Careful attention is reserved to the topic of text translation within the field of linguistics. However, it is true that the translation of classical languages, widely considered as “dead languages”, is still unexplored. This study is based on the syntactic and lexical analysis of scholastic translations from Latin by high school students and proposes to outline the patterns of this typology of texts. The research intends to demon strate how the linguistic code derived from these texts differentiates it self from the common written and spoken Italian and thus is based on own and artificial norms that make this textual production a simple translation exercise rather than the creation of a self-standing and au tonomous text. Secondly, these patterns are analyzed and compared with contact language (pidgins, interlanguage) and with a technical lan guage, scholastic Italian. This study is part of a broader research project on coreference phenom ena in the Latin language, entitled CorefLat. This paper focuses specif ically on the publication as Linked Open Data of a set of coreference annotations performed on a selection of Latin texts. The annotations are applied to texts already available as Linked Open Data within the LiLa Knowledge Base, a collection of interoperable linguistic resources for Latin. CorefLat systematically identifies and tags entities and mentions, establishing relational links between them. The annotated corpus covers various historical periods and literary genres, including Augustine’s Confessiones, Plautus’ Curculio, Caesar’s De Bello Gallico, and Sen eca’s Medea, offering a balanced dataset suitable for broad linguistic analysis. In this paper we also provide quantitative data on the annota tions carried out so far, by showing some patterns and distributions of the linguistic phenomena within the dataset. Building on this, the paper describes how coreference phenomena are encoded as Linked Open Data using standard classes and object properties from the POWLA framework.

CorefLat. Annotazione e modellizzazione per la risoluzione di coreferenze in latino

Eleonora Delfino;Marco Carlo Passarotti;Francesco Mambrini
2025

Abstract

Careful attention is reserved to the topic of text translation within the field of linguistics. However, it is true that the translation of classical languages, widely considered as “dead languages”, is still unexplored. This study is based on the syntactic and lexical analysis of scholastic translations from Latin by high school students and proposes to outline the patterns of this typology of texts. The research intends to demon strate how the linguistic code derived from these texts differentiates it self from the common written and spoken Italian and thus is based on own and artificial norms that make this textual production a simple translation exercise rather than the creation of a self-standing and au tonomous text. Secondly, these patterns are analyzed and compared with contact language (pidgins, interlanguage) and with a technical lan guage, scholastic Italian. This study is part of a broader research project on coreference phenom ena in the Latin language, entitled CorefLat. This paper focuses specif ically on the publication as Linked Open Data of a set of coreference annotations performed on a selection of Latin texts. The annotations are applied to texts already available as Linked Open Data within the LiLa Knowledge Base, a collection of interoperable linguistic resources for Latin. CorefLat systematically identifies and tags entities and mentions, establishing relational links between them. The annotated corpus covers various historical periods and literary genres, including Augustine’s Confessiones, Plautus’ Curculio, Caesar’s De Bello Gallico, and Sen eca’s Medea, offering a balanced dataset suitable for broad linguistic analysis. In this paper we also provide quantitative data on the annota tions carried out so far, by showing some patterns and distributions of the linguistic phenomena within the dataset. Building on this, the paper describes how coreference phenomena are encoded as Linked Open Data using standard classes and object properties from the POWLA framework.
2025
14
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10278/5121659
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact