Main Article Content

Maria Inês Bico
Faculdade de Letras da Universidade de Lisboa
Portugal
https://orcid.org/0000-0002-6280-9417
Jorge Baptista
Universidade do Algarve
Portugal
https://orcid.org/0000-0003-4603-4364
Fernando Batista
ISCTE, Instituto Universitário de Lisboa
Portugal
https://orcid.org/0000-0002-1075-0177
Esperança Cardeira
Faculdade de Letras da Universidade de Lisboa
Portugal
https://orcid.org/0000-0003-4700-9830
Vol. 17 (2025): Estudos de Lingüística Galega (2025), Pescuda
https://doi.org/10.15304/elg.17.9812
Submitted: 2024-03-26| Published: 2025-09-30

Abstract

Based on a set of semi-automatically annotated data from the Corpus of Ancient Texts (CTA), this paper aims at analysing the results obtained on the syncopation of intervocalic -d- in the second-person plural morpheme, resulting in a hiatus resolution, and the past participle ending forms -udo/-ido in verbs with an etymological origin in the 2nd and 3rd Latin conjugations. The novelty of this article lies in the use of Natural Language Processing (NLP) methods to optimise the systematic collection and extraction of relevant data for analysis, contributing to a study that encompasses a larger set of texts. The methodology used for annotating the data and, consequently, extracting the relevant data for analysis is presented, stating the importance of resorting to NLP methods and tools for the purpose of linguistic study and for describing previous stages of the Portuguese language.