Using MonoConc Pro to teach and learn lexical collocations in Xitsonga

Abstract

Few language resources have been developed for indigenous languages in South Africa. Surprisingly, these are also official languages which constitutionally share the same language status as other languages. One of the major challenges for the development of basic language resources is the lack of digital corpora that can be used to train and develop the resources. Such a challenge has impeded the use of technology for research, learning, and teaching domains in indigenous languages. In this study, we used MonoConc Pro, a concordancer, to demonstrate how language users can utilise the software to display lexical collocation from a corpus for teaching and learning purposes. We illustrated how corpus can be used simultaneously with language technology to teach and learn aspects of linguistics in a form of lexical collocations in Xitsonga. An Autshumato Xitsonga Monolingual Corpus (AXMC) that was retrieved from the South African Centre for Digital Language Resources repository was used as data for analysis. The AXMC is a corpus that was collected and semi-automatically aligned at the sentence level during the Autshumato project. To search for lexical collocations, we interrogated the AXMC corpus using the MonoConc Pro program. A semi-automatic search for collocates of Xitsonga adjectives lavakulu, letikulu, lavantsongo, and letinene was conducted. The study found that lexical collocations or words that co-occur with adjectives are nouns, adjectives, possessives, and relatives. It was also observed that each adjective frequently collocates with certain nouns belonging to a specific class. The results obtained suggest a practical way in which language technologies can be used to explore corpora and examine language patterns for teaching and learning. We hope that this line of study will lead to the use of modern language resources to examine linguistic traits in indigenous languages.