Abstract
This thesis is exploring the topic of forwardshifted time reference in reported speech using Sequence of Tense as an angle of approach. This mechanism is often used as a way to distinguish and classify languages into two groups: the SOT and the non-SOT languages. However, our study tries to prove that this theoretical classification is not as clear-cut as it may seem simply looking at languages such as English or Russian, which are typically considered to be canonical SOT and non-SOT languages. Indeed, we hypothesize the existence of a third intermediate group, composed of languages sometimes behaving like SOT languages, sometimes like non-SOT languages; in the context of our study these are German, French and Spanish. Due to the topic of our study (tense morphology and tense interpretation) and due to the nature of our hypothesis (aiming to prove a difference between theory and practice) we decided to conduct a qualitative and quantitative empirical corpus based analysis using the parallel corpus ParaSOL as well as monolingual corpora when the data collected was insufficient to compile statistics. In order to give us a broader perspective, we decided to look at material collected from nine different Indo-European languages, using Russian as a primary language and a primary point of view in general. Indeed, within the heterogeneous field of forwardshifting and reported speech, Russian, given its formal characteristics seems to facilitate the querying process and to represent a good control group. The study seemed to confirm the validity of our hypothesis by putting forward a non-SOT trend under the conditions of forwarshifted time reference in reported speech in German, French and Spanish; as well as the benefits of looking at languages cross-linguistically using parallel corpora.