Abstract
We present two case studies that demonstrate that co-habitation of various (sub)disciplines of Computational Linguistics and Natural Language Pro-cessing can reach novel findings and help the advancement of the field(s). One example covers a study of synonymy by searching for information in different lexical resources with regard to multilinguality, and the other demonstrates on the study of some phenomena concerning information structure and word order in English and Czech how a parallel multilingual and/or multilayered corpus if properly annotated can be used for a study of some aspects of the deep syntactic structure.
Both cases support our convic-tion that the creation of language resources is a fundamental step in the do-main of computational linguistics, that well-founded annotation is an im-portant step forward towards both testing the original linguistic theory and developing it further.