How to Study a Tone Language, with Exemplification from Oku (Grassfields Bantu, Cameroon)

Abstract

UC Berkeley Phonology Lab Annual Report (2010) How To Study a Tone Language, with exemplification from Oku (Grassfields Bantu, Cameroon) Larry M. Hyman University of California, Berkeley DRAFT • August 2010 • Comments Welcome Introduction On numerous occasions I have been asked, “How does one study a tone language?” Or: “How can I tell if my language is tonal?” Even seasoned field researchers, upon confronting their first tone system, have asked me: “How do I figure out the number of tones I have?” When it comes to tone, colleagues and students alike often forget everything they’ve learned about discovering the phonology of a language and assume that tone is somehow different, that it requires different techniques or expertise. Some of this may derive from an incomplete understanding of what it means to be a tone system. Prior knowledge of possible tonal inventories, tone rules, and tone- grammar interfaces would definitely be helpful to a field researcher who has to decipher a tone system. However, despite such recurrent encounters, general works on tone seem not to answer these questions—specifically, they rarely tell you how to start and how to discover. I sometimes respond to the last question, “How do I figure out the number of tones I have?”, by asking in return, “Well, how would you figure out the number of vowels you have?” Hopefully the answer would be something like: “I would get a word list, starting with nouns, listen carefully, transcribe as much detail as I can, and then organize the materials to see if I have been consistent.” (I will put off until §5 commentary concerning the use of speech software as an aid in linguistic discovery.) Although I don’t think the elicitation techniques that one applies in studying segmental vs. tonal phonology are really very different, what is needed is a general discussion and illustration of how tonologists go about their work from beginning to end. My goal here is to share my personal experience with tonal elicitation in hopes that it will be useful to field researchers and students who are lucky enough to face an unknown tone system. I will not claim that every tonologist adopts the same strategies as I do, some of which I learned directly or indirectly from my teacher, Wm. E. Welmers, but I believe that most Africanists do. Logically, there are three separate tasks that one must take up in studying a tone system from scratch. Since these are necessarily ordered, with each one feeding into the next, I will refer to the three tasks as stages: (i) In Stage I the goal is to determine the surface tonal contrasts. This is first done by considering words in isolation. (ii) In Stage II the goal is to discover any tonal alternations (“morphotonemics”) which may exist in the language. This can be done either by putting words together to make short phrases or by eliciting paradigms. (iii) Stage III comprises the tonal analysis itself, the interpretation of what has been discovered in Stages I and II. At this point one typically draws on theoretical constructs and formal devices, e.g. autosegmental notation, to help express one’s insights as to how the tone system works.