Introducing the per-fide project: parallelizing portuguese with six different languages (Español, Russian, Français, Italiano, Deutsch, English)

Abstract

In this paper we present the Per-Fide project1, aimed at the construction of parallel corpora mapping the Portuguese language to six other languages - English, Russian, French, Italian, German and Spanish - in various domains including literary, journalistic and religious texts. First we will focus on the corpus design criteria and its main features, particularly those that distinguish this corpus from existing parallel corpora. Secondly, we will discuss the challenges of elaborating a typology of text-types for the religious domain and problems associated with the encoding of the texts belonging to this category. To conclude, we will demonstrate how the Per-Fide Corpus can be used in contrastive and translation studies with a case study of pronominal causative constructions in a French-Portuguese contrastive perspective.