LINGUIST List 16.642

Fri Mar 04 2005

Sum: Unmarked Contour Tone

Editor for this issue: Jessica Boynton <>

To post to LINGUIST, use our convenient web form at


        1.    Joaquim de Carvalho, Unmarked Contour Tone

Message 1: Unmarked Contour Tone

Date: 04-Mar-2005
From: Joaquim de Carvalho <>
Subject: Unmarked Contour Tone

Regarding query:

I've received six messages so far. Thanks to the people who responded:

Mark Donohue
Mike Cahill
Bart Mathias
Toby Paff
John E. Koontz
Stavros Skopeteas

Here is a brief summary of the replies:

Three points seem to appear from these messages, though, of course, the one
in (1) still lacks extensive empirical support:

1) Concerning languages where contour tones are not limited to long nuclei
(typically eastern Asian languages, but also, as pointed out by Mark
Donohue, Skou, New Guinea), if there is only one contour tone, this tone
shows a falling melody (see Donohue's and Cahill's postings).

2) This (hypothetical) implicational statement does not run for languages
where contour tones suppose bimoraic nuclei and/or (pitch) stress (see the
last replies). In these languages, the contour might be predictable in such
and such environment, but its shape may vary within the same language, and,
if it does not, it may be a rising contour as well as a falling one
depending on the language.

3) However, given bimoraicity and stress, it is not sure that such
languages can be shown to display the ternary tonal paradigm (H / L /
contour tone) I had in mind, which is more clearly illustrated by type (1).

Once again, many thanks for your responses.

Joaquim Brandao de Carvalho
Departement de linguistique
Universite Paris 8 / UMR 7023

Linguistic Field(s): Phonology
Respond to list|Read more issues|LINGUIST home page|Top of issue