One of the currently biggest challenges in NLP is to develop multilingual language technology. Lack of data in low-resources languages poses great difficulty to NLP researchers and limits NLP technology's availability to a small number of resource-rich languages. It has been shown that linguistic typology and the knowledge of language universals can help NLP in the development of multilingual resources. To contribute to this research area, we present a fuzzy approach to language universals. Our proposal combines a constraint-based formalism with fuzzy logic to define a fuzzy-gradient model to characterize linguistic universals. This model will allow us to evaluate linguistic universals and to define a universal grammar. This universal grammar will be integrated into an automatic technique to infer from linguistic data the particular grammar of any understudied natural language.

This paper aims to establish a link between linguistics and fuzzy phenomena. Throughout the history, linguistics has been relying on discrete descriptions to explain Natural Language Processing. This fact has a negative impact in various areas of language and technology since linguistics rejects all the inputs which are out from a discrete rule. Although this paper might approach the subject more from a linguistic than from a mathematical point of view, we present the theoretical considerations and reasoning used in elaborating a formal characterization of fuzziness in natural language grammars. Property Grammars will be used as the formal theory in order to explain Natural Language fuzziness and variability.