Discourse Particles in Contrast

Investigating translation equivalents of discourse particles can be regarded as a typical corpus problem since it involves looking for all the possible equivalents in order to establish rules for how they are used. Translation data show how transfer of the semantic content results in a rich variety of lexical items. The results may be analysed in terms of formal, semantic, stylistic and pragmatic equivalence. For this study I have used the English-Swedish Parallel Corpus, a 2 million word corpus containing translations from English into Swedish and Swedish into English to initiate fine-grained cross-linguistic comparisons between languages and to test hypotheses (Aijmer et al 1996). For example, on the basis of theories of grammaticalisation it is possible to make cross-linguistic predictions that can be tested on the basis of the corpus.