Lexical vs. grammatical tone: Sorting out the differences

In this paper I raise the question of whether there are systematic differences between the lexical vs. grammatical functions of tone. Researchers of tone are largely influenced by the properties of the language(s) on which they work. Since tone has an almost exclusively lexical function in East and Southeast Asian languages, researchers of these languages focus primarily on how languages and speakers distinguish tone on lexical morphemes. Languages such Thai, Vietnamese and the wide range within Chinese lead one to posit a particular typology of lexical, monosyllabic tones which may interact with each or be influenced by the syntax, but which rarely function as direct exponents of grammar. At the other extreme are languages from other parts of the world where tone is primarily grammatical. These languages tend to treat their (sparse) tones syntagmatically, e.g. Somali and Chimwiini, where the grammar determines where a H tone will be realized on a word or phrase, encouraging many to feel they in fact represent a different typology altogether, often identified as “pitch accent”. In between these two endpoints are the majority of languages where tone exhibits both a lexical and grammatical function. While there are languages which segregate their tones by lexical vs. grammatical function, in most cases lexical and grammatical tones overlap and intersect in extremely interesting ways. After a brief discussion of general tonal typology, I will address the basic question of whether lexical tones and grammatical tones realize the same range of phenomena or whether there are things that one can do that the other can’t. I consider both word-level and phrase-level tonal phenomena to show that while there are some striking examples of grammatical tone run amok, there is in principle no difference in what lexical vs. grammatical tone can do.