Inaugural Kiparsky Seminar - Andrew Ollett (UChicago), "Preliminaries for a Generative Analysis of Kannada and Telugu Meter"

Preliminaries for a Generative Analysis of Kannada and Telugu Meter

The Kannada and Telugu languages, neighboring each other on India's Deccan Plateau, have shared a metrical system since the earliest available literature (9th/10th c. CE). Many components of this system are borrowed from other South Asian literatures (especially Sanskrit, Prakrit, and Apabhramsha), but one component appears to have been distinctive to Kannada and Telugu: often called the aṁśagaṇa system in modern scholarship, it includes meters built from a "hybrid" constituents, in which the first constituent is a moraic trochee (two light syllables or one heavy syllable), and the subsequent constituents are syllables unregulated as to weight. This paper will offer some "preliminaries" to a generative analysis of these meters, including the rather narrow empirical basis on which the analysis must be based, and an overview of the traditional analysis, which offers a useful starting point but, as has long been known, diverges from actual metrical practice in a number of important respects. I will cover the basic features of these meters, including their use of second-syllable alliteration and the tendency for metrical and word boundaries to coincide (precisely the opposite of what happens in the Sanskrit- and Prakrit-derived meters), and I will propose the outlines of a generative analysis.