A Corpus-based Study of the Lexical Bundles in the Ph.D Linguistics Dissertations’ Abstracts in Pakistani Universities

The present study analyses the lexical bundles (LBs) used by scholars of PhD linguistics in Pakistani universities. For this purpose, a small corpus of the abstracts of 73 PhD linguistics theses ranging from 2017 to 2022 was built. Antconc software was used to get the frequencies of the LBs. The objectives of the study were to examine the structural and functional taxonomies of those LBs. Biber's (1999) classification was used for the structural analysis, while Hyland's (2008) classification was used for the functional analysis. The study found there is a less tendency toward the use of 4-word LBs. It was disappointing to observe a low frequency of structural analysis for 4-word LBs. On the other hand, the functional classification of the LBs was relatively satisfactory due to the discovery of a higher number of 4-word LBs.The study stresses the importance of teaching the structure and functions of LBs from the intermediate level so that when these students reach the postgraduate level, they should be able to adhere to the rules of cohesion and coherence.