A Comparative Corpus-Based Study: Exploring Pos-Tagged Phraseological Patterns and Lexical Density In Selected Urdu Novels

Abstract

The current study explores phraseological patterns of selected Urdu novels of contemporary Urdu novelists by using the corpus-based approach. Corpus-based approaches offer regular ways of analyzing the common linguistic patterns and stylistic decisions in literary texts. Previous studies in the Urdu corpus have concentrated on the methods of POS tagging, the development of a corpus, and the accuracy of computation, but little attention has been paid to the utilization of POS-tagged phraseological analysis in the study of stylistic variations in Urdu fiction. Hence, the present study is based on the phraseological patterns of the selected Urdu novels using corpus-based research methodology. A qualitative and quantitative approach was used in the study. Some novels of contemporary writers, namely Meher-un-Nisa Shahmeer and Noor Rajpoot, were selected, and Phraseological units were extracted manually, then it was POS tagged manually. The tagged data were then analysed with the aim of investigating POS frequency distributions of both authors, the occurrence of recurring phrases, and the stylistic differences between the two authors in AntConc.In addition, lexical density was computed to compare the ratio of content words of both corpora. It was found that the dominant category in the combined corpus included nouns. The analysis was carried out to reveal the common phraseological structures and stylistic differences between the author's works. A comparison between the two corpora revealed that Mehrunisa Shahmeer’s corpus tended to be more descriptive and lexically dense, whereas Noor Rajpoot’s corpus had more dynamic and action-oriented constructions with a higher rate of verbs. The study revealed that corpus-based methods are effective in revealing authors' stylistic differences in Urdu fiction. Future research can enhance the corpus by incorporating additional Urdu novels and authors, and may employ more advanced corpus tools or automated tagging methods for broader linguistic investigation.