Mapping complexity in language

Description du projet

L’influence de la complexité sur le langage humain

La théorie linguistique soulève un paradoxe: les flexions des noms semblent indiquer que les pluriels sont plus complexes, alors que la manière dont les singuliers rivalisent pragmatiquement avec les pluriels pour créer du sens (par exemple, en sous-entendant «exactement un») suggère le contraire. Cela montre que différents niveaux de complexité influencent la prononciation et le sens. Soutenu par le programme Actions Marie Skłodowska-Curie, le projet COMPLANG vise à élaborer une théorie unifiée en étudiant la complexité morphologique dans différentes langues, en mettant l’accent sur les cas, les pronoms et les nombres. Il permettra d’affiner la théorie des alternatives pragmatiques afin de mieux comprendre la concurrence sémantique et d’intégrer ces résultats dans un modèle expliquant comment les composants grammaticaux accèdent aux différents niveaux de la structure morphosyntaxique. L’objectif de ce projet est d’approfondir notre compréhension de la faculté du langage humain et de faire progresser les technologies de traitement du langage naturel.

Objectif

"This project addresses a fundamental paradox in linguistic theory: different components of grammar appear to rely on conflicting notions of structural complexity. For instance, the way nouns are inflected for number (e.g. singular/plural) suggests that plural forms are more complex than singular ones. However, the way these forms pragmatically compete to create meaning (e.g. implying ""exactly one"") suggests the opposite. This indicates that the processes governing pronunciation (morphological spell-out) and meaning (pragmatic competition) are sensitive to different types of complexity.

The project's main objective is to resolve this paradox by developing a more nuanced, unified theory of linguistic complexity. This will be achieved through three steps. First, we will conduct a large-scale, cross-linguistic investigation to map the notion of complexity relevant to morphological spell-out, expanding beyond number to phenomena like case and pronouns. Second, we will critically examine and refine the theory of pragmatic alternatives to map the notion of complexity relevant to meaning competition. Finally, we will synthesize these findings into a single, coherent formal model that explains how different grammatical components access different levels of morphosyntactic structure.

This research will significantly advance our understanding of the architecture of the human language faculty. By providing a more precise model of the interface between syntax, phonology, and semantics, the project will make key contributions to linguistic theory. In the long term, these fundamental insights can inform the development of more sophisticated and human-like language technologies in fields such as Natural Language Processing. The combined expertise of the applicant and the supervisor is highly suitable for the project: the applicant has a rich background in semantic competition, while the supervisor is a world-renowned expert on the internal structure of words."