Comparative analysis of scholarly works: a linguistic analysis through word frequency between Chen Yinque and Ji Xianlin

This study investigates the intellectual trajectories of Chen Yinque and Ji Xianlin, foundational figures of modern Chinese humanistic scholarship, through computational text analysis of their published works. It argues that lexical evolution in scholarly corpora quantifiably traces how intellectuals reorient conceptual systems under historical pressure. Using Jieba segmentation and TF-IDF weighting to identify characteristic terminology, the research traces diachronic lexical shifts across each scholar’s career, while co-occurrence networks map the structural organization of their intellectual frameworks. Although extensive biographical scholarship exists, no prior study has empirically quantified how their research vocabularies evolved over time, nor established a measurable link between such shifts and 20th-century China’s political and cultural transformations. Results show Chen’s vocabulary shifted from statecraft and institutional history toward literary aesthetics and poetic historiography, whereas Ji’s evolved from Sino-Indian philology and Buddhist scholarship toward national cultural philosophy. These findings are corroborated by contrasting network topologies: Chen’s depth-oriented, discipline-centered clustering reflects a retreat into aesthetic scholarship as philological refuge amid the political transitions of the 1940s–1950s, while Ji’s distributed, bridge-building connectivity reflects post-1978 cosmopolitan expansion into global academic discourse. The analysis operationalizes network structure as a measurable indicator of divergent scholarly strategies (retrenchment versus engagement) and supplies the empirical, full-career comparison of two major Chinese humanists across distinct political eras. This framework applies to other scholarly traditions, especially non-Western languages, advancing comparative intellectual history from biographical interpretation toward reproducible corpus-based analysis.