Simplification in translated Czech: a new approach to type-token ratio

Václav Cvrček

Abstract: The main objective of the paper is to examine whether simplification can be demonstrated to exist in Czech translated texts. In general, simplification as one of the so-called translation universals, is defined as a translators’ tendency to create simpler texts. According to research of English texts, simplification may be manifested e.g. by a lower level of lexical richness. To describe lexical richness, a simple type-token ratio (TTR) is widely used; however, it is very sensitive to text size. To overcome this disadvantage, a standardized type-token ratio (sTTR) has been introduced, which is calculated for every 1000 words in the text. Nevertheless, it also has certain drawbacks. Our method for standardizing type-token ratio (zTTR) is based on comparing the observed TTR with the referential TTR values representing texts of identical size. Inspired by the z-score, this metric is capable of comparing the lexical richness of texts regardless of their length. The analysis carried out on a large comparable corpus of translated and non-translated Czech proved that the non-translated texts tend be lexically richer, although the difference is not as striking as some studies have predicted.