Introduction

With respect to sound symbolism, I accept as a premise the position which appears to me self-evident, and in any case proved in psycholinguistic experiments: that both synesthetic (or: universal) sound symbolism and conventional (or: local) sound symbolism coexist in every language. I suggest that this coexistence is precisely what should be suspected: it is surely natural that the direct links between sound and meaning should be based both on phonetics and on the lexicon.

This is very probably a simplification of the real state of affairs. I suggest, however, that it is worth following this simplified hypothesis to its conclusion: namely, that it is to be expected that different languages will have synesthetic and universal sound symbolism in differing amounts. Over time, the sound changes and the semantic changes that occur may bring the two closer together, or, conversely, may force them further apart, and hence some languages may be expected to have richer overall sound-symbolic associations than others: thus, reputedly, Korean, and some West African languages.

I think that there are more factors at work than this stark dichotomy suggests; in particular, it seems that there are cultural factors that may or may not emphasize the use of certain kinds of sound symbolism for certain purposes. Nevertheless, I suggest that the disparities between the scope of sound symbolism in different languages may be explained at least in part by the “coalition-or-conflict” theory I have outlined.