This paper examines how pronunciation instruction is enacted in online Swedish as a Second Language (SSL) within Municipal Adult Education (MAE). Using Actor–Network Theory (ANT), classroom observations and teacher interviews trace how human and non-human actors (e.g., teachers, students, platforms, cameras, microphones, recordings) shape instruction. Five analytic scenes show that: (1) close-up camera/microphone affordances amplify articulatory modeling; (2) prosody practice becomes an auditory, distributed activity coordinated via audio files and chat functions; (3) corrective feedback is configured by one-speaker-at-a-time interfaces and headphone listening; (4) assessment is redistributed over time; and (5) domestic soundscapes and connectivity contingencies unevenly condition what is hearable and assessable. In the paper, pedagogical implications grounded in these scene-based analyses are articulated. Pronunciation instruction is thus shown to emerge as a sociomaterial practice, enacted through the actors and infrastructures of the online classroom.
Andreas Lagerlöf (Thu,) studied this question.