Princely County of Gorizia and Gradisca, 1873-1914: creation and analysis of a digital corpus of periodicals in their historical context

Abstract

Digital libraries allow not only to improve the preservation of documents and to facilitate access by users, but also to experiment with new methods; for example, it is possible to examine the statistical relationships between the contents of thousands of documents in a short time, an operation almost inaccessible to traditional methods. The key step remains that of converting from analogue support, paper or microfilm, to the digital one, including the transformation of images of the printed text into digital text: only in this way is it possible to statistically analyze those texts, an analysis that cannot be separated from the historical context of their production and from other sources. In this article, we describe in detail the process of creating a digital corpus formed by Italian newspapers published in Gorizia between 1873 and 1914. This includes digitization, editable text extraction, annotation process and statistical analysis of the resulting time series. The data thus obtained are compared with a corpus of Slovenian newspapers printed in the same city and at the same time, already digitized by the Slovene National Library. The analysis of the 47.466 pages of Italian newspapers allows us to demonstrate the type of information that can be extracted from a digital corpus, highlighting the importance of operating within a historical and comparative context. This example of multilingual digital humanism allows us to identify the statistical traces of profound cultural transitions that have taken place in a very complex geographical area and historical period, whose study cannot ignore a particular attention to cultural, technological and social transformations.