predictability. An Information Theoretic Approach. Glottometrics, 39: 38–71. find
5. Richard Futrell, Kyle Mahowald and Edward Gibson. 2015. Quantifying word order freedom in dependency corpora. In: Proceedings of the Third International Conference on Dependency Linguistics (Depling 2015) , 91–100, Uppsala, Sweden, August 24–26, 2015. find
6. Edward Gibson. 1998. Linguistic complexity: Locality of syntactic dependencies. Cognition, 68(1):1–76. ↗
7. John A. Hawkins. 1994. A Performance Theory of Order and Constituency . Cambridge University Press, Cambridge, UK. find
8. Roman Jakobson. 1936[1971]. Beitrag zur allgemeinen Kasuslehre. In: Roman Jakobson, Selected Writings. Vol. II. Word and Language , 23–71. Mouton, The Hague/Paris. find
9. Paul Kiparsky. 1997. The rise of positional licensing. In: Ans von Kemenade & Nigel Vincent (eds.), Parameters of morphosyntactic change , 460–494. Cambridge University Press, Cambridge, UK. find
10. Christian Lehmann. 2015. Thoughts on Grammaticalization. 3 rd edn. Language Science Press, Berlin. find
11. Stephen Levinson. 2000. Presumptive Meanings – The Theory of Generalized Conversational Implicature . MIT Press, Cambridge, MA. find
12. Thomas McFadden. 2003. On morphological case and word-order freedom. In: Proceedings of the Twenty-Ninth Annual Meeting of the Berkeley Linguistics Society: General Session and Parasession on Phonetic Sources of Phonological Patterns: Synchronic and Diachronic Explanations , 295-306. find
13. Joakim Nivre, Željko Agić, Lars Ahrenberg et al. 2017. Universal dependencies 2.0 – CoNLL 2017 shared task development and test data. LINDAT/CLARIN digital library at the Institute of Formal and Applied Linguistics, Charles University. http://hdl.handle.net/11234/1-2184. See also http://universaldependencies.org/ (last access 14.12.2017). ↗
14. Edward Sapir. 1921. Language, an introduction to the study of speech . Harcourt, Brace and Co, New York. find
15. Anna Siewierska. 1998. Variation in major constituent order: a global and a European perspective. In: Anna Siwerierska (ed.), Constituent Order in the Languages of Europe , 475–552. De Gruyter Mouton, Berlin. find
16. Kaius Sinnemäki. 2008. Complexity trade-offs in core argument marking. In: Matti Miestamo, Kaius Sinnemäki & Fred Karlsson (eds), Language Complexity: Typology, Contact, Change , 67–88. John Benjamins, Amsterdam. find
17. Milan Straka and Jana Straková. 2017. Tokenizing, POS Tagging, Lemmatizing and Parsing UD 2.0 with UDPipe. In: Proceedings of the CoNLL 2017 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies , Vancouver, Canada, August 2017. find our card
18. Russel S. Tomlin. 1986. Basic Word Order: Functional Principles . Croom Helm, London. find
19. Jan Wijffels. 2018. udpipe: Tokenization, Parts of Speech Tagging, Lemmatization and Dependency Parsing with the UDPipe NLP Toolkit. R package version 0.7. https://CRAN.R-project.org/package=udpipe ↗

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/ud-2019-levshina-universal-dependencies-galaxy-far-far/