A corpus-based analysis of language ideologies in Hungarian school metalanguage

Abstract

The main goal of this paper is to present a recently built interview corpus called Corpus of Hungarian School Metalanguage – Interview Corpus (CHSM-IC) and its potential in language ideology studies. This corpus was compiled during a broad survey on Hungarian school metalanguage carried out in 2009 and was recently made available for a wider research community within the CESAR (Central and South-East European Resources) project. The study investigates interactional routines used in metadiscourses on language use. Printed texts cited from prestigious handbooks and interview data from CHSM-IC are compared. Thus, widely used, culturally-inherited text fragments are detected and confronted with the interviewees’ narratives on their own communicational experiences. A case study on the discourse marker hát (‘so’, ‘well’) illustrates that there is a conflict and often a controversy between language ideologies disseminated by the Hungarian school system and the linguistic selfrepresentation in the interviewees’ narratives. Combining Language Ideology, Conversation Analysis, Discourse Analysis and Discursive Social Psychology frameworks, the paper presents a detailed description on the emergence of metadiscourses in a school setting. The paper concludes that metalinguistic utterances (e.g., answers on grammaticality, statements on linguistic accuracy, etc.) and observable, spontaneous (or semi-spontaneous) language use patterns are regularly not in accordance with each other.