A ntsomi is built not from a fixed plot but from a core-image: a remembered, emotionally charged cluster of action (a song, a repeated gesture, a confrontation) that the performer carries in memory. The performer expands, contracts, and recombines these images in real time to produce the full tale.
Performers construct narratives by linking and developing self-contained image-units, each expansible to any length through repetition, detailing, and patterning. Length and complexity of a performance come from how many images are joined and how far each is elaborated, not from a memorized script.
In ntsomi, structured repetition (parallel episodes, recurring songs, cyclical returns) is not filler but the primary device that organizes the narrative and conveys its argument. The audience experiences meaning through the rhythm of recurrence and variation rather than through linear plot progression.
Treating oral narrative as a performance event rather than a fixed text shifts attention from a single authoritative version to the live interaction of performer, body, song, and audience. Each telling is a unique realization, so the tradition lives in the process of composition rather than in any one preserved artifact.
The ntsomi performer is not a passive transmitter but an artist who draws on a shared stock of images and recombines them according to individual skill, mood, and audience response. Originality lies in arrangement and development of inherited materials rather than in inventing new content.
Recurring songs and physical gestures function as fixed points that stabilize the otherwise fluid image, serving as mnemonic anchors the performer returns to. These non-narrative elements give the audience recognizable signposts and bind the expansible material into a coherent whole.