Iconic inferences about personality: From sounds and shapes

A general assumption in modern linguistics is that the relationship between sounds and meanings is arbitrary; i.e. that there are no iconic relationships between sounds and meanings. This thesis of arbitrariness was very clearly articulated in the work of Saussure (1916; 1916/1972) as the first principle that governs languages, and it has been influential in modern linguistic theorizing since then. On the other hand, there has also been a large number of studies suggesting that there are some non-arbitrary relationships between sounds and meanings—cases known as “sound symbolism” (Hinton et al., 1994). For example, for speakers of many languages, low and open vowels like [a] are associated with “big” images, whereas high and closed vowels like [i] are associated with “small” images (Sapir, 1929). Most, if not all, cases of sound symbolism can be considered as an iconic projection from some phonetic properties of sounds to meanings—for the case of [a] being bigger than [i], one plausible explanation is that the mouth is more widely open for the articulation of [a] than it is for the articulation of [i] (see Jespersen 1922/1933; Sapir 1929; Shinohara & Kawahara to appear); or alternatively, the resonance cavity for the second formant resonation (F2) is larger for [a] than it is for [i] (Ohala, 1983, 1994; Shinohara & Kawahara, to appear). Regardless of whether sound symbolism is based on articulatory or acoustic characteristics, we see evidence for iconic relationships