Creatures of Habit? What Translators Usually Do with Words

Abstract

This paper focuses on some of the methodological and theoretical challenges presented by the investigation of "sanitisation" in translated texts through the analysis of semantic pros- ody. The main hypothesis is that target texts tend to use toned down vocabulary compared with their sources, and that this results in the creation of a "sanitised version of the original."Cet article se concentre sur certains des défis méthodologiques et théoriques de l'étude de l'"assainissement" des textes traduits par l'analyse de la prosodie sémantique. L'hypothèse principale est que le vocabulaire des textes cibles est généralement plus restreint que celui de l'original; il en résulte ce que nous appelons des "versions assainies de l'original"

Interference and normalization in genre-controlled multilingual corpora

“…In addition, their study demonstrates that the similarity measure based on Kilgarriff (2001) is a powerful tool to differentiate between translated and original language, while the classical type-token ratio and lexical density measures are not particularly helpful in identifying translated language. The results reported in their study correlate with those in Volansky et al (2013), and prove that the goal of differentiating between translated and original language and identifying source-language family (either Romance (French, Spanish, Italian) or Germanic (German, Dutch)) can be achieved by different means, namely sets of features in Volansky et al (2013) and a similarity measure in Cartoni, Zufferey and Meyer.…”

“…non-translated, target language, standardization, in the sense specified above, will result in a lower degree of lexical and syntactic variation (Olohan 2004: 100), such as a higher proportion of high-frequency words and a lower proportion of low-frequency words (Laviosa 2002), fewer contractions (Olohan 2003) and fewer collocations (Kenny 1998) than in non-translated language. As will be seen below, some of these features overlap with the ones usually listed under the 'universal' of simplification.…”

More spoken or more translated?

“…use of parallel corpora -would presumably yield further insights into the reasons for a particular phenomenon, by allowing us observe the items in context. As Malmkjaer (1998) points out, "[…] when the translation-part of the corpus is used in conjunction with a corpus containing the Source Texts (together constituting a parallel text corpus), the method can promote sense-disambiguation, and can help to identify translation norms […]" (535-536; see also Kenny 1998).…”

Using corpora in scientific and technical translation training: resources to identify conventionality and promote creativity

“…Electronic corpora have made such phenomena increasingly observable (Stewart,(74)(75) Conventionality and creativity have drawn the attention of corpus linguists (Hanks, 1996, Kenny, 1998, Stewart, 1998, inter alia) because both conventionality and creativity are closely related in language: "Routine is not such a bad thing […] It is what allows the creative use of language to be identified as such" (Kenny,515). Corpus linguistics can facilitate the identification of patterns in texts (pointing to frequent and standard uses of language), as well as "rare" instances which may indicate creative uses of language.…”