Catala — A Programming Language for the Law the language design underpinning the Law-to-Code Methodology

Why it matters for NETEVO

Catala occupies the layer of the international Rules-as-Code literature that the Australian implementation lineage does not — the question of what a law-to-code language ought to look like as a programming-languages artefact. Where Representing Legislative Rules as Code at Scale and DataLex XAI in Rules-as-Code cover Australian implementation, and the Bertl, Price and Draheim work covers the 2026 LLM-assisted-extraction frontier, Catala holds the DSL-design seat.

Defeasible logic as the design foundation. Catala's language structure is built around defeasible logic — general rule, then exceptions, then exceptions to exceptions — which is the same logical structure the Income Tax Assessment Act 1997 (Cth) and the Corporations Act 2001 (Cth) actually use. Publication in Proceedings of the ACM on Programming Languages places it within a domain-specific-language tradition with peer-reviewed academic provenance, rather than ad-hoc code generation against statute text; the Law-to-Code Methodology rests on that tradition.

A two-jurisdiction literature. The INRIA and Microsoft Research authorship gives Rules-as-Code a European institutional anchor independent of the AustLII and Greenleaf Australian lineage; read together with the Mowbray, Chung and Greenleaf papers, the field spans both Australian implementation practice and European DSL development. The contemporary debate has four distinct corners — implementation, DSL-design archetype, public-law critique, and defeasible-deontic-logic methodology — and Catala holds the language-design corner. A literature that contains its own public-law critique is, read whole, not a case for naive Rules-as-Code maximalism.