A CORPUS-BASED COMPARISON OF LEXICAL BUNDLES FOUND IN RESEARCH PAPERS OF BRITISH AND PAKISTANI RESEARCHERS PUBLISHED IN THE FIELD OF ENGLISH LINGUISTICS

Abstract

The art of persuasive writing is partially contingent upon the formulaic use of language. Lexical bundles, as formulaic sequences, play a crucial role in constituting persuasive and coherent discourse. Research on lexical bundles seems to have grabbed the attention of EAP instructors and researchers. Although it seems that less attention has been paid to the recurrent expressions found in the research discourses of non-native speakers. The current study compares the use of lexical bundles in research papers by British and Pakistani researchers in the discipline of English Linguistics, focusing on frequency, structure, and function. This study employed mix-method approach, incorporating a self-developed corpus of 1,086 research papers. This corpus was further classified into two sub-corpora, COBRA and COPRA. The corpus was further analyzed by using AntConc (version 4.2.4, Anthony, 2023), which generated two lists of four and five-word target bundles consisting of 158 bundles in COBRA and 229 in COPRA. For structural and functional analysis, Biber et al.’s (1999) and Hyland’s (2008a) taxonomies were used. The findings indicated that both corpora were dominated by PP-based, NP-based, and VP-based bundles, with research-oriented bundles being the most frequent functional category. While British writers depicted the preference for NP + of bundles in locative and descriptive contexts, Pakistani writers overused them in procedural, quantification, and topical contexts. This study plays a significant role in understanding the formulaic use of language in research writing by focusing on cross-cultural variations in research discourse.