A Minimal Computing Approach to Southern African Language Resources

Abstract

This new collaboration between a historian of Southern Africa (RC) and a specialist in computational methods (WJT), is designed to draw on our respective backgrounds and provide opportunities to enlist students and other collaborators in research and teaching. Our goal is to create tools that can be used to help explain unfamiliar languaging in historical contexts. We follow the tenets of minimal computing (Risam & Gil 2022) and take the perspective of language as a complex adaptive system (Kretzschmar 2015). We also situate our work within the postcolonial digital humanities generally (Risam 2018) and the specific critique of knowledge production and racism that Fields & Fields (2012, pp. 5-6) identified as ‘racecraft’, which “highlights the ability of pre- or non-scientific modes of thought to hijack the minds of the scientifically literate”. As practitioners of academic language research and computing, we must be attentive to the history of colonizers trying to not only kill ‘native languages’ but their speakers and cultures (Ngũgĩ wa Thiong'o 2009). To date, we have partially implemented one prototype for automating interlinear morphemic glossing of chiShona and English as shown in Figure 1 (Charumbira et al 2023). Here our intent is speculative design: to imagine a more inclusive space of computational tools and practices that jettisons some of the assumptions that have shaped the digital cultural record in the Global North.