Elamur Research⌕
☰
home / papers / 2022 / Automatic Lemmatization of Old English Class III Strong Verbs (L-Y) with ALOEV3

Automatic Lemmatization of Old English Class III Strong Verbs (L-Y) with ALOEV3

Paper · 2022 · Roberto Torre Alonso

Summary

Old English has no lemmatized corpus, which blocks reliable form-to-lemma links and the training of taggers for this low-resource historical language. Torre Alonso builds ALOEV3, a lemmatizer for Class III strong verbs from L to Y. It works by Morphological Generation rather than by analysis. Rules covering inflection, prefixation and morpho-phonological and spelling alternations produce candidate form-types for simplex and single-prefixed verbs. Each candidate is then checked against the DOEC and the YCOE, and only forms actually found in those corpora are kept and tied to a lemma. The work is type-based, meaning it lemmatizes abstract form-types and leaves the in-context lemmatization of individual tokens to a later step. About 97% of the validated forms receive exactly one lemma. The 38 forms with two competing lemmas cluster in a small number of lemma pairs and their derivatives. Even where it cannot decide, the tool shrinks manual lexicographic work by handing reviewers a short, closed list of candidates.

Key points

  • ALOEV3 generates paradigms for OE Class III strong verbs (L–Y) and validates the generated forms against the DOEC (about 3M words) and the YCOE (about 1.5M words).
  • Of 1,256 validated inflectional forms, 38 remain ambiguous between two lemmas. The rest, roughly 97%, get a single lemma.
  • Earlier semi-automatic work relied on wildcard query strings over the corpus indexes in The Grid knowledge base. That approach returned many unwanted hits, such as nouns turning up in a verb's results, and needed heavy filtering. Metola Rodríguez reported 80% validation before manual revision.
  • Two problem cases are discussed:
    • mearn is linked to both medrinnan and murnan, but corpus contexts show it belongs to murnan.
    • gelimpe (58 DOEC occurrences) collides with forms of the noun gelimp.
  • Out of scope: recursive prefixation (e.g. onawinnan), letters before L, and other verb classes. The author pairs ALOEV3 with the lemmatized ParCorOEv2 for confirming lemmas in context.

Interesting ideas

  • I1. Generate every licensed inflected form from rules, then keep only those attested in a corpus; this filters overgeneration with real data instead of hand-written exclusions.
  • I2. Split lemmatization into a context-free type pass and a later contextual token pass, so ambiguity is detected once per form-type and then resolved per occurrence.
  • I3. Treat a form whose lemma link needs a long chain of unusual rules (syncope plus metathesis plus a rare vowel) as suspect, and prefer the analysis that needs fewer rules.
  • I4. Detect homographs across word classes, such as a verb form that is also a noun form, at the type level as a map of overlap zones.
  • I5. When the system cannot decide, hand the human a small closed list of candidates instead of an open search.

For Mova

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

  • I1 (tokens): The English lemmatizer in crate en could build its irregular and variant form tables by generating from rules and checking them against treebank vocabularies, leaving a list of unattested forms for review.
  • I2 (tokens): The lemma module could precompute, for each word type, its closed set of candidate lemmas and send only multi-candidate types to the tagger or parser for disambiguation in context.
  • I3 (learn): Self-induced repair rules could be scored by derivation length, so analyses built from many rare transformations are marked as low-confidence silver data.
  • I4 (tree): A table of form-types shared across UPOS classes could tell the parser where to examine context closely and where a negative-control evaluation should focus.
  • I5 (learn): LLM teacher tiers could be asked to pick from a short candidate set, which makes their labels cheaper, easier to check, and explainable.

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

🎓Universidad de La RiojaLogroño · Spain24 works · 125 citations

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksLiterature and Literary Theory · 519 works

🧭 Topics 3Research subjects and languages the work is about.English · 2,263 worksMorphology · 405 worksHistorical linguistics & language change · 249 works

Cite

Cite as Roberto Torre Alonso. 2022. Automatic Lemmatization of Old English Class III Strong Verbs (L-Y) with ALOEV3. Journal of English Studies 20.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18172/jes.5324

@article{alonso2022automatic,
  title = {{Automatic Lemmatization of Old English Class III Strong Verbs (L-Y) with ALOEV3}},
  author = {Roberto Torre Alonso},
  journal = {Journal of English Studies 20},
  year = {2022},
  doi = {10.18172/jes.5324},
  url = {https://doi.org/10.18172/jes.5324}
}

References

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

  1. 1971 (1899). The Holy Bible Translated from the Latin Vulgate (Douay Rheims Version). London: Tan books and publishers. find
  2. Biber, D., S. Conrad and R. Reppen 1998. Corpus Linguistics: Investigating Language Structure and Use. Cambridge: Cambridge University Press Bosworth, J. and T.N. find
  3. Toller. 1973 (1898). An Anglo-Saxon Dictionary. Oxford: Oxford University Press. find
  4. Brunner, K. 1965. Altenglische Grammatik, nach der Angelsachsischen Grammatik von Eduard Sievers. Berlin: Max Niemeyer. find
  5. Campbell, A. 1987 (1959). Old English Grammar. Oxford: Oxford University Press. find
  6. Clark Hall, J. R. 1996. A Concise Anglo-Saxon Dictionary. Supplement by Herbert D. Merritt. Toronto: University of Toronto Press. find
  7. Ferrés, D., A. AbuRa’ed and H. Saggion. 2017. “Spanish morphological generation with wide-coverage lexicons and decision trees”. Procesamiento del Lenguaje Natural 58: 109-116. find
  8. Forcada, M.L., M. Ginestí-Rosell, J. Nordfalk, J. O’Regan, S. Ortiz-Rojas, J.A. Pérez-Ortiz, F. Sánchez-Martínez, G. Ramírez-Sánchez and F. Tyers. 2011. “Apertium: A free/open-source platform for translation”. Machine Translation 25 (2): 127-144. find
  9. García Fernández, L. 2020. Lemmatising Old English on a Relational Database. Preterite-Present, Contracted, Anomalous and Strong VII Verbs. Munich: Utzverlag. ↗
  10. García García, L. 2019. “The basic valency orientation of Old English and the causative ja- formation: A synchronic and diachronic approach”. English Language and Linguistics 24 (1): 153-177. http://doi.org/10.1017/S1360674318000345 ↗
  11. García García, L. and E. Ruiz Narbona. 2021. “Lability in Old English verbs: Chronological and textual distribution”. Anglia: Journal of English Philology 139 (2): 283-326. http://doi.org/10.1515/ang-2021-0022 ↗
  12. Healey, A., ed., with J. Price and X. Xiang. 2009. The Dictionary of Old English Web Corpus. Dictionary of Old English Project, Centre for Medieval Studies, University of Toronto. find
  13. Healey, A., ed. 2018. The Dictionary of Old English: A to I. Dictionary of Old English Project, Centre for Medieval Studies, University of Toronto. find
  14. Hogg, R. and R.D. Fulk. 2011. A Grammar of Old English. Oxford: Wiley-Blackwell. ↗
  15. Hostetter, A.K. n.d. The Old English Narrative Poetry Project: Beowulf. Accessed September 2. https://oldenglishpoetry.camden.rutgers.edu/beowulf/ Kastovsky, D. 1992. “Semantics and vocabulary”. The Cambridge History of the English Language I: The Beginnings to 1066. Ed. R. Hogg. Cambridge: Cambridge University Press. 290-408. ↗
  16. Khemakhem, A., B. Gargouri, A. Ben Hamadou and G. Francopoulo. 2015. “ISO standard modeling of a large Arabic dictionary”. Journal of Natural Language Engineering 22 (6): 849-879. ↗
  17. Krygier, M. 1994. The Disintegration of the English Strong Verb System. Frankfurt am Main: Peter Lang. find
  18. Levin, S.R. 1964. “A reclassification of the Old English strong verbs”. Language 40 (2): 156-161. ↗
  19. Manjavacas, E., A. Kádár and M. Kestemont. 2019. “Improving lemmatization of non-standard languages with joint”. Proceedings of NAACL-HLT 2019. Ed. J. Burstein, Doran, Christy and T. Solorio. Minneapolis: ACL. 1493-1503. find
  20. Martín Arista, J. 2012a. “Lexical database, derivational map and 3D representation”. RESLA-Revista Española de Lingüística Aplicada (Extra 1): 119-144. find
  21. Martín Arista, J. 2012b. “The Old English prefix ge-: A panchronic reappraisal”. Australian Journal of Linguistics 32 (4): 411-433. http://doi.org/10.1080/07268602.2012.744264. ↗
  22. Martín Arista, J. 2013a. “Recursivity, derivational depth and the search for Old English lexical primes”. Studia Neophilologica 85(1): 1-21. http://doi.org/10.1080/00393274.2013.771829. ↗
  23. Martín Arista, J. 2013b. Nerthus. Lexical database of Old English: From word-formation to meaning construction. Research Seminar, School of English, University of Sheffield. find
  24. Martín Arista, J., ed. 2016. NerthusV3. Online Lexical Database of Old English. Nerthus Project, Universidad de La Rioja. www.nerthusproject.com. find
  25. Martín Arista, J. 2017a. “El paradigma derivativo del inglés antiguo”. Onomázein 37: 144-169. find
  26. Martín Arista, J. 2017b. The design and implementation of a pilot parallel corpus of Old English. Paper presented at the SHELL Session of the 2017 International Medieval Conference, University of Leeds, July 4. find
  27. Martín Arista, J. 2018. “The semantic poles of Old English. Toward the 3D representation of complex polysemy”. Digital Scholarship in the Humanities 33 (1): 96-111. ↗
  28. Martín Arista, J. 2019. “Another look at Old English zero derivation and alternations”. Atlantis 41 (1): 163-182. ↗
  29. Martín Arista, J. 2020a. “Old English rejoice verbs. Derivation, grammatical behaviour and class membership”. Poetica 93: 133-153. find
  30. Martin Arista, J. 2020b. “Further remarks on the deflexion and grammaticalization of the Old English past participle with habban”. International Journal of English Studies 20 (1): 51-71. ↗
  31. Martín Arista, J. 2021a. “The syntax and semantics of the Old English predicative construction”., Language Change and Linguistic Theory in the 21 st Century. Eds. N. Lavidas and K. Nikiforidou. Leiden: Brill. Forthcoming. ↗
  32. Martín Arista, J. 2021b. “Word alignment in a parallel corpus of Old English prose. From asymmetry to inter-syntactic annotation”. Corpora in Translation Research: Recent Advances and Applications. Eds. J. Lavid-López, C. Maíz-Arévalo and J.R. Zamorano. Amsterdam: John Benjamins. 76-100. ↗
  33. Martín Arista, J. and A.E. Ojanguren López. 2018. Doing electronic lexicography of Old English with a knowledge-base. Workshop delivered at the Consolidated Library of Anglo-Saxon Poetry (CLASP) Project, Faculty of English Language and Literature of the University of Oxford. find
  34. Martín Arista, J., S. Domínguez Barragán, L. García Fernández, E, Ruíz Narbona, R. Torre Alonso, and R. Vea Escarza (Comp.) 2021. ParCorOEv2. An Open Access Annotated Parallel Corpus Old English-English. Nerthus Project, Universidad de La Rioja, www.nerthusproject.com. find
  35. Mateo Mendaza, R. 2014. “The Old English adjectival affixes ful-and -ful: a text-based account on productivity”. NOWELE-North-Western European Language Evolution 67 (1): 77-94. http://doi.org/10.1075/nowele.67.1 ↗
  36. Mateo Mendaza, R. 2015. “Matching productivity indexes and diachronic evolution. The Old English affixes ful-, -isc, -cund and -ful”. Canadian Journal of Linguistics 60 (1): 1-24. find
  37. Mateo Mendaza, R. 2016a. “The search for Old English semantic primes: The case of HAPPEN”. Nordic Journal of English Studies 15: 71-99. ↗
  38. Mateo Mendaza, R. 2016b. “The Old English exponent for the semantic prime MOVE”. Australian Journal of Linguistics 34 (4): 542-559. http://doi.org/10.1080/07268602.2016.1169976 ↗
  39. von Mengden, F. 2011. “Ablaut or transfixation? On the Old English strong verbs”. More than Words: English Lexicography and Lexicology Past and Present. Eds. R. Bauer and U. Krischke. Frankfurt am Main: Peter Lang. 123-139. find
  40. Metola Rodríguez, D. 2015. Lemmatisation of Old English strong verbs on a lexical database. Unpublished Ph. D. thesis. University of La Rioja, Spain. find
  41. Metola Rodríguez, D. 2017. “Strong verb lemmas from a corpus of Old English. Advances and issues”. Revista de Lingüística y Lenguas Aplicadas 12: 65-76. ↗
  42. Novo Urraca, C. 2015. “Old English deadjectival paradigms. Productivity and recursivity”. NOWELE-North-Western European Language Evolution 68 (1): 61-80. find
  43. Novo Urraca, C. 2016a. “Old English suffixation. Content and transposition”. English Studies 97: 638-655. ↗
  44. Novo Urraca, C. 2016b. “Morphological relatedness and the typology of adjectival formation in Old English”. Studia Neophilologica 88 (1): 43-55. find
  45. Novo Urraca, C. and A. E. Ojanguren López. 2018. “Lemmatising treebanks. Corpus annotation with knowledge bases”. RAEL: Revista electrónica de Lingüística Aplicada 17 (1): 99-120. find
  46. Oflazer, K. and M. Saraçlar, eds. 2018. Turkish Natural Language Processing. Cham: Springer. http://doi.org/10.1007/978-3-319-90165-7 ↗
  47. Ojanguren López, A. E. 2020. “The semantics and syntax of Old English end verbs”. Atlantis 42 (1): 163-188. ↗
  48. Pintzuk, S. and L. Plug. 2001. The York-Helsinki Parsed Corpus of Old English Poetry. Department of Language and Linguistic Science, University of York. find
  49. Reiter, E. and R. Dale. 1997. “Building applied natural language generation systems”. Natural Language Engineering 3 (1): 57-87. http://doi.org/10.1017/S1351324997001502 ↗
  50. Rissanen, M., M. Kytö, L. Kahlas-Tarkka, M. Kilpiö, S. Nevanlinna, I. Taavitsainen, T. Nevalainen, and H. Raumolin-Brunberg, comp. 1991. The Helsinki Corpus of English Texts. Department of English, University of Helsinki. find
  51. Riyeff, J. 2017. The Old English rule of Saint Benedict with related Old English texts. Collegeville: Liturgical press. find
  52. Taylor, A., A. Warner, S. Pintzuk and F. Beths. 2003. The York-Toronto-Helsinki Parsed Corpus of Old English Prose. York: University of York. find
  53. Tapsai, C., H. Unger and P. Meesad. 2021. Thai Natural Language Processing. Cham: Springer. ↗
  54. Tío Sáenz, M. 2019. The lemmatisation of Old English Weak verbs of a relational database. Unpublished Ph. D. thesis. University of La Rioja: Spain. find
  55. Vea Escarza, R. 2013. “Old English adjectival affixation. Structure and function”. Studia Anglica Posnaniensia 48 (2-3): 5-25. ↗
  56. Vea Escarza, R. 2016a. “Recursivity and inheritance in the formation of Old English nouns and adjectives”. Studia Neophilologica 88: 1-23. http://doi.org/10.1080/00393274.2015.1049830 ↗
  57. Vea Escarza, R. 2016b. “Old English affixation. A structural-functional analysis”. Nordic Journal of English Studies 15 (1): 101-119. find
  58. Vea Escarza, R. 2018. “Las funciones y categorías de los nombres y adjetivos afijados del inglés antiguo”. Onomázein 41: 208-226. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/alonso-2022-automatic-lemmatization-old-english-class/