This phenomenon of "sound symbolism" has received a lot of research attention in the last 10 years or so. For a long time it was considered a curiosity at best, and a total red herring at worst, but a lot of evidence is accumulating that sound symbolic effects are very real and may have profound implications for our understanding of sensorimotor cognition.
That seems to me like it just shifts the problem one level. Why are K's and Kikis spiky and why are B's and Boubas round. Why is it universal too across people with different writing systems and languages.
4 comments