Measuring Proximity Between Source and Target Texts: an Exploratory Study

Abstract

This paper reports the results of an ongoing exploratory study aimed at investigating source-target text relations as computed through statistical methods for a manually annotated representative text sample. The purpose was to compare results obtained from clustering texts based on paragraph, sentence, and word counts with those obtained from clustering-based variables built on theory-informed categories of functions realized in the grammar of each language system. The corpus used is made up of ten different translations of an English source text into Spanish and Portuguese. The results point to different patterns with a common group obtained on the basis of the two methods.