Using Corpus-Assisted Discourse Analysis to Identify Food Ideologies: a Study of Online Product Descriptions

The present paper relies on a combination of corpus linguistics and discourse analysis to reveal how food manufacturers construct very specific product identities through product descriptions, which (1) reflect some cultural and institutional values, and (2) have some implications of food ideology. This investigation primarily focuses on the text functions of bread descriptions, the dominant verbal discourses of bread promotion in Lithuania, and the typical linguistic choices used to disseminate these discourses. The data includes product descriptions available on the websites of six major industrial bread bakers in Lithuania and consists of more than 150 product descriptions (approx. 10,000 words). The lexical choices in these texts are evaluated by referring to the Corpus of Contemporary Lithuanian (tekstynas.vdu.lt) as a general reference corpus. The results show that in Lithuanian bread descriptions two text functions dominate: descriptiveness and prepared expressiveness, which further indicates that the target texts are closest to academic prose and fiction. The analysis of collocations and word frequencies supports the prevalence of features characteristic of both academic prose and fiction. This new digital genre is used by producers to disseminate and emphasise certain values and ideologies through explicit verbal references to tradition, inheritance, and continuity.