The CoreGram project: theoretical linguistics, theory development and verification

Abstract

This paper describes the CoreGram project, a multilingual grammar engineering project that develops HPSG grammars for several typologically diverse languages that share a common core. The paper provides a general motivation for doing theoretical linguistics the way it is done in the CoreGram project and therefore is not targeted at computational linguists exclusively. I argue for a constraint-based approach to language rather than a generative-enumerative one and discuss issues of formalization. Recent advantages in the language acquisition research are mentioned and conclusions on how theories should be constructed are drawn. The paper discusses some of the highlights in the implemented grammars, gives a brief overview of central theoretical concepts and their implementation in TRALE and compares the CoreGram project with other multilingual grammar engineering projects.