Tracing Lexical Complexity Across L2 Italian Proficiency Levels: The Role of Diversity, Density and Sophistication

Vocabulary knowledge has been shown to be a key component in language competence development and a reliable predictor of overall language proficiency (Qian & Lin 2020; Milton 2013). Lexical complexity, in particular, is one of the most popular constructs used in corpus-based analyses of L2 vocabulary knowledge. This contribution reports on a single-word lexical analysis based on the CELI Corpus (Spina et al. 2022, 2024), a learner corpus of Italian with evenly distributed tokens across proficiency levels B1, B2, C1 and C2. We consider diversity, density and sophistication as the three measures of lexical complexity and adopt a pseudo-longitudinal perspective in order to see how they develop across progressively higher proficiency levels. We found diversity to be the measure that better distinguished the different levels. Density, on the other hand, decreases significantly in B2 texts, while remaining stable in the other ones, and sophistication develops non-linearly across the four levels.