Plural verbal agreement with sports teams (“Liverpool have beaten Manchester United 4-0 tonight”) is a distinct feature of British English. Prior corpus-based studies show that plural verbal agreement tends to be more frequent with collective nouns in sport texts than in other news (Hundt 1998; Levin 2001). However, beyond these genre-level differences, there are significant gaps in our understanding, particularly regarding sociolinguistic and individual-level variation in this grammatical domain.

This presentation investigates the verbal agreement patterns of individuals who use English as a second language (L2) in public settings. Our study focuses on foreign-born football managers working in England, specifically in the English Premier League, which has attracted managers from various European countries since the late 1990s. These managers, who are public figures and native speakers of languages other than English, conduct pre- and post-game press conferences in English as part of their managerial duties. The individual informants were chosen because the agreement patterns in their native languages favor singular verbal agreement.

We are interested in whether our these L2 speakers of English use singular or plural verbal agreement when talking about their own teams and their opponents. Additionally, we aim to determine if these agreement forms remain stable or change during their careers in England.

To explore this phenomenon, we utilize an underused source for accessing individual variation: a large set of interviews and press conferences available online. We use automatically-generated transcripts from YouTube recordings, with data obtained through national digital humanities infrastructure projects in Finland and in Sweden. This dataset includes recordings of managers’ interviews and press conferences spanning roughly a decade from the mid-2010s onwards.

The topic is of interest to various audiences, not limited to those studying agreement patterns. Our unique data collection also appeals to people interested in digital humanities. The corpus-based observations not only shed light on verbal agreement patterns in proper noun + verb structures but also contribute to debates on how idiolects change in the short-term post-adolescence (Sankoff 2018). The results could provide insights into how non-native adult speakers handle stable lexico-grammatical changes in society. Prior studies have identified three trajectories of individual change: stability, adoption, and swimming against the community current.