*The phonetics and phonology of intonation in the light of
cross-linguistic evidence*
*Amalia Arvaniti*
/University of Kent/
In recent years, research on intonation has seen a revival but also a
new conflation of phonetics and phonology and the wide adoption of
atheoretical terms, such as /prominence/, which rely entirely on a
phonetic understanding of speech without recourse to phonological
structure. In this talk, I will review cross-linguistic findings on the
phonetics, phonology, and pragmatics of intonation, focusing on recent
research of mine on the phonetics and pragmatics of tunes used with
Greek wh-questions. These findings support the view that a principled
separation of phonetics and phonology is essential for understanding
intonation and disentangling gradient from categorical aspects of
realization. Further, they demonstrate how the study of intonational
pragmatics can shed light onto the phonological structure of intonation,
thereby providing evidence for the need to consider the two in tandem.
These findings will be discussed in light of recent research which
appears to cast doubt on some basic tenets of the Autosegmental-Metrical
theory of intonational phonology, such as the principles of tune-text
association. I argue that this research does not question these
principles to the extent it has been argued; rather, it highlights the
need for extensive cross-linguistic research, and the development of a
credible typology leading to plausible predictions about what is
possible in intonation. Moving away from models designed primarily for
English (or any one language for that matter), and acknowledging the
diverse ways in which intonation, and prosody more generally, is
organized, much as we acknowledge that some segmental phenomena and
associated formalisms do not apply to all languages, is essential for
progress in the study of intonation.
