Self-study Tools of Morphologically Rich Languages: A Prototype of a Nahuatl Learning App




Mobile-Assisted Language Learning (MALL) tools support the self-study of languages and have been linked to positive results in several language skills [1]. However, the majority of the tools were developed for, and are tailored to, the languages most learnt as second languages, such as English and Spanish. This is a limitation because these languages don't adequately represent the richness of the world's languages. Particularly, it is argued [2] that the existing MALLS don’t adequately address the specifics of Morphologically Rich languages (MRLs) and instead focus on vocabulary memorization. 

In computational linguistics, MRLs are languages in which a significant part of the information is encoded at the word level, and the words are composed of multiple morphemes [3]. This corresponds to agglutinative, polysynthetic and fusional morphological types.  An example of a MRL is Nahuatl, an agglutinative polysynthetic language from the Uto-Aztecan language family spoken in central Mexico. The polysynthetic character of Nahuatl can be seen in the example of the incorporation of the object chichik ‘beer’ into the verb nichichiktentlapohketl  ‘I am a person who opens beer’ (ni-: 1st person singular; tentlapoh: to open; -ke-: agentive nominazer; -tl: absolutive ending). 

Research Goals and Results

The present work studies how MALL tools can efficiently support the acquisition of Nahuatl morphology. First, the findings related to the self-study of morphologically rich languages are extracted from a literature review of relevant works in the broader fields of psychology, language teaching, and digital language learning. Second, using the results from the previous step, the features of agglutinative MRLs that could pose the biggest difficulties for learners whose first languages are not agglutinative are identified. Finally, is it tested whether and how the existing Nahuatl learning apps introduce or test the morphology of Nahuatl. The results show that only one of the available MALL tools introduces any kind of morphological explanation, and that even this tool has a lot of potential to improve.


The present work contributes to a better understanding of how to create effective learning apps that are tailored both to the diversity of languages worldwide and to human cognition, through psychological concepts such as motivation and engagement. It also contributes to the visibility of minority languages in the digital world. In a planned follow-up study, the current findings will be used in the creation of a prototype of a Nahuatl learning app.


[1] M. Shortt, S. Tilak, I. Kuznetcova, B. Martens, and B. Akinkuolie, “Gamification in mobile-assisted language learning: A systematic review of Duolingo Literature from public release of 2012 to early 2020,” Computer Assisted Language Learning, vol. 36, no. 3, pp. 517–554, Jul. 2021. doi:10.1080/09588221.2021.1933540 

[2] N. Gilbert and M. Keet, “Automating Question Generation and Marking of Language Learning Exercises for isiZulu,” in Controlled Natural Language, Berlin: Springer, 2010, pp. 31–40 

[3] R. Tsarfaty, et al. “Statistical parsing of morphologically rich languages (SPMRL) what, how and whither,” Proceedings of the NAACL HLT 2010 First Workshop on Statistical Parsing of Morphologically-Rich Languages. [Online]. Available: [Accessed: 30-May-2024]