finding symbolism in r.f. kuang’s yellowface (2023) by chatgpt: a qualitative hermeneutic inquiry

The current study conducts a critical analysis of the interpretative abilities of large language models simply known as llms, specifically openai’s chatgpt-4, in discerning and scrutinizing intricate literary symbolism within culturally and politically nuanced fiction. employing r.f. kuang’s satirical novel yellowface (2003) as our focus, the research uses a qualitative and iterative prompting methodology to evaluate the model’s capacity to engage with key symbolic passages. these passages discuss important themes such as cultural appropriation, digital performance, and the ongoing problem of historical erasure. the findings show that chatgpt-4 can reliably recognize definite symbols and apply established theoretical frameworks when given clear, directive prompts. however, its default interpretive tendencies reveal significant limitations. specifically, the model shows a consistent pattern of generalizing and, in effect, depoliticizing critiques which are racially and culturally specific. this tendency results in interpretations that lack the contextual depth and do not engage with the novel’s satire and its subtle interrogation of racial capitalism. the model’s ability to engage with satire, multivalent or layered meanings, and the necessity for contextual grounding remains notably insufficient.