Investigating Greek Learners’ Interlanguage in Italian through a Custom-Tagged Learner Corpus

Abstract

This paper describes the creation and development of a learner corpus of Italian written by Greek university students, conducted within the framework of an ongoing PhD project. The corpus aims to fill a notable gap in learner corpus research concerning less-represented language pairings, specifically Italian as a second language learned by Greek speakers. The data were collected from authentic written examinations of the State Certificate of Language Proficiency (ΚΠΓ) at the B and C CEFR levels, comprising 293 texts and approximately 53,000 words. To analyse learners’ interlanguage, a custom error-annotation tagset was designed and refined through several trial phases. The tagset includes three hierarchical dimensions—error level, part of speech, and error type—allowing for a detailed categorisation of linguistic deviations. Preliminary observations reveal recurrent difficulties with prepositions, verb morphology, and syntactic agreement, alongside noticeable traces of cross-linguistic influence from Greek, English, and other languages. The project provides both a methodological framework and a valuable linguistic resource for further studies on Italian L2 acquisition in Greek-speaking and multilingual contexts.