On the Computational Modelling of Michif Verbal Morphology

Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics, pages 2631–2636 April 19 - 23, 2021. ©2021 Association for Computational Linguistics 2631 On the Computational Modelling of Michif Verbal Morphology Fineen Davis, Eddie A. Santos National Research Council Canada {firstname.lastname} @nrc-cnrc.gc.ca Heather Souter Prairies to Woodlands Indigenous Language Revitalization Circle p2wilrc@gmail.com Abstract This paper presents a finite-state computa- tional model of the verbal morphology of Michif. Michif, the official language of the Métis peoples, is a uniquely mixed language with Algonquian and French origins. It is spo- ken across the Métis homelands in what is now called Canada and the United States, but it is highly endangered with less than 100 speak- ers. The verbal morphology is remarkably complex, as the already polysynthetic Algo- nquian patterns are combined with French ele- ments and unique morpho-phonological inter- actions. The model presented in this paper, LI VERB KAA-OOSHITAHK DI MICHIF handles this complexity by using a series of composed finite-state transducers to model the concatena- tive morphology and phonological rule alterna- tions that are unique to Michif. Such a rule- based approach is necessary as there is insuffi- cient language data for an approach that uses machine learning. A language model such as LI VERB KAA-OOSHITAHK DI MICHIF fur- thers the goals of Indigenous computational linguistics in Canada while also supporting the creation of tools for documentation, education, and revitalization that are desired by the Métis community. 1 Introduction In recent years there has been an increase in com- putational linguistic analysis of Indigenous lan- guages spoken in what is now called Canada, and in particular Algonquian languages such as East Cree (Arppe et al., 2017a), Plains Cree (Har- rigan et al., 2017), and Odawa (Bowers et al., 2017). This paper adds Michif—a mixed lan- guage of Cree and French origin—to the list with a description of LI VERB KAA-OOSHITAHK DI MICHIF. LI VERB KAA-OOSHITAHK DI MICHIF, which translates as “The Michif verb maker”, is a computational model of the verbal morphology of Michif, implemented in the XFST framework of Beesley and Karttunen (2003). 2 Motivation During the 19th century, marriage between French fur traders and Cree and Anishinaabe women in the Métis homeland was common, and their descendants became known as the Métis peo- ples (Bakker, 1997; Rosen and Souter, 2009). Michif (ISO 639-3: crg), an Algonquian lan- guage, emerged as a mixed language which com- bined elements of French with the Indigenous Al- gonquian languages Cree and Saulteaux—a dis- tinct dialect of Ojibwa (Bakker, 1997; Rosen and Souter, 2009). There are many varieties of the Michif language, however LI VERB KAA- OOSHITAHK DI MICHIF is based on a variety spo- ken mainly in Manitoba, Southern Saskatchewan, North Dakota, and Montana. Lack of linguistic documentation and analy- sis restricts the ability to create formalized teach- ing tools and technologies and hinders the efforts of language learners and activists in Indigenous communities. The number of speakers has de- creased significantly. It is difficult to estimate true …

Content truncated to 500 words