enGrammatica

enGrammatica is a framework and open standard for describing, validating, and applying English grammar in computational contexts. It combines a formal grammar specification, tooling for parsing and annotation, and an online repository of grammar resources intended for researchers, educators, and developers.

Core components include a formal grammar notation for English syntax and morphology, a statistical-augmented parser, and an annotation pipeline supporting tokenization, part-of-speech tagging, constituency and dependency parsing, and semantic role labeling. The project emphasizes interoperability, enabling grammars to be exchanged between systems and integrated with other resources such as treebanks and lexical databases.

History and scope: enGrammatica originated in an academic collaboration started in 2019 and released as an open project in 2021. It seeks to provide an accessible yet rigorous description of English grammar that can be used for research, education, and production NLP. The name blends the English prefix en- with grammatica, reflecting its linguistic focus and its aim as an executable grammar resource.

Applications: The framework has been applied to grammar checking, educational tools, corpus annotation, and the development of NLP pipelines. It also serves as a testbed for grammar induction, evaluation of parsing strategies, and comparison across English varieties.

Reception and challenges: While valued for openness and modularity, enGrammatica faces issues in standardization, coverage of dialectal variation, and performance with large grammars. Ongoing development emphasizes community governance, contribution guidelines, and alignment with broader linguistic resource ecosystems.