Corpas Pastor, Gloria (2008): Investigar con corpus en traducción: los retos de un nuevo paradigma. Frankfurt: Peter Lang, 297 p.

Research on translation universals has been the cornerstone of Descriptive Translation Studies throughout the last decade (Baker 1993; Toury 1995; Laviosa 1996; 1997; Moropa 2000). However, these kinds of studies are characterised by the use of introspection and a lack of empirical data and therefore their results can be considered as very preliminary and with a lack of credibility and thoroughness. By contrast, the book reviewed here explores translation universals in a very systematic, rigorous and consistent way within a coherent theoretical framework. This study can be considered an outstanding example of a scientific corpus-based translation study. The author, Gloria Corpas Pastor, shows how to carry out cutting-edge high-quality research that shatters the preconceptions in preceding studies. She presents an up-to-date study whose specific aim is to check the validity of the three following translation universals: simplification, convergence and transfer. To do so, she employs an empirical methodology based on corpora as the main tool of the study, combined with computational linguistics and natural language processing techniques. Throughout the five chapters of this book, the author offers a linear and exhaustive scientific approach that ultimately yields exceptional results with respect to translation universals. Chapters I and II (Breve recorrido histórico and Metodología de corpus para el establecimiento de la equivalencia) deals with European policies regarding the language industries, corpus-based research in computational linguistics and applied linguistics, studies on equivalence in translation studies, among others. In short, these two chapters constitute the background of the research and serve as an introduction and an update to the use of corpus. In the next chapter (Aportaciones a los estudios de Traducción e Interpretación), the author focuses on the contribution of corpus research in translation studies. As an extension of the previous chapters, where only the theoretical and technological approaches to corpora were explored, she includes the learning-teaching approach to translation and interpreting. In this chapter, the author also examines the most important descriptive studies about norms, laws and universals of translated language (in particular, simplification, convergence and transfer) that have been carried out so far, and mentions the most relevant authors who have worked on this topic. This chapter serves as an introduction to the following chapters. The next two chapters (Investigación de la hipótesis de universalidad: organización de los datos and Investigación de la hipótesis de universalidad: experimentos y resultados) present the empirical studies carried out with regard to the three translation universals (simplification, convergence and transfer). In Chapter IV, the author describes the starting hypothesis and the main objectives, together with the methodology and techniques used in the study. She also provides information about the compilation of the different corpora used, as well as data relevant to the organisation and design of the study. Chapter V is dedicated to the implementation of the study presented in the previous chapter. The author provides an exhaustive description of the experiments carried out to test each of the three universals and the corresponding results. She also analyses the results, questions the results of previous similar studies and demonstrates the existence or non-existence of these three universals. The last chapter is dedicated to the conclusions. Here the author presents a concise summary of all the main ideas appearing in the work and presents possible future research that can be derived from this study. As the main part of this section, she interprets the obtained results and concludes that her research study does not support the existence of the studied translation universals (simplification, convergence and transfer) and therefore, they are far from their universality: “Nuestro estudio […] no apoya …