Swedish vocalic size-sound symbolism in the lexicon and non-linguistic utterances

Phonologists increasingly recognise the existence of imitative, synesthetic, and other iconic, rather than arbitrary, relationships between the sounds and semantics of linguistic items. Abelin (1999) has provided evidence of phonoesthemic strings in Swedish: non-morphemic consonant clusters occurring systematically with specific semantic elements. This paper extends investigation into iconic phonology in Swedish, and considers vocalic size-sound symbolism: a synesthetic phonological relationship between the size of the referent and the relative size of the vowel space created in phonetic production. The respective articulatory parameters for vowels present in Swedish size adjectives show a significant tendency to constitute open/mid-back vowels for ‘large’ (eg. stor - /stɔɾ/ (big)), and high-front vowels for ‘small’ (eg. pytte - /pytɛ/ (tiny)). Six L1 Swedish speakers completed tasks to test sensitivity to size-sound symbolism in pseudowords and non-linguistic utterances. As predicted, when asked to choose between two pseudowords as a name for a larger and smaller creature, “Tol” (/tɔl/) was consistently assigned to the larger, and “Til” (/til/) to the smaller. When presented with the same images, and asked to provide the “noises” that these animals would make, participants uniformly chose open vowels for the “call” of the large animal, and high vowels for the small. Participants were assigned written questionnaires or telephone interviews, with fictionalised drawings, or photographs of real animals. This variation was intended to test whether the ‘performative intensity’ (Dingemanse, 2012) versus ‘naturalness’ of a speaking situation correlated with more or less iconic language use. There were no significant differences across the conditions. These results add to the body of evidence that iconic phonological phenomena are noncoincidental and systematic. This paper analyses these results in context of embodied cognition, and the relationships between language and gesture (Rossini, 2012), the environment (Tsur, 2006), and the neurological organisation of differing sensory modalities (Spence, 2011).