How “U” are “U” words?

Abstract

That U (upper-class) and non-U (non-upper class) speakers are identifiable through their vocabulary is an axiom in England. These claims are repeated in books, in print media, on social media, and in conversations regarding social class. However, such claims are seldom investigated empirically. To redress this, we consider the production and perception of allegedly U and non-U lexis through two studies. In the first, we identify the sociolinguistic distribution of the usage of three variables which are purported to be indicators of socioeconomic status, namely, loo, napkin, and sofa. The second study employs the matched-guise technique to investigate the perception of variants of these three variables. The production results reveal that all three variables exhibit change in apparent-time with limited evidence of class-based variation. In the perception study, we find no systematic class-based indexicalities across the variables. Ultimately, our findings challenge the belief that allegedly U words are shibboleths of upper-classness.