Elamur Research⌕
☰
home / papers / 2025 / Universal Dependencies for Amahuaca

Universal Dependencies for Amahuaca

Summary

The authors build the first UD treebank for Amahuaca. Amahuaca is a Panoan language of the Headwaters subgroup, spoken by about 400 people in the Ucayali region of Peru, most of them older. The main difficulty is clitics. They encode tense, aspect, mood and case, attach to different parts of speech, and some do not need a lexical host word at all. The usual word-level UD annotation loses the relations between such morphemes, so the authors argue for morpheme-level annotation. The corpus has 202 sentences. Native speakers translated them from Spanish, and the sentences were then annotated by hand in two parallel versions. In the morpheme-level version, clitics are separate nodes in the tree. The word-level version was produced by a script that attaches clitics to the preceding word. Labels such as aux:<gloss> are introduced for non-nominal clitics, and nsubj:copy for duplicate pronouns. Automatic parsing quality (UDPipe) has not yet been compared, so the paper reports no metrics such as LAS. The work is valuable as an example of how to reconcile polysynthetic languages with free clitics with lexically oriented UD.

Key points

  • Data: the project compares treebanks of several Amazonian languages of Peru, and 420 sentences were prepared for it. Three native speakers translated them into Amahuaca, and 202 sentences have already been annotated and checked.
  • Size: the word-level version has 1028 words and 1031 dependencies, the morpheme-level version 1928 tokens and 1927 dependencies. The difference comes mostly from the PART tag: 889 clitics in the morpheme-level version.
  • Scheme: 56 dependency types are used. The most frequent labels for clitics are aux:top (=mun, 183 cases), aux:decl (=nu, 178) and case (=n, 93). nsubj:copy is introduced for duplicate pronouns.
  • Reference points: the scheme is based on the UD treebanks of the related languages Shipibo-Konibo and Kakataibo, and the morpheme-level approach draws on Park et al. (2021).
  • Unfinished: evaluation of both schemes in UDPipe is planned but not yet done. The remaining translated sentences are still awaiting review.

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: Linguistics & languages

Topics: Universal Dependencies · Treebanks

Cite

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

@inproceedings{angulo2025universal,
  title = {{Universal Dependencies for Amahuaca}},
  author = {Angulo, Candy and Valenzuela, Pilar and Zariquiey, Roberto},
  booktitle = {Proceedings of the Eight Workshop on the Use of Computational Methods in the Study of Endangered Languages},
  year = {2025},
  url = {https://aclanthology.org/2025.computel-main.17/}
}

References

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

  1. Alonso Vasquez, Renzo Ego Aguirre, Candy Angulo, John Miller, Claudia Villanueva, Željko Agić, Roberto Zariquiey, and Arturo Oncevay. 2018. Toward Universal Dependencies for Shipibo-Konibo. In Proceedings of the Second Workshop on Universal Dependencies (UDW 2018), pages 151– 161, Brussels, Belgium.… find
  2. Cristopher Hewlett. 2014. History, kinship and comunidad: learning to live together amongst Amahuaca people on the Inuya River in the Peruvian Amazon (Doctoral dissertation, University of St Andrews). find
  3. Çağrı Çöltekin. 2016. (When) do we need inflectional groups? In Proceedings of The 1st International Conference on Turkic Computational Linguistics, page (to appear). find
  4. Pilar Valenzuela, Roberto Zariquiey and Candy Angulo. In preparation. A grammar sketch of Amahuaca (Pano, Peru) find
  5. Roberto Zariquiey, Claudia Alvarado, Ximena Echevarría, Luisa Gomez, Rosa Gonzales, Mariana Illescas, Sabina Oporto, Frederic Blum, Arturo Oncevay, and Javier Vera. 2022. Building an Endangered Language Resource in the Classroom: Universal Dependencies for Kakataibo. In Proceedings of the Thirteenth… find
  6. Emily Clem. 2019. Amahuaca ergative as agreement with multiple heads. Natural Language & Linguistic Theory, 37, 785-823. find
  7. Francis Tyers and Karina Mishchenkova. 2020. Dependency annotation of noun incorporation in polysynthetic languages. In Proceedings of the Fourth Workshop on Universal Dependencies (UDW 2020), pages 195–204, Barcelona, Spain (Online). Association for Computational Linguistics. find
  8. Gertrude Dole. 1998. Los amahuaca. Guía etnográfica de la Alta Amazonía, 3, 125-273. find
  9. Hyunji Hayley Park, Lane Schwartz, and Francis Tyers. 2021. Expanding Universal Dependencies for Polysynthetic Languages: A Case of St. Lawrence Island Yupik. In Proceedings of the First Workshop on Natural Language Processing for Indigenous Languages of the Americas, pages 131– 142, Online. Associa… find
  10. Joakim Nivre, Daniel Zeman, Filip Ginter, and Francis Tyers. 2017. Universal Dependencies. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Tutorial Abstracts, Valencia, Spain. Association for Computational Linguistics. find
  11. Joakim Nivre, Marie-Catherine de Marneffe, Filip Ginter, Jan Hajič, Christopher D. Manning, Sampo Pyysalo, Sebastian Schuster, Francis Tyers, and Daniel Zeman. 2020. Universal Dependencies v2: An Evergrowing Multilingual Treebank Collection. In Proceedings of the Twelfth Language Resources and Evalu… find
  12. Margarethe Sparing-Chávez. 2012. Aspects of Amahuaca grammar: An endangered language of the Amazon basin. Dallas: SIL International. find
  13. Pilar Valenzuela. 2003. Transitivity in shipibo- konibo grammar. University of Oregon. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/angulo-2025-universal-dependencies-amahuaca/