Bonnie McLean: Arbitrary by Accident, Iconic by Design: On the role of motivation in the nature and development of lexicons

Abstract

This thesis investigates the role of iconicity, or motivated relations between meaning and form, in the nature and development of lexicons. Papers I and II develop a perspective on iconicity as a subjective process of construal, and detail how a robustness approach can meet the challenges this poses for the operationalisation of iconicity in scientific investigations. Paper III shows how iconicity can act as a factor of attraction affecting stability and change in the development of form-meaning mappings in lexicons, with an experiment exploring the transmission of artificial languages in an iterated learning and interaction paradigm. We see how generations of participants reshape arbitrary languages to become more iconic, and how iconicity is maintained in iconic languages over several generations. The results are explained by iconic mappings being easier to learn and to co-ordinate on. However, it should not be taken from this that iconicity is simple. Papers IV and V show iconicity in natural language is much more complex than it appears in experiments, with an in-depth analysis of ideophones. Paper IV is an fNIRS study of Japanese ideophones, which reveals them to be perfectly honed tools for a variety of complex linguistic, perceptual, and sociopragmatic functions. Paper V explores the challenges ideophones pose for documentary linguistics, presenting a mixed methods toolkit for eliciting, documenting, and analysing ideophones in a way that makes sense of their sophisticated semiotics. I show how, taken together, the papers in the thesis encourage us to reconsider traditional perspectives on language structure that see arbitrariness as the dominant principle, and iconicity as auxiliary. Such perspectives are grounded in a concept of iconicity that is misinformed, and incompatible with the account of iconicity developed in this thesis. With an evolutionary approach, the centrality of iconicity to language structure, and the marginal contributions of arbitrariness become even starker. I argue that it is iconicity that is built into the design of language, while arbitrariness may be merely an accident.