A Database of Modal Typology

This is a database recording (at least) which force-flavor pairs various modal expressions across the world's languages can express. It is designed to be both easy to use and to contribute to. For more details and motivation, please see the following paper:

Qingxia Guo, Nathaniel Imel, Shane Steinert-Threlkeld, "A Database for Modal Semantic Typology", Proceedings of the 4th Workshop on Computational Typology and Multilingual NLP (SIGTYP 2022), pp 42-51.

Structure

Each language has a corresponding directory, with the following minimal structure:

language/
|- modals.csv
|- metadata.yml

modals.csv is a CSV file assumed to have the following four columns:

• expression
• force
• flavor
• can_express: this field will have value in [1,0,?]
• polarity
• syntactically_negated: this field will have value in [high,low,no] to indicate the relation between modal and negation, low means negation scopes lower than modal and vice versa. no means the expression is not syntactically negated.
• full_form

Each row is one observation: can_express has values stating whether or not an expression (orthographically represented) is capable of expressing a particular combination of force and flavor.

metadata.yml is a YAML file containing basic data about the language, the references used to generate the data in modals.csv and the identity of the contributors.

Contributing

See CONTRIBUTING.md