Title: Decoding surrogate speech: Phonetic and phonemic levels in musical surrogate languages
Abstract: Many cultures around the world have traditions of musical surrogate speech, i.e.communication using a musical instrument to encode linguistic structure. Stern (1957) identifies two major types of systems, so-called “abridging” systems that represent elements of phonemic structure and “lexical ideogram” systems that represent concepts directly. This talk focuses on the former. Drawing on case studies from the literature and original fieldwork, I demonstrate that decoding an abridging system means determining not only what contrasts are encoded but also at what level. Remarkably, in many systems, phonemic structure is not encoded uniformly. In the West African Sambla balafon system, for instance, tone is encoded at a morphophonemic level, eschewing postlexical processes common in the spoken language, but rhythmic encoding shows evidence of surface phonetic gradience. A similar situation holds for the Amazonian Bora drumming system, with phonemic encoding of the two tone levels but a tight correlation between interstrike duration and spoken V-to-V intervals. Languages with surrogate systems on more than one instrument offer an opportunity to determine which factors influence how a contrast will be encoded. Yòrubá, for instance, can be encoded on at least two types of drums, the tension drum (“talking drum”) dùndún and a double-headed barrel drum ensemble known as bàtá.The dùndún encodes tone at a surface level, while the bàtá encodes less phonetic detail for tone but encodes more information about vowel quality. In this talk, I show how musical surrogate languages reflect the practitioner’s nuanced understanding of their language’s sound system and offer a preliminary account of how linguistic, instrumental, and cultural constraints shape surrogate encoding.