Skip to content
Mary Bary.

Mary Bary · applied linguistics · creative intelligence

Research that makes language—and its consequences—impossible to ignore.

I work across computational linguistics, legal interpretation, and second-language acquisition—turning complex language data into evidence about meaning, understanding, and communication.

PhD in Applied Linguistics — Northern Arizona University

Editorial assistant — Language Learning

7
peer-reviewed articles
4.2M
words in a genre-balanced TV corpus
3
connected research practices

About

Methodological and substantive research in applied linguistics.

I study how language data can illuminate real questions about interpretation, learning, and communication. My work brings together computational methods, legal language, and second-language acquisition.

Explore the research areas

Current focus

  • Corpus-based sense tagging and genre classification
  • Language complexity, comprehension, and legal meaning
  • Second-language development and crosslinguistic influence

Research

Computational Linguistics

I build models that decode complex language data — leveraging LLM training and statistical analysis to help organizations automate information extraction and understand the hidden patterns in their communications.

  • Large-scale data mining & pattern recognition
  • Word sense tagging & disambiguation
  • Semantic gold-standard annotation
  • Custom LLM training & optimization
  • Automated genre classification
  • Predictive complexity modeling

Image: Legally Blonde (2001) [Photograph]. Warner Bros. Pictures.

Legal Linguistics

I provide empirical, data-driven solutions that resolve linguistic ambiguity and deliver the objective evidence needed to settle high-stakes disputes over legal interpretation, authorship, and the comprehensibility of the law.

  • Miranda-warning comprehension across diverse populations
  • Statutory & contract interpretation — ‘ordinary meaning’
  • Expert testimony & case support
  • Conversation & context analysis

Image: El sueño del caballero — Antonio de Pereda.

Second-Language Acquisition

By analyzing large-scale media data and conducting psycholinguistic experiments, I identify the linguistic and crosslinguistic factors — such as first-language background and feature complexity — that drive learner performance.

  • Crosslinguistic influence analysis
  • Media-based acquisition research
  • Predictive proficiency modeling

Corpus collaboration

American television corpus

A 4.2 million-word corpus balanced by word count and running time across drama, reality TV, sitcoms, and talk shows.

Contact about collaboration
A clustered network visualizing connections across the American television corpus