A corpus analysis of Italian easy-to-read texts to improve the guidelines for the Italian language

Easy-to-read in Italy is still a partially unexplored matter, although its status is difficult to define (Perego 2021: 282). Research on easy-to-read Italian has shown a growing interest in recent years and in different fields (Bleve et al. 2022; Del Bianco 2019; Mastrogiuseppe et al. 2021; 2022; Perego 2021; Span 2016); however, it is still scarce, especially from a linguistic point of view (with the exception of Sciumbata 2020; 2021a; 2021b; 2022). This article will give an insight into the linguistic features of easy-to-read Italian (known as ‘linguaggio facile da leggere e da capire’) by analyzing a corpus of easy-to-read texts in Italian, comparing them to four collections of texts, including articles from two different newspapers, contents extracted from a periodical magazine of simplified news called Due parole, and a corpus of spoken Italian. Corpus and text analysis have already been proven useful for other simplified varieties of standard languages, either for a better understanding of its traits (cf. for instance Hansen-Schirra et al. 2021; Maaß & Rink 2020; 2021; Vandeghinste et al. 2019a; 2019b; Vanhatalo & Lindholm 2020) or for computational applications (Ebling et al. 2022; Falkenjack and Jönsson 2014; Klaper et al. 2013; Liu et al. 2021). However, contributions concerning corpus analyses of easy-to-read Italian are limited to Sciumbata (2020: 87-186). Both quantitative and qualitative methods will be used. The first ones include the application of the Gulpease readability index, which is specific for the Italian language, as well as observing part of speech distributions, which are factors that can affect readability. A more qualitative approach will be also used focusing on the collection of easy-to-read texts to observe in context some phenomena found in the quantitative analysis. The qualitative approach is useful to confirm and interpret the quantitative results. The comparison, which will use a corpus-driven approach (Biber 2015), aims to extract some of the distinctive linguistic features of easy-to-read Italian with respect to other common linguistic varieties, representing both written and spoken Italian. Secondly, the purpose is to understand whether easy-to-read is actually simpler than other written language varieties. Lastly, the goal is to highlight aspects of texts in easy-to-read that can be improved by giving more specific linguistic recommendations. Finally, the article will also illustrate how the findings were used to create the new guidelines for easy-to-read specifically designed for Italian, which were issued only recently (Sciumbata 2022).