Abstract
This contribution compares three electronic corpora with regard to their use in studying eighteenth-century Spanish, a period that has been relatively neglected in diachronic research. On the one hand, we present an unpublished corpus of Jesuit documents from the Rioplatense-Paraguayan region, which can be consulted through a relational database, and on the other hand, two comparative sub-corpora extracted from CHARTA and CORDIAM, one Spanish and the other Novo-Hispanic. To illustrate the advantages of these corpora for corpus linguistics, particularly for philologists, we perform two exemplary searches and conduct a comprehensive comparison of a series of parameters across the three corpora, including computational and philological aspects as well as the comfort and manageability of their user interfaces.
This work is licensed under a Creative Commons Attribution 4.0 International License.
Copyright (c) 2024 Marina Albers