Translating the Untranslatable: An Operationalizable Ontology for Untranslatability
For NLP researchers and MT developers, this provides a framework to analyze and improve MT on linguistically challenging cases where standard benchmarks are insufficient.
The paper introduces a structured ontology and taxonomy of compensation strategies for untranslatability in machine translation, along with a multilingual dataset. Human preference studies show that translations using the Annotation strategy (explanatory context) are preferred.
Untranslatability, cases where meaning cannot be directly preserved across languages, is well-studied in linguistics but underexplored in NLP. As machine translation (MT) systems improve on standard benchmarks, their limitations increasingly concentrate in such cases, where translation cannot be reduced to one-to-one equivalence. We introduce a structured ontology of untranslatability along with a taxonomy of compensation strategies, which are specific techniques to convey meaning under these untranslatable circumstances. We operationalize this framework into a multilingual dataset of untranslatable sentences paired with strategy-based translations, enabling controlled analysis of translation behavior. Initial human preference studies suggest that translation quality depends on the strategy used, with consistent preferences for outputs that include explanatory context, known as the Annotation compensation strategy. Our framework and dataset provide a foundation for studying and modeling strategy-informed machine translation.