Detecting intertextuality between Late Antique and Classical works using Textometry and Deep Learning: the case of Jerome

This contribution aims to assess the relevance and effectiveness of textometry and Deep Learning looking for intertextuality in the corpus of Jerome’s letters. One of the most erudite Fathers of the Church, his corpus of letters is considered the heir to the great classical epistolary writers for the clarity of its Latin and the breadth of the topics covered (Canellis 2002; Hritzu 1939). But to what extent did the great classical authors influence his style? In order to answer this question, we will compare Jerome's letters with a large corpus of comparison consisting of the main prose works of the classical period and use two different but complementary methodologies: 1. Hyperbase: a Textual Data Analysis software which offers statistical tools dedicated to the study of textual corpora (Vanni 2024); 2. Hyperdeep: a Deep Learning model which allows to predict the borrowings of a text from different styles or authors (Vanni et al. 2020).