Looking for lexical patterns in a trilingual corpus of source and interpreted speeches: extended analysis of EPIC (European Parliament Interpreting Corpus)

This article concerns a major research project carried out on the EPIC corpus (European Parliament Interpreting Corpus), a collection of 9 sub-corpora containing transcriptions of original speeches and the corresponding interpreted versions in three languages (English, Italian and Spanish). Here, the authors examine the lexical patterns present in source speeches delivered in Spanish and in target speeches interpreted in Spanish (from English and Italian), looking in particular at lexical density (expressed in terms of the number of lexical words relative to the total number of words contained in each sub-corpus) and lexical variety (expressed as a percentage represented by the 100 most frequent words in each sub-corpus). This methodology was used by Laviosa to study lexical patterns observed in written English texts (originals and translations). The results obtained from our Spanish sub-corpuses are compared with those of Laviosa, as well as with the results of a study we have already carried out on EPIC corpus material in English and Italian. In this way, EPIC has been examined as both a comparable and a parallel corpus. The complex lexical patterns observed in our research enable us to better understand the role played by the translation modality (written translation vs. simultaneous interpreting), by the language combination and by the direction of interpreting, i.e. from which language and into which language interpreting takes place.