Reviving old ideas: parallel and comparable analysis in translation studies - with an example from translation stylistics

One of the most innovative strands of (corpus-based) research in translation studies in the last few years has been the comparative analysis of corpora of texts originally written in one language, and comparable corpora of texts translated into the same language from a variety of source languages (known as Monolingual Comparable Corpora or MCC). The rationale underlying this approach is the belief in the existence of “universal features of translation” characterising translation as a mediated communicative event (Baker 1993). Whilst this typology of corpora and the associated methodology of analysis are undoubtedly of major importance, there are other resources whose relevance to translation studies seems to have been somewhat underestimated so far, namely bilingual parallel corpora (translations and source texts). This paper suggests that parallel corpora might offer the translation scholar a vantage point from which to observe differences in translation norms and practices operating in different cultures. More importantly perhaps, they allow one to control for possible biases and variables in the design of MCC, thus making the evidence in favour or against the existence of translation universals offered by the latter all the more convincing. An overview of translation practices in the English and Italian language communities is provided, and the ways in which these might affect an analysis of the linguistic features of translated texts is discussed. The paper suggests that a careful reconsideration of methodological issues relating to the construction and analysis of MCC, as well as the use of a varied array of corpus resources, including parallel texts, may be necessary if corpus-based translation research is to be pursued on firmer ground.