TheseUsage

TheseUsage is a theoretical framework and dataset collection in linguistics that studies how English demonstratives, notably these and those, are used in natural discourse. It treats demonstratives as markers of discourse salience and referential distance rather than mere spatial indicators.

The framework integrates corpus data, psycholinguistic experiments, and computational modeling to map usage across genres, registers, and dialects. It aims to distinguish when a speaker chooses these versus those based on proximity, focus, and discourse structure.

Methodology involves annotating large corpora for proximity, salience, referential distance, and discourse function, then applying statistical and machine-learning techniques to identify usage patterns and predictive cues.

Key findings include that these often introduces items perceived as near in the discourse or newly accessible to participants, while those signals distance or prior mention; frequency varies with formality and narrative position.

Applications span language education, where TheseUsage informs teaching of demonstratives, and natural language processing, where it improves coreference resolution and generation of coherent references.

Limitations include cross-dialect variation and the challenge of generalizing from English to other languages; ongoing work seeks broader corpora and cross-linguistic comparison to refine the framework.