Abstract
The aim of this paper is to determine what shifts occur in the translations from French into Czech and from Czech into French in segmentation into sentences (the final punctuation), and whether it is possible to consider these changes as a general tendency associated with the translation language, i.e. as a translation universal. The analyzed material is a set of non1: 1 segments extracted from parallel text editor INTERTEXT (86 texts from the core of the parallel InterCorp - 65 translations from French into Czech and 21 translations in the opposite direction).
At the beginning of our analysis, we separate linguistically motivated non1: 1 segments from those segments where there are shifts in segmentation to sentences given by other causes, including technical ones, e. g. differences in the computer tools used for segmentation in Czech and in French. The secondary objective of our research is the analysis of the quality of alignment in this part InterCorp.
The focus of the research is to determine the typology of linguistically motivated shifts in the segmentation into sentences and in particular to explain their causes and consequences for the overall tone of the sentence. In addition to the final punctuation marks (period, exclamation mark, question mark, ellipsis) we also analyze the shifts regarding the semicolon because it is located on the border of the non-terminal and terminal punctuation and because the differences in the no / usage of this punctuation mark in Czech and in French is among translators the subject of a large amount of prejudice.
The corpus research of the semicolon is therefore completed by a short survey among translators from Czech to French.