NARRATIVE IDENTITY AND EXPOSITORY STRUCTURE: AN NLP-BASED GENRE STUDY OF BOOK 2 TEXTS

Abstract

This study examines genre variation in Intermediate English Book 2 through a corpus-based and computational framework, focusing on the distinction between narrative (hero essays) and expository prose. Using techniques from Natural Language Processing, the analysis integrates syntactic, functional, and lexical measures, including sentence length, clause types, tense usage, and lexical patterns. The findings reveal that narrative texts are characterized by past tense dominance, temporal sequencing, and moderate syntactic complexity, resulting in a linear discourse structure that supports accessibility and fluency. In contrast, expository texts exhibit higher levels of subordination, longer sentences, and more abstract vocabulary, producing a hierarchical structure that increases cognitive demands and supports analytical reasoning. The study demonstrates that genre variation in Book 2 is systematic and measurable, with expository texts consistently occupying the higher end of the complexity spectrum, particularly in scientific and historical lessons. Clause-level analysis further shows that dependent clauses play a central role in encoding logical relationships in expository discourse, while narrative texts rely more on sequential organization. These differences highlight the role of genre in shaping both linguistic structure and cognitive processing. The study argues that the integration of narrative and expository texts in Book 2 provides a balanced pedagogical framework, enabling learners to develop both narrative fluency and analytical competence. By applying NLP-based methods to textbook analysis, the research contributes to genre studies and second language acquisition by offering a data-driven approach to evaluating instructional materials. The findings have important implications for curriculum design, emphasizing the need to balance linguistic complexity with pedagogical accessibility.