A survey of experimental evidence for diachronic change

Abstract

Abstract There are synchronic sources of data that can provide an additional line of evidence which may be useful in reconstructing sound changes: patterns of sound change in progress and experimentally induced changes, variation in production, natural errors in production and perception, experimentally elicited errors in perception and production, and experiments and simulations of iterated learning. This article surveys existing studies that have made use of such evidence in support of sound changes and reviews limitations of experimental methods and factors to consider when designing experiments to use these parallels to inform sound change. To demonstrate the parallels between patterns in synchronic data and sound changes, a sample typology of diachronic developments was compared with patterns of categorical errors from experimentally elicited misperception in adverse listening conditions and errors of perception and production in natural speech. All of these correlations are highly significant, demonstrating the potential of such synchronic data as a source of parallels to provide evidence for reconstructed sound changes.