Situated listening..
What began to surface through this work was a tension that already exists within captioning practice. Accessibility guidelines often prioritise clarity and legibility, which can discourage captions from introducing cultural context that is not visibly or audibly present in the film. Yet neutral labels such as ‘[laughs]’ or ‘[street ambience]’ can sometimes flatten relational and cultural nuance as if there is a single universal listener.
Rather than making captions more subjective or explanatory, there is an invitation here to rethink how listening itself is described. Cultural recognition rarely appears through explicitly naming a culture. Instead, it emerges through specificity through attention to rhythm, delivery, texture and social function.
For example, the Southeast Asian example of ‘555’ demonstrates that laughter is not interchangeable across contexts. Laughter can soften hierarchy, signal politeness or create intimacy. When captions reduce these moments to a single bracketed “[laughs]”, they risk erasing relational meaning. However, inserting cultural explanations into captions where they are not present visually shifts accessibility into interpretation. The question becomes how captions might remain grounded in observable sound while still acknowledging relational nuance.
A relational listening approach focuses on what the sound is doing rather than what culture it belongs to. Instead of generic labels, captions can attend to delivery – a brief easing laugh, a restrained exhale, overlapping chatter, distant music leaking through a wall. These are not poetic interpretations but material qualities that situate the viewer within a specific environment.
This lens is not limited to Southeast Asian contexts. Standard English captioning often assumes neutrality, flattening regional dialects, dry humour or class coded sonic environments. A British coastal promenade, a rural pub or a London estate might all be reduced to ‘[crowd noise]’ or ‘[wind]’ even though their textures carry distinct social atmospheres. Situated captioning resists universal categories by foregrounding specificity – not by naming culture directly but by allowing sonic environments to remain particular.
Relational listening therefore shifts captioning away from categorisation toward attentiveness. It asks practitioners to notice density and spatial texture such as rain against metal roofs versus rain against leaves; distant television commentary leaking through a window; layered voices moving between languages. These descriptions remain grounded in what is audible yet acknowledge that sound is always culturally situated.
Seen this way, captions do not become more personal or expressive. Rather, they become more attentive. The Southeast Asian examples function less as content to be replicated and more as a methodological lens, a reminder that neutrality often hides a dominant listening position. By expanding descriptive vocabulary while remaining faithful to the sonic moment, captioning can hold cultural specificity without turning accessibility into explanation.
This research by Celina Loh draws on Southeast Asian communal practices such as eating together as a starting point to explore how relational approaches might enrich UK structural access frameworks, fostering togetherness, mutual hospitality and belonging. From this grounding, the work extends into collaborative listening, sound and facilitation practices, aiming to inform more holistic and culturally responsive approaches to inclusion in the UK arts and cultural sector.
Loh’s research is supported by the British Art Network (BAN). BAN is a Subject Specialist Network supported by Tate and the Paul Mellon Centre for Studies in British Art, with additional public funding provided by the National Lottery through Arts Council England.
