Between Care and Contempt: Corpus-Assisted Sentiment Analysis of Disability Representation in Feryal Ali-Gauhar’s An Abundance of Wild Roses

Abstract

This article examines how the village in Feryal Ali-Gauhar’s An Abundance of Wild Roses (2024) responds to Lasnik, a boy who limps, leans on a staff and stutters. It uses corpus-assisted qualitative sentiment analysis: a script extracts passages and a researcher codes them by hand, and no classifier is used. A pilot extraction, capped at 45 lines per character pair, shrank from 28 coded lines to 14 audited passages and showed how counting artefacts produced a false 12-12 split between positive and negative sentiment. A revised, uncapped extraction (all 226 occurrences of the name, a whole-text lexicon scan and an orthographic search for marked stuttering) yielded 43 coded units. Each unit was coded for stance toward the impairment, stance toward the individual, and narrative position (direct speech, thought, conduct, narration or Lasnik’s own perspective). Character-voiced responses to physical impairment (gait, wounds, bandages, staff) were more often caring than responses to his speech: five of nine physical units were positive against three of twelve speech units. Ridicule and contempt, a category added in revision, appeared only where impairment was fused with judgements of intellect or conduct; all six such fused stances were negative. The clearest cases in which a listener took up what Lasnik said were warm, but they were few. The article argues that the novel stages physical care without communicative recognition, reads this pattern through Fricker’s (2007) account of epistemic injustice, and shows that keyword-based sentiment analysis can locate the pattern but cannot interpret it.