| Autor: |
Mandenaki, Katerina, Sotirakou, Catherine, Mourlas, Constantinos, Moschonas, Spiros |
| Předmět: |
|
| Zdroj: |
International Journal of Communication & Linguistic Studies; Jun2023, Vol. 21 Issue 1, p37-58, 22p |
| Abstrakt: |
In this article, we employ a version of critical discourse analysis augmented with computational tools in order to process texts expressing a neoliberal ideology. We built a corpus of articles and blog entries from the websites of acknowledged think tanks of neoliberal orientation. We implement a series of state-of-the-art algorithms and corpus linguistic methodologies to detect ideological markers and constructions, domain-specific concepts, and seed words associated with the discursive conventions of neoliberalism. Topic modeling uncovered clusters of terms that defined discursive fields, from which we selected a sample of key concepts. Furthermore, we custom trained the word embeddings model in our corpus to detect the semantic fields underpinning neoliberal stereotypes. The vectorization of sentences revealed repetitive patterns formulated around the concepts, thus highlighting the mechanical nature of an ideology in its discursive function. Although our preliminary results confirm and deepen our understanding of expected ideological assumptions such as the establishment of a dominant market as a sole solution and solidarity to all business--and, more specifically, to "small business"--the analysis didn't reveal widespread discourses about specific social strata, for example, people receiving public welfare. However, subtle formulations and metaphors such as the topos of burden as regards the welfare state or the topos of war concerning poverty emerged through our analysis outlining commonsensical discursive frameworks. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
