Elamur Research⌕
☰
home / papers / 2020 / A Small Universal Dependencies Treebank for Hittite

A Small Universal Dependencies Treebank for Hittite

Summary

Hittite is an extinct Anatolian language and the oldest attested Indo-European language. Until now, it had no treebank in the Universal Dependencies format. The authors created the first such resource. The material consists of original Hittite sentences from the textbook by Hoffner and Melchert (2008). These are legal, religious and mythological texts from the Old, Middle and New Hittite periods. They were combined into a single corpus, and their dating, where known, was recorded in the metadata. The annotation was done manually, based on the proposals of Inglese (2015), which the authors updated to the current version of UD. This required decisions on phenomena found in later texts, such as clitic doubling and the particle -za. Since Hittite has no punctuation, sentence boundaries were determined by clausal conjunctions and non-emphatic clitics. Determinatives were split off as separate words. The paper contains no computational experiments or parser evaluation: its contribution is the resource itself and the annotation scheme. The authors present the corpus as a starting point for larger UD corpora of Hittite, computational analysis of the language, and typological research.

Key points

  • The corpus contains 136 sentences, 1309 words and 970 whitespace-separated tokens. All UPOS tags are used except SYM and PUNCT. The most frequent tag is NOUN (29.56 %).
  • The transcription is hybrid: hyphens mark the boundaries of cuneiform signs, as in narrow transcription, and "=" marks clitic boundaries, as in broad transcription. Determinatives are treated as part of a multiword token and attached to the noun via clf.
  • Language-specific relations are introduced: acl:relcl, advmod:emph (the particle -pat), advmod:loc (local particles), discourse:conn (the conjunctions nu/ta/šu in sentence-medial position) and expl:pass (reflexive -za).
  • Morphology: neuter nouns in the nominative-accusative are marked as Abs, and the Aspect feature is not used. Akkadograms and Sumerograms are marked with the feature Language=Akk/Sum. Lemmas are taken from stems, and homonymous stems are distinguished by the labels -#1-/-#2- (as in the Hittite PROIEL).
  • For clitic doubling and scribal "corrections", expl is used, as in the Bulgarian UD treebank. reparandum was rejected because the redundancy of the pronoun cannot be proven.

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{andersen2020small,
  title = {{A Small Universal Dependencies Treebank for Hittite}},
  author = {Andersen, Erik and Rozonoyer, Benjamin},
  booktitle = {Proceedings of the Fourth Workshop on Universal Dependencies (UDW 2020)},
  year = {2020},
  doi = {10.48448/z7x2-rj12},
  url = {https://doi.org/10.48448/z7x2-rj12}
}

References

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

  1. Billie Jean Collins. 2012. The Hittites and Their World. SBL Press. find
  2. Andrew Dalby. 2004. Dictionary of Languages. Columbia University Press, Revised edition. find
  3. Harry A. Hoffner, Jr. and H. Craig Melchert. 2008a. A Grammar of the Hittite Language. Part 1: Reference Grammar. Eisenbrauns. find
  4. Harry A. Hoffner, Jr. and H. Craig Melchert. 2008b. A Grammar of the Hittite Language. Part 2: Tutorial. Eisenbrauns. find
  5. Guglielmo Inglese, Maria Molina, and Hanne Eckhoff. 2018. Incorporating Hittite into PROIEL: a pilot project. In Andrew U. Frank, Christine Ivanovic, Francesco Mambrini, Marco Passarotti, and Caroline Sporlede, editors, Proceedings of the Second Workshop on Corpus-based Research in the Humanities, p… find
  6. Guglielmo Inglese. 2015. Towards a Hittite Treebank. Basic Challenges and Methodological Remarks. In M. Passarotti, F. Mambrini, and C. Sporleder, editors, Proceedings of the Workshop on Corpus-Based Research in the Humanities (CRH). find
  7. Kamil Kopacewicz. 2018. UD Akkadian PISANDUB. find
  8. H. Craig Melchert. 2007. Middle Hittite revisited. In A. Archi and R. Francia, editors, VI Congresso Internazionale di Ittitologia, pages 525–531. find
  9. H. Craig Melchert. 2011. The Problem of the Ergative Case in Hittite. In M. Fruyt, M. Mazoyer, and Dennis Pardee, editors, Grammatical case in the languages of the Middle East and Europe. Acts of the international colloquium Variations, concurrence et evolution des cas dans divers domaines linguisti… find
  10. Maria Molina and Alexei Molin. 2016. In a Lacuna: Building a Syntactically Annotated Corpus for a Dead Cuneiform Language (on the basis of Hittite). In Computational Linguistics and Intellectual Technologies: Proceedings of the International Conference “Dialogue 2016”. find
  11. Maria Molina. 2016. Syntactic Annotation for a Hittite Corpus: Problems and Principles. In Proceedings of the Workshop on Computational Linguistics and Language Science. find
  12. Kiril Simov, Petya Osenova, and Martin Popel. 2015. UD Bulgarian BTB. find

This page in Markdown → https://md.research.elamur.ai/free/en/papers/andersen-2020-small-universal-dependencies-treebank-hittite/