The Role Of Natural Language Process- ing In Creating Ju/’hoansi Stories With Structurally Accurate Cultural Narratives

Natural Language Processing (NLP) has made significant progress in enabling machines to understand and generate human language. However, most advancements have been concentrated on high-resource languages, leaving African indigenous languages, such as Ju/'hoansi, largely under-represented in both research and application. Ju/’hoansi is an endangered language with a rich oral storytelling tradition, yet it lacks the digital resources and computational frameworks required for meaningful NLP engagement. This research moves beyond basic text generation to investigate how narrative structure and cultural context can be modelled and preserved within AI systems. The study focuses on analysing the narrative structures embedded in Ju/’hoansi stories and developing methods to incorporate these structures into a language model. Rather than relying solely on fine-tuning, the research explores how culturally grounded narrative patterns, themes, and storytelling conventions can guide the generation process. This approach aims to produce stories that are not only linguistically coherent but also culturally authentic and structurally faithful to traditional forms. To evaluate this, the research adopts a hybrid assessment strategy that combines computational metrics with community-informed evaluation, ensuring that the generated stories are meaningful and appropriate from both a technical and cultural perspective. Ultimately, this work contributes to the growing field of NLP for low-resource languages by emphasising cultural preservation, narrative integrity, and community involvement. It positions AI not just as a tool for language generation but as a means of supporting the continuity and revitalisation of Indigenous storytelling traditions.