Phonemes as Modeling Devices

Abstract

The study of how sounds in language are used to create words and other linguistic structures that are imitative of some sound property of their referents comes generally under the rubric of sound symbolism theory. While it has been dismissed by various approaches to language, the empirical and anecdotal support for sound symbolism is now too massive to ignore. From the database of evidence it now makes available, various hypotheses can be formulated about originating elements in word formation and, by extension, in seemingly diverse linguistic systems. One of these is the phoneme which bears suggestive meaning in itself as a primary originating element. It is thus a “modeling device” that leads, by extension, to the derivation of larger structures of meaning. Using an adapted version of modeling systems theory as developed by the Tartu School of semiotics, this paper argues that the phoneme is in fact more than a cue for distinguishing words – rather it is an elemental modeling device. This view potentially has some basic implications not only for sound symbolism theory, but also for the study of semiosis itself.