Consensus communities: emergent literary communities and the challenge to genre

Abstract This paper introduces ‘consensus communities’ as a novel computational methodology for understanding literary works within their corpus contexts while challenging the naturalization of basic meaning-making frameworks in literary studies, including genre, period, and style. The methodology enables texts to self-organize based purely on textual affinities, synthesizing these relationships through consensus matrices that integrate Louvain community detection with Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN) density-based clustering to construct a multidimensional representation of the corpus as hybrid field tensors. This hybrid approach, not previously applied to literary analysis, identifies robust structures that persist across algorithmic variations. When applied to twentieth-century Nigerian novels in English, the framework reveals six distinct communities where texts organize independently of traditional literary categories, simultaneously exposing the constructed nature of literary taxonomies while providing radically new interpretive contexts. We further innovate by mapping abstract mathematical communities onto concrete textual passages through Structural Topic Modelling and sentence embeddings. Applied to Chinua Achebe’s Things Fall Apart, this technique identifies specific passages that exhibit pronounced resonance with computational affinities shared across the novel’s community. These computationally identified textual moments not only invite revision of existing scholarship but reveal aspects of literary meaning inaccessible to conventional critical approaches. By prioritizing the dynamic, relational nature of textual meaning within corpus-level structures, this work addresses a central methodological problem in literary studies while expanding its interpretive repertoire. The framework’s corpus-agnostic design enables comparative investigations across diverse literary traditions, offering digital humanities a rigorous, portable tool for understanding literature’s emergent organizational properties beyond predetermined taxonomic assumptions.