Abstract
This study is concerned with the coherence of a newly-emerged standard variety of Italian, i.e., the so-called neo-standard Italian, as compared with the traditional standard variety. Our analysis aims to test whether, and to what extent, linguistic features which are conventionally associated with these two standard varieties co-occur in actual speech data. A principal component analysis on a corpus of spoken Italian reveals coherent patterns of aggregation of linguistic variables across the two language varieties. The results of our analysis are relevant to the more general problem of identifying language varieties in speakers’ usage. On a theoretical level, we suggest that the identification of aggregates of linguistic variables is better captured by a continuum with focal points rather than by a pure continuum of values or by a set of discrete categories. On an empirical level, the analysis indicates that the traditional standard is a more coherent variety which is strongly associated with precise situational contexts (e.g., in our data, oral examinations at the university). In contrast, the new standard appears to be a more dispersed range of linguistic elements which is called on in different verbal interactions.