Iconic hand gestures from ideophones exhibit stability and emergent phonological properties: an iterated learning study

Abstract Ideophones are marked words which depict sensory imagery and are usually considered iconic by native speakers (i.e., ideophones sound like what they mean). Owing to shared cross-linguistic characteristics of expressive prosody, reduplication, and unusual phonological structure, ideophones have been likened to meaning performed. Iconic hand gestures frequently occur alongside ideophones choreographed to the timing of syllables. Given the visual modality’s richness in iconic affordances, these gestures have been supposed to help interlocutors infer semantic nuances and contextualize utterances, especially when an ideophone is polysemous, and may even inform speakers’ mental representations of spoken language as imitative. Such gestures should therefore be learnable and replicable like any unit of language. This is what we indeed find. Using a linear iterated learning paradigm, we investigated the stability of iconic gestures from Japanese and Korean ideophones transmitted across five generations. Despite noise in the visual signal, participants’ hand gestures converged, speaking to the emergence of phonological targets. Handshape configurations over time exhibited finger coordination reminiscent of unmarked handshapes observed in phonological inventories of signed languages. Well-replicated gestures were correlated with well-guessed ideophones from a spoken language study, further highlighting the complementary nature of the visual and spoken modalities in formulating mental representations.