Abstract

The field of study that can be named “temporal semantics” is marked by the domination of grammatical studies. The researchers’ interest is focused on the tense and the aspect, lexical expressions with temporal meaning stay in the background. Obviously, it is grammatical means that for the most part express temporal meanings: they are present in most languages and in every sentence in these languages. However, lexical means can convey the same information. We argue that the cataloging of lexical means and comparing the systems of these means in different languages allow us to outline the boundaries of human perception of time and to better understand what is universal and what is specific in it. Our goal is to conduct a corpus study of temporal lexical expressions in the Russian language. To do this, we developed an annotation scheme and performed a pilot markup on some fragments of the SynTagRus. The paper describes our annotation scheme and provides linguistic grounds for the decisions made during its development. We also compare our annotation scheme with the TimeML scheme, the most used annotation scheme for temporal information in the world. We show that despite the elaborated structure and a large set of elements, TimeML fails to grasp some important fragments of the expressions’ semantics. The proposed annotation scheme is more expressive and flexible and suits for the annotation of the complex examples: sekretarša na minutu vyšla (‘the secretary left for a minute’), delatʹ čto-libo tretij čas podrjad (‘do something more than two hours in succession’ – literally ‘for the third hour in a row’).

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call