A Corpus-Based Analysis of Figurative Language in Travis Scott’s Hip-Hop Lyrics (2013–2023)

Abstract

This study investigated the frequency and function of figurative language, as well as its temporal developmental trends, in the lyrics of Travis Scott, a prominent figure in contemporary hip-hop. A corpus-based analysis was conducted on 132 songs (65,841 words) released between 2013 and 2023, sourced from AZLyrics.com. Using manual annotation grounded in cognitive and pragmatic theories (Lakoff & Johnson, 1980; Grice, 1975; Black, 1962), ten types of figurative language were identified and quantified: simile, metaphor, personification, hyperbole, allusion, onomatopoeia, irony, double entendre, paradox, and symbolism. Results indicated that metaphor (21.99%), hyperbole (15.86%), and symbolism (12.26%) are the most frequently used devices, while irony (2.74%) was the least frequently used.  Functionally, these rhetorical devices work together to create an abstract mode of perception, moving beyond traditional narrative structures toward the central aesthetics of psychedelic hip hop. Temporal analysis further shows that while metaphor and hyperbole remained stable core features throughout the decade, linguistic complexity increased. The growing use of ambiguity reflects linguistic maturation, and the highest frequency of onomatopoeia appeared during 2016–2017, aligning with a significant shift in musical dynamics. The findings indicate that the lyrical style emphasizes sensory imagery and emotional reflection rather than the sequential storytelling typical of traditional hip hop. The study also demonstrates reliability through inter-coder agreement, with a Cohen's kappa value above 0.80 based on a random sample of 13 songs, and contributes to the expansion of corpus-based linguistic analysis in the context of music.