"The goal of the current proposal is to provide a new empirical foundation to our knowledge of sound patterns that are statistically underrepresented (typologically rare) in the world's languages. Typologically rare patterns have been claimed to be more difficult to produce and perceive, and more prone to undergo sound change. Yet this claim is controversial: others maintain that due to the stabilizing force of grammar and learning, within a language, all sound patterns are equally entrenched. Understanding the status and the origins of typologically rare patterns remains one of the most profound challenges in linguistics, phonetics, as much as in psychology and cognitive science. There is, however, an insufficient empirical knowledge of the status of typologically rare patterns in languages that have stabilized these patterns in their synchronic grammars. Whether typologically rare phonotactic patterns can be shown to be relatively less entrenched within a speaker's grammar is an open empirical question and will be the major research focus of this proposal. Based on several languages (Germanic, Slavic, Kartvelian) the current proposal investigates how the licensing of typologically rare phonotactic patterns (reverse sonority clusters, syllabic consonants) interacts with language particular characteristics of articulatory implementation (the coordination of articulator motion in space and time). The detailed production studies are complemented by imitation experiments that allow us to assess whether typologically rare patterns show greater plasticity through language use. The results will make a significant contribution to our understanding of the perception-production link in speech, language typology and sound change, and the reciprocal relationship between cognitive and physical forces in spoken language."