A CORPUS-ASSISTED CRITICAL DISCOURSE ANALYSIS OF COUNTRY REPRESENTATION IN TED TALKS

Abstract

This study investigates how TED represents Kazakhstan in global media discourse through a corpus analysis of TED Talk transcripts. The relevance of the study lies in examining how international media platforms shape the discursive image of Kazakhstan. The data were collected using the TED Corpus Search Engine tool, while the analysis was conducted using a blended critical discourse analysis (CDA) framework. The novelty of the study lies in combining corpus analysis with a tailored CDA framework specifically designed for this research. This framework draws on elements of Fairclough’s Three-Dimensional Model, van Dijk’s Socio-Cognitive Model, and Wodak’s Discourse-Historical Approach. The aim of the study is to analyze the contexts in which Kazakhstan is mentioned and to identify the discursive patterns through which it is represented. This framework was applied to analyze the diverse contexts in which Kazakhstan was mentioned. The methodological approach allowed us to focus on lexical choices, representations of agency, and underlying ideologies in a dataset of selected examples from materials released between 2007 and 2022. The results show that Kazakhstan is often portrayed in TED discourse through several recurring themes. The country’s discursive image on this platform reflects ideas of development, questions of national identity, and the use of natural resources. The theoretical value of the study lies in demonstrating the usefulness of integrating corpus-based methods with discourse analysis for examining global media narratives. The practical value lies in providing insights into how Kazakhstan is framed in influential international communication platforms. Although the article has limitations, such as its focus on English-language transcripts, a relatively small dataset, and a lack of differentiation in the style and length of the materials, it nonetheless contributes to discourse-analytical research by highlighting the value of corpus-based approaches.