Building a Dataset from Inside Black Short Stories

By Kenton Rambsy 

I am building a dataset drawn from more than 200 Black short stories by annotating granular textual features, such as dialogue, geography and settings, and character demographics, to create a level of structured data on creative writing that is still rare in digital humanities and African American literary studies.

The goal is twofold. First, we are constructing one of the most sizable datasets focused specifically on Black short fiction. Second, we are generating insights into recurring narrative features so readers can identify stories based on interests in space, dialogue patterns, character types, or structural tendencies.

For more than a decade, my research tracked circulation and visibility. In “Discovering the Big 7,” from 2019, I identified 7 short story writers most frequently anthologized across decades. In “Studying 500 Black Writers in the New York Times,” from 2022, my brother Howard Rambsy II and I examined how media attention concentrates around a limited group of writers.

Those projects documented where writers appear and how often institutions return to them. This current annotation project shifts the focus from circulation to composition by examining what happens inside the stories themselves. Publication records can be organized into trends, but dialogue, spatial orientation, and character interaction must be identified through line by line reading and interpretive judgment.

By isolating these elements, structural patterns become measurable rather than intuitive. Baldwin and Ellison, across six stories each, both rely mostly on settings and projected spaces, yet Baldwin’s 301 total spatial references compared to Ellison’s 70 show a far denser layering of geography. The value of the dataset lies in making those similarities and differences visible in concrete terms rather than leaving them as generalized claims.

Related:

• The Black Short Story Project