Published online by Cambridge University Press: 19 September 2018
Popular song lyrics constitute an exception to dominant, standard language ideologies of English: nonstandard grammatical forms are common, relatively unstigmatized, and even enregistered in the genre. This project uses song lyrics to test whether genre cues can shift linguistic expectations, influencing how speakers process morphosyntactic variants. In three self-paced reading experiments, participants read sentences from pop songs. Test sentences contained either ‘standard’ NPSG + doesn't or ‘nonstandard’ NPSG + don't. In Experiment 1, some participants were told that the sentences came from lyrics, while others received no context information. Experiment 2 eliminated other nonstandardisms in the stimuli, and Experiment 3 tested for the effect of stronger context information. Genre information caused participants to orient to the sentences differently, which partially—but not straightforwardly—mitigated surprisal at nonstandard don't. I discuss future directions for understanding the effects of context on sociolinguistic processing, which I argue can inform concepts like genre and enregisterment, and the processes underlying language attitudes. (Morphosyntactic variation, genre, invariant don't, language ideology, pop songs, experimental sociolinguistics, sentence processing)*
Thanks to the editor and anonymous reviewer, who provided critical and helpful readings of this article. I would also like to acknowledge Robin Queen and Julie Boland for longstanding conversations that have shaped this research; my research assistants, Bethany Toma and Jamie Smith; and my departmental colleagues, Tracee Mohler and Wayne Lovely, who helped me in various ways to get this work done. I am also grateful to audiences at SVALP, Stanford University, and the Graduate Center at CUNY for their valuable comments and questions on this work as it evolved. All shortcomings are, as always, my own responsibility.